Greenhouse gas emissions and nitrogen pollution in U.S. agriculture: An assessment of current emissions, projections, and mitigation strategies A report for the David and Lucile Packard Foundation April 2012
2 In early 2012, California Environmental Associates conducted a literature review and data synthesis on the topics of nitrogen pollution and greenhouse gas emissions stemming from US agriculture. Specifically, the report sought to answer the following questions: What are the main agricultural sources of greenhouse gas and nitrogen pollution in the US? How are those sources changing, and what trajectory are we on in terms of future emissions? What appear to be the most promising mitigation opportunities available for both greenhouse gas emissions and nitrogen pollution, and what can we say about their cost and feasibility? The following write‐up is a summary of the report, which was submitted to the Packard Foundation in April, 2012. The full report is available upon request. Contact: Amy Dickie – firstname.lastname@example.org Introduction Agriculture has reshaped the face of the planet. Over the past few centuries, as the population has grown to over seven billion people, we have converted over a third of the earth’s terrestrial surface into cultivated or grazing land.1 Moreover, in the past half‐century, we have dramatically intensified our use of this land, increasing phosphate fertilizer use two‐fold, nitrogen fertilizer six‐fold, and pesticides use more than eight‐fold.2 This intensification, combined with more irrigated land and better crop varieties, has more than doubled average crop yields,3 but it has also dramatically altered the earth’s natural environment, threatening a range of natural resources. Biodiversity loss, freshwater pollution, air pollution, and aquifer depletion are often accompanied by an expansion or intensification of agriculture. The United States is one of the world’s most important agricultural producers. The US leads the world in beef, corn, and soybean production, and is among the top producers of dairy products, wheat, and poultry.4 Many of the country’s natural ecosystems have been transformed thanks to agricultural expansion over the last few generations. And despite decades of production efficiency improvements and increasing resources dedicated to conservation practices, the US agricultural sector is still a significant contributor to national greenhouse gas emissions and, in most regions of the country, represents the leading contributor to nitrogen pollution. The science around mitigation opportunities for both agricultural greenhouse gas emissions and nitrogen pollution in the US is improving but remains highly uncertain. The complex nature of agricultural systems makes measurement of emissions and mitigation potential difficult, costly, and often unreliable. Further, while the greenhouse gas mitigation potential in US agriculture appears fairly significant – in the range of 300 to 800 Mt CO2e/year – achieving mitigation in this sector has proven difficult because of its diffuse nature, the behavioral changes required, and the reality that conservation practices are not always in the economic best interest of producers. Agricultural nitrogen pollution mitigation practices face the same challenges. 1 Stienfeld et al. 2006. 2 Green et al. 2005. 3 Green et al. 2005 4 FAO, 2010.
3 That said, many agricultural mitigation options, for both greenhouse gases and nitrogen, cost less than opportunities to reduce emissions from other sectors. And while these options are distributed across many individual actors, they are also fairly concentrated geographically and by commodity. Continued efforts by the conservation community, researchers, and industry associations to shift the agricultural sector towards production practices that reduce greenhouse gas and nitrogen pollution on a per unit basis may prove to be a worthwhile investment. Greenhouse gas emissions The U.S agricultural sector emitted 430 Mt CO2e in 20095, approximately 6% of total US greenhouse gas emissions. Agriculture’s share of emissions has been fairly constant for the last few decades. Greenhouse gas emissions from US agriculture have been rising at a modest pace in recent years: less than 1% growth per year from 1990‐2009. Most recent projections for US agricultural emissions over the next few decades forecast a comparable growth rate, even when accounting for increases in biofuel production. At the national level, agricultural emissions are split roughly 60/40 between livestock and crops. The biggest component of livestock emissions is methane released from the digestive function of animals, particularly cattle, a process called enteric fermentation. Enteric fermentation accounts for about 30% of all US agricultural emissions. In addition, emissions from manure and grazed lands each account for another 14% of total emissions. Cropland emissions are almost entirely driven by releases of nitrous oxide, about 33% of total US agricultural emissions, stemming from synthetic fertilizer application and crop biological fixation. Emissions from rice cultivation are insignificant in the US because rice acreage is very small. Soil carbon in cropped and grazed lands can function as either a source or a sink, depending on weather, usage patterns, and management of the land. Soil carbon from croplands is