• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Univ of Michigan Hydraulic Fracturing Technical Reports
 

Univ of Michigan Hydraulic Fracturing Technical Reports

on

  • 1,561 views

A series of seven reports (and an overview) produced by teams of faculty and students at the University of Michigan, part of a two-year project called the Hydraulic Fracturing in Michigan Integrated ...

A series of seven reports (and an overview) produced by teams of faculty and students at the University of Michigan, part of a two-year project called the Hydraulic Fracturing in Michigan Integrated Assessment. This series of seven reports establishes the current situation and provides an excellent backgrounder for hydraulic fracturing and the process of shale drilling. Michigan has significant quanities of shale gas, particularly in the Utica-Collingwood layer. The reports say that with the low price of natural gas, it will not be economical to mine Michigan's shale gas for some years to come.

Statistics

Views

Total Views
1,561
Views on SlideShare
1,500
Embed Views
61

Actions

Likes
0
Downloads
53
Comments
0

1 Embed 61

http://marcellusdrilling.com 61

Accessibility

Upload Details

Uploaded via as Adobe PDF

Usage Rights

CC Attribution-ShareAlike LicenseCC Attribution-ShareAlike License

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Univ of Michigan Hydraulic Fracturing Technical Reports Univ of Michigan Hydraulic Fracturing Technical Reports Document Transcript

