《森林生态学》课程教学资源（生态学热点）Issues in ecology - 17 生物燃料 Ecological Dimensions of Biofuels.pdf_大学文库

© The Ecological Society of America • esahq@esa.org esa 1 ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 Ecological Dimensions of Biofuels Clifford S. Duke, Richard V. Pouyat, G. Philip Robertson, and William J. Parton SUMMARY Biofuels, liquid fuels derived from biological materials such as crop plants, forest products, or waste materials, have been widely promoted as a means to reduce dependence of our transportation systems on fossil fuels and to reduce emissions of the greenhouse gases (GHG) that contribute to global warming. The primary forms of biofuels are ethanol, made from sugars, starches, cellulose, and other plant structural components, and biodiesel, made from oils produced by plants. Many countries, including the United States and members of the European Union, have adopted production and use targets for biofuels. The promise of biofuels as a renewable, environmentally friendly energy source, combined with these mandates, has driven a worldwide expansion in their production. However, many questions remain about how to produce biofuels without causing new and unanticipated environmental impacts. In this report we summarize the environmental effects of biofuels, illustrate some uncertainties about these effects, and identify topics for an integrated research program aimed at clarifying tradeoffs and reducing uncertainties in planning for sustainable biofuels production. Our considerations include effects on GHG emissions, soil carbon, water supply and quantity, land use, and biodiversity. We conclude: • Estimated net GHG emissions from biofuels production can be lower than those of fossil fuels. However, this is highly dependent on feedstock (raw material) choice, fuel and fertilizer inputs, whether biofuel crops replace native vegetation, and whether the soil is tilled. Further, emissions estimates for a given feedstock vary among studies, contributing to uncertainty about GHG effects. • The effects of biofuels production on water supply and quality are a function of the feedstock choice and production method. High intensity agricultural crops such as fertilized and irrigated corn can contribute nitrogen and phosphorus pollution to adjacent waterways and downstream, and can place substantial demands on regional water supplies. Perennial cellulosic crops such as switchgrass and mixed prairie grasses can substantially reduce these impacts. • Today’s grain-based biofuel crops compete with food crops for prime agricultural land. Pressure is growing to expand grain-based biofuels production onto marginal agricultural lands or land currently in the U.S. Department of Agriculture (USDA) Conservation Reserve Program. These lands support diverse wildlife communities and conversion is likely to affect some species of concern. Land conversion is also a major source of GHG production, especially when native habitats are destroyed. Land use impacts can be reduced by selecting feedstocks that do not displace food crops or require conversion of native habitats for production. • Impacts on wildlife abundance and diversity depend on the feedstock choice and whether production takes place on existing agricultural lands or on newly cleared land. At a landscape scale, more diverse feedstock crops are associated with greater biological diversity, while monocultures decrease it. Some plants being considered as sources of biofuels are potentially invasive, requiring consideration of potential impacts on habitats adjacent to the biofuels crop. Biofuels production presents a wide range of potential impacts and benefits, with substantial uncertainty associated with different choices among sources and production methods. Society must carefully consider the environmental tradeoffs of different biofuels sources, and of biofuels compared with other energy sources, including fossil fuels. An integrated research program that explores optimal crop selection, agricultural landscape design, effects on GHG emissions, soils, and biodiversity, and economic and social factors, is necessary to fully inform decisions about these tradeoffs. Appropriately designed, a biofuels production system can be a sustainable and resilient source of energy for the long term. Cover photos: Examples of biofuel feedstocks. Clockwise starting on the upper left: (a) Switchgrass (Panicum virgatum), a prairie grass native to North America (b) Two species of algae (Cyclotella and Oocystis) (c) Sunflowers (d) Hybrid poplars (crosses between two or more species of Populus), (e) Corn field. Photos credits: (a) Peggy Greb, USDA-ARS (b) Robert O. Megard, University of Minnesota (c) Edward McCain, USDA-ARS (d) Stephen Ausmus, USDA-ARS (e) Flickr user fishhawk