believed to have served as a small net sink in recent years, reducing overall agricultural emissions by approximately 5% per year in the last decade. Manure management has been the only sub‐segment of agricultural emissions that has seen notable growth in recent years, growing approximately 42% from 1990‐2009. The growth in manure emissions is due to a shift from pasture‐based operations to large scale confinements for both dairy and swine. Geographic distribution Dividing national agricultural emissions by region and commodity reveals some noteworthy hot spots. Texas, California, and Iowa lead the country in terms of state agricultural emissions, accounting for 10%, 7%, and 7% of US emissions respectively. Emissions in Texas are mainly tied to enteric fermentation from its large beef cattle population. In Iowa, emissions are split fairly evenly between cropland emissions and emissions from livestock, both from swine and beef cattle. Approximately two thirds of California’s emissions are attributable to the dairy cattle industry, both enteric fermentation and 5 EPA Greenhouse Gas Inventory – Note: does not include carbon fluxes
4 manure, with the remaining third is from nitrous oxide emissions from croplands. Together the Midwestern states account for about 30% of all agricultural greenhouse gas emissions. Figure 1: Agricultural greenhouse gas emissions by state (2008) Livestock emissions Livestock emissions are dominated by cattle – approximately 60% of livestock emissions are from beef cattle and another 25% are from dairy cattle. Of the remaining 15% of livestock emissions, two thirds (10%) are from swine. On a per head basis, dairy cattle have by far the largest emissions because they are more commonly housed in feedlots (vs. grazed systems for beef), and on average are much larger, productive animals (larger animals eat more and therefore create more methane and more manure). Livestock that are kept in confinement have higher manure emissions because manure is often stored on site. The larger the confinement system, the more likely it is to manage manure with a wet system, thus rising the methane emissions of the manure. Conversely, manure that falls onto grazed lands generates very little emissions. Dairy cattle emissions are more than two times that of beef cattle on a per head basis. Beef cattle populations, however, dwarf those of dairy cattle and thus dominate aggregate emissions (see chart below). Interestingly, California’s emissions per head for dairy cattle appear to be much higher than that of other dairy states (e.g. Wisconsin), both because warmer temperatures lead to higher methane emissions from stored manure, and because Wisconsin relies more heavily on pasture systems than feedlot systems. Note also that poultry are left off of the chart
5 because their large population numbers would change the scale. Poultry emissions in the aggregate are approximately 4.5 Mt CO2e per year, less than horses. As expected, their emissions on a per head basis are infinitesimal. Figure 2: Comparative greenhouse gas emissions by animal type (2008) Livestock emissions per head (x‐axis), total head (y‐axis) and total emissions (bubble size). Cropland emissions Approximately 40% of all cropland emissions are attributable to corn, roughly 20% to soybeans, and about 18% to “non major crops” (i.e. specialty crops). Cropland emissions are dominated by corn due both its high fertilizer requirement – corn receives nearly 45% of all nitrogen fertilizer in the US, but only accounts for around 25% of all cropland – and the fact that it accounts for more acres of cropland than any other crop. The dominance of corn leads to a predictable concentration of cropland emissions in the Midwest, though California ranks high as well on a state‐by‐state comparison. 140652360204060801001200 1 2 3 4 5 6Total Head (in Millions)Emissions per Head (Mt CO2e) Beef CattleDairy CattleSwineHorsesSheepGoats
6 Figure 3: Comparative greenhouse gas emissions by crop (2008) Crop emissions per area (x‐axis), total acres planted (y‐axis) and total emissions (bubble size). Greenhouse gas mitigation options The current technical mitigation potential in agriculture is quite large. It probably exceeds the total emissions from the sector thanks to untapped potential to sequester high levels of carbon in agricultural soils. The literature defines a range of approximately 300 to 800 Mt CO2e per year, with the majority of the opportunity in soil carbon sequestration (see Figure 4, below). These numbers may be even higher if the full potential of biochar is considered. Opportunities to reduce emissions from livestock seem to be below 100 Mt CO2e per year, although research and development efforts around diet and feed additives for livestock may able to expand the technical potential in this area. Similarly, the potential to reduce emissions from nitrous oxide through more efficient application of fertilizers is probably in the 100 Mt CO2e per year range. Soil carbon sequestration, both in croplands and grasslands is likely in the 300 to 500 Mt CO2e per year range. Although in the aggregate the mitigation potential of agriculture is quite compelling, progress is slow because of the distributed nature of the emissions and the lack of regulatory levers to work through. 5629202432‐50510152025303540450.0 0.5 1.0 1.5 2.0Millions of HectaresMt CO2e per Millon HectaresCornSoyWheatHayCottonSorghum