    • Overview + Glossary H Y D R A U L I C F R A C T U R I N G I N T H E S T A T E O F M I C H I G A N
    • Participating University of Michigan Units Graham Sustainability Institute Erb Institute for Global Sustainable Enterprise Risk Science Center University of Michigan Energy Institute THIS PUBLICATION IS A RESULT OF WORK SPONSORED BY THE UNIVERSITY OF MICHIGAN Direct questions to grahaminstitute-ia@umich.edu
    • GRAHAM SUSTAINABILITY INSTITUTE INTEGRATED ASSESSMENT REPORT SERIES VOLUME II, REPORT 1 HYDRAULIC FRACTURING IN THE STATE OF MICHIGAN Overview + Glossary SEPTEMBER 3, 2013 WHAT IS THE ISSUE? T here is significant momentum behind natural gas extraction efforts in the United States, with many individual states embracing it as an opportunity to create jobs and foster economic strength. Natural gas extraction has also been championed as a way to move toward domestic energy independence and a cleaner energy supply. First demonstrated in the 1940’s, hydraulic fracturing is now the predom- inant method used to extract natural gas in the U.S. As domestic natural gas production has accelerated in recent years, however, the hydraulic fracturing process has come under increased public scrutiny. Concerns include perceived lack of transparency, chem- ical contamination, water availability, waste water disposal, and impacts on ecosystems, human health, and surrounding commu- nities. Consequently, numerous hydraulic fracturing studies are being undertaken by government agencies, industry, NGO’s, and academia, yet none have a particular focus on Michigan.
    • 2 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: OVERVIEW + GLOSSARY, SEPTEMBER 2013 What is Happening in Michigan? Recent interest from energy developers, lease sales, and permit- ting activities suggest increasing activity around deep shale gas extraction in Michigan. • Roughly 9,800 Antrim Shale wells are currently in production and hydraulic fracturing was used as part of the completion activity in virtually every one of these wells without incident. Most of these wells were drilled and completed in the late 1980s and early 1990s. Some new activity will still take place, and a very small number of the old wells may be hydraulically fractured in the future, but this is a “mature” play and is unlikely to be repeated. • The hydrocarbon resources in the Utica and Collingwood Shales in Northern Michigan will likely require hydraulic fracturing. • A May 2010 auction of state mineral leases brought in a record $178 million—nearly as much as the state had earned in the past 82 years of lease sales combined. Most of this money was spent for leases of State-owned mineral holdings with the Utica and Collingwood Shales as the probable primary targets. • Some ground water zones in Michigan are closer to gas zones than in other shale gas regions. It is significant that the Antrim Formation is only about 100 to about 1000 feet below the potential fresh water zones, and approximately 9,800 wells have been completed with no known contamination of fresh water zones to date. The Utica and Collingwood Shales are 3,000 to 10,000 feet below the fresh water zones. • State representatives have proposed packages of bills to regulate hydraulic fracturing, and state officials are reviewing existing regulations. Recognizing this context and that future hydraulic fracturing treatments will likely be of very high volume suggests a need for Michigan to be as well prepared as possible to manage this trend. What is being Done? Through a research-based partnership of University of Michigan (U-M) institutes, centers, and faculty, we are holistically evaluating the impacts of hydraulic fracturing in Michigan. Hydraulic fractur- ing has the potential to touch issues that all Michigan residents care about - drinking water, air quality, Great Lakes health, water supply, local land use, energy security, economic growth, tourism, and natural resource protection. This project’s technical analysis, stakeholder engagement, and proposed approaches to minimize negative impacts will be important outcomes that guide future decision making on this issue and hopefully help state decision makers avoid some of the pitfalls encountered in other states. The project is based on the premise that natural gas extraction pressures will likely increase in Michigan due to a desire for job creation, economic strength, energy independence, and cleaner fuels. What is Our Approach and Expertise? This project is using Integrated Assessment (IA) (http://graham. umich.edu/knowledge/ia), which is a useful method for analyzing environmental, social, and economic dimensions of challenging sustainability problems. The IA process achieves significant impact by leveraging interdisciplinary faculty expertise and engaging decision makers and stakeholders outside of academia to affect policy analysis and decision making. The figure above illustrates an IA framework focusing on hydrau- lic fracturing and its impact on Michigan’s communities, human health, and ecosystems. The project is: • Leveraging and building upon U-M’s existing relationships to facilitate successful partner and stakeholder engagement. • Drawing on key studies and regulatory approaches from across the country. Because hydraulic fracturing is thus far less con- tentious in Michigan, this project can be a platform to consider multiple stakeholder perspectives. • Acknowledging that hydraulic fracturing is likely to be part of Michigan’s future while providing analysis to address concerns and determine what strategies may be needed to improve the process. Ecosystem  Impacts Human Health  Impacts Community  Impacts Hydraulic  Fracturing in  Michigan Technology Policy Science Economics
    • 3 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: OVERVIEW + GLOSSARY, SEPTEMBER 2013 Currently identified U-M partners with relevant expertise include: • The Graham Environmental Sustainability Institute is a bound- ary organization connecting academics and policy-makers to address challenging sustainability problems. • The Risk Science Center is an interdisciplinary research and communication center in the School of Public Health that supports science-informed decision making on health risks. • The Energy Institute seeks to chart the path to a clean, afford- able and sustainable energy future through multi-disciplinary research. • The Erb Institute for Global Sustainable Enterprise is com- mitted to creating a socially and environmentally sustainable society through the power of business. How Long Will it Take? IA’s typically involve a 2 year timeline, the approach allows for flexi- bility and interim deliverables based on partners’ needs. PROCESS Phase 1: Technical Reports An effective IA in this context first requires compiling technical reports to provide a solid foundation of information for decision makers and stakeholders, and upon which the policy analysis can be built. These reports cover key issues within each topic related to hydraulic fracturing, and conclude with Michigan-specific ques- tions/issues for later analysis in Phase 2. Below are the primary topics which were identified for the technical reports and the lead authors for each report: • Technology: Johannes Schwank, Chemical Engineering; John Wilson, U-M Energy Institute • Geology/hydrogeology: Brian Ellis, Civil and Environmental Engineering • Environment/ecology: Allen Burton, School of Natural Resources & Environment; Knute Nadelhoffer, Department of Ecology and Evolutionary Biology • Human health: Nil Basu, School of Public Health • Policy/law: Sara Gosman, Law School • Economics: Roland Zullo, Institute for Research on Labor, Employment, & the Economy • Social/public perception: Andy Hoffman and Kim Wolske, Erb Institute for Global Sustainable Enterprise Each report considers a range of impacts/issues related to the primary topic. There may be overlaps of impacts/issues analyses as many of the items connect to multiple topics. Below is a non-ex- haustive list of possible impacts/issues which may be considered in the technical reports. While the IA has been developed to focus on High Volume Hydraulic Fracturing (HVHF) in Michigan (defined by the Michigan Department of Environmental Quality as hydrau- lic fracturing activity intended to use a total of more than 100,000 gallons of hydraulic fracturing fluid), data and analyses may cover a range of activity depending on topic or issue. Groundwater Impacts Health Impacts Surface Water Impacts Community Benefits/Impacts Risk Assessment State Economy Impact Air Quality Impacts Indirect Impacts (noise, traffic, roads) Fracturing Materials Catastrophic Events Federal-State-Local Policy Nexus Emergency Preparedness Process Innovations Public Perception Life Cycle Assessment Communications and Messaging Non-regulatory Strategies Local Land Use Policy Terrestrial and Aquatic System Impacts Lease Agreements/Good Neighbor Models Hydraulic Fracturing in Oil Production Management and Reuse of Flowback Water Methane Gas Releases On-site Diesel Emissions Phase 2: Integrated Assessment The IA will build from the technical reports, focus on identifying strategic policy options, and work to address the following guiding question: What are the best environmental, economic, social, and technological approaches for managing hydraulic fracturing in the State of Michigan? The IA will likely be formed around topics identified in the techni- cal reports and faculty authors from Phase 1 will likely be involved with the IA as leaders of topic specific analysis teams. However, new faculty may also become engaged at this point. Key aspects of the IA that will distinguish it from the technical reports include: • focus on the identification of key strategies and policy options, • collaboration and coordination across analysis teams to identify common themes and strategies, • regular engagement with decision makers, and • robust stakeholder engagement process to gauge public concerns and perceptions.
    • 4 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: OVERVIEW + GLOSSARY, SEPTEMBER 2013 Steering Committee The following steering committee has been assembled to guide project efforts including the configuration and structure of the IA during Phase 2: • Mark Barteau, Director, U-M Energy Institute • Valerie Brader, Senior Strategy Officer, Office of Strategic Policy, State of Michigan • John Callewaert, Int. Assessment Program Director, U-M Graham Sustainability Institute • James Clift, Policy Director, Michigan Environmental Council • John De Vries, Attorney, Mika Meyers Beckett & Jones; Michigan Oil and Gas Association • Hal Fitch, Director of Oil, Gas, and Minerals, Michigan Department of Environmental Quality • Gregory Fogle, Owner, Old Mission Energy; Michigan Oil and Gas Association • James Goodheart, Senior Policy Advisor, Michigan Department of Environmental Quality • Andy Hoffman, Director, U-M Erb Institute for Global Sustainable Enterprise • Drew Horning, Deputy Director, U-M Graham Sustainability Institute • Andrew Maynard, Director, U-M Risk Science Center • Tammy Newcomb, Senior Water Policy Advisor, Michigan Department of Natural Resources • Don Scavia, Director, U-M Graham Sustainability Institute • Tracy Swinburn, Managing Director, U-M Risk Science Center • Grenetta Thomassey, Program Director, Tip of the Mitt Watershed Council • John Wilson, Consultant, U-M Energy Institute The role of the steering committee is to provide broad stakeholder input and guidance to the overall IA process and to ensure the scope of study is relevant to key decision makers. Committee members may also provide data and input to research teams throughout the process, but decisions regarding content of project analyses and reports are determined by the researchers. Engagement The IA will be informed by semi-annual meetings with analysis teams and the steering committee for project updates and dis- cussions. Twice during the IA, these meetings will involve a larger group of decision makers and stakeholders. An online comments/ ideas submission site has been established to direct public input to the steering committee and analysis teams: http://graham.umich. edu/knowledge/ia/hydraulic-fracturing Funding At present, the IA is entirely funded by the University of Michigan. The project is expected to cost at least $600,000 with support coming from the University of Michigan’s Graham Institute, Energy Institute and Risk Science Center. Current funding sources are limited to the U-M general fund and gift funds, all of which are governed solely by the University of Michigan. As the project develops, the Graham Institute may seek additional funding to expand stakeholder engagement efforts. All funding sources will be publicly disclosed. Timeline • Mid May 2013: steering committee and technical report leads meet to discuss technical reports and plans for the Integrated Assessment • Early September 2013: technical reports are released with 30 day public comment period for ideas and questions for the Integrated Assessment • Early Fall 2013: plans are developed for the Integrated Assessment • Mid 2014: final Integrated Assessment report released (tentative) Direct comments or questions to: grahaminstitute-ia@umich.edu
    • 5 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: OVERVIEW + GLOSSARY, SEPTEMBER 2013 AIR QUALITY. A measure of the amount of pollutants emitted into the atmosphere and the dispersion potential of an area to dilute those pollutants. AQUIFER. A body of rock that is sufficiently permeable to con- duct groundwater and to yield economically significant quantities of water to wells and springs. BASIN. A closed geologic structure in which the beds dip toward a central location; the youngest rocks are at the center of a basin and are partly or completely ringed by progressively older rocks. BIOGENIC GAS. Natural gas produced by living organisms or biological processes. CASING. Steel piping positioned in a wellbore and cemented in place to prevent the soil or rock from caving in. It also serves to isolate fluids, such as water, gas, and oil, from the surrounding geologic formations. COAL BED METHANE/NATURAL GAS (CBM/CBNG). A clean‐burning natural gas found deep inside and around coal seams. The gas has an affinity to coal and is held in place by pres- sure from groundwater. CBNG is produced by drilling a wellbore into the coal seam(s), pumping out large volumes of groundwater to reduce the hydrostatic pressure, allowing the gas to dissociate from the coal and flow to the surface. COMPLETION. The activities and methods to prepare a well for production and following drilling. Includes installation of equip- ment for production from a gas well. CONVENTIONAL NATURAL GAS. Natural gas comes from both ‘conventional’ (easier to produce) and ‘unconventional’ (more difficult to produce) geological formations. The key difference between “conventional” and “unconventional” natural gas is the manner, ease and cost associated with extracting the resource. Exploration for conventional gas has been almost the sole focus of the oil and gas industry since it began nearly 100 years ago. Conventional gas is typically “free gas” trapped in multiple, relatively small, porous zones in various naturally occurring rock formations such as carbonates, sandstones, and siltstones. CORRIDOR. A strip of land through which one or more existing or potential utilities may be colocated. DISPOSAL WELL. A well which injects produced water into an underground formation for disposal. DIRECTIONAL DRILLING. The technique of drilling at an angle from a surface location to reach a target formation not located directly underneath the well pad. DRILL RIG. The mast, draw works, and attendant surface equip- ment of a drilling or workover unit. EMISSION. Air pollution discharge into the atmosphere, usually specified by mass per unit time. ENDANGERED SPECIES. Those species of plants or animals clas- sified by the Secretary of the Interior or the Secretary of Commerce as endangered pursuant to Section 4 of the Endangered Species Act of 1973, as amended. See also Threatened and Endangered Species. EXPLORATION. The process of identifying a potential subsur- face geologic target formation and the active drilling of a borehole designed to assess the natural gas or oil. FLOW LINE. A small diameter pipeline that generally connects a well to the initial processing facility. FORMATION (GEOLOGIC). A rock body distinguishable from other rock bodies and useful for mapping or description. Formations may be combined into groups or subdivided into members. FRACTURING FLUIDS. A mixture of water and additives used to hydraulically induce cracks in the target formation. GROUND WATER. Subsurface water that is in the zone of sat- uration; source of water for wells, seepage, and springs. The top surface of the groundwater is the “water table.” HABITAT. The area in which a particular species lives. In wildlife management, the major elements of a habitat are considered to be food, water, cover, breeding space, and living space. HYDRAULIC FRACTURING GLOSSARY OF COMMONLY USED TERMS1 1. General sources include: • “Modern Shale Gas Development,” a Department of Energy Report: www.eogresources.com/responsibility/doeModernShaleGasDevelopment.pdf • The Canadian Association of Petroleum Products: www.capp.ca/ CANADAINDUSTRY/NATURALGAS/CONVENTIONAL-UNCONVENTIONAL/ Pages/default.aspx • The Union of Concerned Scientists: www.ucsusa.org/clean_energy/ our-energy-choices/coal-and-other-fossil-fuels/how-natural-gas-works.html
    • 6 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: OVERVIEW + GLOSSARY, SEPTEMBER 2013 HIGH VOLUME HYDRAULIC FRACTURING. High volume hydraulic fracturing well completion is defined by the State of Michigan as a “well completion operation that is intended to use a total of more than 100,000 gallons of hydraulic fracturing fluid.”2 HORIZONTAL DRILLING. A drilling procedure in which the wellbore is drilled vertically to a kickoff depth above the target formation and then angled through a wide 90 degree arc such that the producing portion of the well extends horizontally through the target formation. HYDRAULIC FRACTURING. Injecting fracturing fluids into the target formation at a force exceeding the parting pressure of the rock thus inducing a network of fractures through which oil or nat- ural gas can flow to the wellbore. HYDROSTATIC PRESSURE. The pressure exerted by a fluid at rest due to its inherent physical properties and the amount of pressure being exerted on it from outside forces. INJECTION WELL. A well used to inject fluids into an under- ground formation either for enhanced recovery or disposal. LEASE. A legal document that conveys to an operator the right to drill for oil and gas. Also, the tract of land, on which a lease has been obtained, where producing wells and production equipment are located. NORM (Naturally Occurring Radioactive Material). Low‐level, radioactive material that naturally exists in native materials. ORIGINAL GAS IN PLACE. The entire volume of gas contained in the reservoir, regardless of the ability to produce it. PARTICULATE MATTER (PM). A small particle of solid or liquid matter (e.g., soot, dust, and mist).PM10 refers to particu- late matter having a size diameter of less than 10 millionths of a meter (micrometer)and PM2.5 being less than 2.5 micro‐meters in diameter. PERMEABILITY. A rock’s capacity to transmit a fluid; dependent upon the size and shape of pores and interconnecting pore throats. A rock may have significant porosity (many microscopic pores) but have low permeability if the pores are not interconnected. Permeability may also exist or be enhanced through fractures that connect the pores. PRIMACY. A right that can be granted to state by the federal government that allows state agenciesto implement programs with federal oversight. Usually, the states develop their own set of regulations. By statute, states may adopt their own standards, however, these must be at least as protective as the federal stan- dards they replace, and may be even more protective in order to address local conditions. Once these state programs are approved by the relevant federal agency (usually the EPA), the state then has primacy jurisdiction. PRODUCED WATER. Water produced from oil and gas wells. PROPPING AGENTS/PROPPANT. Silica sand or other particles pumped into a formation during a hydraulic fracturing operation to keep fractures open and maintain permeability. PROVED RESERVES. That portion of recoverable resources that is demonstrated by actual production or conclusive formation tests to be technically, economically, and legally producible under existing economic and operating conditions. RECLAMATION. Rehabilitation of a disturbed area to make it acceptable for designated uses. This normally involves regrading, replacement of topsoil, re‐vegetation, and other work necessary to restore it. SETBACK. The distance that must be maintained between a well or other specified equipment and any protected structure or feature. SHALE GAS. Natural gas produced from low permeability shale formations. SLICKWATER. A water based fluid mixed with friction reducing agents, commonly potassium chloride. SOLID WASTE. Any solid, semi‐solid, liquid, or contained gas- eous material that is intended for disposal. SPLIT ESTATE. Condition that exists when the surface rights and mineral rights of a given area are owned by different persons or entities; also referred to as “severed estate”. STIMULATION. Any of several processes used to enhance near wellbore permeability and reservoir permeability. STIPULATION. A condition or requirement attached to a lease or contract, usually dealing with protection of the environment, or recovery of a mineral. 2. Department of Environmental Quality, Supervisor of Wells Instruction 1-2011 (2011), available at www.michigan.gov/documents/deq/SI_1-2011_353936_7.pdf (effective June 22, 2011). Michigan.
    • 7 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: OVERVIEW + GLOSSARY, SEPTEMBER 2013 SULFUR DIOXIDE (SO2). A colorless gas formed when sulfur oxidizes, often as a result of burning trace amounts of sulfur in fossil fuels. TECHNICALLY RECOVERABLE RESOURCES. The total amount of resource, discovered and undiscovered, that is thought to be recoverable with available technology, regardless of economics. THERMOGENIC GAS. Natural gas that is formed by the com- bined forces of high pressure and temperature (both from deep burial within the earth’s crust), resulting in the natural cracking of the organic matter in the source rock matrix. THREATENED AND ENDANGERED SPECIES. Plant or animal species that have been designated as being in danger of extinc- tion. See also Endangered Species. TIGHT GAS. Natural gas trapped in a hardrock, sandstone or limestone formation that is relatively impermeable. TOTAL DISSOLVED SOLIDS (TDS). The dry weight of dis- solved material, organic and inorganic, contained in water and usually expressed in parts per million. UNCONVENTIONAL NATURAL GAS. Natural gas comes from both ‘conventional’ (easier to produce) and ‘unconventional’ (more difficult to produce) geological formations. The key differ- ence between “conventional” and “unconventional” natural gas is the manner, ease and cost associated with extracting the resource. However, most of the growth in supply from today’s recoverable gas resources is found in unconventional formations. Unconventional gas reservoirs include tight gas, coal bed methane, gas hydrates, and shale gas. The technological breakthroughs in horizontal drill- ing and fracturing are making shale and other unconventional gas supplies commercially viable. UNDERGROUND INJECTION CONTROL PROGRAM (UIC). A program administered by the Environmental Protection Agency, primacy state, or Indian tribe under the Safe Drinking Water Act to ensure that subsurface emplacement of fluids does not endanger underground sources of drinking water. UNDERGROUND SOURCE OF DRINKING WATER (USDW). 40 CFR Section 144.3 An aquifer or its portion: (a) (1) Which supplies any public water system; or (2) Which contains a sufficient quantity of ground water to supply a public water system; and (i) Currently supplies drinking water for human consumption; or (ii) Contains fewer than 10,000 mg/l total dissolved solids; and (b) Which is not an exempted aquifer. WATER QUALITY. The chemical, physical, and biological char- acteristics of water with respect to its suitability for a particular use. WATERSHED. All lands which are enclosed by a continuous hydrologic drainage divide and lay upslope from a specified point on a stream. WELL COMPLETION. See Completion. WORKOVER. To perform one or more remedial operations on a producing or injection well to increase production. Deepening, plugging back, pulling, and resetting the liner are examples of workover operations.
    • © 2013 BY THE REGENTS OF THE UNIVERSITY OF MICHIGAN MARK J. BERNSTEIN, ANN ARBOR JULIA DONOVAN DARLOW, ANN ARBOR LAURENCE B. DEITCH, BLOOMFIELD HILLS SHAUNA RYDER DIGGS, GROSSE POINTE DENISE ILITCH, BINGHAM FARMS ANDREA FISCHER NEWMAN, ANN ARBOR ANDREW C. RICHNER, GROSSE POINTE PARK KATHERINE E. WHITE, ANN ARBOR MARY SUE COLEMAN, EX OFFICIO Please print sparingly and recycle
    • Technology Technical Report H Y D R A U L I C F R A C T U R I N G I N T H E S T A T E O F M I C H I G A N
    • Participating University of Michigan Units Graham Sustainability Institute Erb Institute for Global Sustainable Enterprise Risk Science Center University of Michigan Energy Institute ABOUT THIS REPORT This document is one of the seven technical reports com- pleted for the Hydraulic Fracturing in Michigan Integrated Assessment conducted by the University of Michigan. During the initial phase of the project, seven faculty-led and student-staffed teams focused on the following topics: Technology, Geology/ Hydrogeology, Environment/Ecology, Human Health, Policy/ Law, Economics, and Public Perceptions. These reports were prepared to provide a solid foundation of information on the topic for decision makers and stakeholders and to help inform the Integrated Assessment, which will focus on the analysis of policy options. The reports were informed by comments from (but do not necessarily reflect the views of) the Integrated Assessment Steering Committee, expert peer reviewers, and numerous public comments. Upon completion of the peer review process, final decisions regarding the content of the reports were deter- mined by the faculty authors in consultation with the peer review editor. These reports should not be characterized or cited as final products of the Integrated Assessment. The reports cover a broad range of topics related to hydraulic fracturing in Michigan. In some cases, the authors determined that a general discussion of oil and gas development is important to provide a framing for a more specific discussion of hydraulic fracturing. The reports address common hydraulic fracturing (HF) as meaning use of hydraulic fracturing methods regardless of well depth, fluid volume, or orientation of the well (whether vertical, directional, or horizontal). HF has been used in thousands of wells throughout Michigan over the past several decades. Most of those wells have been shallower, vertical wells using approxi- mately 50,000 gallons of water; however, some have been deeper and some have been directional or horizontal wells. The reports also address the relatively newer high volume hydraulic fracturing (HVHF) methods typically used in conjunction with directional or horizontal drilling. An HVHF well is defined by the State of Michigan as one that is intended to use more than 100,000 gallons of hydraulic fracturing fluid. Finally, material in the technical reports should be understood as providing a thorough hazard identification for hydraulic fracturing, and when appropriate, a prioritization according to likelihood of occurrence. The reports do not provide a scientific risk assess- ment for aspects of hydraulic fracturing.
    • GRAHAM SUSTAINABILITY INSTITUTE INTEGRATED ASSESSMENT REPORT SERIES VOLUME II, REPORT 2 HYDRAULIC FRACTURING IN THE STATE OF MICHIGAN The Application of Hydraulic Fracturing Technologies to Michigan Oil and Gas Recovery SEPTEMBER 3, 2013 Faculty Leads JOHN R. WILSON CONSULTANT, UNIVERSITY OF MICHIGAN ENERGY INSTITUTE JOHANNES W. SCHWANK PROFESSOR, DEPARTMENT OF CHEMICAL ENGINEERING TABLE OF CONTENTS 2 Executive Summary 3 1.0 Introduction 4 2.0 Status and Trends 12 3.0 Challenges and Opportunities 24 4.0 Prioritized Pathways For Phase 2 24 Literature Cited THIS PUBLICATION IS A RESULT OF WORK SPONSORED BY THE UNIVERSITY OF MICHIGAN Direct questions to grahaminstitute-ia@umich.edu
    • 2 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 EXECUTIVE SUMMARY T his report focuses on technical issues related to hydraulic fracturing or “fracking” technologies and related methods of oil and gas recovery with special emphasis on methods that find applications in the State of Michigan. The report also identifies technical issues in the area of hydraulic fracturing that may require additional research. A brief review of the history of oil and gas recovery in Michigan is included, and the Michigan-based activities are discussed and contrasted with other U.S. and Canadian hydraulic fracturing activities. Since Michigan has limited experience with deep and/or directional drilling, this report also draws on the experience devel- oped in other states. Michigan, compared to other states in the U.S., has been a mid- dle-of-the-pack producer of both oil and gas for many years. After the first commercially successful recovery of oil in Michigan in the Saginaw field in 1925, oil was also found near Muskegon, between Midland and Mount Pleasant, in the northern Niagaran Reef struc- ture and along a trend between Albion and Scipio. Oil production in Michigan peaked in 1981-83 at about 32 million barrels/year but has since than declined by about 50%1 . Natural gas production started later in Michigan, mostly in the Antrim Shale and in the northern Niagaran Reef, and peaked in 1996 at almost 300 billion cu. ft./year. Michigan’s natural gas production has since fallen steadily to a little under 150 billion cu. ft./year2,3 . For comparison, the U.S. uses about 18 million barrels of oil daily (6.6 billion barrels annually) and about 23 trillion cu. ft. of natural gas annually4 . In the past 30 years, there have been no significant new oil finds in Michigan. However, considerable reserves of natural gas are believed to exist in deep shale formations such as the Utica- Collingwood, which underlies much of Michigan and eastern Lake Huron and extends well down into Ontario, Canada. Despite attempts dating back as far as 1859 in both Michigan and Ontario to extract gas, gas liquids and even oil from this very tight formation, there has been no successful commercial development to date. A few promising finds in Kalkaska County in Michigan, on Manitoulin Island in Lake Huron, and on the Southern Ontario mainland as far south as Niagara Falls have not yet led to commercial devel- opment. In view of the currently low price of gas, the high cost of drilling these deep shales, and the absence of new oil discoveries, it is unlikely that there will be significant growth of the oil and gas industry in Michigan (or Ontario) in the near-term future. High-pressure (usually deep well) hydraulic fracturing (HF) rep- resents one of many widely used methods of enhancing or ini- tiating oil and gas recovery from deep, tight formations5 . It has not found widespread application in Michigan, except for a few exploratory wells in the Utica/Collingwood and the associated A-1 and A-2 Carbonates. However, HF has been used in the form of low-pressure nitrogen foam fracking and also low-pressure water fracking in the Antrim Shale in the northern Lower Peninsula since the late 1940s (Hal Fitch, Michigan Department of Environmental Quality, pers. comm.). Hydraulic fracking originated in 1947-1949, initially in Kansas, Oklahoma, and Texas as a means of stimulating production from uneconomic gas and (mostly) oil wells, and was quickly success- ful at increasing production rates by 50% or more, typically using hydrocarbon fluids (not water) as the carrier. Fracking now involves water mixed with at least 9-10% of sand or a synthetic ceramic such as calcined bauxite. The sand or ceramic particles are dispersed in the water to help keep the cracks in the formation open; the water also contains about 0.5 % of a total of about 10 chemical additives (such as surfactants and antibacterial agents similar to those used in dishwashing detergents) to help keep the newly-formed cracks open and clean. In the past, far less environmentally benign chem- icals were added but the use of these has been discontinued by all of the major operators and their sub-contractors, partly as a result of public pressure and greater state disclosure requirements. As noted, hydraulic fracturing was first performed experimentally in 1947 and the first commercial “frac job” was performed in 1949. As of 2010, it was estimated that 60% of all new oil and gas wells worldwide were being hydraulically fractured6 . Many of these early fracking jobs were a mixture of stimulation of oil and gas produc- tion from existing under-performing wells and the development of new wells in “tight” formations from which commercially accept- able oil and gas flows could not otherwise be obtained. As of 2012, it is estimated that 2.5 million hydraulic fracturing jobs of all kinds have been performed on oil and gas wells worldwide, over one million of them in the United States7 . To date in the U.S., fracking technologies are estimated to have been applied to more than 1.25 million vertical or directional oil or gas wells. Canadian com- panies have fracked at least another 200,000 wells8 . In many recent cases, a combination of directional drilling and high-pressure multi-stage fracking has been used to access oil or gas trapped in larger ‘drainage volumes’ of a reservoir. Modern high-pressure HF is generally applied to deep, often directional wells and uses what are often perceived as high vol- umes of water (typically up to 7 million gallons per well although in a very small number of cases, including one in Michigan, quantities over 20 million gallons have been reported, usually associated with unusually low flowback water recoveries and apparently associated with abnormal “sinks” for water deep underground). Compared to other industrial or agricultural uses, these volumes of water are not large, but water availability tends to be a local or regional
    • 3 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 problem, and its use for fracking has raised concerns especially in the western U.S. To decrease the use of water, several non-aque- ous fracking methods are now in use or being developed. A more serious problem is disposal or treatment of the often-substantial fraction of the fracking water returned as so-called flowback water and also of any subsequently produced water. Flowback or pro- duced water is now often (as in Michigan) disposed of in Federal or State approved deep injection wells. An increasing and so far partially successful effort is being made to develop better water treatment methods for the often highly saline return water which may also contain small amounts of hydrocarbons, some of them toxic9 . If these treatment methods are effective, the water can be re-used—and in some cases is, in Colorado and Pennsylvania, for example. Another concern for the natural gas industry is potential leakage of methane. Methane is a potent greenhouse gas. Over the years, substantial efforts have been made to gradually decrease the num- ber of both large and small leaks in the national distribution sys- tem. Newly designed pipeline compressors, once a major source of methane leaks, are now essentially leak-proof while gas process- ing plant hardware and instrumentation is improving through the use of welded joints and changes in design. In the past, fracked gas well sites used to be fairly major contributors to methane leaks due to careless handling of flowback water and practices such as open-well liquids unloading and incomplete combustion in flares. Field monitoring of methane emissions from such sites now shows them to be comparable to conventional gas wells producing under reservoir pressure, and field levels of methane leaking from HF sites are now generally low, as was very recently confirmed by the U.S. Environmental Protection Agency (EPA)10 . Although methane leakage remains a concern for the natural gas industry in general, the probability of significant methane leakage in deep shale drill- ing, completion, hydraulic fracturing, testing, and production in Michigan is quite low provided that best practices are adhered to. However, local distribution systems in older cities are still thought to be a major source of methane leakage. Fracking, like oil or gas drilling, involves complex equipment and procedures operated by humans. Errors and accidents do occa- sionally occur, sometimes leading to the escape of fracking water or, much more often, gas into the atmosphere or into groundwater or drinking water aquifers. Fortunately, such events have become increasingly rare over the past ten years as both regulations and industry practices have improved. Most recent incidents have involved faulty equipment or its faulty installation. This report reviews the safety record accumulated over more than 30 years of high-pressure deep well fracking (and a much longer period of all forms of fracking) and arrives at the conclusion that the fracking process has a good safety record. Phase 2 work that is proposed includes a long-overdue study of the adsorption of natural gas components on minerals that are found in Michigan’s gas reservoirs as well as a more quantitative look at the physical characteristics of the Collingwood, Utica, and related shales that are thought to be important to Michigan’s natural gas future. 1.0 INTRODUCTION A lthough Michigan has long been a moderately prolific (albeit now declining) producer of oil and gas, in common with many other states, it is in most ways geologically unique. While it has some charac- teristics in common with neighboring Indiana, Ohio and Ontario, Canada, the history of “fracking” in other states such as Ohio, Pennsylvania, New York, Texas, Colorado or Wyoming has limited relevance in Michigan. Among American states, Michigan has been a middle-of-the-pack producer of both oil and gas for many years. This report will combine that part of out-of-state experience that is relevant to Michigan with the state’s 100+ years of in-state discovery and production of oil and gas. It will provide an analy- sis of the past, present, and likely future of the use of formation drilling and fracturing technologies to enhance natural gas and oil production in the state. The first commercial discovery of oil in Michigan was made in the Saginaw field in 1925. This was followed by many other finds near Muskegon, between Midland and Mount Pleasant, in the northern Niagaran Reef structure and along a trend between Albion and Scipio11 . Oil production state-wide increased steadily and peaked in 1981-83 at about 32 million barrels/year but has declined by more than 50% since that time3 . Natural gas production devel- oped somewhat later in Michigan, mostly in the Antrim Shale and in the northern Niagaran Reef, and grew steadily until 1996 when it peaked at almost 300 billion cu. ft./year. Michigan’s natural gas production has since fallen to a little under 150 billion cu. ft./year. For comparison, the U.S. uses about 18 million barrels of oil daily (6.6 billion barrels annually) and about 23 trillion cu. ft. of natural gas annually7 . No significant new finds of oil have been made in Michigan in the past 30 years. Additional natural gas is thought to exist in deep shale formations such as the Utica-Collingwood, which underlies much of Lower Michigan and Lake Huron and extends well down into Ontario, Canada. Attempts have been made in both Michigan and Ontario to extract gas, gas liquids and even oil from this very tight formation dating back to 1859, but so far there has been no successful commercial development. There have, however, been one or two promising (but so far undeveloped) finds in several
    • 4 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 areas such as Kalkaska County in Michigan and on Manitoulin Island in Lake Huron and on the Southern Ontario mainland as far south as Niagara Falls. Notwithstanding these positive indications, the low price of gas, the high cost of drilling these shales, and the absence of new oil finds do not bode well for the near-term future of the oil and gas industry in Michigan (or Ontario). 2.0 STATUS AND TRENDS 2.1 A Brief History of Oil and Gas in Michigan and Vicinity The following map shows the well-established bedrock geology of Michigan12 . The map shows the irregular “stack of dinner plates” characteristic of Michigan geology which has resulted in formation Figure 1: Geology of Michigan12 KEWEENAW HOUGHTON ONTONAGON BARAGA MARQUETTE GOGEBIC CHIPPEWA LUCE ALGER SCHOOLCRAFT IRON DICKINSON MACKINAC DELTA MENOMINEE EMMET CHEBOYGAN PRESQUE ISLE CHARLEVOIX ALPENA MONTMORENCY LEELANAU OTSEGO ANTRIM GRAND TRAVERSE ALCONAOSCODACRAW FORDKALKASKA BENZIE IOSCOOGEMAWROSCOMMONMANISTEE MISSAUKEEWEXFORD ARENAC MASON GLADWINCLAREOSCEOLALAKE HURON BAY MIDLANDISABELLAOCEANA MECOSTA NEWAYGO TUSCOLA SANILAC SAGINAW GRATIOTMUSKEGON MONTCALM LAPEER KENT GENESEE ST CLAIR OTTAW A SHIAWASSEE CLINTONIONIA MACOMB OAKLAND LIVINGSTONINGHAMEATONBARRYALLEGAN WAYNE WASHTENAWJACKSONCALHOUNKALAMAZOOVAN BUREN BERRIEN MONROE LENAWEEHILLSDALE BRANCHST JOSEPHCASS BEDROCK GEOLOGY OF LOWER PENINSULA SALINA GROUP BASS ISLAND GROUP GARDEN ISLAND FORMATION BOIS BLANC FORMATION MACKINAC BRECCIA SYLVANIA SANDSTONE DETROIT RIVER GROUP DUNDEE LIMESTONE BELL SHALE TRAVERSE GROUP ANTRIM SHALE ELLSWORTH SHALE BEDFORD SHALE BEREA SS & BEDFORD SH SUNBURY SHALE COLDWATER SHALE MARSHALL FORMATION MICHIGAN FORMATION BAYPORT LIMESTONE SAGINAW FORMATION GRAND RIVER FORMATION RED BEDS BEDROCK GEOLOGY OF WESTERN UPPER PENINSULA JACOBSVILLE SANDSTONE FREDA SANDSTONE NONESUCH FORMATION COPPER HARBOR CONGLOMERATE OAK BLUFF FORMATION PORTAGE LAKE VOLCANICS SIEMENS CREEK FORMATION INTRUSIVE QUINNESEC FORMATION PAINT RIVER GROUP RIVERTON IRON FORMATION BIJIKI IRON FORMATION NEGAUNEE IRON FORMATION IRONWOOD IRON FORMATION DUNN CREEK FORMATION BADWATER GREENSTONE MICHIGAMME FORMATION GOODRICH QUARTZITE HEMLOCK FORMATION MENOMINEE & CHOCOLAY GROUPS EMPEROR VULCANIC COMPLEX SIAMO SLATE & AJIBIK QUARTZITE PALMS FORMATION CHOCOLAY GROUP RANDVILLE DOLOMITE ARCHEAN ULTRAMAFIC ARCHEAN GRANITE & GNEISSIC ARCHEAN VOL. & SEDIMENTARY MACKINAC BRECCIA BEDROCK GEOLOGY OF EASTERN UPPER PENINSULA MUNISING FORMATION TREMPEALEAU FORMATION PRAIRIE DU CHIEN GROUP BLACK RIVER GROUP TRENTON GROUP COLLINGWOOD SHALE MEMBER UTICA SHALE MEMBER STONINGTON FORMATION BIG HILL DOLOMITE QUEENSTON SHALE MANITOULIN DOLOMITE CABOT HEAD SHALE BURNT BLUFF GROUP MANISTIQUE GROUP ENGADINE GROUP POINT AUX CHENES SHALE SAINT IGNACE DOLOMITE SALINA GROUP BASS ISLAND GROUP GARDEN ISLAND FORMATION BOIS BLANC FORMATION MACKINAC BRECCIA 0 20 40 MilesDate: 11/12/99 N Michigan MICHIGAN DEPARTMENT O FNATURAL RESOU RCES LAND AND MINERALS SERVICES DIVISION RESOURCE MAPPING AND AERIAL PHO TOG RAPHY Michigan Resource Information System Part 609, Resource Inventory, of the Natural Resources and Environmental Protection Act, 1994 PA 451, as amended. Automated from "Bedrock Geology of Mi chi gan," 1987, 1:500,000 scal e, which was compiled from a vari ety of sources by the Michigan Department of Environmental Quality, Geological Survey Division. SOURCE RMAP 1987 BEDROCK GEOLOGY OF MICHIGAN
    • 5 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 outcrops quite remote from the state itself—in Ontario or under Lake Michigan, for example. Figure 2 shows the location of oil (green) and gas (red) wells in Michigan. The red area at the top of the Lower Peninsula represents the Antrim shale formation. This is still a major gas producer, but at only about half of its former peak rate. Other oil and gas wells are distributed over the state in a manner that more or less follows the geology shown in the preceding map, but its location has been sufficiently unpredictable to have made wildcatting a high-risk occupation in the state for many years! By odd coincidence, the first discovery of oil related to Michigan’s rather unusual geology was made in 1858 at what is now Oil Springs, Ontario, where there were long-known “gum” beds (the gum being the residue left after the lighter fractions of the natu- rally seeping crude had evaporated). This was where Michigan’s Dundee limestone formation (or ‘dinner plate’) outcropped at the edge of the Michigan Basin. A hole was dug to a depth of 13 ft. (later deepened to 39 ft.) and free-flowing crude oil was observed. This occurred a year before Edward Drake’s famous 1859 well at Titusville, PA, which is usually considered the forerunner of the U.S oil industry. Michigan’s first recorded oil field was discovered in St. Clair County in 1886 and also tapped the Dundee formation at 575 ft. The oil recovered was used locally “as was”, mostly for the lubrication of heavy machinery. The last of this field’s oil wells was plugged in 1921. Oil was sought by drilling in the Upper Peninsula (UP) in the very early 1900s, but exploration in the UP has never resulted in a commercially-viable oil or gas find. In Michigan, the Collingwood/Utica formation that lies deep under much of the state (at 10,000–12,000 ft.) has not yet proven to be a commercial source of oil or gas. However, as noted above, the Collingwood also underlies much of Lake Huron as well as Manitoulin Island and the Bruce Peninsula of Ontario and gets its name from the town of Collingwood, ON, on the SE shore of Lake Huron (where the formation forms the shoreline but is overlain just further south by the Blue Mountain shale). Oil was produced from the Collingwood at the tiny town of Craigleith just west of Collingwood in 185914 . The enterprise failed by 1863. The oil in that part of the Collingwood is the high-kerogen variety also found in the Green River Basin in Wyoming, Colorado and Utah15 . It can be extracted only by retorting the shale and condensing the vapor- ized product—a process that even now is not economically viable. There are also numerous developmental gas wells in the Ontario sector of the Collingwood, including some on Manitoulin Island. Fortunately, most of the oil discovered in Michigan has been much more conventional in nature and therefore much more accessible. However, there have been few recent discoveries and the state’s production of both oil and gas is now in sharp decline. The widely-accepted beginning of Michigan’s commercial oil production began with the discovery of the Saginaw Field, just south of the city of Saginaw, in 1925. The period from 1925 to 1955 resulted in numerous oil and gas discoveries throughout the Lower Peninsula, along with a lot of dry holes. Of greatest note was the Muskegon field (1927) that was followed by several additional sig- nificant finds a little farther south and east in Ottawa, Allegan, and Van Buren counties. Oil was also discovered between what are now Mount Pleasant and Midland in a quite prolific area that involved Osceola, Clare, Gladwin, Midland, Isabella, and Mecosta counties. Most of the wells were shallow and produced only oil from the sandstone that underlaid the famous “red beds”; nevertheless, the result made Mount Pleasant the “Oil Capital of Michigan”. The Albion-Scipio Trend in Calhoun and Hillsdale counties was dis- covered in the mid-1950s and produced 125 million bbl. of oil, thus qualifying as a major field. (There was probably at least three times that amount of oil originally in place; enhanced oil recovery meth- ods will eventually recover more of it. A small amount of gas was OIL AND GAS WELLS SOUTHERN PENINSULA OF MICHIGAN ± 0 10 20 30 405 Miles ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! !! !! ! !! !!! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! !! ! !! !!! !! !! ! ! ! !! ! ! ! !! ! ! !! !!!!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! !! ! ! !! ! ! !! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! ! !! ! !! ! !! !! !! ! ! ! ! ! ! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !!! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !!! !! !! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !! !! !! ! !!! ! ! ! ! ! !!! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !!! ! !! !! !! !! ! ! ! ! ! ! ! !! !! !!! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! !!!!! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! !! ! !! ! ! ! !! ! ! !! ! !!!! ! ! !!! !!! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! !!!! !! !! ! !!!! ! ! ! !! !! !! ! !! !!! !! ! !! ! ! ! ! !! !! ! ! !! ! !!! ! ! !! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! !! ! ! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! !! ! !! ! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! !! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !!! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! !!! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !! ! !! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !!! ! ! !!!! ! ! ! !!!!!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! !!! ! !! ! ! !!! ! ! ! ! !!! !! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! !!! ! ! !! ! ! ! !! ! !! !! !! !!! !! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !!! ! ! ! !!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!!! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !!!! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! !! !!! !! ! ! !!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! !! !! !! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !!!! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! !!! ! !! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! !! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !!!! ! ! !!! ! !!! ! !!! ! !! ! !! !!! !! !!!!! ! ! !! ! ! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! !! ! ! ! ! !! !! ! ! !!!! ! ! !! ! !! ! ! ! ! ! !! ! !!!! ! !! !!! !! ! ! !!!!!! ! ! ! ! !! !!! ! !!!! !! !! ! ! !! ! !!! ! ! !! ! ! ! ! ! !!!! ! ! ! ! !!! !!! !!!! !!! ! ! !! !! ! !!! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! ! ! ! !! !! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! !! !! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!! !! ! !! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!!!! ! ! !! ! ! ! ! !!! ! ! ! !! ! !! !! ! ! ! !! !!! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!!!!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! !!!!! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! ! !! ! ! ! ! !!! !!!! ! !! ! ! !! !!! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! ! ! !! !!! !!! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !!! !! ! ! ! !! !!!! !! !! !! !!!! ! ! ! !!! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! !! ! ! !! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! ! ! ! !!! ! !!! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!!! ! ! ! ! ! !!! !! ! ! !! ! ! !! !! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !!! ! !! !! ! !! ! ! !! ! !!! ! !! ! !! ! !! !! ! ! ! ! ! ! ! ! !!! !!! ! ! !! ! ! !!!!! !! !! !! !! ! ! ! !! ! ! ! !! !!! !! ! !! ! !! !! ! !! ! ! ! ! ! ! ! !! ! !!! !! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !!! !!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !!! !! !! ! ! ! ! !!! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !! !! !! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! !!!!!! ! ! ! ! !! ! ! ! ! !!! ! ! !!! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !! ! !! !!! ! ! ! ! !! ! !! !! !!! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!!!! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! !! ! !! !! ! ! ! ! ! !!! !!! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! !! !! !! !! ! ! ! ! !! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! !! ! ! ! ! ! ! !!! !!! ! ! ! ! !! ! ! ! !! ! !!! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !! ! !! ! !!! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! !!!!! !! ! ! ! ! !! !! !! ! ! ! ! ! !!! ! ! ! !!!! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! ! ! !! ! !! !! !! !! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !!!!! ! ! ! ! ! !! !!! ! !! !! ! !!!!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! !!!! ! ! !! ! ! ! !! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! !! !! ! ! !!!! ! ! ! ! ! !! !! !! ! ! ! ! ! !! !! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! !!! ! ! ! ! ! !! !! ! ! !! ! ! !! ! ! !! ! !!!! ! ! !! ! !! ! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !!!! !! ! !! ! ! ! ! ! ! ! !!!! ! !! !!!!! ! ! ! ! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! ! ! !!! !! !!! !!! ! !! ! ! ! ! ! ! !! !!!! ! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! !! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! !!!! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !!! !! ! ! ! ! !! !! !! ! !! ! !! ! ! ! !!! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !!! !! ! !! ! !!! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! ! ! ! !! ! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! !! ! ! ! !! ! !! ! ! !!! !!!! ! ! ! !!! ! ! ! !! ! !! ! ! ! !! ! !!!! !! ! !! !! ! ! !! !!! ! ! ! ! ! !! ! !! !! ! !! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !!! ! !!!! ! ! ! !! ! ! ! !! !!! ! ! ! !! !! !!! ! !!! ! ! ! ! ! !!! !!! ! ! ! ! ! !! ! !! ! !! ! !! ! ! !! ! !! ! !! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! !! ! ! ! ! !!! !!! !! ! ! ! ! !! ! !! !! ! !!!!! ! ! !! !! ! !! ! !! ! ! ! !! !! !!! ! ! ! ! ! ! ! !!! ! ! ! !! ! !!!! ! ! ! !! ! !!! ! !! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !!! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! !! ! ! ! ! !!! !! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! !!! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !! ! !! ! !! !! ! !! !! ! !! ! ! !! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! !! !!! ! ! ! ! ! ! !!! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! ! ! !! ! !! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! !!! ! ! !!!!! ! ! ! !! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! !!! !!! ! !!!!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!!!! ! ! ! !! ! ! !!!! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! ! ! ! !! !!! !! !! ! !! !! ! !! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !!! ! !!!! ! ! !!! ! !! ! ! ! ! !!! !! !! ! ! ! !! ! ! ! ! ! !!!!! ! !!! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! !!! !! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! !! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!! !! !! ! ! ! ! !!!!!! ! ! ! ! !!!!!!! ! ! !!! ! ! ! !!! ! ! ! !!! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! !! ! !! ! ! ! !!! !!!!! !! ! ! !! ! !! !! ! ! !! ! ! !!! !! !! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!!!!!! !!!!!!! ! ! ! ! ! ! !! ! ! !!!!!!!!! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! !!!!! ! !! ! ! ! ! ! !! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!!!!!! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! !! ! ! !! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !! ! ! !! !! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! !!! ! ! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !!! ! !!! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! !!! !! ! ! !! ! !! ! !! !! ! ! !!! !!! ! ! ! !!! !! ! !! ! ! !! ! ! ! ! !! !! !! ! ! ! ! ! !! !! !! ! !! ! !! !! ! !!! ! !! !! !! ! ! ! ! ! !! !! ! ! !!!!!! ! ! !! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! !! !! !!!!! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! !!! !!! ! !!! ! ! ! !! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! !! !! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!!! ! !! ! !!!! !! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! !! ! ! ! ! !! ! ! !!!! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!!! !!!! ! ! !!!! ! ! ! !!! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !!!! ! ! ! ! ! !!! ! ! ! !! !! !!!! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !! ! ! ! !!!! ! ! ! ! ! ! ! ! !!!!!!! !! !! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! !! !! ! ! ! !! ! !! ! !! !! !! !! !! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! !!!! !! ! ! ! !!!!! ! !! ! !!!! !! !! !! ! !! ! !!!! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !!! ! !!!! ! !!! !! ! ! ! !!! !! !! ! ! ! !! ! !! ! ! !! ! ! ! ! !!!! !!!!! !! !!! ! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! ! !! !!! ! !! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! ! ! !! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! !! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! !!! !!!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !!! ! !! !! ! ! !!!! !!! ! !! !!! ! !!! !!!!! ! ! ! !! ! !! ! ! ! !!! ! !! ! ! ! ! !!!! !! ! !! ! !! ! !! ! !!! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! !!! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! !! !! !! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !!!! !! !! !!!! !!!!!!!!! !!!!! !! !!!! !! !! !!!! !! ! !!!!!! !! ! ! !!!! !!!! !!!!! !!! ! ! !! ! !!!!!!!!!!!!!!!!! !!!! !! !!!!! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! !!! !! !! ! ! ! ! ! !!!!!!!! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !!! !! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! !! !! !! ! !! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! !! !! !!!!!! !! !! ! !! !!!! !!!!!!!!!! ! ! !! !!! !! ! !!! !! !! ! ! ! ! ! ! ! !! ! !! !!!!!! !! !! !! ! ! !! ! ! ! !! ! !!! ! !! ! !! ! !! !!! !!! !! !! ! !! ! ! ! ! ! ! ! !! ! !! !! !! ! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! !!! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! !!! !! ! ! ! !!!! ! ! ! ! !! ! ! ! ! !! !! ! ! !!! ! !!!!! ! ! !! !! ! ! ! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! !!!!! ! !! ! ! ! !! ! ! !! ! ! ! !! ! ! !!!!!!! ! ! ! ! ! ! ! ! !!!! !! ! !! ! !!!!! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! !! !!! ! !! !! !!!!!! ! !! ! ! !!! !!!!! ! ! ! ! ! !!! !!! ! ! ! !! !! ! !!!! ! !! !! ! !! ! !! ! ! !!!!!!! !!! !! !! !! ! ! !! !!! !! !! !!! ! !! ! !! ! !! ! ! ! !! !!! !! ! !!!! ! !! !!! ! !! ! ! !! ! ! !!! ! ! ! ! ! ! !!!! !! ! !! ! ! !!! !!!! !!! !!! !!!! !! !! !!! !! !!!! ! ! !! ! ! !! ! !!!! ! ! !!! ! !! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !!!!!!!! ! ! ! ! !! ! ! !! !!!!! !! ! !!! !!! ! ! ! ! ! !!!! !! !! ! ! ! !! ! !!! !!! ! !! ! !! ! !!! !!! !!! ! ! ! ! ! ! ! !! !! ! ! !!! ! ! !!! !! !! ! !! !!! !! ! ! ! ! !! !!!!!! !! ! ! !!! ! ! !! ! ! ! !!! ! ! ! ! ! !!! !!!! ! !!!!! ! ! !! ! !! !! ! !! ! ! ! !!! !! !!!! ! ! !!!! ! ! ! ! ! !! !!!!!! !!!! ! !!! !!!!! !! ! !!! !!! ! ! !! !!! ! ! !! !! !! ! ! !!!! !!! ! !!! !!!!!!!!!!!!!!!!! !!! ! !!!!!! !! ! !! !!! ! !!!! ! ! !! !!! ! ! ! !!! !!! ! ! !!!!!!!!! !! !!!!!!!! ! !!!!! !!! ! !!!! ! ! ! ! ! !! ! !!! ! !!! !!! !!! ! ! ! ! !! !! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! !! !! ! ! ! !!! !! ! ! !! ! ! ! !!! ! !! !! ! !! !!! ! !! ! !! !! ! ! ! ! !!! !! !! !! !!! ! ! ! !! ! !!!! !!!!!!! !!! ! ! ! !! ! ! ! ! ! ! !!!! !!! !!! ! !! ! ! ! ! ! ! ! ! ! !! !!!!! !! !! !! !! !! ! ! ! ! ! ! !! !! !! !! ! ! !!!!! ! ! !!!! ! !!! ! !!! !! !! ! ! !!!!! ! ! ! ! !! !! !!!! ! ! !!!! ! ! !!! !!! ! !!! ! !!! ! !! ! ! !! !! ! ! ! ! ! ! !!!!! !! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! !!! !!! ! ! ! !! ! ! ! !! !! !!! !!!!! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!!!!!! ! ! ! ! ! !! !! !!!! !! !!! !! !! ! ! ! ! !! !! !!! !! !!!!! ! ! ! ! ! ! !! !!! ! ! ! !! !!!! ! ! !!!! !!! !!! ! ! ! ! ! !! !!! ! ! ! ! ! ! !! !!! !!! ! !!! ! ! ! !! !! !! ! !!!! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !!! !!!! !! ! ! !! ! ! ! !! ! ! ! ! !! !!!! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! ! !!!! ! ! ! ! !! !! ! ! !! ! ! ! ! !!! ! !!! ! ! ! ! !! !! ! ! ! ! !!!! ! ! ! ! ! !! ! !! ! !! ! !! !!! ! ! !! !!!! ! ! ! !! ! ! ! ! !!! ! ! ! !!!! !!!!!! ! ! ! !! ! !! ! !!! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!!!! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! !!! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! !! !! !! ! !!! ! ! ! ! !! ! !! ! !! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!! !! ! ! ! !! !!! ! ! !! ! ! ! ! ! !! ! !!! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! !!! ! !!!!!! !!! ! ! !!! ! ! !! !! !!! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!!!! ! ! !! ! ! ! !! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !!! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! !!! ! !! ! ! ! ! ! !!!! ! !!! ! !! ! !! ! ! !! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!! ! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! !! !!! !!!!! ! !! !!!!! !!! ! !!!! ! !! ! !! ! !!! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !! !! !! !! !! !! !!! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! !!!! !!! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! ! !! !! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! !!! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! !!!! !! ! !!! !! ! ! !! !! !! ! ! !! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !!! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! !!!!! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! ! !!!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! !!!!!!!!! !!!! !! ! !!!!! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! !! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! !! !! !!! !! !! !! ! ! ! ! !!! ! ! ! ! ! !!!! !! ! ! !!! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! !! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !!! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! !! !! ! !! ! ! ! ! !!!! !! !! !!! ! ! ! ! ! !!! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! !! !! ! !! !! ! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! !! ! !! !!! ! !! !! ! !! ! ! ! ! !! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! !!!!!! ! ! !!! ! !!! !! !! ! ! !!!! !! !! !!! !!! !! ! ! ! !! !! !!! ! ! ! ! ! !!! ! ! ! ! ! !! !! !!! !!! ! ! !!! ! ! ! ! ! !!!! ! ! !!! ! !!!! ! ! ! ! !! ! !! !! ! !! ! ! !! ! ! !!! ! ! ! ! ! !! !! !! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!!!! !! !! ! ! ! ! ! !! !!! !!! ! !!! !! !!!! !! !! ! ! ! ! ! !! ! !! ! !! !! !!!!! !!!!! !! !!!!!!!!!!! !! !!!!!!!!! ! ! !! !! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! !!! ! !! !! !! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! !! ! !!!! ! !! ! !! ! !! ! ! !! !! ! ! ! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! !!! !! ! ! ! ! ! ! ! !! !!!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! !!!!! ! !!! ! !! !! ! !!! ! !!! ! !!! !! !! ! !!! !!! ! ! ! !!!! ! ! ! ! ! ! ! !!!!!!!!!!!!!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !!! ! !! !!!!!!!!!! ! ! ! ! !!!!!!! ! ! !!!!!!!!! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!!!! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!!! !!!! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! !!! ! !! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! !! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! !! ! !!!! ! !!!! ! !!! ! ! ! !! ! ! !!!! ! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!!!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! !! !! ! ! !!! ! ! ! ! ! !! !! ! ! ! !!!! ! ! !!! ! ! ! !! ! ! !! ! !! !! ! ! ! !! !!! !! ! !!! !! !!! ! ! ! ! ! ! ! !!! !!!! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !!! ! ! ! !! !!!!! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! !! ! !! ! ! !! !!!! ! !!! !! !!!!! !!! !! !!! !!!! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! !! !! ! !!!! ! ! !! ! !!!! ! !!! !! !! ! ! !! !! ! ! !! ! ! !! !!! !! !! ! ! ! ! !! !! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !!! ! ! ! !!! ! !! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! !!! !! !! !! !! ! !!!! ! ! !!! !!!!! ! ! !!!! ! !!!!!!!!!!! !!! ! !! ! ! ! ! !! ! ! !! ! !! ! !!! ! ! ! !!!!! ! ! !!!! !!!!! ! !! ! ! !!!!!! !! ! !!! ! ! ! !! ! ! !! ! !!!!! ! !! !!! ! ! ! !! !! ! ! ! ! ! !! !!!! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! ! ! ! ! ! !! !!!! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! !!!!!!! ! ! ! ! !!! ! !! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! !! !! ! ! ! !!! !!! ! ! ! ! ! !!! !! !!!!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !!! ! ! ! !!!!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! !!! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! !!! ! !! ! ! !! !!!! ! ! ! !!! ! !!! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!!!!! ! ! !!!!! !!! !! ! !! ! !! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! ! ! ! !! ! !! ! ! !!! ! ! ! ! !! !! ! !! ! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !!!! !! ! !! !! !! ! !!! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !!! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !!!!!! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! !! !! ! ! ! !!! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !! !!! ! !!! !! ! ! ! ! !! ! ! ! !! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !!!! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! !! !!!! !!! ! ! ! !! !! !! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! !! !! ! ! ! ! !! ! !! ! ! !!! !!! !! ! !!! !! !!!! ! ! !!!! !!! !!!! ! !! ! ! ! ! !! ! ! ! !!!!! ! !! !! !! !! !! ! ! ! !!!!!!!! !! !! !!!!!!!!!! !!! !!!! !!!!! ! !!! !!!!! ! ! ! !!! ! ! !! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !!! ! ! ! !!!! ! ! !!!!!!! ! ! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! !!!! ! !! !!! !! ! !! ! ! ! ! ! ! !!!! ! !! ! !!! ! ! !!!!!! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! !!! !! !! ! ! !!! !!!! !! !! !! !!! ! ! !!! !! !! ! ! !! !! !!! !!!! ! !! ! ! ! !! ! !! !! !! !! !!!!!! ! ! !!! ! !! ! !!!!! ! ! !!! ! ! !!!! ! ! ! !!!!!!!!!!!! !! !! ! !!! ! !! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !!!! ! !! ! !! ! !! !! ! ! ! ! ! ! !!!!!! ! !!! !!! !!! !!! !!! ! ! !! ! ! ! !! ! ! !!!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! ! !! !! !!!!!!! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! !!!! ! ! !! ! ! ! ! !!! !!!!! ! !! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! !!!! ! !! !!! ! ! !! !! !! ! ! ! ! ! ! !! !!!!! ! !!!! ! ! ! !! !!!!!!! ! ! !! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! !! ! !! ! !! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! ! ! ! !!!!! !! ! ! ! !!! ! ! ! !! !! ! !! !!!!!! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! !!!! !! !! ! ! !! ! ! !! !!! ! ! ! !!! !!!! !!! !! ! !! ! ! ! !! ! ! !!! ! ! ! ! !!! ! ! ! ! !!! !! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !! ! ! ! !!!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! !!!! !! !! !! !! !!! ! !! ! ! ! ! !! !! ! ! ! !! !! !!! ! ! !! !! !!!!! ! !! !!!!!!! ! ! ! !!! !! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! !!!!! !!! !! ! ! !!! !!!!!! !!! ! ! !!!! !!! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !!!! ! ! !!! !! !! ! ! !! ! !!!!!! !! ! ! !!! ! ! !! ! !! !!! !!!! ! !!!!! ! !! !! !!!! !!! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! !! !! ! ! !! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! ! ! !!!!!!!!!!!!! ! !!! ! !! ! !! ! !! !! ! !!!!! !! ! !! ! ! ! !! ! ! ! !! !! !!!! !! !! ! ! ! ! ! ! !!!!!!! !! ! ! !! !!! !!! !!! ! !! !! ! !!! !! ! !! ! ! ! ! !!! !!! ! ! !!!!! !!!! ! !!! ! !!! !! !! !!!!!! ! ! !! ! !!! !!!! ! ! !! ! !!!!! !! !!! !!! !! ! !! ! ! ! !!!!! !! ! ! !! ! ! !! ! ! !! !!! !! !! ! !! ! ! !! !!! ! ! ! !!!! ! ! !! ! !! ! ! !!! !! !! ! !! !! ! ! ! !! !!!!!!!!!! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! !!! !! !! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! !! !! !!!! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! ! !! ! !!! ! !!!!! !!!! !! !!!!!! ! !!!! !! !! ! ! ! ! !! !! ! !!! ! ! ! ! !! !!! ! ! !! !! ! ! ! ! ! ! ! ! !!!! ! !!! ! !!! ! !! ! ! !! ! ! ! ! !! !!!! ! ! ! !! !! ! !! !!! !! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! ! ! !!!!!!!! ! ! ! !! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! !!!!! !!!! ! ! ! !! !! ! ! ! ! ! ! ! ! !! !!!!! !! ! !! !! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !!!!! !! ! ! ! ! ! ! ! !!!! !!!! !! !! ! !! !!! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! !!!! ! ! ! !!!! ! !! !! ! ! ! ! !!!! ! ! ! !! !!! ! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! !! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !!!!! ! !! ! ! !!! ! ! ! !!! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !!! !!!! !! !! ! !! ! ! !! !!!!!! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! !! ! ! ! !! !! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! !! !! ! !! ! !! ! ! ! !!!!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!!! ! ! !! !!! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! !! ! ! !! !!! !!!! ! ! !! !! !! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!! ! !!!!!!!!!! !! ! !!!! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!!! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! !! !! ! ! ! ! !! !! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! !! ! ! !! !! ! ! !! ! !! !!! ! ! ! !!! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !!! !! !!!!!! ! ! ! ! !!!! !! ! ! !! !! ! ! !!!!! !! !!!!!!!!!! ! ! ! ! !!! ! ! !!! !! ! !!!!!! ! ! ! !!! !! ! !!! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !! !! ! !! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !! ! !! ! !! ! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !!!!! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !!!!!!!!! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !!! !!!! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !!!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! !!! ! !!!! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !!! ! ! ! !! ! !!! ! ! !! !! !! !! ! ! !!!!!! ! ! ! ! !!! ! !! ! !! ! ! ! ! !!! ! ! ! ! !! !! ! ! ! !!!!! !! ! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! ! !! !!!!!!! ! ! ! ! !! ! !! !! !! ! ! !!!!!! ! ! ! !! ! ! ! ! !! !! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! !! !! !!!!! ! !! !! !! ! !! ! !!!! ! !! ! !! ! ! ! !! !! !! ! ! !!!!!!!! !!! ! ! !!! !! ! !!!! ! !!!!! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !!!! !!! !! !! ! ! !! ! ! ! !!! !!!! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! !!!! !!! ! !!! ! ! ! ! ! !! !!!!!! ! !! !!!!! !! !!!! ! !!!!!!! !! !! ! !!! ! ! !!! ! !! !!! !!!! !! !!! ! ! !!! ! ! ! !! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !! !!!! ! ! ! !! ! ! ! !! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!!! ! ! !! ! ! ! !!!! ! ! !! ! ! ! !! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! !!! !!! !! ! ! ! ! ! !! !!! ! ! ! ! !!!!! !! !! !!! ! ! ! ! ! ! ! !!!!! !! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! !!!! ! !!!!!!!!!!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! !!!!!!!!! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!!! ! !!!!!!! ! ! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! !!!!!! ! ! ! ! !!! ! ! !! !! ! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! !! ! !! !! ! !!! ! ! ! ! ! ! ! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! !!! ! !! ! ! ! !! ! !!!!!!!! !! ! ! ! ! ! ! !! ! !!! !!! ! ! ! !! ! ! !! ! ! ! ! ! ! !! ! !!! ! !!!!!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! !! ! !! ! !! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! ! ! !!! ! ! ! ! !! ! !! ! !! !!!!!! ! ! !! ! !!! ! !! ! ! !! ! ! !! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! !!!!! !! !! ! !!! !! ! ! ! ! !! ! !! !!!!!! !!! !!!! !!!! ! ! !! ! ! !! ! ! !! ! !!! ! !! !!! !! !!!! ! !! !! ! !! ! ! ! ! !! !! ! !! !! ! !!!!!!!!!! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!!! ! ! ! ! ! !! !! !!! !!! ! !! ! !!!!!!!!!! !!!!! !!!! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !!! !! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !!!!! ! ! !! ! !! ! ! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !!!!! !!! ! ! !! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! !! !!!!! !!!!! !!!! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! !!!!!! ! ! ! ! !!! !!!! !!!!!! ! ! ! ! !! ! ! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! !! !! ! !!!! ! ! !! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!! ! !!!!! !!!! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !! ! ! ! !! !! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! !!! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !!!! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! !! ! ! ! !!!! !! ! ! ! ! !!! !! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! !! !! ! !! ! !! ! ! ! ! !! ! !! ! !!! ! ! !! !! !! !! !! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!!!!! ! ! !!! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! !! ! ! ! ! ! ! !! !!!! !! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! !! ! !! !! ! !!!! !!!!! !!! ! !!! !! ! !!! ! !!! ! ! !!!! ! ! ! !!! ! ! !! ! ! ! ! !! ! !!!!! ! ! !!!! ! ! ! ! !!! !! ! !! ! ! !!!!!!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !! !!!! ! !!! !! !!! ! ! ! !! ! ! !! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! !! !! ! ! !!!!!! ! ! ! ! ! !!!!! !! !!! !!!!!!! ! !!! ! ! !! ! ! !! !!! ! ! ! ! !!! ! ! !! !!! ! ! ! !!! !! ! ! !!!! ! ! ! ! ! ! ! !! ! !! ! !!!!!!! ! !! ! ! ! ! ! ! ! ! ! !!! !! !! ! !!!! ! !! !! !!!!!!!! ! ! ! ! !!! !! ! ! !! ! ! !!!! ! !! !! ! ! ! ! ! !!! ! ! ! ! ! ! !! !!!!!! !! ! !!!!! !! ! ! ! ! !!!! ! ! !! !!! ! !! !!! ! ! ! ! !! !!!! ! ! !!!! !!! !! !!!! ! ! ! ! !! ! !! ! ! ! ! !! ! !!! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! !! !!! ! !!!!!! !! ! ! ! ! ! !!!! !!! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! !! ! ! !!! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !! ! ! !! !! ! !! !! !! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! ! ! !!! !! ! !! ! ! ! !!! ! ! !! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! !!! ! !!! ! ! ! !! !!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! !!!! ! !!!!!!!! ! !!!! ! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! !! !! !! !! !!!!!!! !!!!! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! !! !! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !!! ! !!!!!! ! ! ! !!!!!!! !!!!!!!! ! ! !! ! ! ! ! !!! ! ! !! ! !! !!! !! ! ! ! ! ! !! !! ! ! ! ! ! !! !!! ! !!! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! !! ! !! !!! ! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! !!! !!!! ! ! ! ! !! !! ! !!! !! !! ! ! !! ! ! ! !!!!!!! !!! ! !! !! ! ! !! !! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! !! !! ! ! ! ! !! !! ! ! !! !! !! ! !! !! !! ! ! !! ! ! ! ! ! !! ! ! !!! !! ! ! ! !!! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!! ! ! !!! ! ! ! ! !! ! ! !!!!! !!!!!!!! !! ! ! !! ! ! ! !!!!!!!!!!!!!! ! ! !!! ! ! ! !!! ! ! ! !!! !!! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !!!!!!!! !!! ! !! !!! ! ! ! ! ! ! ! ! !!!!!! ! !!! !!! ! ! ! ! !!!!! !!!!!!!! ! !!!! ! !! ! ! !! ! !! !! !!!! !!! ! ! !!! ! !!!!!!!! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !!!! ! ! ! ! ! !! !!!! ! ! ! !! !! ! ! !!!! !! ! !! ! !!!!!!! ! !!!!!! ! ! ! !! ! !! ! ! !!! !! !!!!!!!! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! !!!! ! ! !! ! !!!! ! ! ! ! !!! ! ! !!!!!! ! ! ! ! ! !!!! ! !! !! ! ! !! ! !!!!!!!!!!!! !!! ! !! ! ! ! ! !!!!! !! ! ! ! ! ! !! !! !!!!!!! ! !!!!!! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! !! ! !!! !! !! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! !! ! !!!!!!! ! !! ! !! ! ! ! !! !!! ! !! ! !! ! ! ! ! ! !!!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! !!! ! ! !! !!! ! ! !! ! ! !!! !!!!! ! !!!!! !! !! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! !!!!!! ! ! !!!!! ! !! ! ! ! !!! ! !!! !!! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !!!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !!! ! ! !! !!! ! !! ! !!!! !!! ! ! !! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! !! ! !! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !!! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! ! ! ! ! ! !! ! !! ! ! ! ! !!!! !! !! ! ! ! !! !! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! !!!! !!! !!! ! ! ! !!! ! ! !!! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! !!! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! !!! !! ! ! !! !!! ! !!! ! !!!! !! !! !! !! ! !! ! ! !!! ! ! ! !!!! !!!! ! ! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! !! ! !!! !! !!!! ! ! ! ! !! ! ! !!!! !!! ! !!!! ! ! ! ! !! !! ! ! ! !! ! ! !! ! !!!!!! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! !! ! !! !!! ! ! ! !! !!!!!!! !! ! !!! !! !!!!! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!!! !! !! ! !! ! ! ! ! ! !!!!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !!!! ! ! ! !!!!!! !! !!!! ! !! ! !!! !!! !!! !!! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !!! ! !! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! !!! ! ! ! ! ! !!!!! !!!! !!!!! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !! ! ! !! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! !!!! !! ! ! !!!! !!! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !!!!!! !!!! ! ! ! ! ! ! ! !!! ! !!! ! ! ! !!! ! !!! ! ! ! !!! ! ! !!! ! !! ! ! ! ! !!!! ! !! ! ! !! ! ! !!! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! !! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! !! ! ! !! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! ! !!! ! ! !! ! ! ! ! !!! ! !! !!!! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !!!! ! !!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! !! !!! ! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! !! !!! !!!!! !!!! ! ! ! ! ! ! ! ! !! !!!! ! !! ! ! ! !! ! !! ! !! ! ! ! ! !! !! ! !! ! ! !!!! ! ! ! ! ! ! ! !!! !! !! !! !!!!! ! ! ! ! ! ! !!! !!!!!! !! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! !! !!! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! !!!!!!! !! ! !! ! ! !!! ! ! ! ! !!!! ! !!!! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!!!!! ! !! ! !!! ! ! ! ! ! ! ! ! !!! ! ! ! !! !! ! !!! ! ! ! ! ! !! ! ! ! !!! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !!!!!!!!! ! !!!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! !!! !! ! !! ! !! ! ! ! ! !! ! ! ! !!! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!!!! !!!! !!!!!!!!!! !! ! !! !! ! ! ! ! ! !! !!! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! ! ! ! !!! ! ! !!! ! ! ! !! !! !! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! !! !! ! !!!!! ! ! !! ! ! !! ! ! ! ! ! !!!! ! ! !! ! !! ! ! ! !!!!! ! !!! ! ! !! !! ! ! !!! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! !! !! ! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! !!!! ! ! !!! !!! ! ! !!! !! ! !! ! !! !! ! ! !! !! ! !! !! ! ! ! ! ! ! !! !! ! !! !!! !! ! !!!!! ! !!!!!! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!!! !! !! ! ! !! !!!!! !!!!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! !!! !! ! !!! !!! ! ! ! ! !!!! !!! !!!!! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! !! !!! ! ! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! !! !!!! ! ! !!! !! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! !!!!!! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! !! ! !! ! ! ! !!!! ! ! ! ! ! ! ! !!!! !! !!!!! ! !! ! !! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! ! !! ! !!!!! ! !!! !!!! ! ! !!! !! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! !!!! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!!! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! !! ! ! !! !! !! ! !!! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! !!! !!! !!!! ! !! ! ! !! !!! ! !! !! ! !!!!! ! ! ! ! ! !!!! ! !!!! ! ! !! !! ! ! ! ! ! !!! !! ! ! ! ! ! !!! ! !!! ! ! !!! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! !! ! ! ! !! ! !!! ! ! !! !!!! ! !! ! ! ! ! ! ! ! !! ! !! !! !! ! !! !!! !! ! ! !!! ! ! ! ! ! ! !! ! !! ! ! ! !!! ! ! ! !! ! !!! ! ! ! !! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! ! ! ! ! ! ! !!! !!!!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! !!!! !!!! ! !! ! !! ! !!! ! ! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!!! !!! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! !! ! !! ! !! ! !!! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!!! !! ! !! ! ! ! ! ! !!! ! !! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! !! !!!!!! !! ! !!! !!!!!!! !! !! ! ! ! ! !! ! !!! !!!!! !!! !!!! !!! !!!!! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! !!!!!!! ! ! ! ! ! ! !!! !!!!!!!! ! ! ! !! ! ! !! !!!! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! !!!! !!! !! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! ! !! !! ! !!! ! ! !! !! ! !! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! !!! ! !!!!! !! !!! ! ! ! ! ! !!! ! !! !!! ! ! !! ! ! ! !! ! ! ! !! !! ! ! ! !! !! ! ! ! !!!!!!! ! ! ! ! ! !! ! !! ! !! ! ! ! !! ! !!!! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! !! M D m E m Q O O G M S m O G W G F gure 2 Loca on o o g een and gas we s ed n M ch gan
    • 6 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 recovered from the northern end of the trend in Calhoun County). It was followed in the late 1960s by the Niagaran Reef Trend which resulted, by the 1970s, in a tripling of Michigan oil production and multiplying natural gas production by 6 times. By 1979, Michigan’s total oil production had reached 35 million barrels annually. It had declined to about 6.5 million barrels by 2010 in the absence of any additional major finds, but increased slightly to almost 7 million barrels by 2011, the last year for which data are available, mostly because of aggressive workovers and a few new but small discoveries. Since this total is far less than Michigan’s annual oil consumption, the state imports about 97% of its total petroleum needs, mostly from western Canada via pipelines that pass through the Chicago area. In 2009, Michigan consumed 163.6 million barrels of petroleum product. Relative to petroleum, Michigan’s natural gas production is more substantial and accounts for about 18% of the state’s demand for natural gas. The following Figure 3 shows monthly production data. In the early 1980s, deep-strata natural gas production (for example from the northern Niagaran Trend) began to make a significant contribution to Michigan’s economy while there was also a major expansion of drilling activity in the long-known Antrim shale for- mation in the northern Lower Peninsula. All of these more recent fields, including the Antrim, are still significant producers, primar- ily of gas. However, the production from all of them is in decline. Production in the Antrim shale stood at 131 billion cubic feet per year (bcf/y) in 2008, but this figure has been steadily declining, only reaching 85 bcf/y last year (2011). It is projected to continue falling and to stabilize at approximately 62 bcf/y by 2020. Total in-state natural gas production peaked at 280 bcf/year in 1997 but had declined to 141 bcf/year by 2010 due to decreased well produc- tivity. Again, note that U.S. annual consumption is about 23 trillion cu. ft./year while Michigan’s annual consumption was about 765 billion cu. ft. in 20104 . The following figure (Figure 4) shows monthly production data of Michigan shale gas. Very recently, the discovery of gas and gas liquids in the Collingwood and Utica deep shales (which also happen to outcrop in Ontario along the southeastern edge of Lake Huron - primarily under the Blue Mountain Shale - as does the Antrim Shale in the southwestern corner where it becomes the Kettle Point shale in Ontario) have caused a flurry of lease sales in the state. Activity has dropped, however, perhaps temporarily, as a result of the very low price of natural gas18 and what has turned out to be relatively intractable “tight” shale. The potentially productive part of the Collingwood shale in Michigan is at least 10,000 ft. deep (about 2 miles) and will typically require the drilling of several wells and laterals from the same pad to achieve acceptable production rates. These are costly requirements. At current prices, very few opera- tors working in deep shale can hope to make money, and a return of major activity will have to wait until gas prices reach and stabilize at a minimum of $6-$8/MCF. The price of gas appears to have sta- bilized recently after peaking at about $4.00 and falling somewhat and is now (May, 2013) again trading at about $4.00/MCF18 . As a sidebar, the Kettle Point (Antrim extension) shale that runs roughly from east of Sarnia almost down to Windsor/Chatham along the east side of the St. Clair and Detroit Rivers has been a local gas producer in the past19 . It and other shale formations in the vicinity (including the Hamilton Group, which is mostly lime- stone, and also the northern extremity of the Marcellus shale which extends under Lake Erie) provided the gas that started what is now a very large petrochemical and oil refining complex in Sarnia, ON. Figure 3: Michigan Crude Oil Production16 0 500 1,000 1,500 2,000 2,500 3,000 3,500 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 1,000 Barrels Michigan Crude Oil Production January 1981 to December 2010 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 Jan‐2007 Jul‐2007 Jan‐2008 Jul‐2008 Jan‐2009 Jul‐2009 Jan‐2010 Jul‐2010 Jan‐2011 Jul‐2011 Million Cubic Feet Michigan Natural Gas Gross Withdrawals from Shale Gas Figure 4: Michigan Natural Gas Production from Shale Gas17
    • 7 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 Numerous water wells were also drilled in the vicinity and almost all have produced gas along with the water, but certainly not as a con- sequence of gas drilling in the area (which many of them predate). It is also worth noting that shallow-water drilling has also long taken place in the Canadian side of Lake Erie near Erieau, ON, tapping the extreme northern edge of the Marcellus formation. There are still several active gas plants in the area that refine gas from Lake Erie and sell it to Union Gas of Chatham, ON. Otherwise, there has been little or no attempt to commercialize gas from the Kettle Point, Hamilton Group, Collingwood, Blue Mountain/Utica and other gas-bearing structures in Ontario. Recently, the Government of Ontario has been criticized for its failure to capitalize on these probable resources, but that seems to be changing20 . Curiously, there are few reports of gas in areas west of the Detroit and St. Clair Rivers. These areas are now highly populated. Very modest quantities of both oil and gas have been produced in Oakland County, apparently from the southern Niagaran Trend, but by far most wells drilled were dry. Wayne County has recently supported a ban on fracking (of little note, since the county has no known commercial oil or gas fields and lies well south of the Collingwood/Utica shale). As noted earlier, St. Clair County was the site for Michigan’s first oil field in 1886 (in the Dundee Formation) and still has producing wells. There were also a few at one time in Macomb County, but most wells drilled there were dry, as was the case in Washtenaw County. 2.2 Gas Storage in Michigan During the summer months when gas demand is usually low, gas is pipelined to Michigan and stored underground in specifically appropriate locations, often former brine wells or depleted gas wells. At 649 billion cu. ft., Michigan has more gas storage capacity than any other state. During the winter heating season, this gas is withdrawn and used both in-state and by neighboring states. 2.3 Realistic limits on resource recovery In the case of natural gas, most of the gas in a conventional reser- voir can usually be recovered given sufficient natural permeability, porosity (or fracturing, natural or man-made) of the supporting rock formation. A much smaller proportion may be recovered from “tight” formations with limited permeability such as tight sandstones or coal bed formations. Most formations, especially shales with high organic content, also retain a certain amount of methane that is adsorbed on to mineral surfaces. This amount is unpredictable and often indeterminate and in some cases may be replaced by carbon dioxide or other compounds as the methane is withdrawn. In very tight formations with low permeability and low interconnected porosity, very little gas (or oil) flow may be estab- lished regardless of how the formation is drilled or fractured. In the case of oil resources, the story is quite different. In the early days of oil production in Michigan, in the Spindletop field in East Texas, and in many other states, wells were in many cases drilled too close together and oil was pulled from what was a common reservoir too quickly, often leading to water flooding, which can effectively stop oil flow in small channels in the formation through surface tension effects. For example in East Texas, only 10% of the oil now known to have been in place in several major reservoirs was recovered. The remainder may be lost forever or at least until a new technology is developed to recover it. Even today with opti- mum drilling practices, it is unusual to obtain more than 35% of the oil in place by primary production methods. Another 15-25% can sometimes be recovered by enhanced oil recovery (EOR) technol- ogies, some of which are discussed below, but that still leaves a lot of oil in the ground and currently inaccessible. 2.4 Recovering More of the Resource There are many ways by which more of the energy resource in the ground can be recovered. They are summarized briefly here because most have been important at times in Michigan. Hydraulic fracturing (HF) can be viewed as one of these methods. It has seen little use in Michigan in the manner that is currently practiced in, say, Pennsylvania, North Dakota or Texas, other than in a few exploratory wells in the Utica/Collingwood and the associated A-1 Carbonates. However, it has been used in the form of low-pressure nitrogen foam fracking in the Antrim Shale in the northern Lower Peninsula since the late 1940s. All methods of fracking now involve the use of a high pressure fluid, typically water, with 9-10% (and often up to 20%) of sand or a synthetic ceramic such as calcined bauxite dispersed in the fluid to help keep the cracks in the formation open after fracturing; the fluid also contains a total of about 0.5 % of chemical additives (such as surfactants and antibacterial agents, most of which are used in other commercial or domestic operations such as dish- washing) to help keep the newly-formed cracks open and clean. At one time, and especially in the 1940s, 1950s, and even as late as 2000, far less environmentally benign chemicals were employed (Napalm or gelled gasoline was one very early example). A long list of chemicals once, and in some cases still, in use can be found at the Fracfocus.com website21 . A more current list of chemicals commonly in use can be found at the recent ASTM Jacksonville meeting proceedings22 . In the past, far less environmentally benign chemicals were added, but the use of these has been discontinued by all of the major operators and their sub-contractors, partly as a result of public pressure and greater state disclosure requirements. Hydraulic fracturing was first performed experimentally in 1947, and the first commercial “frac job” using hydrocarbon fluids, mostly locally-produced crude oil, with some rather unusual additives, was