2 esa © The Ecological Society of America • esahq@esa.org Introduction Policy makers are increasingly looking to renewable energy sources as environmentally friendly and sustainable replacements for fossil fuels used for transportation. Biofuels, which are liquid fuels derived from a variety of sources—for example, row crops, trees, algae, and food waste— appear to hold promise to reduce our dependence on fossil fuels and to reduce net emissions of greenhouse gases (GHG) from mobile sources that contribute to global warming. Many countries, including the United States, have set targets for biofuels production (ethanol and biodiesel) and mandated the blending of ethanol with gasoline. These policies, combined with economic incentives and tariffs, have driven a worldwide expansion in the production of various crops for use as transport fuels. At present, biofuels are primarily derived from a small number of plant materials, or feedstocks, primarily corn and sugar cane, but the variety of materials is expanding. Worldwide biofuel use for transport is expected to nearly double by 2017 over 2005–2007 levels.1 A target adopted by the European Union (EU) in 2009 requires 10% of fuels for transport to be from renewable sources by 2020. In the U.S., ethanol, which accounts for more than 99% of biofuels produced, is made almost exclusively from corn grown on prime agricultural land. With rising demand and legislative targets for biofuels, increasing output will require boosting yields of existing crops, bringing more land into biofuel crop production, and/or developing new feedstocks (see Box 1). Provisions in U.S. legislation, including the U.S. Energy Independence and Security Act of 2007 and the Food, Conservation and Energy Act of 2008 (known as the Farm Bill), set new targets and incentives for biofuels production. In amending an earlier Renewable Fuel Standard (RFS), the Energy Independence and Security Act called for a nine-fold increase in renewable fuel production by 2022. Gasoline for road transportation would have to consist of 20% biofuels by this date. The U.S. Environmental Protection Agency (EPA) in 2010 issued new production targets for various biofuels and detailed schedules for meeting them (see Box 2.) The 2008 Farm Bill increased targets in the U.S. and established tax credits, grants, and other provisions to encourage the expansion and use of transport biofuels, with new emphasis on fuels derived from cellulosic sources. If produced in a sustainable fashion, biofuels could reduce demand for fossil fuels, in turn reducing needs for imported oil and mitigating climate change by limiting GHG emissions. As with any agricultural crop, however, biofuel crops can affect water supply and quality, soil biogeochemistry, land use, and biodiversity. Erosion, nutrient runoff, habitat loss, the loss of beneficial species, and the spread of invasive species are all potential risks of expanded production. Scientists and policy makers are examining how potential adverse effects on natural resources can be avoided or mitigated in order to meet biofuels production goals while enhancing environmental, social, and economic sustainability. Research has also focused on the impacts of various economic policies intended to stimulate biofuels production, the economics of growing crops for transportation fuel, and the potential impact on worldwide food production. As the need for food increases with a burgeoning global human population, biofuel crops compete for arable land and may lead to higher food prices. While acknowledging the potential economic and social implications, the objectives of this report are to summarize the environmental effects of biofuels, illustrate some of the uncertainties about those effects, and identify topics for an integrated research program that could reduce uncertainties in planning for sustainable biofuels production. We first summarize the potential direct and indirect effects of biofuels production on GHG emissions and soil carbon. Subsequent sections describe impacts on water use and quality, biodiversity, including the effects of land conversion, monoculture agriculture, invasive species, and the potential tradeoffs between biofuel energy yield and biodiversity. We propose elements of an integrated research program that could reduce the identified uncertainties. Finally, we link a set of princiEcological Dimensions of Biofuels Clifford S. Duke, Richard V. Pouyat, G. Philip Robertson, and William J. Parton ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013

© The Ecological Society of America • esahq@esa.org esa 3 ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 Box 1. Types of Biofuels and the World’s Major Producers Biofuels are a renewable energy source derived from biological materials such as crops, wood, or algae grown specifically for fuel purposes or from wastes such as forest or agricultural residues and municipal waste. Biofuels come in two primary forms: ethanol and biodiesel. Ethanol, the most widely used renewable transportation fuel, is an alcohol fuel made from sugars found in grains or derived from starches, cellulose, and other plant components. Virtually all of the transportation biofuel produced in the U.S. today is ethanol derived from corn grain. The U.S. also imports ethanol, largely sugarcane-based ethanol from Brazil, which together with the U.S. supplies 90% of the world’s fuel