7 Figure 4: The mitigation potential of agriculture There are many different conceptual ways to approach agricultural mitigation opportunities (see Figure 5, below). We have identified four overarching approaches: 1) reduce or change consumptions patterns, 2) reduce US agricultural production, 3) shift US agricultural production toward less carbon intensive commodities, and 4) reduce the greenhouse gas footprint of current production systems. Reduce or change consumptions patterns – This approach focuses on reducing the demand for carbon intensive agricultural commodities through efforts to discourage red meat consumption, encourage vegetarianism, or reduce food waste. The challenge with these approaches is that this is an incredibly diffuse problem that largely defies regulation and lacks obvious leverage points. Reduce US agricultural production ‐ The second approach would be a simple attack on the supply side of the equation, reducing overall agricultural commodity production through mechanisms like expanding the Conservation Reserve Program or implementing a production tax. In addition to the challenges of doing this at scale, the main risk involved here is that without a simultaneous reduction in demand, production is likely to shift to other parts of the world, typically to less efficient regions, potentially causing a net rise in agricultural greenhouse gas emissions, particularly when land use change is factored in. Shift production to less greenhouse gas intensive commodities ‐ This approach could be triggered by economic incentives that support an increase in the use of perennials, the conversion of cropland to pastureland, or simply more diverse crop rotations. Such changes 0100200300400500600700800900all measurescropland soil carbongrassland soil carbonLand conversion, soil carbon livestockMt CO2e / yr
8 might lead to a net reduction in emissions, but without a careful assessment of their impact on commodity markets, it is difficult to state conclusively whether such changes would have a positive or negative impact. A negative impact is possible if production of the more greenhouse gas intensive commodities simply move elsewhere. Reduce the greenhouse gas intensity of production – With this approach, we can keep the same production patterns, but reduce the associated emissions by incentivizing practices such as conservation tillage, winter cover crops, nutrient use efficiency, better management of grazed lands, and improved manure management. This approach seems to be the obvious place to focus efforts, and has been the main area of investigation within the scientific literature. While it is not without its challenges, it seems to have the least risk of unintended consequences, and also some of the lowest implementation costs. An extensive literature is emerging around the issue of agricultural greenhouse gas mitigation. Research is split roughly into two basic categories: 1) studies that document the mitigation potential of specific practices in specific locations, typically of a per hectare basis and usually derived from field level studies, and 2) research that applies sectoral economic models to determine the economic potential of different broad categories of practices (e.g., afforestation, conservation management practices, nutrient management, biofuel production), depending on different prices of carbon. The former tend to be difficult to apply widely, and the latter are often too aggregated in their application to assess the nationwide biophysical potential of some individual practices. The recent publication of the Nicholas Institute’s (T‐AGG) “Greenhouse Gas Mitigation Potential of Agricultural Land Management in the United States: A Synthesis of the Literature” provides an extremely useful data set. This report provides mean estimates as well as high and low ranges for the soil carbon sequestration potential, and methane, nitrous oxide, and process and upstream emissions reductions potential on a per hectare basis for 42 mitigation practices, as well as an assessment of the maximum area available for each mitigation practice. This data set is the most comprehensive available in documenting the biophysical potential of a range of practices on a per hectare basis. Unfortunately, to date an economic analysis at a comparable level of detail does not exist. The most promising practices are those that combine the following characteristics: 1) have a high biophysical potential on a per hectare basis, 2) can be widely applicable, and 3) have a low implementation cost. According to this synthesis report the practices with the highest biophysical potential (CO2e per hectare) include the set aside and management of histosol cropland (i.e. protecting organic soils), application of biochar, restoring wetlands, switching to short‐rotation woody crops, and agroforestry. Unfortunately, there is not great overlap between those high intensity opportunities and those that are most widely applicable. The opportunities that are widely applicable include conservation tillage, winter cover crops, and grazing lands management. In the aggregate, the later set of opportunities appear more compelling, though their per hectare potential is lower. There are many important considerations when choosing individual or sets of mitigation practices to promote.
