for Forestry Spring 2013
Better Forest Damage Assessment
ArcGIS Saves Time, Improves Accuracy
By Barbara Shields, Esri Writer
Thunderstorms, powerful winds, and deadly tornadoes tore through
39 counties in Alabama in April 2011. The city of Tuscaloosa was flat-
tened. Across the state, people were killed and homes and businesses
were destroyed. Forests suffered loss as well. The Alabama Forestry
Commission (AFC) used ArcGIS to assess forest resource damage.
AFC estimated that 26,733 acres of forestland were impacted by the
April 15 tornadoes, with an assessed value of $37,728,175. The April 27
tornadoes were estimated to have damaged 177,857 acres, with an
Alabama forests were damaged by tornadoes in April 2011.
assessed value of $228,360,576. Impact to the forestland in Alabama
from the April storms totaled 204,590 acres, with an assessed value of
“The ability to use GIS technologies to integrate cadastral, remotely
sensed, and observational data greatly reduced the time frame neces-
sary for developing the assessment information,” said Patrick A. Glass,
assistant state forester, AFC. “Geoprocessing models developed in
continued on page 3
response to the April tornadoes allowed for unprece-
dented speed and precision in producing results that are
portable and scalable to any significant geographic area.”
AFC collected GPS Exchange Format (GPX) files
on the tornado paths and converted them to feature
classes for GIS analysis to meet agency objectives. AFC
GIS specialist Abi Dhakal was tasked to create maps for
analyzing and assessing the total forest damage by acre,
by county, and for the entire state. Dhakal created a GIS
model that significantly reduced the time to produce
these geospatial products.
Manually geoprocessing data for 39 counties individu-
ally and generating maps for tornado damage analysis
would have taken weeks, if not months, to complete.
Instead, Dhakal built a GIS model that included batch
processing, which allowed him to quickly and efficiently
render processes on multiple datasets. He easily ac-
cessed ArcGIS tools—such as Select, Intersect, Buffer,
Clip, and Tabulate Area—in the geoprocessing models.
Dhakal designed more than a dozen geoprocesses
and steps to produce results. By slightly tweaking the
models, he could adapt subtle changes in analysis
objectives. Initially, it took him a few days to develop the GIS model,
but upon completion, the model ran data for multiple counties
through geoprocesses and created damage analysis maps in just a few
Next, Dhakal and Glass used Forest Inventory & Analysis (FIA) data
outside the model to calculate the volume of timber loss. Finally, they
applied Timber-Mart South standing timber prices to the volume loss
and calculated tornado damage to the forests in US dollar amounts.
The results of the analysis were used by Alabama’s congressional
delegation to document a supplemental funding request for the
USDA Farm Service Agency’s Emergency Forest Restoration Program.
Results were also used by AFC, Alabama Forestry Recovery Task
Force, Alabama Cooperative Extension Service, and other agencies to
manage recovery of the forest resources.
For more information about the
forest damage assessment model,
contact Abi Dhakal at
To assess forest types impacted by two tornadoes, a model used tornado
path polygons derived from aerial reconnaissance, the National Land Cover
Dataset (NLCD), and parcel data. Data from Esri, AFC, NOSS NWS, and the
US Geological Survey (USGS) was also used.
The volume of timber damage was calculated for
each of the 39 Alabama counties impacted by the
tornadoes. Data is from AFC, Esri, NLCD, and USGS.
3Spring 2013 esri.com/forestry
Better Forest Damage Assessment continued from cover
Clinton Climate Initiative Uses
GIS to Preserve and Regrow Forests
security, viability of coastal cities, and water
availability around the world.
To reduce emissions, governments and
economies must use less fossil fuels and
increase the use of energy-efficient and
renewable technologies. The CCI Forestry
Program focuses on helping developing
countries reverse deforestation and plant new
trees. By showing that they can monitor and
verify the reduction of their carbon dioxide
emissions, countries become eligible for fund-
ing to manage their forest programs and other
low-carbon economic activities.
CCI uses GIS technology as a centerpiece
of forest carbon measurement, reporting,
and verification (MRV) systems for develop-
ing countries. GIS is one of three legs of the
platform—data, models, and GIS—that allow
CCI used GIS to measure carbon
and compute carbon credits that were
used to preserve this forest in Guyana.