ethanol.a Owing to increased domestic production and tariffs on imported ethanol, imports have declined in recent years.b Ethanol is generally blended with gasoline, although some cars can run on pure ethanol. In 2009, ethanol made up about eight percent of the U.S. motor vehicle gasoline market, which is nearly double the market in 2006. A wide variety of perennial grasses, including both native prairie species such as switchgrass and exotic species such as miscanthus, are being studied for their potential as biofuel crops. Because cellulosic ethanol can be made from any number of plant species, including trees, mixtures of species can potentially be grown to optimize environmental benefits in addition to fuel production. Such mixtures might comprise two to three species of native grasses and forbs grown together to enhance prairie restoration. Cellulosic biofuels constitute less than a half-percent of current biofuels production in the U.S., although this proportion is expected to grow rapidly, in part due to incentives for development that acknowledge their superior environmental benefits. Biodiesel, now made largely from palm and vegetable oils, fats, or greases, can be blended with diesel fuel or used directly in diesel engines. Biodiesel is currently produced largely in the EU and particularly in Germany, although biodiesel production is expanding worldwide, particularly in Southeast Asia, the U.S., and parts of South America and Europe. Biodiesel from the EU is made primarily from canola (also known as rapeseed), although the U.S. and Brazil, among others, have also used soybeans for biodiesel. Oil palm plantations have expanded in Southeast Asia for production for biodiesel (see Table 1). Algae constitute another potential biofuel feedstock. Algae grown in outdoor ponds or enclosed in containers (“photobioreactors”) are typically harvested, dewatered, and dried for their lipids and oils, which can be used to make biodiesel, ethanol, or other hydrocarbons. Some producers are developing systems in which the algae excrete the desired product, for example ethanol, into the culture medium and the product is then extracted from the medium without the need for harvesting of the algae. a World Bank. 2008. Biofuels: the promise and the risks, in World Development Report 2008. http://siteresources.worldbank.org/INTWDR2008/Resources/2795087-1192112387976/WDR08_05_Focus_B.pdf b Renewable Fuels Association. 2012. www.ethanolrfa.org/pages/statistics viewed 7 December 2012. Figure 1. Examples of different biofuel types. a) field corn (conventional bioethanol); b) switchgrass (advanced bioethanol); c) sunflower (conventional biodiesel); d) green alga Botryococcus braunii (advanced biodiesel, bioethanol, biobutanol, aviation fuels). Photo credits: a) Warren Gretz / NREL b) Bob Nichols / USDA c) Peggy Greb / USDA. d) Tim Devarenne / Texas AgriLife Research. Table 1. Major biofuels and their sources. “Conventional” biofuels are those that dominate today’s marketplace. “Advanced” biofuels are new-generation fuels under experimental production. Type of Biofuel Feedstock Where produced Conventional bioethanol Corn United States, Canada Sugarcane South America (primarily Brazil), Central America, Asia, Africa Sugar beets Europe Cereals (e.g. milo, wheat, barley) Europe, Canada Cassava Asia, South and Central America Advanced bioethanol Cellulosic biomass • Grass (e.g. switchgrass, miscanthus, mixed species) In development • Short-rotation woody crops (e.g. poplar) • Plant waste (e.g. corn stover, wood waste) Conventional biodiesel Rapeseed (canola) Europe, Canada, Asia Soybean Europe, Canada, South and Central America, Africa, Asia, United States Sunflower Europe, Canada, Africa, Asia Palm South and Central America, Africa, Asia Jatropha South and Central America, Africa, Asia Castor South and Central America Advanced biodiesel, bioethanol, Algae In development biobutanol, aviation fuels Source: United Nations Environment Programme. 2009. Towards sustainable production and use of resources: assessing biofuels. http://www.unep.org/PDF/Assessing_Biofuels.pdf (a) (b) (c) (d)

ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 ples for biofuels and sustainability to a landscape approach designed to meet social, economic, and energy needs. Potential Environmental Effects The potential environmental effects of biofuels production have been examined using field measurements, laboratory experiments, computer models, and combinations of two or more of these methods. Not all studies agree with each other, and we discuss some of the reasons for differing conclusions. The effects of biofuels on the environment are many and complex; they vary greatly depending on initial conditions and assumptions. Greenhouse Gas Emissions and Soil Carbon GHG emissions resulting from biofuel production depend on 1) land clearing, if necessary, and whether the feedstock crop replaces native vegetation or another existing crop; 2) feedstock choice; 3) fuel and energy use for crop growth, harvest, and biofuels production; 4) water use and source; 5) the use of nitrogen