9 Figure 5: Agricultural greenhouse gas emissions mitigation logic model Category Sub‐category Intervention options Risks, limitations & co‐benefits Reduce demand for carbon intensive agricultural commodities • Reduce per capita meat consumption • Reduce % of food waste • Vegetarianism campaign• Food service campaign • Change in expiration date protocols • Solutions difficult to scale• Difficult to develop mandates or incentives Reduce agricultural commodity production • Afforestation • Restoration of wetlands, organic soils • Convert land to set‐asides or buffers • Production tax• Expand CRP • No grazing on fed lands • Stricter CWA regulations • Decrease commodity subsidies • End biofuels subsidies • Pay farmers not to farm • Leakage: Although these measures will reduce emissions regionally or nationally, without a simultaneous shift in demand, production will likely just shift elsewhere, possibly to a less carbon efficient location. Shift production to less greenhouse gas intensive commodities • Use more perennials • Increase production of woody crops, agroforestry • Convert cropland to pastureland • Diversify crop rotation • Subsidize the lowest greenhouse gas crops • Revenue neutral tax on top greenhouse gas agricultural products (e.g. dairy and corn) • May also be a risk of leakage with these interventions. The dynamics of specific changes in production patterns would need to be modeled. Change practices to reduce greenhouse gas intensity of production • Improve productivity and management of grazed lands • Improve productivity and management of croplands (e.g. tillage, cover crops, nutrient use efficiency) • Improve livestock efficiency • Improved manure management • USDA programs • Supply chain pressure • Carbon markets • Other PES markets • Some of the practices in this category may increase intensity (positive leakage effects) and/or have positive environmental co‐benefits. • Some may have negative impacts on other environmental resources (e.g. water, toxics).