    • 8 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 performed (with only modest success) in 1949. Water was used for hydraulic fracturing only after 1953. As of 2010, it was estimated that 60% of all new oil and gas wells worldwide were being hydrau- lically fractured6 . As of 2012, it is estimated that 2.5 million hydrau- lic fracturing jobs of all types (not all of them deep or involving directional drilling) have been performed on oil and gas wells worldwide, more than half of them in the United States7 . To date in the U.S., fracking technologies are estimated to have been applied to more than 1.25 million vertical or directional oil or gas wells. Canadian companies are said to have fracked at least another 200,000 wells. In many recent cases, a combination of directional drilling and high-pressure multi-stage fracking has been used to access oil or gas trapped in larger ‘drainage volumes’ of otherwise unproductive reservoirs. Currently, about 35,000-40,000 U.S. wells are being hydraulically fractured annually with a far greater propor- tion of them directionally drilled than previously. A major issue with modern high-pressure, deep formation HF is its use of what are often seen as high volumes of water. Based on information posted on FracFocus for 16 wells, Michigan has seen a wide range in terms of water use from as low as 14,000 to over 21,000,000 gallons of water21 . These volumes are not in fact large compared to other industrial or agricultural uses (for example golf courses in the arid U.S Southwest), but flowback water in Michigan is disposed of via deep well injection. Water availability tends to be a rather emotional local or regional problem, especially in the western U.S. and reduction in its use is always desirable. Water typ- ically costs $0.10 to $0.25/gallon (up to $0.75/gallon under drought conditions) which is a significant incentive to limit use at current low gas prices. Several non-aqueous fracking methods are now in use or being developed, but they are more costly than water- based hydraulic fracking. A more serious problem is disposal or treatment of the often-sub- stantial amount of the fracking water returned as so-called flowback water and also of any subsequently produced water. In Michigan, this water is sent to disposal wells regulated and permitted under the EPA Safe Drinking Water Act through the Underground Injection Control as Class II wells23 . These wells are also regulated under Michigan’s Oil and Gas Regulations24 . An increasing effort is being made to develop water treatment methods appropriate to the often highly saline return water which may also contain small amounts of hydrocarbons, some of them toxic. If these efforts are successful, as they have been in a limited number of cases in PA and CO, the water will be re-used. 2.4.1 Directional Drilling Conventional vertical drilling can access only that part of a gas- or oil-bearing formation with which the drilled hole intersects. Since most such formations have a significant lateral dimension, usually extending 360° around the vertical well, drilling one, or preferably several, lateral (directional) wells into the producing formation can provide much greater access to oil and/or gas. If exclusively ‘dry’ gas (free of hydrocarbon liquids) is being produced, the lateral can be nominally horizontal because no liquid drainage is required. (Nevertheless, some liquids often accumulate at the bottom of the vertical part of the well; it eventually blocks the flow of gas and must be removed from time to time by a process known as ‘liquids unloading’.) In many cases, the well produces some liquids, either water or gas liquids (typically C2 to C6 alkanes or a gasoline-like condensate). In this case, the lateral segment of the well is often drilled at a slight angle to facilitate drainage and pumps may be installed to remove the liquids. In many cases, the formation being accessed is not exactly horizontal and the slope of the lateral well may be designed to follow the formation. Although the term “horizontal” is often applied to such laterals, very few of them are precisely horizontal. The success of any drilled hole, whether vertical or directional, depends on the pore size and/or the size of flow channels in the surrounding formation. Most pores or channels are in the 1 to 100 millidarcy rangei . Even with extensive directional drilling of laterals, only a fraction of the gas- or oil-bearing formation can be accessed. The volume of rock that can be fracked around any one lateral is limited, and there are large potential ‘drainage volumes’ that are not accessed. Of course, more holes and more laterals (only one per vertical hole is possible if casing integrity is to be maintained although many vertical wells can be drilled from a single pad) can improve access, but that is very costly. 2.4.2 Explosive Fracturing of the Energy-Bearing Formation Explosive fracturing was widely employed at one time during the nineteenth and first half of the twentieth century. It was used only for what were intended to be vertical holes (early cable-tool and similar drilling technologies often resulted in holes that deviated significantly from the vertical; even modern deep rotary drilling can suffer that problem as a result of contact of the rotary drill bit with large, hard rocks) and consisted of lowering or dropping a dynamite charge into the well which was then ignited with a red- hot rod or, later, with an electrical charge via a wireline. The result, while often spectacular, fractured the rock in only a limited volume of the formation around the bottom of the hole. Nevertheless, gas or oil flow was enhanced although the process often had to be repeated frequently. Explosive charges are still used to perforate a i. The darcy is a measure of permeability. A gravel bed has a permeability of about 100,000 darcys while sand beds approximate 1.0 darcy. At the other extreme, granite typically has a permeability of 0.01 micro-darcys or 10 nano-darcys.
    • 9 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 well casing in preparation for hydraulic fracturing but are not used for formation fracturing per se. 2.4.3 Hydraulic Fracturing (HF) of the Formation Both vertical and horizontal wells generally require some form of formation fracturing process to provide greater access to the gas or oil contained therein. In recent years, some 85% of wells drilled in North America have been subjected to some form of fractur- ing process, usually hydraulic, but not always involving water use. Many of them have involved directional drilling. A few rock formations, such as the Antrim shale in the northern lower peninsula of Michigan and the New Albany shale in Indiana/ Illinois, are already highly fractured and flow gas without much additional fracturing once the substantial amount of naturally-oc- curring water is removed from the formation. In some cases, pres- surized nitrogen foam (in water) may be used, primarily to clean drilling mud and rock chips out of natural fractures that intersect with the well walls, and hence casing, in the production zone. Most other shale formations, while often naturally fractured, must be additionally drilled and fractured to achieve sufficient flow of gas or liquids. ‘Tight’ rock formations such as sandstones or limestones (carbonates) may require fracturing if the porosity is very small or not well interconnected. In some extremely tight formations with porosity in the micro-darcy range (see earlier footnote), even extensive fracturing may recover only a small proportion of the gas or oil in place in the fractured zone. Gas will flow through solids that are of low porosity more easily than will oil, which requires at least 10-100 times the porosity. In both cases, the presence of water may effectively block the pores. Hydraulic fracturing can be very effective in formations that are nat- urally cracked but in which the cracks are too narrow to permit flow, especially of oil. In such cases, water at a pressure of up to 15,000 psi is forced into the formation through holes in the steel well liner (casing) that is cemented to the bore wall; the water opens the cracks. Up to 20% of sand or ceramic in the water, assisted by a small number of chemicals at high dilution (roughly 0.5% in total), acts as a ‘proppant’ to hold the cracks open and to permit the flow of gas and oil. The water that is injected into the well contains only a modest amount of chemicals, predominantly very dilute hydrochloric acid. In most cases, the injected water collects additional chemical com- pounds from the formation (the amount picked up may be small in true dry gas wells) such as highly saline water containing mostly sodium, calcium, and magnesium chlorides and some hydrocar- bons. This “flowback water” presents a disposal problem and must either be treated for re-use (this can be difficult due to the high salinity) or disposed of in deep wells. In Michigan, these disposal wells are regulated and permitted under the EPA Safe Drinking Water Act through the Underground Injection Control as Class II wells23 . These wells are also regulated under Michigan’s Oil and Gas Regulations24 . Flowback water can be handled only rarely in conventional waste water treatment systems (such handling is pro- hibited in Michigan). Increasingly, specialized waste water treat- ment systems are being developed that involve either distillation or reverse osmosis to reduce salinity to levels that permit re-use. They are in common use in Pennsylvania’s Marcellus shale and in Colorado’s Denver foothills area. 2.4.4 Formation Fracturing with Minimum or Zero Water The possibility of using compressed nitrogen or CO2 foam to frac- ture “easy” formations that contain already substantial amounts of natural fractures and cracks was mentioned above. This method is environmentally appealing since no chemicals other than a foam- ing agent and sand are used. The method has proved very effec- tive in vertical wells with no laterals such as those commonly used in Michigan’s Antrim shale formation but has been less effective elsewhere. Liquid nitrogen and CO2 must be brought to the drilling site by truck, which adds to traffic disruption and road damage. An alternative method developed by GasFrac Energy Services Inc., a Canadian company headquartered in Calgary, uses gelled LPG (liquefied petroleum gas) as the fracking agent. Once it is used, the gel ‘disintegrates’, and the LPG Components (propane, butane, etc.), are recovered in the well product. The method has now been used in over 1,000 frac jobs in Canada and the U.S. (including Ohio’s Utica shale as well as in the Eagle Ford shale formation in Texas). No problems have been reported, but given the explosive nature of the LPG components when mixed with air, all oxygen and ignition sources must be eliminated. The method uses no water but costs about 20% more than water-fracking a well of similar depth and is thus being used primarily for oil-producing wells, given pres- ent low gas prices. Several competitors for this technology have recently appeared, at least one (eCorp Stimulation Technologies LLC of Houston) using only propane in the Eagle Ford Shale. 2.4.5 Vacuum Application Several operators have sought to improve declining gas well pro- ductivity by applying a modest vacuum (typically 2-3 inches of mer- cury or about -0.1 bar) at the top of the well. In most states, vacuum extraction requires a special permit and very few have been issued. Concern has been expressed that the application of a vacuum will draw gas from neighboring properties. The technology has been so little practiced that it is not clear if it is effective. 2.4.6 Liquids Unloading Even in so-called dry gas wells, liquids may accumulate over time
    • 10 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 at the bottom of the well and inhibit gas flow. The liquid may be water or, in wells producing under significant reservoir pressure, can be gas liquids (typically C2 through C6 hydrocarbons or pos- sibly natural gasoline/condensate). In some vertical wells, a sub- mersible pump may be placed in a sump - usually an extension of the vertical well bore - to remove these liquids. The old practice of opening the well and letting the gas flow blow the liquid out of the well has supposedly been discontinued everywhere since both the gas and the gas liquids have significant value and, if released, have a negative environmental impact. However it is carried out, the practice is known as ‘liquids unloading’ and plays an essential role in maintaining the productivity of wells, but may contribute significantly to methane emissions25 . 2.4.7 Water Flooding Intentional and well-designed water flooding of declining oil (not normally gas) wells can enhance production of oil by pushing it through the formation toward the production hole. This requires the drilling of one or more water injection wells around the produc- tion well—a significant added cost. Care is necessary to prevent water breakthrough to avoid water replacing oil in the producing formation. This was a common reason for the premature failure of many early oil wells in over-produced fields (including some in central and western Michigan); in those cases, the water can fill the capillaries or pores through which oil is flowing and block its movement. That oil cannot then be produced using any currently available technique. 2.4.8 Polymer flooding In this case, water gelled with an added proprietary polymer for- mulation is used in lieu of “straight” water. The effect is similar to water flooding, but the gelled water is less likely to “get ahead” of the oil front to block it. Instead, it pushes the oil ahead of it. We are not aware of its use in Michigan, perhaps because of its high cost. 2.4.9 Steam flooding In this case, steam is injected into the formation to heat the oil and reduce its viscosity. The oil then flows more easily toward the production well. This is a preferred technique for heavy oils such as those found in, for example, California’s Santa Barbara region and in southern Alberta. However, in deep formations, the heat loss from the steam on its way to the point of use may make the method ineffective. Solvent (e.g., kerosene or gas liquids) may be added to the steam to further enhance recovery. 2.4.10 Carbon Dioxide Flooding Carbon dioxide injected at moderately high pressure (3,000 psi) into oil-bearing formations can greatly reduce the viscosity of some (but not all) crude oils and also, at least temporarily, increase the volume (and reduce the density) of the oil. As a result, the oil flows more easily in the reservoir and more—sometimes much more - may be produced than in the absence of CO2 . This method of enhanced oil recovery has been widely used with good effect, especially in the Permian Basin of West Texas (using CO2 from gas wells in New Mexico) and has also been promoted as a method of sequestering carbon dioxide. However, little work has been done on the eventual disposition of the CO2 ; for example, it is not known whether it combines chemically with the oil and thus remains sequestered or desorbs from the oil during and after production. In the latter case, it would have to be re-captured and re-seques- tered with little net benefit to the CO2 emissions problem. There is a modest amount of evidence that a combination of methane and carbon dioxide may be slightly more effective than CO2 alone. A few enhanced oil recovery (EOR) experiments using CO2 have been conducted in Michigan’s northern Niagaran (Middle Silurian) Pinnacle Reef structure. Elsewhere in the world, numerous CO2 — EOR projects exist, some developmental and some commercial. For example, in Alberta, Shell is building a pipeline to take com- bustion-generated CO2 from its northern Alberta oil sands opera- tion, first of all to underground storage and then on demand south to the Cold Lake area to be used in enhanced conventional heavy oil recovery. Power station combustion CO2 injection is also being explored for North Sea oilfield use by both Norway and the UK. For those who are interested, the Oil & Gas Journal publishes a list of EOR projects every other year. There are many such projects, not all involving CO2 injection. 2.4.11 Solvent Flooding Another, relatively costly, method of enhanced oil recovery that does not appear to have been used commercially in Michigan involves the injection into the formation of a heated solvent such as kerosene, naphtha (natural gasoline), C2 -C6 gas liquids or even LPG into the formation to thin the crude oil and thus stimulate flow. The solvent is sometimes mixed with steam to provide an addi- tional incentive to flow. Versions of this technology (for example, the N-Solv Process) are increasingly being used in the Canadian Oil Sands to enhance the productivity of steam-assisted gravity drain- age (SAGD) bitumen production but have seen relatively limited use elsewhere. 2.4.12 Fire Flooding or In-situ Combustion To the best of our knowledge this method of enhanced oil recovery has not been widely used in Michigan. It consists of using in-hole combustion, using injected air and gas or other combustible fuel to generate a flame front at the foot of the injection well; a com- bustion wave is driven through the reservoir where it heats the oil and lowers its viscosity and drives it toward a parallel production well. This method has been used for recovery of heavy or waxy crudes in areas such as southern and central Alberta and has been
    • 11 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 experimented with in California and elsewhere where heavy crudes are common. 3.0 CHALLENGES AND OPPORTUNITIES I n the following section, the technical aspects of current methods for hydraulic fracturing will be reviewed, and where appropriate, technical challenges and opportunities for improved techniques will be discussed. Furthermore, areas where further research is needed will be identified. 3.1 Hydraulic Fracturing (also Fracking or HF) This topic has been the subject of a separate report by the authors26 ; that report looked at HF from a national perspective. It detailed much of the technology in common use and that will therefore only be summarized here. Michigan has not been typical in its use of HF (as compared to states like Texas, Arkansas, Colorado and North Dakota, for example) so its fracking history and current practice is somewhat unique. Fracking or HF is a technology used (in a variety of different ways) to open or create cracks in an oil- or gas-bearing formation so that more product flows from the well. Almost all wells can benefit to some degree from fracking; about 85% of the wells drilled in the U.S. in the past decade have been fracked. In total, about 1.25 million U.S. wells have been fracked as well as about 200,000 in Canada, and at least 700,000 in other foreign countries and off- shore, so there is ample experience on which to rely in determining the most appropriate method to use and the environmental and safety rules that are appropriate. These have been documented in very substantial detail by the American Petroleum Institute27 . The fracking operations by the oil and gas industry and its many sub- contractors are regulated by the governments of oil- and gas-pro- ducing states and nations. 3.1.1 Fracking Methods Most fracking begins with the construction of a drilling pad that may be 1-4 acres in area. The pad is now often covered with a thick polyethylene sheet and a thin layer of absorbent material (often just sand or soil) to minimize the impact of spills. The location of the pad site and the position of the drilling rig are primarily deter- mined from a variety of information on the geological substructure and the estimated probability of striking oil and/or gas, but a wide range of environmental factors are also considered. A drilling rig is brought in and situated over the intended well site. Vertical drill- ing is then begun. In the case of formations like Michigan’s Antrim shale, the hole is drilled down into the production zone, the rig is removed and preparations are made to frack the well (see below). A drilling rig requires a lot of energy to turn the rotary drill bit and is usually powered by high-torque diesel-electric motors but, in response to environmental concerns, more and more rigs are using engines powered by compressed or even liquefied natural gas (both must be trucked in, however, with the potential for damage to local roads). In some cases, lateral wells in shale may also be drilled using directional drilling technology originally developed in the 1980s by Mitchell Energy of The Woodlands, TX with some assistance from DOE’s National Energy Technology Labs near Pittsburgh28 . The lat- eral penetrates the hydrocarbon-bearing formation and provides more routes for product to enter the well. In the case of dry gas wells with no production of water or gas liquids, the lateral may be close to horizontal. In cases where liquids drainage must be man- aged or if the formation itself is not horizontal (common in basin structures), the lateral may be inclined to the horizontal. Laterals are typically 10-20,000 ft. in length but a few have been as long as 40,000 ft. Once the well is drilled (or more usually concurrently with drilling) all of the well is cased throughout in one or more layers of high-strength steel tubing (to withstand the overburden pressure) that are sealed to one another and to the well wall with cements developed for the purpose. This is especially true if the well passes through an aquifer, as most do, or through a part of the formation that may have low strength and therefore might collapse. Because the tubing must withstand fracking pressures (especially the lon- gitudinal stresses set up in the vertical bore), it is also normally constructed of high-strength steel and joints between tubing seg- ments are strengthened and may even be welded, although that is rare. Nevertheless, one of the most common reasons for well failures, usually during fracking when the internal pressure is high, is tube joint failure or even tubing failure. In severe cases this can result in the ejection of a section of tubing from the well along with the “Christmas Tree”, the complex arrangement of tubing at the top of the well that is designed to handle the produced gas or oil and that usually includes the blowout preventer(s). Very little fluid leaks under these circumstances because the fracking pumps immediately detect the pressure drop and shut down. Fracking of deep and/or directional wells is most often done with several hundred thousand to several million gallons of high-pres- sure (up to ~15,000 psig) water that contains about 10-20% of sharp sand or an equivalent ceramic with controlled mesh size and about 0.5% of five to ten chemicals that are used to promote flow both into and subsequently out of the fractured formation. The list of chemicals includes hydrochloric acid to dissolve minerals and initiate cracks in the formation. Biocides such as glutaraldehyde or quaternary ammonium chloride may be added to eliminate bacteria that produce corrosive byproducts. Choline chloride, tetramethyl ammonium chloride, or sodium chloride may be added as clay stabilizers. Corrosion inhibitors such as isopropanol, methanol,
    • 12 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 formic acid, or acetaldehyde may be dissolved in the water, along with friction reducing compounds, for example polyacrylamide. In some cases, scale inhibitors are mixed in, for example acrylamide/ sodium acrylate copolymer, sodium polycarboxylate (commonly used in dishwasher detergents), or phosphoric acid salt. Surfactants such as lauryl sulfate are added to prevent emulsion formation, and in some cases, the surfactant is dispersed in a carrier fluid such as isopropyl alcohol. To adjust the pH, sodium or potassium hydrox- ide or carbonate is used. All of these, of course, are present at very low levels. Table 1 gives an overview of typical fracking fluid components. The sand or ceramic acts as a so-called “proppant” and helps to prop the cracks open. Sometimes, more complex proppants are used - rigid fibers, for example, or ceramic particles of controlled size and geometry. Calcined bauxite is common since it has very high crushing strength. To facilitate fracking, the steel casing that is inserted into the well is typically penetrated with pre-placed explosive charges (shaped charges are common). The fracking mixture flows into the forma- tion through the resulting holes, and these holes subsequently provide a route for product flow back into the production tubing. In deep wells with long laterals, the fracking may be done in stages, beginning at the far end of the well bore, with the later stages sep- arated by a temporary plug to isolate the section being fracked. Once each section is fracked, the plug is removed and the same fracking solution may be used for the next segment. Once the well is fracked, the fracking water that can be recovered (usually between 25 and 75% of the total used) is pumped out of the well or (if gas flows from the well under sufficient pressure) flows out of the well along with the produced gas. Wells in oil-bearing formations, especially those involving shale, are much more likely to require pumping. The ‘lost water’ disappears into areas around the fracked formation or enters deep aquifers in which it is diluted and eventually lost. At the concentrations typically used for HF, most of the chemicals employed today are not considered toxic or carcinogenic. However, not much is known about how these chem- icals interact with the various constituents of the formations deep underground, and it is conceivable that under certain conditions new compounds, some of them not benign, may be formed. TABLE 1: Typical Fracking Fluid Components NOTE: Not all components may be used in every well Component Concentration Reason Common Uses Fresh Water 80.5% Solvent or carrier Drinking Sand or ceramic 10-20% Proppant – keeps fractures open to permit oil/gas flow Playground sand, drinking water filtration Acids (usually HCl) 0.12% Helps dissolve minerals, initiate fractures in rock Swimming pool cleaner Petroleum Distillates 0.088% Dissolves polymers, reduces friction Mineral Oil – laxative, makeup remover, candy Isopropanol 0.081% Viscosity increaser Antiperspirant, glass cleaner, first aid antiseptic Potassium chloride 0.06% Creates brine carrier fluid Low-sodium table salt substitute Guar gum 0.056% Water thickener for sand suspension Thickener used in cosmetics, baked goods, ice cream….. Ethylene Glycol 0.043% Prevents scale deposits in pipe(s) Automotive antifreeze, household cleansers, deicer, caulk. Sodium or Potassium Carbonate 0.011% Improves the effectiveness of other components such as cross-linkers Washing detergents, soaps, water softeners, glass, ceramics Sodium chloride 0.01% Stabilizes gel polymer chains Table salt Polyacrylamide 0.009% Minimizes friction between fluid and pipe Water treatment, soil conditioner Ammonium bisulfite 0.008% Oxygen remover to prevent pipe corrosion Cosmetics, food and beverage processing, water treatment Borate salts 0.007% Maintains fluid viscosity as T increases Laundry detergents, hand soaps, cosmetics Citric acid 0.004% Prevents precipitation of metal oxides Food additive, foods and beverages, lemon juice N,N-dimethyl formamide 0.002% Prevents pipe corrosion Pharmaceuticals, acrylic fibers, plastics Glutaraldehyde 0.001% Eliminates bacteria from produced water Disinfectant, sterilizer for medical or dental equipment
    • 13 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 3.1.2 Water Acquisition and Disposal The so-called “flowback water” presents a disposal problem. Not only does it contain the generally harmless chemicals that were introduced with it, it also may contain chemicals picked up from the formation during fracking. It also contains rock debris from the drilling process, some of which may be slightly but not dangerously radioactive in cases where the sediments that formed some of the layers in the shale contained 238 uranium or even small amounts of 226 radium. It also contains drilling mud (mostly insoluble barium sulfate). Usually the flowback water is stored temporarily in closed tanks or open lagoons (tanks have the advantage that any methane and other gases emitted can be collected and used; open lagoons are not allowed in Michigan). Technologies are being developed that will allow the processing of this water stream and prepare it for legal disposal in injection wells or, more likely, re-use, with the solids residue going to an approved landfill29 . Most formations also contain water prior to fracking. This espe- cially true of the Antrim (MI) and New Albany (IL, IN, KY) shales, which must almost always be pumped dry after drilling and frack- ing before gas can be produced (the Antrim shale contains no sig- nificant oil and C2 -C6 gas liquids are found in commercial quantities only in the eastern extremity of the onshore part of the shale—it continues under Lake Huron). In many cases, this water may be highly saline and also presents a disposal problem. Interestingly, in the Antrim, and also in the geologically similar New Albany shales in Indiana, Illinois and Kentucky, millions of years of rain and glacial water (‘meteoric water’) seem to have greatly diluted the water in the shale and effectively pushed the high salinity water down to lower levels. As a result, the produced water is more easily dis- posed of but may still be too saline for conventional water treat- ment plants. In Michigan, in the Antrim and the northern Niagaran Reef structures, the primary dissolved solid is sodium chloride but significant concentrations of calcium and magnesium ions are also present, along with barium and strontium. Antrim fracking requires only a relatively small amount of water since the most commonly-used fracking technology is nitrogen foam fracking. In the much deeper Collingwood and Utica shales and especially in the A-1 Carbonate that overlies them, it appears that deep vertical wells accompanied by extensive and possibly multiple lateral wells are required to access enough gas to justify drilling. This leads to a substantial need for water. In all such sit- uations, the greater volumes used (and hence also produced as flowback water) are becoming a cause for local concern. In addi- tion to disposal and spillage issues (in the latter case, especially of concentrated chemicals on the well pad), there is a concern that sourcing such “large” amounts of water could have an impact on the sustainability of local water resources. This is a somewhat mis- placed concern since many other much larger uses of the same water exist—crop irrigation, for example. However, as mentioned previously, flowback water in Michigan is disposed of via deep well injection. Of course, concerns over water are far more severe in the western U.S. where water is less readily available than in Michigan. Water Treatment Several efforts are being made to develop a treatment methodol- ogy for fracking flowback water. The primary challenge in doing so is the extreme variability in the chemical composition of the flow- back stream, mostly due to the wide range of possible compounds that is picked up from the formation. High salinity, organics and total dissolved solids are the greatest problems. While the ingo- ing fracking solution consists of 90% water with up to 20% of sand or other inert proppant and about 0.5% of chemicals that, for the most part, are identical to those used in many other above-ground applications, the return water has an unpredictable amount of salinity and organic content that can vary quite widely, even among neighboring wells (the proppant mostly remains in the formation). The best that can be expected is a generic treatment basic water treatment system that must be “tweaked” for every individual case with the objective of producing re-usable fracking water. 3.1.3 Well Completion Issues The often-postulated percolation upward of fracking water used in deep, long lateral well extensions to contaminate drinking water aquifers near the surface through the intervening impermeable rock formations is highly unlikely and has never reliably been shown to have occurred. There is a lot of impervious rock between a deep directional (lateral) well and the surface. Leakage, even if some of the intervening rock is fissured vertically, is geologically very unlikely. Other routes, however unlikely, are remotely possible. These include leakage via neighboring wells that are too close to the active well (a problem that can easily be avoided). In an early 2012 case in Alberta, a vertical well was inexplicably drilled and fracked closer than regulations (and common knowledge) allowed to an existing well in the same formation and fracking water at high pressure moved horizontally through the formation made its way up the existing well which was not designed for the fracking pres- sure used. Fortunately, damage was minimal and no water supplies were affected (Wilson JR, personal observation). Faulty well drilling or (more commonly) well completion issues are a different issue. The following schematic (Figure 5) shows a typical drilling rig configuration with a list of major components, using the terminology most common in the industry. It is important to understand that almost all wells, whether or not they are intended for fracking, are drilled in this way—included any directional seg- ment. The drilling rig is then removed and fracking, which requires different equipment, is begun. However, while the rig is in place, a wide range of equipment, including casings of various sizes to
    • 14 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 production tubing and sealing cement is placed in the well bore along with various centralizing spacers, packers, and other devices that assist in subsequent stabilization of the well during and after fracking. While the American Petroleum Institute attempts to set standards to be followed by the industry, there is a wide variety of different practices in use. Each well can present different chal- lenges and may require a different approach or set of procedures or equipment. Thus, drilling and fracking is far from the simple process that many seem to believe. An added factor is that much of the work is not automated and is still done by people, and people tend to make errors, espe- cially under the difficult working conditions common on a drilling rig, even when they have been highly trained. This is far less of a problem in the industry than it was, say, two decades ago, and catastrophic blowouts and spills are now very, very rare given that some 35,000-40,000 wells are drilled and fracked in the U.S. alone every year—for a total of well over 1.25 million to date. Changes in equipment design have served to minimize problems, but mis- takes still occur. As noted earlier, drinking water aquifers near the surface are pro- tected by a series of layers of high strength steel tubing (casing) and specially formulated cement. Once the cement has set, it should be impossible, even during the extreme pressures applied during fracking, for any of the fracking solution or “slickwater” to reach the aquifer. However, problems can occur in well construc- tion. This is usually the result of faulty materials, faulty down-hole equipment, faulty casing joints or welding, faulty materials or (much less often than was once the case) faulty cementing. The following table shows the frequency of occurrence of the most common errors in well construction. Note that “poor cement” is a minor offender. Note also that Michigan has been relatively free of the problems listed. The establishment of a reliable barrier against gas leakage is critical to all gas wells, whether or not they are subjected to HF. Methane is a major global warming agent during its relatively long lifetime in the atmosphere. Cement is the major barrier used, and historically this was often a source of gas leaks at the wellhead or into drink- ing water aquifers. Changes in cement formulation and methods of use have greatly reduced the problem, but it still occurs occa- sionally, especially, for some reason, in Canada. Gas is distributed nation-wide through a very complex pipeline system which pres- ents numerous opportunities for leaks. Diligent work over many years has reduced these to less than 2% of the gas produced in some cases but there is clearly room for further improvement30 . The simplified schematics on the following pages show the well- head equipment that is typically required for a well producing oil Figure 5: Schematic of an oil or gas drilling rig31,32 1. Mud Tank 2. Shale shakers 3. Suction line (mud pump) 4. Mud pump 5. Motor or power source 6. Vibrating hose 7. Draw-works 8. Standpipe 9. Kelly hose 10. Goose-neck 11. Traveling block 12. Drill line 13. Crown block 14. Derrick 15. Monkey board 16. Stand (of drill pipe) 17. Pipe rack (floor) 18. Swivel (On newer rigs this may be replaced by a top drive) 19. Kelly drive 20. Rotary table 21. Drill floor 22. Bell nipple 23. Blowout preventer (BOP) Annular type 24. Blowout preventer (BOP) Pipe ram & blind ram 25. Drill string 26. Drill bit 27. Casing head or Wellhead 28. Flow line
    • 15 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 Figure 6: Typical pump jack for oil33
    • 16 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 from a well not producing under its own pressure (as is the case with almost all shale oil wells) with gas as a by-product31,32 . In the case of gas-only production, there is no walking-beam pump and production comes from the natural gas pressure in the well. In all cases, there are typically two additional casing runs at the top of the hole that are not shown here—the conductor casing which is wide and only very short and is designed to provide a foundation for subsequent well casing; and a longer run of casing, separated from the conductor casing by a cement seal and designed to sep- arate the well from any aquifers through which it passes. This lining extends below the lowest aquifer used for drinking water. The main well casing (shown in pink) is separated from the bore wall (and any other wider casings) with cement (shown as the “annulus”). Although rare, problems associated with fracking were more often than not due to human error in the selection and operation of down-hole equipment (spacers, centralizers, packers, tubing joints, blowout preventers) and, of course, a wide variety of well maintenance and workover operations. The well and its hardware must be built for the specific fracking operation intended (with an appropriate margin of error) and must therefore, in most cases, withstand high internal pressures without failure. Table 2 lists some of the failures that occurred over a 36-year period between 1960 and 1996 in wells drilled and in most (but not all) cases fracked in Louisiana, Texas and nearby shallow-wa- ter offshore locations. While few of these failures would occur today due to improved personnel training, better-designed pro- cedures, improved equipment, and better technology/materials, they emphasize the need to maintain constant vigilance over these issues if disasters such as the recent Deep-Water Horizon/ Macondo blowout (BO) in the Gulf of Mexico are to be avoided. Human errors are not limited to those listed. Very often, especially TABLE 2: Results of a Survey of Blowout Modes, 1960 to 1996 (36 years), Louisiana + Texas + Offshore Barrier Problems Primary Barrier # BOs Comments Swabbing 158 One of several methods of ‘liquids unloading’ a well. A plunger that seals to the wall of the production tubing is used to lift water or other fluids out of the well. Mishandling of this often leads to costly loss of well control (LWC). Low drilling fluid weight 50 Drilling fluid may have too much water or not enough high-density components such as barium sulfate (barites), thus failing to hold gas down in the well and/or resulting in lost fluids circulation. Drilling break/unexpectedly high pressure 45 Very common, normally handled by blowout preventers. This may have triggered BP’s Macondo disaster where two BOPs in series apparently malfunctioned. (BOP = Blowout Preventer) Formation Breakdown/Lost Circulation 43 May lead to complete loss of well or at least the directional section. Wellhead Failure 40 More common than it should be. Leaking fluids should be controlled by blowout preventers, now more reliable than during this survey period. Trapped/Expanding Gas 40 May get past drilling fluid but should be stopped by BOP at wellhead. Older BOPs were often unreliable. Gas Cut Drilling Fluid 33 Effectively results in loss of drilling fluid density and hence blowout – should be prevented by BOP. Christmas Tree Failure 23 Usually due to errors in fabrication such as defective welds or valves. Stopped by actuation of wellhead BOP. While Cement Setting 20 Caused by failure to allow enough time to set – gas (usually) gets by cement seal while it is still in slurry form or ‘plastic’ – establishes ‘channels’ that are difficult to fill or plug. Unknown Reason 19 Usually a blowout resulting from a cascade of events of which first step is unknown. In some cases, even after detailed analysis, a reason may not be identified. Poor Cement 16 Cement formulations have improved greatly since 1996 and “poor cement” problems are now very rare but still occur occasionally where inexperienced or cost-cutting subcontractors are involved. Tubing Leak 15 Once common due to poor materials, manufacturing and joint design, now rare. Tubing Burst 10 See above for “Tubing Leak”. Now extremely rare. Tubing Plug Failure 9 Can still occur, usually because of poor installation. Plugs are used to isolate sections of tubing during perforation and fracking. Can be difficult to ‘fish’ out. Packer Leakage 6 Caused by improperly designed packer or improper installation. Now very rare. Annular Losses 6 A catch-all for gas leakages past cement that should fill and seal the annulus. Usually the result of a poor cement job. Still occurs, albeit rarely, in the U.S. Uncertain Reservoir Depth/ Pressure 6 Inexcusable - there is no reason why this information should not be available and used in process design.
    • 17 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 Secondary Barrier # BOs Comments Failed to Close BOP 78 Can be mechanical problem or, more often, human error BOP Rams not seated 14 Due to fouling, wear, corrosion….. Unloaded to quickly 13 Too-quick liquids unloading may not provide operator time to close off well or operate BOP. One of the dangers of the old open-well liquids unloading practice. DC/Kelly/TJ/WL (see Figure 5) stuck in BOP 5 Any hardware still in the BOP will prevent closure. BOP Failed After Closure 66 Poor selection of BOP for the pressures involved, poor BOP construction, poor manual operation of BOP. Now very rare. BOP not in place. 43 Inexcusable effort to cut costs. Fracture at Casing Shoe 38 Can permit annular gas leakage. Rare, but still occurs. Casing Leakage 23 Once common due to poor materials, manufacturing or joint design, now very rare. Diverter Activation 19 Used to divert emergency fluid flows away from the drilling rig in cases where the well bore cannot be shut in until more permanent remedial action can be taken. String Safety Valve Failed 19 Drill string safety valve is designed to control pressure kicks (a form of annular flow) during drilling. Failure results in gas leakage unless the BOP can be closed. Now rare. Formation Breakdown/Lost Circulation 15 Serious, can result in loss and subsequent plugging of well. String Failure 13 A major problem because the lost (lower) part of the string must be fished out of the hole. Alternatively, the original hole may be plugged and a new one drilled. Both costly. Casing Valve Failed 11 Now almost unheard of due to improved design and materials. Wellhead Seal Failed 10 Usually a cement failure, now rare. Failed to Operate Diverter 7 Usually an operator failure in an undisciplined “every man for himself” culture. Very rare now. Christmas Tree Failed 7 Usually due to errors in fabrication such as defective welds. Stopped by actuation of wellhead BOP. Very rare as a secondary failure. Diverter Failure 17 More common 20 years ago, now very rare. in deep drilling (10,000 ft. or more), little is known about the shale formation that is to be accessed or how it varies in the directions parallel to the plane of the shale that will be directionally drilled. This lack of knowledge can lead to costly problems such as dry wells or difficulties in accessing sufficient gas, gas liquids or oil to justify the very high cost of drilling long laterals. The following record is taken from the work of Skalle and Podio34 . The comments are those of the present writers, based on some 40 years of experience. The list will provide the reader with some evidence that well-drilling is not a simple operation. In fact, it is technically very complex (and costly) with numerous opportunities for human error and equipment failure. Fortunately, most of these have been overcome by diligent engineering studies and steadily improving technology, materials and construction methods and, above all, operator training. As noted earlier, a short but sturdy conductor casing is common to almost all well installations. This provides a foundation for the con- struction of the rest of the well. Thus, in addition to the conductor casing, there is a surface casing whose primary “duty” is to protect any surface aquifer, especially one used for drinking water. It must therefore extend below the deepest aquifer of interest—usually only a few hundred feet. The well is drilled initially at a somewhat larger diameter to accommodate it. This is then followed by the main well casing, which extends the full length of the well. All of the conductor casing, the surface casing, the well bore casing and the production tubing (shown in blue in the preceding figure) are separated from one another and from the bore hole by an annulus filled with cement sealing materials. Many types of cements are used, each designed for a specific set of down-hole conditions, so it is important that these conditions be known before the cement is selected. For example, some cement contains special ingredients such as latex to enhance bore and annulus sealing but cannot be used at the greatly elevated down-hole temperatures (>200o C) found in a few areas35 . For some reason, many shallow- to moderate-depth gas wells in Canada (Alberta and Saskatchewan) and in the northern states of the U.S. (e.g., ND, MT) suffer from a gas migration problem. The TABLE 2 (Continued): Results of a Survey of Blowout Modes, 1960 to 1996 (36 years), Louisiana + Texas + Offshore Barrier Problems
    • 18 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 gas from the well (whether fracked or not) somehow flows up the cement-sealed annulus and either enters an aquifer or leaks into the atmosphere. Most are not serious, but they indicate a problem with cement formulation or installation. All well operators and their subcontractors are required to adhere to API standards36 , so the frequent failures are a mystery. These errors in what is part of well completion are still of concern but can be prevented by good engineering and wellhead design accompanied by thorough inspection. It is more common for gas, rather than fracking solution, to leak past the casing and cement and enter an aquifer or reach the surface simply because the low-viscosity gas finds an easier path than the much higher viscos- ity water. This reportedly occurred recently due to a casing failure in a well just SW of Traverse City, MI37 . Hal Fitch, Director of Oil, Gas, and Minerals, for the Michigan Department of Environmental Quality (DEQ) has stated that no fluid spilled and there was no damage done to the environment as a result of this incident (Hal Fitch, MDEQ, pers. comm.). There have been a few reports of gas contamination but so far no confirmed reports of fracking water contamination of drinking water aquifers. As noted in our earlier report, contamination of water wells by naturally-occurring gas is common in many areas of the U.S. and Canada and is easily dealt with except in cases when the gas contains large amounts of sulfur (in the form of H2 S). Gas in water can occur when the water well is unintentionally drilled into a gas-bearing stratum such as coal or a non-commercial shale bed. Therefore, it is considered prudent to collect baseline data before drilling for oil and gas begins, to assess to what extent the existing water wells are contaminated by naturally-occurring gas. Baseline data collection requires great care to avoid misleading contamination of the source water. For example, the EPA drilled two wells in Pavillion, a small town in Fremont County, Wyoming. The objective was to assess whether hydraulic fracturing had contami- nated drinking well water in the vicinity of the town. Unfortunately, as was subsequently pointed out by the USGS and many others, the EPA apparently knew little of water well drilling and contaminated the samples obtained with drilling fluid, well drilling residue, and hydrocarbons. They also used drill pipe and well casing that was not rust-resistant (stainless steel is normally required, especially in test wells). In that area, as in most of Wyoming, natural gas occurs very commonly in well water, perhaps derived from coal beds; thus the presence of gas does not indicate contamination by hydraulic fracking. The EPA results were therefore inconclusive38-394041 . 3.2 Well Stimulation Technologies Used In Michigan 3.2.1 Directional Drilling Recently, there has been only very limited activity in terms of directionally drilled wells in Michigan with the exception of a few exploratory wells drilled in deep formations such as the A-1 and A-2 Carbonates and the Collingwood/Utica shales beneath themii . Many of the wells in the Antrim formation are vertical since the shale there is naturally fractured horizontally, and directional drill- ing would not offer much advantage. Historically, Michigan was a pioneer in directionally drilled “slant holes” and in the develop- ment of micro-resistivity dipmeter analysis and side-tracking of wells targeting the Brown Niagaran (now referred to as the Guelph Dolomite/Ruff Formation) pinnacle reef formation. Today, one can still find hundreds of such directionally drilled boreholes in these areas, some of them made for formation drainage purposes. Since 2008, Michigan’s DEQ has issued more than 50 active permits for “high volume” (those using more than 100,000 gallons of hydrau- lic fracturing fluid) hydraulically fractured wells, and more than half a dozen applications are pending42 . Most have been drilled (or will be drilled) in the A-1/A-2 Cs and Collingwood, Utica, generally in the northern LP and also in the Black River/Van Wert shale in Hillsdale County on the Ohio Border (most of this shale formation lies in Ohio). However, at the time of this writing, there was only one drilling rig in Michigan, and it is not clear if it is currently active. So far, drilling in Michigan’s deep shales and carbonate forma- tions has produced disappointing results, but these are early days (Encana Corporation, pers. comm.). A high proportion of the wells drilled in the Collingwood shale, once considered so promising that it led to the sale of a record number of leases at record prices on state land in 2010, have been dry or have been permanently or temporarily abandoned. The high cost of drilling these wells is not compatible with the current low price for the dry gas that most of them produce. Nevertheless, a handful of wells in the Collingwood are producing gas and at least one has reported a show of oil43 . Oil has also been found in the A-1 Carbonate, but that well is recorded as shut-in, presumably indicating that it was not commercial. Almost 10,000 wells, almost all of them vertical, have been drilled in the Antrim shale and almost all have been fracked. While little or no new drilling is possible, workovers of existing wells are common. Like the original well completions, this usually involves fracking, sometimes with gelled nitrogen foam, less often with slickwater, to ii Note that while the Collingwood lies well below the A-1 and A-2 carbonate formations in Michigan, it surfaces to form the shore of southeastern Lake Huron in Ontario (before it disappears under the Blue Mountain Shale). It would be interesting to have access to the details of the geology of the Collingwood under Lake Huron.
    • 19 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 stimulate the productivity of the well. This method has been quite successful, and the Antrim shale still continues to produce signif- icant amounts of gas44 . As the Antrim field as a whole declines, the middle portion, mostly in Montmorency and Otsego Counties, produces an increasing amount of CO2 . The current level in the gas being produced is 30-35% by volume. This CO2 is separated from the methane and any other hydrocarbon gases that may be pres- ent and is then (a) released to atmosphere (no longer considered a good practice), (b) sequestered underground, often in depleted gas wells or (c) used for enhanced oil recovery (EOR). Several trials of CO2-induced EOR have been run in the Niagaran Reefs, appar- ently with good success for a first attempt and, as a result, more oil may still be recovered from this already quite prolific formation45 . The northern trend of the Niagaran reef crosses and underlies the Antrim shale and is still a major producer, depending on where drilled, of gas, water or oil. It has been Michigan’s greatest con- tributor of energy products since its discovery in the late 1960s and appears likely to continue to be a major factor for some years to come. Its geology is very complex; we leave that discussion to others in this series of reports. The gas found in the Antrim formation, like that in Ontario’s Kettle Point extension of the Antrim and also the New Albany shale in Indiana/Illinois/Kentucky is known to be biogenic in origin—i.e., it is derived (and may still be being derived) from anaerobic bacte- rial action on organic matter deposited long ago when the shales were formed. Methanogenesis in all of these cases has apparently been aided by the “washing” of the formation by glacier and rain/ snow-derived meteoric water which has pushed what were once high salt levels down to greater depths46 . This has prevented the inhibition of biological action by what would otherwise have been high salinity. It may also explain the very high water content of these shales. Many other shales appear to have produced meth- ane by thermal action (thermogenesis)—i.e., by intense heating of the organic matter, usually through geothermal activity. The source of most of the organic matter in most shale appears to have been algae that were carried into relatively quiescent areas by the flow of streams originating from bodies of fresh water (most shale was apparently formed in saline waters, presumably on estu- aries or sea beds). Other contributions were possibly made by small vegetable particles such as leaves or roots ground up by the water flow over stones or sand. Whether the methane is biogenic or ther- mogenic in origin then was determined by the subsequent history of the deposits over a period of many million years, during which layers of mineral sediment and clays were often also deposited. Many of the large number of other oilfields in Michigan have been important but relatively small and shallow. Most if not all are now in decline or have been plugged but few have yet been subjected to the aggressive EOR methods that have become common in other states such as Texas. This means that most (up to 90% in some areas but an average of 60-65%) of Michigan’s original oil (but not gas) remains in the ground. A field-by-field study to determine the probable effectiveness of EOR technology seems long overdue. While some of the 14,542 oil wells and 13,269 gas wells drilled in Michigan (not to mention a few of the 22,067 dry holes) as of September 2012 were subjected to fracking47 , it seems likely that not only directional drilling from existing producing oil wells but also HF workovers, where allowed by the MDEQ, could be a source of additional oil. Gas fields do not suffer as much from this limitation. Provided that a formation can retain enough porosity and does not become water- flooded, gas production will continue until the fall in well pressure makes the well uneconomical to operate. As mentioned earlier, vacuum extraction may obtain a little more gas but is frowned on in most jurisdictions 3.3 Current Oil and Gas Industry Practices in Michigan 3.3.1 Current Practices In general, Michigan oil companies have not been technology lead- ers in oil and gas exploration and production. They have followed much the same conservative (but safe and usually environmentally sound) pathway of many other mid-range producing states such as Ohio and Indiana. This may change with the recent discovery of probable gas and perhaps oil in formations such as the A-1 and A-2 Carbonates and perhaps even the deeper Collingwood and Utica shales (including the Utica in Ohio), but little appears to be known about these on a micro-geological scale and they will be costly to explore and develop based on the few results obtained so far. Directional drilling and fracking will be required, based on what is known of the limited permeability of these formations and the laterals will probably have to be of unusual length to ensure rea- sonable gas production. Even with substantial laterals, it still may not be possible to open pathways to enough gas or gas liquids to make wells in these formations economically viable. Significantly higher gas prices may be needed. This is discussed further in the next section. So far, there has been very little experience with the A-1, A-2, or the Collingwood and Utica shales. Encana, Chevron Michigan, Merit Energy and several others all have permits to drill in Michigan’s Collingwood formation, but there have been few holes completed to date and, as previously noted, there is now only one rig in total drilling (potentially) in the state. Given the limited experience to date with the Collingwood, most of it negative or neutral, it seems unlikely that any company will establish an aggressive drilling
    • 20 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 TABLE 3: Challenges Encountered in Well Drilling and Completion Key Variable Comments and Notes Shale Shale types vary widely and are often surprisingly heterogeneous. There is often more than one shale layer interspersed with other rocks such as sandstone, limestone or clays. Black shales are usually high in Total Organic Carbon (TOC) while white or grey shales are often high in limestone and/or quartz but may nevertheless contain natural gas or even gas liquids (which may have migrated from elsewhere). Shale gas is typically found in fine-grained reservoir rocks in which the gas is self-sourced from the TOC present. Some of the gas is stored in the adsorbed state, predominantly in the organic fraction. Total Organic Carbon (TOC) Can be obtained by wireline logs but the data are often not an accurate representation of TOC. Confirmation is required by core sampling and analysis (costly!). TOC may vary widely in any shale bed, even over distances of a few feet. Organic geochemistry may not be a good indicator of formation prospects. Mapping TOC in a Formation Has the same limitation as TOC, above. Confirmation may require many exploratory drill holes – very costly in deep formations. Better methods of assessing formation potential are needed. Wireline Logs Only occasionally a good guide to organic content or formation prospects (see “TOC”). Results should be confirmed by other means. Shale area Shale beds are often highly variable, inconsistent, not always productive where expected, may have variable thickness, structure and TOC. This seems to be especially true of the Collingwood Shale. Shale thickness Will vary widely throughout a large shale bed such as the Collingwood. Productive shale may be present in multiple layers separated by non-oil (or gas)-bearing layers. Geochemistry Often unknown or very hard to determine. Even TOC history may not be clear (see below). Organic-rich layers may contain multiple minerals (quartz, alumina, aluminosilicates and carbonates) which make history obscure. Diagenesis (formation of sediments and conversion to sedimentary rocks) is usually clearer. Maturity Often unclear. Many gas shales, like the Antrim and Kettle Point, are relatively immature, contain biogenic methane (and may still be producing it) while others, even in the same formation may be thermogenic and/or much more mature. Thermal maturation structurally modifies the organic fraction, creating more macroporosity – and hence more adsorption sites. Presence of oil v. gas v gas liquids v. condensate determined by many factors including T, P, TOC and formation history. Adsorbed Gas Methane adsorbs preferentially in organic-rich shales, is often replaced by carbon dioxide (if present) as it is depleted by production (as in the central Antrim formation). Other organic and non-organic gas may compete for adsorption space, usually on the interior of pores. This can make determination of original gas in place (OGIP) difficult. In general desorption testing will indicate more gas than is actually adsorbed; the balance are probably free ‘pore gases. OGIP = Free Gas + Adsorbed Gas + Solution Gas…..but still not clear what it means since OGIP is often seriously overestimated by testing. Moisture Complicates analysis of core samples, which must be dried first. Competes with methane, etc., for adsorption “space”, may give spurious gas content results. Free Gas Most shale contains some non-adsorbed gas which may be trapped in closed pores until accessed by some form of fracking. Gas Capacities (see also OGIP) Often very difficult to determine precisely prior to actual production. Many promising wells are still a disappointment for this reason. Solution Gas Gas dissolved in liquids present (gas liquids, crude oil, condensate, water…..) that may be hard to determine prior to production. Pressure Usually, deep formations = high pressure due to overburden load above. Can be a problem during lateral drilling (due to well bore shape distortion) and in fracking deep laterals. May limit producibility. Temperature In geological time, may have had a major impact on, e.g., the conversion of TOC to gas, oil, etc. Higher temps tend to favor oil production, but not always. Producibility Some “tight” shale formations (as well as sandstones, coal beds, etc.) may be non-viable economically despite promising test data. The causative factors are those listed below. Such wells may initially produce strongly immediately after fracking but production soon declines. The problem? Poor access to the contained gas. Permeability This and the next ‘challenge’, porosity, go together. Permeability in rock is extremely difficult to measure precisely and as a result permeability data are often wildly erroneous. Permeability is highly dependent on porosity and whether pores are closed or open.
    • 21 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 Porosity Porosity of rock varies very widely. Pore sizes are generally larger in oil-bearing formations. If only closed porosity is present (as in a closed –cell urethane foam) this has a strongly negative effect on permeability. If partially interconnected pores are present, measured permeability may be much higher since it does not have to depend as much on slow diffusion through the formation rock. “Tight” reservoirs may be very porous (in terms of pore count) but may still be very unproductive if no way can be found to connect or access the pores. The pores typically contain some free gas and rather more adsorbed gas. The amount of adsorbed gas depends on the local pressure, temperature and TOC – a higher temperature results in less adsorbed gas. But if the pore is closed, the gas cannot be recovered. NOTE: • Porosity measurements using skeletal density measured by helium are always too high. • With other gases, correction for sorption is mandatory • Correction for pore compressibility is essential All in all, permeability and porosity are not easily quantified. Sedimentology Shales are initially formed by sedimentation, but are almost always later modified by temperature and pressure and sometimes by complex geochemistry. The organic matter is typically derived from algae and small plant particles and is the primary source of oil and/or gas in the shale. Some shallow shale can lose some or all of its organic matter by conversion to methane and seepage into the atmosphere. Others, like Michigan’s Collingwood shale, apparently have a much more complex history and have become very deeply buried by further sedimentation and subsequent geological action. The porosity, and hence permeability of these shales is probably established during these early formative processes. If the shale contains moisture or even free water containing dissolved species, this can materially alter its physical characteristics. Silica Content; Alumina Content CaO Content Most shale contains fairly high levels of silica, sometimes as quartz, and variable amounts of alumina and other oxides. In general, shales with high silica (70-80%) and low alumina (5-7%) exhibit relatively large pore sizes (10,000-100,000 nanometers) in a coarse grain structure while those with higher alumina (10-20%) and lower silica (50-70%) have much smaller pore sizes (1-10 nanometers) in a fine grain structure. Gas contained in high alumina shale is therefore likely to be much more difficult to access. Calcium oxide appears to have little effect until it reaches high levels, as in carbonate rock. Diagenesis The history of formation of the shale or other gas/oil-bearing rock from the original sediments is important in determining its ability to deliver the gas or oil to the operator, but relatively little is known of the relationship between diagenesis and well productivity other than the relationships described above. Porosity (above) decreases with diagenesis and effective stress due to compaction. Fracturing The purpose of formation fracturing, however it is done, is to gain access to more of the oil or gas contained in the formation. In formations with a coarse grain structure and large-scale porosity, such as the Antrim, this may be relatively easy but as the depth and hence pressure increases and especially in high-alumina formations with fine porosity, fracturing or ‘fracking’ may not achieve the desired result. Additionally, the original fracture orientation will depend on depth and hence pressure. In some formations, even when heavily fracked, only some 10% of the gas (and especially liquids) in place may be accessed, depending on the pore structure of the formation. Simply fracturing the rock into big pieces may have little effect. It may be necessary to reduce it to small-size particles, and that may not be practically achievable in deep wells. NOTE 1: microseismic data show what fracks, not what produces. NOTE 2: gas released from the rock matrix is strongly stress-dependent. In many reservoirs with widely-spaced fractures, the resulting low rate may be production-limiting. More Challenges Quantitative assessment of exploration targets. Determining intervals at which to frack or drill laterals. Predicting production rates. Predicting decline rates Predicting Estimated Ultimate Recoveries (EURs) If intervals of shale are thick, predicting drainage areas (spacing units) Remaining Unknowns for most reservoirs. What is the OGIP? Most data viewed as highly unreliable (see “adsorbed gas”). Production data seldom matches the OGIP number. What is the optimum interval to perforate and frack? Optimal Frack Design – e.g., # of stages, length of lateral? What is the drainage area/volume of each well? What is the recovery factor (percentage of oil or gas originally in place that is recovered)? What is the optimum well spacing and lateral spacing unit?
    • 22 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 program at present drilling and water costs and gas prices. The Collingwood requires wells of 10-12,000 ft. depth and laterals of roughly equal size unless some new method of fracking can be developed that will be more effective at fragmenting these very tight, low-porosity formations. Such wells can cost up to $10,000,000 if capital costs for gas refining and pipelines are included. At cur- rent refined gas spot prices of $3.50/1,000 cu. ft., over 3 billion cu. ft. of gas must be produced from the well just to break even. This means that it is doubtful that current gas prices will support the very high drilling cost. For the foreseeable future, we are likely to see only enough drilling to ensure lease retention in cases where agreements have a “use it or lose it” clause. This especially true in areas such as Kalkaska County where exploratory directional wells drilled to date have required far more water than is normal (an esti- mated 20 million gallons per well instead of the usual 5-7 million). The reason for this is not known but a large demand for water on this scale adds considerably to the cost to the well. Meanwhile, the Antrim shale will continue to be addressed with stimulation methods such as additional fracking to ensure the recovery of the maximum possible amount of gas. Drilling will continue on the limited number of remaining inactive leases on the Niagaran Reef northern trend (which stretches between Manistee and Presque Isle Counties in the northern LP) with most of the focus being on oil except in Manistee County, where the reef has always been a moderately good gas producer. Additional attempts at enhanced oil recovery can be expected in the north- ern Niagaran Reef. The southern trend of the Niagaran Reef which runs roughly from western Oakland County through Livingston and Eaton Counties to the northern end of the Albion-Scipio trend in Calhoun County) is unlikely to generate much that is new absent an aggressive enhanced oil recovery program. 3.3.2 Permeability and Producibility The challenge facing a would-be producer in formations like the A1 and A2 Carbonates and the underlying Collingwood is the very large number of variables that need to be understood and in some cases quantified before drilling can begin with a reasonable chance of economic success. If the formation to be drilled is two miles or more deep and if the laterals to be drilled cover in total a large area, the information that is required is very difficult to obtain. The following table attempts to summarize some of the key challenges facing the operator. Radioactivity Because uranium-238 is so widespread, albeit in very small concentrations, it accumulates in many sedimentary rocks – including shales. It has a half-life of 4.47 billion years and emits only harmless α-radiation. It is accompanied by about 1% of U-235 (half-life of 704 million years) and a minute amount of U-234. All are only very slightly radioactive and emit only weak α-radiation, which cannot penetrate even a thin paper tissue. Both flowback and produced water from some wells may be slightly radioactive or contain radioactive rock chips but are completely harmless. In a few black shale deposits, radium may also be found and typically shows up in the solid residue well from drilling, along with flowback water and drilling mud. The most common isotope is radium 226 which has a half-life of 1,600 years and emits only harmless α-radiation. Argonne Labs have determined that drilling residue containing uranium and radium can be landfilled if more than 10’ deep and suitably capped with clay. See Ellis (this series) for additional information. Flowback and Produced Water Flowback water is very unpredictable. It may contain all or only part of the dilute chemical cocktail pumped into the well, plus a lot of inorganic and organic material picked up from the formation, some of which may be geologically very old. It also carries a heavy load of drilling mud and rock chips, some of which may be mildly (and harmlessly) radioactive – see above. Increasingly, flowback water is being reused after removal of solids and organics and at least some salinity. The flowback water leaving the well gradually transitions to produced water. The latter originates in the formation and may contain high salinity (although not generally in biogenic shales since the salinity interferes with the biological action; there the salinity has often been driven lower by meteoric water) and dissolved or suspended organics. Dry gas wells (those producing no gas liquids or condensate) often produce little or no water. Produced water from oil-rich wells can be very complex and difficult to treat. While common, deep well disposal should be approached with caution – there is too much chance of future and unpredictable movement of the waste ‘plumes’, perhaps over very long times Methane Management Methane is a major Global Warming agent (25-100X as effective as CO2). Every effort must be made not to release it into the atmosphere. This has already caused many changes in well management methods, especially during well drilling and completion and liquids unloading, but more is needed. The natural gas industry, which has historically lost 2-5% of its product between wellhead and point of use, is making major efforts, for economic as well as environmental reasons, to capture this lost gas, e.g., by using improved compressor seals, more welded joints (especially to mount instruments) and reduced flaring during plant upsets. Much more focus is needed on this issue. Bacterial Activity In shallow black shales in particular, methane is produced under anaerobic conditions by the action of certain bacteria on the organic matter (algae, for example) in the shale. The product is termed biogenic methane. In other shales, some methane may be produced by the thermal decomposition of the organic matter and is termed thermogenic methane. In black shales, some of the bacteria may survive and can be present in the flowback or produced water. The amounts are small and the bacteria are harmless. The same or similar bacteria are found in stagnant ponds and marshland, for example, where they also produce methane anaerobically from rotting vegetation in the mud at the bottom of the wetland.
    • 23 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 3.3.3 Methane Leakage Another major issue for the entire natural gas industry is methane leakage. Successful efforts have been made for many years to gradually reduce the number of both large and small leaks in the vast and complex national distribution system because methane is a potent greenhouse gas. New-design pipeline compressors, once a major source, are now essentially leak-proof while instrumenta- tion is improving through the use of welded joints and changes in design that return operating gases to the plant. Field monitoring of fracked gas well sites, which were once fairly major contributors to methane leaks due to careless handling of flowback water and practices such as open-well liquids unloading, now shows them to be comparable to conventional gas wells producing under reservoir pressure. While it is quite difficult to arrive at accurate estimates of methane emissions48,49,50 , field levels of methane reported for HF sites are now typically very low, in the range of 0.4—0.6 vol%51 . The American Petroleum Institute (API) and the American Natural Gas Alliance (ANGA) recently issued a report on sources of methane emissions form natural gas production52 . URS Corporation and the University of Texas at Austin recently issued a report on methane emission factors for selected processes and equipment used in the natural gas industry53 . Wells for which the methane emitted during liquids unloading or collection of flowback water is captured and either used or included in the gas delivered to the well gathering system are increasingly referred to as “green completion” wells. Despite the fact that these practices have been used for several years by responsible operators, the development of “green completions” seems to be regarded as new. Regardless of its novelty, the prac- tice is clearly one that should be adopted as widely as possible. As far as Michigan is concerned, the Utica/Collingwood and A1 formations are extremely tight and so far the drilling mud returned contained only small amounts of methane. However, in other loca- tions, it is conceivable that the returned drilling mud and subse- quent fracking flowback water may contain a higher proportion of methane. Regardless of the amount or source, it must be captured. In Michigan, all drilling operations must adhere to the requirement that during HVHF operations all hydraulic fracture fluid must be contained in pipelines and steel tanks; the methane that is sepa- rated from the liquids is sent by pipeline to the gathering system. Today’s best practices for green well completions and testing would result in virtually no methane being flared or vented during the entire drilling, HVHF and testing and production operations. 3.3.4 Safety Fracking, like oil or gas drilling, involves complex equipment and procedures operated by humans rather than being automated. Errors and accidents do occasionally occur, sometimes leading to the escape of fracking water or, much more often, gas into the atmosphere or into groundwater or drinking water aquifers, but such events have become increasingly rare as all of regulations and industry practices and personnel training have improved, especially over the past ten years. Most recent errors have involved faulty equipment or its faulty installation. A preliminary review of the safety record accumulated over more than 30 years of high-pres- sure deep well fracking (and a much longer period of all forms of fracking) indicates that so far fracking operations in Michigan have had a relatively good safety record, but a more thorough examination of documentation regarding safety related incidents in Michigan’s oil and gas industry will be required before a defin- itive conclusion can be reached. With over 40,000 wells annually now being subjected to fracking in the U.S. alone and many more overseas, an occasional accident can be expected, just as it can for other industrial processes or the transportation sector, despite the extraordinary precautions and oversight that are now involved to ensure safety. People, however well trained, make mistakes. 3.4 Future Practices in Michigan For many in the oil and gas industry, the oil and gas rush in Michigan is over and has been for several years. Most of the majors like Shell or Exxon have pulled out of the state and, at current gas prices, deep drilling in the Collingwood and Utica seems unlikely to be commercially viable, at least in the near term. Unfortunately, only a modest amount of gas liquids and, so far, no oil, seems to be associated with those formations, at least in Michigan; thus, unless there is something much deeper, it is unlikely that Michigan will follow North Dakota in discovering a highly productive Bakken-like formation. Most of the other oil and gas resources in the state are in decline (although oil production saw a burst of activity between early 2010 and August 2012), and the amount of those commodi- ties produced by the state as a whole has been diminishing for 1-2 decades. While exploration for oil and gas is, almost by definition, an expensive hit-or-miss process, even with the availability of modern exploration tools (‘wildcatter’ is still a relevant description of many current-day small drilling companies), Michigan now seems to be unlikely territory for a major success. The state was extensively and intensively explored in the middle years of the 20th Century and experienced many small to medium sized successes in both gas and oil. The last notable finds—all at least 40 years ago - were in the northern Niagaran Reefs and the Albion-Scipio trend. The Antrim Shale, while still producing, has been known for almost 70 years. Given the extensive exploration history of the state and its now well understood geology, it seems very unlikely that Michigan will ever again see an oil or gas boom. The one possible exception, however apparently remote, is that significant recoverable gas will be found
    • 24 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 in the carbonate formations and the underlying Utica-Collingwood formation. These undoubtedly contain gas, gas liquids, and pos- sibly oil (albeit high-kerogen oil in the southern extremity of the Collingwood which can currently be recovered only by expensive retorting technology) but, absent new technology developments, seem unlikely to be developed in the next decade or so. The one possibility of enhancing production from Michigan oil wells seems to be enhanced oil recovery (EOR) technology that has so far been underutilized in the state. Only a limited experimental program with CO2 injection has been tried (with notable success) in the northern Niagaran Reef. There appear to be many old wells in Michigan, some of them still producing with the help of walking beam pumps that could possibly benefit from an EOR program. While the CO2 used for EOR in the northern Niagaran Reef came from the Antrim formation, that for southern wells would proba- bly have to come from much-needed (but not yet implemented) power station CO2 sequestration programs. This would be costly but should pay for itself in the greater amount of oil recovered and sold. 4.0 PRIORITIZED PATHWAYS FOR PHASE 2 T he objective of Phase 2 will be to begin addressing some of the science and technology gaps relevant to fracking by conducting a series of experiments. The experimental approach is to develop a fundamental understanding of the adsorption and release of hydrocarbons, carbon dioxide, and other chemical species of interest. These experiments will be carried out both on well-defined model sys- tems relevant for hydraulic fracking, as well as on actual shale sam- ples that are representative for geological formations in Michigan and Ontario, including samples of “black shale” that may contain organic residues. The topics to be addressed include: • Fundamental investigation of hydrocarbon adsorption/desorp- tion from sand and shale samples • Investigation of adsorption and surface reactions of typical individual fracking chemicals with sand and shale samples as a function of temperature and pressure. • Comparative study of water-based and water-free fracking methods • Study of the relationship between fracturing methods, fracture propagation and permeability in materials with different forms and distributions of gas-bearing porosity at high formation pressures LITERATURE CITED 1. Distribution of Oil and Gas Production by State [Internet]. Washington, DC.: U.S. Energy Information Administration, DOE; 7 January 2011. Available from: http://www.eia.gov/pub/oil_gas/petrosystem/petrosysog.html 2. Michigan’s Oil and Gas Industry Statistics [Internet]. Kalamazoo, MI: Western Michigan University, Michigan Geological Repository for Research and Education. Available from: http://wsh060.westhills.wmich.edu/MGRRE/reports/mi_oilgas_sum2007.shtml 3. Oil and Gas Database [Internet]. Michigan Department of Environmental Quality. Available from: http://www.michigan.gov/deq/0,4561,7-135-3311_4111_4231-98518--,00.html 4. United States Energy Information Administration [Internet]. Washington, DC: U.S. Energy Information Administration, DOE; December 2012. Available from: http://www.eia.gov/ 5. Schultz A, editor. Hydraulic Fracturing and Natural Gas Drilling. New York, NY: Nova Science Publishers; 2012. 158 p. 6. Montgomery CT, Smith MB. Hydraulic Fracturing: History of an Enduring Technology. Journal of Petroleum Technology [Online]. 2010 [cited 2012 Nov 7]; 26-41. Available from: http://www.spe.org/jpt/print/archives/2010/12/10Hydraulic.pdf 7. King GE. Hydraulic Fracturing 101: What Every Representative, Environmentalist, Regulator, Reporter, Investor, University Researcher, Neighbor and Engineer Should Know About Estimating Frac Risk and Improving Frac Performance in Unconventional Gas and Oil Wells. SPE Paper 152596. SPE Hydraulic Fracturing Technology Conference; 2012 February 6-8; The Woodlands, TX. Society of Petroleum Engineers; 2102. 80 p. Available from: http://www.kgs.ku.edu/PRS/Fracturing/Frac_Paper_SPE_152596.pdf 8. Industry establishes Canada-wide operating practices for shale, tight natural gas hydraulic fracturing [Internet]. Canadian Association of Petroleum Producers; c2009. Available from: http://www.capp.ca/aboutUs/mediaCentre/NewsReleases/Pages/operating-practices-for-hydraulic-fracturing. aspx?utm_source=English+Distribution+List&utm_campaign=cf0de957c5-Fracking+vs.+Beer&utm_medium=email&utm_term=0_cfc48e9fb7-cf- 0de957c5-59999997