fertilizers; and 6) soil turnover effects on carbon and nitrogen emissions (Box 3). The use of fossil fuels and nitrogen fertilizer has direct effects on emissions. Indirect effects can occur when biofuels production displaces another agricultural activity (e.g., cattle grazing in tropical regions which then expands into native rainforest). Whether biofuels production causes net GHG emissions, has no net GHG emissions, or takes up GHGs from the atmosphere (e.g., by storing carbon in plant roots and soil) depends on the entire life cycle of production and use. For this reason, researchers studying the effects of biofuels production on GHG emissions generally conduct life cycle analyses (or assessments), known as LCAs (Box 4). An LCA describes the impacts of a product at every step from start to finish— 4 esa © The Ecological Society of America • esahq@esa.org Terms and Definitions biofuels Liquid fuels derived from biological materials such as crop plants, forest products, or waste materials carbon debt The amount of carbon released as a result of land use conversion, for example from grassland or forest to crops for biofuels production. The carbon debt can be repaid over time if the biofuel produced has lower greenhouse gas emissions than the fossil fuel that it replaces (see reference 8). cellulosic Refers to fuel derived from vegetative plant tissue, composed primarily of cellulose, hemicellulose, and lignin (for example, crop residues, wood, and grass not harvested for grain); contrast to grain-based or algae-based biofuels. CO2 Carbon dioxide CRP Conservation Reserve Program of the U.S. Department of Agriculture EPA United States Environmental Protection Agency EU European Union feedstock The source material for biofuel, for example, algae, crop plants, waste materials, or wood foregone sequestration The carbon that would otherwise have been stored by an ecosystem in the absence of its conversion to biofuel cropping GHG Greenhouse gas hypoxia Oxygen deficiency indirect land use change Refers to the carbon cost of converting grassland or forest to food crops in order to replace the food production lost when cropland elsewhere is diverted to biofuels production no-till Farming without plowing (tillage); the prior crop’s residue is left on the soil surface to decompose. N2O Nitrous oxide RFS Renewable fuel standard. The first RFS was established by the U.S. Environmental Protection Agency under the Energy Policy Act of 2005. RFS2, an expanded version of the standard, was developed in response to the Energy Independence and Security Act of 2007. stover Corn leaves and stalks, a potential cellulosic feedstock USDA United States Department of Agriculture

© The Ecological Society of America • esahq@esa.org esa 5 ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 in the case of biofuels, from field to tailpipe. Results of LCAs depend partly on the spatial and temporal boundaries of the analysis, for example on the inclusion (or not) of factors such as the GHG emissions resulting from the construction and operation of the biofuel refinery. Conclusions of different studies therefore vary depending on the boundaries selected and the models used. The studies summarized below are based primarily on LCAs. Direct Effects A growing body of environmental evidence on GHG production has dampened enthusiasm for corn as a feedstock. This is due in part to the way corn is cultivated and in part to the biology of the plant itself. Corn, like many other annual crops, depends on fossil-fuel inputs for planting, harvesting, and ethanol production. Moreover, a major factor in the GHG balance of biofuel systems is how much carbon remains stored in the soil. Soil organic carbon is also necessary to maintain soil productivity. Conventional farming practices for annual crops involve tilling the soil, which substantially reduces soil organic carbon from pre-conversion levels. When the soil is tilled, microbes are stimulated to release more stored carbon into the atmosphere as carbon dioxide (CO2). According to the U.S. Economic Research Service, about 70% of the land dedicated to corn is currently tilled. Growing corn without tillage (no-till), alternatively, conserves soil carbon. Crop residue (corn cobs and corn stover - the leaves and stalks), left on the field will also conserve soil carbon. Corn production requires high inputs of nitrogen fertilizer. High nitrogen fertilization can cause corn stover to degrade more readily to CO2. 