10 Opportunity costs ‐ Practices are economically viable if they have low opportunity and low transaction costs. Practices that take land out of production (e.g., set‐asides) and/or significantly change cropping patterns (e.g., agroforestry, short rotation woody crops, perennials) generally have a high opportunity cost and are usually not widely viable unless there is some form of economic compensation, such as a high price for carbon or tax incentive. If the price of carbon gets high enough, these options, and others including afforestation and biofuel production become rational, at least conceptually. Indirect land use change – Although opportunities that take land out of production in the US are compelling in terms of the greenhouse gas mitigation potential they can offer domestically, these options may not be beneficial on a net global greenhouse gas basis. Baker et al. 2011 finds that “climate mitigation opportunities increase the demand for land for nonfood benefits, reduce commodity supply, and result in significant commodity market impacts.” Recent studies from both Iowa State University (Elobeid et al. 2011) and Nicholas Institute (Mosnier et al. 2012) find that taking land out of food production or diminishing yields in the US can lead to a net gain in greenhouse gas emissions on a global basis because the demand for agricultural commodities is fairly inelastic and production moves elsewhere. Mitigation practices that do not change land use or cropping patterns, but rather change the greenhouse gas intensity of production, are almost certainly the options with lowest risk of unintended consequences. Transaction costs - Transaction costs are lowest for those practices that are easily monitored and widely applicable (e.g., tillage, winter cover crops, fallow mgmt.) That said, opportunities that are widely distributed and have a smaller per hectare opportunity may be more difficult, and more costly, to implement. Soil carbon sequestration ‐ The potential to sequester more carbon in agricultural soils, both on cropland and grazed land, provides a greater mitigation opportunity than reducing nitrous oxide or methane emissions. However, soil carbon sequestration has complexities that need to be understood and controlled for when designing agricultural mitigation or offset programs. Soil carbon sequestration is reversible, meaning that a change in practice can release, or begin to release, carbon that has been stored up over several years. Secondly, there is a limit to the amount of carbon that can be stored in our agricultural soils, meaning that over a 30 – 50 year time horizon, soils will become saturated and the annual mitigation opportunity it provides will decline, and eventually disappear entirely. Scientific certainty – There is a high level of uncertainty with a number of the mitigation practices currently being explored. Biochar application is one practice that ranks very high both in terms of its per hectare potential and applicable hectares, however there is still great uncertainty associated with this opportunity. Basic questions persist regarding longevity and mitigation potential, lifecycle concerns and economic factors that require further research. Grazing land management is another set of practices with very high mitigation potential and a great deal of scientific uncertainty. These practices also deserve further research. Additionality ‐ Virtually all agricultural emissions face a common challenge with respect to regulated or voluntary offset markets; determining baselines and additionality. Common practices vary greatly by region, and even by producer. They are also always changing. Determining what constitutes baseline conditions or practices, and, conversely, what level of
11 adoption of mitigation practices are additional, is a serious challenge, and risks keeping some agricultural mitigation opportunities from being included in carbon markets. Nitrogen pollution Agricultural nitrogen is the largest source of new reactive nitrogen annually in the US. Agricultural nitrogen is split approximately 60/40 between synthetic fertilizers and crop biological fixation. After steep growth in the 1960s and 70s, synthetic fertilizer use has leveled off dramatically and is now growing at ~1% per year. This growth rate is not expected to increase materially even with the current biofuels mandate. Corn is the largest user of nitrogen fertilizer in the U.S., accounting for over 40% of use. However, on a per acre basis, some of the specialty crops are bigger nitrogen users. Crop biological fixation is growing at about 2.5 times the rate of the synthetic fertilizers (2.4% and 0.9% per year respectively). Soybeans account for about 40% of nitrogen from crop biological fixation, and are the major crop that has grown the fastest over the last 20 years in terms of planted acreage. We did not study the relative impact on nitrogen between various crop rotations, so cannot say whether the growth in soy acres is a positive or negative trend with respect to nitrogen fluxes and nitrous oxide emissions. Further inquiry is advised. Once nitrogen is applied to fields, its pathway is difficult to track and measure, and varies greatly by region and specific site. Some nitrogen is used by crops and turned into food for humans and livestock as well as crop residues. Other nitrogen is neutralized in the denitrification process. But a significant amount of nitrogen causes air and water pollution. In many parts of the country as much as 20‐30% of applied agricultural nitrogen ends up in aquatic systems. Only ~1% is released as nitrous oxide, but it is such a potent greenhouse gas (~300 times more impact per unit weight than CO2) that these small volumes have very a very big impact. The Mississippi River Basin is one watershed with particularly high fluxes of nitrate into the river system, in part due to the tiling system the drains much of the Midwestern agricultural lands. As much as 59% of the spring nitrate loading in the Mississippi River Basin is due to fertilizer run off.6 Nitrogen mitigation The approaches to nitrogen pollution mitigation largely follow the same logic as those available for agricultural greenhouse gas mitigation: Reduce or change consumptions patterns ‐ Nitrogen pollution can be reduced by shifting human diet to commodities that convert feed to food more efficiently and thus make better use of nitrogen applied to feed. The same challenges to modifying human behavior apply here as with the effort to reduce demand for greenhouse gas intensive food commodities – few obvious points of leverage. 6 Booth et al. 2007.