    • 25 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 9. Henricks M. Energy industry works to recycle hydrofracking waste water. American Recycler [Internet]. 2012 May. Available from: http://www.americanrecycler.com/0512/1517energy.shtml 10. Associated Press. EPA lowered estimates of methane leaks during natural gas production Fuelfix [Internet]. 2013 April 28. Available from: http://fuelfix.com/blog/2013/04/28/epa-lowered-estimates-of-methane-leaks-during-natural-gas-production/ 11. Westbrook JR. Images of America: Michigan Oil and Gas. South Carolina: Arcadia; 2006. 128 p. 12. Michigan Geology Maps [Internet]. Michigan DEQ; c2001-2013. Available from: http://www.michigan.gov/deq/0,1607,7-135-3304-116670--,00.html 13. Michigan online oil and gas information system [Internet]. Michigan Department of Environmental Quality. Available from: http://ww2.deq.state.mi.us/mir/ 14. Ontario’s Historical Plaques [Internet]. Ontarioplaques.com. Available from: http://www.ontarioplaques.com/Plaques_GHI/Plaque_Grey04.html 15. Oil shale in Colorado, Utah and Wyoming [Internet]. Environmentally Conscious Consumers for Oil Shale; c2011. Available from: http://www.eccos.us/index.php?option=com_content&view=article&id=10&Itemid=6 16. Michigan Field Production of Crude Oil [Internet]. Washington, DC.: U.S. Energy Information Administration, DOE; 2013 May 5. Available from: http://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=mcrfp_smi_1&f=m 17. Michigan Natural Gas Gross Withdrawals from Shale Gas [Internet]. Washington, DC: U.S. Energy Information Administration, DOE; 2013 May 5 [cited 2013 May 6]. Available from: http://www.eia.gov/dnav/ng/hist/ngm_epg0_fgs_smi_mmcfm.htm 18. See Barchart [Internet]. Barchart.com Inc.; c2013. Available from: www.barchart.com 19. Otis CB, Carter TR, Forner L. Preliminary results of a shale gas assessment project in Ontario, Canada. #50390. Adapted from oral presentation at AAPG Eastern Section Meeting; 2010 Sept. 25-29; Kalamazoo, MI. Association of American Petroleum Geologists. 2011. 27 p. Available from: http://www.searchanddiscovery.com/documents/2011/50390belandotis/ndx_belandotis.pdf 20. Hamilton T. Alberta firm eyes Ontario’s untapped shale gas. Toronto Star [Internet]. 2010 March 20. Available from: http://www.thestar.com/business/2010/03/20/alberta_firm_eyes_ontarios_untapped_shale_gas.html 21. FracFocus Chemical Disclosure Registry [Internet]. Groundwater Protection Council and Interstate Oil & Gas Compact Commission. c2013 Available from: http://www.fracfocusdata.org/ 22. Hydraulic Fracturing [Internet]. Bismarck, ND: North Dakota Department of Mineral Resources. Available from: https://www.dmr.nd.gov/oilgas/presentations/ActivityUpdate20132013-01-29ASTMJacksonville.pdf 23. U.S. Environmental Protection Agency. Underground injection wells in region 5. [cited 2013 March 10]. Available from: http://www.epa.gov/r5water/uic/r5uicwells.htm#michigan 24. Michigan Department of Environmental Quality. Michigan’s Oil and Gas Regulations. 2006 [cited 2013 February 14]. Available from: http://www.michigan.gov/documents/deq/ogs-oilandgas-regs_263032_7.pdf 25. Shires T, Lev-On M. Characterizing Pivotal Sources of Methane Emissions from Natural Gas Production: Summary and Analysis of API and ANGA Survey Responses. Final Report. American Petroleium Institute and America’s Natural Gas Alliance; 2012 Sept 21 [Updated 2012 Sept; cited 2013 May 6]. 50 p. Available from: http://www.api.org/news-and-media/news/newsitems/2012/oct-2012/~/media/Files/News/2012/12-October/API-ANGA-Survey-Report.pdf 26. Wilson JR, Schwank JW. Environmental and other issues related to the use of fracking for the enhanced recovery of natural gas and oil from shale, tight rock and coal beds. Unpublished draft manuscript. 27. American Petroleum Institute. Overview of Industry Guidance/Best Practices on Hydraulic Fracturing. American Petroleum Conference; 2012 April 18; Traverse City, MI. 28. Gold R. The man who pioneered the shale-gas revolution. Wall Street Journal [Internet]. 2012 Oct 23. Available from: http://online.wsj.com/article/SB10001424052970203630604578075063683097342.html 29. Smith KP, Blunt DL, Williams GP, Arnish JJ, Pfingston M, Herbert J, Haffenden RA. (Argonne National Laboratory). An Assessment of the Disposal of Petroleum Industry Norm in Nonhazardous Landfills. Final Report. Tulsa, OK: National Petroleum Technology Office, DOE; 1999 Oct. W-31-109-ENG-38. 92 p. Available from: http://www.netl.doe.gov/kmd/cds/disk23/G-Soil%20Projects/NORM%5CW-31-109-ENG-38%5CDOE-BC-W-31-109-eng-38-8.pdf 30. National Greenhouse Gas Emission Data [Internet]. U.S. Environmental Protection Agency [cited 2013 May 6]. Available from: http://www.epa.gov/climatechange/ghgemissions/usinventoryreport.html 31. Tosaka. Oil Rig NT.PNG. Oil field engineering beginner’s guide. ISBN 4-303-73420-9. Available from: http://commons.wikimedia.org/wiki/File:Oil_Rig_NT.PNG 32. Tosaka. Oil Rig NT.PNG. Story of petroleum. ISBN 4-542-90229-3. Available from: http://commons.wikimedia.org/wiki/File:Oil_Rig_NT.PNG 33. TastyCakes. Pump Jack labeled.png. Available from: http://commons.wikimedia.org/wiki/File:Pump_Jack_labelled.png
    • 26 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: TECHNOLOGY TECHNICAL REPORT, SEPTEMBER 2013 34. Skalle P, Podio AL. Trends Extracted from 800 Gulf Coast blowouts during 1960-1996. SPE/IADC 39354. IADC/SPE Drilling Conference; 1998 March 3-6; Dallas, TX. 1998.  Society of Petroleum Engineers; 1998. 8 p. 35. American Petroleum Institute. Isolating Potential Flow Zones During Well Constructipn: API Standard 65—Part 2. 2nd ed. Washington DC: API Publishing Services; 2010 Dec. 96 p. 36. American Petroleum Institute. Overview of Industry Guidance/Best Practices on Hydraulic Fracturing. American Petroleum Conference; 2012 April 18; Traverse City, MI. 37. Associated Press. Leak shuts down drilling operation at well in Northwest Michigan. Mlive [Internet]. 2011 Feb 11. Available from: http://www.mlive.com/news/index.ssf/2011/02/leak_shuts_down_drilling_opera.html 38. U.S. Geological Survey (USGS). USGS Releases Reports on Groundwater-Quality Sampling Near Pavillion, Wyo. [Internet]. USGS. [2012 Sept 26; updated 2012 Oct 4]. Available from: http://www.usgs.gov/newsroom/article.asp?ID=3410&from=rss_home 39. U.S. Environmental Protection Agency (USEPA). Pavillion [Internet]. USEPA. Available from: http://www2.epa.gov/region8/pavillion 40. Enormous Differences between USGS and EPA on Pavillion [Internet]. Independent Petroleum Association of America; c2013. [2012 Oct 16]. Available from: http://www.energyindepth.org/enormous-differences-between-epas-pavillion-data-and-usgss/ 41. Bush B. New USGS tests cast doubt on EPA water quality testing at Pavillion, Wyoming [Internet]. American Petroleum Institute; c2013. Available from: http://www.api.org/news-and-media/news/newsitems/2012/oct-2012/new-usgs-tests-cast-doubt-on-epa-water-quality-testing-at-pavillion.aspx 42. High volume hydraulically fractured well completion active permits and applications [Internet]. Michigan DEQ; 2013 May 5 [cited 2013 May 6]. Available from: http://www.michigan.gov/documents/deq/utica.collingwood_spreadsheet1_358438_7.pdf 43. Michigan Department of Environmental Quality. Downloads and available data. MDEQ; c2013. Available from: http://www.michigan.gov/deq/0,4561,7-135-3311_4111_4231-188295--,00.html 44. Dolton GL, Quinn JC. An Initial Resource Assessment of the Upper Devonian Antrim Shale in the Michigan basin. Denver, CO: USGS; 1996. U.S. Geological Survey Open-File Report 95-75K. 10 p. Available from: http://www.michigan.gov/documents/deq/GIMDL-USGSOFR9575K_303059_7.pdf 45. CO2 EOR—Green Oil [Internet]. Core Energy; LLC. Available from: http://coreenergyholdings.com/CO2EOR-GreenOil.html 46. Curtis JP. Fractured Shale-Gas Systems. American Association of Petroleum Geologists Bulletin 2002; 86:1921-1938. 47. Questions and answers about hydraulic fracturing in Michigan [Internet]. Michigan DEQ. Available from: http://www.michigan.gov/documents/deq/deq-FINAL-frack-QA_384089_7.pdf 48. Tollefson J. Methane leaks erode green credentials of natural gas. Nature [serial online]. 2013 Jan 2;493(12). Available from: http://www.nature.com/news/methane-leaks-erode-green-credentials-of-natural-gas-1.12123. 49. Levi MA. Comment on “Hydrocarbon emissions characterization in the Colorado Front Range: A pilot study” by Gabrielle Pétron et al. J. Geophys. Res. 2012;117(D21203) doi: 10.1029/2012JD017686. 50. Levi MA. Reply to “Reply to ‘Comment on “Hydrocarbon emission characterization in the Colorado Front Range—A Pilot Study”’ by Michael A. Levi” by Gabriella Pétron et al, J. Geophys. Res. 2013;118:3044-3046 doi:10.1002/jgrd.50299. 51. O’Sullivan F, Paltsev S. Shale gas production: potential versus actual greenhouse gas emissions. Environ. Res. Lett. 2012;7(2012):044030. 52. Shires T, Lev-On M. Characterizing Pivotal Sources of Methane Emissions from Natural Gas Production: Summary and Analysis of API and ANGA Survey Responses. Final Report. American Petroleum Institute and America’s Natural Gas Alliance; 2012 Sept 21 (Updated 2012 Sept). 50 p. Available from: http://www.api.org/news-and-media/news/newsitems/2012/oct-2012/~/media/Files/News/2012/12-October/API-ANGA-Survey-Report.pdf 53. Harrison MR, Galloway K, Hendler A, Shires TM, Allen D, Foss M, Thomas J, Spinhirne J. (URS Corporation and University of Texas at Austin). Natural Gas Industry Methane Emission Factor Improvement Study. Draft Final Report. Washington DC: US Environmental Protection Agency; 2011 Dec. Assistance Agreement No. XA-83376101. 53 p.
    • © 2013 BY THE REGENTS OF THE UNIVERSITY OF MICHIGAN MARK J. BERNSTEIN, ANN ARBOR JULIA DONOVAN DARLOW, ANN ARBOR LAURENCE B. DEITCH, BLOOMFIELD HILLS SHAUNA RYDER DIGGS, GROSSE POINTE DENISE ILITCH, BINGHAM FARMS ANDREA FISCHER NEWMAN, ANN ARBOR ANDREW C. RICHNER, GROSSE POINTE PARK KATHERINE E. WHITE, ANN ARBOR MARY SUE COLEMAN, EX OFFICIO Please print sparingly and recycle
    • Geology/ Hydrogeology Technical Report H Y D R A U L I C F R A C T U R I N G I N T H E S T A T E O F M I C H I G A N
    • Participating University of Michigan Units Graham Sustainability Institute Erb Institute for Global Sustainable Enterprise Risk Science Center University of Michigan Energy Institute ABOUT THIS REPORT This document is one of the seven technical reports com- pleted for the Hydraulic Fracturing in Michigan Integrated Assessment conducted by the University of Michigan. During the initial phase of the project, seven faculty-led and student-staffed teams focused on the following topics: Technology, Geology/ Hydrogeology, Environment/Ecology, Human Health, Policy/ Law, Economics, and Public Perceptions. These reports were prepared to provide a solid foundation of information on the topic for decision makers and stakeholders and to help inform the Integrated Assessment, which will focus on the analysis of policy options. The reports were informed by comments from (but do not necessarily reflect the views of) the Integrated Assessment Steering Committee, expert peer reviewers, and numerous public comments. Upon completion of the peer review process, final decisions regarding the content of the reports were deter- mined by the faculty authors in consultation with the peer review editor. These reports should not be characterized or cited as final products of the Integrated Assessment. The reports cover a broad range of topics related to hydraulic fracturing in Michigan. In some cases, the authors determined that a general discussion of oil and gas development is important to provide a framing for a more specific discussion of hydraulic fracturing. The reports address common hydraulic fracturing (HF) as meaning use of hydraulic fracturing methods regardless of well depth, fluid volume, or orientation of the well (whether vertical, directional, or horizontal). HF has been used in thousands of wells throughout Michigan over the past several decades. Most of those wells have been shallower, vertical wells using approximately 50,000 gallons of water; however, some have been deeper and some have been directional or horizontal wells. The reports also address the relatively newer high volume hydraulic fracturing (HVHF) methods typically used in conjunction with directional or horizontal drilling. An HVHF well is defined by the State of Michigan as one that is intended to use a total of more than 100,000 gallons of hydraulic fracturing fluid. The reports indicate if the text is addressing oil and gas development in general, HF, or HVHF. Finally, material in the technical reports should be understood as providing a thorough hazard identification for hydraulic fracturing, and when appropriate, a prioritization according to likelihood of occurrence. The reports do not provide a scientific risk assess- ment for aspects of hydraulic fracturing.
    • GRAHAM SUSTAINABILITY INSTITUTE INTEGRATED ASSESSMENT REPORT SERIES VOLUME II, REPORT 3 HYDRAULIC FRACTURING IN THE STATE OF MICHIGAN Geology Technical Report SEPTEMBER 3, 2013 Faculty Lead BRIAN R. ELLIS ASSISTANT PROFESSOR, DEPARTMENT OF CIVIL AND ENVIRONMENTAL ENGINEERING, MICHIGAN SOCIETY OF FELLOWS TABLE OF CONTENTS 2 Executive Summary 2 1.0 Introduction 11 2.0 Status, Trends & Associated Challenges 19 3.0 Prioritized Directions for Phase 2 21 Literature Cited THIS PUBLICATION IS A RESULT OF WORK SPONSORED BY THE UNIVERSITY OF MICHIGAN Direct questions to grahaminstitute-ia@umich.edu
    • 2 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 EXECUTIVE SUMMARY T his report provides a survey of the Michigan Basin geology by discussing basin hydrogeological charac- teristics within the context of evaluating the impact of high-volume hydraulic fracturing. It also identifies existing knowledge and current practices related to extraction of oil and natural gas. In some instances, these practices are com- pared and contrasted to those of other states in order to provide perspective on the unique characteristics of the Michigan Basin. The goal of this report is to (1) guide the Phase 2 integrated assessment efforts as they pertain to evaluating questions spe- cific to Michigan’s unique geological characteristics and (2) serve as a resource for the non-technical reader who wishes to gain a fundamental understanding of Michigan Basin geology in the context of hydraulic fracturing practices and concerns within the state of Michigan. Although a thorough review of the existing data and literature was undertaken as part of this study, this report is not structured as a pure literature review. It is instead written with the purpose of providing an overview of the current knowledge as it relates to practices associated with hydraulic fracturing and to identify opportunities for additional data collection during the second phase of the integrated assessment. Topics covered In addition to providing a general background to the geology of the Michigan Basin and current unconventional resource plays within the state, this report focuses on several key issues related to evaluating hydraulic fracturing practices within the state of Michigan. These topics include: • Regional fluid transport regimes within the basin • Potential fluid migration pathways • Proximity of hydraulically fractured wells to drinking water resources • Factors controlling chemical composition of flowback waters • Water use and disposal of flowback fluids Summary of prioritized pathways for Phase 2 of integrated assessment Several opportunities for additional data collection and improved technical understanding are identified in this report. These include: • Establish baseline water quality for freshwater resources throughout the state • Evaluate impact of hydraulic fracturing fluids on the release and transport of trace toxic metals and naturally occurring radionu- clides from organic-rich shales • Monitor extent of fracture propagation during hydraulic fracturing • Evaluate potential for induced fluid migration associated with deep well brine disposal In addition to addressing data gaps within the areas identified in the above list, it is suggested that better communication between all involved stakeholders be facilitated in order to best utilize resources and coordinate efforts related to these investigations. This includes improved knowledge sharing among stakeholders and the establishment of a common language for use in discus- sions related to hydraulic fracturing in the state of Michigan, the latter of which will be especially beneficial when communicating with the public on this topic. 1.0 INTRODUCTION G rowth in the extraction of oil and natural gas from unconventional reservoirs has drawn considerable attention to the practice of hydraulic fracturing. The oil and natural gas industry has long used the practice of hydraulically fracturing the rock surrounding a wellbore to enhance hydrocarbon extraction from conventional reservoirs. However, recent advances in directional drilling coupled with high-volume hydraulic fracturing stimulation techniques have made extraction of hydrocarbons from unconventional reservoirs, such as shale formations, economically viable. The rapid application of this tech- nology has drawn the attention of many stakeholders, including local communities in areas where this extraction technique is being applied. The public discourse on the topic of hydraulic fracturing is highly polarizing and may often be driven by incomplete infor- mation, miscommunication or misunderstandings on both sides of the issue1 . In seeking to address the topic of hydraulic fracturing within the limited context of the state of Michigan, the Graham Environmental Sustainability Institute has initiated an integrated assessment of hydraulic fracturing within Michigan. This report is designed to provide a broad discussion of the current knowledge pertaining to the unique geologic characteristics of the state of Michigan within this assessment framework, and to identify exist- ing knowledge gaps that could be explored further during the second phase of the assessment. 1.1 Scope and usage of ‘hydraulic fracturing’ within the context of this report As mentioned above, much of the attention focusing on the use of hydraulic fracturing to extract hydrocarbons from unconventional reservoirs is related the rapid expansion of shale gas drilling in states like Texas and Pennsylvania. Production of hydrocarbons from shale formations falls within the definition of ‘unconventional’ oil and natural gas extraction. This term is applied to hydrocar- bon extraction from formations that do not possess naturally high permeabilities and often require extensive stimulation in order to be productive reservoirs2 . Unconventional reservoirs, such as shale
    • 3 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 formations, may serve as both the source and the reservoir for the hydrocarbons stored within the rocks. Conventional reservoirs are formations with permeabilities sufficient to allow for economic flows of hydrocarbons to a production well without the need of extensive reservoir stimulation and often contain hydrocarbons that migrated from underlying source rocks. Common conventional reservoirs include sandstone and carbonate formations, however, low permeability reservoirs of these same lithologies (e.g. tight carbonates) can also be considered unconventional reservoirs. Currently, there is minimal drilling activity within the state of Michigan that qualifies as high-volume hydraulic fracturing. There are fewer than 60 existing permits or active permit applications for high-volume completions in Michigan3 (and some of these permits are for pilot wells that will not be high-volume completions). The Michigan Department of Environmental Quality (MDEQ) defines high-volume hydraulic fracturing as any hydraulic fracturing com- pletion intended to use more than 100,000 gallons of hydraulic fracturing fluid4 . If this report were to limit its discussion to only these instances it would be very narrow in its scope and discussion of Michigan’s geology within the context of hydraulic fracturing practices in the state. As such, this report will focus on unconven- tional reservoirs in the Michigan Basin where hydraulic fracturing is utilized and not exclusively on high-volume hydraulic fracturing completions. The two primary unconventional reservoirs addressed in this report are the Antrim and the Utica-Collingwood shale for- mations. Hydraulic fracturing is used to enhance gas production from the Antrim, but high-volume completions are not necessary. This is in contrast to the Utica-Collingwood formation, which is currently being developed exclusively via high-volume hydraulic fracturing completions. Production from the A-1 Carbonate is also briefly discussed because of recent high-volume hydraulic fractur- ing well completions within this formation. A recent flurry of min- eral rights acquisitions in 2010 associated with exploratory drilling in the Utica-Collingwood suggests anticipated potential growth in unconventional reservoir production via hydraulic fracturing within the state of Michigan, although only a handful of wells have been drilled in this reservoir since 20105 . Michigan is thus in a unique position to assess the future of high-volume hydraulic fracturing before the gas boom begins and learn from experiences in other states like Pennsylvania. The following sections provide background coverage of the Michigan Basin geology and introduce several unconventional hydrocarbon reservoirs that are examined in this report. 1.2 Michigan Basin geology The state of Michigan sits squarely in the center of the Michigan Basin. The Michigan Basin is a bowl-shaped, intracratonic sedi- mentary basin. The Michigan Basin sits atop crystalline basement rocks of Precambrian age, which lie at depths of approximately 16,000 feet below the surface in the thickest section of the basin near the center of Michigan’s Lower Peninsula and 3,300 feet near the basin margins. The specific deformational history that led to the development of the Michigan Basin is still debated6 . However, geodynamic modeling of basin development by Howell and van der Pluijm7,8 suggests the Michigan Basin developed via several different stress-induced subsidence mechanisms (trough-shaped, regional tilting, narrow basin-centered, and broad basin-centered) occurring over discrete periods of time. These subsidence events coupled with deposition of sedimentary material created the basin as we see it today. Because much of the sedimentary strata formed under a shallow marine environment, the basin strata is dominated by carbonate and evaporite formation lithologies8,9 . Past glaciation events have left the majority of the land surface in the state of Michigan covered by a thin layer of unconsolidated glacial deposits that overlay the sedimentary strata covering the entire Lower Peninsula and much of the southern half of the Upper Peninsula10 . In northern and western sections of the Upper Peninsula igneous rocks outcrop at the surface, providing the cop- per and iron ore that are mined in areas such as the Keweenaw Peninsula and Marquette. A formation outcrop is described as a location at which it is visible at the surface. But in the case of the Michigan Basin where the majority of the land surface is covered in a veneer of glacial sed- iments, the sedimentary formations within the basin are seldom visible at the surface. The location at which a formation would be visible at the surface if not for being covered by unconsolidated sediment is referred to as a formation subcrop. Figure 1(a) provides an overview of the basin geology, showing the characteristic ‘bull- seye’ outcrop/subcrop pattern found in the basin. When viewed in cross-section, as shown in Figure 1(b), it is observed that the sedimentary strata dip toward the center of the state from all direc- tions. This is why the basin is described as being bowl-shaped. Figures such as 1(b) are exaggerated in the vertical direction to clearly show the bowl shape of the basin as the actual formation dip is very gradual, generally being on the order of 1° or less. Due to extensive hydrocarbon exploration and production within the state, Michigan’s geology is well characterized. It is believed that the basin has been tectonically stable since the Jurassic9 . At its thickest point near the center of the Lower Peninsula, the sedimen- tary strata is approximately 16,000 feet thick and is underlain by Precambrian igneous and metamorphic rocks. Common terminol- ogy refers to the underlying crystalline rocks as the basement rocks because they represent the end of the sequence of sedimentary strata.
    • 4 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 KEWEENAW HOUGHTON ONTONAGON BARAGA MARQUETTE GOGEBIC CHIPPEWA LUCE ALGER SCHOOLCRAFT IRON DICKINSON MACKINAC DELTA MENOMINEE EMMET CHEBOYGAN PRESQUE ISLE CHARLEVOIX ALPENA MONTMORENCY LEELANAU OTSEGO ANTRIM GRAND TRAVERSE ALCONAOSCODACRAW FORDKALKASKA BENZIE IOSCOOGEMAWROSCOMMONMANISTEE MISSAUKEEWEXFORD ARENAC MASON GLADWINCLAREOSCEOLALAKE HURON BAY MIDLANDISABELLAOCEANA MECOSTA NEWAYGO TUSCOLA SANILAC SAGINAW GRATIOTMUSKEGON MONTCALM LAPEER KENT GENESEE ST CLAIR OTTAW A SHIAWASSEE CLINTONIONIA MACOMB OAKLAND LIVINGSTONINGHAMEATONBARRYALLEGAN WAYNE WASHTENAWJACKSONCALHOUNKALAMAZOOVAN BUREN BERRIEN MONROE LENAWEEHILLSDALE BRANCHST JOSEPHCASS BEDROCK GEOLOGY OF LOWER PENINSULA SALINA GROUP BASS ISLAND GROUP GARDEN ISLAND FORMATION BOIS BLANC FORMATION MACKINAC BRECCIA SYLVANIA SANDSTONE DETROIT RIVER GROUP DUNDEE LIMESTONE BELL SHALE TRAVERSE GROUP ANTRIM SHALE ELLSWORTH SHALE BEDFORD SHALE BEREA SS & BEDFORD SH SUNBURY SHALE COLDWATER SHALE MARSHALL FORMATION MICHIGAN FORMATION BAYPORT LIMESTONE SAGINAW FORMATION GRAND RIVER FORMATION RED BEDS BEDROCK GEOLOGY OF WESTERN UPPER PENINSULA JACOBSVILLE SANDSTONE FREDA SANDSTONE NONESUCH FORMATION COPPER HARBOR CONGLOMERATE OAK BLUFF FORMATION PORTAGE LAKE VOLCANICS SIEMENS CREEK FORMATION INTRUSIVE QUINNESEC FORMATION PAINT RIVER GROUP RIVERTON IRON FORMATION BIJIKI IRON FORMATION NEGAUNEE IRON FORMATION IRONWOOD IRON FORMATION DUNN CREEK FORMATION BADWATER GREENSTONE MICHIGAMME FORMATION GOODRICH QUARTZITE HEMLOCK FORMATION MENOMINEE & CHOCOLAY GROUPS EMPEROR VULCANIC COMPLEX SIAMO SLATE & AJIBIK QUARTZITE PALMS FORMATION CHOCOLAY GROUP RANDVILLE DOLOMITE ARCHEAN ULTRAMAFIC ARCHEAN GRANITE & GNEISSIC ARCHEAN VOL. & SEDIMENTARY MACKINAC BRECCIA BEDROCK GEOLOGY OF EASTERN UPPER PENINSULA MUNISING FORMATION TREMPEALEAU FORMATION PRAIRIE DU CHIEN GROUP BLACK RIVER GROUP TRENTON GROUP COLLINGWOOD SHALE MEMBER UTICA SHALE MEMBER STONINGTON FORMATION BIG HILL DOLOMITE QUEENSTON SHALE MANITOULIN DOLOMITE CABOT HEAD SHALE BURNT BLUFF GROUP MANISTIQUE GROUP ENGADINE GROUP POINT AUX CHENES SHALE SAINT IGNACE DOLOMITE SALINA GROUP BASS ISLAND GROUP GARDEN ISLAND FORMATION BOIS BLANC FORMATION MACKINAC BRECCIA 0 20 40 MilesDate: 11/12/99 N Michigan MICHIGAN DEPARTMENT O FNATURAL RESOU RCES LAND AND MINERALS SERVICES DIVISION RESOURCE MAPPING AND AERIAL PHO TOG RAPHY Michigan Resource Information System Part 609, Resource Inventory, of the NaturalResources and Environmental Protection Act, 1994 PA 451, as amended. Automated from "Bedrock Geology of Mi chi gan," 1987, 1:500,000 scale, which was compiled from a variety of sources by the Michigan Department of Environmental Quality, Geological Survey Division. SOURCE RMAP 1987 BEDROCK GEOLOGY OF MICHIGAN Figure 1a 1987 Bedrock Geology of Michigan Figure 1b Figure 1: (a) Michigan Basin outcrop geology; (b) Cross- section (from the northwest to the southeast) of the Michigan Lower Peninsula demonstrating the bowl- shaped basin stratigraphy (vertical exaggeration). From MDEQ.
    • 5 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 1.2.1 Fundamental hydrogeology concepts and terminology Sedimentary rocks can be either aquifers or aquitards. A simple way to think about aquifers versus aquitards is that an aquifer is a permeable geologic formation from which water can be extracted in good quantity, whereas an aquitard is a low permeability geo- logic formation with insufficient permeability to allow for signifi- cant water production. Aquicludes have essentially no flow of water whatsoever and can be thought of as the extreme end of impermeable aquitards. Examples of typical aquifer lithologies are sandstone and limestone formations, while shales are common aquitards. It is important to note that not all aquifers are freshwater aquifers. In the Michigan Basin the vast majority of water contained within the sedimentary strata is saline water not suitable for drink- ing. Potable drinking water is found at shallower depths in the glacial aquifers and near the recharge zones of some sedimentary formations. Porosity is the percent void space within a unit volume of rock. Two good examples of porosity in common objects are (1) the voids within a sponge and (2) the void space within a jar filled with marbles. In both cases, there is a unit of volume delineated by the outer edges of the object (sponge dimensions, total jar volume). Within this unit volume is space occupied by both solid material (sponge, marbles) and air. The relative volume of air to the total volume, represented mathematically as (air[m3 ]/VolumeTOT [m3 ]), is the porosity of the system (Φ). It is therefore unitless and is given as a percentage. Porosity alone will not dictate whether a given formation will be an aquifer or an aquitard, as the key determinant here is the formation permeability. Permeability is a measure of the ease with which a substance moves through a porous medium and reflects the degree of connectedness of the pores within geologic media. 1.2.2 Michigan Basin stratigraphy The lowest sedimentary unit in the Michigan Basin is the Mt. Simon sandstone, which is a regional saline aquifer. Due to its high poros- ity and permeability it has been used as a storage formation for waste injection and is considered to be a target storage formation for the underground sequestration of CO2 11 . Overlying the Mt. Simon is the Eau Claire confining unit, which is part of the Munising Group that consists of a series of interbedded sandstones, dolo- mite, and mudstones12,13 . Above the Munising Group is the early Ordovician Prairie Du Chien Group consisting primarily of dolo- mite and sandstone with some minor shale lenses12 . The St. Peter sandstone is a saline aquifer that sits on top of the Prairie Du Chien Group strata14 . The middle Ordovician age Trenton and Black River carbonate formations overlie the Glenwood shale immediately above the St. Peter sandstone. The Trenton is overlain by the Utica shale throughout the state and by the organic-rich Collingwood limestone in portions of the north- ern Lower Peninsula. The Utica-Collingwood formation is recog- nized as major source rocks for hydrocarbons within the Michigan Basin13 . A thick sequence of Silurian-age strata lies unconformably above the Utica beginning with the Manitoulin dolomite and the Cabot Head shale. Above the Cabot Head shale lie the Burnt Bluff Group and the Manistique Group carbonates. The Niagaran Group is found above the Manistique Group and is composed predominantly of micritic limestone and dolostone. This group is a major hydrocarbon reservoir in the state of Michigan and is gener- ally sub-divided by the oil and gas industry into three units (Brown (top), Gray and White (bottom)) based on color, texture, and faunal properties12 . The Salina Group is found above the Niagaran and is composed of a thick sequence of evaporite (e.g. halite, anhydrite and gypsum) and mixed carbonates. The evaporites are basin-cen- tered, thinning toward the outer margins of the Salina Group where they are no longer present. Above the Salina Group is the Bass Islands Group, a dolomitic saline aquifer, which is overlain by the regionally confining units of the Bois Blanc (cherty carbonate) and Detroit River Group. The Detroit River Group was formed during the middle Devonian and is composed of a mixed series of carbonates, evaporites, and sandstones. The Dundee and Rogers City limestone formations sit atop of the Detroit River Group. The Dundee is a hydrocarbon-bearing reservoir in the central portion of the Michigan Basin and is often targeted for wastewater injection along its periphery. The Traverse Group is another regional saline aquifer dominated by a vuggy limestone lithology. The Bell shale sits at the base of the Traverse group and serves as a confining unit separating the Traverse and Dundee Groups15 . The organic-rich Antrim shale is above the Traverse Group and serves as another primary source rock for oil and natural gas within the Michigan Basin13 . A series of confining shale units lies atop of the Antrim. Moving up the strata we find the Ellsworth, which represents of the end of Devonian strata, followed by the Mississippian-aged Sunbury, and Coldwater shales. In eastern por- tions of the basin the Berea sandstone, a regional saline aquifer, lies between the Antrim and Coldwater shales. Another sandstone aquifer, the Marshall sandstone lies above the Coldwater shale. The Michigan formation, which is considered a confining unit14 , consists of a mix of dolomite, shale, and evaporites12 . The final units in the Michigan Basin stratigraphy are the Saginaw formation, a regional aquifer, and the Ionia formation, a regional confining unit14 . This extensive list presents a comprehensive representation of the Michigan Basin stratigraphy, but for the sake of discussing basin hydrology many of these individual units are lumped together into what can be considered general aquifer and aquitard units, as will be discussed later in this report14 .
    • 6 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 1.3 Oil and gas drilling in Michigan Hydraulic fracturing of oil and natural gas wells is common practice in Michigan. The MDEQ reports that more than 75% of recent wells drilled in Michigan have been hydraulically fractured and estimates that a total of 12,000 wells have been hydraulically fractured to date within the state of Michigan16 . Given that the focus of the broader integrated assessment effort is on high-volume hydraulic fractur- ing, for the purposes of this report it is important to draw a distinc- tion between the process of hydraulic fracturing, in general, and high-volume hydraulic fracturing, in particular. Most of the wells in Michigan that have been hydraulically fractured do not qualify as high-volume completions (i.e. they have used <100,000 gallons of hydraulic fracturing fluid). The technique of hydraulic fracturing has been used since the 1940’s by the oil and gas industry to enhance the production of wells17 . High-volume slickwater hydraulic fractur- ing is a relatively new practice that often combines directional drill- ing and hydraulic fracturing at larger scales, thus requiring larger volumes of water for well completions. The term ‘slickwater’ refers to the hydraulic fracturing fluid mixture, which includes friction reducers in addition to other chemical additives designed to opti- mize the hydraulic stimulation process. Directional drilling refers to the process where the direction of the drill bit is gradually deviated from the vertical, allowing the well to follow a formation in the hori- zontal direction. In the case of shale-gas wells, such drilling may be coupled with hydraulic fracturing to provide increased access to the stored natural gas within the impermeable rock matrix. To be clear, not all horizontal wells are high-volume completions, and not all high-volume completions include horizontal drilling. It is no surprise that organic-rich shales hold economic quanti- ties of oil and natural gas, as these same formations serve as the source rocks for the hydrocarbons found in more conventional res- ervoirs (e.g. sandstones). However, the low permeability of these formations prevented economic production via traditional drilling and well completion methods2 . Within the past decade the interest in and exploration of unconventional reservoirs has increased dra- matically due to successful applications of high-volume hydraulic fracturing in shale reservoirs. The increased application of direc- tional drilling coupled with hydraulic fracturing and its impact on productivity of unconventional gas production is evident when examining Figure 2, which shows the production of natural gas from the Barnett shale in Texas from 1997–2010 as a function of the type of well drilled (vertical vs. horizontal). This horizontal drilling and hydraulic fracturing activity also led to increased water consumption. Between 2009 and June 2011 the median amount of water used to hydraulically stimulate horizontal wells in the Barnett shale was 2.8 million gallons per well18 . The growth in U.S. domestic shale-gas reserves has led the U.S. Energy Information Administration to predict that natural gas extracted from shale formations will total 49% of all domestic natural gas production by 203519 . Because extraction of natural gas from shales has experienced such rapid growth, it seems appropriate to provide some back- ground on shale lithology and hydrocarbon generation. Shale is a fine-grained sedimentary rock rich in clay minerals and often contains elevated concentrations of organic matter. Plant and animal organic matter can be deposited within sedimentary rocks, and when this occurs rapidly in anoxic marine environments, the organic material may be preserved within the rock. Overtime, further sedimentation and burial may expose these sediments to higher temperatures and pressures found deeper within the Earth’s crust. Thermal breakdown of this organic material can lead to the formation of kerogen, which is the primary ingredient in the generation of hydrocarbons. Under the right temperature and pressure conditions, kerogen may be further transformed into nat- ural gas, oil or CO2 . Organic-rich shales serve as the source rocks for traditional oil and gas reservoirs when the oil or gas is able to migrate beyond the shale and accumulate under an overlying stratigraphic or structural trap. However, due to the inherently impermeable nature of shale formations these products are often trapped within the shale and until recently, were not considered to be economically recoverable resources21 . It is thus only due to the recent advances in directional drilling and slickwater chemical formulations that unconventional extraction of natural gas from organic-rich shale formations has become economically viable via high-volume hydraulic fracturing completions. In deep shale gas reservoirs, natural gas exists primarily as free gas within the shale porosity (both interstitial and fracture porosity). Additional gas may also be adsorbed to the surfaces of clay min- erals and organic matter. The fracture networks created during the hydraulic fracturing process allow the trapped gas to flow to the Figure 2: Annual Barnett shale natural gas production by well type. Figure from U.S. Energy Information Administration20 .
    • 7 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 well. The fractures also generate increased surface area pathways within the formation allowing for enhanced gas desorption. 1.4 Unconventional hydrocarbon reservoirs in the Michigan Basin The following section presents background information and characterization of several unconventional reservoirs within the Michigan Basin. The order of presentation will begin with a dis- cussion of the most developed natural gas play, the Antrim shale. This will be followed by discussion of the Utica-Collingwood shale, which is a developing target reservoir where high-volume hydraulic fracturing has been utilized in recent drilling efforts. Lastly, a brief discussion of the A-1 Carbonate formation will be presented given that there is also interest in high-volume hydraulic fracturing within this formation3 . Although not discussed in detail in this report, the Trenton-Black River carbonate formation has also been a target for high-volume hydraulic fracturing completions. The Trenton-Black River would be considered a conventional reservoir with the mature Albion-Scipio field having been actively produced for nearly 50 years. Although uncommon, high-volume completions in the newer Napoleon field may be used to overcome inconsistent and highly variable reservoir quality22 . Wells may be in the vicinity of good reservoir rock that could be accessed more easily through well stimulation as opposed to drilling a second well. These completions tend to be just above the high-volume threshold of 100,000 gallons of hydraulic fracturing fluid. In com- piling this report it was decided to limit discussion to unconventional reservoirs, that is, those reservoirs that can be consid- ered ‘continuous plays’ and are especially amenable to high-volume hydraulic fractur- ing completions. Data on formation total organic content (TOC), which serves as a proxy for potential hydrocarbon content, and lithology charac- teristics were collected from the Michigan Geologic Repository for Research and Education in Kalamazoo, Michigan. 1.4.1 Antrim shale The Antrim shale was one of the first economic shale-gas plays in the U.S. and has been actively developed since the 1980’s. It is a naturally jointed formation with significant gas production occurring at relatively shallow depths (~1,000–2,000 feet below the surface)23,24 . These two aspects have allowed for economic produc- tion of the natural gas in place through use of traditional extraction methods (i.e. vertical wells). The Antrim is somewhat unique among shale-gas reservoirs in that it produces significant volumes of water25 . This is due to its highly naturally fractured nature26 and proximity to subcrop recharge, which consequently has also led to biogenic methane production in much of the active play near the northern formation subcrop region27,28 . The majority of Antrim natural gas production wells are located in what is referred to as the Northern Producing Trend centered near Otsego County and OIL AND GAS WELLS SOUTHERN PENINSULA OF MICHIGAN ± 0 10 20 30 405 Miles ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! !! ! !! !!! !! !! ! ! ! !! ! ! ! !! ! ! !! !!!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! !! ! ! !! ! ! !! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! ! !! ! !! ! !! !! !! ! ! ! ! ! ! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !!! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !!! !! !! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !! !! !! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !!! ! !! !! !! !! ! ! ! ! ! ! ! !! !! !!! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! !!!!! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! !! ! !! ! ! ! !! ! ! !! ! !!!! ! ! !!! !!! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! !!!! !! !! ! !!!! ! ! ! !! !! !! ! !! !!! !! ! ! ! ! ! ! ! !! !! ! ! !! ! !!! ! ! ! ! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! !! ! ! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! !! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !!! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! !!! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !! ! !! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !!! ! ! !!!! ! ! ! !!!!!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! !! !!! ! !! ! ! !!! ! !! ! !!! !! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! !!! ! ! !! ! ! ! !! ! !! !! !! !!! !! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !!! ! ! ! !!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !!!! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! !! !!! !! ! ! !!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! !! !! !! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !!!! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! !!! ! !! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! !! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !!!! ! ! !!! ! !!! ! !!! ! !! ! !! !!! !! !!!!! ! ! !! ! ! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! ! ! ! ! !! !! ! ! !!!! ! ! !! ! !! ! ! ! ! ! !! ! !!!! ! !! !!! !! ! ! !!!!!! ! ! ! ! !! !!! ! !!!! !! !! ! ! !! ! !!! ! ! !! ! ! ! ! ! !!!! ! ! ! ! !!! !!! !!!! !!! ! ! !! !! ! !!! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! !! !! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!! !! ! !! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!!!! ! ! !! ! ! ! ! !!! ! ! ! !! ! !! !! ! ! ! !! !!! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!!!!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! !!!!! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! ! !! ! ! ! ! !!! !!!! ! !! ! ! !! !!! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! ! ! !! !!! !!! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !!! !! ! ! ! !! !!!! !! !! !! !!!! ! ! ! !!! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! !! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! ! ! ! !!! ! !!! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!!! ! ! ! ! ! !!! !! ! ! !! ! ! !! !! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !!! ! !! !! ! !! ! ! !! ! !!! ! !! ! !! ! !! !! ! ! ! ! ! ! ! ! !!! !!! ! ! !! ! ! !!!!! !! !! !! !! ! ! ! !! ! ! ! !! !!! !! ! !! ! !! !! ! !! ! ! ! ! ! ! ! !! ! !!! !! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !!! !!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !!! !! !! ! ! ! ! !!! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !! !! !! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! !!!!!! ! ! ! ! !! ! ! ! ! !!! ! ! !!! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !! ! !! !!! ! ! ! ! !! ! !! !! !!! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!!!! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! !! ! !! !! ! ! ! ! ! !!! !!! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! !! !! !! !! ! ! ! ! !! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !!! !! ! ! ! ! ! ! !!! !!! ! ! ! ! !! ! ! ! !! ! !!! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !! ! !! ! !!! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! !!!!! !! ! ! ! ! !! !! !! ! ! ! ! ! !!! ! ! ! !!!! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! ! ! !! ! !! !! !! !! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !!!!! ! ! ! ! ! !! !!! ! !! !! ! !!!!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! !!!! ! ! !! ! ! ! !! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! !! !! ! ! !!!! ! ! ! ! ! !! !! !! ! ! ! ! ! !! !! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! !!! ! ! ! ! ! !! !! ! ! !! ! ! !! ! ! !! ! !!!! ! ! !! ! !! ! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !!!! !! ! !! ! ! ! ! ! ! ! !!!! ! !! !!!!! ! ! ! ! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! ! ! !!! !! !!! !!! ! !! ! ! ! ! ! ! !! !!!! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! !! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! !! ! !!!! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! !! ! ! ! ! !! !! !! ! !! ! !! ! ! ! !!! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !!! !! ! !! ! !!! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! ! ! ! !! ! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! !! ! ! ! !! ! !! ! ! !!! ! !!! ! ! ! !!! ! ! ! !! ! !! ! ! ! !! ! !!!! !! ! !! !! ! ! !! !!! ! ! ! ! ! !! ! !! !! ! !! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !!! ! !!!! ! ! ! !! ! ! ! !! !!! ! ! ! !! !! !!! ! !!! ! ! ! ! ! !!! !!! ! ! ! ! ! !! ! !! ! !! ! !! ! ! !! ! !! ! !! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! !! ! ! ! ! !!! !!! !! ! ! ! ! !! ! !! !! ! !!!!! ! ! !! !! ! !! ! !! ! ! ! !! !! !!! ! ! ! ! ! ! ! !!! ! ! ! !! ! !!!! ! ! ! !! ! !!! ! !! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !!! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! !! ! ! ! ! !!! !! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! !!! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !! ! !! ! !! !! ! !! !! ! !! ! ! !! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! !!! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! ! ! !! ! !! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! !!! ! ! !!!!! ! ! ! !! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! !! ! !!! !!! ! !!!!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!!!! ! ! ! !! ! ! !!!! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! ! !! !!! !! !! ! !! !! ! !! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !!! ! !!!!! ! !!! ! !! ! ! ! ! !!! !! !! ! ! ! !! ! ! ! ! ! !!!!! ! !!! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! !!! !! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! !! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!! !! !! ! ! ! ! !!!!!! ! ! ! ! !!!!!!! ! ! !!! ! ! ! !!! ! ! ! !!! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! !! ! !! ! ! ! !! ! !!!!! !! ! ! !! ! !! !! ! ! !! ! ! !!! !! !! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!!!!!! !!!!!!! ! ! ! ! ! ! !! ! ! !!!!!!!!! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! !!!!! ! !! ! ! ! ! ! !! ! ! !!! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!!!!!! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! !! ! ! !! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !! ! ! !! !! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! !!! ! ! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !!! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! !! ! ! !! ! !! ! !! !! ! ! !!! !!! ! ! ! !!! !! ! !! ! ! !! ! ! ! ! !! !! !! ! ! ! ! ! !! !! !! ! !! ! !! !! ! !!! ! !! !! !! ! ! ! ! ! !! !! ! ! !!!!!! ! ! !! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! !! !! !!!!! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! !!! !!! ! !!! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !!! ! ! !! !! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!!! ! !! ! !!!! !! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! !! ! ! ! ! !! ! ! !!!! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!!! !!!! ! ! !!!! ! ! ! !!! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !!!! ! ! ! ! ! !!! ! ! ! !! !! !!!! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !! ! ! ! !!!! ! ! ! ! ! ! ! ! !!!!!!! !! !! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! !! !! ! ! ! !! ! !! ! !! !! !! !! !! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! !!!! !! ! ! ! !!!!! ! !! ! !!!! !! !! !! ! !! ! !!!! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !!! ! !!!! ! !!! !! ! ! ! !!! !! !! ! ! ! !! ! !! ! ! !! ! ! ! ! !!!! !!!!! !! !!! ! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! ! !! !!! ! !! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! ! ! !! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! !! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! !!! !!!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !!! ! !! !! ! ! !!!! !!! ! !! !!! ! !!! !!!!! ! ! ! !! ! !! ! ! ! !!! ! !! ! ! ! ! !!!! !! ! !! ! !! ! !! ! !!! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! !!! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! !! !! !! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !!!! !! !! !!!! !!!!!!!!! !!!!! ! ! !!!! !! !! !!!! !! ! !!!!!! !! ! ! !!!! !!!! !!!!! !!! ! ! !! ! !!!!!!!!!!!!!!!!! !!!! ! ! !!!!! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! !!! !! !! ! ! ! ! ! !!!!!!!! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !!! !! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! !! !! !! ! !! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! !! !! !!!!!! !! !! ! !! !!!! !!!!!!!!!! ! ! !! !!! !! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! !! !!!!!! !! !! !! ! ! !! ! ! ! !! ! !!! ! !! ! !! ! !! !!! !!! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! !!! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! ! ! !!!! ! ! ! !! ! !!!!! ! ! !! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !!!!! ! !! ! ! ! !! ! ! !! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !!!! !! ! !! ! !!!!! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! !! !!! ! !! !! !!!!!! ! !! ! ! !!! !!!!! ! ! ! ! ! !!! !!! ! ! ! !! !! ! !!!! ! !! !! ! !! ! !! ! ! !!!!!!! !!! !! !! !! ! ! !! !!! !! !! !!! ! !! ! !! ! !! ! ! ! !! !!! !! ! !!!! ! !! !!! ! !! ! ! !! ! ! !!! ! ! ! ! ! ! !!! ! !! ! !! ! ! !!! !!!! !!! !!! !!!! !! !! !!! !! !!!! ! ! !! ! ! ! ! ! ! !!! ! ! !!! ! !! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !!!!!!!! ! ! ! ! !! ! ! !! !!!!! !! ! !!! !!! ! ! ! ! ! !!!! !! !! ! ! ! !! ! !!! !!! ! !! ! !! ! !!! !!! !!! ! ! ! ! ! ! ! !! !! ! ! !!! ! ! !! ! !! !! ! !! !!! !! ! ! ! ! !! !!!!!! !! ! ! !!! ! ! !! ! ! ! !!! ! ! ! ! ! !!! !!!! ! !!!!! ! ! !! ! !! !! ! !! ! ! ! !!! !! !!!! ! ! !!!! ! ! ! ! ! !! !!!!!! !!!! ! !!! !!!! ! !! ! !!! ! ! ! ! ! !! !!! ! ! !! !! ! ! ! ! !!!! !!! ! !!! !!! !! !!!!!!!!!!!! !!! ! !!!!!! !! ! ! ! !!! ! !!!! ! ! ! ! !!! ! ! ! !!! !!! ! ! !!!!!!!!! !! !!!!!!!! ! !!!!! !!! ! !!!! ! ! ! ! ! !! ! !!! ! !! ! !!! !!! ! ! ! ! !! !! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! !! !! ! ! ! !!! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! !! !!! ! !! ! !! !! ! ! ! ! !!! !! !! !! !!! ! ! ! !! ! !!!! !!!!!!! ! !! ! ! ! !! ! ! ! ! ! ! !!!! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! !! !!!!! !!!! !! !! !! ! ! ! ! ! ! !! !! !! !! ! ! !!!!! ! ! !!!! ! !!! ! !!! !! !! ! ! !!!!! ! ! ! ! !! !! !!!! ! ! !!!! ! ! !!! !!! ! !!! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! !!!!! !! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! !!! !!! ! ! ! !! ! ! ! !! !! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!!!! !! ! ! ! ! ! !! !! ! !!! !! !!! !! !! ! ! ! ! !! !! !!! !! !!!!! ! ! ! ! ! ! !! !!! ! ! ! !! !!!! ! ! !!!! !!! !!! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !!! !! ! ! !!! ! ! ! !! !! !! ! ! !!! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !!! !!!! !! ! ! ! ! ! ! ! !! ! ! ! ! !! !!!! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! ! !!!! ! ! ! ! !! !! ! ! !! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! !! ! !! ! !! ! !! !!! ! ! !! !!!! ! ! ! !! ! ! ! ! !!! ! ! ! ! !!! !!!!!! ! ! ! !! ! !! ! !!! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!! !! ! ! ! ! !! !! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! !!! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! !! !! !! ! !!! ! ! ! ! !! ! !! ! !! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!! !! ! ! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !!! ! !!!!!! !!! ! ! !!! ! ! !! !! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!!!! ! ! !! ! ! ! !! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! !!!! ! !!! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!! ! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! !! !!! !!!!! ! !! !!!!! !!! ! !!!! ! !! ! !! ! !!! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !! !! !! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! !!! !!!! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! !! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! !!! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! !!!! !! ! !!! ! ! ! ! !! !! !! ! ! !! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !!! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! !!!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! ! !!!! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! !!!!!!!!! !!!! !! ! !! !!! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! !! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! !! !! !!! !! !! !! ! ! ! ! !!! ! ! ! ! ! !!! ! !! ! ! !!! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! !! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! !! !! ! !! ! ! ! ! !!!! ! ! !! !!! ! ! ! ! ! !!! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! !! !! ! !! !! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! !! ! !! !!! ! !! !! ! !! ! ! ! ! !! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! !!!!!! ! ! !!! ! !!! !! !! ! ! !!!! !! ! ! !!! !!! !! ! ! ! !! !! !!! ! ! ! ! ! !!! ! ! ! ! ! !! !! !! ! !!! ! ! !!! ! ! ! ! ! !!!! ! ! ! !! ! !!!! ! ! ! ! !! ! !! !! ! !! ! ! !! ! ! !!! ! ! ! ! ! !! !! !! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!!!! !! !! ! ! ! ! ! !! !!! !!! ! !!! !! !! !! !! !! ! ! ! ! ! !! ! !! ! !! !! !!!!! !!!!! !! !!!!!!!!!!! !! !!!!! !!!! ! ! !! !! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! !! !! !! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! !! ! !! ! ! !! !! ! ! ! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! !!! !! ! ! ! ! ! ! ! !! !!!! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! !!!!! ! !!! ! !! !! ! !!! ! !!! ! !!! !! !! ! !!! !!! ! ! ! !!!! ! ! ! ! ! ! ! !!!!!!!!!!!!!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !!! ! !! !!!!!!!!!! ! ! ! ! !!!!!!! ! ! !!!!!!!!! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!!!! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! !!!! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! !! ! !! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! !! !! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! !! ! !!!! ! !!!! ! !!! ! ! ! !! ! ! !!!! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!!!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! !! !! ! ! !!! ! ! ! ! ! !! !! ! ! ! !!!! ! ! !!! ! ! ! !! ! ! !! ! !! !! ! ! ! !! !!! !! ! !!! !! !!! ! ! ! ! ! ! ! !!! !!!! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !!! ! ! ! !! !!!!! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! !! ! !! ! ! !! !!!! ! !!! !! !!!!! !! ! !! !!! !!!! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! !! !! ! !!! ! ! ! !! ! !!!! ! !!! !! !! ! ! !! !! ! ! !! ! ! !! !!! !! !! ! ! ! ! !! !! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !!! ! ! ! !!! ! !! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! !!! !! !! !! !! ! !!!! ! ! !!! !!!!! ! ! !!!! ! !!!! !!!!!!! !!! ! !! ! ! ! ! !! ! ! !! ! !! ! !!! ! ! ! !!!!! ! ! !!!! !!!! ! ! !! ! ! !!!!! ! !! ! !!! ! ! ! !! ! ! !! ! !!!!! ! !! !!! ! ! ! !! !! ! ! ! ! ! !! !!!! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! ! ! ! ! ! !! !!!! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! !!!!!!! ! ! ! ! ! !! ! !! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !! !!! ! ! ! ! ! ! !! !! !!!!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !!! ! ! ! !! !!! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !!! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! !! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!! ! ! ! !! ! ! !! ! ! ! ! ! !!! ! ! !!! ! ! ! ! ! !!! ! !! ! ! !! !!!! ! ! ! !! ! ! !!! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!!!!! ! ! !!!!! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! !! ! !! ! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! !! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! !!! !! ! !! !! !! ! !!! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !!!!!! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! !! !! ! ! ! ! !! ! ! ! ! ! !! !! !! ! ! ! ! ! !!! ! !! ! !!! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !!!! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! !! !!!! !!! ! ! ! !! !! !! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! !!! !! !! ! ! ! ! !! ! !! ! ! !!! !!! !! ! !!! !! !!!! ! ! !!!! !!! !!!! ! !! ! ! ! ! !! ! ! ! !!!!! ! !! !! !! !! !! ! ! ! !!!!!!!! !! !! !!!!!!!!!! !!! !!!! !!!!!! !!! !!!!! ! ! ! !!! ! ! !! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !!! ! ! ! !!!! ! ! !!!!!!! ! ! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! !!!! ! !! !! !!! ! !! ! ! ! ! ! ! !!!! ! !! ! !!! ! ! !!!!!! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! !!! !! !! ! ! !!! !!!! !! !! !! !!! ! ! !!! !! !! ! ! !! !! !!! !!!! ! !! ! ! ! !! ! !! !! !! !! !!!!!! ! ! !!! ! !! ! !!!!! ! ! !! ! ! ! !!!! ! ! ! ! !!!!!!!!!!! !! !! ! ! !! ! !! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! !!!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! !!! !!! ! !! !!! !!! ! ! !! ! ! ! !! ! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! ! !! !! !!!! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! !!!! ! ! !! ! ! ! ! !!! !!!!! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! !!!! ! !! !!! ! ! !! !! !! ! ! ! ! ! ! !! !!!!! ! !!!! ! ! ! !! !!!!!!! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! !! ! !! ! !! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! ! ! ! !!!!! !! ! ! ! !!! ! ! ! !! !! ! !! !!!!!! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! !!!! !! !! ! ! !! ! ! !! !!! ! ! ! !!! !!!! !!! !! ! !! ! ! ! !! ! ! !!! ! ! ! ! !!! ! ! ! ! !!! !! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! !!!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! !!! !! !! !! !! !!! ! !! ! ! ! ! !! !! ! ! ! !! !! !!! ! ! !! !! ! !!!! ! !! !!!!!!! ! ! ! !! ! !! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! !!!!! !! ! !! ! ! !!! ! ! !!!! !! ! ! ! !!!! !!! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !!!! ! ! !!! !! !! ! ! !! ! !!!!!! !! ! ! !!! ! ! !! ! !! !!! !!!! ! !!!!! ! !! !! !!!! !!! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! !! !! ! ! !! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! ! ! !!!!!!!!!!!!! ! !!! ! !! ! !! ! !! !! ! !!!! ! !! ! !! ! ! ! !! ! ! ! !! !! !!!! !! !! ! ! ! ! ! ! !! !! !!! !! ! ! !! !!! !!! !!! ! !! !! ! !!! !! ! !! ! ! ! ! !!! !!! ! ! !!!!! !!!! ! !!! ! !!! !! !! !!!!!! ! ! !! ! !!! !!!! ! ! !! ! !!!!! !! !!! !!! !! ! !! ! ! ! !!!!! ! ! ! ! !! ! ! !! ! ! !! !!! !! !! ! !! ! ! !! !!! ! ! ! !!!! ! ! !! ! !! ! ! !!! !! !! ! ! ! !! ! ! ! !! !!!!!!!!!! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! !! !! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! !! !! !!!! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! !! ! !!! ! !!!!! !!!! !! !!!! !! ! !!!! !! !! ! ! ! ! !! !! ! !!! ! ! ! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! !!! ! ! !! ! !! ! ! !! ! ! ! ! !! !!!! ! ! ! !! !! ! !! !!! !! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !!!!!!!! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! !!!!! !!!! ! ! ! !! !! ! ! ! ! ! ! ! ! !! !!!!! ! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! ! ! ! ! ! ! !!!! !!!! !! !! ! !! !!! ! !! !! ! ! ! !! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! !!!! ! ! ! !!!! ! !! !! ! ! ! ! !!!! ! ! ! !! !!! ! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! !! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !!!!! ! !! ! ! !!! ! ! ! !!! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !!! !!!! !! !! ! !! ! ! !! !! ! !!! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! !! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! ! !!! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! !! ! ! !! !!! !!!! ! ! !! !! !! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!! ! !!!!!!!!!! !! ! !!!! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!!! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! !! !! ! ! ! ! !! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! !! ! ! !! !! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !!! !! !!!!!! ! ! ! ! !!!! !! ! ! !! !! ! ! !!!!! !! !!!!!!!!!! ! ! ! ! !!! ! ! !!! !! ! !!!!!! ! ! ! !!! !! ! !!! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! !!!! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !! !! ! !! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !!!!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!!!!!!! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !!! !!!! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !!!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! !!! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! !!! ! ! !! !! !! !! ! ! !!!!!! ! ! ! ! !!! ! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! !! !! ! !!!!! !! ! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! ! !! !!!!!!! ! ! ! ! !! ! !! !! !! ! ! !!!!!! ! ! ! !! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! !! !! !!!!! ! !! !! !! ! !! ! !!!! ! !! ! !! ! ! ! !! !! ! ! ! ! !!!!!!!! !!! ! ! !!! !! ! !!!! ! !!!!! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !!!! !!! !! !! ! ! !! ! ! ! !!! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! !!!! !!! ! !!! ! ! ! ! ! !! !!!!!! ! !! !!!!! !! !!!! ! !!!!!!! !! !! ! !!! ! ! !!! ! !! !!! !!!! !! ! !! ! ! !!! ! ! ! !! !! ! ! !! ! !! ! !! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !! !!!! ! ! ! !! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!!! ! ! !! ! ! ! !!!! ! ! !! ! ! ! !! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! !!! !!! !! ! ! ! ! ! !! !!! ! ! ! ! !!! ! ! !! !! !!! ! ! ! ! ! ! ! !!!!! !! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! !!!! ! !!!!!!!!!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! !!!!!!!!! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!!! ! !!!!!!! ! ! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! !!!!!! ! ! ! ! !!! ! ! !! !! ! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! !!! ! !! ! ! ! !! ! !!!!!!!! !! !! ! ! ! ! !! ! !!! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! !! ! !!! ! !!!!!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! !! ! !! ! !! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! ! ! !!! ! ! ! ! !! ! !! ! !! !!!!!! ! ! !! ! !!! ! !! ! ! !! ! ! !! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! !!!!! !! !! ! !!! !! ! ! ! ! !! ! !! !!!!!! !!! !!!! !!!! ! ! !! ! ! !! ! ! !! ! ! !! ! !! !!! !! !!! ! ! !! !! ! !! ! ! ! ! !! !! ! !! !! ! !!!!!!!!!! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!!! ! ! ! ! ! !! !! !!! !!! ! !! ! !!!!!!!!!! !!!!! !!!! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !!! !! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !!!!! ! ! !! ! !! ! ! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !!!!! !!! ! ! !! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! !! !! ! !! !!!!! !!!! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! !!!!!! ! ! ! ! !!! !!!! !!!!!! ! ! ! ! !! ! ! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! ! ! !!!! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! !! !! ! !!!! ! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!! ! !!!!! !! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !! ! ! ! !! !! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !!!! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! ! ! !!! !! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! !! !! ! !! ! !! ! ! ! ! !! ! !! ! !!! ! ! !! !! !! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!!!!! ! ! !!! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! !! ! ! ! ! ! ! !! !!!! !! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! !! ! !! !! ! !!!! !!!!! !!! ! !!! !! ! !!! ! !!! ! ! !!!! ! ! ! !!! ! ! !! ! ! ! ! !! !!!!!! ! ! !!!! ! ! ! ! !!! !! ! !! ! ! !!!!!!! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !! !!!! ! !!! !! !!! ! ! ! !! ! ! !! ! !! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!!!!! ! ! ! ! ! !!!!! !! !!! !!!!!!! ! !!! ! ! !! ! ! !! !!! ! ! ! ! !!! ! ! !! !!! ! ! ! !!! !! ! ! !!!! ! ! ! ! ! ! ! !! ! !! ! !!!!!!! ! !! ! ! ! ! ! ! ! ! ! !!! !! !! ! !!!! ! !! !! !!!!!!!! ! ! ! ! !!! !! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! !!! ! ! ! ! ! ! !! !!!!!! !! ! !!!!! !! ! ! ! ! !!!! ! ! !! !!! ! !! !!! ! ! ! ! !! !!!! ! ! !!!! !!! !! !!!! ! ! ! ! !! ! !! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! !! ! !! ! !!!!!! !! ! ! ! ! ! !!!! !!! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! !! ! !!!! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !! ! ! !! !! ! !! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! !!! !! ! !! ! ! ! !!! ! ! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! !!! ! !!! ! ! ! !! !!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! !!!! ! !!!!!!!! ! !!!! ! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! !! !! !! !! !!!!!!! !!!!! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! !! !! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !!! ! !!!!!! ! ! ! !!!!!!! !!!!!!! ! ! ! !! ! ! ! ! !!! ! ! !! ! !! !!! !! ! ! ! ! ! !! !! ! ! ! ! ! !! !!! ! !!! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! !! ! !! !!! ! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! !!! !!!! ! ! ! ! !! !! ! !!! !! !! ! ! !! ! ! ! !!!!!! !!!! ! !! !! ! ! !! !! ! ! ! !!! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! !! ! ! !! !! !! ! !! !! !! ! ! !! ! ! ! ! ! !! ! ! !!! !! ! ! ! !!! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!! ! ! !!! ! ! ! ! !! ! ! !!!!! !!!!!!!! !! ! ! !! ! ! ! !!!!!!!!!!!!!! ! ! !!! ! ! ! !!! ! ! ! !!! !!! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !!!!!!!! !!! ! !! !!! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !!! ! ! ! ! !!!!! !!!!!!!! ! !!!! ! !! ! ! !! ! !! !! !!!! !!! ! ! !!! ! !!!!!!!! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !!!! ! ! ! ! ! !! !!!!! ! ! !! !! ! ! !!!! !! ! !! ! !!!!!!! ! !!!!!! ! ! ! !! ! !! ! ! !!! !! !!!!!!!! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! !!!! ! ! !! ! !!!! ! ! ! ! !!!! ! ! ! ! !!! ! ! ! ! ! !!!! ! !! ! !! ! !! ! !!!!!!!!!!!! !!! ! !!! ! ! ! !!!!! !! ! ! ! ! ! !! !! !!!!!!! ! !!!!!! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! !! ! !!! !! !! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! !! ! !!!!!!! ! !! ! !! ! ! ! !! !!! ! !! ! !! ! ! ! ! ! !!!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! !!! ! ! !! !!! ! ! !! ! ! !!! !!!!! ! !!!!! !! !! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! !!!!!! ! ! !!!!! ! !! ! ! ! !!! ! !!! !!! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! !!!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !!! ! ! !! !!! ! !! ! !!!! !!! ! ! !! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! !! ! !! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !!! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! ! ! ! ! ! !! ! !! ! ! ! ! !!!! !! !! ! ! ! !! !! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! !!!! !!! !!! !! ! !!! ! ! !!! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! !!! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! !!! !! ! ! !! !!! ! !!! ! !!!! !! !! !! !! ! !! ! ! !!! ! ! ! !!!! !!!! ! ! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! !! ! !!! !! !!!! ! ! ! ! ! !! ! !!! ! !!! ! !!!! ! ! ! ! !! !! ! ! ! !! ! ! !! ! !!!!!! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! !! ! !! !!! ! ! ! !! !!!!!!! !! ! !!! !! !!!! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!!! !! !! ! !! ! ! ! ! ! !!!!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !!!! ! ! ! !!!!!! !! !!!! ! !! ! !!! !!! !!! !!! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !!! ! !! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! !!! ! ! ! ! ! !!!!! !!!! !!!!! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !! ! ! !! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! !!! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! !!!! !! ! ! !!!! !!! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !!!!!! !!!! ! ! ! ! ! ! ! !!! ! !!! ! ! ! !!! ! !!! ! ! ! !!! ! ! !!! ! !! ! ! ! ! !!!! ! !! ! ! !! ! ! !!! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! !! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! !! ! ! !! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! ! !!! ! ! !! ! ! ! ! !!! ! !! !!!! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !!!! ! !!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! !! ! !!!! ! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! !!!! !!!!! !!!! ! ! ! ! ! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! !! !! ! !! ! ! !!!! ! ! ! ! ! ! ! !!! !! !! !! !!!!! ! ! ! ! ! ! !!! !!!!!! !! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! !! !!! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! !!!!!!! !! ! !! ! ! !!! ! ! ! ! !!!! ! !!!! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!!!!! ! !! ! !!! ! ! ! ! ! ! ! ! !!! ! ! ! !! !! ! !!! ! ! ! ! ! !! ! ! ! !!! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !!!!!!!!! ! !!!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! !!! !! ! !! ! !! ! ! ! ! !! ! ! ! !!! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!!! !!!!! !!!!!!!!!! !! ! !! !! ! ! ! ! ! !! !!! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! ! ! ! !!! ! ! !!! ! ! ! ! ! !! !! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! !! !! ! !!!!! ! ! !! ! ! !! ! ! ! ! ! !!!! ! ! !! ! !! ! ! ! !!!!! ! !!! ! ! !! !! ! ! !!! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! !! !! ! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! !! !! ! ! !!! !!! ! ! !!! !! ! !! ! !! !! ! ! !! !! ! !! !! ! ! ! ! ! ! !! !! ! !! !!! !! ! !!!!! ! !!!!!! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!!! !! !! ! ! !! !!!!! !!!!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! !!! !! ! !!! !!! ! ! ! ! !!!!!! !!!!!! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! !! !!! ! ! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! !! !!!! ! ! !!! !! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!!!!! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !!!! ! ! ! ! ! ! ! !!!! !! !!!!! ! !! ! !! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! ! !! ! !!!!! ! !!! !!!! ! ! !!! !! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! !!!! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!!! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! !! ! ! !! !! !! ! !!! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! !!! !! ! !!!! ! !! ! ! !! !!! ! !! !! ! !!!!! ! ! ! ! ! !!! ! ! !!!! ! ! !! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! !!! ! ! !!! ! ! ! ! ! !!!!! ! ! ! !! ! ! ! !! ! ! ! !! ! !!! ! ! !! !!!! ! !! ! ! ! ! ! ! ! !! ! !! !! !! ! !! !!! !! ! ! !!! ! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! !! ! !!! ! ! ! !! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! ! ! ! ! ! ! !!! !!!!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! !!!! !!!! ! !! ! !! ! !!! ! ! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!!! !!! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! !! ! !! ! !! ! !!! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! !!! ! !!! ! ! !! !! !!! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! !! !!!!!! !! ! !!! !!!!!!! !! !! ! ! ! ! !! ! !!! !!!!! !!! !!!! !!! !!!!! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! !!!!!!! ! ! ! ! ! ! !!! !!!!! ! !! ! ! ! !! ! ! !! !!!! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! !!!! !!! !! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! ! !! !! ! !!! ! ! !! !! ! !! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! !!!!! !! !!! ! ! ! ! ! !!! ! !! !!! ! ! !! ! ! ! !! ! ! ! !! !! ! ! ! ! !!! ! ! ! !!!!!!! ! ! ! ! ! !! ! !! ! !! ! ! ! !! ! !!!! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! !! !!! ! ! ! ! !!! ! !!! ! !! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!! !! ! ! ! ! !! !!!!!! !! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! !! !! !!!!!!!!! !!!! !!!!!!! ! !!!!! ! !! ! !! ! ! ! !! ! !!!! !!!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !!! !!!! ! !!!!!! !! ! !!!! ! ! ! ! !! ! ! ! ! !!!!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! !!! !!!! ! ! ! !! ! ! !!! ! !!!!!! ! ! !! !! ! ! ! !! !! !! !! !! ! ! !! !! ! ! ! ! !!!!! !! !! ! !! !! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!!!! ! !!!!! ! !!!!! !! !! ! ! !! !! !!! !! ! !!!! ! ! ! ! ! ! ! ! M ch gan Depa men o Env onmen a Qua y O ce o O Gas and M ne a s Sep embe 2012 O L AND GAS WELL LEGEND O W G W O G O G W O W W O W O W ! ! F gure 3 O and gas we s n M ch gans Lowe Pen nsu a The No he n P oduc ng T end o he An m s ev den by he h gh dens y o na u a gas we s ed do s n he no he n Lowe Pen nsu a F om MDEQ
    • 8 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 extending to its neighboring counties in the east and west. This region of Antrim production wells is evident in Figure 3 by the high density of natural gas wells in the northern half of the Lower Peninsula. The Antrim comprises four distinct zones, as shown in Figure 4(b). The main producing zones are in the lower Antrim, specifically, the Lachine and Norwood zones, which have elevated concentrations of organic matter. The occurrence of natural radioactive elements such as U correlates well with the organic content of black shales29 . This makes gamma ray logs, which measure the gamma radiation from the bulk rock along a well segment, quite a useful tool in identifying potentially gas-rich zones with high organic content. Measured TOC content in the Antrim ranges from <1–25%wt, with an average of 8%24 . The Antrim shale is a naturally highly fractured formation26,30 . The average porosity in the Antrim is 9%, and it is assumed that natural gas fills roughly half of this porosity31 . Given that the Antrim shale contains a substantial network of natural frac- tures, economic quantities of natural gas can be extracted through use of vertical wells and low-volume hydraulic fracturing stimula- tion that opens horizontal fractures to better connect the natural fractures with the production well. While organic thermal maturity in the Antrim supports the possi- bility for hydrocarbon development near the deeper portions of the Michigan Basin34,35 , a large portion of the produced gas along the margins of the Antrim play, specifically that of the Northern Producing Trend, is of biogenic origin27,28,36 . The reason for this biogenic methane production is that the Antrim experienced sig- nificant Pleistocene freshwater intrusion that facilitated the growth of methanogenic microbes in situ32,37,38 . Although directional drilling has been used to extract natural gas from the Antrim, it is not the most common practice as vertical wells often provide sufficient contact with production zones. It is standard practice to fracture wells drilled in the Antrim prior to completion. Nitrogen foam fracturing was often used in the past, but slickwater hydraulic fracturing has recently become more com- mon in this play (Wilson and Schwank, this series). Recent efforts have investigated use of directional drilling techniques to access very shallow Antrim production zones and natural gas under areas where surface access is limited39 . These test wells were not high-volume hydraulic fracturing well completions, yet they do demonstrate opportunities for future growth in the use of direc- tional drilling techniques to extract additional natural gas from the Antrim shale. There is an overall decreasing trend in the net production of natural gas from the Antrim shale and the number of new wells being drilled in this play. 1.4.2 Utica-Collingwood formation The Utica shale is present throughout the Michigan Basin and is composed primarily of compacted clay minerals and quartz, in addition to minor carbonate minerals. Common TOC values for the Utica are found to be in the range of 0.5–1.3%wt. Water satu- ration in the lower Utica is approximately 44% and porosity ranges between 3–4%. The Collingwood formation is an organic-rich car- bonate with moderate clay mineral content (~20%vol) and would most accurately be described as a limestone, not a shale. Average TOC values in the Collingwood range from 1–6%wt. Total poros- ity estimates in the Collingwood are between 2–5% with almost no water saturation. The Collingwood is located at the top of the Trenton formation, near the bottom of the Utica shale and is pres- ent over only a small section of the northern Lower Peninsula, as is indicated in Figure 5. The Collingwood outcrops in Collingwood, Ontario, Canada, near the shores of Lake Huron. The Collingwood facies is quite thin. At its thickest, the Collingwood is only found to be around 35 feet in total thickness and narrows toward the basin center prior to dis- appearing. It is thus often grouped with the overlying organic-rich zone of the Utica shale because production wells will likely capture both facies within the hydraulically stimulated zones. Together, Figure 4: (a) Stratigraphic column focusing on the Antrim shale; (b) Antrim units and total organic carbon (TOC); (c) the cor- responding gamma ray log for the Antrim shale. Figures modified from McIntosh et al.32 and Curtis33 .
    • 9 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 these two formations are referred to as the Utica-Collingwood formation. The Utica-Collingwood is a newly developing reservoir, having gained much attention from industry after the successful drilling of one deep horizontal well by Encana in late 20095 . Figure 5 also shows the current and pending permits for high-volume hydraulic fracturing in Michigan. A key take-away point related to Figure 5 is that the majority of applications for high-volume well completions in Michigan are located in the Utica-Collingwood. 1.4.3 A-1 Carbonate The A-1 Carbonate is part of a sequence of evaporite and carbon- ate formations that form the Salina Group. Historically, the A-1 Carbonate in Michigan has been a primary hydrocarbon reservoir. It is bounded from above and below by evaporite formations and has three distinct areas of hydrocarbon production within the basin. In the northern and southern portions of the Lower Peninsula, several production wells target the A-1 overlying the Niagaran Pinnacle Reef. In the southwestern region of the Lower Peninsula there is some hydrocarbon production from the A-1 where the res- ervoir displays solution-enhanced porosity and fracturing due to salt dissolution. Much of the potential production from the A-1 via high volume hydraulic fracturing is within the central portion of the Michigan Basin, including the south central and western central areas of the Lower Peninsula. Higher porosity and permeability within the A-1 in this region occurs as intercrystaline porosity due to dolomitization and provides a more continuous resource play conducive to development via high volume hydraulic stimulation techniques40 . Porosity in this region is on the order of 6–7% and TOC values tend to be <1%wt. Figure 5: Map showing the issued and pending permits for high-volume hydraulically fractured wells3 . The extent of the Collingwood is also shown with formation thickness (in feet) given by the isopach map.
    • 10 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 2.0 STATUS, TRENDS & ASSOCIATED CHALLENGES T he main body of this report is split across five primary topics of interest: 1. Michigan Basin hydrogeology 2. Potential fluid migration pathways 3. Unconventional reservoir proximity to drinking water resources 4. Flowback water chemistry 5. Water withdrawal and disposal In the following section a review of each topic is presented, includ- ing identification of challenges and opportunities to address data gaps associated with each topical area. 2.1 Michigan Basin hydrogeology 2.1.1 Hydrogeologic units The Michigan Basin is composed of a series of sedimentary for- mations stacked upon one another, each having different charac- teristics in terms of porosity, permeability, mineralogy, and overall levels of heterogeneity across all of these properties. Even within a single formation these characteristics may change as one moves from the basin center outward toward the margins. In some cases (e.g. Collingwood) a given formation may not be continuous throughout the basin. The majority of sedimentary formations within the Michigan Basin are either water saturated or contain a mixture of water, oil, natural gas, and CO2 . Formations with very low porosity and permeability may not always be fully fluid satu- rated. A formation is described as being fully saturated when a fluid occupies all of the available porosity. If the permeability is sufficient to allow flow of water within the formation and/or allow water to be extracted in significant quantities, the formation is described as an aquifer. Figure 6 illustrates an approximate group- ing of aquifer and aquitard units in the Michigan Basin. Michigan Basin formation waters tend to be highly saline. Formation water salinity is due to a combination of formation water origin being derived from evaporated Paleozoic seawater and due to dissolu- tion of evaporite deposits41–43 . Formation water salinity increases rapidly with depth, reaching total dissolved solids concentrations of >350 g/L at depths as shallow as 2,600 feet37 . 2.1.2 Regional subsurface flow Fundamentally, groundwater flow is driven by hydraulic gradients with water flowing from areas of high pressure to areas of low pres- sure. In the Michigan Basin modern groundwater flow occurs from areas of higher elevation to areas of lower elevation, generally following surface topography with flow from topographic highs in the north, west, and south toward the topographically lower Saginaw Bay area15 . Meteoric water recharges the glacial aquifers at the surface and maintains the water level in these open aquifers. At topographic highs, meteoric water also recharges sedimentary formations where they subcrop below the glacial till15 . In shallower aquifers such as the Marshall sandstone, flow from recharge zones in the northern and southern subcrop regions discharges into Lake Michigan and Lake Huron in the western and eastern subcrop zones44 . A study by Vugrinovich45 found a correlation between lower sub- surface temperatures and zones of aquifer recharge in Devonian and Ordovician aquifers within the Michigan Basin. Positive devi- ations away from the geothermal gradient for the Michigan Basin of 19°C/km, which assumes that heat transfer occurs only due to Figure 6: Generalized hydro- logic units within the Michigan as defined by Lampe14 . Figure modified from Lampe14 .
    • 11 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 conductance, were found near aquifer discharge zones. This sug- gests that regional groundwater flow and subsurface heat transfer are correlated. Deviations away from the expected geothermal gradient, as evidenced by Vugrinovich45 , are associated with subsurface convection and correlate with regional groundwater flow patterns—for example, cooler waters correspond to aquifer recharge zones. Evidence of subsurface temperature gradients and measured hydraulic head suggest that groundwater flow in many of the deeper aquifers in the Michigan Basin occurs with recharge at formation subcrops in the southern and northern portions of the basin with discharge near the Saginaw Bay region15,45 . The following paragraph draws heavily on data presented by Vugrinovich15 , who completed a very thorough study of regional flow characteristics of deep brines in the Michigan Basin. Although many of the aquifers in the Michigan Basin present hydraulic head distributions that mimic surface topography, suggesting gravi- ty-driven flow conditions, some aquifers present deviations from expected nominal pressure gradients corresponding with depth. When aquifers are bounded by aquitards (e.g. shales), in situ pres- sure conditions may be higher than those expected for an equiva- lent column of water at a given depth (nominal expected pressure), which means the aquifers have not reached equilibrium with the current land surface topography. An example of one such aquifer is the Berea sandstone, which is bounded from above by the Sunbury shale and below by the Bedford and Antrim shales. Evidence that the Berea remains overpressured supports the existence of good confining characteristics (i.e. low permeability) of the shales that bound it. The Antrim is also found to be overpressured over much of the basin and is believed to leak into the underlying Traverse limestone15 . The Dundee formation is substantially underpressured near the center of the Michigan Basin, with recharge occurring throughout the basin from both the overlying Traverse limestone and the underlying Detroit River Group aquifers. The confining units that separate these aquifers from the Dundee are considered to be leaky aquitards. The hydraulic low in the central portion of the Michigan Basin is believed to be associated with artificial fluid discharge related to hydrocarbon production from the Dundee within this region. The A-1 Carbonate is overpressured in the cen- tral portion of the Lower Peninsula as it is bounded from above and below by impermeable evaporite formations. The fact that the A-1 remains overpressured and has not re-equilibrated with the current surface topography demonstrates the good confining character- istics of the overlying strata. Where the A-1 is unconfined near the edges of the basin, flow is thought to be gravity-driven, with discharge occurring in the southern end of the Lower Peninsula and under Lakes Huron, Michigan, and Erie. Finally, flow within the Trenton-Black River occurs with recharge at the subcrop and flow is toward the center of the basin. Data suggest that the Trenton-Black River aquifer is confined by the bounding aquitards. This supports the basin-wide impermeable nature of the Utica-Collingwood that overlies the Trenton limestone. Measured elevated concentrations of Br- and He in shallow aqui- fers provide evidence of vertical leakage of deeper brines into shallower formation waters (Marshall sandstone, glacial drift)44 . Higher concentrations of Br- are representative of evaporated sea- water46 and help differentiate the origin of brine salinity from that of dissolution of evaporites that would result in very little Br- com- ponent47 . Near the recharge zones of the Marshall, the water in the sandstone is potable and serves as the primary source of municipal drinking water for much of Jackson County48 . By the time the water discharges from the Marshall sandstone into the Saginaw Bay it is nearly as saline as the underlying Antrim and Traverse brines. This observation coupled with the evidence of elevated concentrations of the two conservative tracers (Br- , He) support upward migration of deeper basinal brines from the Traverse limestone into the over- lying Marshall sandstone44 . 2.2 Potential fluid migration pathways Concern of water quality impacts and specifically, groundwater con- tamination, is often cited as top issue regarding the environmental implications of hydraulic fracturing49 . This is understandable given that clean drinking water is a resource that we cannot live without. Migration of methane, the dominant component of natural gas, into groundwater reservoirs has received significant attention in the context of unconventional natural gas extraction. Natural gas migration beyond source rock formations is commonplace and is what leads to the development of conventional natural gas plays. The reason for this is that methane gas is much less dense than water and therefore will rise due to buoyant forces if a pathway allowing vertical migration exists. Reports of natural gas migration into drinking water aquifers50 and investigations of contamination of drinking water by hydraulic fracturing chemicals51 have fueled public concern of drinking water contamination by hydraulic frac- turing-related activity. A study by Osborne et al.50 cited evidence for natural gas migra- tion into shallow drinking water aquifers associated with natural gas drilling in Pennsylvania. These authors used carbon isotopes to differentiate between thermogenic (i.e. ‘deep’) methane and biogenic (i.e. ‘shallow’) methane in order to determine the origin of methane found in the drinking water wells. The assumption was that methane of thermogenic origin could only be found in well water if it were to have escaped from deeper sources such as (but not exclusively) from the Marcellus shale, whereas methane derived from methanogenic microbial activity would be found in shallow reservoirs where meteoric water recharge takes place. A conclusion from this study was that water wells located nearby nat- ural gas production wells had a higher contribution of thermogenic