2 A recent modeling study shows that when conservation reserve or native grassland is converted to corn production using conventional tillage, there are large losses of soil carbon. The use of no-till practices greatly reduces soil carbon loss. Other agricultural practices that can reduce GHG emissions from grassland conversion include the use of slow-release fertilizer and nitrification inhibitors that have the potential to reduce soil nitrous oxide (N2O) fluxes by more than 50% (Box 3). There are roughly similar N2O emissions for corn and soybeans, and lower values for fertilized switchgrass, according to EPA. Estimates of GHG emissions of both corn ethanol and soybean biodiesel production are often only slightly lower, and sometimes higher1 , than petroleum, although some analyses suggest that best practices, if enacted, could provide GHG reductions relative to petroleum of up to 50%. In contrast to grain cultivation, cellulosic feedstocks require fewer fertilizer inputs and, because they are perennial rather than annual plants, require no tilling. Consequently they store or “sequester” carbon in the soil. Perennial biofuel crops such as switchgrass or mixed prairie grasses actually can reduce Box 2. Types of Biofuels Defined by EPA For regulatory purposes, the U.S. Environmental Protection Agency defines biofuels. Fuels developed to meet U.S. production targets must meet the requirements of those definitions. Renewable fuel: Fuel produced from renewable biomass and that is used to replace or reduce the quantity of fossil fuel present in a transportation fuel. It must also achieve a life cycle GHG emission reduction of at least 20%, compared to the gasoline or diesel fuel it displaces. [Ethanol derived from corn grain currently would fall in this category.] Target: Total of 36 billion gallons by 2022 (minimum of 15 Bgal additional). Advanced biofuel: Renewable fuel other than ethanol derived from corn starch, with life cycle GHG emissions at least 50% less than the gasoline or diesel fuel it displaces. Includes sugarcane ethanol, cellulosic ethanol, and algal biodiesel, for example. Target: Total of 21 billion gallons by 2022 (minimum of 4 Bgal additional). Cellulosic biofuel: Renewable fuel derived from any cellulose, hemicellulose, or lignin (components of stems, stalks, and woody parts of plants) each of which must originate from renewable biomass. It must also achieve a life cycle GHG emission reduction of at least 60%, compared to the gasoline or diesel fuel it displaces. Cellulosic biofuel generally also qualifies as both “advanced biofuel” and “renewable” fuel. Target: 16 billion gallons by 2022. Biomass-based diesel: Any diesel fuel made from renewable biomass feedstocks or from vegetable oils or animal fats. Its life cycle GHG emissions must be at least 50% less than the diesel fuel it displaces, and it cannot be co-processed with a petroleum feedstock. Target: 1 billion gallons by 2012 and beyond

ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 6 esa © The Ecological Society of America • esahq@esa.org atmospheric GHG concentrations by transforming CO2 in the atmosphere to stored soil organic carbon. Soil organic carbon increased in switchgrass fields harvested annually over a five-year period.3 Similar carbon storage findings have also been reported for “low input, high diversity” vegetation.4 These crops, which require little or no fertilizer input, stored more than 30 times as much carbon in soil and roots as monoculture soybean and corn crops. After accounting for the release of CO2 from fossil fuel combustion during all phases of production and processing, the GHG emissions reduction attributable to low input high diversity biomass was 6 to 16 times greater than for corn or soybeans. An additional benefit to the GHG balance is that N2O output by cellulosic crops can be half that for grain feedstocks. Substantial increases in soil carbon sequestration could occur if the 30-40% of current U.S. lands used for corn were replaced with cellulosic perennial grasses, according to modeling studies. Miscanthus, a perennial grass native to Africa and Asia, for example, has high soil carbon storage, because it has a high rate of root biomass production and does not need much fertilizer. Projections for native North American switchgrass, on the other hand, are less optimistic. Growing switchgrass without fertilization significantly decreases soil carbon and nitrogen. Algal biofuels production systems are in the early stages of development, and while a number of pilot scale production facilities are being constructed, it is difficult to make definitive statements about the GHG emissions of commercial-scale production systems. A comparison of the potential environmental impacts of algae with other biofuel feedstocks found that algae had much higher GHG emissions than corn, switchgrass, or rapeseed.5 This is attributable to the use of petroleum-based fertilizers for algal culture and to energy required to produce the CO2 that is bubbled into the water as a carbon source for the growing algae. The GHG balance might be improved by finding ways to recycle flue gas from fossil fueled power plants to provide CO2 and wastewater from water treatment plants as a source of nitrogen and phosphorus for the algae. Co-location of algae production systems with power plants could also increase the overall energy produced per CO2 from the combined power and fuel production. According to EPA, air pollutant emissions associated with algal biodiesel production are lower than those of other biodiesel feedstocks and much lower than emissions from corn ethanol production. Waste material from algal biofuels production could potentially be