4 Esri News for Forestry Spring 2013
The Clinton Climate Initiative (CCI) Forestry
Program develops forestry projects and
carbon measurement systems that help
governments and local communities receive
compensation for preserving and regrowing
forests. CCI uses GIS technology to help
countries monitor their carbon levels.
Global warming is caused by increased
carbon dioxide in the atmosphere from burn-
ing fossil fuels, and deforestation accounts
for about 15 percent of total carbon dioxide
emissions in the world. Scientists predict that
if governments and communities don’t take
action to reduce carbon dioxide emissions,
our world will face increasingly drastic con-
sequences ranging from stronger heat waves
to more droughts and floods to increasing
sea level. All these affect agriculture, food
countries to determine how much carbon they
have, how it is changing, and how the drivers
of deforestation and forest degradation can
be monitored and adjusted as required.
With GIS in place for forestry, developing
countries can be eligible for direct payments
through international agreements based on
the effectiveness of their MRV systems. Once
in place, GIS can be used more broadly for
other resource development, land surveys,
and the determination of land tenure.
For example, tree farming projects in Kenya
help make forest conservation and restoration
profitable for local communities. CCI is cur-
rently working on 10 sustainable-forest man-
agement projects, encompassing 644,000
hectares of land, that will benefit more than
353,000 people in forest-dependent commu-
nities around the world.
Esri awarded CCI the 2012 Special
Achievement in GIS Award for helping the
country of Guyana become eligible for
$70 million in forest-based payments from the
government of Norway. Guyana is now using
this funding to facilitate specific elements of
a low-carbon development plan envisioned
and put in place by former Guyana president
Bharrat Jagdeo. This project is part of the CCI
Forestry Program and has been supported by
the Rockefeller Foundation and the govern-
ments of Norway and Australia.
Read more about CCI’s
forestry projects at
The Clinton Climate Initiative supports tree farming projects
in Kenya. This Kenyan tree farmer is working in a nursery to
conserve a local forest.
Quickly and Easily Use
Landsat Data with ArcGIS Online
Landsat imagery is one of several remote sensor systems land analysts use to study vegetation,
land use, soil conditions, and terrain. ArcGIS Online image services provide quick and simple
access to free United States Department of the Interior’s US Geological Survey (USGS) Landsat
data. These image services are based on the Global Land Survey (GLS) datasets created by
USGS and the National Aeronautics and Space Administration (NASA). Landsat data supports
global assessments of land cover, land cover change, and ecosystem dynamics such as
disturbance and vegetation health.
Landsat represents the world’s longest continuously acquired collection of space-based,
moderate-resolution, land remote-sensing, data change research. It is more efficient than any
other technology used to meet decision support requirements.
Esri provides Landsat image services that enable users to explore important imagery
information such as band combination analysis for natural color, infrared, vegetative change,
and Normalized Difference Vegetation Index (NDVI). The user can also compare many years of
imagery to see changes over time.
Two Landsat viewers are available online to help users do just that—LandsatLook Viewer
from USGS and ChangeMatters Viewer from Esri. Both of these viewers use ArcGIS technology
to serve Landsat as web services, accessing many terabytes of imagery. LandsatLook Viewer
provides access to the complete archive of USGS Landsat scenes going back to 1972.
ChangeMatters Viewer provides access to the best cloud-free Landsat scenes from the five
epochs, circa 1975, 1990, 2000, 2005, and 2010. It uses ArcGIS as its underlying technology to
additionally process the Landsat GLS dataset on the fly into multiple data products. These are
served as World Landsat Services on ArcGIS Online. The image service platform is also host to
the Landsat community, a group that shares maps and applications using Landsat imagery.
This map is a mosaicked composite of data
from the image service of the Landsat Global
Land Survey (GLS) 2010 dataset and can be
accessed on ArcGIS Online. This data has
been enhanced with radiometric correction
and histogram stretching to make it more
5Spring 2013 esri.com/forestry
Researchers Develop an Effective Approach to
Forest Cover Analysis
By Aaron E. Maxwell, Remote Sensing Analyst, Natural Resource Analysis Center, West Virginia University
Overwatch. The outcome was a statewide
forest cover and forest fragmentation map
created from raster data at a nine-meter cell
size for the state of West Virginia.