    • 12 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 methane than wells that were further away from natural gas drill- ing sites. The association of shallow water thermogenic methane contamination with hydraulically fractured natural gas wells led the authors to suggest a link between the practice of hydraulic fractur- ing and methane seepage to the surface. To be clear, the study by Osborne et al.50 does not definitively ascribe methane leakage to hydraulic fracturing, but inferences are made that suggests such a link may exist. It was noted that another possible, if not more likely, leakage pathway could be flow up the wellbore due to incomplete well cementation. Such leakage would be non-unique to hydrauli- cally fractured wells and therefore does not necessarily provide any evidence directly related to hydraulic fracturing-derived leakage. A different study by Molofsky et al.52 suggested that methane leak- age occurs naturally and is not correlated with natural gas drilling activity, but instead, topography. Others have also pointed out that natural seepage of methane is known to occur and have argued that without baseline measurements of methane concentrations prior to drilling activity one is unable to definitively attribute meth- ane in drinking water to drilling activities53–55 . As these competing studies demonstrate, assessing leakage can be quite complicated and requires having well established baseline water quality data49 . The key issue regarding leakage is the existence of leakage path- ways coupled with a driving force promoting vertical migration of fluids beyond the stimulated reservoir. This report does not attempt to assess the likelihood of groundwater contamination from hydraulic fracturing in Michigan. However, theoretical leak- age pathways and the parameters controlling movement of flu- ids beyond the fractured reservoir are discussed in the following sections. 2.2.1 Fracture propagation One concern surrounding the practice of hydraulic fracturing is that the induced fracture network will extend beyond the target forma- tion. If this were to occur then flow pathways would exist between the target reservoir and overlying formations, possibly allowing for migration of fracturing fluids beyond the production reservoir. The topic of hydraulically-induced fractures has been studied exten- sively, as understanding how the fracture network develops is key to both evaluating the enhanced productivity of a well and ensur- ing the safety of overlying sources of potable water56,57 . A study by Fisher and Warpinski58 provides information on the relative height of induced fractures as a function of depth for several unconven- tional resource plays across the U.S. collected over a 9-year period ending in 2010. Data gathered from hydraulically stimulated wells in other states does not show evidence of hydraulically-induced fractures extending into overlying fresh water aquifers58 . This data does not support the scenario for propagation of fractures all the way to the surface; however, no data was collected for drilling activ- ity in the Michigan Basin. Collection of similar data for hydraulic fracturing operations in Michigan would be beneficial in helping evaluate the extent and direction of fracture propagation within the unconventional plays in the Michigan Basin. 2.2.2 Existing faults, natural fracture networks, diffuse leakage Natural leakage of natural gas and oil from shale source rocks is common and is the mechanism by which economic deposits within permeable reservoirs develop. In the case of conventional natural gas exploration, gas plays are often identified by the existence of natural stratigraphic or structural traps above permeable reservoirs that overlie organic-rich source rocks. The fundamental force driv- ing the upward migration of oil and natural gas is that these fluids are less dense than formation brines and rise due to natural buoy- ant forces. Vertical migration of these fluids takes place over geo- logic timeframes and is often retarded or completely blocked by overlying impermeable formations. In areas where impermeable caprocks are not present above source rocks, for example at loca- tions where the Antrim subcrops beneath glacial drift in Michigan, evidence of natural gas migration into potable aquifers has been found59 . As previously discussed in section 2.1.2, widely distributed cross-formational flow from deeper formations to shallower forma- tions has been documented in the Michigan Basin based on high He fluxes and elevated Br- concentrations in the Marshall sand- stone44 , confirming previous explanations for the occurrence of higher salinity brines at shallow depths within the basin60 . Evidence for natural leakage of deeper brines into shallower formations is not unique to the Michigan Basin. Warner et al.61 recently found evidence for natural migration of Marcellus shale brines into shal- lower aquifers. One important issue to point out here is that the timeframe for this natural vertical brine leakage is not well known, and it is thus difficult to relate such leakage, which may have taken place over geologic timescales, to the short timescales of recent subsurface energy extraction practices. Additionally, low water sat- urations in many shale gas formations may create strong capillary tension trapping mechanisms that prevent upward migration of injected fluids and formation brines62 . Many of the existing oil and gas fields within the Michigan Basin are associated with structural features such as low displacement faults30 . The two largest of these structural features are the Howell Anticline and the associated Monroe-Lucas Anticline structures in southeast Michigan. Low displacement faulting is observed across the Michigan Basin in a general northwest-southeast trending direc- tion. These features can be closely linked to the oil and gas plays shown in Figure 3 by noticing the spatial trends in oil wells across the state. Many of these structural features can be interpreted as small anticline folds due to minimal vertical displacement of the
    • 13 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 geologic strata. Although the deformation events that generated these anticline structures within the basin were not severe enough to cause significant sediment displacement, they are believed to have allowed vertical migration of hydrothermal fluids that lead to dolomitization across much of the Michigan Basin30 . The natural jointing within the Antrim shale is also dominantly northwest-south- east and in some cases may be associated with stresses from these features26 . It is not clear, however, that the jointing in the Antrim can be attributed solely to structural trends within the basin63 . 2.2.3 Leakage along well bores Existing wells may serve as leakage conduits if they are improperly sealed or if their seals have been degraded over time64,65 . When wells are drilled, completed, or abandoned, drilling companies must abide by regulations that specifically state what efforts that must be undertaken to ensure that no contamination of potable aquifers occurs due to fluid leakage up the wellbore or along the well annulus (exterior of the well casing)66 . These include guidelines for well casing, cementation, and plugging. However, these types of stringent guidelines aimed at ensuring the safety of overlying drinking water resources were less common or non-existent prior to the 1950’s. In the past, abandoned wells may not have been properly plugged due to previously lax regulatory enforcement, and these wells may be poorly documented67 . This means that there is a level of uncertainty regarding the integrity of wells drilled and/or abandoned prior to this time period. Undocumented abandoned wells are often described as being orphan wells. To address the issue of orphan wells in Michigan, the Orphan Well Program (Act 308) was established in 1994. This program provides funding for the identification and remediation of improperly abandoned wells located in Michigan. This proactive approach to addressing leakage concerns related to abandoned wells is a step in the right direction toward reducing the likelihood of leakage up existing wellbores within the state of Michigan. To address the issue of existing wells in close proximity to newly per- mitted wells, the MDEQ requires the operator of a proposed well to evaluate any preexisting wells within up to a ¼ mile radius of a new proposed well. If potential conduits are identified then there are several options the permit applicant can take. One option is to relocate the well where no other wells fall within the radius of influ- ence. Another option is to demonstrate that the hydraulic fractur- ing activity will not cause fluid migration up the neighboring well. In addition to abandoned or degraded wells, there is a possibility for incomplete cementation of the wellbore during well construc- tion. The integrity of the cement bond is evaluated through use of a cement bond log after the well is cemented. Two challenges associated with cementing horizontal wells are associated with (1) solids settling within the drilling slurry resulting in a mud channel along the base of the horizontal leg of a well and (2) excess water in the cement slurry resulting in a water channel along the top of the horizontal leg68 . Both of these events may compromise the cement casing bond quality along the horizontal segment of the wellbore. These potential challenges highlight the importance of ensuring good cement casing bond quality along the vertical segment in order to prevent leakage along the wellbore. 2.2.4 Driving force for upward fluid migration A study by Myers69 suggested a possibility for vertical migration of hydraulic fracturing fluids and formation brines from the Marcellus shale. This finding was related in part to the fact that the study region of the Marcellus is naturally overpressured and the model assumed the existence of a single leakage pathway from the Marcellus to the surface. That is, there is a preexisting natural pres- sure gradient favoring upward advection of fluids and this gradient was exacerbated by the assumption that hydraulic fracturing would lead to prolonged elevated pressures within the reservoir. Finally, Myers69 further assumed that the injection of fresh water associ- ated with hydraulic fracturing would dilute the formation water brines to the point that there would also be density-driven vertical flow of the less-dense mixed hydraulic fracturing/formation fluids. However, given that the volume of injected fluid is small relative to that held within the reservoir and overlying formations, prolonged density driven flows induced by the injection of the slickwater would be unlikely. In a comment on the Myers69 study, Saiers and Barth70 argue that the model used by Myers69 neglects many critical hydrogeological properties and is over simplified to the point that its findings are not applicable to modeling the fate of hydraulic fracturing fluids in the Marcellus. The validity and applicability of Myers’ model69 is further challenged in a comment by Cohen et al.71 The current technical report does not attempt to evaluate these studies, but points to this academic discussion as evidence of the challenges in accurately modeling complex systems such as that presented by Myers69 . An important point of discussion here is that the key factor controlling vertical leakage of hydraulic fracturing fluids beyond the target reservoir is the existence (and persistence) of a driving force for upward fluid movement. This is of course in addition to the existence of a potential leakage pathway. Such a driving force could be due to either pressure or density gradients promoting vertical fluid movement. Accurate leakage risk assessment will thus be improved if modeling assumptions can be validated by field data. Another energy technology that presents many similar environ- mental concerns in the context of leakage of injected fluids back to the surface is geologic carbon sequestration. The main concept behind carbon sequestration is that CO2 captured from point source
    • 14 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 emission sites is injected into the subsurface in an effort to reduce the emissions of CO2 to the atmo- sphere. Two key differences between CO2 leakage risk and the risk of migration of natural gas related to hydraulic fracturing of unconventional reservoirs, are that (1) CO2 injection creates an overpressured system in the injection aquifer, whereas natural gas production creates an underpressured system and (2) CO2 injection relies on an intact impermeable caprock to prevent CO2 leakage, whereas shale- gas production relies on fracturing impermeable formations through hydraulic stimulation. A large body of research has been undertaken in an effort to understand the fate of the injected CO2 and to assess the risk of CO2 leakage back to the sur- face72–75 . Much attention has been paid to assess- ing leakage up abandoned wells as these wells can present a conduit to the surface, bypassing layers of impermeable rock64,65 . It may be beneficial to examine the risk analysis framework used within the CO2 sequestration research community in future efforts to assess leakage risk related to hydraulic fracturing activities. 2.3 Unconventional reservoir proxim- ity to drinking water resources Michigan is fortunate in that it has abundant sup- plies of fresh water. The Great Lakes alone hold approximately 21% of the world’s fresh surface water76 . Added to the fact that fresh water is abun- dant in Michigan, throughout the Lower Peninsula past glaciation has left the majority of the state’s surface covered in a thin layer of unconsolidated sediments. These sediments consist of layers of sand, gravel, and clay. The sand and gravel layers serve as the fresh water aquifers for approximately 21% of residential wells and municipal water withdrawals within the state of Michigan77 . Larger municipalities, such as Detroit, draw water directly from the Great Lakes and its surrounding waterways (e.g. Detroit River). However, in more rural areas of the state where residential densities are lower there is a higher reliance on shallow subsurface glacial aquifers. For example, the village of Kalkaska has municipal drinking water wells that draw water from glacial aquifers at a depth of 102 feet below the surface48 . In estimating the proximity of unconventional reservoirs to drinking water resources, it is important to note that the glacial cover is vari- able and dependent on the exact location of interest. For example, Figure 7: Schematic showing approximate relative locations of Antrim and Utica-Collingwood reservoirs in relation to glacial drinking water aquifers. Distances are approximate and representative of averages depths to top of formation near Kalkaska County, MI. Exact depths will depend on specific location of wells. Figure modified from MDEQ and Michigan Basin Geological Society
    • 15 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 some of the thickest zones of the glacial sediment occur near the Northern Producing Trend with thicknesses on the order of 1,000 feet. As discussed previously, most drinking water in Michigan is either sourced from surface waters or from shallow wells that tap into the glacial aquifers at depths generally less than 300 feet. There are also locations in Michigan where drinking water is pumped from the lithic strata, such as in Jackson County where the municipal wells draw drinking water from the Marshall sandstone. But even in this case, the drinking water is pumped from shallow depths (~220 feet) and is due in part to the fact that the glacial sediments are very thin in this area48 . Given the bowl shape of the Michigan Basin, formations dip downward toward the center of the Lower Peninsula. This means that the exact proximity of each potential target unconventional hydrocarbon reservoir to overlying drinking water aquifers is a function of location within the basin. The Antrim shale is present through most of the Lower Peninsula, subcropping as far north as Cheboygan County and as far south as Lenawee County. The Utica-Collingwood contact is found at depths near 4,600 feet in Charlevoix County and is at its deepest near the intersection of Gladwin, Clare, and Roscommon Counties where it is found at depths just over 10,000 feet below the surface. Moving further south, the Collingwood continues into Gratiot County where it is found at around 8,000 feet deep before it eventually discontinues (see Figure 5). The A-1 Carbonate is found throughout the basin and is also deepest near the center of Michigan’s Lower Peninsula, where it is found at approximately 8,000 feet. It is commonly found at depths on the order of 4,500 feet near the northern producing zones of the Niagaran Pinnacle Reef and a recent high-volume completion drilled in Oceana County (Alta-Riley #1-22HD1) tar- geted the A-1 at a depth of 4,000 feet. Figure 7 provides a depiction of the relative location of glacial drinking water aquifers to the Antrim, A-1, and Utica-Collingwood. The cross section shown in Figure 7 identifies all formations pres- ent in the Michigan Basin. The relative depth to a given formation decreases as one moves outward from the basin center and some of the shallower formations disappear completely (see Figure 1). Figure 7 demonstrates the relative distances between the uncon- ventional reservoirs discussed in this report and is used simply to provide a general idea of where these formations lie with respect to the freshwater aquifers. As discussed above, the exact distance between a given formation and glacial aquifers will be dependent on the specific location within the basin. 2.4 Flowback water chemistry Hydraulic fracturing of deep horizontal wells often requires large quantities of water (on the order of several million gallons per deep horizontal well18,78 ) that are mixed with small amounts of chemicals (<1%wt) and physical proppants (e.g., sand). This water mixture, referred to as the hydraulic fracturing fluid or slickwater, is injected into the shale to create and hold open zones of high permeability to increase contact with the gas-containing rock. Of the total volume of hydraulic fracturing fluids injected into a well, amounts varying from 10 to 70% may return to the surface Permit # Well Name Well No County Target Formation Water Utilized (gal) *Flowback water (gal) 59112 SCHULTZ 1--36 SANILAC A1 Carbonate 154,600 121,800 60041 HUBBEL 2-22 HD1 MONTMORENCY Niagaran 220,000 (est) 34,524 60170 STATE KOEHLER & KENDALL 1-27 HD1 CHEBOYGAN Collingwood 3,256,596 1,000,902 60212 KELLY ET AL 1-26 HD1 HILLSDALE Black River 228,312 188,152 60305 STATE WILMOT 1--21 CHEBOYGAN Collingwood 109,410 15,330 60360 STATE EXCELSIOR 1-13 HD1 KALKASKA Collingwood 5,860,764 505,386 60380 CRONK 1-24 HD1 GLADWIN A1 Carbonate 758,454 725,046 60389 STATE EXCELSIOR 1-25 HD1 KALKASKA Collingwood 8,461,614 615,972 60537 MCNAIR ET AL 1-26 HD1 HILLSDALE Black River 350,448 3,893 60452 WILEY 1-18 HD1 GLADWIN A1 Carbonate 1,420,939 1,674,960 60545 STATE EXCELSIOR 2-25 HD1 KALKASKA Collingwood 12,562,096 64,451 60546 STATE EXCELSIOR 3-25 HD1 KALKASKA Collingwood 21,112,194 35,202 60560 STATE RICHFIELD 1-34 HD1 ROSCOMMON Collingwood 4,811,940 1,328,250 60579 STATE GARFIELD 1-25 HD1 KALKASKA Collingwood 12,539,639 NA 60575 RILEY 1-22 HD1 OCEANA A1 Carbonate NA NA TABLE 1: High-volume well completions in the state of Michigan *Reported flowback volumes account for fluid collected over a period ≤60 days after well completion and this collection period may vary between individual wells. Data from MDEQ23 , accurate as of March 13, 2013. State Garfield data from FracFocus81 .
    • 16 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 as ‘flowback water’ after the pressure is reduced and gas begins to flow toward the wellhead78,79 . As shown in Table 1, the average amount of flowback water returning to the surface for high-volume hydraulically fractured wells in Michigan is around 37% of the total injected volumes. This volume may include some amount of native formation brines and is the sum of fluid collected over a period of less than 60 days following completion of the well. The collection periods are not necessary consistent from one well to another, so the flowback data in Table 1 should be considered as approximate values. The addition of proppants to hydraulic fracturing fluids is to ensure that the fractures remain open after the pressure is released, allow- ing for continued gas flow in the stimulated area. The chemicals are added to alter the fluid properties of the hydraulic fracturing water in ways that enhance the fracturing process, thus optimizing fracture efficiency and well productivity. Table 2 provides several examples of typical chemical additives used in hydraulic fracturing fluids in Michigan wells. This list is not comprehensive, as the EPA has identified more than 1,000 possible chemicals that may be used in hydraulic fracturing fluids within the U.S.80 However, most well completions use around 10 different chemical additives in their particular slickwater formulations81 . Disclosure to regulatory authorities of the particular chemicals and amounts used in the slickwater is not currently required in all states. However, many companies are beginning to voluntarily release this information through the website www.FracFocus.org. As of January 2013, more than 35,000 chemical disclosures have been reported on FracFocus. A side effect of hydraulic fracturing is the exposure of previously hydraulically isolated minerals that may contain naturally occur- ring radioactive material and toxic metal elements82–85 . Some of the possible radionuclides that may be released into formation and hydraulic fracturing fluids include U, Th, and their daughter products 226 Ra and 228 Ra82,86 . The half-life of 226 Ra, the most stable Ra isotope, is 1600 years. In addition to enhanced levels of radio- nuclides, flowback fluids may contain elevated concentrations of naturally-occurring salts, metals, organics, and methane50,87–89 . In Michigan, native formation brine salinities can be on the order of 200–400 g/L total dissolved solids37 . Evaporite dissolution is believed to be the major driving force in controlling the high brine salinities found in Michigan Basin sedimentary formations.42,90 Organic-rich shales in close proximity to drinking water reservoirs have been associated with low-level water contamination due to natural weathering processes causing the release of trace metals to groundwater91–95 . In trying to identify specific contaminants of concern, the EPA has identified a list of target chemicals to monitor when assessing the adequacy of flowback water treatment. This list includes monitoring of As, Ra, Sr, Ba, U, and BTEX components (benzene, toluene, ethylbenzene, and xylene), among others80 . Some of these contaminants may be naturally present in the for- mation brines, so assessment of flowback water quality must take this into account, requiring that baseline measurements be taken prior to drilling activities. 2.4.1 Characteristics of organic-rich shale formations Black shales, so named for their higher organic carbon content Application Chemical Name Maximum Concentration (% by mass) Crosslinker Ethylene Glycol 0.3 Acid Hydrochloric Acid 0.36 Acid Corrosion Inhibitors Isopropyl Alcohol 0.3 Methanol 0.6 Biocides 2,2-Dibromo-3-Nitrilopropionamide 0.6 Friction Reducer Polyacrylamide, Hydrotreated Light Petroleum Distillate 0.3 Surfactants & Foamers Glycol Ethers 0.07 2-Butoxyethanol .003 Gel Breaker Ammonium Persulfate 0.6 Sodium Persulfate 1 Gelling Agent Guar Gum 1 Iron Control Agent Citric Acid 1 Proppant Crystalline Silica, Quartz 0.88 Non-emulsifier Isopropyl Alcohol 0.4 Methanol 0.13 TABLE 2: Summary of Common Hydraulic Fracturing Chemical Additives Used in Michigan Table 2: Common hydraulic fracturing fluid additives used in the state of Michigan. Data were compiled from MSDS reports filed with MDEQ and data posted on FracFocus.org for high-volume fracturing completions. This list is not comprehensive of all potential additives.
    • 17 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 and dark color, maintain reducing environments in the subsurface. These reducing conditions can lead to immobilization and subse- quent concentration of trace toxic metals and naturally occurring radioactive elements such As and U, respectively85 . The elements are often bound in sulfide (e.g., FeAsS) and U(IV)-bearing minerals. The naturally occurring higher radioactive content of organic-rich shales has been utilized by the oil and gas industry for detecting promising formations when prospecting for potential shale source rocks during hydrocarbon exploration. The enhanced gamma radi- ation emission from the shales serves as an indicator for elevated organic content. Figure 4(b,c) shows the correlation of TOC and gamma radiation for the Antrim shale in Michigan. These organic-rich shales also contain increased concentration of U and Th decay daughter products 226 Ra and 228 Ra82,83 . Uranium content of Antrim shale is on the order of 10–40 ppm and scales proportionally with TOC96 . Trace rare earth and heavy metals are also found at higher abundances in black shales relative to other sedimentary rocks84,85 . The highest natural concentrations of As (20–200 ppm) tend to be found in organic-rich or sulfide-rich shale formations91 . Higher concentrations of 226 Ra in produced fluids are often correlated with higher salinities, as at higher salinities other metals such as Ca2+ outcompete Ra2+ for ion exchange sites of the reactive clay mineral surfaces, thereby enhancing the mobility of Ra2+ 82 . Positive correlations have also been shown for high concen- trations of innocuous salts (e.g., NaCl, CaCl2 ) and elevated con- centrations of trace heavy metals such as Ba87 . Flowback fluids that come back to the surface include portions of the injected fluids that have reacted with host rock minerals and portions of native formation brines. These native brines may contain elevated con- centrations of such elements as Ra and Ba that far exceed drinking water standards97 . Analyses of flowback waters from hydraulically fractured Marcellus shale wells have found 226 Ra concentrations that exceed EPA minimum standards for safe discharge to the environment by a factor of 250 or more98,99 . This is in addition to the high concentrations of salts and the many hydraulic fracturing chemical additives that return to the surface78 . 2.4.2 Potential contaminant release mechanisms Although oxygen scavengers are often added to the hydraulic frac- turing fluids to prevent pipe corrosion, it is still possible that the injection of oxygen-rich water may lead to oxidative dissolution of reduced mineral phases containing radioactive elements or heavy metals or metalloids, and thereby release more soluble oxidized forms of these elements to the aqueous phase. For example, under reducing conditions, U forms the highly insoluble uranium oxide phases, such as uraninite (UO2(s) ), but in the presence of oxy- gen-bearing water UO2(s) can be readily oxidized to highly soluble U(VI) uranyl complexes100–102 . Because U(VI) forms highly soluble carbonate-calcium- hydroxide complexes in water103–105 , the rate and extent of the oxidative dissolution will also be influenced by the concentration of calcium and carbonate and pH100 . Likewise, reduced sulfide minerals that contain toxic elements (e.g., AsS or HgS) may be subject to oxidative dissolution, and also serve as a source for increased toxic element release to the flowback fluids. Some examples of this mechanism for iron sulfide phases include the release of As by the oxidation of arensopyrite (FeAsS)106 or As associated with mackinawite (FeS)107 , depending on pH. Yet another pathway may result from the dissolution or desorption of reduced As from Fe-(oxyhydr)oxide minerals as a result of the change in system redox conditions associated with hydraulic fracturing.108 Many of the toxic metals found in shales may also sorb onto the surfaces or into the interlayers of clay minerals85,108 . Chemical addi- tives that alter clay mineral surface properties may play a key role in determining the fate of sorbed metals. For example, surfactants can alter the mobility of sorbed metal cations. In a study by Hayes et al.109 , divalent cations sorbed to fixed-charge interlayer sites in swelling smectite clays could be released by a series of alkyl- trimethylammoniumchloride surfactants (C8TMAC, C12TMAC, and C16TMAC), while those sorbed to external clay surface hydroxyl sites were not impacted. Mono- and poly-cationic amine compounds (e.g. triethanol amine methyl chloride) are also often used in the hydraulic fracturing fluid mixtures as clay stabilizers to prevent swelling of 2:1 clay minerals. These additives compete for sorption sites on the negatively charged clay surfaces and may also impact the release of sorbed metals in shale-gas reservoirs110 . 2.5 Water use in hydraulic fracturing: withdrawal and disposal 2.5.1 Water withdrawal Typical volumes of water used for high-volume hydraulic fracturing completions are on the order of 2-5 million gallons per well18,80 ; however, a recent high-volume completion in Michigan (State Excelslor 3-25 HD-1) used a reported 21.1 million gallons81 , high- lighting the variation among individual well completions. Small hydraulic fracturing completions, such as those used in vertical Antrim wells, use approximately 50,000 gallons of water. To put this water use into perspective, an Olympic size swimming pool holds approximately 660,000 gallons of water. Specific water use for completed high volume hydraulically fractured wells in Michigan is given in Table 1. The actual amount of water needed to finish hydraulically fracturing a given well is a function of the well depth, lateral extent, and number of hydraulic fracturing stages. The topic of water withdrawal as it relates to high-volume hydraulic fracturing activity is covered in greater detail within the Environment/Ecology report (Burton and Nadelhoffer, this series). Although the amount of water utilized in hydraulic fracturing may be comparable to (or far less than) that used in other industries18 ,
    • 18 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 the volume and short timeframe of extraction may lead to local impacts. Water that is withdrawn from shallow subsurface reser- voirs may influence local public water supplies, ecosystems, and compete with other local industries111 . The MDEQ has issued an instruction requiring use of an online water resource impact assess- ment tool to evaluate the area of potential adverse resource impact (ARI) prior to any high-volume water withdrawal from subsurface water reservoirs. MDEQ defines an ARI as any withdrawal that (1) reduces stream flow to a point where sensitive fish populations may be adversely impacted due to the reduced flow conditions or (2) reduces surface levels of a body of water to the point where its ability to support characteristic fish populations is functionally impaired. Any well completion that requires withdrawal of more than 100,000 gallons of water per day (averaged over a thirty-day period) must complete such a risk assessment prior to approval or issuance of a drilling permit. See the Environment/Ecology report (Burton and Nadelhoffer, this series) for a more thorough discussion on water withdrawal issues and use of MDEQ’s Water Withdrawal Assessment Tool. 2.5.2 Flowback disposal Disposal of flowback and produced brine fluids in Michigan occurs via deep well injection into brine disposal wells. This method for disposal of produced oilfield brines is very common throughout the U.S.112 These brine disposal wells are regulated and permitted under the EPA Safe Drinking Water Act through the Underground Injection Control as Class II wells. They are also regulated by the MDEQ under Michigan’s Oil and Gas Regulations66 . Michigan has 1,460 Class II injec¬tion wells in current operation113 . Of these 1,460 wells, brine disposal wells constitute about half, with the remainder serving other uses such as water injection for enhanced oil recovery or gas injection for natural gas storage. Brine disposal wells are often co-lo- cated alongside oil and natural gas production wells. This means that producers in Michigan have a much easier time disposing of hydraulic fracturing flowback fluids than producers in Pennsylvania where only five brine disposal wells are currently in operation49 . The MDEQ requires that all flowback and produced fluids be con- tained in aboveground steel containers and reinjected back into the subsurface via brine disposal wells66 . Open pits are not used for flowback water or produced brine storage66 . Since flowback is treated as equivalent to produced brine, handling of hydraulic fracturing flowback fluids is no different than handling of produced brine. Most oil and natural gas production wells in Michigan pro- duce copious amounts of brine. This produced water amounted to more than 4.2 billion gallons of brine that was disposed of via brine injection wells across Michigan in 2011. Table 3 relates this total volume of injected brine to the cumulative volume of flow- back water associated with high-volume hydraulic fracturing in all completed Michigan Basin wells. A topic of concern often cited in relation to flowback disposal via deep well injection is a risk of induced seismicity. Recent events such as that in Youngstown, Ohio, where flowback disposal served to lubricate a nearby fault and cause a seismic event, are motivat- ing these concerns114 . The Michigan Basin has been tectonically stable since the Jurarssic9 and contains many formations capable of receiving fluid, as evidenced by the prevalence of brine disposal wells throughout the state. Given that brine disposal wells have been actively injecting fluids into the Michigan Basin for many years without any reported incidents of induced seismicity, Michigan has the infrastructure to easily handle the current volume of flowback disposal. A recent study by Keranen et al.115 identified the poten- tial for a time lag (on the order of years) between the occurrence of fluid injection-induced seismic events and the start of fluid injection in wells located in Oklahoma. These findings support a need for continued monitoring efforts and careful site selection for future brine disposal wells. One other question related to brine disposal is whether these disposal wells may be altering the local hydraulic gradients and possibly discharging fluids beyond the target injection reservoir. Nearly half of the 712 brine disposal wells in current operation in Michigan inject fluids into the Dundee limestone23 . As demon- strated in Table 3, this may amount to well over a billion gallons of fluid injected into the Dundee. Closer examination of the ultimate fate of injected fluids and the impact on the local injection forma- tion hydrodynamics would help assuage any concerns surrounding the continued use of these disposal wells. The amount of flowback water collected from an individual well is highly variable. Table 1 reflects this fact, as can be seen when com- paring the volume of water utilized with the volume of flowback collected. As shown in Table 3, high-volume hydraulic fracturing flowback waters currently make up less than 1% of the annual brine disposal volumes (compared to 2011 cumulative disposal volumes). This percentage could change rapidly if there is an accelerated rate of application of high-volume hydraulic fracturing in the state, as was the case for Pennsylvania116 . TABLE 3: Brine Disposal Wells (BDW) in Michigan Total operational BDW 712 Total annual volume of injected brine (106 gal)* 4200 Total volume flowback water from all high-volume hydraulic fracturing wells (106 gal) 6.3 Flowback as % of total brine injection 0.15 *Cumulative 2011 BDW injection data, from MDEQ23 .
    • 19 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 2.5.3 Surface contamination (spills, improper disposal) Surface contamination likely carries the greatest risk for negative water quality impacts associated with hydraulic fracturing due to the proximity to potable water resources when handling these waste fluids at the surface117 . However, since in Michigan all flow- back is disposed of via deep-well injection and it is not allowed to sit in open pits, the risk of this type of contamination will be lower than in other states without such disposal opportunities and regu- lations. The one exception to deep well brine disposal in Michigan is the spreading of produced oilfield brines on dirt roads to control dust. In a probability bounds analysis by Rozell and Reaven117 , the risk of potable water contamination associated with hydraulic fracturing waste water disposal was found to be several orders of magnitude larger than contamination from other pathways such as contami- nant migration through fracture networks. This study was specif- ically examining the risk of water contamination associated with hydraulic fracturing activities in the Marcellus shale. Handling of waste and production fluids from hydraulically fractured wells in Pennsylvania has been a continuing challenge for natural gas pro- ducers. This is primarily due to the fact that the state of Pennsylvania has only five operating brine disposal wells49 , of which only two are commercial wells that accept brine from more than one company (all others are privately owned and operated). This means that producers in Pennsylvania must find alternative solutions to han- dle flowback water at the surface. Possible options include on-site treatment and reuse or trucking flowback fluids to a site capable of disposing of these fluids (e.g. Ohio brine disposal wells). Flowback waters are no longer accepted at municipal water treatment plants because these plants are not equipped to handle the high salinity brines. Past attempts to treat flowback brines in Pennsylvania have led to contamination of surface waters, prompting the discour- agement of this practice116,118 . Recently, increased application of on-site water recycling and reuse has lessened the surface water handling challenges faced by operators in Pennsylvania119 . 2.6 Common challenges One challenge related to assessing the impact of hydraulic fractur- ing on Michigan’s water resources is the potentially long timeframe in which the effects of such activity may be recognized. Although a detailed analysis of fluid residence times within the Michigan Basin has not been conducted in this report, the most active groundwa- ter flow regime likely exists in the shallow Quaternary glacial aqui- fers with minimal interaction with deeper saline formation waters37 . Additionally, vertical migration of fluids beyond the production reservoirs may also be a slow process even in the presence of flow pathways because of a general lack of a pressure gradient driving force. Hydraulic fracturing of oil and gas wells has taken place in Michigan for over half a decade without any reported contamination issues. This observation lends credibility to sound industry practices and state regulations within the state of Michigan. However, there may be an opportunity to better characterize subsurface flow regimes in order to build Michigan-specific basin-scale models to investi- gate fluid residence times and predict the impact on local aquifer hydrodynamics due to brine disposal. This type of modeling effort might serve to better determine the ultimate fate of flowback waters injected into brine disposal wells. One key challenge that is non-unique to Michigan is the need to determine what chemicals or chemical signatures to test for when investigating migration of fluids beyond the target reservoir. Methane isotopic signatures have been used to evaluate methane origin when testing water wells in Pennsylvania50 , and stable iso- topes such as those of Sr have been used to determine the origin of saline fluids118 . Studies have also investigated the migration of native brines into potable water reservoirs61 and within flowback water97 . The study by Halusczak et al.97 identified Ra within flow- back waters at similar levels to that which would be expected of native formation brines. Advanced analytical techniques (e.g. LC-MS) may be required to identify the degradation byproducts of hydraulic fracturing fluid chemical additives that may accompany leakage of fluids in relation to hydraulic fracturing practices. This is an emerging area of research and will require a concerted effort by industry and the academic research community to identify new tracers to test for when assessing potential cases of potable water contamination. 3.0 PRIORITIZED DIRECTIONS FOR PHASE 2 T his last section of the report lists several research areas and topics that could be addressed during the second phase of the integrated assessment. Beyond simply addressing data gaps, it would be beneficial to foster opportunities for improved knowledge transfer and commu- nication among relevant stakeholders. 3.1 Opportunities for additional data collection 3.1.1 Establish baseline water quality data In order to assess whether there are any water quality impacts associated with hydraulic fracturing in Michigan, baseline water quality data must be gathered from all potentially impacted reser- voirs (subsurface drinking water, surface waters). Existing data (e.g. USGS water quality data120 ) could be gathered and aggregated in an effort to establish where specific data gaps exist and prioritize locations for additional water sampling. Implicit within this effort is an assumption that water quality impacts will be detectable if they
    • 20 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 occur—this requires the establishment of water quality param- eters to monitor and determination of chemical tracers explicitly linked to hydraulic fracturing practices. Example parameters might include pH, turbidity, electrical conductance (measure of bulk salin- ity), and analysis of specific analytes such as Cl- , Sr2+ , Br- and Ba2+ 121 . The Shale Network (www.shalenetwork.org) provides a great example of such an effort to establish a database of baseline water quality data in the Appalachian Basin. The Shale Network seeks to develop a comprehensive water quality database for the region of Marcellus shale development through engaging community volunteers to gather new water quality data and aggregating exist- ing data from government, industry, and academic studies122 . All of this data is then made publically available through the CUAHSI HydroDesktop server. 3.1.2 Evaluate the impact of hydraulic fracturing chemicals on the release and transport of toxic metals and naturally occurring radionuclides While flowback fluids are expected to be contaminated to some degree, e.g., it is known that these fluids will be highly saline and contain elevated levels of heavy metals and radioactive elements, in addition the original chemical additives78,97,99,123–125 , their poten- tial to be leached as a function of the shale mineralogy, reservoir conditions (pH, pe, T and P), and chemical additives present in hydraulic fracturing fluid formulations has not yet been deter- mined. Current efforts to monitor flowback and produced water chemistry in other states have focused on characterization of the fluids returning to the surface via standard analytical methods such as ICP-MS, ICP-AES and IC61,87,99,118,125–127 . Two main conclusions can be drawn from analysis of flowback brines in other states: (1) flowback fluid salinity and concentration of trace inorganic contam- inants (e.g., Ba, Ra) increases as a function of time; (2) flowback fluids represent a mixture of injected waters and native brines. One area that has received limited attention is the contribution of the water-rock interactions in controlling flowback fluid chemistry. Prediction of the chemical evolution of remnant hydraulic fracturing fluids or the chemistry of flowback water returning to the surface will be improved if the hydraulic fracturing fluid-shale geochemical reactions occurring in the subsurface are better characterized. It is therefore important to assess whether hydraulic fracturing fluids may enhance the leaching and mobilization of naturally occurring trace elements that may be associated with adverse water quality impacts. If water-rock interactions are found to contribute to the increased toxic metal and radionuclide concentrations in flowback fluids, then it may also be beneficial to determine specific geo- chemical and mineralogical characteristics of potential reservoir formations that may lead to elevated levels of these constituents. 3.1.3 Monitor fracture propagation Work to implement standard measurement techniques (e.g. microseismic) for evaluating the extent and direction of major fracture networks during hydraulic fracturing. Gather data from Michigan well completions similar to that presented by Fisher and Warpinski58 . Industry experience may suggest that shallow Antrim fracturing completions occur only horizontally (so called ‘pancake fracs’), but collection of data to verify this would strengthen this claim and better elucidate whether fracture propagation beyond the reservoir is an area of concern for overlying aquifer water quality. 3.1.4 Conduct modeling studies to assess subsurface flow, fluid residence times, and leakage risk Assessment of leakage risk up existing wells will require some stan- dardized means for structural integrity analysis. The existing efforts by the MDEQ to identify abandoned or improperly sealed wells demonstrate a commitment to addressing leakage risks related to existing wellbores. Additional work on leakage risk assessment and subsurface migration of hydraulic fracturing fluids would be benefi- cial to understanding the likelihood for fluid migration beyond the target reservoir. Furthermore, it may be worthwhile to take a closer examination of local perturbations of hydraulic gradients associ- ated with brine disposal wells to determine whether brine injection may be pushing fluid beyond the target injection reservoir. 3.2 Reevaluate current regulatory definition of ‘produced water’ The key question here is whether flowback fluids should continue to be classified as ‘produced water’? Although flowback water associated with high-volume hydraulic fracturing completions cur- rently only makes up a small fraction of the total water produced from oil and natural gas extraction within the state of Michigan, treating it as equivalent to produced brine may be an area that could be reevaluated. It is true that a portion of the flowback fluids is native formation brines97 , but it may also be enriched in other trace contaminants at higher levels than are typically found in produced brines, such as elevated levels of Ba87 . In addition to containing potentially higher concentrations of toxic metals and radionuclides, many of the chemical additives will return to the surface with the flowback fluid78 . As a first step, it may be advisable to analyze the flowback water chemistry and compare it with that of the produced brine from older wells nearby. Building a database of native brine chemistry, for instance by aggregating currently avail- able brine data, would also support this effort.
    • 21 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 3.3 Facilitate improved dialogue and knowledge transfer among stakeholders 3.3.1 Open dialogue with industry One obvious shortcoming of this report is the lack of clairvoyance in assessing the potential growth in the application of high-volume hydraulic fracturing within the Michigan Basin. This statement is not meant to be tongue-in-cheek, but instead reflects the author’s lack of knowledge regarding industry expectations and plans for future drilling operations in the state of Michigan. Industry players make significant investments to evaluate and develop new resource plays. This knowledge is of course not shared publically nor is it complete. That is, there is always risk that an anticipated reservoir turns out to be less productive than originally hoped. If growth in the application of high-volume hydraulic fracturing can be antic- ipated, then efforts investigating the potential for adverse envi- ronmental impacts could be focused toward the topics or locales of greatest importance. Establishing an open dialogue between industry stakeholders, state-level regulators, and the research community will help guide research efforts moving forward. 3.3.2 Knowledge sharing Phase 2 of this integrated assessment should leverage the knowl- edge and experience of seasoned experts within the fields of petroleum engineering and sedimentary/structural geology. Those with the most practical experience are often members of the oil and gas industry. Although developing reports on fundamental topics related to hydraulic fracturing within the Michigan Basin, such as that provided here, represents a good first step toward addressing this complex issue, coordinated involvement of a diverse group of experts will be essential to fully address many of the existing data and knowledge gaps. One suggestion might be to hold a Michigan-centric research colloquium bringing together experts from across the many sub disciplines of relevant fields (e.g. geol- ogy, engineering) with interests in and knowledge of the Michigan Basin. Such a meeting would allow for knowledge sharing and consolidation of existing data, thereby enabling a more thorough discussion on the topic of Michigan’s geology within the context of hydraulic fracturing. 3.3.3 Need to justify scope of assessment This report has taken a broad scope when addressing the topic of hydraulic fracturing and has thus included low-volume hydraulic fracturing completions within the context of the report. As was described in the introduction, this was done to provide a more thorough review of formations within the Michigan Basin rele- vant to unconventional hydrocarbon production. What was not assessed in this report was whether there is or should be a dis- tinction between high-volume and what can be described as more ‘traditional’ lower-volume well stimulations. Making this distinction clear in the Phase 2 report and assessing the rationale for sepa- rating these two practices will make for a more robust discussion of the environmental costs and benefits associated with hydrau- lic fracturing in general. Advancements in slickwater chemical formations have aided the successful application of high-volume hydraulic fracturing completions. It is the recent application of high-volume hydraulic fracturing well completions, often coupled with directional drilling, that has drawn so much attention to the topic of hydraulic fracturing. By the same logic, one might argue that high-volume hydraulic fracturing is the only technology that should be investigated. Indeed, high-volume hydraulic fracturing is the primary focus of the technical reports compiled during the Phase I portion of this assessment. The general public may not understand the differences between the two practices or may sim- ply choose not to disaggregate one from the other. It is therefore incumbent upon the second phase of the integrated assessment to make a clear case for isolating high-volume hydraulic fracturing within the scope of its analysis. Doing so will provide a necessary transparency to the efforts of the integrated assessment and help educate the lay public as to whether, and why, a distinction should be drawn between the two practices. 1. Ernstoff AS, Ellis BR. Clearing the waters of the fracking debate. Michigan Journal of Sustainability. in press;1(1). 2. Arthur JD. The modern practices of hydraulic fracturing: A focus on Canadian resources. 2012. 3. Michigan Department of Environmental Quality. MDEQ High-volume Hydraulic Fracturing Permits and Aplications. 2012 December 20. Available from: www.michigan.gov/documents/deq/utica.collingwood_spreadsheet1_358438_7.pdf 4. State of Michigan Department of Environmental Quality. High volume hydraulic fracturing well completions. MDEQ; 2011. 5. Alexander J. Special Report: Modern gas rush stirs controversy in Michigan. The Center of Michigan [Internet]. 2011 July 13 [cited 2013 January 27]. Available from: thecenterformichigan.net/special-report-modern-gas-rush-stirs-controversy-in-michigan/ 6. Leighton MW. Interior cratonic basins: A record of regional tectonic influences van der Pluijm B, Catacosinos PA, editors. Geological Society of America Special Papers. 1996 January 1;308:77–93. 7. Howell PD, Pluijm BA van der. Early history of the Michigan basin: Subsidence and Appalachian tectonics. Geology. 1990 December 1;18(12):1195–1198. LITERATURE CITED
    • 22 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 8. Howell PD, Pluijm BA van der. Structural sequences and styles of subsidence in the Michigan basin. Geological Society of America Bulletin. 1999 July 1;111(7):974–991. 9. Dorr JA, Eschman DF. Geology of Michigan. University of Michigan Press: Ann Arbor, Michigan; 1970. 10. Bergquist SG. The glacial history and development of Michigan. [cited 2013 March 18]. Available from: www.michigan.gov/documents/deq/Glacial_History_Bergquist_306034_7.pdf 11. Medina CR, Rupp JA, Barnes DA. Effects of reduction in porosity and permeability with depth on storage capacity and injectivity in deep saline aquifers: A case study from the Mount Simon Sandstone aquifer. International Journal of Greenhouse Gas Control. 2011 January;5(1):146–156. 12. Catacosinos PA, Harrison WB III, Reynolds RF, Westjohn DB, Wollensak MS. Stratigraphic Lexicon for Michigan. MDEQ Geological Survey Division; 2001. 13. Swezey C, Geological Survey (U.S.). Regional stratigraphy and petroleum systems of the Michigan Basin, North America. 2008. 14. Lampe DC. Hydrogeologic framework of bedrock units and initial salinity distribution for a simulation of groundwater flow for the Lake Michigan Basin.; 2009 p. 49. 15. Vugrinovich R. Patterns of regional subsurface fluid movement in the Michigan basin. Michigan Geological Survey Division; 1986. 16. Michigan Department of Environmental Quality. Questions and answers about hydraulic fracturing in Michigan. [cited 2013 January 4]. Available from: www.michigan.gov/documents/deq/deq-FINAL-frack-QA_384089_7.pdf 17. Montgomery CT, Smith MM. Hydraulic Fracturing: History of an Enduring Technology. Journal of Petroleum Technology. 2010;62(12):26–32. 18. Nicot J-P, Scanlon BR. Water Use for Shale-Gas Production in Texas, U.S. Environmental Science & Technology. 2012 March 20;46(6):3580–3586. 19. U.S. Energy Information Administration. Annual Energy Outlook 2012.; 2012 p. 239. Available from: www.eia.gov/forecasts/aeo/pdf/0383(2012).pdf 20. U.S. Energy Information Administration. Technology drives natural gas production growth from shale gas formations. 2011 July 12 [cited 2013 March 12]. Available from: www.eia.gov/todayinenergy/detail.cfm?id=2170 21. Rahm D. Regulating hydraulic fracturing in shale gas plays: The case of Texas. Energy Policy. 2011 May;39(5):2974–2981. 22. Harrison W. Michigan Geologic Repository for Research and Education. (pers. comm.). 23. Michigan Department of Environmental Quality. MDEQ online oil and gas database. [cited 2012 December 5]. Available from: ww2.deq.state.mi.us/mir/ 24. Dolton GL, Quinn JC. An Initial Resource Assessment of the Upper Devonian Antrim Shale in the Michigan Basin. U.S. Geological Survey; 1996 p. 10. 25. Boyer C, Kieschnick J, Suarez-rivera R, Lewis R, Walter G. Producing gas from its source. Oilfield Review. 2006:36–49. 26. Ryder RT. Fracture patterns and their origin in the Upper Devonian Antrim Shale gas resevoir of the Michigan basin: A review. U.S. Geological Survey; 1996 p. 30. 27. Martini AM, Budai JM, Walter LM, Schoell M. Microbial generation of economic accumulations of methane within a shallow organic-rich shale. Nature. 1996 September 12;383(6596):155–158. 28. Martini AM, Walter LM, Ku TCW, Budai JM, McIntosh JC, Schoell M. Microbial production and modification of gases in sedimentary basins: A geochemical case study from a Devonian shale gas play, Michigan basin. AAPG Bulletin. 2003 August;87(8):1355–1375. 29. Fertl W, Chilingar G. Total Organic Carbon Content Determined From Well Logs. SPE Formation Evaluation. 1988;3(2). 30. Wood JR, Harrison WB III. Advanced characterization of fractured reservoirs in carbonate rocks: The Michigan basin. Michigan Technological University; 2002 p. 63. 31. Hayden J, Pursell D. The Barnett Shale: Visitors guide to the hottest gas play in the US. Pickering Energy Partners, Inc.; 2005. 32. McIntosh JC, Walter LM, Martini AM. Pleistocene recharge to midcontinent basins: Effects on salinity structure and microbial gas generation. Geochimica Et Cosmochimica Acta. 2002 May;66(10):1681–1700. 33. Curtis JB. Fractured shale-gas systems. Aapg Bulletin. 2002 November;86(11):1921–1938. 34. Cercone K. Thermal History of Michigan Basin. Aapg Bulletin-American Association of Petroleum Geologists. 1984;68(2):130–136. 35. Cercone KR, Pollack HN. Thermal maturity of the Michigan basin. 1991;256:1–11. 36. Martini AM, Walter LM, McIntosh JC. Identification of microbial and thermogenic gas components from Upper Devonian black shale cores, Illinois and Michigan basins. Aapg Bulletin. 2008 March;92(3):327–339. 37. McIntosh JC, Garven G, Hanor JS. Impacts of Pleistocene glaciation on large-scale groundwater flow and salinity in the Michigan Basin. Geofluids. 2011 February;11(1):18–33. 38. Person M, McIntosh J, Bense V, Remenda VH. Pleistocene hydrology of north America: The role of ice sheets in reorganizing groundwater flow systems. Reviews of Geophysics. 2007 September 21;45(3). 39. Wood JR, Quinlan WC. An Approach to Recover Hydrocarbons from Currently Off-Limit Areas of the Antrim Formation, MI Using Low-Impact Technologies. Michigan Technical University; 2008 p. 61.
    • 23 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 40. Harrison WBI. Facies, fabrics and organic geochemisty of the A-1 carbonate in Michigan. In: Mount Pleasant, MI; 2013. 41. Wilson T, Long D. Geochemistry and Isotope Chemistry of Ca-Na-Cl Brines in Silurian Strata, Michigan Basin, Usa. Applied Geochemistry. 1993 September;8(5):507–524. 42. Wilson T, Long D. Geochemistry and Isotope Chemistry of Michigan Basin Brines - Devonian Formations. Applied Geochemistry. 1993 January;8(1):81–100. 43. Hanor JS, McIntosh JC. Are secular variations in seawater chemistry reflected in the compositions of basinal brines? Journal of Geochemical Exploration. 2006;89(1-3):153–156. 44. Ma L, Castro MC, Hall CM, Walter LM. Cross-formational flow and salinity sources inferred from a combined study of helium concentrations, isotopic ratios, and major elements in the Marshall aquifer, southern Michigan. Geochemistry Geophysics Geosystems. 2005 October 8;6. 45. Vugrinovich R. Subsurface temperatures and surface heat flow in the Michigan Basin and their relationships to regional subsurface fluid movement. Marine and Petroleum Geology. 1989 February;6(1):60–70. 46. Stueber AM, Walter LM. Origin and Chemical Evolution of Formation Waters from Silurian-Devonian Strata in the Illinois Basin, Usa. Geochimica et Cosmochimica Acta. 1991 January 1;55:1. 47. McIntosh JC, Walter LM, Martini AM. Extensive microbial modification of formation water geochemistry: Case study from a Midcontinent sedimentary basin, United States. Geological Society of America Bulletin. 2004 June;116(5-6):743–759. 48. Michigan Department of Environmental Quality. Water Well Viewer [Internet]. [cited 2013 March 16]. Available from: wellviewer.rsgis.msu.edu 49. Vidic RD, Brantley SL, Vandenbossche JM, Yoxtheimer D, Abad JD. Impact of Shale Gas Development on Regional Water Quality. Science. 2013 May 17;340(6134). 50. Osborn SG, Vengosh A, Warner NR, Jackson RB. Methane contamination of drinking water accompanying gas-well drilling and hydraulic fracturing. Proceedings of the National Academy of Sciences of the United States of America. 2011 May 17;108(20):8172–8176. 51. DeGiulio DC, Wilkin RT, Miller C, Oberley G. Investigation of groundwater contamination near Pavillion, Wyoming. U.S. Environmental Protection Agency; 2011. 52. Molofsky LJ, Connor JA, Farhat SK, Wylie AS, Wagner T. Methane in Pennsylvania water wells unrelated to Marcellus shale fracturing. Oil & Gas Journal. 2011 December 5;109(19):54–67. 53. Davies RJ. Methane contamination of drinking water caused by hydraulic fracturing remains unproven. Proceedings of the National Academy of Sciences of the United States of America. 2011 October 25;108(43):E871–E871. 54. Schon SC. Hydraulic fracturing not responsible for methane migration. Proceedings of the National Academy of Sciences of the United States of America. 2011 September 13;108(37):E664–E664. 55. Saba T, Orzechowski M. Lack of data to support a relationship between methane contamination of drinking water wells and hydraulic fracturing. Proceedings of the National Academy of Sciences of the United States of America. 2011 September 13;108(37):E663–E663. 56. Davies RJ, Mathias SA, Moss J, Hustoft S, Newport L. Hydraulic fractures: How far can they go? Marine and Petroleum Geology. 2012 November;37(1):1–6. 57. Mahrer KD. A review and perspective on far-field hydraulic fracture geometry studies. Journal of Petroleum Science and Engineering. 1999 November;24(1):13–28. 58. Fisher K, Warpinski N. Hydraulic-Fracture-Height Growth: Real Data. Spe Production & Operations. 2012;27(1):8–19. 59. Harrison WBI. Chapter 10 - Hydrocarbon Resources. In: Schaetzl RJ, Darden JT, Brandt D, editors. Geography and Geology of Michigan. Upper Saddle River, NJ: Pearson Custom Publishers; 2009. 60. Weaver TR, Frape SK, Cherry JA. Recent cross-formational fluid flow and mixing in the shallow Michigan Basin. Geological Society of America Bulletin. 1995;107(6):697–707. 61. Warner NR, Jackson RB, Darrah TH, Osborn SG, Down A, Zhao K, White A, Vengosh A. Geochemical evidence for possible natural migration of Marcellus Formation brine to shallow aquifers in Pennsylvania. Proceedings of the National Academy of Sciences of the United States of America. 2012 July 24;109(30):11961–11966. 62. Engelder T. Capillary tension and imbibition sequester frack fluid in Marcellus gas shale. Proceedings of the National Academy of Sciences. 2012 December 26;109(52):E3625–E3625. 63. Holst TB, Foote GR. Joint orientation in Devonian rocks in the northern portion of the lower peninsula of Michigan. Geological Society of America Bulletin. 1981 February 1;92(2):85–93. 64. Nordbotten JM, Celia MA, Bachu S, Dahle HK. Semianalytical solution for CO2 leakage through an abandoned well. Environmental Science & Technology. 2005 January 15;39(2):602–611.
    • 24 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 65. Watson TL, Bachu S. Evaluation of the Potential for Gas and CO2 Leakage Along Wellbores. Spe Drilling & Completion. 2009 March;24(1):115–126. 66. Michigan Department of Environmental Quality. Michigan’s Oil and Gas Regulations. 2006 [cited 2013 February 14]. Available from: www.michigan.gov/documents/deq/ogs-oilandgas-regs_263032_7.pdf 67. National Petroleum Council. Plugging and abandonment of oil and gas wells.; 2011 p. 21. Available from: www.npc.org/Prudent_Development-Topic_Papers/2-25_Well_Plugging_and_Abandonment_Paper.pdf 68. Keller S, Crook R, Haut R, Kulakofsky D. Deviated-Wellbore Cementing .1. Problems. Journal of Petroleum Technology. 1987 August;39(8):955–960. 69. Myers T. Potential Contaminant Pathways from Hydraulically Fractured Shale to Aquifers. Ground Water. 2012 December;50(6):872–882. 70. Saiers JE, Barth E. Potential Contaminant Pathways from Hydraulically Fractured Shale Aquifers. Ground Water. 2012;50(6):826–828. 71. Cohen HA, Parratt T, Andrews CB. Potential Contaminant Pathways from Hydraulically Fractured Shale to Aquifers. Groundwater. 2013;51(3):317–319. 72. Celia MA, Nordbotten JM. Practical Modeling Approaches for Geological Storage of Carbon Dioxide. Ground Water. 2009 October;47(5):627–638. 73. Gaus I. Role and impact of CO2–rock interactions during CO2 storage in sedimentary rocks. International Journal of Greenhouse Gas Control. 2010 January;4(1):73–89. 74. Lewicki JL, Birkholzer J, Tsang C-F. Natural and industrial analogues for leakage of CO2 from storage reservoirs: identification of features, events, and processes and lessons learned. Environmental Geology. 2007 April;52(3):457–467. 75. Shukla R, Ranjith P, Haque A, Choi X. A review of studies on CO2 sequestration and caprock integrity. Fuel. 2010 October;89(10):2651–2664. 76. U.S. Environmental Protection Agency. Great Lakes Facts and Statistics. 2012 July 5 [cited 2013 January 3]. Available from: www.epa.gov/greatlakes/lakestats.html 77. Nathan VR. Drinking water in Michigan: source, quality, and contaminants. Journal of water and health. 2006;4 Suppl 1:67–73. 78. Arthur JD, Bohm B, Layne M, Cornue D. Hydraulic fracturing considerations for natural gas wells of the Marcellus shale. In: Cincinnati, OH: The Groundwater Protection Council; 2008. pp. 1–16. 79. Ground Water Protection Council, ALL Consulting. Modern shale gas development in the United States: A primer.; 2009 p. pp. 95. 80. U.S. Environmental Protection Agency. Study of the potential impacts of hydraulic fracturing on drinking water resources: Progress Report.; 2012 p. 262. 81. Ground Water Protection Council, Interstate Oil and Gas Compact Commission. FracFocus. FracFocus [Internet]. Available from: fracfocus.org 82. Fisher RS. Geologic and Geochemical Controls on Naturally Occurring Radioactive Materials (NORM) in Produced Water from Oil, Gas, and Geothermal Operations. Environmental Geosciences. 1998;5(3):139–150. 83. Leventhal J. Pyrolysis-Gas Chromatography-Mass Spectrometry to Characterize Organic-Matter and Its Relationship to Uranium Content of Appalachian Devonian Black Shales. Geochimica Et Cosmochimica Acta. 1981;45(6):883–889. 84. Brumsack HJ. The trace metal content of recent organic carbon-rich sediments: Implications for Cretaceous black shale formation. Palaeogeography Palaeoclimatology Palaeoecology. 2006 March 22;232(2-4):344–361. 85. Vine JD, Tourtelo EB. Geochemistry of Black Shale Deposits - a Summary Report. Economic Geology. 1970;65(3):253–&. 86. Smith KP, Blunt DL, Williams GP, Tebes CL. Radiological dose assessment related to management of naturally occuring radioactive materials generated by the petroleum industry. Department of Energy; 1996 p. 69. 87. Blauch M, Myers R, Moore T, Lipinski B, Houston N. Marcellus Shale Post-Frac Flowback Waters - Where is All the Salt Coming from and What are the Implications? In: Society of Petroleum Engineers; 2009. 88. Gregory KB, Vidic RD, Dzombak DA. Water Management Challenges Associated with the Production of Shale Gas by Hydraulic Fracturing. Elements. 2011 June 1;7(3):181–186. 89. Pritz ME, Kirby CS. Geochemical investigations of Marcellus Shale natural gas hydrofracturing waters. In: Abstracts with Programs. Vol. 42 (1). ; 2010. p. 121. 90. McIntosh JC, Walter LM. Volumetrically significant recharge of Pleistocene glacial meltwaters into epicratonic basins: Constraints imposed by solute mass balances. Chemical Geology. 2005 November 5;222(3-4):292–309. 91. Smedley PL, Kinniburgh DG. A review of the source, behaviour and distribution of arsenic in natural waters. Applied Geochemistry. 2002 May;17(5):517–568. 92. Peng B, Song Z, Tu X, Xiao M, Wu F, Lv H. Release of heavy metals during weathering of the Lower Cambrian Black Shales in western Hunan, China. Environmental Geology. 2004 June 1;45(8):1137–1147. 93. Woo NC, Choi MJ, Lee KS. Assessment of Groundwater Quality and Contamination from Uranium-Bearing Black Shale in Goesan–Boeun Areas, Korea. Environmental Geochemistry and Health. 2002 September 1;24(3):264–273.
    • 25 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 94. Loukola-Ruskeeniemi K, Kantola M, Halonen T, Seppänen K, Henttonen P, Kallio E, Kurki P, Savolainen H. Mercury-bearing black shales and human Hg intake in eastern Finland: impact and mechanisms. Environmental Geology. 2003 January 1;43(3):283–297. 95. Lee J-S, Chon H-T, Kim K-W. Migration and dispersion of trace elements in the rock-soil-plant system in areas underlain by black shales and slates of the Okchon Zone, Korea. Journal of Geochemical Exploration. 1998;65(1):61–78. 96. Swanson VE. Oil yeild and uranium content of black shales.; 1960. 97. Haluszczak LO, Rose AW, Kump LR. Geochemical evaluation of flowback brine from Marcellus gas wells in Pennsylvania, USA. Applied Geochemistry. 2013 January;28(0):55–61. 98. Lustgarten A. Is New York’s Marcellus Shale Too Hot to Handle? ProPublica [Internet]. 2009 November 9. Available from: www.propublica.org/article/is-the-marcellus-shale-too-hot-to-handle-1109 99. NYSDEC. Draft Supplemental Generic Environmental Impact Statement On The Oil, Gas and Solution Mining Regulatory Program, Well Permit Issuance for Horizontal Drilling And High-Volume Hydraulic Fracturing to Develop the Marcellus Shale and Other Low-Permeability Gas Reservoirs.; 2009. 100. Bi Y, Hyun SP, Kukkadapu RK, Hayes KF. Oxidative dissolution of UO2 in a simulated groundwater containing synthetic nanocrystalline mackinawite. Geochimica et Cosmochimica Acta. 2013 February 1;102:175–190. 101. Pierce EM, Icenhower JP, Serne RJ, Catalano JG. Experimental determination of UO2(cr) dissolution kinetics: Effects of solution saturation state and pH. Journal of Nuclear Materials. 2005 October 15;345(2-3):206–218. 102. De Pablo J, Casas I, Gimenez J, Clarens F, Duro L, Bruno J. The oxidative dissolution mechanism of uranium dioxide. The effect of pH and oxygen partial pressure. In: Oversby VM, Werme LO, editors. Scientific Basis for Nuclear Waste Management Xxvii. Vol. 807. Warrendale: Materials Research Society; 2004. pp. 83–88. 103. Guillaumont R, Agency ONE, Mompean FJ. Update on the chemical thermodynamics of uranium, neptunium, plutonium, americium and technetium. Elsevier; 2003. 104. Dong W, Brooks SC. Determination of the formation constants of ternary complexes of uranyl and carbonate with alkaline earth metals (Mg2+, Ca2+, Sr2+, and Ba2+) using anion exchange method. Environmental Science & Technology. 2006 August 1;40(15):4689–4695. 105. Stewart BD, Mayes MA, Fendorf S. Impact of Uranyl-Calcium-Carbonato Complexes on Uranium(VI) Adsorption to Synthetic and Natural Sediments. Environmental Science & Technology. 2010 February 1;44(3):928–934. 106. Suess E, Planer-Friedrich B. Thioarsenate formation upon dissolution of orpiment and arsenopyrite. Chemosphere. 2012 November;89(11):1390–1398. 107. Jeong HY, Han Y-S, Park SW, Hayes KF. Aerobic oxidation of mackinawite (FeS) and its environmental implication for arsenic mobilization. Geochimica Et Cosmochimica Acta. 2010 June 1;74(11):3182–3198. 108. Paikaray S. Environmental hazards of arsenic associated with black shales: a review on geochemistry, enrichment and leaching mechanism. Reviews in Environmental Science and Bio-Technology. 2012 September;11(3):289–303. 109. Hayes KF, Chen C, McAvoy DC. Quaternary Ammonium Surfactant Effects on Sorption of Trace Metals onto Quartz and Aluminosilicates. Soil Science Society of America Journal. 1995 4/01;59(2):380–387. 110. Patel A. Design and Development of Quaternary Amine Compounds: Shale Inhibition With Improved Environmental Profile. In: Society of Petroleum Engineers; 2009. Available from: www.onepetro.org/mslib/servlet/onepetropreview?id=SPE-121737-MS 111. Cooley H, Donnelly K. Hydraulic fracturing and water resources: separating the frack from the fiction. 2012. 112. Veil J, Clark C. Produced water volume estimates and management practices. SPE Production & Operations. 2011;26(3):234–239. 113. U.S. Environmental Protection Agency. Underground injection wells in region 5. [cited 2013 March 10]. Available from: www.epa.gov/r5water/uic/r5uicwells.htm#michigan 114. Hopey D. Quakes in Ohio tied to area shale operations. Pittsburgh Post-Gazette [Internet]. 2012 March 10 [cited 2013 June 1]. Available from: www.post-gazette.com/stories/local/marcellusshale/quakes-in-ohio-tied-to-area-shale-operations-221762/ 115. Keranen KM, Savage HM, Abers GA, Cochran ES. Potentially induced earthquakes in Oklahoma, USA: Links between wastewater injection and the 2011 Mw 5.7 earthquake sequence. Geology. 2013 March 26. 116. Lutz BD, Lewis AN, Doyle MW. Generation, transport, and disposal of wastewater associated with Marcellus Shale gas development. Water Resources Research. 2013;49:1–10. 117. Rozell DJ, Reaven SJ. Water Pollution Risk Associated with Natural Gas Extraction from the Marcellus Shale. Risk Analysis. 2012 August;32(8):1382–1393. 118. Chapman EC, Capo RC, Stewart BW, Kirby CS, Hammack RW, Schroeder KT, Edenborn HM. Geochemical and Strontium Isotope Characterization of Produced Waters from Marcellus Shale Natural Gas Extraction. Environmental Science & Technology. 2012 March 20;46(6):3545–3553.
    • 26 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: GEOLOGY/HYDROGEOLOGY TECHNICAL REPORT, SEPTEMBER 2013 119. Maloney KO, Yoxtheimer DD. Production and Disposal of Waste Materials from Gas and Oil Extraction from the Marcellus Shale Play in Pennsylvania. Environmental Practice. 2012;14(4):278–287. 120. U.S. Geological Survey. Michigan Water Science Center: Water Quality Data. 2012 [cited 2013 June 1]. Available from: mi.water.usgs.gov/qwdata.php 121. Olmstead SM, Muehlenbachs LA, Shih J-S, Chu Z, Krupnick AJ. Shale gas development impacts on surface water quality in Pennsylvania. Proceedings of the National Academy of Sciences. 2013 March 26;110(13):4962–4967. 122. Brantley SL, Wilderman C, Abad J. Workshop discusses database for Marcellus water issues. Eos, Transactions American Geophysical Union. 2012;93(34):328–328. 123. U.S. Environmental Protection Agency. Plan to study the potential impacts of hydraulic fracturing on drinking water resources.; 2011 p. 174. 124. Simon JA, Fleming ME. Editor’s perspective—Shale gas development: Environmental issues and opportunities. Remediation Journal. 2011;21(4):1–10. 125. Barbot E, Vidic NS, Gregory KB, Vidic RD. Spatial and Temporal Correlation of Water Quality Parameters of Produced Waters from Devonian-Age Shale following Hydraulic Fracturing. Environmental Science & Technology. 2013 March 19;47(6):2562–2569. 126. Boyer EW, Swistock BR, Clark J, Madden M, Rizzo DE. The impact of Marcellus gas drilling on rural drinking water supplies. The Center for Rural Pensylvania; 2012 p. 26. 127. Hayes T. Sampling and analysis of water streams associated with the development of marcellus shale gas. Marcellus Shale Coalition; 2009 p. 249. Acknowledgements This publication is a result of work sponsored by the University of Michigan. BRE also acknowledges partial funding support from the Michigan Society of Fellows and from the National Science Foundation under Grant No. 1214416.
    • © 2013 BY THE REGENTS OF THE UNIVERSITY OF MICHIGAN MARK J. BERNSTEIN, ANN ARBOR JULIA DONOVAN DARLOW, ANN ARBOR LAURENCE B. DEITCH, BLOOMFIELD HILLS SHAUNA RYDER DIGGS, GROSSE POINTE DENISE ILITCH, BINGHAM FARMS ANDREA FISCHER NEWMAN, ANN ARBOR ANDREW C. RICHNER, GROSSE POINTE PARK KATHERINE E. WHITE, ANN ARBOR MARY SUE COLEMAN, EX OFFICIO Please print sparingly and recycle
    • Environment/ Ecology Technical Report H Y D R A U L I C F R A C T U R I N G I N T H E S T A T E O F M I C H I G A N
    • Participating University of Michigan Units Graham Sustainability Institute Erb Institute for Global Sustainable Enterprise Risk Science Center University of Michigan Energy Institute ABOUT THIS REPORT This document is one of the seven technical reports com- pleted for the Hydraulic Fracturing in Michigan Integrated Assessment conducted by the University of Michigan. During the initial phase of the project, seven faculty-led and student-staffed teams focused on the following topics: Technology, Geology/ Hydrogeology, Environment/Ecology, Human Health, Policy/ Law, Economics, and Public Perceptions. These reports were prepared to provide a solid foundation of information on the topic for decision makers and stakeholders and to help inform the Integrated Assessment, which will focus on the analysis of policy options. The reports were informed by comments from (but do not necessarily reflect the views of) the Integrated Assessment Steering Committee, expert peer reviewers, and numerous public comments. Upon completion of the peer review process, final decisions regarding the content of the reports were deter- mined by the faculty authors in consultation with the peer review editor. These reports should not be characterized or cited as final products of the Integrated Assessment. The reports cover a broad range of topics related to hydraulic fracturing in Michigan. In some cases, the authors determined that a general discussion of oil and gas development is important to provide a framing for a more specific discussion of hydraulic fracturing. The reports address common hydraulic fracturing (HF) as meaning use of hydraulic fracturing methods regardless of well depth, fluid volume, or orientation of the well (whether vertical, directional, or horizontal). HF has been used in thousands of wells throughout Michigan over the past several decades. Most of those wells have been shallower, vertical wells using approxi- mately 50,000 gallons of water; however, some have been deeper and some have been directional or horizontal wells. The reports also address the relatively newer high volume hydraulic fracturing (HVHF) methods typically used in conjunction with directional or horizontal drilling. An HVHF well is defined by the State of Michigan as one that is intended to use more than 100,000 gallons of hydraulic fracturing fluid. The reports indicate if the text is addressing oil and gas development in general, HF, or HVHF. Finally, material in the technical reports should be understood as providing a thorough hazard identification for hydraulic fracturing, and when appropriate, a prioritization according to likelihood of occurrence. The reports do not provide a scientific risk assess- ment for aspects of hydraulic fracturing.
    • GRAHAM SUSTAINABILITY INSTITUTE INTEGRATED ASSESSMENT REPORT SERIES VOLUME II, REPORT 4 HYDRAULIC FRACTURING IN THE STATE OF MICHIGAN Environment/Ecology Technical Report SEPTEMBER 3, 2013 Faculty Leads G. ALLEN BURTON SCHOOL OF NATURAL RESOURCES & ENVIRONMENT, EARTH & ENVIRONMENTAL SCIENCES DEPARTMENT KNUTE J. NADELHOFFER ECOLOGY & EVOLUTIONARY BIOLOGY Research Assistant KATHLEEN PRESLEY EARTH & ENVIRONMENTAL SCIENCES DEPARTMENT TABLE OF CONTENTS 2 Executive Summary 2 1.0 Introduction 3 2.0 Status and Trends 8 3.0 Challenges and Opportunities 9 4.0 Prioritized Pathways for Phase 2 9 Literature Cited THIS PUBLICATION IS A RESULT OF WORK SPONSORED BY THE UNIVERSITY OF MICHIGAN Direct questions to grahaminstitute-ia@umich.edu
    • 2 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 EXECUTIVE SUMMARY A s hydraulic fracturing operations expand, we seek to scientifically assess the potential impacts of hydrau- lic fracturing operations on ecosystems of varying scales and compositions. Generally, the closer geo- graphical proximity of the “susceptible” ecosystem to a drilling site or a location of related industrial processes, the higher the risk of that ecosystem being impacted by the operation. Although the actual “hydraulic fracturing” process targets geologic forma- tions well below surface level, potential impacts of infrastructure development and drilling operations (including groundwater withdrawals and wastewater processing) associated with hydraulic fracturing on surface terrestrial and aquatic ecological systems are great. This review of potential ecological effects applies to high volume hydraulic fracturing (HVHF) and also shallow/low volume fracturing. Both types of fracturing operations have similar “foot- prints”, where the greatest potential for ecosystem impacts exists. This study is not a risk assessment but rather is identifying potential hazards associated with HVHF that may pose a risk to the environ- ment. This study is also not a comparison of HVHF to other energy and oil, gas or coal extraction technologies, which is beyond the scope of the University of Michigan study. Michigan’s dense, interconnected aquatic ecosystems (e.g., streams, rivers, lakes, inland and coastal wetlands) and the hypor- heic zones and aquifers with which they exchange water, chemicals, and organisms are of particular concern. Hydrologic connectivity of these aquatic networks to lowland and upland landscape features and associated plant, microbial and animal communities (includ- ing wildlife) can lead to impacts on terrestrial ecosystems as well. The landscape-scale connectivity, therefore, which is mediated by hydrologic flows across through watersheds and between surface and ground water bodies, can lead to impacts distant from, as well close to drilling sites. Building the necessary roads, product transportation lines, power grid, and water extraction systems, together with the siting of drilling equipment and increased truck traffic, produces varying site-specific environmental externalities. Potential effects include: increased erosion and sedimentation, increased risk of aquatic contamination from chemical spills or equipment runoff, habitat fragmentation and resulting impacts on aquatic and terrestrial organisms, loss of stream riparian zones, altered biogeochemical cycling, and reduction of surface and hyporheic waters available to aquatic communities due to lowering groundwater levels. In December 2012, the US Environmental Protection Agency (EPA) panel on the potential impacts of hydraulic fracturing on drinking water resources suggested that the impact of hydraulic fracturing on aquatic resources is heavily influenced by the proximity of the well site location to water resources (http://epa.gov/hfstudy/pdfs/ hf-report20121214.pdf ). Also of note is the EPA’s suggestion that the density of wells in a specific geographic region strongly cor- relates to the potential for degradation of a particular ecosystem. Michigan is fortunate to have a Wetland Protection Program and also a Water Withdrawal Assessment Tool (WWAT). These could allow for effective evaluations of potential ecological impacts from fracturing operations by considering their proximity and density in relation to sensitive and vulnerable wetlands and fisheries, such as trout streams. The focus of the WWAT is on long term ground- water withdrawal impacts to surface waters. However, questions have been raised about the ability of the tool to address short- term intensive withdrawals such as those associated with hydraulic fracturing operations and the need for periodic revisions to better account for important considerations such as streamflow, stream- flow gaging, and rare ecosystems1,2 . This tool, moreover, cannot assess the potential impacts of establishing the infrastructure and operations on habitat, wildlife, and nearby waters receiving site run- off. A surface water ecosystem relies upon a myriad of factors for its proper function. While the groundwater-surface water interchange is a key factor, other very important ecological considerations are: amount and timing of precipitation and runoff. For example, the water withdrawal tool will not measure potential changes in surface runoff patterns due to the clearing of land and road construction for fracturing operations. However, GIS-based modeling and site monitoring could allow for these potential impacts to be evaluated by ensuring proper siting and operational controls are established. 1.0 INTRODUCTION S hale oil and gas development, if not properly man- aged, could adversely affect water quality due to surface water and groundwater contamination as a result of 1) spills and releases of produced water, chem- icals, and drill cuttings, 2) erosion from ground disturbances, or 3) underground migration of gases and chemicals. Oil and gas development, whether conventional or from fracturing to extract shale oil and gas, can contribute to erosion, carrying varying loads of sediments and /or chemicals of concern pollutants into sur- face waters3–14 . Spills into surface waters can result from spills or releases of toxic chemicals and waste that occur as a result of tank ruptures, blowouts, equipment or impoundment failures, overfills, vandalism, accidents (including vehicle collisions), ground fires, or operational errors. For example, tanks storing toxic chemicals or hoses and pipes used to convey wastes to the tanks could leak, or impoundments containing wastes could overflow as a result of
    • 3 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 extensive rainfall3,4 . Wastewater impoundments are not allowed in Michigan for these operations. In addition to the hazards of the leaks of natural gas itself, the fluids used in the hydraulic fracturing fluids can be toxic in their own right. Hydraulic fracturing fluids are composed of proppants, gelling agents, solvents, and biocides. The proppants, gener- ally silica-based sand, are necessary to prop open the hydraulic fracturing cracks in the rock. In order for the proppants to seep into these cracks, a gel must first be formed that is then removed using solvents. The biocide added to hydraulic fracturing chemi- cals keeps the cracks from being clogged by bacterial growth and biofilm formation. Around 30 different chemicals are still used for hydraulic fracturing of tight gas and conventional gas reservoirs, but result in a mass fraction of approximately 0.5% chemicals in hydraulic fracturing fluid15 . The chemical additives in fracturing fluid, if not properly handled, pose risks to water quality if they come into contact with surface water or groundwater. Some additives used in fracturing fluid are known to be toxic, but toxicological data are limited for other additives, Michigan law does not require disclosure of all additives (See Table 4 in the Technical Report on Policy & Law), and not all end products of reacting additives injected in shale formations are known. 2.0 STATUS AND TRENDS 2.1 Factors of Potential Concern The industrial nature of shale oil and gas development requires operators to undertake a number of earth-disturbing activities, such as clearing, grading, and excavating land to create a pad to support the drilling equipment3 , or other necessary industrial process materials. One specific example of this equipment used in fracturing operations is the implementation of small rigs to “flare” excess gas into the atmosphere. While this practice is declining, it still occurs. These are particularly utilized when the market demand for gas is low, contributing to low natural gas prices, and thus operators’ hesitance to spend capital selling gas. If necessary, operators may also construct access roads to transport equipment and other materials to the site. In general, these pads and roads are not paved, thus increasing the potential for sediment erosion on and off location9,10,16,17 . If sufficient erosion controls to contain or divert sediment away from surface water are not established then surfaces exposed to precipitation and runoff could carry sediment and other harmful pollutants into nearby rivers, lakes, and streams. Sediment clouds water, decreases photosynthetic activity, and destroys organisms and their habitat3 . In addition, nutrients and other chemicals tend to sorb to sediments where they accumulate and can contaminate overlying waters and biota . Construction of the well pad, access road, and other drilling facil- ities requires substantial truck traffic. Up to 96% of the fleet of on-road and off-road vehicles employed in a particular hydraulic fracturing operation are diesel trucks and trailers; however, many of these trucks are being converted to natural gas resulting in reduced emissions. These trailers function to transport equipment and chemicals on-site, transport product or waste by-products off-site, and power the massive fracturing operation itself16 . The increased traffic creates a risk to air quality as engine exhaust that contains air pollutants such as nitrogen oxides (which react to form ground-level ozone) and particulate matter that are of concern to human, environmental, and ecological health3 . According to the EPA’s National and Environmental Effects Research Laboratory, defining ecosystem health can be a nebulous effort. However, it can be equated to human health, or rather, the environment in which a human would be healthy. The Research Laboratory points out that, “Most people envision instinctively a ‘healthy’ ecosystem as being pristine, or at least minimally altered by human action19 .” Thus, an ecosystem with extended human impacts from industrial processes could be an unhealthy ecosys- tem. The vast infrastructure requirements for fracturing operations —from individual well bores, to pipeline networks—imply enor- mous industrial processes, and consequent significant impacts on ecosystems. Particular ways in which ecosystems may be affected are discussed later in this paper. Increased networks of pipelines must be constructed to move product to storage and/or processing facilities20,21 . Much of Michigan’s shale play activity is in the northern region of the lower peninsula (Figure 1). This should be kept in mind while considering the potential ecosystems at risk. Hydraulic fracturing chemicals are transported to drilling sites in tank trucks, and are stored and mixed at the sites. More than 750 distinct chemicals, ranging from benign to toxic, have been used in hydraulic fracturing solutions; however, usually only several are used in each operation. Although these additives are approxi- mately 0.5 % by volume of the total fracturing fluid, hydraulic frac- turing is a water-intensive process and at least 13,000 gallons of chemicals would be used for a typical 2.6 million gallon hydraulic fracturing project. Chemical and wastewater transport vehicles can potentially be involved in traffic accidents, and it is estimated that a 30 ton tank truck will have an accident every 207,000 miles. And while this does not necessarily mean that chemical emissions will occur each time, they can potentially occur nonetheless. Moreover, truck accidents that occur on public roads could result in chemicals