used as cattle or fish food, or further digested to make syngas or other biofuels. A National Research Council report, Sustainable Development of Algal Biofuels in the United States, addresses the sustainability of algal biofuels in detail. Liquid fuels produced from forest products such as wood or other biomass residues may reduce net GHG emissions more than fuels produced from agricultural sources like fastgrowing poplars or other short-rotation woody plants. GHG reductions reported for ethanol produced from woody biomass, compared with gasoline, range from 51% to 107%.6 However, Box 3. Biofuels and Nitrous Oxide Emissions In theory, biofuels should not contribute to GHG buildup in the atmosphere because the plants grown for fuel take up carbon dioxide (CO2) as they grow, offsetting carbon released to the atmosphere when the fuels are burned. However, although CO2 gets most of the attention, there is another GHG of serious concern (see Figure 2). Only about half of the nitrogen applied to a grain crop is taken up by plants while the remainder is lost to the environment. Some of the nitrogen ends up in the atmosphere as nitrous oxide (N2O), a greenhouse gas 300 times more potent than CO2. Most of the N2O is lost from the field itself, but some is lost indirectly after nitrate leached from the fields is transformed to N2O in downstream waterways and wetlands. Row crop agriculture is the largest human source of N2O globally, and N2O loss is often the biggest GHG source in annual cropping systems. Figure 2. Nitrous oxide (N2O) emissions from agriculture. Nitrogen Sources Direct N2O Emissions Indirect N2O Emissions Synthetic N Fertilizer N Fixing Crops Other Sites Urine and dung Crop Residues Residue Burning N Volatilization and Re-deposition, and N Leaching and Runoff

© The Ecological Society of America • esahq@esa.org esa 7 ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 in order for wood harvested for biofuel to have no net carbon emissions, or net carbon storage, it must be grown and harvested in such a way that the landscape-level carbon captured equals or exceeds that which the forest would have stored without being used for biofuels.7 Indirect Effects Plant biomass and soils store large amounts of carbon, so land conversion that destroys these stores and accelerates the decomposition of carbon creates a “carbon debt” by releasing CO2 into the atmosphere. Converting unfarmed perennial vegetation to annual crops grown for biofuels loses not only much of the carbon currently in the soil but can also lose the future carbon that would have been stored if the land had been left unconverted. Repayment of this “carbon debt” can occur once the net GHG emissions from production and combustion of the biofuels drop below the GHG emissions of the fossil fuel being replaced. Conversion of native lands (tropical rainforest, peatland rainforest, native Brazilian grasslands, and grasslands in the U.S.) was estimated to incur large carbon debts that would take decades to centuries to pay off, depending on the crop and the type of land being converted.8 Conversion of grasslands in the central U.S. to production of corn for ethanol would create a carbon debt that would take 93 years to repay. The estimated payback time for the carbon debt associated with converting Conservation Reserve Program (CRP) land in Michigan to biofuels under five different scenarios ranged from 29 years for cornsoybean rotations managed without tillage to Box 4. What Is Life Cycle Analysis? Life cycle analysis or assessment (LCA) provides a “cradle to grave” picture of all environmental impacts of biofuels and the processes that go into producing them. The U.S. Environmental Protection Agencya describes LCA as a systematic, phased approach with four components: • Goal definition and scoping describes the product, process or activity; establishes the context for the assessment; and identifies the boundaries and environmental effects to be reviewed. • Inventory analysis identifies and quantifies the energy, water, and materials use and environmental releases (e.g., air emissions, solid waste disposal, wastewater discharges). • Impact assessment assesses the potential human and ecological effects of energy, water, and material use and the environmental releases identified in the inventory analysis. • Interpretation evaluates the results of the inventory analysis and impact assessment. An LCA incorporates data on many aspects of the life cycle, including fertilizer use, changes in crops or acreage, and energy used for growing and transporting feedstocks and for processing the biofuel. These data are then used in economic and environmental models to assess net effects on GHG generation (see Figure 3). While an important tool, like any analysis LCA does not guarantee agreement among investigators. One reason for the variation in results is disagreement about what factors should go into an LCA. For example, one criticism of LCAs is that they often leave out various types of information, such as how changes in prices will affect other variables and feed back into