The project goal was to capture the
spectral, textural, and land-use variability
with defined classes. WVU researchers relied
on object-based image analysis, which uses
spectral and textural information within an
image to extract thematic data. The pro-
ject’s imagery was from the United States
Department of Agriculture (USDA) Farm
Service Agency. This was four-band, leaf-on,
one-meter pixel size, uncompressed imagery
State resource agencies model wildlife habitat
to support the planning and management of
natural resources. The most widely available
land cover for states to use is the Landsat-
derived National Land Cover Dataset (NLCD).
Because of its coarse resolution (30 meters)
and temporal lag to the current conditions,
using the NLCD is a challenge for wildlife
modeling at a local scale. Researchers from
West Virginia University (WVU) created a more
effective approach for modeling habitat. They
implemented a GIS and remote-sensing meth-
odology for creating statewide forest cover
and forest fragmentation data layers.
Figure 1. This training data was derived from interpreting photo imagery as polygons in ArcGIS. This base imagery is 2011 NAIP
orthophotography displayed in true color.
The approach was to use publicly avail-
able orthophotography from the National
Agriculture Imagery Program (NAIP). This pro-
vided better resolution for raster data layers,
increased accuracy of forest cover analysis,
and delineation of fragmented wildlife habi-
tat. NAIP orthophotography has a one-meter
cell size and four-band spectral information
(true color bands and an infrared band). The
photography is captured on a two-year cycle,
so datasets can be regularly updated.
The analyst used Esri’s ArcGIS 10 Desktop
image analysis tools and tools from the
Feature Analyst 5 extension for ArcGIS by
6 Esri News for Forestry Spring 2013
representing forest conditions during the 2011
The bulk of the researchers’ time was spent
extracting cover from each image and classify-
ing it as either forested/woody, grasslands/
herbaceous, or barren/nonvegetated cover.
It was necessary to collect a large number
of samples to accurately extract the cover of
interest. Researchers spent from one to three
hours creating training data, which contains
examples of the cover types of interest. They
manually interpreted photographs within the
ArcGIS editor, using its functionality to create
training data as vector and polygon features.
This training data process allowed user input
of vector data and generated examples of the
cover of interest. Figure 1 shows an example
of the training data digitized throughout the
To extract cover, researchers used Feature
Analyst 5 to process each image. The integra-
tion of its extraction tools with ModelBuilder
enabled users to complete mapping tasks in
a timely manner. They visually inspected all
outputs for accuracy and, when necessary,
Having completed this mapping task, re-
searchers merged the resultant raster data to
produce a statewide grid at a nine-meter cell
size. They did this by using the Mosaic To New
Raster tool in the ArcGIS Data Management
toolbox. The outcome was a statewide forest
cover map representing the 2011 growing-
season conditions at a higher resolution than
is currently available from existing datasets,
such as the NLCD. Figure 2 shows the result-
To assess the accuracy of the forest cover
data, researchers compared the cover extrac-
tion to manual photograph interpretation at
randomly selected point locations across the
state. They streamed the 2011 NAIP ortho-
photography through ArcGIS for Server, which
was hosted by the Aerial Photography Field
Office (APFO) of the USDA. This allowed them
to quickly and easily access photography and
assess the accuracy of the resultant cover.
This approach yielded a forest cover map with
accuracy that was greater than 90 percent.
Forest fragmentation was created as a deriv-
ative of the raster forest cover. First, researchers
smoothed the resultant cover to provide a
more general representation of fragmentation
and to remove small canopy interruptions that
were deemed too small to fragment the forest.
To create the forest fragmentation data, they
employed morphological image analysis, ap-
plying mathematical morphology to analyze the
shape and form of objects.
Running the downloadable Landscape
Fragmentation Tool (LFT) version 2.0 in ArcGIS,
researchers mapped the types of fragmenta-
tion present in specified land cover types,
including patch, edge, perforated, and core,
which is based on a specified edge width.
To complete the processing, researchers
segmented the state into manageable units
and then merged the final results, ultimately
producing a raster grid of statewide forest
fragmentation with a nine-meter cell size.
Figure 3 shows the fragmentation data.
Aaron E. Maxwell, Natural Resource Analysis Center, West Virginia University, Morgantown,
West Virginia; Michael P. Strager, Division of Forestry and Natural Resources, West Virginia
University, Morgantown, West Virginia; Elise M. Austin, Natural Resource Analysis Center,
West Virginia University, Morgantown, West Virginia
Figure 2. Researchers created thematic tree cover for West Virginia.