    • 4 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 KALKASKAKALKASKA HIGHVOLUMEHYDRAULICFRACTURING 0 2010 Miles AS OF 5/09/2013 HIGH VOLUME HYDRAULICALLY FRACTURED WELL COMPLETIONS ARE DEFINED IN SUPERVISOR OF WELL INSTRUCTION 1-2011 AS A 'WELL COMPLETION OPERATION THAT IS INTENDED TO USE A TOTAL OF MORE THAN 100,000 GALLONS OF HYDRAULIC FRACTURING FLUID'. WE MADE ALL EFFORTS TO TRACE BACK THE WELL COMPLETION RECORDS THRU 2008 TO COMPLILE THIS MAP AND LIST. THIS INFORMATION PROVIDED HEREIN IS ACCURATE TO THE BEST OF OUR KNOWLEDGE AND IS SUBJECT TO CHANGE ON A REGULAR BASIS, WITHOUT NOTICE. WHILE THE DEPARTMENT OF ENVIRONMENTAL QUALITY - OFFICE OF OIL, GAS, AND MINERALS (DEQ-OOGM) MAKES EVERY EFFORT TO PROVIDE USEFUL AND ACCURATE INFORMATION, WE DO NOT WARRANT THE INFORMATION TO BE AUTHORITATIVE, COMPLETE, FACTUAL, OR TIMELY. IT IS SUGGESTED THAT THIS INFORMATION BE COMBINED WITH SECONDARY SOURCES AS A MEANS OF VERIFICATION. INFORMATION IS PROVIDED "AS IS" AND AN "AS AVAILABLE" BASIS. THE STATE OF MICHIGAN DISCLAIMS ANY LIABILITY, LOSS, INJURY, OR DAMAGE INCURRED AS A CONSEQUENCE, DIRECTLY OR INDIRECTLY, RESULTING FROM THE USE, INTERPRETATION, AND APPLICATION OF ANY OF THIS INFORMATION. Legend !. !. ISSUED ACTIVE PERMITS (52) PENDING ACTIVE APPLICATIONS (17) !. !. !. !. !. !.!. !. !. PN60133 VERTICAL PN60170 HORIZONTAL PN60328 VERTICAL PN60305 VERTICAL PN60183 VERTICAL PN60360 HORIZONTAL PN60138 VERTICAL PN60198 HORIZONTAL PN60161 HORIZONTAL PN59919 VERTICAL PN59979 HORIZONTAL PN59449 VERTICAL PN60137 VERTICAL A110068 HORIZONTAL !. PN60562 VERTICAL PN60606 HORIZONTAL A120046 VERTICAL PN60546 HORIZONTAL PN60545 HORIZONTAL !. PN60685 VERTICAL PN60686 HORIZONTAL !.PN60559 VERTICAL PN60560 HORIZONTAL !. A130031 HORIZONTAL !.PN60579 HORIZONTAL !. !. PN60600 VERTICAL PN60601 HORIZONTAL !. PN60617 HORIZONTAL !. PN60379 VERTICAL PN60380 HORIZONTAL !. PN60451 VERTICAL PN60452 HORIZONTAL !. PN60525 VERTICAL PN60526 HORIZONTAL !. PN60574 VERTICAL PN60575 HORIZONTAL PN60581 VERTICAL !. !. !. !.!. !. PN60041 VERTICAL !. PN59173 HORIZONTAL !.!. !. PN60620 VERTICAL PN60621 HORIZONTAL !. PN60670 VERTICAL PN60672 HORIZONTAL !. PN60389 HORIZONTAL PN60357 VERTICAL A130032 HORIZONTAL A130033 HORIZONTAL A130034 HORIZONTAL A130035 HORIZONTAL !.!.!. PN60582 HORIZONTAL !. !.A130053 VERTICAL A130054 HORIZONTAL !.!.!. A130037 HORIZONTAL A130038 HORIZONTAL A130039 HORIZONTAL A130043 HORIZONTAL A130044 HORIZONTAL A130045 HORIZONTAL A130046 HORIZONTAL A130047 HORIZONTAL !.!.!.!.!.!.!.!.!. KENTKENT HURONHURON SANILACSANILAC BAYBAY LAKELAKE IONIAIONIA OAKLANDOAKLAND CASSCASS ALLEGANALLEGAN WAYNEWAYNE SAGINAWSAGINAW IOSCOIOSCO TUSCOLATUSCOLA EATONEATON CLARECLARE NEWAYGONEWAYGO ALCONAALCONA BARRYBARRY LAPEERLAPEER ST.CLAIRST.CLAIR LENAWEELENAWEE JACKSONJACKSON ALPENAALPENA CALHOUNCALHOUN MENOMINEEMENOMINEE OTTAWAOTTAWA INGHAMINGHAM MASONMASON ANTRIMANTRIM GENESEEGENESEE EMMETEMMET OSCODAOSCODA CLINTONCLINTON BERRIENBERRIEN GRATIOTGRATIOT OCEANAOCEANA MONTCALMMONTCALM CHEBOYGANCHEBOYGAN OGEMAWOGEMAW MONROEMONROE OTSEGOOTSEGO BRANCHBRANCH ISABELLAISABELLA OSCEOLAOSCEOLA MECOSTAMECOSTA MIDLANDMIDLAND WEXFORDWEXFORD VANBURENVANBUREN WASHTENAWWASHTENAW HILLSDALEHILLSDALE GLADWINGLADWIN MANISTEEMANISTEE MACOMBMACOMB MISSAUKEEMISSAUKEE LIVINGSTONLIVINGSTON PRESQUEISLEPRESQUEISLE BENZIEBENZIE KALAMAZOOKALAMAZOO ST.JOSEPHST.JOSEPH MUSKEGONMUSKEGON ARENACARENAC ROSCOMMONROSCOMMON SHIAWASSEESHIAWASSEE MONTMORENCYMONTMORENCY LEELANAULEELANAU CHARLEVOIXCHARLEVOIX DELTADELTA GRANDGRAND TRAVERSETRAVERSE PN60614 VERTICAL PN60615 HORIZONTAL PN59912 VERTICAL PN60536 VERTICAL PN60537 HORIZONTAL PN60212 HORIZONTAL PN60588 HORIZONTAL PN60587 HORIZONTAL * ± ACTIVEAPPLICATIONSANDISSUEDPERMITS-SINCE2008* NOTE: PERMIT NUMBER AND APPLICATION NUMBER CAN BE CROSS REFERENCED BETWEEN THE MAP AND SPREADSHEET. PN60674 HORIZONTAL PN60662 HORIZONTAL # Permit # Company Name Well Name Well No County Wellhead T R S comments target formation Well type Well status Confidential 1 59112 BEACON EXPLORATION AND PRODUCTION CO LLC SCHULTZ 1--36 SANILAC 12N 15E 36 well completed Feb. 2012 A1 Carbonate Oil Shut-in NO 2 59173 CIMAREX ENERGY CO SOPER 1-25 HD1 OSCEOLA 17N 10W 25 well completed Aug. 2008 Antrim Gas Plugging complete NO 3 59449 O I L NIAGARAN LLC HENKEL D4-24 MISSAUKEE 21N 6W 24 proposed deepening of Antrim permit Utica-Collingwood Dry Hole Plugging complete NO 4 59979 ENCANA OIL AND GAS USA INC PIONEER 1-3 HD1 MISSAUKEE 24N 7W 3 well completed Feb 2010 Utica-Collingwood Gas Shut-in NO 5 60041 MERIT ENERGY COMPANY HUBBEL 2-22 HD1 MONTMORENCY 29N 1E 22 well completed June. 2011 Niagaran Oil Producing NO 6 60133 ENCANA OIL AND GAS USA INC KENDALL 1-33 CHEBOYGAN 35N 2W 33 permit for vertical well Utica-Collingwood Dry hole Well complete NO 7 60137 ATLAS RESOURCES LLC STATE MANCELONA 1-28 ANTRIM 29N 5W 28 permit for vertical well PILOT - Not to be Hydraulic Fractured Dry Hole Temporarily abandoned NO 8 60138 ATLAS RESOURCES LLC LUCAS 1-13 KALKASKA 26N 8W 13 permit for vertical well PILOT - Not to be Hydraulic Fractured Dry Hole Plugging complete. HD permitted NO 9 60161 ATLAS RESOURCES LLC STATE NORWICH 1-6 HD1 MISSAUKEE 24N 6W 6 permit for horizontal well Utica-Collingwood Dry Hole Temporarily abandoned NO 10 60170 ENCANA OIL AND GAS USA INC STATE KOEHLER & KENDALL 1-27 HD1 CHEBOYGAN 35N 2W 33 well completed Oct 2010 Utica-Collingwood Oil Temporarily abandoned NO 11 60183 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 1-24 KALKASKA 27N 6W 24 permit for vertical well PILOT - Not to be Hydraulic Fractured Dry Hole Plugging complete. HD drilled NO 12 60198 ATLAS RESOURCES LLC LUCAS 1-13 HD1 KALKASKA 26N 8W 13 permit for horizontal well (60138) Utica-Collingwood Not available Temporarily abandoned NO 13 60212 CONTINENTAL RESOURCES INC KELLY ET AL 1-26 HD1 HILLSDALE 6S 2W 26 well completed Sept. 2011 Black River (Van Wert) Oil Producing NO 14 60305 ENCANA OIL AND GAS USA INC STATE WILMOT 1--21 CHEBOYGAN 33N 3W 21 well completed July 2011 Utica-Collingwood Oil Plugging complete NO 15 60328 ENCANA OIL AND GAS USA INC STATE TUSCARORA 1--34 CHEBOYGAN 35N 5W 34 permit for vertical well Utica-Collingwood Location Permitted Well YES 16 60357 ENCANA OIL AND GAS USA INC STATE OLIVER 1-1 KALKASKA 26N 6W 1 permit for vertical well PILOT - Not to be Hydraulic Fractured Dry Hole Well complete. HD drilled NO 17 60360 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 1-13 HD1 KALKASKA 27N 6W 24 well completed Nov 2011 Utica-Collingwood Gas Producing NO 18 60379 DEVON ENERGY PRODUCTION COMPANY LP CRONK 1-24 P GLADWIN 19N 1W 24 permit for vertical well PILOT - Not to be Hydraulic Fractured Dry Hole Well Complete NO 19 60380 DEVON ENERGY PRODUCTION COMPANY LP CRONK 1-24 HD1 GLADWIN 19N 1W 24 permit for horizontal well A1 Carbonate Gas Well Complete NO 20 60389 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 1-25 HD1 KALKASKA 26N 6W 1 well completed Nov 2011 Utica-Collingwood Gas Producing NO 21 60451 DEVON ENERGY PRODUCTION COMPANY LP WILEY 1-18 P GLADWIN 18N 2W 18 permit for vertical well PILOT - Not to be Hydraulic Fractured Other Well Complete NO 22 60452 DEVON ENERGY PRODUCTION COMPANY LP WILEY 1-18 HD1 GLADWIN 18N 2W 18 well completed may/june 2012 A1 Carbonate Gas Temporarily abandoned NO 23 60525 DEVON ENERGY PRODUCTION COMPANY LP SCHICK 1-7P CLARE 19N 3W 7 permit for vertical well PILOT - Not to be Hydraulic Fractured Other Well Complete NO 24 60526 DEVON ENERGY PRODUCTION COMPANY LP SCHICK 1-7HD1 CLARE 19N 3W 7 well not hydraulic fractured A1 Carbonate Other Well Complete NO 25 60536 CONTINENTAL RESOURCES INC MCNAIR ET AL 1-26 P HILLSDALE 6S 2W 26 permit for vertical well Black River (Van Wert) Other Plugging complete NO 26 60537 CONTINENTAL RESOURCES INC MCNAIR ET AL 1-26 HD1 HILLSDALE 6S 2W 26 permit for horizontal well Black River (Van Wert) Oil Producing NO 27 60545 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 2-25 HD1 KALKASKA 26N 6W 1 permit for horizontal well Utica-Collingwood Gas Drilling complete NO 28 60546 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 3-25 HD1 KALKASKA 26N 6W 1 permit for horizontal well Utica-Collingwood Location Drilling complete NO 29 60559 DEVON ENERGY PRODUCTION COMPANY LP STATE RICHFIELD 1-27P ROSCOMMON 22N 1W 27 permit for vertical well PILOT - Not to be Hydraulic Fractured Other Well Complete NO 30 60560 DEVON ENERGY PRODUCTION COMPANY LP STATE RICHFIELD 1-34 HD1 ROSCOMMON 22N 1W 27 permit for horizontal well (60559) Collingwood Gas Well Complete NO 31 60562 ENCANA OIL AND GAS USA INC STATE MENTOR 1-17 CHEBOYGAN 34N 3W 17 permit for vertical well Utica-Collingwood Location Permitted Well YES 32 60574 ALTA ENERGY OPERATING LLC RILEY 1-22 OCEANA 15N 18W 22 permit for vertical well PILOT - Not to be Hydraulic Fractured Other Plugged back NO 33 60575 ALTA ENERGY OPERATING LLC RILEY 1-22 HD1 OCEANA 15N 18W 22 permit for horizontal well (60574) A1 Carbonate Location Drilling complete NO 34 60579 ENCANA OIL AND GAS USA INC STATE GARFIELD 1-25 HD1 KALKASKA 25N 6W 36 permit for horizontal well Utica-Collingwood Location Drilling complete NO 35 60581 DEVON ENERGY PRODUCTION COMPANY LP DAVID'S ACRES, LLC 1-19 P OGEMAW 22N 4E 19 permit for vertical well PILOT - Not to be Hydraulic Fractured Other Well Complete NO 36 60582 DEVON ENERGY PRODUCTION COMPANY LP DAVID'S ACRES, LLC 1-19 HD1 OGEMAW 22N 4E 19 well not hydraulic fractured A1 Carbonate Other Well Complete NO 37 60587 COUNTRYMARK ENERGY RESOURCES LLC ARNO 1-25 HD1 HILLSDALE 6S 2W 25 permit for horizontal well Black River (Van Wert) Location Permitted Well NO 38 60588 COUNTRYMARK ENERGY RESOURCES LLC ARNO & TIMMONS 1-24 HD1 HILLSDALE 6S 2W 24 permit for horizontal well Black River (Van Wert) Location Permitted Well NO 39 60600 ENCANA OIL AND GAS USA INC WESTERMAN 1-29 KALKASKA 28N 8W 29 permit for vertical well PILOT - Not to be Hydraulic Fractured Location Permitted Well YES 40 60601 ENCANA OIL AND GAS USA INC WESTERMAN 1-32 HD1 KALKASKA 28N 8W 29 permit for horizontal well Utica-Collingwood Location Permitted Well YES 41 60606 ENCANA OIL AND GAS USA INC STATE MENTOR 1-17 HD1 CHEBOYGAN 34N 3W 17 permit for horizontal well (60562) Utica-Collingwood Location Permitted Well YES 42 60614 ROSETTA RESOURCES OPERATING LP CHRISTENSEN 1-21 P IONIA 6N 6W 21 permit for vertical well PILOT - Not to be Hydraulic Fractured Location Permitted Well YES 43 60615 ROSETTA RESOURCES OPERATING LP CHRISTENSEN 1-21 HD1 IONIA 6N 6W 21 permit for horizontal well A1 Carbonate Location Permitted Well YES 44 60617 DEVON ENERGY PRODUCTION COMPANY LP YOUNKMAN 1-29 HD1 MISSAUKEE 21N 8W 29 permit for horizontal well Utica Location Permitted Well YES 45 60620 ENCANA OIL AND GAS USA INC STATE BEAVER CREEK C3-11 CRAWFORD 25N 4W 11 permit for vertical well PILOT - Not to be Hydraulic Fractured Location Permitted Well YES 46 60621 ENCANA OIL AND GAS USA INC STATE BEAVER CREEK 1-23 HD1 CRAWFORD 25N 4W 11 permit for horizontal well Utica-Collingwood Location Permitted Well YES 47 60662 COUNTRYMARK ENERGY RESOURCES LLC STIVERSON & FRENCH 1-25 HD1 HILLSDALE 6S 2W 24 permit for horizontal well Black River (Van Wert) Location Permitted Well NO 48 60670 ENCANA OIL AND GAS USA INC STATE ROSCOMMON D1-17 ROSCOMMON 21N 4W 17 permit for vertical well PILOT - Not to be Hydraulic Fractured Location Permitted Well YES 49 60672 ENCANA OIL AND GAS USA INC STATE ROSCOMMON 1-7 HD1 ROSCOMMON 21N 4W 17 permit for horizontal well Utica-Collingwood Location Permitted Well YES 50 60674 MUZYL OIL CORPORATION BURNS A1-23 HD1 HILLSDALE 6S 2W 23 permit for horizontal well Black River (Van Wert) location Permitted Well NO 51 60685 ENCANA OIL AND GAS USA INC STATE GARFIELD 1-26 KALKASKA 25N 6W 26 permit for vertical well PILOT - Not to be Hydraulic Fractured Location Permitted Well YES 52 60686 ENCANA OIL AND GAS USA INC STATE GARFIELD 1-23 HD1 HILLSDALE 25N 6W 26 permit for horizontal well Utica-Collingwood Location Permitted Well YES # App # Company Name Well Name Well No County Wellhead T R S target formation 1 A110068 MERIT ENERGY COMPANY STATE MANCELONA 8-33 ANTRIM 29N 5W 33 Utica-Collingwood 2 A120046 ENCANA OIL AND GAS USA INC STATE WILMONT 1-6 CHEBOYGAN 33N 3W 6 Utica-Collingwood 3 A130031 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 1-14 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 4 A130032 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 1-12 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 5 A130033 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 1-11 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 6 A130034 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 2-14 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 7 A130035 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 2-12 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 8 A130037 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 3-12 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 9 A130038 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 4-12 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 10 A130039 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 5-12 HD1 KALKASKA 27N 6W 24 Utica-Collingwood 11 A130043 ENCANA OIL AND GAS USA INC STATE OLIVER 3-13 HD1 KALKASKA 26N 6W 1 Utica-Collingwood 12 A130044 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 4-25 HD1 KALKASKA 26N 6W 1 Utica-Collingwood 13 A130045 ENCANA OIL AND GAS USA INC STATE OLIVER 2-13 HD1 KALKASKA 26N 6W 1 Utica-Collingwood 14 A130046 ENCANA OIL AND GAS USA INC STATE OLIVER 1-13 HD1 KALKASKA 26N 6W 1 Utica-Collingwood 15 A130047 ENCANA OIL AND GAS USA INC STATE EXCELSIOR 5-25 HD1 KALKASKA 26N 6W 1 Utica-Collingwood 16 A130053 ROSETTA RESOURCES OPERATING LP SWANSON TRUST 1-1 P MUSKEGON 9N 14W 1 A1-Carbonate 17 A130054 ROSETTA RESOURCES OPERATING LP SWANSON TRUST 1-1 HD1 MUSKEGON 9N 14W 1 A1-Carbonate HIGH VOLUME (>100,000 gallons) HYDRAULIC FRACTURING PROPOSALS - ACTIVE APPLICATIONS comments HIGH VOLUME (>100,000 gallons) HYDRAULIC FRACTURING SINCE 2008 - ACTIVE PERMITS application for horizontal well application for horizontal well application for vertical well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for horizontal well application for pilot - not to be hydraulically fractured
    • 5 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 being spilled on unpaved areas and draining into surface and groundwaters16 . Evaluations of fracturing operations in central Arkansas found that surface water quality violations at site operations were due to erosion (22%), illegal discharges (10%) and spills (10%)5 . Impacts to receiving water streams and their biota were significantly linked to well and pad densities, rate of installations, inverse flow path length, pipeline density, and a combination of roads-pasture and well density proximity5 . These recent findings were presented at the annual meeting of the Society of Environmental Toxicology and Chemistry in November 2012 and support concerns that have previously been identified by Entrekin and others7–14 . One critical factor is that gas wells are often located adjacent to rivers and streams. In shale basins with a high density of operations, numer- ous well pads may be located within the same watershed, thus compounding the cumulative impacts of industrial activity within that particular watershed. To date most research focusing on envi- ronmental concerns of hydraulic fracturing focus on contamination of groundwater and contamination of drinking water sources. However, fewer data are available to address concerns associated with surface water and terrestrial ecosystems. The ongoing studies of Entrekin5,7 represent one of the best and most comprehensive scientific evaluations of the impacts of frac- turing operations on receiving waters and should be considered as applicable for Michigan in regards to runoff issues associated with site development. Although Entrekin’s report focuses on the Fayetteville Shale in Arkansas, it is applicable to basins in Michigan if site development increases as it has in Arkansas. The comparison can be made in the broad similarities of vegetation percentage, surface cover type, moisture availability, and amount of runoff. Both Arkansas and Michigan are prone to high amounts of precip- itation, and have slightly rolling topography with high percentages of vegetation cover. Produced water will be a significant waste stream during the pro- duction phase, requiring extensive trucking to offsite injection wells. Regulations govern the disposal of this waste stream; most is disposed of by underground injection either in disposal wells or, in mature producing fields, in enhanced oil recovery wells (i.e., wells through which produced water and other materials are injected into a producing formation in order to increase formation pressure and production)23 . In locations where naturally occurring radioactive materials (NORM)-bearing produced water and solid wastes are gener- ated, mismanagement of these wastes can result in radiological contamination of soils or surface water bodies24–26 . In some loca- tions, produced water may carry NORM to the surface. Typically, the NORM radionuclides (primarily radium-226, radium-228, and their progeny) are dissolved in the produced water25 . Proper man- agement of NORM-bearing produced water and solid wastes are critical to prevent both occupational and public human health risks and environmental contamination. NORM waste problems are generally associated with long-term operations of oil gas fields. The NORM Technology website (http://norm.iogcc.state.ok.us/ reg/dsp_statereg.cfm) provides information about the regulation of NORM bearing wastes on a state by state basis as generated by the petroleum industry25 . Exposure of wildlife to light and noise is an additional concern, and impacts on wildlife will likely vary among types of wildlife and species (e.g. game species, migratory birds, amphibians). The main sources of noise during the production phase would include compressor and pumping stations, producing wells (including occasional flaring), and vehicle traffic. Compressor stations pro- duce high noise levels. Use of remote telemetry equipment would reduce daily traffic and associated noise levels within the oil and gas field area. The primary impacts from noise would be localized disturbance to wildlife, livestock, recreationists, and residents. Flooding an ecosystem with excessive light can disrupt feeding, breeding, and rest patterns in micro- and mega- flora and fauna, providing a potential for ecosystem degradation. The risks that hydraulic fracturing poses to susceptible ecosystems were studied in the adjacent Marcellus Shale region5 . This is appli- cable to Michigan as models describing cumulative probability and contamination volume per well are developed. The Marcellus Shale region studies point to the need for monitoring8–11 . A useful way to assess the potential impacts of hydraulic fracturing opera- tions is through geographic information system (GIS)-based mod- els that incorporate ecological, political, and fracturing features6 . The USEPA estimates that 5 million gallons of fracturing solution is consumed per month, along with 1.5 million pounds of proppant. In the Marcellus, the EPA undertook a biological assessment of the Allegheny and Monongahela Rivers. To design their study, they first evaluated conditions via probabilistic survey for the following: fish, fish habitat, macro-invertebrates (such as mussels), water chemis- try, plankton, and sediment. Data assisted in risk assessment from potential stressors, as well as aided in analyzing the potential sea- sonal and yearly variability. In these formations, process waters may be discharged to wastewater treatment plants; however, that will not occur in Michigan where these waters are deep-well injected. Figure 1: Locations of High Volume Hydraulic Fracturing in Michigan according to Michigan Department of Environmental Quality22 .
    • 6 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 Another tool used by the EPA in their 2008 Marcellus study was “RAIN,” or, River Alert Information Network (http://www.3rain. org/). RAIN integrates information from water treatment, source water protection, and distribution system maintenance into a mul- tiple barrier approach. The goal of RAIN is to employ protection measures to form a first barrier to a multiple-barrier approach to drinking water protection. This includes providing information and tools to aid water suppliers in making decisions, and improving communication between water suppliers about water quality events. RAIN implements these goals by installing monitoring equipment at appropriate locations and providing operational training. The EPA RAIN administrators will develop a secure web- site to share information about water quality, as well as improve communication between water suppliers, the US Army Corps of Engineers (USACE) and emergency responders. RAIN covers the areas of the Allegheny, Monongahela, Youghiogheny, and headwaters of the Ohio River. The particular communication system is governed on spill alerts, alarm noti- fications, and water supplier roundtables. RAIN’s monitoring systems are based on-line, continuous monitoring equipment, and operator training. The RAIN website employs water qual- ity data, both historical and current, and provides links to other applicable websites that provide monitoring data such as the US Geological Survey (USGS), National Oceanic and Atmospheric Administration, National Weather Service, and the USACE. The RAIN’s Monongahela Total Dissolved Solids Project Monitoring Effort is composed of ten RAIN facilities that measure conductivity, pH and temperature. Additionally, RAIN has four remote tributary sites with data readings. The remote sites measure conductivity, pH and temperature. RAIN has proposed 11 more 15-member moni- toring facilities. Monitoring efforts will focus on the environmental constituents of concern: nitrate, ammonia, dissolved oxygen, UV organics, suspended solids and turbidity, as well as ORP. As a tandem effort to RAIN, the EPA initiated a waste character- ization study to measure TDS, metals, organics and TENORM. The study is dual-phased, with Phase I focusing on site-specific characteristics across the region. In Pennsylvania, the rapid pace of Marcellus Shale drilling has outstripped Pennsylvania’s ability to document pre-drilling water quality, even with some 580 organiza- tions focused on monitoring the state’s watersheds. More than 300 are community-based groups that take part in volunteer stream monitoring. Unlike the Marcellus Shale region, there will not be discharges of process waters to wastewater treatment plants or surface impoundments in Michigan; however, there is a need for similar surface water monitoring programs as described above, both pre- and post-drilling operations. 2.2 State of Michigan Programs State specific regulations concerning surface waters and hydrau- lic fracturing operations in Michigan are driven by the Michigan Department of Environmental Quality (DEQ) (drilling permit) and Michigan Department of Natural Resources (DNR) (Well-site per- mit for State of Michigan owned surface lands) (see the Technical Report on Policy & Law for details). Before permits are issued, DNR and DEQ personnel evaluate any potential sensitive ecosystems, considering endangered and threatened species, streams and fisheries, and other relevant issues. The State permitting process dictates that all hydraulic fracturing operations reduce their potential impact on-site through a variety of measures. These include construction of the well-pad at least 1320 feet from the nearest stream for State leases. For private properties, the DEQ requires optimal location that protects surface water while considering a host of other property and environmen- tal issues. The State’s considerations also include land elevations, avoiding hillsides, and always using silt curtains. All pervious site grounds are covered in plastic to capture any potential spillage. Permitted sites are for a drilling unit (a tract which the DEQ has determined can be efficiently drained by one well), which is gener- ally a minimum of 80 acres in size but often much larger, while the working pad area is usually less than 5 acres regardless of unit size. Lined-berms are put in place to contain tank or pipe spills. The DEQ (and the DNR where State acreage is involved) also evaluates where roads may be constructed. Well operators are required to have spill pollution prevention plans. After site operations cease, the owners are required to reclaim the site using native species of vegetation. All of these procedures are encouraging if imple- mented and if monitored routinely by State personnel. The primary hazard of operations appears to be that of trucking production brine waters from the fracturing process. This leads to the possi- bility of vehicle related accidents, and increased dust and erosion from dirt roads. Some of the public health issues related to this are covered in a subsequent chapter. Michigan DEQ has developed a fairly robust Wetlands Protection program, stemming from Part 303 of the Natural Resources and Environmental Protection Act (NREPA), PA 451 of 1994 (NREPA). The statute requires protection of wetlands under private and public land, without respect to zoning or ownership. However, with respect to wetland protection, it is important to remember that even though the acreage sizes of wetlands may be small, they are generally interconnected systems. Even very small wetlands can still be important surface water sources and reserves. With this scenario in mind, when considering the accidental spills or unintentional impacts of any hydraulic fracturing operations, it is important to remember that there is a connection between water quantity and quality. Taking water from a small stream concentrates
    • 7 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 any contaminants in the stream water. If stream flow is reduced by groundwater withdrawals, the lower dilution rate of any solids loadings or other contaminants from the watershed can damage ecosystems and harm aquatic life. Michigan’s WWAT2 is designed to estimate the likely Adverse Resource Impact of a water withdrawal on nearby streams and rivers (http://www.miwwat.org/). Use of the WWAT is required of anyone proposing to make a new or increased large quantity withdrawal (over 70 gallons per minute) from the waters of the state, including all groundwater and surface water sources, prior to beginning the withdrawal. The Michigan Senate and House enacted new legislation to manage large water withdrawals in the state using science as the basis for policy development, including a water withdrawal assessment process for high capacity wells. These Public Acts became part of Michigan Compiled Laws and amended Part 327 (Great Lakes Preservation) of the NREPA and the Safe Drinking Water Act (SDWA), PA 399 of 1976. Fracturing operations will be assessed during the permit review process by the WWAT to determine if a proposed withdrawal is likely to cause an Adverse Resource Impact. This system allows for an evaluation of potential impacts to many sensitive ecosystems but has limita- tions, as discussed below. Currently the State is not maintaining a registry of the hydraulic fracturing operation water withdrawals or entering them into the water withdrawal accounting system to be subtracted from the available water balance. However, the DEQ does account for prior withdrawals for hydraulic fracturing. The hydraulic fracturing withdrawals are kept in a separate database and included in later assessments on a case-by-case basis. After the WWAT analyses is conducted and approved for a hydraulic fracturing operation per- mit, the withdrawals are no longer considered in relation to other applications or operations in the area. Given that fracturing oper- ations can be dense and adjacent to each other, this creates the possibility for negative cumulative impacts from high volume water withdrawals. Indeed recent operations will be in the tens of millions of gallons extracted for each operation. It can take months to years for groundwater aquifers to replenish after large extractions, there- fore the impacts of multiple operations within the same aquifer with not be assessed by the WWAT. Some of the State’s surface waters most sensitive to groundwa- ter withdrawals are classified as cold transitional and cold waters. Streams whose headwaters are shallow are particularly at risk during drought and low flow periods. The WWAT may not fully account for the shallow stream morphology. Based on a limited analysis, Jocks and Bzdok (2010) question using the tool in its current form for analyzing the impacts of withdrawals on small streams and riv- ers28 . In addition, this tool does not account for water withdrawal impacts to wetlands and lakes. The WWAT estimates surface water flows across the state from fewer than 150 USGS river and stream gauges, which tend to be located on medium and large sized streams. Sensitive headwaters are rarely monitored; therefore the WWAT model has high associated uncertainties. It was primarily designed to account for long term withdrawals, such as agricultural irrigation. However, questions have been raised about the ability of the tool to address short-term intensive withdrawals such as those associated with hydraulic fracturing operations1 . It is also a concern that the massive quantities being removed from the aquifer are not being replaced, but rather deep-well injected. Some relevant questions that arise from the hydraulic fracturing operations that the current version of WWAT cannot answer are: Will local hydrologic cycles be altered? How long before they recover? How do water withdrawals during winter conditions impact fish during this sensitive time for their survival? Are stream base flow estimates accurate? What are base flows for critical headwater streams? When the Legislature approved the WWAT, the authors of the approach stressed that “Any implementation […of the recommended WWAP] must include a plan for ongoing, periodic field testing and review and revision of the process and tool.”2 The WWAT has never been updated and remains as Version 1. There is a critical need to update the model to better account for wetlands, shallow streams, and high volume water withdrawals. The prospects of a changing regulatory environment that could result in shifting priorities from protection of resources to produc- tion of renewable and non-renewable commodities impact risks to ecological systems. For example, Michigan Senate Bill 7829 recently passed by the state Senate and headed for a vote in the House of Representatives would modify Michigan’s Natural Resources and Environmental Protection Act to explicitly prohibit state agencies from designating or classifying an area of land for the purpose of achieving or maintaining biological diversity. In addition, the bill redefines the goals of biological conservation to be more aligned with managing public lands for economic interests, eliminates text about managing forests for sustainability, and eliminates text saying that biodiversity loss is primarily caused by humans. While biodiversity is not one of the specific criteria used by the DNR for nondevelopment lease classification30 , this bill raises questions about the likelihood that hydraulic fracturing operations could be allowed in natural/wild/wilderness areas. The associated well- pad and road construction, habitat loss, and associated human activities would undoubtedly pose threats to sensitive ecosystems that have historically been protected. Further analysis is needed regarding the potential impact of this legislation.
    • 8 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 3.0 CHALLENGES AND OPPORTUNITIES O ne of the greatest challenges in quantifying the eco- logical effects of hydraulic fracturing is the enormous potential for variation within and among different eco- systems and the differing hydraulic fracturing opera- tion sizes, pad densities and quality control measures. Additionally, as multiple well sites are established within watersheds, there is potential for the ecological effects of these fracturing operations to interact. Upstream wells, for example, could impact water flows, turbidity or nutrient and TDS loadings of aquatic communities far downstream, particularly if impacts of downstream wells are addi- tive or synergistic. Another challenge lies in the examination of the effects of fracturing operations before, during, and after the actual hydraulic fracturing occurs. Typically wells will only be actively fractured in a one to two month time frame. However, the ecological effects of the fracturing begin as soon as infrastructure construction is initiated, and last through the fracturing phase, and for an un-established period of time after fracturing is finished. Related to this is the inability to assess whether an actual ecological impact has occurred due to the lack of monitoring. Very few sites exist across the nation where baseline (reference condition) environmental monitoring has occurred prior to hydraulic fracturing operations commenc- ing. From both scientific and practical perspectives it is difficult to establish “impacts” if the baseline is unknown, particularly if these operations are occurring in human dominated watersheds. It is essential that at least a subset of hydraulic fracturing operations have pre- and post-monitoring of environmental conditions to establish whether or not detrimental impacts are occurring. One example of an ongoing effort to characterize the effects of the development of shale gas resources through hydraulic frac- turing is by a forest conservation organization, called the Pinchot Institute for Conservation31 . The Pinchot Institute is in the process of implementing non-partisan research on the future of sustainable resource management. Their interdisciplinary workshop in 2011 outlined ways in which scientific methods are being established to cumulatively assess the various facets of shale gas development that impact any ecosystem32 . Three specific key factors analyzed by this report are: the location planning for water withdrawals, tim- ing of water withdrawals, and the centralization of infrastructure. Presumably, if the timing of water withdrawals could be initiated during a time of groundwater recharge, the effect to surface water ecosystems would be diminished. Additionally, if the location of groundwater withdrawals could be located at a feasible distance from ecologically susceptible biodiversity, possible harm to surface water ecosystems could also be diminished. And, lastly, if the infra- structure utilized to perform the withdrawals could be centrally located, the infrastructure operator could reduce the potential area for ecological impact31 . Although chemical spills are less frequent than chronic habitat disturbance and erosion, it is important to begin to understand the toxicity of the wide-range of hydraulic fracturing chemicals and combinations of these chemicals that may be released in produced waters, in addition to any pure chemical products stored on-site. Full assessment of the complex task of determining whether eco- logical systems are at risk from hydraulic fracturing operations requires a comprehensive, watershed-based research and man- agement approach. An appropriate analogy that may be useful is the Total Maximum Daily Loading (TMDL) program, used widely by EPA and the States, which offers a useful watershed-based framework for this task and accounts for the cumulative contribu- tions of multiple sources to receiving waters. Although oil and gas operations are not granted surface water discharges, the idea of considering environmental and groundwater “loadings or use” on a watershed by watershed basis is appropriate. The TMDL is a useful tool in establishing particular watersheds, water bodies, or water basins that may be impaired. The TMDL was developed under section 303(d) of the Clean Water Act that requires states or territories to develop lists of waters that are “impaired” or other- wise too degraded to meet water quality standards32 . The TMDL actually calculates a maximum amount of pollutant that a body of water can maintain, while still adhering to the approved water quality standards32 . The TMDL tool provides curves that aid in a calculation of the duration that a particular pollutant or chemical of concern can last in a certain water body. Thus, an industrial oper- ator, or monitoring agency, could use this approach to evaluate how to assess the potential terrestrial and surface water impacts of multiple HVHF operations within a watershed. The WWAT must be modified to consider cumulative withdrawal impacts from oper- ations drawing on the same aquifer, at extremely high volumes, during biologically sensitive seasonal periods. There are extensive studies ongoing to determine the environ- mental and ecological impacts of hydraulic fracturing. For exam- ple, EPA provided a progress report on their study of the potential impacts of hydraulic fracturing on drinking water resources in December 2012 (http://www.epa.gov/hfstudy/). In January of this year, Glenn Paulson, EPA’s science advisor, was hopeful that the agency’s nationwide project examining natural gas hydraulic frac- turing and potential drinking water impacts will provide compre- hensive guidelines. The report will be released in 2014. EPA, DOE and DOI signed a MOU in April 2012 to align their research in 2013. They are also including CDC, NIOSH and other HHS agencies to be engaged on their steering committee.
    • 9 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 Michigan is fortunate to have a Wetland Protection Program and also a WWAT and a comprehensive permitting program by the DNR and DEQ that consider the local environmental impacts and source controls. These could allow for effective evaluations of potential ecological impacts from fracturing operations by con- sidering their proximity and density in relation to sensitive and vulnerable wetlands and fisheries, such as shallow trout streams and cold transitional waters. The focus of the WWAT is on long- term, growing season, groundwater withdrawal impacts to surface waters but may require updating to address questions associated with short-term extremely high volume extractions, multiple opera- tions in the same watershed, shallow streams, and headwater base flow conditions1,2 . This screening tool, currently designed, may not identify ecologically sensitive situations. This tool will also not assess the potential impacts of establishing the infrastructure and operations on habitat, wildlife, and nearby waters receiving site runoff. Routine site inspections will be required to ensure site ero- sion is minimal and spill prevention plans are being followed. GIS- based modeling and site monitoring will allow for these potential impacts to be evaluated ensuring proper siting and operational controls are established and followed. 4.0 PRIOTIZED PATHWAYS FOR PHASE 2 • Establish a decision-matrix that guides decision making on establishing hydraulic fracturing operations in “sensitive/sus- ceptible” ecosystems. • Establish baseline (reference condition) ecosystem monitoring in susceptible areas that continues through post-operation periods to establish whether or not detrimental impacts occur. • Assess the cumulative impacts of multiple hydraulic fracturing operations within a watershed for downstream surface waters and groundwater. Update the WWAT Version 1 to reduce critical uncertainties identified above. • Establish to what degree other likely stressors in watershed, unrelated to fracturing operations, impact aquatic communities. • Identify areas for improved quality control / best practices in fracturing operations, especially near riparian zones, surface waters and shallow aquifers. • Establish a publically available database for HVHF studies and data. • It is important that close attention be paid to the findings pub- lished in the “peer-reviewed” scientific literature in the coming months to years to improve decision-making. • Any assessment of ecological health impacts from this ener- gy-driven activity, should in turn, evaluate how these potential impacts compare to the environmental impacts of energy- related activities, such as coal mining, that it may be replacing. LITERATURE CITED 1. Michigan Department of Environmental Quality [Michigan DEQ]; Water Use Advisory Council Agendas and Meeting Notes [Internet]. Michigan DEQ [Cited 25 July 2013]. Available from: http://www.michigan.gov/deq/0,4561,7-135-3313_3684_64633---,00.html. Added by report editor with approval of report authors. 2. Hamilton DA, Seelbach PW. Michigan’s Water Withdrawal Assessment Process and Internet Screening Tool. Fisheries Division Special Report 55. Lansing, MI: Michigan Department of Natural Resources; 2011 May. 37p. 3. US Government Accountability Office [GAO]. Oil and Gas: Information on Shale Resources, Development, and Environmental and Public Health Risks. Report to Congressional Requesters. Washington, DC: GAO; 2012 Sept. 4. US Government Accountability Office [GAO]. Unconventional Oil and Gas Development: Key Environmental and Public Health Requirements. Report to Congressional Requesters. Washington, DC: GAO; 2012 Sept. 5. Entrekin S. Presentation at the Annual Meeting of the Society of Environmental Toxicology & Chemistry. Long Beach, CA. 2012 Nov. and personal communication 2013 May 1. 6. Davis JB, Robinson GR. A geographic model to assess and limit cumulative ecological degradation from Marcellus Shale exploitation in New York, USA. Ecology and Society 2012;17(2):25. 7. Entrekin S, Evans-White M, Johnson B, Hagenbuch E. Rapid expansion of natural gas development poses a threat to surface waters. Frontiers Ecol. 2011;9:503-511. 8. Weltman-Fahs M, Taylor JM. Hydraulic fracturing and brook trout habitat in the Marcellus Shale region: Potential impacts and research needs. Fisheries 2013;38:4-15. 9. Williams HFL, Havens DL, Banks KE, Wachal DJ. Field-based monitoring of sediment runoff from natural gas well sites in Denton County, Texas, USA. Environ Geol. 2008;55:1463-1471.
    • 10 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: ENVIRONMENT/ECOLOGY TECHNICAL REPORT, SEPTEMBER 2013 10. Drohan PJ, Brittingham M, Bishop J, Yodeer K. Early trends in landcover change and forest fragmentation due to shale-gas development in Pennsylvania: A potential outcome for the Northcentral Appalachians. Environ Management. 2012 May;49(5):1061-1075. 11. Batley GE, Kookana RS. Environmental issues associated with coal seam gas recovery: managing the fracking boom. Environ Chem. 2012;9:425-428. 2012. 12. Rahm BG, Riha SJ. Toward strategic management of shale gas development: Regional, Collective impacts on water resources. Environ Sci & Policy 2012;17:12-23. 13. Rozel DA, Reaven SJ. Water Pollution Risk Associated with Natural Gas Extraction from the Marcellus Shale. Risk Analysis 2012;32(8): 1382-1393. 14. Vidic RD, Brantley SL, Vandenbossche JM, Yoxtheimer D, Abad JD. Impact of shale gas development on regional water quality. Science 2013 May 17;340(6134):1235009. 15. Michigan Department of Environmental Quality [Michigan DEQ]; Annual Wastewater Reporting [Internet]. [2012; cited 2013 Feb]. Available from: http://www.michigan.gov/deq/0,4561,7-135-3313_3682_4136---,00.html 16. Ewen C, Borchardt D, Richter S, Hammerbacker R. Hydraulic fracturing Risk Assessment: Study concerning the safety and environmental compatibility of hydraulic fracturing for natural gas production from unconventional reservoirs. Team Ewen - Darmstadt; ExxonMobil Production Deutchland GmbH. 2012 March. 17. STAC (Chesapeake Bay Program Scientific and Technical Committee). Exploring the environmental effects of shale gas development in the Chesapeake Bay Watershed. Edgewater, MD; 2013. STAC Publ. #13-01. 30 pp. 18. Burton GA, Johnston EL. Assessing contaminated sediments in the context of multiple stressors. Environ Toxicol Chem 2010;29: 2625-2643. 19. Lackey RT. Values, Policy, and Ecosystem Health. Bioscience 2001;51(6): 437-443. 20. Conradi R. A Frackin’ Shame. The Washtenaw Voice [Internet]. 2013 Feb 2. Available from: http://www.washtenawvoice.com/tag/antrim-shale/ 21. Shale Navigator; Pipeline Infrastructure Map, Antrim / Collingwood Shale, Michigan [Internet]. [Cited 2013 Feb 25]. Available from: http://www.shalenavigator.com/datastoreimages/StatePipelineMaps_MI.jpg 22. Michigan DEQ; High Volume Hydraulic Fracturing, Active Applications and Issued Permits Since 2008 [Internet]. Michigan DEQ [Cited 2013 March]. Available from: http://www.michigan.gov/documents/deq/utica-_collingwood_activity_map3_354847_7.pdf 23. Pennsylvania State University. College of Agricultural Sciences, Cooperative Division. Water Withdrawals for the Development of Marcellus Shale Gas in Pennsylvania. 2010. 24. US Environmental Protection Agency [EPA]. Study of the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources: Progress Report. Washington, DC: EPA; 2012 Dec. 262 p. 25. NORM Technology Connection [Internet]. Interstate Oil and Gas Compact Commission [Cited 2013 May 3]. Available from: http://norm.iogcc.state.ok.us/index.cfm 26. US Geological Survey. Naturally Occurring Radioactive Materials (NORM) in produced water and Oil-Field Equipment – An Issue for the Energy Industry. FS-142-99. 1999 Sept. 4p. 27. Michigan DEQ; Wetlands Protection [Internet]. Michigan DEQ [Cited 2013 Feb]. Available from: http://www.michigan.gov/deq/0,4561,7-135-3313_3687---,00.html 28. Jocks JL, Bzdok CM. The impact of Michigan’s water withdrawal legislation on small streams and small stream riparian owners. Michigan Real Property Review. Spring 2010;37(1): 28-33. 29. Michigan Bill 78 [Internet]. Available from: http://www.legislature.mi.gov/documents/2013-2014/billengrossed/Senate/pdf/2013-SEBS-0078.pdf 30. Michigan Department of Natural Resources; 27.23-15 – Oil and Gas Lease Classification Procedure (Issued: 07/11/2005) [Internet]. Michigan DNR [Cited 25 July 2013]. Available from: http://www.midnr.com/Publications/pdfs/InsideDNR/publications/DNRPolProc/27.23.15.htm. Added by report editor with approval of report authors. 31. Pinchot Institute for Conservation; Assessing the Effects of Marcellus Shale Gas Development: The State of Science [Internet]. [2011 April]. Available from: http://www.pinchot.org/gp/EffectsofMarcellusShale 32. US Environmental Protection Agency [EPA]; Impaired Waters and Total Maximum Daily Loads, Clean Water Act Section 303(d) [Internet]. EPA [Updated 2012 Oct.18]. Available from: http://water.epa.gov/lawsregs/lawsguidance/cwa/tmdl/index.cfm
    • © 2013 BY THE REGENTS OF THE UNIVERSITY OF MICHIGAN MARK J. BERNSTEIN, ANN ARBOR JULIA DONOVAN DARLOW, ANN ARBOR LAURENCE B. DEITCH, BLOOMFIELD HILLS SHAUNA RYDER DIGGS, GROSSE POINTE DENISE ILITCH, BINGHAM FARMS ANDREA FISCHER NEWMAN, ANN ARBOR ANDREW C. RICHNER, GROSSE POINTE PARK KATHERINE E. WHITE, ANN ARBOR MARY SUE COLEMAN, EX OFFICIO Please print sparingly and recycle
    • Public Health Technical Report H Y D R A U L I C F R A C T U R I N G I N T H E S T A T E O F M I C H I G A N
    • Participating University of Michigan Units Graham Sustainability Institute Erb Institute for Global Sustainable Enterprise Risk Science Center University of Michigan Energy Institute ABOUT THIS REPORT This document is one of the seven technical reports com- pleted for the Hydraulic Fracturing in Michigan Integrated Assessment conducted by the University of Michigan. During the initial phase of the project, seven faculty-led and student-staffed teams focused on the following topics: Technology, Geology/ Hydrogeology, Environment/Ecology, Human Health, Policy/ Law, Economics, and Public Perceptions. These reports were prepared to provide a solid foundation of information on the topic for decision makers and stakeholders and to help inform the Integrated Assessment, which will focus on the analysis of policy options. The reports were informed by comments from (but do not necessarily reflect the views of) the Integrated Assessment Steering Committee, expert peer reviewers, and numerous public comments. Upon completion of the peer review process, final decisions regarding the content of the reports were deter- mined by the faculty authors in consultation with the peer review editor. These reports should not be characterized or cited as final products of the Integrated Assessment. The reports cover a broad range of topics related to hydraulic fracturing in Michigan. In some cases, the authors determined that a general discussion of oil and gas development is important to provide a framing for a more specific discussion of hydraulic fracturing. The reports address common hydraulic fracturing (HF) as meaning use of hydraulic fracturing methods regardless of well depth, fluid volume, or orientation of the well (whether vertical, directional, or horizontal). HF has been used in thousands of wells throughout Michigan over the past several decades. Most of those wells have been shallower, vertical wells using approxi- mately 50,000 gallons of water; however, some have been deeper and some have been directional or horizontal wells. The reports also address the relatively newer high volume hydraulic fracturing (HVHF) methods typically used in conjunction with directional or horizontal drilling. An HVHF well is defined by the State of Michigan as one that is intended to use a total of more than 100,000 gallons of hydraulic fracturing fluid. The reports indicate if the text is addressing oil and gas development in general, HF, or HVHF. Finally, material in the technical reports should be understood as providing a thorough hazard identification for hydraulic fracturing, and when appropriate, a prioritization according to likelihood of occurrence. The reports do not provide a scientific risk assess- ment for aspects of hydraulic fracturing.
    • GRAHAM SUSTAINABILITY INSTITUTE INTEGRATED ASSESSMENT REPORT SERIES VOLUME II, REPORT 5 HYDRAULIC FRACTURING IN THE STATE OF MICHIGAN Public Health and Hydraulic Fracturing in Michigan SEPTEMBER 3, 2013 Faculty Lead NILADRI BASU, PHD ASSISTANT PROFESSOR OF ENVIRONMENTAL HEALTH SCIENCES, UNIVERSITY OF MICHIGAN SCHOOL OF PUBLIC HEALTH Research Assistants MARK BRADLEY, CASEY McFEELY, MARIE PERKINS UNIVERSITY OF MICHIGAN SCHOOL OF PUBLIC HEALTH TABLE OF CONTENTS 2 Executive Summary 3 1.0 Introduction 6 2.0 The Workplace Environment 14 3.0 The Surrounding Environment 21 4.0 The Community Environment 25 5.0 Challenges and Opportunities 28 6.0 Proposed Prioritized Pathways for Future Work 28 7.0 Acknowledgment 29 Literature Cited THIS PUBLICATION IS A RESULT OF WORK SPONSORED BY THE UNIVERSITY OF MICHIGAN Direct questions to grahaminstitute-ia@umich.edu
    • 2 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 EXECUTIVE SUMMARY Overview T he purpose of this report is to document and discuss public health issues related to hydraulic fracturing in the State of Michigan. As a report focused on public health, it is guided by the definition put forth by the American Schools of Public Health, which states that public health is “the science and art of protecting and improving the health of communities through education, promotion of healthy lifestyles, and research for disease and injury prevention… public health works to prevent health problems before they occur”. Therefore, this report is concerned with potential hazards that are not only chemical (i.e., toxicants) but also those that are physical, biological, and psychosocial. In many ways, this report may be considered a hazard identification piece. While a number of hazards have been identified, largely from studies performed outside of Michigan, there exists limited data linking these hazards with human expo- sures, thus limiting the opportunity to conduct a risk assessment. This report is focused on reviewing public health risks and benefits in three primary environments: 1) the hydraulic fracturing work- place; 2) the ecological environments that surround facilities; and 3) the communities that are situated near facilities. For each of these three settings, a number of potential hazards are identified, perti- nent human exposure routes are described, and associated human health effects are listed. Possible hazards in the workplace include accidents and injuries, exposure to excessive noise and inade- quate lighting conditions, exposures to silica and industrial chem- icals, and shift or night work. Possible hazards in the surrounding environment include impaired local and regional air quality, water pollution, and degradation of ecosystem services. Possible hazards in nearby communities include increased traffic and motor vehicle accidents, stress related to risk perception amongst residents, and boomtown-associated effects such as a strained healthcare system and road degradation. Public health benefits exist, and these may include improved economic conditions in communities and energy or pollution tradeoffs related to a shift away from coal and oil. The risks and benefits can vary from local to global scales. Key Challenges and Opportunities The greatest challenge to understanding the potential public health risks of hydraulic fracturing in Michigan is the lack of State- specific data. While thousands of wells in Michigan have been pro- duced (or are currently in production), the potential public health risks related to these facilities have been poorly documented. Further, most of these are not considered high-volume operations. The lack of objective, scientific-data specifically from Michigan related to high-volume hydraulic fracturing provides a challenge for local risk assessors and decision makers. Another challenge in any human health study is related to epide- miological causation. Even if a hazard is identified at a hydraulic fracturing facility, linking it to an adverse health outcome through an exposure-disease model requires carefully designed epidemi- ological studies that include, for example, a robust sample size, state-of-the art exposure assessments, an appropriate control population, temporal sampling to assess potential latent effects, and consideration of cumulative impacts from multiple stressors. Ideally, these efforts would already be underway given the extent of existing hydraulic fracturing in the State. One major criticism of risk assessments from other States concerns a lack of pre-frac- turing, baseline data. Baseline data is needed in order to discern whether or not hydraulic fracturing operations are causing health impacts on workers, ecosystems, or communities. Timely action towards gathering baseline public health data may allow Michigan the opportunity to overcome challenges faced in other states related to poor or missing data. There has been some disclosure of the chemicals used at hydraulic fracturing sites in Michigan, perhaps representing a small sample of Michigan-specific data. However, this disclosure is minimal, with only a few facilities reporting upon a small number of drilling events. In addition, the disclosed information does not include data regarding the amount of chemicals used. Thus, it is of limited use in evaluating potential risks and weighing them against poten- tial benefits. A survey of Michigan residents found that a majority of respon- dents have heard about hydraulic fracturing, and that many (45%) had a negative reaction to the term ‘fracking’. However, more than 50% of respondents believe natural gas drilling will provide more benefits to the State, and more than 50% either ‘strongly’ or ‘some- what’ support the activity. The lack of objective, science-based evidence may be leading to skewed risk perceptions and debate among stakeholders. Any public health assessment of hydraulic fracturing in Michigan needs to be conducted with careful consideration of other energy sources, relative tradeoffs, and associated public health risks and benefits. These risks and benefits will vary from the local to regional/state, national, and global levels. Proposed Prioritized Pathways for Phase 2 • A need for Michigan-specific public health data that ranges from the health of workers to ecosystems to communities. • There is a need to increase understanding of the fluids used in the hydraulic fracturing process, such as which chemicals are used, their exact volumes, recoveries at different stages of extraction and associated processes, disposal, etc.
    • 3 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 • Public health education and outreach to increase understand- ing by stakeholders. • Consideration of the risks-benefits of hydraulic fracturing in Michigan versus other energy sources, in terms of both health outcomes and economic value. 1.0 INTRODUCTION 1.1 Background H ydraulic fracturing is growing rapidly across many regions of the United States and worldwide1 . While the rate of hydraulic fracturing development in Michigan is not as great as other States, as elabo- rated by Wilson and Schwank2 , Michigan has long been known to have plentiful natural gas reserves. These reserves are largely in the Antrim Shale, though with increased geo- chemical knowledge and newer technologies there has been interest in deeper reserves sit- uated within the Utica and Collingwood Shales (Figure 1). As discussed by both Wilson and Schwank (2013) and Zullo and Zhang (2013) in this series, it is currently not economically feasi- ble to develop these deeper shale gas reserves, though increases in the price of natural gas could change this. More than 12,000 wells have been hydraulically fractured in Michigan3 . Most of these have been shallow, vertical hydraulic fractures, requiring relatively low volumes of hydraulic fracturing fluid4 . Stimulation of such reserves is not new, but rather, the activity in certain regions has increased in recent years because new technologies (e.g., horizontal or directional drilling, multi-stage platforms) have become available and proven to increase extraction, enabling production of previously inaccessible natural gas reserves in an uncon- ventional manner. Furthermore, hydraulic frac- turing is being hailed as an activity that will help grow the economy, create jobs, and lessen the dependency upon foreign energy suppliers. Hydraulic fracturing using high volumes of water and new technol- ogies has proliferated in recent years, though there exists a dearth of knowledge concerning its public health impacts5 . Much of the public health information available is based on anecdotes and non-peer-reviewed science. There are some emerging peer-re- viewed publications and a number of active studies being pursued by governmental and non-governmental agencies, industry, and academics. Thus, in coming years the outputs of these studies will be important to risk assessors and decision makers. 1.2 Objective The purpose of this report is to document and discuss public health issues concerning hydraulic fracturing in the State of Michigan. In doing so, it is guided by core principles in public health and human health risk assessment, both of which are briefly reviewed below. Given that hydraulic fracturing in Michigan consists of both high-volume and lower volume operations, and that at this Figure 1: Oil and gas wells in the Lower Peninsula of Michigan. Source: Michigan Department of Environmental Quality. OIL AND GAS WELLS SOUTHERN PENINSULA OF MICHIGAN ± 0 10 20 30 405 Miles ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! !! !! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! !! ! !! !!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !!!!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! !!! ! !! ! ! !! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! ! !! ! !! ! !! !! !! ! ! ! ! ! ! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !!! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !!! !! !! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !! !! !! ! !!! ! ! ! ! ! !!! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !!! ! !! !! !! !!! ! ! ! ! ! ! !! !! !!! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! !!!!! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! !! ! !! ! ! ! !! ! ! !! ! !!!! ! ! !!! !!! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! !!!! !! !! ! !!!! ! ! ! !! !! !! ! !! !!! !! ! ! ! ! ! ! ! !! !! ! ! !! ! !!! ! ! !! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! !! ! ! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! !! ! !! ! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!! ! !! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !!! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! !!! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !!! ! ! !!!! ! ! ! !!!!!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! !! !!! ! !! ! ! !!! ! !! ! !!! !! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! !!! ! ! !! ! ! ! !! ! !! !! !! !!! !! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !!! ! ! ! !!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!!! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !!!! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! !! !!! !! ! ! !!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !!!! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! !!! ! !! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! !! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !!!! ! ! !!! ! !!! ! !!! ! !! ! !! !!! !! !!!!! ! ! !! ! ! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! ! ! ! ! !! !! ! ! !!!! ! ! !! ! !! ! ! ! ! ! !! ! !!!! ! !! !!! !! ! ! !!!!!! ! ! ! ! !! !!! ! !!!! !! !! ! ! !! ! !!! ! ! !! ! ! ! ! ! !!!! ! ! ! ! !!! !!! !!!! !!! ! ! !! !! ! !!! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! ! ! ! !! !! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! !! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! !! !! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!! !! ! !! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !!!!! ! ! !! ! ! ! ! !!! ! ! ! !! ! !! !! ! ! ! !! !!! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!!!!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! !!!!! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! ! !! ! ! ! ! !!! !!!! ! !! ! ! !! !!! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! ! ! !! !!! !!! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !!! !! ! ! ! !! !!!! !! !! !! !!!! ! ! ! !!! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! !! ! ! !! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! ! ! ! !!! ! !!! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!!! ! ! ! ! ! !!! !! ! ! !! ! ! !! !! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !!! ! !! !! ! !! ! ! !! ! !!! ! !! ! !! ! !! !! ! ! ! ! ! ! ! ! !!! !!! ! ! !! ! ! !!!!! !! !! !! !! ! ! ! !! ! ! ! !! !!! !! ! !! ! !! !! ! !! ! ! ! ! ! ! ! !! ! !!! !! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !!! !!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !!! !! !! ! ! ! ! !!! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !! !! !! ! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! !!!!!! ! ! ! ! !! ! ! ! ! !!! ! ! !!! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !! ! !! !!! ! ! ! ! !! ! !! !! !!! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!!!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! !! ! !! !! ! ! ! ! ! !!! !!! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! !! !! !! !! ! ! ! ! !! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !!! !! ! ! ! ! ! ! !!! !!! ! ! ! ! !! ! ! ! !! ! !!! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !!! !! ! !!! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! !!!!! !! ! ! ! ! !! !! !! ! ! ! ! ! !!! ! ! ! !!!! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! ! ! !! ! !! !! !! !! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !!!!! ! ! ! ! ! !! !!! ! !! !! ! !!!!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! !!!! ! ! !! ! ! ! !! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! !! !! ! ! !!!! ! ! ! ! ! !! !! !! ! ! ! ! ! !! !! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! !!! ! ! ! ! ! !! !! ! ! !! ! ! !! ! ! !! ! !!!! ! ! !! ! !! ! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !!!! !! ! !! ! ! ! ! ! ! ! !!!! ! !! !!!!! ! ! ! ! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! ! ! !!! !! !!! !!! ! !! ! ! ! ! ! ! !! !!!! ! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! !! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! !! ! !!!! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !!! !! ! ! ! ! !! !! !! ! !! ! !! ! ! ! !!! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !!! !! ! !! ! !!! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! ! ! ! !! ! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! ! !! ! ! ! !! ! !! ! ! !!! !!!! ! ! ! !!! ! ! ! !! ! !! ! ! ! !! ! !!!! !! ! !! !! ! ! !! !!! ! ! ! ! ! !! ! !! !! ! !! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !!! ! !!!! ! ! ! !! ! ! ! !! !!! ! ! ! !! !! !!! ! !!! ! ! ! ! ! !!! !!! ! ! ! ! ! !! ! !! ! !! ! !! ! ! !! ! !! ! !! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! !! ! ! ! ! !!! !!! !! ! ! ! ! !! ! !! !! ! !!!!! ! ! !! !! ! !! ! !! ! ! ! !! !! !!! ! ! ! ! ! ! ! !!! ! ! ! !! ! !!!! ! ! ! !! ! !!! ! !! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !!! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !!! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! !! ! ! ! ! !!! !! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! !!! !! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !! ! !! ! !! !! ! !! !! ! !! ! ! !! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! !! !!! ! ! ! ! ! ! !!! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! ! ! !! ! !! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! !!! ! ! !!!!! ! ! ! !! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! !! ! !!! !!! ! !!!!! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!!!! ! ! ! !! ! ! !!!! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! ! ! ! !! !!! !! !! ! !! !! ! !! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !!! ! !!!! ! ! !!! ! !! ! ! ! ! !!! !! !! ! ! ! !! ! ! ! ! ! !!!!! ! !!! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! !!! !! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! !!! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!! !! !! ! ! ! ! !!!!!! ! ! ! ! !!!!!!! ! ! !!! ! ! ! !!! ! ! ! !!! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! !! ! !! ! ! ! !! ! !!!!! !! ! ! !! ! !! !! ! ! !! ! ! !!! !! !! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!!!!!! !!!!!!! ! ! ! ! ! ! !! ! ! !!!!!!!!! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! !!!!! ! !! ! ! ! ! ! !! ! ! !!! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !!!!!!! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! !! ! ! !! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !! ! ! !! !! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! !!! ! ! !! ! ! !! !!! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !!! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! !! ! ! !! ! !! ! !! !! ! ! !!! !!! !! ! !!! !! ! !! ! ! !! ! ! ! ! !! !! !! ! ! ! ! ! !! !! !! ! !! ! !! !! ! !!! ! !! !! !! ! ! ! ! ! !! !! ! ! !!!!!! ! ! !! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! !! !! !!!!! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! !!! !!! ! !!! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !!! ! ! !! !! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!!! ! !! ! !!!! !! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! !! ! ! ! ! !! ! ! !!!! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!!! !!!! ! ! !!!! ! ! ! !!! !!! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !!!! ! ! ! ! ! !!! ! ! ! !! !! !!!! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !! ! ! !!!!! ! ! ! ! ! ! ! ! !!!!!!! !! !! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! !! !! ! ! ! !! ! !! ! !! !! !! !! !! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! !! !!!! !! ! ! ! !!!!! ! !! ! !!!! !! !! !! ! !! ! !!!! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !!! ! !!!! ! !!! !! ! ! ! !!! !! !!! ! ! !! ! !! ! ! !! ! ! ! ! !!!! !!!!! !! !!! ! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! ! !! !!! ! !! !! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! !! ! ! ! !! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! !! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! !!! !!!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! !!! ! !! !! ! ! !!!! !!! ! !! !!! ! !!! !!!!! ! ! ! !! ! !! ! ! ! !!! ! !! ! ! ! ! !!!! !! ! !! ! !! ! !! ! !!! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! !! ! ! ! ! !!! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! !! !! !! ! !!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !!!! !! !! !!!! !!!!!!!!! !!!!! !! !!!! !! !! !!!! !! ! !!!!!! !! ! ! !!!! !!!! !!!!! !!! ! ! !! ! !!!!!!!!!!!!!!!!! !!!! !! !!!!! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! !!! !! !! ! ! ! ! ! !!!!!!!! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! !!! !! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! !! !! !! ! !! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! !! !! !!!!!! !! !! ! !! !!!! !!!!!!!!!! ! ! !! !!! !! ! !!! !! !! ! ! ! ! ! ! ! !! ! !! !!!!!! !! !! !! ! ! !! ! ! ! !! ! !!! ! !! ! !! ! !! !!! !!! !! !! ! !! ! ! ! ! ! ! ! !! ! !! !! !! ! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! !!! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! !!! !! ! ! ! !!!! ! ! ! ! !! ! ! ! ! !! !! ! ! ! !! ! !!!!! ! ! !! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !!!!! ! !! ! ! ! !! ! ! !! ! ! ! !! ! ! !!! !!!! ! ! ! ! ! ! ! ! !!!! !! ! !! ! !!!!! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! !! !!! ! !! !! !!!!!! ! !! ! ! !!! !!!!! ! ! ! ! ! !!! !!! ! ! ! !! !! ! !!!! ! !! !! ! !! ! !! ! ! !!!!!!! !!! !! !! !! ! ! !! !!! !! !! !!! ! !! ! !! ! !! ! ! ! !! !!! !! ! !!!! ! !! !!! ! !! ! ! !! ! ! !!! ! ! ! ! ! ! !!!! !! ! !! ! ! !!! !!!! !!! !!! !!!! !! !! !!! !! !!!! ! ! !! ! ! ! ! ! !!!! ! ! !!! ! !! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !!!!!!!! ! ! ! ! !! ! ! !! !!!!! !! ! !!! !!! ! ! ! ! ! !!!! !! !! ! ! ! !! ! !!! !!! ! !! ! !! ! !!! !!! !!! ! ! ! ! ! ! ! !! !! ! ! !!! ! ! ! !! !! !! ! !! !!! !! ! ! ! ! !! !!!!!! !! ! ! !!! ! ! !! ! ! ! !!! ! ! ! ! ! !!! !!!! ! !!!!! ! ! !! ! !! !! ! !! ! ! ! !!! !! !!!! ! ! !!!! ! ! ! ! ! !! !!!!!! !!!! ! !!! !!!!! !! ! !!! ! ! ! ! ! !! !!! ! ! !! !! !! ! ! !!!! !!! ! !!! !!!!!!!!!!!!!!!!! !!! ! !!!!!! !! ! ! ! !!! ! !!!! ! ! ! ! !!! ! ! ! !!! !!! ! ! !!!!!!!!! !! !!!!!!!! ! !!!!! !!! ! !!!! ! ! ! ! ! !! ! !!! ! !! ! !!! !!! ! ! ! ! !! !! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! !! !! ! ! ! !!! !! ! ! !! ! ! ! !!! ! !! !! ! !! !!! ! !! ! !! !! ! ! ! ! !!! !! !! !! !!! ! ! ! !! ! !!!! !!!!!!! ! !! ! ! ! !!! ! ! ! ! ! !!!! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! !! !!!!! !!!! !! !! !! ! ! ! ! ! ! !! !! !! !! ! ! !!!!! ! ! !!!! ! !!! ! !!! !! !! ! ! !!!!! ! ! ! ! !! !! !!!! ! ! !!!! ! ! !!! !!! ! !!! ! !!! ! !! ! ! !! !! ! ! ! ! ! ! !!!!! !! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! !!! !!! ! ! ! !! ! ! ! !! !! !!! !!!!! ! ! ! ! ! ! ! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!!!!!! ! ! ! ! ! !! !! !!!! !! !!! !! !! ! ! ! ! !! !! !!! !! !!!!! ! ! ! ! ! ! !! !!! ! ! ! !! !!!! ! ! !!!! !!! !!! ! ! ! ! ! !! !!! ! ! ! ! ! ! !! !!! !! ! ! !!! ! ! ! !! !! !! ! !!!! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !!! !!!! !! ! ! ! ! ! ! ! !! ! ! ! ! !! !!!! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! ! !! !! ! ! ! ! !! !! ! ! !!! ! ! ! !!! ! !!! ! ! ! ! !! !! ! ! ! ! !!! ! ! ! ! ! ! !! ! !! ! !! ! !! !!! ! ! !! !!!! ! ! ! !! ! ! ! ! !!! ! ! ! !!!! !!!!!! ! ! ! !! ! !! ! !!! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!!!! ! ! ! ! !! !! ! ! ! ! !!!! ! ! ! ! ! ! ! ! !! !! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! !!! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! !! !! !! !! ! !!! ! ! ! ! !! ! !! ! !! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!! !! ! ! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! !!! ! !!!!!! !!! ! ! !!! ! ! !! !! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !!! !!!!! ! ! !! ! ! ! !! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !!! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! !!! ! !! ! ! ! ! ! ! !!! ! !!! ! !! ! !! ! ! !! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! !!! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!! ! !! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! !! !!! !!!!! ! !! !!!!! !!! ! !!!! ! !! ! !! ! !!! ! !! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !! !! !! !! ! ! !! !!! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! !! ! ! ! ! ! ! !!!!!!! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!! ! !!! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! !!!! !! ! !!! !! ! ! !! !! !! ! ! !! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !!! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !!!!! ! ! ! ! ! !!!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! ! !!!! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! !!!!!!!!! !!!! !! ! ! !!!! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! !! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! !! !! !!! !! !! !! ! ! ! ! !!! ! ! ! ! ! !!!! !! ! ! !!! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! !! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !!! ! ! ! !!! !! ! !! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! !! !! ! !! ! ! ! ! !!!! !! ! ! !!! ! ! ! ! ! !!! ! ! ! !! !! !! ! ! ! ! ! ! ! ! ! ! !! !! ! !! !! ! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !! !! ! !! ! !! !!! ! !! !! ! !! ! ! ! ! !! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! !!!!!! ! ! !!! ! !!! !! !! ! ! !!!! !! ! ! !!! !!! !! ! ! ! !! !! !!! ! ! ! ! ! !!! ! ! ! ! ! !! !! !!! !!! ! ! !!! ! ! ! ! ! !!!! ! ! ! !! ! !!!! ! ! ! ! !! ! !! !! ! !!! ! !! ! ! !!! ! ! ! ! ! !! !! !! !! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!!!! !! !! ! ! ! ! ! !! !!! !!! ! !!! !! !!!! !! !! ! ! ! ! ! !! ! !! ! !! !! !!!!! !!!!! !! !!!!!!!!!!! !! !!! !!!!!! ! ! !! !! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! !! !! !! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! !! ! !!!! ! !! ! !! ! !! ! ! !! !! ! ! ! ! !! ! !! ! ! !! ! ! !!! ! ! ! ! ! ! !!! ! ! ! ! !! !!! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! !!! !! ! ! ! ! ! ! ! !! !!!! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! !!!!! ! !!! ! !! !! ! !!! ! !!! ! !!! !! !! ! !!! !!! ! ! ! !!!! ! ! ! ! ! ! ! !!!!!!!!!!!!!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! !!! ! !! !!!!!!!!!! ! ! ! ! !!!!!!! ! ! !!!!!!!!! !! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!!!!! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!!!! !!!! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! !!! ! !! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! !! !! ! ! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! !! ! !! ! !!!! ! !!!! ! !!! ! ! ! !! ! ! !!!! ! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! !!!!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! !! !! ! ! !!! ! ! ! ! ! !! !! ! ! ! !!!! ! ! !!! ! ! ! !! ! ! !! ! !! !! ! ! ! !! !!! !! ! !!! !! !!! ! ! ! ! ! ! ! !!! !!!! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! ! ! !! ! ! ! !!! ! ! ! !! !!!!! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! !! ! ! !! ! !!!!!!! ! ! !! ! ! ! ! ! ! ! ! ! !!!!! !!!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! !! ! !! ! ! !! !!!! ! !!! !! !!!!! !!! !! !!! !!!! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! !! !! ! !!! ! ! ! !! ! !!!! ! !!! !! !! ! ! !! !! ! ! !! ! ! !! !!! !! !! ! ! ! ! !! !! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! !! ! ! !!! ! ! ! !!! ! !! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! !!! !! !! !! !! ! !!!! ! ! !!! !!!!! ! ! !!!! ! !!!! !!!!!!! !!! ! !! ! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! !!!!! ! ! !!!! !!!!! ! !! ! ! !!!!!! !! ! !!! ! ! ! !! ! ! !! ! !!!!! ! !! !!! ! ! ! !! !! ! ! ! ! ! !! !!!! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! ! ! ! ! ! !! !!!! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! !!!!!!! ! ! ! ! ! !! ! !! ! ! !!! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! !! !!! ! ! ! ! ! ! !! !! !!!!! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! !!!!! ! ! ! !!! ! !! ! ! !! ! ! ! ! !!! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! !! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!! ! ! ! !! ! ! !! ! ! ! ! ! !!! ! ! !!! ! ! ! ! ! !!! ! !! ! ! !!!!!! ! ! ! !! ! ! !!! ! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! !! ! !! ! ! ! !! !!!!!! ! ! !!!!! !!! !! ! !! ! !! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! !! ! !! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! !! !! !! ! !!! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! !! !! ! ! ! ! !! ! ! ! ! ! !! !! !! ! ! ! ! ! ! !! ! !! ! !!! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !!!! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! !! !!!! !!! ! ! ! !! !! !! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! !!! !! !! ! ! ! ! !! ! !! ! ! !!! !!! !! ! !!! !! !!!! ! ! !!!! !!! !!!! ! !! ! ! ! ! !! ! ! ! !!!!! ! !! !! !! !! !! ! ! ! !!!!!!!! !! !! !!!!!!!!!! !!! !!!! !!!!!! !!! !!!!! ! ! ! !!! ! ! !! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !!!! ! ! ! !!!! ! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! !!!! ! !! !! !!! ! !! ! ! ! ! ! ! !!!! ! !! ! !!! ! ! !!!!!! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! !! ! ! !! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !!! !!! !! !! ! ! !!! !!!! !! !! !! !!! ! ! !!! !! !! ! ! !! !! !!! !!!! ! !! ! ! ! !! ! !! !! !! !! !!!!!! ! ! !!! ! !! ! !!!!! ! ! !!! ! ! !!!! ! ! ! !!!!!!!!!!!! !! !! ! !!! ! !! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! !!!! ! !! ! !! ! !! ! ! ! ! ! ! ! ! !! !!!! ! !!! !!! !!! !!! !!! ! ! !! ! ! ! !! ! ! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!! ! ! !! !! !!!! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! !!!! ! ! !! ! ! ! ! !!! !!!!! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! !!!! ! !! !!! ! ! !! !! !! ! ! ! ! ! ! !! !!!!! ! !!!! ! ! ! !! !!!!!!! ! ! ! ! !! ! ! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! !! ! !! ! !! ! ! ! !! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! ! ! ! !!!!! !! ! ! ! !!! ! ! ! !! !! ! !! !!!!!! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! !! ! ! !! !! !!!! !! !! ! ! !! ! ! !! !!! ! ! ! !!! !!!! !!! !! ! !! ! ! ! !! ! ! !!! ! ! ! ! !!!! ! ! ! !!! !! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! !!!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !!! ! ! !!! !! !! !! !! !!! ! !! ! ! ! ! !! !! ! ! ! !! !! !!! ! ! !! !! !!!!! ! !! !!!!!!! ! ! ! !!! !! ! ! !! ! !! ! ! ! ! ! ! !! !! ! ! !!!!! !! ! !! ! ! !!! ! ! !!!! !! ! ! ! !!!! !!! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! !!!! ! ! !!! !! !! ! ! !! ! !!!!!! !! ! ! !!! ! ! !! ! !! !!!!!!! ! !!!!! ! !! !! !!!! !!! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! !! !! ! ! !! ! ! !! ! !! ! ! ! ! ! !!! ! ! ! ! ! !!!!!!!!!!!!! ! !!! ! !! ! !! ! !! !! ! !!!!! !! ! !! ! ! ! !! ! ! ! !! !! !!!! !! !! ! ! ! ! ! ! !! !! !!! !! ! ! !! !!! !!! ! !! ! !! !! ! !!! !! ! !! ! ! ! ! !!! ! !! ! ! !!!!! !! !! ! !!! ! !!! !! !! !!!!!! ! ! !! ! !!! !!!! ! ! !! ! !!!!! !! !!! !!! !! ! !! ! ! ! !!!!! !! ! ! !! ! ! !! ! ! !! !!! !! !! ! !! ! ! !! !!! ! ! ! !!!! ! ! !! ! !! ! ! !!! !! !! ! ! ! !! ! ! ! !! !!!!!!!!!! ! !!! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! !! !! ! !! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! !! ! ! ! !! !! !!!! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! !! !! ! !! ! !!! ! !!!!! !!!! !! !!!! !! ! !!!! !! !! ! ! ! ! !! !! ! !!! ! ! ! ! !! !!! ! ! !! !! ! ! ! ! ! ! ! ! !!!! ! !!! ! !!! ! !! ! ! !! ! ! ! ! !! !!!! ! ! ! !! !! ! !! !!! !! !!!!!!! ! ! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !!!!!!!! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! ! !!!!! !!!! ! ! ! !! !! ! ! ! ! ! ! ! ! !! !!!!! ! ! ! !! !! !! ! !! ! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !!!!! !! ! ! ! ! ! ! ! !!!! !!!! !! !! ! !! !!! ! !! !! ! ! ! !! ! ! ! ! ! !! ! !! !! !! ! ! ! ! ! !! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! !!!! ! ! ! !!!! ! !! !! ! ! ! ! !!!! ! ! ! !! !!! ! !! ! !! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! !! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! !!!! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !!!!! ! !! ! ! !!! ! ! ! !!! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !!! !!!! !! !! ! !! ! ! !! !!!!!! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! !! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! !! !! ! !! ! !! ! ! ! !!!!! ! !! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! !!!!! ! ! !! !!! ! ! ! !! !! ! ! ! ! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! !! ! !! ! ! !! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! !! ! ! !! !!! !!!! ! ! !! !! !! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!! ! !!!!!!!!!! !! ! !!!! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!!! ! ! !! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! !! ! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! !! ! !! !! ! ! ! ! !! !! ! ! !! ! ! !!! ! ! ! ! !! ! ! ! ! !! ! ! !! ! ! !! !!! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !! ! !! ! ! !! !! ! ! !! ! !! !!! ! ! !!! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !!! !! !!!!!! ! ! ! ! !!!! !! ! ! !! !! ! ! !!!!! !! !!!!!!!!!! ! ! ! ! !!! ! ! !!! !! ! !!!!!! ! ! ! !!! !! ! !!! !! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! !!!! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!!! !! !! ! !! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! !! ! !! ! !! ! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! !! ! ! ! !!!!! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!!!!!!!! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !!! !!!! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !!!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! !!! ! !!!! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !!! ! ! ! !! ! !!! ! ! !! !! !! !! ! ! !!! !!! ! ! ! ! !!! ! !! ! !! ! ! ! ! !!! ! ! ! ! !! !! ! ! ! !!!!! !! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! !! !!!!!!! ! ! ! ! !! ! !! !! !! ! ! !!!!!! ! ! ! !! ! ! ! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!! ! !! !! !!!!! ! !! !! !! ! !! ! !!!! ! !! ! !! ! ! ! !! !! !! ! ! !!!!!!!! !!! ! ! !!! !! ! !!!! ! !!!!! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !!!! !!! !! !! ! ! !! ! ! ! !!! !!!! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! !!!! !!! ! !!! ! ! ! ! ! !! !!!!!! ! !! !!!!! !! !!!! ! !!!!!!! !! !! ! !!! ! ! !!! ! ! ! !!! !!!! !! ! !! ! ! !!! ! ! ! !! !! ! ! !! ! !! ! !! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !!! !! !!!! ! ! !!! ! ! ! !! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!!! ! ! !! ! ! ! !!!! ! ! !! ! ! ! !! ! !! ! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! !!! !!! !! ! ! ! ! ! !! !!! ! ! ! ! !!! ! ! !! !! !!! ! ! ! ! ! ! ! !!!!! !! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! !!!! ! !!!!!!!!!!! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! !!!!!!!!! !! !! !! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! !! ! ! !!!!! ! !!!!!!! ! ! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! ! !!!!!! ! ! ! ! !!! ! ! !! !! ! ! !!! ! ! !! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! !!! ! !! ! ! ! !! ! !!!!!!!! !! !! ! ! ! ! !! ! !!! !!! ! ! ! !! ! ! !! ! ! ! ! ! ! !! ! !! ! ! !!!!!!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! !!! !! ! ! ! !! ! !! ! !! ! !!!!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! !! ! ! ! ! !!! ! ! ! ! !! ! !! ! !! !!!!!! ! ! !! ! !!! ! !! ! ! !! ! ! !! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! !!!!! !! !! ! !!! !! ! ! ! ! !! ! !! !!!!!! !!! !!!! !!!! ! ! !! ! ! !! ! ! !! ! ! !! ! !! !!! !! !!!! ! !! !! ! !! ! ! ! ! !! !! ! !! !! ! !!!!!!!!!! ! ! ! ! ! !!!!! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!!!! ! ! ! ! ! !! !! !!! !!! ! !! ! !!!!!!!!!! !!!!! !!!! !! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! !!!! !! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! !!! !! ! !! !! ! ! !! ! ! ! ! ! ! !! ! !!!!! ! ! !! ! !! ! ! ! !! ! !! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !!!!! !! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! !! ! !! ! !! !!!!! !!!!! !!!! ! ! !! !!! ! ! ! ! !! ! ! ! ! ! !!!!!! ! ! ! ! !!! !!!! !!!!!! ! ! ! ! !! ! ! ! ! ! ! !!!!! ! !! ! ! ! ! ! ! ! ! !!!! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !! !! ! ! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! !! !! ! !!!! ! ! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!! ! !!!!! !! !! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !! ! ! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !!! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !!!! !! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !!! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!!! !! ! !! ! ! ! ! !! ! ! !! !! ! ! ! ! ! ! !! ! ! ! !!!! !! ! ! ! ! !!! !! ! !! !! ! ! ! ! !! ! ! ! ! ! ! !! !! !! ! !! ! !! ! ! ! ! !! ! !! ! !!! ! ! !! !! !! ! ! !! !!! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !! !! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! !! ! ! !!! ! ! ! ! !! ! ! !! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! ! ! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! !! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !!!!!!! ! ! !!! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! ! !! ! ! ! ! ! ! !! !!!! !! !!! ! ! ! ! !! ! !! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! !!!! !!!!! !!! ! !!! !! ! !!! ! !!! ! ! !!!! ! ! ! !!! ! ! !! ! ! ! ! !! ! !!!!! ! ! !!!! ! ! ! ! !!! !! ! !! ! ! !!!!!!! ! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !! !!!! ! !!! !! !!! ! ! ! !! ! ! !! ! !! ! !! !! !! ! ! ! !! ! ! ! ! ! ! ! !! ! ! !!!!!! ! ! ! ! ! !!!!! !! !!! !!!!!!! ! !!! ! ! !! ! ! !! !!! ! ! ! ! !!! ! ! !! !!! ! ! ! !!! !! ! ! !!!! ! ! ! ! ! ! ! !! ! !! ! !!!!!!! ! !! ! ! ! ! ! ! ! ! ! !!! !! !! ! !!!! ! !! !! !!!!!!!! ! ! ! ! !!! !! ! ! ! !! ! !!! ! ! !! !! ! ! ! ! ! !!! ! ! ! ! ! ! !! !!!!!! !! ! !!!!! !! ! ! ! ! !!!! ! ! !! !!! ! !! !!! ! ! ! ! !! !!!! ! ! !!!! !!! !! !!!! ! ! ! ! !! ! !! ! ! ! ! !! ! !! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! !! !!! ! !!!!!! !! ! ! ! ! ! !!!! !!! !! ! ! !! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! !! ! !!!! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!!! ! ! ! ! !! ! ! !! ! ! !! !! ! ! ! ! ! !!! ! ! ! !! !! ! ! !! ! ! !! !! ! !! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! !!! !! ! !! ! ! ! !!! ! ! !! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! !!! ! !!! ! ! ! !! !!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! !! !! ! ! !! ! ! ! ! ! ! ! ! !! ! !! ! ! !! !! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! !!!! ! !!!!!!!! ! !!!! ! ! ! ! ! ! !! !!! ! ! ! ! ! !! ! !! !! !! !! !!!!!!! !!!!! ! ! ! ! !!!! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! !!! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! !! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !!!!!! ! ! ! !!!!!!! !!!!!!!! ! ! !! ! ! ! ! !!! ! ! !! ! !! !!! !! ! ! ! ! ! !! !! ! ! ! ! ! !! !!! ! !!! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! ! !!!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !!! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !! !! ! !! !!! ! !! ! ! ! ! !!!! ! ! !! ! ! ! ! ! !!! !!!! ! ! ! ! !! !! ! !!! !! !! ! ! !! ! ! ! !!!!!! !!!! ! !! !! ! ! !! !! ! ! ! !!! ! ! !! !! ! ! ! !! ! ! ! !! !! ! ! ! ! !! !! ! ! !! !! !! ! !! !! !! ! ! !! ! ! ! ! ! !! ! ! !!! !! ! ! ! !!! ! ! ! ! !! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !!!! ! ! !!! ! ! ! ! !! ! ! !!!!! !!!!!!!! !! ! ! !! ! ! ! !!!!!!!!!!!!!! ! ! !!! ! ! ! !!! ! ! ! !!! !!! ! !!!! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! !!!!!!!! !!! ! !! !!! ! ! ! ! ! ! ! ! !!!! ! ! ! !!! !!! ! ! ! ! !!!!! !!!!!!!! ! !!!! ! !! ! ! !! ! !! !! !! !! !!! ! ! !!! ! !!!!!!!! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! ! !!!! ! ! ! ! ! !! !!!!! ! ! !! !! ! ! !!!! !! ! !! ! !!!!!!! ! !!!!!! ! ! ! !! ! !! ! ! !!! !! !!!!!!!! ! ! ! ! !! ! ! ! ! ! !!!!! ! ! !!!! ! ! !! ! !!!! ! ! ! ! !!! ! ! !! ! !!! ! ! ! ! ! !!!! ! !! ! !! ! !! ! !!!!!!!!!!!! !!! ! !!! ! ! ! !!!!! !! ! ! ! ! ! !! !! !!!!!!! ! !!!!!! ! !!!! ! ! ! ! ! ! !! ! ! ! ! !!! ! ! ! ! ! ! !!!! ! !! ! !!! !! !! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! !! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! !!! !! !! ! !!!!!!! ! !! ! !! ! ! ! !! !!! ! !! ! !! ! ! ! ! ! !!!! !! !! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! ! ! ! ! !!! ! !!! !!! ! ! !! ! ! !!! !!!!! ! !!!!! !! !! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! !!! ! !! ! ! ! !! ! ! ! !! ! ! ! ! ! !!!! ! ! ! ! ! ! !!!!!! ! ! !!!!! ! !! ! ! ! !!! ! !!! !!! ! ! ! !!! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! !! ! !! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !! ! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !!!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! ! !!! ! ! !! !!! ! !! ! !!!! !!! ! ! !! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! !! ! !! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! ! !! ! ! !! ! ! ! !!! ! !!! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! ! ! ! ! ! !! ! !! ! ! ! ! !!!! !! !! ! ! ! !! !! ! ! !! !!!! ! ! ! ! ! !!! ! ! ! !!!!!!! !!! !! ! !!!! ! !!! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! !!! ! ! !! ! !!! ! ! ! !!! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! !! ! ! ! !! !!! ! ! ! ! !! ! ! ! !! ! ! !! ! ! ! ! ! !!! !! ! ! !! !!! ! !!! ! !!!! !! !! !! !! ! !! ! ! !!! ! ! ! !!!! !!!! ! ! ! ! ! !! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! ! ! !! ! ! ! ! !! ! !!! !! !!!! ! ! ! ! ! !! ! !!! ! !!! ! !!!! ! ! ! ! !! !! ! ! ! !! ! ! !! ! !!!!!! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! !! ! !! !!! ! ! ! !! !!!!!!! !! ! !!! !! !!!!! !! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! !!! !!! ! !! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !! !! !! ! ! ! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! !!!! !! !! ! !! ! ! ! ! ! !!!!! ! ! ! !! ! !! ! ! ! ! ! ! ! ! !!!! ! ! ! !!!!!! !! !!!! ! !! ! !!! !!! !!! !!! !!! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! !! !!! ! !! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! !!! ! ! ! ! !! ! ! ! ! ! !! ! ! !! ! ! !!! ! ! ! ! ! !!!!! !!!! !!!!! ! ! ! ! ! ! ! !!!! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! !! ! ! !! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! !!! !!! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! !! ! !! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! ! !!!! !! ! ! !!!! !!! ! ! ! ! !! !! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! !!!!!! !!!! ! ! ! ! ! ! ! !!! ! !!! ! ! ! !!! ! !!! ! ! ! !!! ! ! !!! ! !! ! ! ! ! !!!! ! !! ! ! !! ! ! !!! ! ! ! !!!!! ! ! ! ! ! ! ! !! ! !! ! !!! ! ! ! ! !! !! ! !! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! ! ! ! ! !! !! ! !! ! ! ! ! ! ! ! !! ! ! !! !! !! ! ! ! ! ! ! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! !!! ! !!! ! ! !! ! ! ! ! !!! ! !! !!!! ! ! ! ! !! ! ! ! !!! !! ! ! ! ! ! ! ! !! ! !! ! ! !!! ! ! ! ! ! ! ! ! ! !!! !! !! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !!!! ! !!!! !! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! !! ! ! ! ! ! ! ! ! !!!! ! ! !!! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! !! ! ! !!! ! ! !!!!! ! ! ! ! ! ! ! !!!! !!!!! !!!! ! ! ! ! ! ! ! ! ! !!!!! ! !! ! ! ! !! ! !! ! !! ! ! ! ! !! !! ! !! ! ! !!!! ! ! ! ! ! ! ! !!! !! !! !! !!!!! ! ! ! ! ! ! !!! !!!!!! !! ! !! ! ! !! ! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! ! ! !! !!! ! !!!! ! ! ! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! !!!!!!! !! ! !! ! ! !!! ! ! ! ! !!!! ! !!!! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! !!!!!! ! !! ! !!! ! ! ! ! ! ! ! ! !!! ! ! ! !! !! ! !!! ! ! ! ! ! !! ! ! ! !!! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !!!!!!!!! ! !!!! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! !!! !! ! !! ! !! ! ! ! ! !! ! ! ! !!! !!! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! !! ! !!! !! ! ! !! ! ! ! ! ! ! ! ! !! ! !!! ! ! ! !! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! !! ! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! !!!! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!!! !!!!! !!!!!!!!!! !! ! !! !! ! ! ! ! ! !! !!! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! !! ! ! ! !!! ! ! !!! ! ! ! ! ! !! !! ! !! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !!!! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! !! !! !! ! !!!!! ! ! !! ! ! !! ! ! ! ! ! !!!! ! ! !! ! !! ! ! ! !!!!! ! !!! ! ! !! !! ! ! !!! ! ! !! ! ! ! ! ! ! !!!! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! !! !! ! !! !! ! !! !! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! ! ! ! !! !! ! ! !!! !!! ! ! !!! !! ! !! ! !! !! ! ! !! !! ! !! !! ! ! ! ! ! ! !! !! ! !! !!! !! ! !!!!! ! !!!!!! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! !! ! ! ! ! ! ! !!!! !! !! ! ! !! !!!!! !!!!! ! ! ! ! ! ! !! ! ! !! ! ! ! ! ! !! !! ! ! ! !! ! ! ! !!! ! ! ! ! ! ! ! ! !! ! ! ! ! !!! ! !!! !! ! !!! !!! ! ! ! ! !!!!!! !!!!!! !! ! ! ! ! ! ! !! !! ! ! ! !! ! ! ! ! !! !!! ! ! !! ! !! ! !!! ! ! ! ! ! ! !! ! ! ! !! !!!! ! ! !!! !! !! !! ! ! ! !! ! ! !! ! !! ! ! ! ! !! ! !! ! !! ! ! ! ! ! ! ! ! ! ! !! ! !!! ! ! !!!!!! ! !! ! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! ! !! ! !! ! ! ! !!!! ! ! ! ! ! ! ! !!!! !! ! !! ! ! ! !! ! !! !!! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !! !!! ! !! ! !! ! !!!!! ! !!! !!!! ! ! !!! !! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! !!!! ! ! ! ! !! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! !! !! !!! ! ! ! ! ! ! ! ! ! ! ! !! !! !!!! ! !! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! ! !! ! !!!! ! !!! ! !! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! !! ! ! !! !! !! ! !!! ! ! !! !!! ! ! ! ! ! ! !! ! ! ! !! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! !! ! !!! !!! !!!! ! !! ! ! !! !!! ! !! !! ! !!!!! ! ! ! ! ! !!! ! ! !!!! ! ! !! !! ! !! ! ! !!! !! ! ! ! ! ! !!! ! !!! ! ! !!! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! !! ! ! ! !! ! !!! ! ! !! !!!! ! !! ! ! ! ! ! ! ! !! ! !! !! !! ! !! !!! !! ! ! !!! ! ! ! ! ! ! !! !!! ! ! ! !!! ! ! ! !! ! !!! ! ! ! !! !!!! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! !! ! ! !! ! !!! ! ! ! ! ! ! ! ! ! !!!! ! !! ! !! ! ! ! ! ! ! !!! !!!!! ! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !!! ! ! !!!! !!!! ! !! ! !! ! !!! ! ! !! ! ! ! !!! ! ! ! !! ! ! ! ! ! !!!! !!! ! ! !! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! !! !! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! !! ! ! ! !! ! ! ! !! ! !! ! !! ! !!! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! !! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!! ! !! ! !! ! ! ! ! ! !!! ! !!! ! ! !! !! !!! ! ! ! ! ! ! ! ! !!! ! ! ! ! ! !! ! ! ! ! ! !! !! ! ! ! !! !!!!!! !! ! !!! !!!!!!! !! !! ! ! ! ! !! ! !!! !!!!! !!! !!!! !!! !!!!! !!! ! ! ! ! ! ! ! ! ! ! ! ! !!! ! !! ! !!! ! ! ! ! ! ! ! !! ! ! !!!!!!! ! ! ! ! ! ! !!! !!!!! ! !! ! ! ! !! ! ! !! !!!! ! ! ! !! ! ! ! !!! ! !! ! ! ! ! !!!! !!! !! ! ! ! !! ! ! ! ! ! !!!! !! ! ! ! ! ! !! !! ! !!! ! ! !! !! ! !! ! !! ! ! ! ! ! ! !! ! ! ! ! !! ! !! ! ! !!!!! !! !!! ! ! ! ! ! !!! ! !! !!! ! ! !! ! ! ! !! ! ! ! !! !! ! ! ! ! !!! ! ! ! !!!!!!! ! ! ! ! ! !! ! !! ! !! ! ! ! !! ! !!!! ! ! ! ! !! ! ! ! !! ! ! ! ! !!! !! ! ! ! ! ! !! !!! ! ! ! ! !!! ! !!! ! !! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! !!!!!!!!! !! ! ! ! ! !! !!!!!! !! ! ! !! ! ! ! ! !! ! ! ! !! ! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! !! !! ! ! !!! !! ! ! ! ! ! ! ! ! !! !!! ! ! ! ! ! ! ! ! ! ! !! ! ! !! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!!!! ! ! ! ! !! !! !!!!!!!!! !!!! !!!!!!! ! !!!!! ! !! ! !! ! ! ! !! ! !!!! !!!! ! ! ! ! ! !! !! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !!! !! ! ! ! ! !! ! ! ! ! ! !! ! !! ! ! ! ! !! ! ! ! !! !! ! ! ! ! ! ! ! ! !! ! ! ! ! ! ! !!!! ! ! !! ! ! ! ! !! ! ! ! ! ! !! ! ! ! ! !! !! ! ! !!! !!!! ! !!!!!! !! ! !!!! ! ! ! ! !! ! ! ! ! !!!!! ! !! !!! ! ! ! ! ! ! ! ! ! ! ! !!! !!!! ! ! ! !! ! ! !!! ! !!!!!! ! ! !! !! ! ! ! !! !! !! !! !! ! ! !! !! ! ! ! ! !!!!! !! !! ! !! !! ! ! !! !!! ! ! ! !! ! ! ! ! ! !!!!! ! !!!!! ! !!!!! !! !! ! ! !! !! !!! !! ! !!!! ! ! ! !! !! !! ! ! ! !! ! ! ! !! ! ! ! !!!!! !! ! !! ! ! ! ! !! !! ! ! ! !! !! ! !!!! !! ! ! !! ! ! ! ! ! ! !!! ! ! ! ! ! !!! ! ! ! ! ! M h gan Depa men o En onmen a Qua O e o O Ga and M ne a Sep embe 2012 O L AND GAS WELL LEGEND W W W W W W W ! !