the cycle. Such indirect effects are perhaps the most difficult to quantify and thus the most contentious. LCAs are imperfect, but they are incorporated into legislation such as the Energy Independence and Security Act. EPA’s RFS2 standard, for example, uses indirect land use analysis in determining the life cycle GHG emissions of various biofuels sources. As noted in the regulation, “Congress specified that: The term ‘lifecycle greenhouse gas emissions’ means the aggregate quantity of greenhouse gas emissions (including direct emissions and significant indirect emissions such as significant emissions from land use changes), as determined by the Administrator, related to the full fuel lifecycle . . .” (40 Code of Federal Regulations Part 80). There are challenges for both scientists and policymakers when there are no accepted protocols or rules for deciding which components should be included or excluded in LCAs. Calls for standardized approaches for biofuels LCA thus seem particularly pertinent given the importance of these analyses for policy.b Sources: a U.S. Environmental Protection Agency. 2006. Life cycle assessment: principles and practice. EPA/600/R-06/060, USEPA, Cincinnati, OH. b Mitchell, R.B., L.L. Wallace, W. Wilhelm, G. Varvel, and B. Wienhold. 2010. Grasslands, rangelands, and agricultural systems. Biofuels and Sustainability Reports, Ecological Society of America, Washington, D.C. http://esa.org/biofuelsreports/ Figure 3. GHG life cycle analysis of biofuels. Adapted from SC Davis et al. 2009. Trends in Plant Science 14: 140-146. Policy GHG Economic incentive GHG GHG GHG GHG GHG Energy Fertilizers Pesticides Herbicides Seeds Biofuel crop yield Liquid fuel Fuel used Energy Energy Energy GHG GHG Land-use conversion Manufacture/ transport Biofuel cultivation Conversion processing Transport

ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 8 esa © The Ecological Society of America • esahq@esa.org 123 years for continuous corn conventionally tilled.9 In contrast, using CRP land directly for cellulosic biofuel incurred no debt, providing immediate GHG mitigation benefits. In Brazil, where biofuels production is an important agricultural industry, land use change from rangeland to cropland is estimated to have a small impact on carbon emissions.10 However, the indirect impacts could offset the carbon savings from biofuels if displaced ranches expand into the Amazonian rainforest, replacing native forests with pastures. A similar problem could occur where production of sugarcane has moved into areas where soybeans are currently grown; displacement of soybean farming may put further pressure on Amazonian forests.11 In the United States, where biofuels crops are more likely to be grown on land already in agricultural production, indirect land use change is still a risk. Global demand for the agricultural commodities that are displaced by new energy crops may lead to the expansion of agriculture in new and distant areas, with associated environmental and economic costs of land use change in those areas.12 This is the argument of Searchinger and colleagues,13 who predict that as prices for biofuels rise, farmers around the world will convert forest or grassland to make up for land or grain that has been diverted for biofuels. Using a worldwide agricultural model, they estimated the additional GHG emissions that would result from this change in land use. Production of cornbased ethanol, rather than reducing GHG emissions, would nearly double them over a 30-year period and increase them for 167 years. Biofuels produced from switchgrass, which as mentioned earlier is widely considered to be environmentally preferable to corn, could increase GHG emissions by 50% if grown on land used for food production. However, some researchers argue that the effects predicted by Searchinger and colleagues depend on uncertain parameters, such as how prices drive changes in land use and ethanol consumption and production in the United States and Brazil.14 Others believe these authors underestimated several important parameters: yields per acre of food in developing countries, the potential for raising those yields, the availability of high-yield cellulosic feedstocks in the U.S., and the global availability of unused or underutilized cropland.14 Another study likewise found that GHG projections from corn ethanol in the U.S. would be significantly reduced if market-mediated responses and by-product use were factored into the analysis.15 A recalculation using these factors reduced cropland conversion of land used for the ethanol feedstock by 72%. Consequently, the estimated GHG release was roughly one-quarter of Searchinger and colleagues’ estimate of releases attributable to changes in indirect land use. Nonetheless, it was enough to cancel out the benefits of corn ethanol for avoiding global warming. While these studies had different detailed results, none were close to zero, and all were large enough to make the indirect land use emissions a large contribution to the life cycle costs. Biofuels thus can be produced in such a way