They derived this coverage from 2011 NAIP orthophotography by
using object-based image analysis tools in the Feature Analyst 5
extension for ArcGIS.
Figure 3: Forest researchers were able to derive this forest
fragmentation data for West Virginia by using a script within
ArcToolbox to extract forest cover thematic data layers.
7Spring 2013 esri.com/forestry
Forest managers can now use lidar data in combination with GIS to
help them assess forest inventories and create forest plans. The USDA
Forest Service’s (USFS) FUSION software, combined with Esri’s ArcGIS,
provides lidar analytic tools, a streamlined workflow, and functionality
for storing, organizing, and sharing lidar .las files. Foresters can under-
stand, explore, and analyze lidar data point clouds and interactively
view them in 3D.
Foresters use FUSION to quantify vegetation by extracting lidar
point clouds and correlating them with forest inventory plots. It calcu-
lates various canopy metrics, such as height statistics, to describe the
canopy distribution and cover density ratios. FUSION then summarizes
these ratios at the plot level or in a continuous grid cell format. It also
performs a quality routine to assess the appropriateness of lidar point
data for a forestry application.
FUSION provides a robust ability to perform extensive point cloud
analytics of forest inventory variables across large landscapes. The in-
formation it extracts to describe the forest canopy is easily exported as
an ASCII grid and imported into ArcGIS along with other GIS datasets
USDA Forest Service FUSION Offers
Powerful Lidar Tools
Figure 1. The user exported FUSION-derived
lidar canopy metrics into ArcGIS to produce this
map showing the first return percent canopy cover
grid metric, produced at a 25-meter cell size, across
approximately 85,000 acres in southeast Arizona.
Figure 2. The GIS
grid layer (25 m cell
the Basal Area
model applied at
the landscape level.
This GIS layer is
one of the end-user
products that will
be used for future
monitoring for the
Pinaleño Sky Island
8 Esri News for Forestry Spring 2013
Using ArcGIS Spatial Analyst, users apply conditional logic to the
FUSION-derived canopy structure grids to extract pixels meeting
certain criteria. For example, foresters can locate areas containing tall
trees with relatively low canopy cover.
Over the past several years, a workflow incorporating ArcGIS and
FUSION was implemented for a forest restoration effort in the Pinaleño
Mountains of the Coronado National Forest in southeastern Arizona.
To model forest inventory parameters, the team used regression
analysis to determine the correlation between the parameters that
were measured on field plots and the lidar canopy metrics that were
summarized in FUSION and classified as subsets for each plot.
Before applying the resultant statistical models to the landscape,
team members used GIS to ensure the models were applied appropri-
ately and successfully across the landscape. First, they applied a forest/
nonforest mask to ensure the models were applied only in forested
areas. This was accomplished using the Spatial Analyst Conditional
tool in ArcGIS and the canopy height and cover structure grid layers
output from FUSION. Each pixel had to meet a minimum vegetation
height of three meters and 2 percent canopy cover. All pixels that did
not meet the criteria were masked out before models were applied
across the study area.
The final step was to apply the regression models created in the
initial modeling steps to the appropriate ASCII grid layers. This
generated continuous inventory parameter GIS layers covering the
entire study area. Each calculation produced a new grid in which
each 25-meter cell spatially represents the estimated forest inventory
parameter of interest such as biomass, basal area, Lorey’s mean height,
and timber volume (figure 2). The resultant GIS inventory layers were
qualitatively validated by local experts and conformed well to trends
known to occur on the landscape.
The project team estimated that the cost for obtaining data suf-
ficient to implement the Pinaleño Ecosystem Restoration Project using
lidar was half the cost of traditional methods. Furthermore, capturing
the lidar data to measure sample areas was easier than deploying field
crews that could not safely measure trees in extreme terrain. Lidar
made it possible for the team to create continuous coverage of all
FUSION software is currently managed by Robert McGaughey of the
USDA Forest Service—Pacific Northwest Research Station. Download
USFS FUSION software free of charge at http://forsys.cfr.washington
For more information, tutorials, and
sample lidar datasets, visit the USFS
Remote Sensing Applications Center at
Learn more about FUSION by contacting
Ron Behrendt, managing member,
Behron LLC, at email@example.com.