    • 4 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 stage we cannot necessarily discriminate between them in terms of public health relevant data, this paper aims to discuss the issue in broader terms. Public health is concerned with preventing illness and improving the health of populations. The American Schools of Public Health (ASPH) describes public health as “the science and art of protect- ing and improving the health of communities through education, promotion of healthy lifestyles, and research for disease and injury prevention… public health works to prevent health problems before they occur”6 . The view from the World Health Organization, “a state of complete physical, mental and social well-being, and not merely the absence of disease or infirmary”, is also compre- hensive7 . Therefore, a public health assessment of hydraulic frac- turing in Michigan needs to be inclusive and holistic, and consider a range of disciplines including occupational safety and health, air and water quality, public perceptions, and health economics. Risk assessment aims to estimate the nature and probability of adverse human health effects8 (Figure 2). The first step is Hazard Identification, which aims to identify potential threats and to qual- itatively address any adverse consequences associated with these threats. It is important to note that hazard and risk are not equiv- alent terms; hazard refers to a broadly defined agent that could potentially cause harm, while risk refers to the probability of that agent actually causing harm. Exposure Assessment aims to quan- tify hazard exposure for individuals. Dose-Response Assessment aims to link an exposure or dose with the probability of an effect. The information gathered via Hazard Identification, Exposure Assessments, and Dose-Response Assessment is used to charac- terize risk and eventually make management decisions. As elaborated in the next section, substantial data gaps exist for hydraulic fracturing in the State of Michigan, thus a majority of this report may be considered Hazard Identification. This report is concerned with potential hazards that are not only chemical (i.e., toxicants) but also those that are physical, biological, and psy- chosocial. For organizational purposes, the report is broadly sep- arated into hazards found primarily in the workplace (Part 2), the surrounding environment (Part 3), and in nearby communities (Part 4). In reality, the same hazard may exist in different environments, and hazards from one environment may interact with a hazard from another environment. It is also important to realize that despite a long history of hydraulic fracturing in the State of Michigan, most of this activity has occurred with low volumes of water and via vertical drilling (compared to the high-volume, horizontal drilling that is of contemporary concern). As such, this report does not necessarily focus on potential public health impacts of one form of drilling over another, but aims to review information from both. 1.3 Limitations and Assumptions Substantial gaps in data availability, not only in Michigan but elsewhere, prevent a full assessment of public health risks associ- ated with hydraulic fracturing. Public policy should be grounded in strong, objective peer-reviewed science rather than anecdotes and beliefs. Speculative conclusions and opinions about possible hazards based solely upon anecdotes and oversimplified chronol- ogies are not a sufficient foundation to advance state regulatory reforms or policies. Nevertheless, health concerns expressed by community members, especially those with scientific plausibility and those recurring across temporal and spatial scales need to be taken seriously. In this report, all currently available evidence was reviewed and considered. A majority of the evidence reviewed in this report was obtained from studies undertaken in other States, and while we recognized that risks are often site-specific, for the purposes of this paper we maintain broad generalizations (for example, Table 1 provides some generalized parameters related to a typical high-volume hydraulic fracturing operation that may help orient the reader). The lack of exposure assessment information for any hydraulic fractur- ing site limits the ability to perform meaningful risk assessments. Therefore, the review of several hazards in this report should not lead the reader to conclude that all individuals, communities, or ecosystems associated with hydraulic fracturing are exposed to all of these hazards and are at health risk, as hazard is not the same as risk. The limited data also prevents us from adequately distin- guishing between low-volume vertical wells and large-volume and/ or directional drilling operations. Table 2 provides a snapshot of the public health issues covered in this report and the strength of evidence from the broader field and from Michigan specifically. Figure 2: Key steps in human health risk assessment. Hazard  Identification Exposure  Assessment Risk  Characterization Risk Management Exposure – Response Assessment Policy Science: • Research • Validation • Monitoring
    • 5 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 Figure 3: Report framework. WORKPLACE ENVIRONMENT COMMUNITY • noise, smells, light • dust, silica • chemicals (fluid,  others, gases) • water quality • air quality • ecosystem services &  landscapes • traffic, noise • env justice • boomtown culture • jobs/income • accident & injuries • CVD, respiratory • cancer • renal disease • others (dermal,  neuro) • risk perception • mental/behav health • distrust • living std & lifestyle • access to services • MV accidents 1. AREAS OF CONCERN 2. HAZARDS & EXPOSURES [real & perceived] 3. PUBLIC HEALTH  IMPACTS TABLE 1: Estimated base case and range values for a variety of activities related to high-volume hydraulic fracturing operations. Table modified from Jiang et al. 20119 and is reflective of operations in the Marcellus Shale. These numbers may not be representative of Michigan, but may serve as a basis for developing Michigan-specific numbers. Parameter Units Base case Range Area of access road Acres 1.43 0.1-2.75 Wells per pad Number 6 1-16 Area of well pad Acres 5 2-6 Vertical drilling depth Feet 8500 7000-10,000 Horizontal drilling length Feet 4000 2000-6000 Fracturing water MMgal/well 4 2-6 Flowback fraction Percent 37.5 35-40 Recycling fraction Percent 45 30-60 Trucking distance between well site and water source Miles 5 0-10 Trucking distance between well site and deep well injection facility Miles 80 3-280 Well completion time with collection system in place Hours 18 12-24 Well completion time without collection system in place Days 9.5 4-15 Fraction of flaring Percent 76 51-100 Initial 30 day gas flow rate MMscf/day 4.1 0.7-10 Average well production rate MMscf/day 0.3 0.3—10 Well lifetime Years 25 5-25
    • 6 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 1.4 Relation to Integrated Assessment The goal of the Integrated Assessment is to collate and analyze the environmental, social, and economic dimensions of hydraulic fracturing and its impact on Michigan communities, human health, and ecosystems. Accordingly, public health is an integral compo- nent of the overall Integrated Assessment as the scientific findings and tools of public health offer a scheme to integrate across disci- plines. As this report is principally focused on identifying hazards, Phase 2 of the Integrated Assessment should consider such public health hazards (and associated risk-benefits) when addressing the question - “What are the best environmental, economic, social, and technological approaches for managing hydraulic fracturing in the State of Michigan”. 2.0 THE WORKPLACE ENVIRONMENT T he occupational environment or workplace is a unique setting in public health, as it is a central environment where people may spend most of their time outside the home. In the workplace people may be chron- ically exposed to a diverse array of stressors, some of which can potentially occur in high concentrations10 . Hydraulic fracturing workers are thus a sensitive sub-group in public health that warrant focused study. Given that job creation is one argument in favor of the expansion of hydraulic fracturing and natural gas production in Michigan (though Zullo and Zhang (2013) in this series indicate the validity of this claim may be questionable, particularly in relation to the benefit received by local communities), the health hazards for the workers filling those jobs should be examined to inform an assessment of occupational cost-benefits. In this section we cover what is known about the hazards that may exist in a typical hydraulic fracturing operation, and where possible, we provide information specific to Michigan. Further, some of the hazards that exist within the worksite may also extend into the nearby environment (Section 3) and communities (Section 4). For introductory purposes, Table 2 provides an estimate of parameters concerning various activities in a hydraulic fracturing site that may help orient the reader. 2.1 Employment Any discussion of the workplace must also consider the employ- ment landscape. This topic is covered by Zullo and Zhang (this series) and briefly mentioned here. According to the US Bureau of Labor Statistics, in August of 2012 the national rate of unemploy- ment was 8.1%. Michigan had an even higher 2012 unemployment average of 9.4%11 . In regions of economic decline, community members often embrace energy development efforts because of the economic benefits12 . In October 2012, the Center for Local, State and Urban Policy (CLOSEUP) at the University of Michigan Ford School, conducted a telephone survey of 415 Michigan residents. Of those interviewed, via an open-ended question on the ‘primary potential benefit of fracking’, 20% were in support of hydraulic fracturing due to the industry’s ability to create jobs and stimulate local investments. Further, 82% of survey respondents indicated that natural gas was ‘very important’ (36%) or ‘somewhat important’ (46%) to Michigan’s economy13 . According to a 2010 IHS Global Insight report, the hydraulic fracturing industry has created over 600,000 jobs nationwide to date14 . This number is projected to reach 1.6 million by the year 2035. In Michigan, a 2012 IHS Global Insight report found that unconventional natural gas development accounted for 28,063 total jobs in 2010 and is projected to increase to 63,380 jobs by 203515 . Jobs within the hydraulic fracturing industry may fall into eight categories13 : general services, retail and wholesale trade, TABLE 2: Possible public health issues related to hydraulic fracturing from Sections 2, 3, and 4 of this paper. For each, we qualitatively assessed the available evidence and scored each hazard with an “X” mark (plausibility score). “X” = hazard is plausi- ble, anecdotes exist but no scientific studies found; “XX” = scientific evidence exists and is suggestive but also limited (e.g., few studies, poor design, con- founders); “XXX” = scientific evidence exists and is strong (e.g., many studies, good design, causality). The assigned scores are based on the expert judg- ment of the authors. In the final column, we indicate Yes (Y) or No (N) if any anecdotes, reports, studies or datasets were found in Michigan concerning that particular hazard, but did not judge the strength of the evidence. Issue Paper Section Plausibility Score Michigan Evidence? Injuries 2.2 XX N Noise 2.3 XX Y Light 2.4 X N Odor 2.5 XX Y Silica 2.6 XXX Y Intentional-Use Chemicals 2.7 XXX Y By-product Chemicals 2.8 XXX Y Transportation 3.1 XXX N Air Quality 3.2 XXX N Water Quality 3.3 XXX N Habitat and Wildlife Impacts 3.4 XXX N Food and Animal 3.5 XX N Earthquakes 3.6 XX N Public Perception 4.1 XXX Y Environmental Justice 4.3 XX N Health & Social Services 4.4.1 XX N Local Socio-Economics 4.4.2 XX N Bust or Recovery 4.4.3 XX N
    • 7 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 transportation and utilities, manufacturing, construction, mining, agriculture, and government. The largest percentage of workers falls in the general services category, which includes an estimated 45% of the 601,348 workers nationally. The lowest categories of job creation are in government and agriculture, both estimated at about 1% of workers nationally14 . When broken down by individual well operation, there is an average of 420 workers per location across 150 occupations. Approximately 90% of these jobs occur during the development and drilling phases of a hydraulic fractur- ing operation. Typically pre-drilling requires a time period of about six to nine months, with actual drilling occurring over two months16 . Thus the vast majority of jobs associated with hydraulic fracturing operations are short-term temporary positions, but workers may move across sites and thus these positions may be longer-term. A majority of hydraulic fracturing jobs require little post-second- ary education; however, most do require experience-driven skills, unique industry related knowledge, and are highly intensive12,16,17 . Long-term residents of rural communities often find the prerequi- site drilling experience an employment barrier, and thus workers are typically brought in from outside the community to fill such positions12 . While there is a lack of data tracking the residency of workers, many are thought to be transient or of a migrant worker population. According to a report on the history of Pennsylvania’s energy industry and the 2010 census data, about 40% of drilling companies utilized non-resident, temporary employees18 . This is especially true for the larger energy companies who hire from an international pool and have access to supply-chain services16 . These trends can be changed if local training and education facil- ities were more widely available to residents of targeted natural gas rich communities12 . It is important to also note that migrant workers may experience unique health effects19 . 2.2 Fatal and Non-Fatal Injuries and Illnesses Occupational hazards found at a typical hydraulic fracturing operation are similar to those present at most construction sites. Notable physical hazards may include slips and falls, hand and finger injuries, muscle strains, exposure to extreme temperatures and inadequate lighting conditions, fires and explosions, and inju- ries caused by moving vehicles, heavy equipment, high pressure lines, pinch points, and working in confined spaces20 . Long work hours and shift work can increase workers risk of fatigue, which can negatively affect safety and performance and lead to a number of health conditions such as cardiovascular disease and diabetes21 . Properly training workers in the safety protocols related to each job preformed can aid in avoiding many of these hazards. Also, every worker should have the appropriate personal protective equipment for the job that they will be performing and be properly trained on how to use this equipment. There is a lack of readily available datasets concerning rates of fatal and non-fatal accidents and injuries specific to the hydraulic fracturing industry. The U.S. Department of Labor, Bureau of Labor Statistics (BLS), reports fatal and non-fatal injuries and illnesses for the oil and gas industry as a whole, which includes natural gas extraction and hydraulic fracturing. Fatal injuries for the oil and gas industry between 2004 and 2008 ranged from 98 to 125 deaths annually. In 2008, the major causes of fatal injuries were transporta- tion (49%), contact with objects and equipment (30%), and fires and explosions (18%)22 . Based on BLS data, between 2003 and 2009 there were 202 motor vehicle fatalities for workers in the oil and gas sector, and the fatality rate for motor vehicle accidents is approx- imately 8.5-times greater than for other occupational sectors23 . In Michigan, there were no reported fatal injuries from 2008–2011 for the mining industry, which includes oil and gas extraction24 . In terms of non-fatal injuries, in 2011 the injury and illness rate for the oil and gas industry in the United States was 0.9 incidents per 100 workers, with over half of these incidents resulting in days spent away from work or job transfer or restriction25 . In Michigan, the non-fatal injury and illness rate for 2011 was 1.8 incidents per 100 workers for the mining industry as a whole24 . There are a few news reports on injuries and deaths from outside of Michigan. For example, in North Dakota a hydraulic fracturing worker was killed and another worker was injured on January 19th, 2013 at a site located north of Watford City. Reports indicate that a pipe became disconnected and fatally struck the worker in the head26 . There have also been reports that the traumatic injury rate in the Watford City region has increased 200% since 2007 from inci- dents related to the hydraulic fracturing operations there27 . Note that we were unable to find any news reports concerning accidents or injuries in Michigan. 2.3 Noise Pollution According to a 2008 literature review done by the University of Colorado School of Public Health, there were no studies published which focused on noise pollution associated with the oil and gas industry from 2003 to 200828 . However, throughout the process of hydraulic fracturing there are stages in which workers and nearby communities may experience heightened noise in the area. These activities include noise from drilling, well pumps, and compres- sors28 . Noise has been associated with negative health effects such as annoyance, stress, irritation, unease, fatigue, headaches, and adverse visual effects29,30 . Due to the fact that some hydraulic frac- turing operations happen 24 hours a day31 , all of the above activities also have the potential to interfere with sleep of area residents28 . Another potential noise source is truck traffic. The New York State Department of Environmental Protection estimates that 800 - 1200 trucks are required for an individual well32 , and thus residents living
    • 8 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 closest to hauling routes, especially children and the elderly, may be particularly vulnerable to increased noise pollution exposure33 . In Michigan, residents have complained of noise from oil and gas operations. A 2003 survey of newspaper articles and informal inter- views from Manistee and Mason Counties reported complaints about constant noise, such as from routine machine operation, and sudden noise from equipment malfunction or warning devices34 . We emphasize that this survey was not scientifically conducted or peer-reviewed, contains accounts disputed by State agencies, and does not pertain specifically to hydraulic fracturing operations. Nevertheless, we reference it here and in later sections because the anecdotal evidence may be relevant in future assessments, useful in anticipating potential issues, and pertinent to individuals’ perceptions of health-related risks. 2.4 Light Pollution Based on the Fracfocus website, during the first year (or more) of a hydraulic fracturing operation, the process can run 24 hours a day and seven days per week35 , with large industrial lights surrounding the well pad to keep it lit (note, in Michigan operations typically last 2 days in an Antrim well and up to three weeks for a deep well; Fitch and Goodheart, Michigan Department of Environmental Quality, pers. comm.). If drilling and activities occur at night, this may also cause stress and affect sleep in nearby communities. Light pollu- tion from artificial light has become a relatively new focus of public health due to its potential connection to increased breast cancer incidence28 . Light-at-night or LAN has been shown to decrease the production of melatonin and result in the increase of estrogen production36 . A study conducted in 2001 found that there was a 60% increased risk for women diagnosed with breast cancer if they worked a graveyard shift, defined as working between 7:00pm and 9:00am, once in the ten years before diagnosis37 . This becomes a health risk factor in the hydraulic fracturing industry because during early (i.e., 1-3 months) production, crews may be employed 24 hours a day35 . On the other hand, like other construction sites there may also be concerns related to poorly lit work environments. 2.5 Chemical-Related Odor A reported short-term effect of the hydraulic fracturing industry is a rotten egg-like odor commonly associated with odorous hydro- carbons2,27,38 . Hydrogen sulfide (H2S) can enter the air naturally from hydraulic fracturing sites, and at low levels of concentration it has a rotten egg smell28 . In the occurrence of a wastewater spill, Bamberger and Oswald39 documented local families becoming sick due to chronic exposure to sewage gas. It should be noted that this evidence was based on selected interviews and that the report has come under scrutiny by others, but nonetheless relevant people were interviewed and the paper was subject to scientific peer-review. While no odor threshold has been legally established in Michigan40 , in article R 324.1013 of Michigan’s 2006 Oil and Gas Regulations, nuisance odors are to be prevented in all stages of energy related development or production. Nuisance odors are defined as any single or combination of gas, vapor, fume, or mist emission in any quantity, which cause injurious health or quality of life effects. If any facility with the presence of hydrogen sulfide receives one or more complaints about its’ odor, the permittee of the facility must conduct numerical modeling in order to determine ambient air concentrations H2S. In Michigan oil and gas facilities must have H2S detection monitors, emergency breathing apparatus, and all free gas must be flared and tested for H2S40 . No scientific or peer-reviewed studies of chemical-related odor were identified with respect to hydraulic fracturing in Michigan, but some anecdotal evidence associated with drilling has been docu- mented. The 2003 survey of newspaper reports and interviews from Manistee and Mason Counties described reported cases between 1979 and 2002 where residents indicated detecting odors, in some instances allegedly persisting a week and causing nausea and headaches34 . While these reports were not from high-volume hydraulic fracturing sites, it is important to anticipate and recog- nize that some baseline odors may exist and that these may come from various aspects of oil and gas drilling. Again, we note that this survey was not published in the scientific peer-reviewed literature. 2.6 Silica Exposure On June 21, 2012 U.S. Department of Labor, Occupational Safety and Health Administration (OSHA) issued a hazard alert concern- ing the potential of hydraulic fracturing workers to be exposed to respirable crystalline silica41 . This hazard alert was the result of a study conducted by the National Institute of Occupational Health and Safety (NIOSH), which collected a total of 116 full shift air sam- ples from hydraulic fracturing sites in Arkansas, Colorado, North Dakota, Pennsylvania, and Texas. This research documented that 47% of the samples collected had silica levels greater than the per- missible exposure limits (PEL) set by OSHA and that 79% had levels greater than the NIOSH recommended exposure limit (REL)41 . In Michigan, to our knowledge, air samples from hydraulic fracturing sites have never been collected for the purposes of determining concentrations of silica. There is a long history of silica mining in the Upper Midwest, and dramatic increases in silica mining have occurred in both Wisconsin and Minnesota42,43 . While there exists no Michigan data concerning airborne silica levels, disclosure infor- mation from some hydraulically fractured Michigan wells includes “MI Specific 20/40 mesh sand” and “MI Specific 40/70 mesh sand” (trade names), indicating that at least some of the silica used in hydraulic fracturing in Michigan may be sourced from Michigan35 .
    • 9 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 The NIOSH study identified multiple sources of silica dust exposure from a typical hydraulic fracturing operation. Silica sand is used in large quantities during a hydraulic fracturing operation because it is added to hydraulic fracturing fluid as proppant. Proppants are pumped underground during hydraulic fracturing so that they become inserted into the fissures, keeping the fissure open and allowing the natural gas to flow out of the well. Large quantities of silica sand must be mixed with the hydraulic fracturing fluid. Once on site it is transferred using sand movers and conveyor belts, all of which produce silica dust. Silica dust is also produced by the heavy truck traffic that occurs at a hydraulic fracturing site. Silica expo- sure was shown to be the greatest for workers who serve as sand movers and blender operators, as well as individuals who may work downwind of such operations41 . However, workers located upwind and not in the immediate vicinity of the produced silica dust were also exposed to silica, likely through the dust generated from truck traffic. The exposure of workers upwind and distant from the site raises the possibility of silica exposures for community members (Section 4) as well. Prolonged inhalation of silica can lead to the lung disease silico- sis. There are different types of silicosis, which are dependent on the amount of silica a worker is exposed to and the duration of exposure. Chronic silicosis occurs from long-term inhalation of low quantities of silica and can take 20-40 years to develop44 . Acute and accelerated silicosis can occur in a much shorter time frame, 1-3 years, when workers are exposed to high levels of silica. Symptoms can include cough, shortness of breath, massive fibrosis, reduced lung function; silicosis can result in respiratory failure44 . There are multiple ways to reduce the amount of silica that hydraulic fractur- ing workers are exposed to; these include changes to how silica is transported, added engineering controls and protective equip- ment, and using an alternative proppant in hydraulic fracturing fluid (though OSHA also recommends evaluation of alternative proppants for possible health effects)41 . The 2012 OSHA Hazard Alert41 outlines many specific steps that can be taken to reduce the amount of silica dust produced and worker exposure to silica. Silica exposure has also been related to a range of other cardiovascular and respiratory diseases. 2.7 Intentional-Use Industrial Chemicals Workers may be exposed to a variety of industrial chemicals that are intentionally used in hydraulic fracturing. Chemicals are an inte- gral component of the hydraulic fracturing process and perform a number of functions as summarized in Table 3. In order to increase understanding of the potential health impacts associated with exposure to hydraulic fracturing fluids, studies have identified constituent chemicals and cross-referenced them with known or suspected health effects. The diversity of chemicals and formulations, coupled with varying site-specific factors, proves challenging when conducting risk assessments. For example, the number of different formulations (mean: 13, range: 3-37), prod- ucts (mean: 226, range: 67-450), and chemicals (mean: 132, range: 61-304) reported by service companies is wide-ranging49 . Here we briefly review two recent assessments, and also provide some data concerning chemicals disclosed to have been used in Michigan hydraulic fracturing facilities. In 2004 a project was initiated by The Endocrine Disruption Exchange (TEDX)50 to identify and classify chemicals used in the hydraulic fracturing industry45 . This research identified the use of 944 products, with a total of 632 chemicals reported to be in these products. A list of the five chemicals found to be included in the greatest number of products is provided (Table 4). Of the 632 chemicals reported, 56% (or 353 chemicals) had unique Chemical Abstracts Service (CAS) numbers assigned to them. As CAS numbers identify specific chemical substances (single chemicals, mixtures, etc.) which may have many different names, chemicals without unique CAS numbers provided can be difficult or impos- sible to identify as specific chemical additives, posing problems for evaluation of their health effects. Next, the researchers queried each chemical with a CAS number against various databases (e.g., MSDS sheets, TOXNET) to increase understanding of plausible health effects. In doing so, more than 75% of the chemicals were shown to possibly affect the respiratory and gastrointestinal sys- tems as well as eyes, skin, and other sensory organs. Nearly half (40-50%) of the chemicals could affect the neurological, immune, cardiovascular, and renal systems. One-quarter of the chemicals were either known, probable, or possible carcinogens. For exam- ple, benzene is a known human carcinogen and acrylamide is a probable human carcinogen. Finally, 37% of the identified chemi- cals could have effects on the endocrine system. The researchers also noted that 44% of the chemicals were not evaluated because they were not disclosed or they did not have adequate toxicolog- ical data. Waxman et al.51 , via the U.S. Congressional Committee on Energy and Commerce, also conducted a study to identify and classify chemicals used in the hydraulic fracturing industry. The Committee requested 14 leading oil and gas service companies to disclose the types and volumes of products used in hydraulic fracturing between 2005 and 2009, as well as the chemical composition of those prod- ucts. All companies approached voluntarily provided information. Collectively they reported using more than 2,500 products in hydraulic fracturing, and that these products contained 750 chem- icals. As indicated in Table 4, methanol was found in the greatest number of products, followed by isopropanol, 2-butoxyethanol, and ethylene glycol. During the time period under investigation, these companies also reported collectively using 780 million
    • 10 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 gallons of products (not including water) in hydraulic fracturing. Of this amount, 94 million gallons of 279 products contained a com- ponent that was deemed proprietary. In some cases, but not all, this information was disclosed. The study by Waxman et al.51 also revealed that products used in hydraulic fracturing included 29 chemicals of concern that are known or possible carcinogens, regulated under the Safe Drinking Water Act (SDWA), or listed as hazardous pollutants under the Clean Air Act (CAA). BTEX compounds (benzene, toluene, eth- ylbenzene, and xylene) were included within this list of chemical components of concern and were reported to be present in 60 products. There were also notable carcinogens including diesel (51 products), naphthalene (44 products), formaldehyde (12 products), sulfuric acid (9 products), and thiourea (9 products). Ninety-five products contained 13 different carcinogens, and more than 10.2 million gallons of hydraulic fracturing products included at least one carcinogen. The aforementioned studies conducted by Colborn et al.45 and Waxman et al.51 document that a large number of chemicals are used in hydraulic fracturing and that many of these chemicals have intrinsic toxic properties. We explored the FracFocus database to determine if these findings can be extended to Michigan. FracFocus is a national hydraulic fracturing chemical registry managed by the Ground Water Protection Council (a nonprofit for groundwater protection) and the Interstate Oil and Gas Compact Commission (a government agency, for oil and gas development and protection of health and the environment). More than 35,000 well sites have been registered nationally thus far, with 13 well sites in Michigan having reported data as of March 16, 201335 . Sometimes, a single well may need to be fractured more than once, thus requiring TABLE 3: Categories of chemicals used in hydraulic fracturing, their purposes, and an example of a commonly used chemical. Information in this Table was compiled from Colborn et al.45 , DOE report46 , Gosman et al.47 , and Encana48 . Functional Category Purpose Example(s) of Chemical Diluted acids Improve injection and penetration; dissolve minerals and clays to minimize clogging, open pores, and aid gas flow Hydrochloric acid Biocide Minimize bacterial contamination of hydrocarbons; reduce bacterial pro- duction of corrosive byproducts to maintain wellbore integrity and prevent breakdown of gellants Glutaraldehyde Breaker Added near end of sequence to assist flowback from wellbore; breaks down gel polymers Ammonium persulfate Clay Stabilizer Establishes fluid barrier to prevent clays in formation from swelling; keeps pores open; creates a brine carrier fluid Potassium chloride Corrosion Inhibitor Maintain integrity of steel casing of wellbore by preventing corrosion of pipes and casings N,N-dimethylformamide Crosslinker Thickens fluid to hold proppant Borate salts Defoamer Lowers surface tension and allows gas escape Polyglycol Foamer Reduces fluid volume and improves proppant carrying capacity Acetic acid (with NH4 and NaNO2) Friction Reducer Improves fluid flow efficiency through wellbore by reducing friction between fluid and pipe; alleviates friction caused by high pressure conditions Polyacrylamide Gel/Gellant Thickens fluid (water) to suspend proppant Guar gum Iron Control Prevents materials from hardening and clogging wellbore; prevents metal oxide precipitation Citric acid Oxygen Scavenger Maintains integrity of steel casing of wellbore; protects pipes from corrosion by removing oxygen from fluid Ammonium bisulfate pH Adjusting Agent/ Buffers Controls pH of solution; protects pH-dependent effectiveness of other chemicals (e.g., crosslinkers) Sodium carbonate, potassium carbonate Proppant Holds open (props) fractures to allow gas to escape from shale Silica, sometimes glass beads Scale Control Prevents mineral scale formation which can clog wellbore, block fluid or gas flow Ethylene glycol Solvents Improve fluid wettability or ability to maintain contact between the fluid and the pipes Stoddard solvent Surfactant Improves fluid flow through wellbore by reducing surface tension Isopropanol
    • 11 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 greater use of water and chemical components. Of the 13 report- ing sites in Michigan, 11 wells reported to have a single hydraulic fracturing event, 1 reported 2 events, and 1 reported 3 events. The average number of chemicals disclosed for these 13 Michigan wells was 25 chemicals per well (range: 13-55). In addition to the data provided in Tables 3 and 4, below we provide further information on five intentional-use chemicals (methanol, isopropanol, ethylene glycol, 2-butoxy ethanol, hydrotreated light petroleum distillates) that are prevalent in a number of hydraulic fracturing products. For each chemical, we briefly describe their role in hydraulic fracturing, relevant exposure routes, toxicoki- netics, target organs, and potential health effects. A summary of potential health effects is also offered (Table 5) for these five inten- tional-use chemicals and a few other selected hazards. In terms of susceptible populations, workers in close proximity are at highest risk of possible exposures. Exposures to these identified chemicals would likely be in the form of inhalation and/or dermal contact. 2.7.1 Methanol (CAS 67-56-1) Methanol is a colorless, volatile, and flammable liquid. It is used in hydraulic fracturing fluid broadly as a stabilizing and/or winteriz- ing agent, corrosion inhibitor, crosslinker, friction reducer, gelling agent, and surfactant35 . Individuals may be exposed to metha- nol through inhalation, ingestion, or dermal routes. Methanol is water-soluble and can distribute rapidly to a variety of organs, with about 60% of an inhaled dose absorbed52 . Methanol toxicity is generally attributed to the metabolism of methanol to formic acid, which can reach levels high enough to cause acidosis in methanol poisoning; targets of toxicity include the optic nerve, kidneys, and other organs with high oxygen demand such as the brain and the heart53 . In occupational settings methanol exposure can be moni- tored by measuring formic acid in the urine of workers35 ; however, since formic acid is also an ingredient in hydraulic fracturing fluid, and since other compounds used in hydraulic fracturing fluid (methyl ethers, esters, and amides) can also be metabolized into formic acid53 , this is likely not an adequate measure of methanol exposure in hydraulic fracturing workers. Other exposure monitor- ing methods should be sought. 2.7.2 Ethylene Glycol (CAS 107-21-1) Ethylene glycol is an organic chemical that is odorless, color- less, and sweet-tasting. It is used in hydraulic fracturing fluid as a product stabilizer and/or winterizing agent, crosslinker, friction reducer, gelling agent, and non-emulsifier35 . Ethylene glycol can be ingested, inhaled, or dermally absorbed, and workers are likely to be exposed by dermal or ocular routes. It is an ocular, but not a dermal, irritant54 . Within the body, the chemical can be rapidly dis- tributed54,55 . Ethylene glycol may have cardiovascular, neurological, and renal effects56 . Ethylene glycol exerts health effects primarily by two metabolites, glycolate and oxalate54 ; both metabolites con- tribute to renal effects57 . TABLE 4: Notable chemicals intentionally used in hydraulic fracturing fluids. Information is based on reports by Waxman et al.51 and Colborn et al.45 The most commonly used chemicals in Michigan are also reported following a search of FracFocus.org35 . Chemicals found in highest number of products (Chemical Name; CAS #; # products); Ref: Waxman et al.51 Chemicals found in highest number of products (Chemical Name; CAS #; # products); Ref: Colborn et al.45 Chemicals associated with the greatest number of possible health effects (Chemical Name; CAS #; # possible health effects); Ref: Colborn et al.45 Number of times chemical ingredi- ent is reported to have been used in a Michigan well sites (n=12 sites reporting); Ref: fracfocus.org35 Methanol; 67-56-1; 342 products; Isopropanol; 67-63-0; 274 prod- ucts; Crystalline silica, quartz; 14808-60- 7; 207 products; 2-Butoxyethanol; 111-76-2; 126 products; Ethylene glycol; 107-21-1; 119 products Crystalline silica, quartz; 14808-60- 7; 125 products; Methanol; 67-56-1; 74 products; Isopropanol; 67-63-0; 47 products; Petroleum distillate hydrotreated light; 64742-47-8; 26 products; 2-Butoxyethanol; 111-76-2; 22 products 12 possible health effects each: Naphthalene; 91-20-3 Ethanol; 64-17-5 Petroleum distillate naphtha; 8002- 05-9; 12; 11 possible health effects: methanol; 2-Butoxyethanol; glutar- aldehyde; fuel oil #2; formic acid; 2-ethylhexanol; ethylbenzene; Ammonium persulfate; 7727-54-0; 14 times; Crystalline silica, quartz; 14808-60-7; 11 times; 2,2-dibromo-3-nitrilopropionamide; 10222-01-2; 10 times; Methanol; 67-56-1; 10 times; Ethylene glycol; 107-21-1; 9 times; Acetic acid; 64-19-7; 9 times; Hydrogen chloride; 7641-01-1; 9 times Potassium chloride; 7447-40-7; 8 times
    • 12 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 TABLE 5: Summary of potential health effects for a number of physiological systems by notable chemicals. No information was found concerning endocrine, immunologic, or musculoskeletal effects. Data was summarized by reviewing ATSDR Toxicological Profiles and, when Toxicological Profiles were absent, MSDS sheets. Based on ATSDR Toxicological Profiles and The Globally Harmonized System of Classification and Labeling of Chemicals (GHS; a system for harmonizing various rating systems for chemical hazards) provided in MSDS for each chemical, the amount and quality of evidence for effects of notable chemicals on various physiological systems was assessed, with X = less evidence and lower quality, XX = intermediate amount and quality of evidence, and XXX = more evidence and better quality. These classifications were assigned by the authors based on their judgments. TABLE 2: Setback Requirements for Well Location Chemical Name CAS Carcinogen Cardiovascular Dermal Developmental Ecotoxicity Gastro- intestinal Hemato- logical Methanol 67-56-1 X XXX XXX XX Isopropanol 67-63-0 XX XXX XXX XX XXX Ethylene Glycol 107-21-1 XXX XXX 2-Butoxyethanol 111-76-2 XX XX XXX XX XX XXX Hydrotreated Light Petroleum DIstillate 64742-47-8 XXX XXX Hydrogen Sulfide 7783-06-4 XXX XXX Silica 112945-52-5 XX XXX Diesel *** XXX XXX XXX Chemical Name CAS Hepatic Neurological Ocular Renal Reproductive Respiratory Irritant/ Corrosive Methanol 67-56-1 XXX XXX XX XX XXX Isopropanol 67-63-0 XXX XXX XXX XXX XXX Ethylene Glycol 107-21-1 XXX XXX XXX XXX XX XXX XXX 2-Butoxyethanol 111-76-2 XXX XXX XXX XXX XX XX XXX Hydrotreated Light Petroleum DIstillate 64742-47-8 XXX Hydrogen Sulfide 7783-06-4 XXX XXX XXX XXX XXX Silica 112945-52-5 XXX XXX Diesel *** XXX XXX XXX XXX XXX XXX
    • 13 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 2.7.3 2-Butoxyethanol (CAS 000111-76-2) 2-Butoxyethanol (also called ethylene glycol monobutyl ether) is an organic solvent that is colorless with a sweet odor. It is used in hydraulic fracturing fluid as a product stabilizer, given its sur- factant properties35 . 2-Butoxyethanol can be ingested, inhaled, or dermally absorbed, and within the body it is rapidly distributed58 . Dermal absorption is highest when the chemical is present in an aqueous solutions (as would be encountered in fracturing fluids), and absorption can be enhanced with greater temperature and humidity58 . Within the body, 2-butoxyethanol is metabolized to butoxyacetic acid (and other alkoxyacetic acids) by alcohol dehy- drogenase58 . Ethylene glycol may also result from the metabolism of 2-butoxyethanol. Case reports of accidental or intentional ingestion have reported hematuria, metabolic acidosis, pulmonary symptoms, and/or anemia, with patient recovery in each case58 . The central nervous system is a target for 2-butoxyethanol, though this appears to manifest primarily as respiratory depression only at particularly high doses59 . Exposures in humans have been associ- ated with nose, throat, and eye irritation, headaches, and reports of a metallic taste in the mouth58 . 2.7.4 Isopropanol (CAS 67-63-0) Isopropanol or isopropyl alcohol is a colorless and flammable liq- uid that has a strong odor. It is used in hydraulic fracturing fluid as a product stabilizer and/or winterizer, corrosion inhibitor, non-emul- sifier, and surfactant35 . Isopropanol can be ingested, inhaled, or dermally absorbed. Isopropanol may vaporize from solution thus indicating a possible inhalation source directly from hydraulic frac- turing fluids or produced water. The main metabolite of isopropa- nol is acetone, which can be excreted rapidly in urine and expired air60 . Based on its MSDS, Isopropanol can be a skin or eye irritant, and the target organs of toxicity include nerves, kidney, cardiovas- cular system, gastrointestinal tract, and liver61 . 2.7.5 Hydrotreated Light Petroleum Distillates (CAS 64742-47-8) Hydrotreated light petroleum distillates (HLPD) are used in hydrau- lic fracturing fluid as a carrier fluid for borate or zirconate cross- linkers, polyacrylamide friction reducers, and guar gum in liquid gels35 . HLPD are defined as a mixture of hydrocarbons, primarily C9-C15, and their composition may vary greatly depending on the exact source and production method62 , thus posing difficulties for assessment of health. HLPD can be ingested, inhaled, or dermally absorbed. Inhalation absorption is greater during exercise, indi- cating possible higher risk for workers, depending on their specific duties. It is recommended that any HLPD gas, fume, vapor, and/ or spray not be inhaled. HLPD are an eye and skin irritant, and are potentially fatal if swallowed or enters airways. HLPD can also have a narcotic effect following a single exposure, with symptoms including drowsiness and dizziness; unconsciousness and death are possible with continued exposure. Some components of HLPD have also been implicated in neurotoxic effects such as balance, memory, coordination, fatigue, and reduced motivation at low concentrations for short times, which generally did not persist long after exposure. Such neurological effects may impact worker performance and contribute to injuries, should an exposure occur. Hematologic and kidney effects may also result from exposure. Some studies indicate a higher risk of certain cancers (prostate, central nervous system and lymphatic) among workers exposed to HLPD, but this higher risk has often been attributed to other components of complex mixtures to which workers are exposed. Rheumatoid arthritis has also been connected to HLPD by two studies, including one of women in Michigan and Ohio; however, exposures to other stressors existed and causal linkages were dif- ficult to make62 . Cable plant workers exposed to mist and vapor of petroleum distillates have developed pulmonary fibrosis, though again disentangling the effects of HLPD is difficult63 . Long-term handling of HLPD may lead to dry and cracked skin. Susceptible populations may include those with skin disease, chronic respi- ratory disease, liver disease, or kidney disease. The presence of benzene in distillates may increase the magnitude of the hazard64 . 2.8 By-product Chemicals In addition to chemicals used intentionally to facilitate drilling and well stimulation, there are a number of chemicals that may be released ‘unintentionally’. Machinery at the worksite may release chemicals (e.g., diesel exhaust and particulate matter) as part of normal functioning or from malfunctions. The extracted natural gas may contain chemicals such as hydrocarbons (e.g., methanol, benzene) and hydrogen sulfide. These chemicals in the natural gas can be removed following treatment with ethylene (or triethylene) glycol and heat, though the resulting “produced water” (i.e., water boiled off) and “condensate water” (i.e., volatile byproducts) would need to be managed accordingly. In addition, the water that returns to the well after fracturing has been com- pleted (“flowback” water) is likely to be highly saline owing to the interaction of hydraulic fracturing fluids with naturally occurring minerals. This saline flowback water may also contain a number of toxic metal(loid)s, hydrocarbons, and naturally occurring radio- active substances. Another potential by-product are solid wastes that may be generated from drill cuttings, pit liners, sludge from recovered water, and other processes. Below we further describe some of the aforementioned chemicals. We found one disputed report describing exposure of people to a number of chemicals in Michigan34 . The 2003 survey from Manistee and Mason Counties alleges that a well leak of gas and fluids on October 28, 1998 from a natural gas operation not employing hydraulic fracturing resulted in a number of hospitalizations from exposure to methane, hydrogen sulfide, and other chemicals.
    • 14 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 This report was not published in the scientific peer-reviewed lit- erature, and the Michigan Department of Environmental Quality (Michigan DEQ) and Department of Community Health concluded that petroleum condensate vapors, not the gas leak, were likely the cause of the complaints and health effects65 . Nevertheless, we provide this information here, as aspects may be relevant in future assessments and thus help decision makers and first responders anticipate and recognize potential issues. 2.8.1 Diesel A variety of machinery at a hydraulic fracturing site may be die- sel powered, including trucks, drilling rigs, and pumps. Given this prevalence, workers may be exposed to large quantities of diesel emissions on a regular basis. In general, there is a lack of informa- tion concerning the health effects of exposure to fuel oils, such as diesel, though inhalation of some fuel oils can lead to nausea, eye irritation, increased blood pressure, headache, light-headedness, loss of appetite, poor coordination, and difficulty concentrating66 . In particular, long-term inhalation of diesel fuel vapors can damage kidneys and reduce blood clotting66 . While there has been a great deal of research investigating the link between diesel exhaust exposure and cancer risk, it is inconclusive whether exposure leads to increased lung cancer in humans67 . In addition to releasing die- sel, this machinery may also emit other air pollutants (e.g., particu- late matter) and contribute to noise pollution. 2.8.2 Methane Exposure to methane, as a gas, usually occurs via inhalation. According to the U.S. Environmental Protection Agency’s (EPA) Hydraulic Fracturing Study Plan49 , Encana Oil and Gas USA has released information that indicates that methane is either used in hydraulic fracturing fluid or is present in flowback/produced water. Aside from being a risk for fire and explosions, methane is an asphyxiant, particularly in confined spaces68 . Methane may also act as an irritant of the respiratory tract, skin, and eyes69 . Data on health effects from ingestion of methane are scarce, as this is not a common route of exposure, though it could occur since methane is a component in hydraulic fracturing fluids (likely flowback/pro- duced water). 2.8.3 Hydrogen Sulfide Hydrogen sulfide is a particularly dangerous chemical that can naturally occur in natural gas and thus it may be released during a hydraulic fracturing operation70 . Hydrogen sulfide is a colorless gas that has the odor of rotten eggs; however smell cannot be relied upon to detect H2S as the gas can quickly impair an individual’s sense of smell. It is denser than air, so can accumulate in low laying areas20 . Hydrogen sulfide acts as an irritant and a chemical asphyx- iant and is deadly at concentrations of 100 ppm. Exposure at lower concentrations can cause difficulty breathing and irritation of eyes, nose, and throat and can also result in long-term neurological effects including reduced motor function, poor memory and attention span, and headaches71 . Hydrogen sulfide is also highly flammable and produces sulfur dioxide, another toxic gas, when it burns70 . The OSHA Permissible Exposure Limit (PEL) for workers is 20 ppm and NIOSH recommends limiting hydrogen sulfide exposure to 10 ppm for a maximum of 10 minutes71 . Monitoring of H2S is commonplace in hydraulic fracturing operations. In Michigan, there have been reports of H2S releases from oil and gas industry operations in Manistee and Mason Counties occurring from 1980–200234 . Members of the surrounding com- munities reported that self-evacuations, medical treatment, and hospitalizations resulted from these H2S releases, but as earlier, we acknowledge that this report was not published in the scientific peer-reviewed literature. 3.0 THE SURROUNDING ENVIRONMENT T he environment that surrounds a hydraulic fracturing worksite may be the recipient of a number of hazards that originate from, or are caused by, the industry. In this section we discuss how hydraulic fracturing sites may impact ecosystem quality at the local and regional scales. Clean air and drinking water, as well as ecosystem services such as hunting and fishing, have direct linkages to human health and well-being72 . 3.1 Transportation Many aspects of hydraulic fracturing operations lead to an increase in truck traffic in and around the site vicinity, including the use of trucks to transport drilling equipment, workers, and water to and from the site73 . Road congestion was identified by shale gas indus- try experts as a high priority risk to the public and environment in a recent survey in which a variety of experts were asked to identify the most important risks related to hydraulic fracturing opera- tions74 . While truck traffic is likely to be elevated in and around a hydraulic fracturing site for the lifetime of the well, the majority of truck traffic occurs during the construction of the well pad, the hydraulic fracturing process, and the waste fluid and equipment removal process73 . The number of heavy trucks needed during a hydraulic fracturing operation depends on the number of wells and well pads established at a site. For example, New York State Department of Environmental Conservation (NYSDEC)73 estimates that 3950 one-way truck trips (with 1148 of those being heavy, fully loaded trucks) would be required for a newly created well and well pad (Table 6). This estimate is for a horizontal drilling hydraulic frac- turing operation requiring ~5 million gallons of water, all of which would be transported to the site by truck.
    • 15 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 The availability and location of water can affect amount of truck traffic. Sites that are able to pipe water to the site can reduce the number of heavy trucks needed from an estimated 500 trucks to 6073 . This estimate may be more applicable to Michigan, where a number of facilities have reported obtaining water from onsite groundwater wells (e.g., State Excelsior 1-13HD1 from fracfocus. org35 ). TABLE 6. Estimated number of one-way loaded trips for a single, horizontal well requiring a vertical and horizontal rig. Table modified from NYSDEC 201173 , source: All Consulting 201075 . This information is specific for New York, but elements could be used to inform Michigan. Well Pad Activity Early Well Pad Development (all water transported by truck) Heavy Truck Light Truck Drill pad construction 45 90 Rig mobilization 95 140Drilling fluids 45 Non-rig drilling equipment 45 Drilling (rig crew, etc.) 50 140 Completion chemicals 20 326 Completion equipment 5 Hydraulic fracturing equip- ment (trucks and tanks) 175 Hydraulic fracturing water hauling 500 Hydraulic fracturing sand 23 Produced water disposal 100 Final pad prep 45 50 Miscellaneous - 85 Total one-way loaded trips per well 1,148 831 Truck traffic can have multiple impacts on the worksite and sur- rounding community, with noise, air pollution, and accidents of greatest concern73,76 . Many roads, particularly those in rural areas, are not designed for the frequency and weight of heavy trucks, and thus road wear and erosion are likely to result73 . Damage to roads in rural areas may put a strain on local and county finances, which may have fewer resources than more populated regions. For instance, the 2010 road budgets of 8 northern Michigan counties, where hydraulic fracturing is common, range from $3,099,902 (Montmorency) to $7,207,539 (Otsego)77 . These were much lower than other urban and suburban Michigan counties such as Oakland ($93,817,982) and Washtenaw ($31,079,725). Increased traffic can also lead to an increase in the risk of traffic accidents. In 2011, there were 284,049 traffic crashes in the state of Michigan, 834 of which were fatal crashes78 . Only 17 of these fatal crashes occurred in the 8 northern Michigan counties mentioned above. A large increase in heavy truck traffic may thus greatly impact these rural areas. Traffic accidents involving vehicles trans- porting hydraulic fracturing fluid is a particular concern, because accidents involving these trucks may also lead to spills that could potentially contaminate surface water and soil in the area49,73,76, though in our search we did not find any records of accidents in Michigan. 3.2 Air Pollution Many aspects of the hydraulic fracturing process can result in air pollution emissions. While some air pollutants, such as diesel, methane, and hydrogen sulfide mentioned in earlier sections, may disproportionately affect hydraulic fracturing workers, these pol- lutants, as well as a wider variety of volatile organic compounds, criteria pollutants, and metals can also reduce air quality in the sur- rounding communities and ecosystems. The cumulative impacts of these are not clear. Hydraulic fracturing operations are known to release many of the most common air pollutants, known as criteria pollutants, includ- ing nitrogen oxides (NOx), particulate matter (PM), carbon mon- oxide (CO), and sulfur dioxide (SO2), as well as volatile organic compounds (VOC) and other hydrocarbons46,79 . Other contami- nants can result from the release of certain pollutants, for instance VOC’s and NOx emitted during hydraulic fracturing can combine with sunlight to create ozone (O3), and this has been estimated to occur in Wyoming. The cumulative impacts of multiple air pollut- ants being released from several facilities warrant attention. The amount and types of emissions released is dependent on the stage of the hydraulic fracturing operation46 . For each major stage in hydraulic fracturing, an estimated release of these pollutants is provided (Table 7). In addition, a table outlining key machinery and the types of air pollutants they may release is also provided (Table 8). Combustion engines emit major air pollutants, and are used during all stages of a hydraulic fracturing operation to run key machinery, such as compressors, pumps, trucks, and drilling rigs46 . Air pollutants are also released during the well completion process, which vents or flares natural gas from the well, releasing methane and a variety of other air contaminants80 . The EPA estimates that uncontrolled well completion processes for hydraulically fractured wells can vent approximately 23 tons of VOCs, which is 230 times more than conventional natural gas well completion81 . A number of hydrocarbons may also be released into the air during a hydraulic fracturing operation. While hydrocarbons are often released during well completion, the exact source of measured hydrocarbons is often unknown, and they likely originate from multiple sources79 . For methane, in Michigan this gas is to be captured in the tanks
    • 16 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 and flared. A recent survey asked a variety of experts to identify the most important environmental risks related to hydraulic fracturing operations, experts agreed that well venting during the well com- pletion process was a high priority risk to the public and environ- ment74 . These experts also agreed that well flaring during the well completion process was not a priority risk. Flowback operations are also likely to be a key source of hydrocarbon emissions during the hydraulic fracturing process79 , though Michigan requires meth- ane to be captured and flared at this stage. The air pollutants released during hydraulic fracturing operations are known to have a range of adverse effects on human health. Inhalation is the dominant exposure route, though deposition into water and food may also occur. Respiratory and cardiovascular effects are best studied, though negative effects of air pollution on neurological, immune, reproductive, and developmental systems have also been documented. Cumulative effects of releases from multiple wells may affect regional air quality. Below we provide some key details for the relevant air pollutants. 3.2.1 Nitrogen Oxide (NOx) Nitrogen oxide emissions come from motor vehicle exhaust and burning of natural gas and other fuel sources82 . Low level NOx exposure can irritate the eyes, nose, and throat, as well as the lungs, which may result in coughing, shortness of breath, and fluid in the lungs. Inhalation of NOx at high levels can cause severe respiratory damage that can lead to reduced oxygenation and death. OSHA limits exposure of nitrogen oxide for workers to 25 ppm for an 8-hour day, for 40 hours per week82 . The EPA standards for annual ambient nitrogen dioxide concentrations are set at 53 parts per billion (ppb) and 100 ppb averaged over one hour83 . TABLE 7. Estimated annual emissions from various stages of hydraulic fracturing. Table assumes dry gas is present and the flowback gas is flared (units in tons per year). Table modified from NYSDEC, 201173 . Drilling Completion Production Flowback Gas Total PM 0.5 0.2 0.2 1.4 0.9 NOx 15.1 5.8 3.8 4.9 29.6 CO 8.3 3.2 9.2 24.5 45.2 VOC 0.8 0.2 2.4 0.7 4.1 SO2 0.02 0.01 0.07 0.0 0.1 TABLE 8. Summary of air pollutants and sources, which have been modeled for a hydraulic fracturing operation. Table modified from NYSDEC, 201173 . Health Effect SO2 NO2 PM10 & PM2.5 CO Non-criteria combustion emissions H2S and other gas constituents Reduces lung function Lung irritation and damage Reduces lung function Reduces blood oxygen Source Engines for drilling X X X X Compressors for drilling X X X X X Engines for Hydraulic fracturing X X X X X Line heaters X X X X X Off-site compressors X X X X X Flowback gas flaring X X X X X Gas venting X Mud-gas separator X Glycol dehydrator X X
    • 17 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 3.2.2 Particulate Matter (PM) Particulate matter consists of very fine grain particles or liquid mixed within air. It is often made up of a variety of different com- ponents such as dust, smoke, and organic chemicals84 . Inhalation of these particles, particularly very small particles, can cause them to become imbedded in lung tissues and negatively affect the lungs and heart. Studies have shown that inhalation of PM can cause heart attacks, reduce lung function, and exacerbate asthma. Children, the elderly, and people with heart or lung diseases are particularly susceptible to the negative health effects of PM84 . The EPA standard for fine PM (PM2.5) is an annual average concentra- tion of 12 micrograms per cubic meter (µg/m3). 3.2.3 Carbon Monoxide (CO) Carbon monoxide can be produced when a natural gas well is flared during the completion process, as well as from combustion engines46 . Inhalation of CO reduces oxygen delivery within the body, making people with low blood oxygen, caused by certain medical conditions, particularly vulnerable to CO effects85 . Effects of CO inhalation include headache, nausea, vomiting, dizziness, blurred vision, confusion, chest pain, weakness, heart failure, dif- ficulty breathing, seizures, and coma; exposure at high levels can be fatal86 . 3.2.4 Sulfur Dioxide (SO2) Sulfur dioxide emissions are produced from the combustion of fos- sil fuels containing sulfur, such as gasoline and diesel46 . Exposure to high levels of SO2 can be deadly, causing burns to the nose and throat and obstructing breathing87 . Long-term exposure of low SO2 concentrations can cause changes in lung function and short- term exposure has been known to cause bronchoconstriction and increase symptoms of asthma88 . The EPA standard for SO2 expo- sure is 1 hour at 75 ppb88 . 3.2.5 Volatile Organic Compounds (VOC) and Other Hydrocarbons Volatile organic compounds are a group of organic compounds that are easily converted into gas form; burning gasoline and nat- ural gas, and the dehydration of natural gas, can emit VOCs46,89 . Little is known about exposure to low levels of VOCs, but expo- sure to high levels can cause long-term health problems related to brain, blood, and liver. Exposure to VOCs may increase cancer risks89 . McKenzie et al.79 sampled a variety of VOCs and other hydrocar- bons including trimethylbenzenes, xylenes, aliphatic hydrocarbons, benzene, ethylbenzene, and toluene, during well development, production, and completion at a hydraulic fracturing operation in Garfield County, CO. Hydrocarbon emissions were detected with greater frequency and at higher concentrations in samples collected specifically during well completion activities than in samples collected in the general area of the well. McKenzie et al.79 used the hydrocarbon levels found in the collected samples to determine chronic and subchronic non-cancer risk and cancer risk for residents living > ½ mile and < ½ mile from the well site. Residents living < ½ mile from the well had greater risk for chronic and subchronic non-cancer risk and cancer risk than residents living > ½ mile from the well. Chronic non-cancer risks were attributed to neurological effects, while subchronic non-cancer risks included neurological, respiratory, hematologic, and developmental effects. Benzene and ethylbenzene emissions were determined to be the greatest contributors for cancer risk. Colborn et al.90 also collected air samples around a hydraulic frac- turing operation in Garfield County, CO. Samples were analyzed for VOCs, as well as polycyclic aromatic hydrocarbons (PAHs) and carbonyls. VOCs detected in 100% of collected samples included methane, ethane, propane, and toluene. Those detected in the greatest mean concentrations included methane, methylene chloride, ethane, methanol, ethanol, acetone, and propane. The carbonyls formaldehyde and acetaldehyde were also found in all collected samples, with the greatest concentrations detected for crotonaldehyde and formaldehyde. The PAH found at the highest concentration was naphthalene, and it was also detected in every sample. While concentrations of these air pollutants were detected at levels lower than government limits such as the REL and PEL, the effects associated with chronic, low-level exposure to air pollutant mixtures is not well known. To our knowledge, air samples have never been collected in the vicinity of hydraulic fracturing operations in Michigan for the pur- poses of determining air pollutant concentrations. 3.2.6 Greenhouse Gases (GHG) Many of the air pollutants released during a hydraulic fracturing operation are greenhouses gases (GHG). Notable ones include methane, carbon dioxide, and nitrous oxide. Jiang et al.9 used emissions of these 3 gases to estimate the lifetime GHG produc- tion of a typical Marcellus Shale well. They included emissions from numerous GHG producing sources including trucking water and equipment, running drilling and fracturing equipment, methane leaks, emissions associated with gas production, processing, trans- mission, distribution, as well as natural gas combustion. In doing so, they estimated that a well produces approximately 5500 tons of carbon dioxide equivalent emissions. This estimate represents a 3% increase in lifetime GHG emissions for a hydraulically frac- tured natural gas well compared to a conventional natural gas well. Stephenson et al.91 calculated a similar number, and estimated that hydraulic fracturing operations produce approximately 1.8 – 2.4% more GHG emissions than conventional natural gas wells. However,
    • 18 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 Howarth et al.92 determined methane emissions from hydraulically fractured wells to be 30% greater than those from conventional natural gas wells, largely from methane emissions during the flowback and drill out processes. While there is uncertainty in the amount of GHG produced by hydraulically fractured natural gas wells, Weber and Clavin93 examined 6 studies, including Jiang et al.9 , Stephenson et al.91 , and Howarth et al.92 , and could not deter- mine whether hydraulically fractured or conventional wells pro- duced a larger carbon footprint. Hydraulic fracturing operations could significantly reduce their GHG emissions by capturing gases that would normally be flared or vented during the well completion process9,91 . It is also important to consider the amount of GHG emitted during both the hydraulic fracturing process and during combustion. While GHG are produced during the hydraulic fracturing process, combustion of natural gas during energy production has been esti- mated to be more efficient and produce less greenhouse gas emis- sions than other energy sources such as coal9,91 . For the purpose of electricity production, Jiang et al.9 estimated that hydraulically fractured natural gas produces 20-50% less GHG emissions during the lifetime of the well than coal. However, additional methane leaks during the natural gas production, transportation, and use may dramatically increase GHG emissions, particularly since meth- ane is a more potent GHG than carbon dioxide produced from burning coal94 . 3.3 Water Quality Of all public health issues, matters related to water quality and quantity appears to be of paramount concern. For example, a sur- vey of 215 experts from NGO, government, and academia revealed that effects on surface water quality were a dominant concern, which is an interesting finding given that most in the public and regulatory world have expressed concern about ground water95 . The potential impact on water quality and quantity from hydraulic fracturing in the State of Michigan is a focus of the paper by Burton et al. (2013) and Ellis et al. (2013) in this series. Water is integral to hydraulic fracturing in a number of ways. For example, water is needed for key activities such as drilling, cooling and lubricating drill bits, and producing the hydraulic fracturing fluid. Wells situated in major shale plays, such as Barnett and Marcellus, may use an average ~5 million gallons of freshwater per well96 . To put this into perspective, 5 million gallons of water is approximately the amount used by 50,000 people in one day. In Michigan, the amount of water needed to fracture a horizontal well may also be around ~5 million gallons97 . For example, the State Pioneer 1-3HD1 well (Utica-Collingwood discovery well, Missaukee County) has used 6,720,000 gallons of water98 . It should also be noted, that a typical shallow, vertical well in the Antrim Shale requires ~40,000-100,000 gallons of water97,98 . When compared to other users of water, such as agriculture, the amount used in hydraulic fracturing is much less99 (see Ellis, 2013 in this series). Nonetheless, concern may exist when activities are conducted in drought-prone regions or when water tables and river flows sub- side. Removal of water for hydraulic fracturing is viewed from a region’s hydrologic cycle. Given the current reported low water levels of the Great Lakes (record low levels for Lakes Michigan and Huron since record keeping began in 1918)100 the extra water use from expanded hydraulic fracturing operations in Michigan may further alter hydrologic conditions and may increase resident con- cern about water issues. Below Table 9 shows the wide range of reported water use for wells with disclosure data in Michigan (as of March 10, 2013). Notably, the 5 wells with the greatest disclosed water use (State Excelsior 1-25, 2-25, and 3-25; State Excelsior 1-13; State Garfield 1-25) all reside in Kalkaska County; the State Excelsior wells 1-25, 2-25, and 3-25 are particularly closely clustered. Kalkaska County thus may be a priority for examining water resource impacts from high-vol- ume hydraulic fracturing. The potential impacts of hydraulic fracturing on drinking water resources are being studied by the EPA under its “Plan to Study the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources”. This plan is focused on addressing possible impacts on drinking water sources related to water acquisition, chemical mixing, well injection, flowback and produced water, and waste- water treatment and disposal. As outlined in their December 2012 progress report96 , a number of reviews of existing data, computer simulations, laboratory and field studies, and toxicity assessments are currently underway. The outcome of this overall study is expected to significantly improve our understanding of the potential risks of hydraulic fracturing to drinking water resources. However, it appears that Michigan is not included in the EPA’s plan; thus, even after the EPA completes their study, there may be substantial gaps in our understanding of the potential impacts of hydraulic fracturing on water resources in Michigan. 3.3.1 Source of Chemicals into Water Chemicals may be released from various stages of a hydraulic fracturing operation into the surrounding ecosystem. Chemicals may enter the environment following ruptures or mishandling of impoundment tanks or equipment, operator errors, and accidents. The drill cuttings may contain naturally occurring radioactive mate- rials (NORM), and if not handled properly these may be a source of contamination. About 90% of the hydraulic fracturing fluid is water96 , though a number of chemicals are added to hydraulic frac- turing fluids (Table 3) to optimize its properties as discussed above.
    • 19 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 About 1% of the hydraulic fracturing fluid is composed of various chemicals (~50,000 gallons for a typical well that uses 5 million gal- lons of hydraulic fracturing fluid). These intentional-use chemicals may represent a contaminant source during use or flowback. There are reports documenting underground migration of contaminants, particularly in cases of inadequate construction of well casings and cementing, and plugging of abandoned wells101 . It is important to realize that not all the injected water is recovered, 30 to 90% (1.5 to 4.5 million gallons of the 5 million estimated) may remain under- ground96 . Wilson and Schwank (2013) in this series report some- what better recovery, with only 25%-75% remaining underground. In addition to intentional-use chemicals, the produced water may pick up, depending on local geochemistry, a variety of salts, toxic metals, hydrocarbons, and NORM. In Michigan flowback water cannot be sent to wastewater treatment plants or be stored in open impoundments, rather they must be stored in large steel tanks or injected deep into the ground licensed deep disposal wells3 . Further, as discussed by Ellis (2013) in this series, production of natural gas and underground injection of CO2 or flowback/pro- duced water can alter the pressure gradients that drive groundwa- ter flow. Though the vertical distances involved are large, this could potentially lead to altered water supplies for nearby communities particularly in the case of improperly capped wells. 3.3.2 Fate of Water-borne Chemicals Hydraulic fracturing activities may contaminate water resources, though it is not clear whether people could be exposed to contaminants since exposure assessments and biomarker stud- ies are lacking in Michigan and elsewhere. A 2011 study from the Marcellus Shale (Pennsylvania) analyzed 48 water samples from private wells before and after hydraulic fracturing to determine if water quality was affected102 . The analyses revealed no signifi- cant changes in a number of water contaminants, such as chlo- ride, barium, strontium, and methane. Another 2011 study from the Marcellus Shale (Pennsylvania) and Utica Shale (New York) analyzed 60 drinking water samples from aquifers that overlie the formations affected and not affected by hydraulic fracturing103 . The analyses revealed that methane could be detected in 51 of 60 (85% of all) drinking water wells in the study region but that concentrations tended to be higher in samples taken closer to the natural gas wells. The research was not able to pinpoint the source of methane. Though this methane may be of thermogenic or biogenic origin, its origin may not matter greatly if its presence in drinking water is indeed a result of processes associated with gas extraction. Gathering baseline data prior to establishment and fracturing of future gas wells in Michigan could help to clarify whether or not such a relationship exists. This study did not find any evidence of chemicals used in hydraulic fracturing fluid in the samples tested, based on geochemical and isotopic features of tested water. Another study analyzed flowback water from 24 locations in the Marcellus and Barnett Shales; from each location influent (day 0) water was collected along with flowback water at 1, 5, 14, and 90 days following the fracturing event104 . The analysis revealed that a number of general water quality parameters such as alkalinity, total dissolved solids, total organic carbon, sodium, calcium, barium, and iron differed between the flowback water and the influent water. In general, the author indicates that the water quality parameter values measured are normally seen in waters from conventional oil and gas operations. In addition to studies that have directly analyzed water samples for potential contaminants, other studies have queried State ground- water investigation reports and regulatory officials. In a study per- formed by the Ground Water Protection Council concerning Ohio and Texas, there was only one horizontal hydraulic fracturing well in Ohio (prior to 2008) and more than 16,000 wells were completed in Texas over a 16 year period (1993-2008)105 . The review of investiga- tive reports from both states did not find any cases of groundwater contamination from hydraulic fracturing. The authors of this report also met with officials from eight States (Arkansas, Colorado, Louisiana, North Dakota, Ohio, Oklahoma, Pennsylvania, and Texas) and determined that, based on state investigations, no report of groundwater contamination in these states was associ- ated with hydraulic fracturing. Given the widespread lack of base- line data on groundwater contamination near hydraulic fracturing sites, there is still substantial uncertainty regarding contamination because even elevated levels of contaminants near hydraulically TABLE 9. Total Water Volume (gallons) for wells with disclosure information in Michigan35 . Well Total Water Volume (gallons) State Excelsior 3-25 HD1 21,112,194 State Excelsior 2-25 HD1 12,562,096 State Garfield 1-25 HD-1 12,539,639 State Excelsior 1-25 HD1 8,461,635 State Excelsior 1-13 HD1 5,860,777 State Richfield 1-34 HD 4,804,620 Wiley 1-18 HD 1,420,939 Cronk 1-24 HD1 705,390 State Wilmot 1-21 33,306 State Marion & Riverside 14-36 25,914 State Marion & Riverside 14-36 23,023 State Marion & Riverside 14-36 20,918 Clayton Unit 44-31 20,701 Riverside 32-24 17,805 Wineman Unit B2-9 16,263 Wineman Unit B2-9 14,460
    • 20 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 fractured wells cannot be attributed to hydraulic fracturing without baseline data for comparison. 3.4 Habitat and Wildlife Impacts The construction, operation, and maintenance of a hydraulic frac- turing site may affect the landscape in a number of ways. Such land- scape effects can have consequences for wildlife and ecosystems, as well as public health when opportunities for recreation, natural areas, and cultural/spiritual practices are affected. According to the Michigan DEQ, hydraulic fracturing has not been responsible for environmental damage in the State of Michigan3 . The impact of hydraulic fracturing on ecological health is covered in this series by Burton et al. (2013). The development of a hydraulic fracturing site requires the manip- ulation of land (e.g., via clearing and excavating), development and maintenance of primary (e.g., fracking pad) and associated infrastructure (e.g., pipelines, storage tanks), and construction and maintenance of roads to access the site. These types of activities are known to cause habitat loss. For example, oil and gas wells in the Kenai National Wildlife Refuge (Alaska) have removed nearly 1,000 acres of habitat106 . Habitat quality may also be affected. The erosion created from aforementioned activities, if carried into sur- face waters, can have a number of adverse effects on local aquatic resources. Development may also fragment habitats and reduce forest and agricultural land cover107 , which is known to affect biodiversity and ecosystem health. For example, concerns have been raised about the possible effects of oil and gas-related hab- itat fragmentation on avian feeding and nesting106 . Due to edge effects, habitat fragmentation has greater effects on smaller seg- ments of intact habitat as compared to larger ones; thus the rate of degradation of habitat due to fragmentation can be expected to increase as development proceeds. The source of water for hydraulic fracturing activities may come from local surface waters, municipal sources, groundwater aqui- fers, or even via recycling of hydraulic fracturing wastewater106 . In Michigan, for example, the State Excelsior 1-13 HD1 well has drawn water from an onsite groundwater well for use in hydraulic fracturing fluid; a number of other Michigan wells have also obtained water from onsite groundwater wells35 . Increased groundwater utilization could stress local water supplies. Surface water is not allowed to be used for hydraulic fracturing in Michigan. Such withdrawals from surface may affect stream flow and water quality108 , and this may affect the heath of aquatic organisms (fish, invertebrates) and ripar- ian vegetation, as well as recreational opportunities such as fishing and canoeing. It is also possible that extensive withdrawals could affect water availability for other uses. A U.S. Government Accountability Office (GAO) study found that of the 575 National Wildlife Refuges (NWR) in the U.S., 105 ref- uges contain a total of 4,406 oil and gas wells. Though rigorous scientific studies are lacking, the information available reveals that construction-related activities as well as spills and accidents have had a detrimental effect on wildlife and habitat106 . A number of chemicals used intentionally in hydraulic fracturing, such as organic hydrocarbons, as well as produced brine water are known to be toxic to wildlife. In Michigan, two NWRs have had oil and gas activ- ities. Kirtland’s Warbler National Wildlife Refuge, one of few small areas with suitable summer habitat for the endangered Kirtland’s Warbler, has two active oil and gas wells, 15 inactive wells, and multiple pipelines; oil and gas exploration activities continue on the Refuge. Shiawassee NWR has 4 inactive wells. It is noted that the intensity of oil and gas activity in Michigan NWRs is limited compared to other regions of the US, such as Louisiana. 3.5 Food and Animal Impacts The possibility exists that hydraulic fracturing may affect companion and production animals. Bamberger and Oswald39 interviewed ani- mal owners near hydraulic fracturing sites in Colorado, Louisiana, New York, Ohio, Pennsylvania, and Texas, and report upon 24 cases of animals being affected. The interviewees reported witnessing adverse effects in a number of animals (e.g., deer, cows, fish, horses, dogs, chickens, song birds, and amphibian) in relation to hydraulic fracturing activities. In one case, an interviewee reported that hydraulic fracturing fluids were released onto a nearby cow pasture thus resulting in the death of 17 cows (necropsy report indicated most likely cause of death was respiratory failure). In another case, hydraulic fracturing fluid leaked onto a pasture in which goats exhibited reproductive impairments following expo- sure. A number of cases of exposure in farm production animals, mainly beef cattle and cows, were also documented. Exposures largely occurred via the spilling or leaking of hydraulic fracturing fluids or flowback/produced water into well water, ponds, and springs. In general, death was reported by farmers one to three days following exposures and reproductive impairment was the most commonly reported ailment. Effects were also documented in companion animals, including cats, dogs, horses, llamas, goats, and fish. Exposures in these animals likely arose due to ingestion of water from a contaminated source, and a range of neurological, reproductive, gastrointestinal, and dermatological effects were reported. Animals may serve as sentinels of human health risks, and in the report by Bamberger and Oswald39 a number of health effects were also reported in several animal owners. 3.6 Earthquakes The concept of anthropogenic earthquakes, in regard to hydrau- lic fracturing effects, has gained increased attention in recent years109 . While the Michigan DEQ indicates that the state “does