12 Take “leaky” land out of production ‐ Cropland that is most susceptible to nitrogen loss can be taken out of production. The risk here is that this land may be highly productive, so not a good choice for set‐asides. Shift production to less nitrogen intensive commodities ‐ Cropping patterns could shift to less nitrogen intensive crops. Depending on the level of intervention necessary, this approach could lead to indirect land use change and a rise in net greenhouse gas emissions. However, some studies suggest that just a small amount of perennials integrated into major crops can have a significant impact on nitrogen losses. Reduce the nitrogen intensity of production ‐ Nitrogen losses could be decreased by enhancing the efficiency of fertilizer use. This approach is promising because it does not require a change in production patterns and in many cases should lead to economic gains for the producers. However, it requires a behavioral change which may be difficult to scale quickly. Adoption of conservation practices ‐ Nitrogen losses could be decreased through an increased use of conservation practices that filter nitrogen (e.g., buffer strips, tile bioreactors, wetlands, grassed waterways). These practices can be very effective at reducing nitrogen losses, but are generally not economically advantageous to the producers and require voluntary actions by a large number of actors. Progress to date The last two approaches, increasing nutrient use efficiency through best management practices for fertilizer management and increasing the use of conservation measures that filter nitrogen, are commonly championed by conservation groups, and supported by a number of USDA programs. Efforts to date have achieved some success in reducing nitrogen losses, but there is still plenty of room for further improvement. The USDA’s Conservation Effects Assessment Project (CEAP) studied the impact of voluntary adoption of conservation measures by producers across the US from 2003‐2006 in the Upper Mississippi River Basin, the Great Lakes Region, and the Chesapeake Bay. The CEAP found that nitrogen losses have been reduced by 18‐29% during these years, with result varying by region. It further determined that additional reductions of 27‐41% are possible by increasing the level of treatment, particularly to acres that are the most vulnerable to nitrogen losses. There is likewise a great opportunity to reduce nitrogen losses through increased nutrient use efficiency with respect to fertilizer application. Many producers apply more fertilizer than is necessary for crop growth thanks to a combination of two primary factors: 1) it is a low‐cost way to hedge against the risk of low yields, and 2) the information provided by extension agents and crop advisors is often not tailored to specific sites so producers lack sufficient information to manage fertilizers in an adaptive and precise manner. A recent study by the USDA’s Economic Research Service finds that a majority of acres planted in major commodity crops do not adhere to best management practices, as defined by the USDA, leading to hundreds of thousands of tons of excess nitrogen application, primarily concentrated in the Midwest. The study further found that some of the most vulnerable land, tile drained land in the
13 Midwest, are the least likely to comply with best management practices. In fact, data indicate that over 70% of tiled acres do not meet all three nitrogen management criteria (rate, timing, and method of application), which, together, define best management practices.7 Lessons from Europe The European experience shows that controlling agricultural nitrogen application, through both better management of manures and the reduced use of synthetic fertilizers, can greatly improve air and water quality. These measures have lead to a ~20‐30% reduction in nitrogen losses to the environment in the both Denmark and the Netherlands.8 Though Denmark and the Netherlands have had good success in reducing nitrogen pollution through better control of agricultural nitrogen over a relatively short time period, other research in Europe shows that in large, complex watersheds with deep ground water systems, the response time can be very slow. Historical data from the Thames River in demonstrates that a large step change in nitrogen loading in the years 1940 – 1945 from a dramatic rise in fertilizer use was followed by a large step change in nitrate concentrations in the river in the 1970s, indicating a response time of ~30 years. The reaction time in Denmark and the Netherlands was much faster because neither country has a deep ground water system. Conclusions Years of intensification of agricultural productivity and efficiency, along with the promotion of conservation best practices, have slowed the growth rate of agricultural greenhouse gas emissions and nitrogen pollution over the last decades to almost zero. Still, agriculture remains the leading source of nitrogen pollution in most parts of the country and is a notable contributor to US greenhouse gas emissions. Further mitigation of agricultural greenhouse gas emissions and nitrogen pollution poses major challenges: the opportunities are diffuse, not always economically attractive, generally difficult to measure, and largely unregulated. However, the ability of agricultural soils to store additional carbon makes the sector’s mitigation potential disproportionately significant compared with its share of emissions. And certain practices are already gaining traction (e.g. conservation tillage), providing welcome encouragement to researchers, conservation practitioners, and producers working to lessen the environmental footprint of agriculture. Continued efforts to encourage and support research, experimentation, and market adoption, as well as creative efforts to attack the problem in new ways (e.g. shifting consumption patterns), should prove worthwhile. 7 USDA, 2011. Nitrogen in Agricultural Systems: Implications for Conservation Policy 8 Erisman et al. 2005.