9Spring 2013 esri.com/forestry
Create a Map in Seven Steps
ArcGIS Online is the mapping platform for your organization. CEOs,
staff, and contractors can access maps and data for their work. See
opportunities and gain insight into your data. Do this quickly with no
data to install.
Create an ArcGIS Online forest map in seven steps at arcgis.com.
Click Sign In and note the ribbon on the top of the page. Click the word
ArcGIS to get to the ArcGIS Online page.
Let’s get started.
1. Open a map: Click Make a Map, and you see a map of the
2. Create the basemap: Click Basemap and select Light Gray Canvas.
3. Add a data layer: Click the Add button on the ribbon; select Search
for layers; and in the Find field, type “USFS Ecological Subregions”
(the In field should say “ArcGIS Online”) and click Go. When the layer
appears, click Add beside the layer title.
4. To add another layer package, go to the Find field and type “USGS
Forest Fragmentation”. Add it to your map.
5. At the bottom of the Search for layers to add section, click the Done
Adding Layers button.
6. See your layers in the table of contents. Point to a layer title and click
the arrow to its right. Play with the zoom, transparency, and visibility
tools and set them to your preference.
7. Click Save to save your map in your account folder.
Explore your map. Click the layer title to see the map’s data layer
contents and select what you want to see. Above the Contents line,
pause your mouse pointer over an icon to see its function. Click Show
Map Legend to open it and click again to close it. Click the data layer
to select subsets on the map. Zoom to see greater detail. Click on the
map to get specific information. Click the Details button on the ribbon
to turn the contents column on and off.
ArcGIS Online is a service that lets you use GIS software and data.
Create maps that tell your story, build applications, and more.
Sign up for a 30-day free trial subscription
The USDA Forest Service published this map of ecological subregions.
This map shows forest fragmentation risk.
By combining two layers, the reader can see forest fragmentation
risk by ecological regions. The result has been saved and published for
10 Esri News for Forestry Spring 2013
Join forestry professionals from around the world at the Esri Forestry
GIS Conference, which is hosted by the Esri Forestry Group. You will
meet with a community of GIS for forestry users, share ideas, hear best
practices, and talk to experts.
This year’s theme is Benefiting from Change: Realizing the Value of
GIS in Forest Management. The conference offers a hands-on work-
shop for you to develop your forestry GIS skills. You can also attend
sessions that explore these topics:
•• Forest spatial optimization
•• Integrating data and systems
•• Automated workflows
•• A complete desktop, web, and mobile system
•• Image management, classification, and analysis
•• Web applications and services for many user types
Esri Forestry GIS Conference
May 14–16, 2013 | Esri Headquarters, Redlands, California
For more information and to register,
Mark Your Calendar
Western Forestry Leadership Coalition (WFLC)
April 29–May 1, 2013
Denver, Colorado, USA
Esri Forestry GIS Conference
May 14–16, 2013
Redlands, California, USA
2013 Southern Group of State Foresters (SGSF)
June 3–6, 2013
Savannah, Georgia, USA
June 5–8, 2013
The Association of Consulting Foresters
June 21–25, 2013
Keystone, Colorado, USA
Esri International User Conference
July 8–12, 2013
San Diego, California, USA
Remsoft Modeling Conference and User Group
September 9–13, 2013
Fredericton, New Brunswick, Canada
National Association of State Foresters
September 22–26, 2013
Hot Springs, Virginia, USA
Society of American Foresters National Convention
October 23–27, 2013
Charleston, South Carolina, USA
Western Forestry Leadership Coalition (WFLC)
October 28–30, 2013
San Diego, California, USA
Southern Forestry and Natural Resource
December 8–10, 2013
Athens, Georgia, USA
On the RoadAttend a Conference
Just for You
Esri invites you to join the Esri Forestry Group (EFG). Your participation
in this dynamic group will help you get more from your GIS and your
forest and land management data. Meet like-minded professionals,
share experiences, and exchange knowledge.
Become a Member
•• Community: Connect with a community of professionals
passionate about GIS.
•• Information: Stay current on Esri forestry products, events,
webinars, and resources.
•• Solutions: Learn about forestry GIS tools, applications, and projects.
Join today or renew your membership
Esri Invites You to Join
the Forestry Group
11Spring 2013 esri.com/forestry