    • 21 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 not have the conditions necessary for this to occur”3 , the University of Michigan CLOSEUP survey found that 2% of the Michigan resi- dents questioned were concerned about the possibility of seismic activity. Small earthquakes, no matter the size, can be surprising to a community with no history of such events110 . Concerns are not only regarding personal property damage, but also public infra- structure such as water and sewer lines38 . To date, there is a lack of supportive data identifying hydraulic fracturing as a direct cause of earthquakes109 . However, in 2011 seismic activity was reported in relation to oil and gas drilling by both the National Research Council in Blackpool, England and the Oklahoma Geological Survey. In addition, there is evidence of earthquakes in Ohio in relation to sub-surface fluid injection111 , the sole disposal method in Michigan according to Ellis (2013) in this report. 4.0 THE COMMUNITY ENVRIONMENT 4.1 Public Perceptions A technical report in this series by Hoffman and Wolske (2013) is focused on social and public perception of hydraulic fracturing in Michigan. Here, we briefly highlight the salient points, given that an individual’s comprehension of health-related choices lies at the core of public health112 . The Center for Local, State, and Urban Policy (CLOSUP) conducted a telephone survey of 415 Michigan residents13 between October 21 and 25, 2012, and found that a majority of residents have heard about hydraulic fracturing (82% have heard ‘a lot’ or ‘a little’ of fracking). The general reaction to the term ‘fracking’ was positive for 31% of respondents, negative for 45%, and neutral for 17%. Fifty two percent of respondents believe natural gas drilling will provide more benefits to the State of Michigan as opposed to 24% who believed that it would cause more problems. In terms of ‘general stance on fracking’, more than 50% either ‘strongly’ (32%) or ‘somewhat’ (22%) supported the activity. These numbers are sim- ilar to what has been found nationally, with 50-60% of Americans believing that the benefits of hydraulic fracturing outweigh the risks. Via open-ended questions, Michigan respondents indicated the top three ‘most important risks’ of hydraulic fracturing to be: water contamination (18%), health issues (14%), and other (11%). One-quarter (25%) of the respondents were ‘not sure’ or ‘didn’t know’. The results from this CLOSUP survey are instructive at the State- level though they may not necessarily reflect opinions in regions where gas development is actively being pursued or amongst other sub-populations. There have been a number of news articles from across Michigan documenting public concerns about hydrau- lic fracturing (Table 9). While this Table is not an extensive review nor is it scientifically compiled, it does highlight that Michigan residents have a range of concerns and beliefs. In addition, it is important to note that in the Michigan CLOSUP survey, while many respondents have formed strong opinions in favor and against the activity, there are also a number of individuals who remained neutral or uninformed. For example, 17% of respondents indicated to have ‘never’ heard the term ‘fracking’, and >50% are ‘not too closely’ (32%) or ‘not at all’ (20%) following the discussion in Michigan. It is thus imperative to ensure that public health messages are based on sound science and that misinformation (especially in today’s hyper-connected world) is immediately addressed113 . Moving ahead, there is a need in both Michigan and nationally for an open and constructive dialogue among stakeholders (e.g., public, industry, government, academics, NGOs) who have different per- ceptions of the risks and benefits posed by hydraulic fracturing. A recent survey that asked a variety of experts to identify the most important environmental risks related to hydraulic fracturing oper- ations74 is an important first step in this regard. It is also important to conduct focused studies on individuals that reside amongst hydraulic fracturing development sites. For example, the work by Bamberger and Oswald39 is limited because individuals self-selected to participate; however, it does highlight a variety of concerns raised by local residents. There is a rich literature documenting individuals who believe they have been exposed to industrial chemicals; these individuals often develop chronic stress with health outcomes including depression, lack of control of the environment, increased family quarrels, increased health worries, and increased intrusive and avoidant thoughts114,115,116,117 . In such scenarios, helplessness and fear of the unknown are common complaints118 . Children of parents who report chronic stress from the uncertainty of toxic exposures also tend to report increased stress114 . Trust in both governmental agencies and scientific experts erode when communities perceive a failure to adequately respond to toxic contamination118 . In relation to hydraulic fracturing, polls reveal that many respondents and other stakeholders have con- cerns about the potential for groundwater contamination, ecolog- ical degradation, traffic accidents, and reduced quality of life13 . To our knowledge, in Michigan there has yet to be a poll conducted specifically on people living in close proximity to hydraulic fractur- ing sites. 4.2 Michigan Counties An understanding of the potential impact of hydraulic fracturing on Michigan communities requires some knowledge of the Michigan counties where the activity is taking place. Based on our review of well locations, 8 counties were identified that housed a major- ity of Michigan’s hydraulic fracturing activity (Table 11; Figure 4).
    • 22 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 TABLE 10. Selected news articles from Michigan concerning public concerns over hydraulic fracturing. These are for illustrative purposes only. Date Outlet Title & Lede Sentence Source Jan 16, 2013 Cheboygan Tribune Frack flowback used on county roads. An environmental group says it has learned that frack flowback was spread on roads in Cheboygan County in 2012… http://www.cheboygannews.com/arti- cle/20130116/NEWS/130119634 Jan 30, 2013 Crawford County Avalanche Hydraulic fracturing well pad construction has some residents worried, http://www.crawfordcountyavalanche. com/articles/2013/01/30/news/doc51094b- 883c50f634275933.txt December 27, 2012 The Hastings Banner Fracking top local story of 2012 http://hastingsbanner.com/fracking-top-lo- cal-story-of-p4557-84.htm December 20, 2012 The Detroit News Study predicts Michigan fracking boon - Job gains from gas extraction touted, but some warn of risks http://www.detroitnews.com/arti- cle/20121220/BIZ/212200359 November 12, 2012 Crain’s Detroit Business Splitting the state - Hydraulic fracturing frees gas and oil; opponents say environmental cost too high http://www.crainsdetroit.com/arti- cle/20121111/FREE/311119998/splitting-the- state November 8, 2012 MLive Environmentalists plan ‘fractivisim’ meeting in oppose oil and gas exploration in West Michigan http://www.mlive.com/business/west-mich- igan/index.ssf/2012/11/environmentalists_ plan_fractiv.html October 25, 2012 The Grand Rapids Group files suit to annul state auction of oil and gas leases http://www.mlive.com/business/west-michi- gan/index.ssf/2012/10/group_files_lawsuit_ to_invalid.html Oct 24, 2012 Lansing City Pulse Frack this – six arrested, five face felony charges at anti-fracking protest at Lansing Center http://www.lansingcitypulse.com/lansing/ article-8021-frack-this.html October 12, 2012 MLive State officials move oil and gas auction to larger venue to accommodate anti-fracking protesters http://www.mlive.com/business/west-michi- gan/index.ssf/2012/10/state_officials_move_ oil_and_g.html September 16, 2012 Bay City Times Controversy over ‘ fracking ‘ intensifies - Gas exploration ramping up in state  http://www.mlive.com/business/mid-mich- igan/index.ssf/2012/09/michigans_natural_ gas_industry.html TABLE 11. Demographics and pertinent statistics in Michigan Counties with hydraulic fracturing. State of Michigan Alcona Alpena Antrim Crawford Kalkaska Montmorency Oscoda Otsego Population, 2011 estimate 9,876,187 10800 29386 23316 14014 17160 9653 8608 24078 % of state population (Jul 2009) 100 0.11 0.29 0.24 0.14 0.17 0.10 0.09 0.23 % population change (2000-2009) +0.3 -5.4 -6.5 +3.1 -0.5 +1.9 -2.1 -7.5 +0.5 Mean Household Income, 2007-2011 $48,669 $35,490 $38,081 $42,440 $39,597 $39,130 $34,490 $32,838 $46,303 % Below poverty 15.70% 15.30% 17.10% 16.70% 17% 16.50% 18.60% 20.10% 12.50% % Bachelors Degree or Higher 25.3% 12.80% 15.40% 23.40% 15% 10.40% 10.40% 9.70% 20% Arrests in 2001 187 1029 271 329 924 110 424 840 Arrests in 2011 162 1071 401 695 564 166 367 937
    • 23 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 Collectively, these counties account for 1.37% of Michigan’s total population. As shown below, the majority of natural gas extraction in Michigan is occurring in lower socio-economic areas. Pro- fracking arguments will likely present this as positive, framing it as providing higher wage jobs in low socio-economic areas; depend- ing on the location and duration of added jobs, this may or may not be valid (as discussed by Zullo and Zhang (2013) in this series). Anti-fracking arguments will likely present this as negative, such as by arguing that lower socio-economic areas will bear the burden of fracking’s negative effects while other distant communities receive the benefit (a form of environmental injustice, see Section 4.3). 4.3 Environmental Justice Environmental justice “is the fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income, with respect to the development, implementation, and enforcement of environmental laws, regulations, and policies”119 . Stakeholders involved in the EPA Water Study report have raised environmental justice concerns, and the possibility that hydraulic fracturing activities may disproportionately occur in or near suscep- tible communities. In the case of Michigan, it is not clear whether environmental justice is occurring, or is of concern. However, it is important to note that the mean household income is lower in each of the 8 counties identified earlier than the State-average, and the percent of individuals below poverty is slightly higher than the Michigan average for most Counties. This is of particular con- cern since Michigan ranks in the bottom half in the U.S. in terms of per capita income and other measures of socioeconomic success. 4.4 Boomtown Impact Model A Boomtown Impact Model or social disruption model refers to the socio-cultural changes experienced by communities as a result of rapid population increase due to emergence of new indus- try120,121,122,123 . Utilizing rich literature from the 1970s and 1980s, when rural regions in the United States underwent significant energy development, Jacquet illustrates this concept as it pertains to nat- ural gas development and the hydraulic fracturing culture17,123,124 . While quantitative data is limited due to the general lack of pre- boom baseline studies121,125 , the effects experienced by impacted communities provide various opportunities and challenges in rela- tion to population, demographics, governance, geography, and local infrastructure. From Table 11, there is no clear evidence of a population “boom” in the Michigan Counties in which hydraulic fracturing is occurring. Nonetheless, there are a number of potential public health con- cerns at the community-level that need to be considered if devel- opment activities are to rapidly increase. These are highlighted in the real-world examples that follow in the sections below. 4.4.1 Health and Social Services In the Boomtown Impact Model, the majority of recorded nega- tive impacts are in regard to community distress and demands on social services28 . This is because a population influx can extend community services beyond their capability. For the residents who already experience barriers to access, whether geographi- cal, financial, or physical, a population influx may further hinder access126 . Fire protection, traffic control, road maintenance, park- ing, zoning, and subdivision regulations are examples of factors that may be affected127 . If workers are accompanied by their fami- lies, local school enrollment and recreational opportunities may be affected28 . Enumerating community health can be difficult though factors such as mental health, crime, substance abuse, and sexu- ally transmitted infection (STI) rates can be more easily identified28 . First coined by Dr. Eldean Kohrs in 1974, Albrecht120 describes the above lack of social services and public resources as the “Gillette Syndrome” in reference to the excessive social problems experi- enced in Gillette, Wyoming during an energy development boom. In this community, family tensions, emotional breakdowns, divorce, and increased alcoholism were all associated with the rapid com- munity growth. Demands on medical services have often been reported in indus- trial boomtowns. If new workers are insured these heightened demands can actually help to support the healthcare system. However, if workers are uninsured, additional strain on limited resources and effects on healthcare quality could occur. Statistics on health insurance coverage rates for natural gas workers are Figure 4. Map detailing Michigan Counties identified in Table 11.
    • 24 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 unavailable28 . New skills are also needed for emergency respond- ers to most effectively respond and adequately assist on the drilling site120 . For example, Jacquet123 compared the number of monthly Emergency Medical Service calls in Sublette County (Wyoming) with the number of drilling rigs over the period of 2001 and 2007, and found a highly significant correlation (p<0.001; r2 = 0.695). Due to long working hours and possible geographic isolation, on-site workers may become apathetic towards their own health status, especially when symptoms are not present. In addition, the lack of accessible physical and mental health resources poses barriers128 . Workers may also have a tendency to binge drink and use drugs to relieve stress and help with cope with irregular and long work hours128 . These can also be direct factors in the increase in STIs amongst workers. STI’s tend to be perceived as highly stigmatized in these communities due to potential hyper-masculinity, sexism, and general apathy especially when no symptoms are present. For example, chlamydia rates amongst youth 15-24 years old working and/or living near the oil and gas industry “boom” area of British Columbia (Canada) were 22% higher than the provincial average, which could be a low estimate due to a lack of testing and sexual health resources128 . Rapid growing communities have documented disproportionate rises in crime. A classic sociological viewpoint of this phenomenon is that the criminal activity is likely fueled by the growing transient workforce120 . Freundenberg129 questions this assumption and cate- gorizes law enforcement records as subjective data, vulnerable to agency manipulation and variations in record keeping practices. It is also noted that crime rates are based on census population num- bers and may not include the temporary workers in non-traditional living arrangements. However, Jacquet argues that proven cau- sality is challenging but that reported criminal activity does have a positive association with increasing population123 . For example, Jacquet’s assessment of Sublette County (Wyoming) shows that between 1995 and 2004 the population grew by 21% and during that same time there was a 270% increase in the number of arrests. 4.4.2 Local Socio-Economics Most studies have focused on macro-economic issues surrounding hydraulic fracturing. More data is needed which focuses on the community- and individual-level. At the community-level, inflation- ary issues, such as increase in rent, and property and employment pressures, are common33,123 . This can be pared with higher taxes for long-term residents, in order to accommodate the heightened need for social services and industrial tax-break incentives17,120 . Further, non-local workers will pay income taxes in their home community versus the community in which the wells are located18 . Economic benefits may also be realized at the community and individual levels. Land lease payments and natural gas production royalties are one way in which local residents may earn income. In a case study done in Bradford County, Pennsylvania, where there are high levels of established Marcellus Shale development, leasing rates were reported anywhere from $5 to $5,000 per acre. The state minimum for royalties is 12.5 percent, with some resi- dents receiving up to 20 percent12 . As mineral rights ownership in Pennsylvania remain private, it is difficult to conclude whether or not residents who are directly affected by the industry are being adequately compensated. Even for local residents who are being compensated, these royalties generally only span a few years18 . 4.4.3 The “Bust” As the development phases of hydraulic fracturing come to an end, communities begin to experience the “bust” portion of a “boom and bust” model. This period is characterized as a downturn in economic growth and reduction in employment. In order to miti- gate the effects, it is thought that if other non-gas industry related investors become attracted to the area during the boom period, economic status could remain consistent. However, this has shown to be relatively difficult from a planning standpoint123 . Effects experienced by the impacted communities provide var- ious opportunities and challenges in relation to population, demographics, governance, geography, and local infrastructure. Typically, larger communities are better able to absorb the rapid development and population influx due to previously established infrastructure. Large transformations within a less populated com- munity have greater potential for social disruption. Brown et al.125 proposed to add to the “boom-bust” model a period of “recovery”, thus resulting in a “boom-bust-recovery” model. Via longitudinal studies, Brown et al.121,125 , found that long-term residents of the developed region typically experience increased community satisfaction during the post-boom years. They suggest older residences have a greater commitment and attachment to the area and have seen it through multiple transitions, therefore are less likely to be dissatisfied. Hunter et al.122 refer to this transi- tion as post-boom rebound. 5.0 CHALLENGES AND OPPORTUNITIES 5.1 Lack of Michigan Data I n Michigan approximately 12,000 wells have been hydrauli- cally fractured, and the practice has been used on the vast majority of wells drilled in recent years (78%)3 . High-volume hydraulic fracturing is less common, but the operation does exist in the State4 . According to a Michigan DEQ factsheet, hydraulic fracturing has had no “consequence to the environment
    • 25 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 or public health”3 . We found limited scientific evidence to discount nor support this statement, and this is problematic especially if development activities are to increase. Data are needed to estab- lish baseline conditions, understand potential cumulative risks, assess trends over time and space, and to determine whether or not activities are proceeding in a safe manner. The greatest challenge in assessing the potential public health risks of hydraulic fracturing in Michigan is the lack of State-specific data. While this challenge is specific to Michigan, it also applies to many other regions as concluded at the 2012 Institute of Medicine’s Roundtable. Nonetheless, decisions need to be made and a number of potential hazards related to hydraulic fracturing exist; these have been reviewed in Sections 2, 3, and 4 of this report. In Michigan, however, there exists limited data on public health indicators (Table 2). While Michigan has had a long history of hydraulic fracturing, compared to other natural gas deposits, such as the Barnett and Marcellus Shales, the intensity and mag- nitude of contemporary hydraulic fracturing in Michigan is much less (i.e., most wells are low-volume, and high-volume hydraulic fracturing is not commonplace). A great opportunity exists for Michigan to learn lessons from other States, and put into place a mechanism by which baseline data may be properly collected, archived, and analyzed before high-volume hydraulic fracturing development intensifies. Without local and State-specific, objec- tive, and peer-reviewed scientific evidence, it will be a challenge for risk assessors and decision makers to make sound judgments. In terms of establishing a public health tracking system for Michigan’s hydraulic fracturing industry, a number of models exist. In the Institute of Medicine’s (IOM) report “The Future of Public’s Health in the 21st Century”, the U.S. Centers for Disease Control’s (CDC) ten essential public health services were identified as part of a fundamental framework of activities that should be undertaken by all communities (Table 12). Many of these, such as monitor- ing community health status, and informing and empower peo- ple about health issues may be achieved via existing programs. Another example is the U.S. CDC’s Environmental Public Health Tracking Program, which facilitates ongoing collection, integration, analysis, interpretation, and dissemination of data from environ- mental hazard monitoring and from human exposure and health effects surveillance130 . Much of the discussion to date has focused on groundwater studies, though as realized via our review here, there exist a number of other important indicators that could be tracked in addition to water quality. TABLE 12. The 10 essential public health services. From the U.S. CDC131 . Monitor health status to identify and solve community health problems. Diagnose and investigate health problems and health hazards in the community. Inform, educate, and empower people about health issues. Mobilize community partnerships and action to identify and solve health problems. Develop policies and plans that support individual and community health efforts. Enforce laws and regulations that protect health and ensure safety. Link people to needed personal health services and assure the provi- sion of health care when otherwise unavailable. Assure competent public and personal health care workforce. Evaluate effectiveness, accessibility, and quality of personal and popu- lation-based health services. Research for new insights and innovative solutions to health problems. 5.2 Exposure Assessment and Epidemiological Causation Much of this report may be viewed as a “Hazard Identification” exercise, which is the first step in human health risk assessment. As emphasized in the beginning of this report, the presence of a hazard does not necessarily mean that there will be a health risk to workers or community residents. Recall that a hazard is any biological, chemical, physical or psychosocial agent/condition that has the potential to cause harm. While a number of hazards clearly exist in hydraulic fracturing operations (=Hazard Identification), it next needs to be determined whether or not people are exposed to them (=Exposure Assessment). Such is the foundation of risk assessment (Figure 2). While hazards have been identified, few exposure assessments have been conducted in hydraulic fracturing sites, and even fewer have tried to account for their possible cumu- lative health impacts. Despite studies now measuring a number of chemicals and agents in air and water, to our knowledge research has not yet tied these hazards to any internal dose measurement via biomarkers (e.g., urine, blood, breath). To our knowledge, no such activity (even basic ones to assess air and water quality) has been published in Michigan, and this represents a challenge. Another challenge in any human health study concerns epidemio- logical causation. Even if a hazard is identified at a hydraulic frac- turing facility and people become exposed, linking it to an adverse health outcome (=Dose-Response Study) through an expo- sure-disease model will require carefully designed epidemiological studies that include, for example, a robust sample size, state-of-the art exposure assessments, temporal sampling to assess potential latent effects, and replication by others. Foundational elements described in Hill’s criteria for causation should be followed, as would be expected in any rigorous epidemiological study. Ideally, these efforts would already be underway given the extent of
    • 26 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 existing hydraulic fracturing in the State of Michigan. One major criticism of risk assessments from other States concerns a lack of baseline data. While this criticism also applies to Michigan, com- pared to other States such as Pennsylvania and Texas, the mag- nitude and intensity of high-volume, horizontal drilling is much less in Michigan and thus an opportunity may still exist to initiate studies that may qualify as baseline. There exist a myriad of hazards in the hydraulic fracturing industry, and not surprisingly many have advocated for the precautionary principle. Historically, action against harmful environmental and public health hazards has only occurred after the scientific commu- nity has proven its danger. In our report we were unable to find very many documented public health issues related to hydraulic frac- turing (low-volume or high-volume) in Michigan, either because they have not been reported upon or they have not occurred. Nonetheless, before high-volume, horizontal drilling intensifies in Michigan, an opportunity should be seized to initiate basic yet important public health studies. The lack of these in other regions is leading to much criticism and debate, which Michigan has the chance to avoid. 5.3 Chemicals and Disclosures There has been some disclosure of chemicals used in hydraulic fracturing fluids at sites in Michigan, perhaps representing a small sample of Michigan-specific data. The Michigan DEQ requires operators of high-volume hydraulically fractured wells to submit a Material Safety Data Sheet (MSDS)i for each chemical additive in the hydraulic fracturing fluid used, and the DEQ posts this infor- mation on its website. Based on the MSDSs for the 15 well sites posted on the website as of July 24, 2013ii , an average of 18 chem- ical additives were used per well (range: 8-32)132 . A subset of the high-volume hydraulically fractured well operators and other well operators have submitted information to FracFocus voluntarily. Via the FracFocus database, 12 Michigan well sites uploaded data as of January 29, 2013. From these, the average number of chemicals disclosed per well was 25 (range: 13-55). This disclosure is, however, minimal with only a few facilities report- ing upon a small number of drilling events out of more than 12,000 past and presently operating wells that have undergone hydraulic fracturing in the State. In the case of reported additives, the chem- ical identity of certain additives is withheld as a trade secret on the MSDSs. In addition, the information available on either website does not include data regarding the amount of chemicals used, even though operators of high-volume hydraulically fractured wells are required to report the total volume of each additive to the DEQ. Beyond the MSDSs and FracFocus database we are not aware of any published studies that report upon levels of chemicals in air and water near Michigan facilities. This lack of information represents a challenge. Considering that ~1% of the hydraulic fracturing fluid is composed of various chemicals (~50,000 gallons for a typical well that uses 5 million gallons of hydraulic fracturing fluid), chemicals are potentially being released into Michigan water resources. Chemicals intentionally used in hydraulic fracturing serve a number of functions, but few of them have undergone rigorous toxicologi- cal testing. Moreover, the most prominent chemicals (Tables 4 and 5) are proven human health hazards. In addition, an outstanding feature in the toxicological sciences, which is of clear relevance to hydraulic fracturing, is the lack of understanding concerning tox- icant-toxicant interactions (mixture effects), how these toxicants may change with varying temperatures and other industrial condi- tions, and how toxicants may interact with non-chemical stressors (e.g., built environment, socioeconomic status, food security) to influence health. Given the aforementioned challenges, an opportunity exists for Michigan State officials and industry to adopt schemes to minimize potential adverse effects associated with the use of chemicals. For example, Encana133 has a Responsible Products Program which “helps ensure that the hydraulic fracturing fluid products we use in our operations are as safe, effective, and as environmentally responsible as possible”. This resulted in a risk-based product assessment and management program, and has led the company to stop using 2-butoxyethanol, benzene, and diesel in its hydraulic fracturing fluids. Chesapeake Energy, via the Green Frac134 pro- gram they initiated in 2009, has a similar mission. Green Frac and other programs are also embracing green chemistry principles that seek to remove the most hazardous chemicals from operation and replace them with environmentally friendly alternatives. 5.4 Public Health Outreach and Education Surveys indicate that the Michigan public is divided, where some residents support hydraulic fracturing and others do not, and res- idents also have some confusion regarding hydraulic fracturing13 . An increase in public health education and outreach regarding hydraulic fracturing could aid Michigan residents. The 2012 CLOSUP survey of 415 Michigan residents revealed that a majority of respondents have heard about hydraulic fracturing and that most have a negative or neutral reaction to the term ‘hydraulic fractur- ing’. Though, more than half the respondents believe natural gas drilling will provide more benefits to the State, and more than half i The Michigan DEQ Guide to Environmental, Health, and Safety Regulations (2012) Appendix E provides a summary of what is included in an MSDS. Available: http://www.michigan.gov/deq/0,4561,7- 135-3310_4148-15820--,00.html ii Information added by report editor in collaboration with report author.
    • 27 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 either ‘strongly’ or ‘somewhat’ support the activity. Many respon- dents also expressed uncertainty for a number of questions, which seems to be the case across many States, such as Pennsylvania135 . Given the divide in public opinion responses, coupled with a range of media reports, anecdotes, and mis- or lack of information, a lim- ited number of objective and peer-reviewed scientific studies, and a growing industry, there is a need to better communicate risks and benefits to all stakeholders, including the general public. Providing inclusive community access to information and mean- ingful participation is a foundational principle in public health136 . In order to address this issue, taskforce committees could be formed at a variety of levels given the varying opinions and diverse stake- holder groups137 . At the community level, this taskforce could be composed of neighborhood residents, local government officials, public health professionals, businesses, industry representatives, and landowners. Such a taskforce would then be responsible for creating a fair and transparent forum and could also oversee base- line, local studies. To our knowledge, such taskforces do not exist in Michigan though could be achieved with limited resources. 5.5 Public Health Benefits and Tradeoffs There exist no risk-free energy development schemes, and all activities (renewable and non-renewable) pose some degree of risk to public health. Therefore, any public health assessment of hydraulic fracturing in Michigan needs to be conducted with care- ful consideration of other energy sources, relative tradeoffs, and associated public health risks and benefits. It must also be realized that risks and benefits can vary from the local to regional/State, national and international levels. The energy marketplace in the U.S. is shifting with the reliance on coal and hydroelectric diminishing and reliance on natural gas increasing (Figure 5). This is being driven by abundant and rela- tively cheap natural gas reserves, coupled with new rules governing power plants such as the Mercury and Air Toxics Standard (MATS). The opportunities associated with this shift include new jobs and increased economic activity, a more diverse and stable energy base, and decreased reliance upon foreign sources. In addition, utilizing natural gas instead of coal or oil may also yield improved air quality. For example, the deceased dependency upon coal (48% to 36% from 2008 to 2012) has been estimated to reduce car- bon dioxide releases by 300 million tons or 13%. Release of other notable air pollutants has been estimated to be lower from natural gas operations when compared to oil and coal (Table 13). However, there still exists some debate concerning the release of methane, and thus contributions to greenhouse gas emissions92,93,94 . The der- ivation of such numbers needs to be performed at the level of the State, nation, and world. Figure 5. Proportion of energy arising from various sources. Coal-fired electricity genera- tion is declining rapidly as the use of natural gas and renewable energy expands. Source: EIA, “Annual Energy Review”, September 2012 and EIA “Electric Power Monthly,” October 2012, cited in JISEA, November 2012.
    • 28 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 TABLE 13. Combustion emissions (pounds/billion BTU of energy input) from natural gas in relation to oil and coal. Adapted from DOE report. Natural Gas Oil Coal Carbon Dioxide 117,000 164,000 208,000 Carbon Monoxide 40 33 208 Nitrogen Oxides 92 448 457 Sulfur Dioxide 0.6 1,122 2,591 Particulate Matter 7.0 84 2,744 Formaldehyde 0.750 0.220 0.221 Mercury 0.000 0.007 0.016 There seems to be some consensus in the public opinion that hydraulic fracturing may offer more ‘benefits’ than ‘problems’ (in Michigan, 52% indicating ‘more benefits’ versus 24% indicating ‘more problems’13 ). Via an open-ended question on the ‘primary potential benefit of fracking’, 27% of respondents in Michigan indicated that it is a mechanism to attain energy independence and 15% said that it will reduce carbon emissions. Results from Pennsylvania were similar to Michigan. TABLE 14. Potential public health trade-offs when shifting from coal and oil to natural gas as princi- ple energy source. Pros Cons • calculations suggest less air pollutants will be released; • may promote development of new generation of machinery and technologies; • terrestrial footprint of a horizontal hydraulic fracturing operation may be less than a conventional vertical oil & gas operation • cheap natural gas may hinder development of potentially “greener” renewable energy technologies; • foreign energy suppliers may shift exports to countries less able to cope with increased development and pollution; • methane release and global warming potential remains debated 6.0 PROPOSED PRIORITIZED PATHWAYS FOR FUTURE WORK 6.1 Michigan-Specific Evidence B ased on Sections 5.1 and 5.2, empirical data is needed in Michigan concerning a number of public health indicators, such as air and water quality, exposure assessments in workers, and health of fish and wildlife. Such data is needed to help establish baseline measurements, make judgments against acceptable thresholds, and compare to other hydraulic fracturing regions. There are some important data- sets available (e.g., well locations), and to broadly assess poten- tial for risk these could be overlaid with datasets such as location of homes, agricultural fields, hospitals, and schools. The current report can be viewed as one that has identified a number of poten- tial hazards. 6.2 Chemical Disclosures Based on Section 5.3, there needs to be much greater understand- ing of what chemicals are being used in every well, with information related to volumes, amounts, disposal plans, etc. made available. Only a handful of wells have disclosed. 6.3 Public Opinion, Education, and Outreach Based on Section 5.4, more public opinion surveys are needed, particularly local-scale surveys from affected communities. In addi- tion, meetings engaging all stakeholders should be promoted and facilitated to gain a deeper understanding of varying perspectives. Trusted social scientists should be enlisted to carry out such activi- ties, rather than those viewed to have biases. 6.4 Tradeoffs Based on Section 5.5, all energy development schemes have inher- ent risk and thus a health economist should be enlisted to help enumerate risks-benefits of hydraulic fracturing in Michigan versus alternate energy sources, both in terms of health outcomes but also economic value. 7.0 ACKNOWLEDGMENT This publication is a result of work sponsored by the University of Michigan. We thank comments by four anonymous reviewers.
    • 29 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 1. Schenk C. An estimate of undiscovered conventional oil and gas resources of the world, 2012: U.S. Geological Survey Fact Sheet 2012–3042. 2012;6. Available from: http://pubs.usgs.gov/fs/2012/3042/ 2. Wilson J, Schwank J. Environmental and other issues related to the use of fracking for the enhanced recovery of natural gas and oil from shale, tight rock and coal beds. 2012. 3. Michigan Department of Environmental Quality (US). Questions and answers about hydraulic fracturing in Michigan [Internet]. [cited 2013 March 2]. Available from: http://www.michigan.gov/documents/deq/deq-FINAL-frack-QA_384089_7.pdf 4. Michigan Department of Environmental Quality (US). High Volume Hydraulic Fracturing: Active Applications and Issued Permits-Since 2008 [Internet]. MDEQ [revised 2013 May 9]. Available from: http://www.michigan.gov/documents/deq/High_Volume_Hydraulic_Fracturing_Activity_MAP_423435_7.pdf 5. Kormacher K, Jones W, Malone S, Vinci L. Public health and high volume hydraulic fracturing. New Solutions. 2013;23(1):13-31. 6. American Schools of Public Health [Internet]. What is public health?; c2013. [cited 2013 March 2]. Available from: http://www.asph.org/document.cfm?page=300 7. World Health Organization [Internet]. Frequently asked questions; c2013. [cited 2013 March 2]. Available from: http://www.who.int/suggestions/faq/en/index.html 8. National Research Council (US). Science and judgment in risk assessment [Internet]. Washington (DC): National Academy Press; c2013 [cited 2013 Feb 5]. Available from: http://www.nap.edu/openbook.php?record_id=2125&page=R1 9. Jiang M, Griffin WM, Hendrickson C, Jaramillo P, VanBriesen J, Venkatesh A. Life cycle greenhouse gas emissions of Marcellus shale gas. Environ Res Lett. 2011;6(3):034014. 10. Frumkin H. Environmental health; from global to local. 2nd ed. San Francisco (CA): John Wiley & Sons, Inc; 2010. p. 49-78. 11. United States Bureau of Labor Statistics. Local area unemployment statistics. [Internet]. United States Department of Labor. [modified December, 2012, cited March 10, 2013]. Available from http://www.bls.gov/lau/ 12. Braiser K, Filteau M, McLaughlin D, Jacquet J, Stedman R, Kelsey T. Residents’ perceptions of community and environmental impacts from develop- ment of natural gas in the Marcellus Shale: a comparison of Pennsylvania and New York cases’. J Rural Soc Sci. 2001;26(1):32-61. 13. Borick C. Public opinion regarding fracking: the Michigan and Pennsylvania perspectives. The National Surveys on Energy and Environment. In: Center for Local, State, and Urban Policy. 2012 Dec 3; Ann Arbor, MI: Gerald R. Ford School of Public Policy; University of Michigan; Muhlenberg College. [cited 2013 Jan 5]. Available from: http://www.closup.umich.edu/files/2012_12_03_fracking_borick.pdf 14. IHS Global Insight (USA) Inc. The economic and employment contributions of shale gas in the United States [report on the Internet]. Washington (DC): America’s Natural Gas Alliance; 2011 [cited 2012 Nov 15]; Available from http://www.anga.us/media/content/F7D1441A-09A5D06A9EC93BBE46772E12/files/shale-gas-economic-impact-dec-2011.pdf 15. American Natural Gas Alliance [Internet]. Michigan. c2013 [cited 2013 Feb 2]. Available from: http://anga.us/why-natural-gas/jobs/natural-gas-in-my-state/mi#.UT0NPXxtXKx 16. Brundage T, Kelsey T, Lobdell J, Michael L, Jacquet J, Ladlee J, Lorson J, Murphy T. Pennsylvania statewide Marcellus Shale workforce needs assess- ment [report on the Internet]. Marcellus Shale Education and Training Center; 2011 [cited on 2013 Jan 10]. Available from: http://www.shaletec.org/docs/PennsylvaniaStatewideWorkforceAssessmentv1_Final.pdf 17. Jacquet J. Community and economic impacts of Marcellus Shale natural gas development. In: Cornell Cooperative Extension Webinar. 2010 May 12. [cited on 2013 Jan 10]. Available from: http://cce.cornell.edu/EnergyClimateChange/NaturalGasDev/Documents/PDFs/CCE%20webinar%20social%20 and%20economic%20impacts%20051210.pdf 18. Black B, Ladson M. The legacy of extraction: reading patterns and ethics in Pennsylvania’s landscape of energy. Penn His: J Mid-Atl Stud. 2012;79(4):377-394. 19. Crawford J, MacCalman L, Jackson C. The health and well-being of remote and mobile workers. Occup Med-c. 2011;61:385-394. 20. Occupational Safety and Health Administration (US) [Internet]. Oil and gas well drilling and servicing e-tool. [cited 2013 March 2]. Available from: http://www.osha.gov/SLTC/etools/oilandgas/index.html 21. Lerman SE, Eskin E, Flower DJ, George EC, Gerson B, Hartenbaum N, Hursh SR, Moore-Ede M. Fatigue risk management in the workplace. J Occup Environ Med. 2012;54(2):231–258. 22. Bureau of Labor Statistics (US) [Internet]. Washington (DC): Oil and gas industry fatal and nonfatal occupational injuries. 2010 [cited 2013 Feb 5]. Available from: http://www.bls.gov/iif/oshwc/osh/os/osar0013.htm 23. Retzer KD, Hill RD, Pratt SG. Motor vehicle fatalities among oil and gas extraction workers. Accid Anal Prev. 2013;51:168–74. LITERATURE CITED
    • 30 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 24. Bureau of Labor Statistics (US) [Internet]. Washington (DC): State occupational injuries, illnesses, and fatalities. 2012 [cited 2013 February 5]. Available from: http://www.bls.gov/iif/oshstate.htm#CO 25. Bureau of Labor Statistics (US) [Internet]. Washington (DC): Work place injuries and illnesses. 2011 [cited 2013 Feb 5]. Available from: http://www.bls.gov/iif/oshwc/osh/os/ostb3191.pdf 26. Dalrymple A, Passi P. Minn. man killed, Halliday man injured in accident at hydraulic fracturing site. The Dickinson Press [Internet]. 2013 January 21 [cited 2013 Feb 11]. Available from: http://www.thedickinsonpress.com/event/article/id/64955/ 27. Eligon J. An oil boom takes a toll on health care. The New York Times [Internet]. 2013 January 27 [cited 2013 Feb 11]. Available from: http://www.nytimes.com/2013/01/28/us/boom-in-north-dakota-weighs-heavily-on-health-care.html?nl=todaysheadlines&emc=edit_th_20130128&_r=0 28. Witter R, Stinson K, Sackett H, Putter S, Kinney G, Teitelbaum D, Newman L. Potential exposure-related human health effects of oil and gas develop- ment: a literature review (2003-2008), Colorado School of Public Health. University of Colorado Denver. 2008 Aug 1. 29. Job RFS. The influence of subjective reactions to noise on health effects of the noise. Environ Intern. 1996;22(1): 93–104. 30. Stansfeld S, Matheson MP. Noise pollution: non-auditory effects on health. Brit Med Bull. 2003;68(1): 243–257. 31. Vaughan A, Pursell D. Frac attack: risks, hype, and financial reality of hydraulic fracturing in the shale plays. 2010 July 8. 32. New York City Department of Environmental Protection (US). Final impact assessment report; mpact assessment of natural gas production in the New York City water supply watershed. Hazen and Sawyer. 2009 Dec 22. 33. Witter R, McKenzie L, Towle M, Stinson K, Kenneth S, Newman L, Adgate J. Health impact assessment for battle mesa, Garfield County Colorado. Colorado School of Public Health, University of Colorado Denver. September 2010. 34. Schindler D. Survey of accidental and intentional hydrogen sulfide (H2S) releases causing evacuations and/or injury in Manistee and Mason Counties from 1980 to 2003. 2003 March 7. 35. Frac Focus Chemical Disclosure Registry [Internet]. c2013 [cited on 29 January 2013]. Available from: http://fracfocus.org/ 36. Stevens R, Davis S, Thomas D, Anderson L, Wilson B. Electric power, pineal function, and the risk of breast cancer. J Fed Am Soc for Exp Bio. 1992;6:853-860. 37. Davis S, Mirick D, Stevens R. Night shift work, light at night, and risk of breast cancer. J Natl Cancer I. 2001;93(20):1557-1562. 38. Dutzik T, Ridlington E, Rumpler J. The costs of fracking; the price tag of dirty drilling’s environmental damage. Environment Maryland Research & Policy Center. 2012. 39. Bamberger M, Oswald R. Impacts of gas drilling on human and animal health. New Solutions. 2012;22(1):51-77. 40. Michigan Department of Environmental Quality (US). Hydrogen sulfide (H2S) – Q&A. c2001-2013 [cited 2013 Jan 6]. Available from: http://www.michigan.gov/deq/0,4561,7-135-3311_4111_4231-9162--,00.html 41. Occupational Safety and Health Administration (US). Worker exposure to silica during hydraulic fracturing hazard alert. 2012;6:7. 42. Minnesota Department of Natural Resources (US). Industrial silica sands of Minnesota. Saint Paul (MN): 2012;43:2. 43. Wisconsin Department of Natural Resources (US). Silica sand mining in Wisconsin. Madison (WI): 2012. 44. National Institute for Occupational Safety and Health (US). (1986). Occupational respiratory diseases. Washington (DC): U.S. Department of Health and Human Services, Public Health Services, Center for Disease Control. 1986;86-102. 45. Colborn T, Kwiatkowski C, Schultz K, Bachran M. Natural gas operations from a public health perspective. Int J Hum Ecol Risk Assessment. 2011;17(5):1039–1056. 46. United States Department of Energy. Modern shale gas development in the United States: a primer. Oklahoma City (OK). 2009. 47. Gosman S, Robinson S, Shutts S, Friedmann J. Hydraulic fracturing in the Great Lakes Basin: the state of play in Michigan and Ohio. A legal analysis by The National Wildlife Federation. 2012. 48. Encana Corporation [Internet]. Chemical use. c2013. [cited 2013 March 2]. Available from: http://www.encana.com/environment/water/fracturing/chemical-use.html 49. United States Environmental Protection Agency. Plan to study the potential impacts of hydraulic fracturing on drinking water resources. Washington (DC). 2011. 50. TEDx The Endocrine Disruption Exchange [Internet]. c2013. [cited 2013 March 12]. Available from: http://www.endocrinedisruption.com/home.php 51. Waxman H, Markey E, DeGette D. Chemicals used in hydraulic fracturing. United States House of Representatives Committee on Energy and Commerce Minority Staff. 2011 April. 52. Sedivec V, Mráz M, Flek J. Biological monitoring of persons exposed to methanol vapours. Int Arch Occ Env Hea. 1981;48(3):257–271. 53. Liesivuori J, Savolainen H. Methanol and formic acid toxicity: biochemical mechanisms. Pharmacol Toxicol. 1991;69(3):157–163.
    • 31 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 54. Gomes R, Liteplo R, Meek M. Concise international chemical assessment document 45; ethylene glycol: human health aspects. Geneva. 2002. 55. Jacobsen D, Hewlett T, Webb R, Brown S, Ordinario T, McMartin K. Ethylene glycol intoxication: evaluation of kinetics and crystalluria. Am J Med. 1988;84(1):145–152. 56. Faroon O, Tylenda C, Harper C, Yu D, Cadore A, Bosch S, Wohlers D, Plewak D, Carlson-Lynch H. Toxicological profile for ethylene glycol. Atlanta (GA). 2010. 57. Hess R, Bartels M, Pottenger L. Ethylene glycol: an estimate of tolerable levels of exposure based on a review of animal and human data. Arch Toxicol. 2004;78(12):671–680. 58. Reinhart P, Gift J, Howard A. Toxicological review of ethylene glycol monobutyl ether. (CAS No . 111-76-2). Washington (DC). 2010. 59. National Institute for Occupation Safety and Health (US). NIOSH: criteria for a recommended standard - ethylene glycol monobutyl ether and ethylene glycol monobutyl ether acetate. Cincinnati (OH). 1990. 60. Flowers L, Broder M, Forsyth C. Toxicological review of acetone. Washington (DC). 2003. 61. Sigma-Aldrich. Isopropanol: MSDS No. I9516 [Internet]. St. Louis (MO): Sigma-Aldrich; 2012 [cited 2013 March 15]; Available from: http://www.sigmaaldrich.com/MSDS/MSDS/PleaseWaitMSDSPage.do?language=&country=US&brand=SIGMA&product- Number=I9516&PageToGoToURL=http%3A%2F%2Fwww.sigmaaldrich.com%2Fcatalog%2Fsearch%3Finterface%3DAll%26ter- m%3Disopropanol%26lang%3Den%26region%3DUS%26focus%3Dproduct%26N%3D0%2B220003048%2B219853269 %2B219853286%26mode%3Dmatch%2520partialmax 62. Searl A, Galea K. Toxicological review of the possible effects associated with inhalation and dermal exposure to drilling fluid production streams. Institute of Medicine Research Report. 2011 August. 63. Skyberg K, Rønneberg A, Kamøy JI, Dale K, Borgersen A. Pulmonary fibrosis in cable plant workers exposed to mist and vapor of petroleum distillates. Environ Res. 1986;40(2):261–73. 64. Department of Health and Human Services, & Department of Labor (US). Occupational health guideline for petroleum distillates (Naphtha). 1987. 65. Michigan Department of Environmental Quality, Geological Survey Division. Incident Report Murray – State et al 1-8 (PN 37657) Gas Well Michigan Production Company, LLC Manistee Township, Manistee County: Gas Release of October 28, 1998. MDEQ; n.d. Murray State 1-8 Report. 63p. 66. Agency for Toxic Substances and Disease Registry (US). Toxicological profile for fuel oils [Internet]. Atlanta (GA): U.S. Department of Health and Human Services, Public Health Service; 1996. Available from: http://www.atsdr.cdc.gov/toxfaqs/tfacts75.pdf. 67. Hesterberg T, Long C, Bunn W, Lapin C, McClellan R, Valberg P. Health effects research and regulation of diesel exhaust: an historical overview focused on lung cancer risk. Inhal Toxicol. 2012;24(1):1–45. 68. Terazawa K, Tomii S, Takator T, Nakano K. Methane asphyxia: Coal mine accident investigation of distribution of gas. Am J Forensic Med Pathol., 1985;6(3): 211–214. 69. Sigma-Aldrich. Methane: MSDS No. 295477 [Internet]. St. Louis (MO): Sigma-Aldrich; 2013 [cited 2013 March 15]; Available from: http://www.sigmaaldrich.com/MSDS/MSDS/DisplayMSDSPage.do?country=US&language=en&productNum- ber=295477&brand=ALDRICH&PageToGoToURL=http%3A%2F%2Fwww.sigmaaldrich.com%2Fcatalog%2Fsearch%3Finter- face%3DAll%26term%3Dmethane%26lang%3Den%26region%3DUS%26focus%3Dproduct%26N%3D0%2B220003048%2B219853269 %2B219853286%26mode%3Dmatch%2520partialmax 70. Occupational Safety and Health Administration (US). Hydrogen sulfide (H2S) Washington (DC). 2005;2. 71. Agency for Toxic Substances and Disease Registry (US). Toxicological profile for hydrogen sulfide (update) [Internet]. Atlanta (GA): U.S. Department of Health and Human Services, Public Health Service; 2006. Available from: http://www.atsdr.cdc.gov/tfacts114.pdf 72. Percy S, Lubchenco J, Almeida F, Armsworth A, Bennett A, Ebi K, Ehrmann J, Griffith J, Kumar P, Pagiola S, et al. Ecosystems and human well-being; opportunities and challenges for business and industry [Internet]. Millennium Ecosystem Assessment. [cited 2013 March 12]. Available from: http://www.millenniumassessment.org/documents/document.353.aspx.pdf 73. New York State Department of Environmental Conservation (US). Supplemental generic environmental impact statement on the oil, gas and solution mining regulatory program; well permit issuance for horizontal drilling and high-volume hydraulic fracturing to develop the Marcellus Shale and other low-permeability gas. Albany (NY). 2011. 74. Krupnick A, Gordon H, Olmstead S. Pathways to dialogue: what the experts say about the environmental risks of shale gas development. Washington (DC). 2013;78. 75. ALL Consulting. NYDEC information requests, prepared for the Independent Oil & Gas Association of New York. 2010 September. Project No. 1284. 76. Cooley H, Donnelly K, Ross N, Luu P. Hydraulic fracturing and water resources: separating the frack from the fiction. Oakland (CA). 2012. 77. Accessmygov.com [Internet]. Municipal directory. c2013. [cited 2013 March 2]. Available from: https://www.accessmygov.com/MunicipalDirectory?uid=2075
    • 32 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 78. Michigan Office of Highway Safety Planning (US). Reported traffic crashes by county in Michigan. Lansing (MI). 2011;4. 79. McKenzie L, Witter R, Newman L, Adgate J. Human health risk assessment of air emissions from development of unconventional natural gas resources. Sci Total Environ. 2012;424: 79–87. 80. United States Environmental Protection Agency. Greenhouse gas emissions reporting from the petroleum and natural gas industry. Washington, (DC). 2010;144. 81. United States Environmental Protection Agency. Regulatory impact analysis: final newsource performance standards and amendments to the National Emissions Standards for hazardous air pollutants for the oil and natural gas industry. 2012. 82. Agency for Toxic Substances and Disease Registry (US). Toxicological profile for nitrogen oxides [Internet]. Atlanta (GA): U.S. Department of Health and Human Services, Public Health Service; 2002. Available from: http://www.atsdr.cdc.gov/toxfaqs/tfacts175.pdf 83. United States Environmental Protection Agency [Internet]. Nitrogen dioxide. [updated 2013 Feb 2; cited 2013 March 2]. Available from: http://www.epa.gov/airquality/nitrogenoxides/index.html 84. United States Environmental Protection Agency [Internet]. Particulate matter (PM). [updated 2013 Feb 8; cited 2013 March 2]. Available from: http://www.epa.gov/airquality/particlepollution/index.html 85. United States Environmental Protection Agency [Internet]. Carbon monoxide. [updated 2012 Feb 12; cited 2013 March 2] Available from: http://www.epa.gov/airquality/carbonmonoxide/index.html 86. Agency for Toxic Substances and Disease Registry (US). Toxicological profile for carbon monoxide [Internet]. Atlanta (GA): U.S. Department of Health and Human Services, Public Health Service; 2012. Available from: http://www.atsdr.cdc.gov/toxfaqs/tfacts201.pdf 87. Agency for Toxic Substances and Disease Registry (US). Toxicological profile for sulfur dioxide [Internet]. Atlanta (GA): U.S. Department of Health and Human Services, Public Health Service; 1999. Available from: http://www.atsdr.cdc.gov/tfacts116.pdf 88. United States Environmental Protection Agency [Internet]. Sulfur dioxide. 2013 [update 2013 Feb 5; cited 2013 Feb 6]. Available from: http://www.epa.gov/airquality/sulfurdioxide/basic.html 89. Centers for Disease Control and Prevention (US). Community fact sheet: VOCs in blood. [cited 2013 Feb 5]. Available from: http://emergency.cdc.gov/gulfoilspill2010/voc_coastalresidents.asp 90. Colborn T, Schultz K, Herricka, L, Kwiatkowskia C. An exploratory study of air quality near natural gas operations. Human and Ecological Risk Assessment: An International Journal [Internet]. 2012 Nov 25 [cited 2013 Feb 5]. Available from: http://www.tandfonline.com/doi/abs/10.1080/10807039.2012.749447 91. Stephenson T, Valle JE, Riera-Palou X. Modeling the relative GHG emissions of conventional and shale gas production. Environ Sci Technol. 2011;45(24):10757–10764. 92. Howarth RW, Santoro R, Ingraffea A. Methane and the greenhouse-gas footprint of natural gas from shale formations. Climatic Change. 2011;106(4):679–690. 93. Weber CL, Clavin C. Life cycle carbon footprint of shale gas: review of evidence and implications. Environ Sci Technol. 2012;46(11):5688–5695. 94. Alvarez RA, Pacala SW, Winebrake JJ, Chameides WL, Hamburg SP. Greater focus needed on methane leakage from natural gas infrastructure. Proc Natl Acad Sci U.S.A. 2012;109(17):6435–6440. 95. Resources for the Future. Center for Energy Economics and Policy. Pathways to dialogue: what the experts say about the environmental risks of shale gas development [Internet]. Washington (DC). 2013 February. [cited 2013 March 17]. Available from: http://www.rff.org/Documents/RFF-Rpt-PathwaystoDialogue_FullReport.pdf 96. United States Environmental Protection Agency. Office of Research and Development. Study of the potential impacts of hydraulic fracturing on drinking water resources; progress report [Internet]. Washington (DC). 2012 December. [cited on 2013 March 12]. Available from: http://www.epa.gov/hfstudy/pdfs/hf-report20121214.pdf 97. Michigan Department of Environmental Quality (US). Office of Geological Survey. Hydraulic fracturing of natural gas wells in Michigan [Internet]. [updated 2011 May 31; cited 2013 March 12]. Available from: http://www.michigan.gov/documents/deq/Hydrofrac-2010-08-13_331787_7.pdf 98. Michigan Department of Environmental Quality (US). Resource Management Division. Office of Oil, Gas, and Minerals [Internet]. Regulatory response to hydraulic fracturing in Michigan. [cited 2013 March 12]. Available from: http://www.michigan.gov/documents/deq/Regulatory_Responce_376699_7.pdf 99. Moniz E, Jacoby H, Meggs A, Armstrong R, Cohn D, Connors S, Deutch J, Ejaz Q, Hezir J, Kaufman G. Future of natural gas; an interdisciplinary MIT study [Internet]. [cited 2013 March 12]. Available from: http://mitei.mit.edu/system/files/NaturalGas_Report.pdf 100. United States Army Corps of Engineers. Lake Michigan-Huron sets all-time record for lowest monthly water level [Internet]. 2013 February. [cited 2013 March 15]. Available from: http://www.lre.usace.army.mil/_kd/Items/actions.cfm?action=Show&item_id=8703&destination=ShowItem 101. United States General Accounting Office. Report to Congressional Requesters. Oil and gas; information on shale resources, development, and environmental and public health risks. [Internet]. 2012 September. [cited 2013 March 12]. Available from: http://www.gao.gov/assets/650/647791.pdf
    • 33 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 102. Boyer E, Swistock B, Clark J, Madden M, Rizzo D. The impact of Marcellus gas drilling on rural drinking water supplies [Internet]. Pennsylvania State University. 2011 October. [cited 2013 March 12]. Available from: http://www.rural.palegislature.us/documents/reports/Marcellus_and_drinking_ water_2011_rev.pdf 103. Osborn SG, Vengosh A, Warner NR, Robert B. Methane contamination of drinking water accompanying gas-well drilling and hydraulic fracturing. Proc Natl Acad Sci USA. 2011;108(20):8172–8176. 104. Hayes T. Characterization of Marcellus and Barnett Shale flowback waters and technology development for water reuse. In: Hydraulic fracturing technical workshop #4 USEPA meeting facilities. Arlington (VA). 2011 March 30. 105. Kell S. State oil and gas agency groundwater investigations and their role in advancing regulatory reforms; a two-state review: Ohio and Texas. Ground Water Protection Council. 106. United States General Accounting Office. Report to Congressional Requesters. National wildlife refuges; opportunities to improve the management and oversight of oil and gas activities on federal lands [Internet]. 2003 August. [cited 2013 March 12]. Available from: http://www.gao.gov/assets/240/239441.pdf 107. Slonecker E, Milheim L, Roig-Silva C, Malizia A, Marr D, Fisher G. Landscape consequences of natural gas extraction in Bradford and Washington Counties, Pennsylvania, 2004-2010. United States Department of the Interior. United States Geological Survey. 2012. [cited 2013 March 12]. Available from: http://pubs.usgs.gov/of/2012/1154/of2012-1154.pdf 108. Entrekin S, Evans-White M, Johnson B, Hagenbuch E. Rapid expansion of natural gas development poses a threat to surface waters. Front Ecol Environ. 2011;9(9):503-511. 109. Holland A. Examination of possibly induced seismicity from hydraulic fracturing in the Eola Field, Garvin County Oklahoma. Oklahoma Geological Survey, Open-File Report. 2011. 110. Weinhold B. Energy development linked with earthquakes. Environ Health Persp. 2012;120(10):A388. 111. Seeber L, Armbruster J, Kim W. A fluid-injection-triggered earthquake sequence in Ashtabula, Ohio: implications for Seismogenesis in stable continental regions. B Seismol Soc Am. 2004;94(1):76-87. 112. Fischhoff B, Bostrom A, Quadrel M. Risk perception and communication. Annu Rev Publ Health. 1993;14:183-203. 113. Bier V. On the state of art: risk community to the public. Reliab Eng Syst Safe. 2001;71(2):139-150. 114. Edelstein M. Contaminated communities: the social and psychological impact of residential toxic exposure. Boulder (CO): Westview Press; 1988. 115. Unger D, Wandersman A, Hallman W. Living near a hazardous waste facility: coping with individual and family distress. Am J Orthopsychiat. 1992;62:55-70. 116. Matthies E, Hoger R, Guski R. Living on polluted soil: determinants of stress symptoms. Environ Behav. 2000; 32:270–286. 117. Downey L, Van Willigen M. Environmental stressors: the mental health impacts of living near industrial activity. J Health Soc Behav. 2005;46:289-305. 118. Kroll-Smith JS, Couch S. As if exposure to toxins were not enough: the social and cultural system as a secondary stressor. Envir Health Persp. 1991;95:61-66. 119. States Environmental Protection Agency [Internet]. Environmental justice. [modified 2013 Feb 11; cited 2013 March 1]. Available from: http://www.epa.gov/environmentaljustice/ 120. Albrecht S. Socio-cultural factors and energy resource development in rural areas in the west. In: The Annual Meetings of the Rural Sociological Society. New York, NY. Brigham Young University. 1976 August. 121. Brown R, Geertsen H, Krannich R. Community satisfaction and social integration in a boomtown: a longitudinal analysis. Rural Sociol. 1989;45(4):568-586. 122. Hunter L, Krannich R, Smith M. Rural Migration, rapid growth, and fear of crime. Rural Sociol. 2002;67(1):71-89. 123. Jacquet J. Energy boomtown and natural gas: implications for Marcellus Shale local governments and rural communities. The Northeast Center for Rural Development and Pennsylvania State University. 2009. 124. Jacquet J. Sublette County, Wyoming: a brief history of drilling 1995 – 2005, the socioeconomics of gas. 2006. 125. Brown R, Dorius S, Krannich. The boom-bust-recovery cycle: dynamics of change in community satisfaction and social integration in Delta, Utah. Rural Sociol. 2005;70(1):28-49. 126. Weigle J. Resilience, community, and perceptions of Marcellus Shale development in Pennsylvania wilds: reframing the discussion. Sociol Viewpoints. 2011. 127. Kelsey T. New residents and the demand for public services in growing communities: Local officials’ perceptions and responses. Econ Dev Q. 1998;12: 279-288.
    • 34 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: PUBLIC HEALTH TECHNICAL REPORT, SEPTEMBER 2013 128. Goldberg S, Shoveller J, Ostry A, Koehoorn M. Youth sexual behavior in a boomtown: implications for the control of sexually transmitted infections. Sex Transm Infect. 2008;84:220-223. 129. Freudenburg W, Jones R. Criminal behavior and rapid community growth: examining the evidence. Rural Sociol. 1991;56(4):619-645. 130. Centers for Disease Control and Prevention (US) [Internet]. National environmental public health tracking program. [updated 2012 May 31; cited on 2013 Feb 14]. Available from: http://www.cdc.gov/nceh/tracking/ 131. Centers for Disease Control and Prevention (US) [Internet]. National public health performance standards program. [updated 2010 Dec 9; cited 2013 March 3]. Available from: http://www.cdc.gov/nphpsp/essentialservices.html 132. Michigan Department of Environmental Quality (US). Hydraulic Fracturing in Michigan [Internet]. MDEQ; c.2013 [cited 2013 July 24]. Available from: http://www.michigan.gov/deq/0,4561,7-135-3311_4111_4231-262172--,00.html 133. Encana Corporation. c2013. Responsible products program [Internet]. [cited 2013 March 2]. Available from: http://www.encana.com/environment/water/fracturing/products.html 134. Askcheasapeak.com. Green Frac Program [Internet]. c2013. [cited 2013 March 12]. Available from: http://www.askchesapeake.com/Pages/Green-Frac.aspx 135. Rabe B, Borick, C. Fracking for natural gas: public opinion on state policy options. The Center for Local, State, and Urban Policy. Gerald R. Ford School of Public Policy. University of Michigan. 2011 November. 136. Masuda J, Poland B, Baxter J. Reaching for environmental health justice: Canadian experiences for a comprehensive research, policy and advocacy agenda in health promotion. Health Promot Int. 2010;25(4):453-463. 137. Goldstein B, Kriesky J, Pavliakova B. Missing from the table: role of the environmental public health community in governmental advisory commission related to Marcellus Shale drilling. Envir Health Persp. 2012;120(4):483-486.
    • © 2013 BY THE REGENTS OF THE UNIVERSITY OF MICHIGAN MARK J. BERNSTEIN, ANN ARBOR JULIA DONOVAN DARLOW, ANN ARBOR LAURENCE B. DEITCH, BLOOMFIELD HILLS SHAUNA RYDER DIGGS, GROSSE POINTE DENISE ILITCH, BINGHAM FARMS ANDREA FISCHER NEWMAN, ANN ARBOR ANDREW C. RICHNER, GROSSE POINTE PARK KATHERINE E. WHITE, ANN ARBOR MARY SUE COLEMAN, EX OFFICIO Please print sparingly and recycle
    • Policy/Law Technical Report H Y D R A U L I C F R A C T U R I N G I N T H E S T A T E O F M I C H I G A N
    • Participating University of Michigan Units Graham Sustainability Institute Erb Institute for Global Sustainable Enterprise Risk Science Center University of Michigan Energy Institute ABOUT THIS REPORT This document is one of the seven technical reports com- pleted for the Hydraulic Fracturing in Michigan Integrated Assessment conducted by the University of Michigan. During the initial phase of the project, seven faculty-led and student-staffed teams focused on the following topics: Technology, Geology/ Hydrogeology, Environment/Ecology, Human Health, Policy/ Law, Economics, and Public Perceptions. These reports were prepared to provide a solid foundation of information on the topic for decision makers and stakeholders and to help inform the Integrated Assessment, which will focus on the analysis of policy options. The reports were informed by comments from (but do not necessarily reflect the views of) the Integrated Assessment Steering Committee, expert peer reviewers, and numerous public comments. Upon completion of the peer review process, final decisions regarding the content of the reports were deter- mined by the faculty authors in consultation with the peer review editor. These reports should not be characterized or cited as final products of the Integrated Assessment. The reports cover a broad range of topics related to hydraulic fracturing in Michigan. In some cases, the authors determined that a general discussion of oil and gas development is important to provide a framing for a more specific discussion of hydraulic fracturing. The reports address common hydraulic fracturing (HF) as meaning use of hydraulic fracturing methods regardless of well depth, fluid volume, or orientation of the well (whether vertical, directional, or horizontal). HF has been used in thousands of wells throughout Michigan over the past several decades. Most of those wells have been shallower, vertical wells using approximately 50,000 gallons of water; however, some have been deeper and some have been directional or horizontal wells. The reports also address the relatively newer high volume hydraulic fracturing (HVHF) methods typically used in conjunction with directional or horizontal drilling. An HVHF well is defined by the State of Michigan as one that is intended to use a total of more than 100,000 gallons of hydraulic fracturing fluid. The reports indicate if the text is addressing oil and gas development in general, HF, or HVHF. Finally, material in the technical reports should be understood as providing a thorough hazard identification for hydraulic fracturing, and when appropriate, a prioritization according to likelihood of occurrence. The reports do not provide a scientific risk assessment for aspects of hydraulic fracturing.
    • GRAHAM SUSTAINABILITY INSTITUTE INTEGRATED ASSESSMENT REPORT SERIES VOLUME II, REPORT 6 HYDRAULIC FRACTURING IN THE STATE OF MICHIGAN Technical Report on Policy/Law SEPTEMBER 3, 2013 Faculty Lead SARA GOSMAN LECTURER, LAW SCHOOL Research Assistants SCOTT ROBINSON & SUSIE SHUTTS LAW SCHOOL TABLE OF CONTENTS 2 Executive Summary 2 1.0 Status and Trends 16 2.0 Challenges and Opportunities 18 3.0 Prioritized Pathways for Phase 2 20 Literature Cited THIS PUBLICATION IS A RESULT OF WORK SPONSORED BY THE UNIVERSITY OF MICHIGAN Direct questions to grahaminstitute-ia@umich.edu
    • 2 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: POLICY/LAW TECHNICAL REPORT, SEPTEMBER 2013 EXECUTIVE SUMMARY A s high-volume hydraulic fracturing and public con- cern have grown in the last few years, governments have begun to make policies specifically addressing hydraulic fracturing, and in some cases high-volume hydraulic fracturing. The details of these policies may be presented in informal statements of policy or guidance, or may be made binding in law through legislative action or agency rulemaking. Courts have also been called upon to resolve disputes, creating an additional source of law. This report is designed to inform the integrated assessment by describing the existing policies and laws that apply to oil and gas wells utilizing high-volume hydraulic frac- turing in Michigan, explaining trends in policy approaches across the United States, and identifying challenges and opportunities inherent in policymaking in this area. The report then offers five pathways for interdisciplinary assessment of policy options. The report takes a broad view, examining high-volume hydraulically fractured wells in light of the overall policy and legal framework governing oil and gas wells. This is for two reasons. It is import- ant to understand the ways in which the general framework, when applied to high-volume hydraulically fractured wells, addresses the increased scale of the wells. In addition, the policies and laws specific to high-volume hydraulic fracturing supplement the gen- eral framework, and thus cannot be fully understood without that context. The first section of the report discusses the status of the policies and laws that govern high-volume hydraulically fractured wells in Michigan, and then examines recent policy trends in the United States. The section begins by describing the federal, state, and local policies and laws—both general and specific in nature— that are applicable to a high-volume hydraulically fractured well in Michigan. After presenting the law on mineral ownership and leasing and providing an overview of the primary state permitting program, the section follows the arc of a well: from spacing and pooling; to siting and natural resource use; to well construction; to high-volume hydraulic fracturing and chemical use; to production and plugging; and finally to waste management. The section then describes policy trends at the federal, regional, state, and local government levels. Five areas of state policymaking are examined in further detail—public participation, land use, water withdrawals, chemical disclosure, and special protections for landowners— together with Michigan’s current and proposed policies. The second section of the report identifies three primary axes that must be considered in making policy on high-volume hydraulic fracturing: the level of government, the policy or legal actor, and the policy approach. The section also identifies the underlying challenge of uncertainty—both about risk and the direction of policy and law. To inform the analysis of policy options, the sec- tion presents the opportunities and challenges of the points along each axis. Based on the above sections, the third section offers five priori- tized pathways for an analysis of policy options. The first pathway— and the one the authors suggest should be the highest priority—is an analysis of the policy options for addressing public concern about chemical use. The second pathway is an analysis of the pol- icy options for public participation in governmental decisions on high-volume hydraulic fracturing. The third pathway is an analysis of the policy options for addressing contamination, both prospec- tively and retrospectively. The fourth pathway is an analysis of the policy options for water withdrawals and potential impacts. And the fifth pathway is an analysis of the policy options for land use, both at the state and local level. The authors would like to thank the five anonymous peer reviewers for their very helpful comments. STATUS AND TRENDS T his section of the report presents the current sta- tus of the federal, state, and local policies and laws governing high-volume hydraulically fractured oil and gas wells in Michigan, and the trends at each level. Historically, the policy and legal framework did not distin- guish between hydraulically fractured wells and non-hydraulically fractured wells, other than to require reporting of the completion method. Nor did the framework distinguish between high-volume hydraulically fractured wells and low-volume hydraulically fractured wells. But as high-volume hydraulic fracturing has increased, gov- ernments have begun to adopt specific policies and laws.1 1.1 Status of Policies and Laws The legal system treats oil and gas development as a unique activity in many ways, treatment that begins with the nature of the property right in oil and gas. This part of the section begins with the scope of the right and the way in which the right is conveyed through leasing. The section then introduces the primary regula- tory program, the state oil and gas well permitting program, and the 2011 state policy on high-volume hydraulically fractured wells. Building on this foundation, the section describes two sets of laws that determine the location of wells: the state conservation mea- sures designed to efficiently “conserve” or drain the reservoir; and the federal, state, and local measures that seek to protect natural resources and other land uses. From there, the section reviews the policies and laws governing the life of the well itself: state
    • 3 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: POLICY/LAW TECHNICAL REPORT, SEPTEMBER 2013 well construction standards, the bulwark of the state regulatory program; state review and reporting requirements on high-volume hydraulic fracturing; and state production reporting requirements and well plugging standards. This part ends with a discussion of the federal and state laws governing the back end of the life cycle, wastewater management. 1.1.1 Mineral Rights and Leasing A private or public landowner traditionally owns both the “surface” of the land and the “mineral interest” in the oil and gas beneath the land.2 Mineral rights may be severed from the surface, however, resulting in what is known at law as a “split estate.” In this case, the surface owner is not the same as the mineral interest owner. When there is a split estate, the mineral interest is the “dominant” inter- est.3 The mineral interest owner has the right to reasonably use the surface to extract the oil and gas underneath, even if the surface owner is opposed.4 Under the Dormant Minerals Act, any oil or gas interest not developed for more than twenty years will be deemed abandoned and will revert to the surface owner, unless the mineral interest owner files a notice of intent to retain the rights.5 In Michigan, the state Department of Natural Resources (DNR) is the largest owner of mineral interests. The state owns both the surface and the mineral interests of just over 4 million acres of state land, and also holds in trust mineral interests under 25 mil- lion acres of Great Lakes bottomland.6 In addition, the state owns mineral interests below 2.3 million acres of private lands.7 These private-public split estates are primarily the result of a historical artifact: the state became the owner of the lands during the Great Depression, and reserved the mineral interest when it later sold the property.8 In 1998, Michigan enacted a law that allows surface own- ers to petition the DNR to buy the underlying mineral interest at fair market value.9 The DNR is prohibited from selling the interest when there is already oil and gas production, the interest is leased or permitted for production, or there are unusual or sensitive envi- ronmental features.10 Unlike other natural resources such as coal or timber, the right in oil and gas is limited. The mineral interest owner has the reasonable opportunity to extract oil and gas, not a right to the specific oil or gas underneath the property. According to the “rule of capture,” recognized by Michigan and other states, a person who “captures” oil or gas is entitled to the extracted minerals even if the well drains oil or gas from an adjoining property.11 Confronted with drainage, the adjacent owner’s “remedy under such circumstances is that of self-help: ‘go and do likewise.’”12 This rule is balanced by the “fair share” principle, which restricts the rights of those who access a common reservoir. An owner has a duty not to act negligently in extracting gas or oil, or to act in a way that injures the reservoir itself.13 As discussed below, the rule of capture is also softened by state conservation measures that promote efficient extraction by limiting the placement and production of wells. The owner of oil and gas rights generally leases those rights to an oil and gas exploration and production company with the expertise to drill wells and manage production.14 Leases are negotiated between the owner and a “landman” who represents the company. Some of the standard provisions in a mineral lease include a description of the property interest being conveyed; the primary term in which the operator must begin specified activities or potentially lose the lease; a royalty to the mineral owner on any produced minerals; and a bonus, paid to a mineral owner when a lease is signed.15 While the lessee has the right to reasonable use of the surface, the lease may include a provision on compensation for damage to crops or trees.16 Unlike some other states, Michigan does not require lessees to pay for surface damages by statute. Most leases also contain a provision giving lessees the right to consolidate leased premises with adjoining leased tracts to form a “pool” or “unit,” which may be necessary to meet the minimum acreage requirement for a well permit under the state regulatory program.17 The DNR has its own leasing program for state mineral interests, whether below state or private lands. As the owner of the interests held for the benefit of the public, the DNR faces a careful balanc- ing act: it seeks to maximize revenue and ensure that the oil and gas is not being drained by wells on adjacent properties, and at the same time to protect the environmental, archaeological, and historical features on the surface.18 The state primarily leases its interests through a public auction held twice a year; more rarely, the state enters into direct leases.19 Before an auction, the state solicits nominations of specific parcels from oil and gas companies. The state then reviews the surface features of the properties and classifies them into four categories: leasable surface development, leasable surface development with restrictions, leasable non-sur- face development, and non-leasable.20 The private surface owners and the broader public are given thirty days to comment on the proposed tracts, and the state then makes the final decision about which interests to auction.21 An oil and gas company may conduct exploration activities, such as seismic surveys, after it obtains a lease or through a separate agreement with the mineral owner. The DNR generally requires a company to enter into a state lease in order to survey state-owned minerals.22 Under the terms of the general state lease, the company must request permission from the DNR for surveys on state land.23 When there is a split estate, Michigan courts have not addressed whether a mineral owner has the right to use the surface to con- duct seismic surveys. Other jurisdictions, however, have decided in favor of the mineral owner.24 Neither state law nor the general
    • 4 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: POLICY/LAW TECHNICAL REPORT, SEPTEMBER 2013 state lease requires a company to notify a private surface owner of exploration activities. A review of the general state lease and a form lease used by pro- ducers in Michigan reveals significant differences in their terms. For example, companies must pay the state one-sixth of the gross proceeds from the sale of the extracted oil and gas, while the producer lease offers private landowners one-eighth of the pro- ceeds.25 Companies must also pay a rental or “delay” payment to the state if no minerals are produced and no royalty payments are being made, while the producer lease offers private landowners a “paid-up” lease with no rental payments.26 Companies must pro- vide a lease performance bond of $10,000 to $50,000 to the state and pay for all damages or losses to the surface, while the producer lease offers private landowners compensation for certain damages, such as to crops.27 Finally, the state lease requires companies to attempt to negotiate a surface use and damage agreement, while, not surprisingly, there is no similar provision for different surface owners in the producer lease.28 The final terms in a lease are dependent on the bargaining power and knowledge of the parties. Because of the state’s greater bar- gaining power, the final terms in a state lease obtained through auction will be the same as those offered by the state.29 In contrast, the final terms in a private lease may vary significantly from those offered by the producer.30 When there is a rush to lease and the owner has more bargaining power, one would expect the final terms to be more favorable to the owner; in contrast, when leasing activity is slow and owners have little bargaining power, one would expect the final terms to reflect the terms offered by the producer. At the same time, during periods of slow activity, knowledgeable owners may negotiate more environmental protections because the financial benefit of leasing is smaller.31 A review of recorded leases in Michigan would be necessary to determine the specific differences in leases. Disputes about leases are resolved through judge-made common law rather than by statute or regulation. To protect the owner of the mineral interests and encourage production, courts presume that the lessee has agreed to several “implied covenants” not present in the lease, such as the covenant to prevent drainage and the covenant to reasonably develop the minerals.32 The Michigan Legislature has stepped in, however, to require lessees who drill gas wells to provide monthly revenue statements to lessors and to prohibit the deduction of certain costs from royalties unless there is a specific provision in the lease.33 Since 2010, when the Petoskey Pioneer well spurred interest in deep, directionally drilled, high-volume hydraulically fractured wells, nineteen such wells are known to have been completed.34 Of these, nine are on state land and were developed pursuant to state mineral leases, eight are on private land and were developed pursuant to private leases, and two are on private land and were developed pursuant to both state and private leases.35 A review of recorded leases would be necessary to determine whether any of the private leases are for severed mineral interests. While the increase in leasing due to high-volume hydraulic frac- turing has received some attention, the nature of property rights and the means of conveyance are rarely discussed in the debate surrounding environmental impacts. But this private law plays an important role in driving development. The limited property inter- est in oil and gas incentivizes oil and gas leasing and extraction, as does the separate interests in a split estate. To the extent that high-volume hydraulic fracturing opens up areas in Michigan that have not historically seen oil and gas development, the legal framework encourages mineral interest owners to sign leases to gain the economic benefits and avoid drainage, and companies to lease and develop in the least-cost way. Incentives to lease played out in the 2009-2010 rush to develop the Utica/Collingwood play in Michigan. The promising initial pro- duction of a high-volume hydraulically fractured well in Missaukee County drove record lease sales at the state auction in May 2010. The state received $178 million in bonus payments, almost as much as the state had collected in the previous eighty-one years of auc- tions.36 Meanwhile, many private owners signed leases, some with very favorable terms. At the height of the leasing rush, for example, owners negotiated leases with terms such as a bonus of $3,000 per acre, a 25% royalty on gross proceeds, and greater protections for the surface.37 Since 2010, however, leasing activity has slowed due to disap- pointing production and low natural gas prices. This boom and bust in leasing resulted in allegations of anti-trust violations and extensive litigation. Chesapeake Energy Corporation and Encana Corporation, the two companies most interested in the Utica/ Collingwood, are being investigated by the U.S. Department of Justice and the state Attorney General for alleged collusion to divide bidding and reduce the cost of leasing.38 A private mineral owner sued the companies in February 2013, alleging similar col- lusion on a 2010 sale.39 Private owners have also sued Chesapeake Energy for its decision not to go forward with many of its leases because of a mortgage on the property; these cases have been amicably resolved.40 1.1.2 The State Oil and Gas Well Permitting Program In Michigan, Part 615 of the state Natural Resources and Environmental Protection Act (NREPA) governs oil and gas explo- ration, development, and production.41 The statute gives broad
    • 5 HYDRAULIC FRACTURING IN MICHIGAN INTEGRATED ASSESSMENT: POLICY/LAW TECHNICAL REPORT, SEPTEMBER 2013 authority to the “Supervisor of Wells,” the director of the Michigan Department of Environmental Quality (DEQ), to prevent waste.42 “Waste” includes physical loss of oil and gas through inefficient extraction and surface escape, damage to the environment and public health caused by underground or surface operations, and excess market supply caused by overproduction.43 Using this stat- utory authority, the DEQ issues specific rules and guidance, sets permitting conditions, and enforces requirements on the location, construction, completion, operation, and plugging and abandon- ment of oil and gas wells, including high-volume hydraulically frac- tured wells.44 Once an oil and gas company ob