14 Bibliography 1) Baker et al., “Net Farm Income and Land Use under a U.S. Greenhouse Gas Cap and Trade,” Policy Issues (2010). 2) Booth and Campbell, “Spring Nitrate Flux in the Mississippi River Basin: A Landscape Model with Conservation Applications,” Environmental Science and Technology (2007). 3) Davidson et al., “Excess Nitrogen in the U.S. Environment: Trends, Risks, and Solutions,” Issues in Ecology 15 (2011). 4) Eagle et al., “Addressing Greenhouse Gas Mitigation Potential of Agricultural Land Management in the United States: A Synthesis of the Literature,” Nicholas Institute (2012). 5) Elobeid et al., “Greenhouse Gas and Nitrogen Fertilizer Scenarios for U.S. Agriculture and Global Biofuels,” Iowa State University (2011). 6) Environmental Protection Agency, “Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990‐2009,” 2011. 7) Environmental Protection Agency Science Advisory Board, “Reactive Nitrogen in the United States: An Analysis of Inputs, Flows, Consequences, and Management Options,” (2011). 8) Erisman et al., “The Dutch Nitrogen Cascade in the European Perspective,” Science in China (2005). 9) Food and Agriculture Organization of the United Nations – FAOSTAT 10) Green et al., “Farming and the Fate of Wild Nature,” Nature (2005) 307: 550‐555. 11) Howden et al., “Nitrate pollution in intensively farmed regions,” Water Resources Research Vol. 47 (2011). 12) McKinsey & Company, “Greenhouse Gas Abatement Cost Curves,” (2009). 13) Mosnier et al., “The Net Global Effects of Alternative U.S. Biofuel Mandates,” Nicholas Institute, (2012). 14) Murray et al., “Greenhouse Gas Mitigation Potential in U.S. Forestry and Agriculture,” Environmental Protection Agency (2005). 15) NRCS, “Assessment of the Effects of Conservation Practices on Cultivated Cropland in the Upper Mississippi River Basin,” (2010). 16) NRCS, “Assessment of the Effects of Conservation Practices on Cultivated Cropland in the Great Lakes Region,” (2011). 17) NRCS, “Assessment of the Effects of Conservation Practices on Cultivated Cropland in the Chesapeake Bay,” (2011). 18) Paustian et al., “Agriculture’s Role in Greenhouse Gas Mitigation,” Pew Center on Global Climate Change (2006). 19) Smith et al. “Greenhouse Gas Mitigation in Agriculture,” Philosophical Transactions of the Royal Society (2008) 363, 789–813 20) Stienfeld et al., “Livestock’s Long Shadow,” FAO (2006). 21) Tomich, T., T. Rosenstock, D. Liptzin, S. Scow, R. Dahlgren, D. Sumner, S. Brodt, K. Thomas, A. White, C. Bishop. California Nitrogen Assessment. Unpublished data. Agricultural Sustainability Institute, University of California, Davis. 22) USDA, “Nitrogen in Agricultural Systems: Implications for Conservation Policy,” (2011).