that they have lower net GHG emissions than fossil fuels, but estimated emissions vary widely among studies. Some of these differences are associated with feedstock choice, plant growth and harvest methods, and production processes, while other differences reflect varying analytical methodologies. Using consistent methodologies to develop systematic comparisons would help clarify the consequences of different choices among biofuels sources and growth and production methods. Water Use and Quality Biofuels production can affect both water availability and water quality. The impacts of different biofuel crops on water depend on many factors, including fertilizer application, tillage, soil type, and whether the crop is replacing a different type of vegetation. Water Use The greatest water use in biofuels production is for growth of the crops. For biofuels crops such as corn and sugar cane, 99% of the water needed is for growth of the feedstock.16 The amount of water used to produce biofuel corn varies greatly depending on climate, with the total water required for irrigation and processing in the U.S. ranging from 5 liters per liter ethanol in Ohio to 2,138 liters in California17 and up to 2,570 liters globally (Table 2). Refining ethanol from the feedstock requires just 3–6 liters of water per liter of ethanol. Thus, although processing plants may have localized impacts on water supply, the water demand for growing biofuel crops is a bigger concern than for processing. These numbers, however, mask wide variation in irrigation practices. Although increased biofuels production is not expected to alter national water

© The Ecological Society of America • esahq@esa.org esa 9 ISSUES IN ECOLOGY NUMBER SEVENTEEN SPRING 2013 availability in the next 5 to 10 years, regional and local impacts can be expected in areas where water resources are already stressed. Consumptive water use for biofuel between 2005 and 2008 grew at almost twice the rate of growth in ethanol production, indicating that biofuel agriculture had expanded into areas where enhanced irrigation was required. In some Western states groundwater resources are being depleted rapidly for irrigation of biofuel corn.17 There are even more water-intensive crops than corn being grown throughout the world for biofuels (Table 2), and in the next four decades irrigation for biofuel crops is expected to increase by 14 to 45%.18 The water requirements of biofuel crops should factor into any policy to implement a robust and environmentally sustainable national biofuels program. Biofuels production can be consistent with sustainable water use, given energy conservation, water use planning, and careful agricultural practices. One strategy is to limit expansion of dedicated bioenergy crops to areas where irrigation is not required and locate biorefineries in areas that do not depend on irrigated corn feedstock.17,19 Even where irrigation is intermittent, groundwater irrigation can quickly reduce any GHG emissions benefits of biofuels because of the fossil energy needed to pump water from deep underground to the surface. Water use efficiency varies from crop to crop. Switchgrass, like other perennial biofuel plants with deep roots, generally uses water more efficiently than do shallow-rooted annual crops. Switchgrass can be 1.8 to 5.0 times more efficient than corn, depending on the site and assuming that all the corn stover is not removed.20 However, the total consumptive water use is higher for switchgrass and miscanthus, another perennial grass, because of the longer growing season for perennial biofuel crops. Algae can be cultivated either in ponds or enclosed, transparent containers called photobioreactors. As illustrated by the values in Table 2, pond production requires far more water. Optimizing where algae are grown—concentrating production in sunny, humid climates where evaporation is minimized—could reduce water use by 75%.21 Cultivation of marine algae either in coastal areas or using saline groundwater could also reduce freshwater use. Water Quality Water quality can be degraded by expansion of biofuels production from high intensity monoculture agricultural practices. In the U.S., industrial agriculture causes soil erosion and runoff of nitrogen, phosphorus, and pesticides into local waterways and groundwater. Among the consequences is long-distance transport of nutrients to estuaries such as the Gulf of Mexico where it leads to hypoxia and anoxia across thousands of square miles. Hypoxia (low oxygen) and anoxia (no oxygen) create dead zones where the oxygen level is too low to support animal life. Reducing hypoxia in the Gulf will be increasingly difficult given the ambitious targets set for renewable fuel production. Agricultural sources, especially corn, are estimated to contribute more than 70% of the total nitrogen and phosphorus delivered to the Gulf of Mexico22. Meeting the grain-based biofuels target could increase the annual dissolved inorganic nitrogen carried into the Gulf of Mexico by the Mississippi and Atchafalaya Rivers by 10–34%.23 Because agrochemicals bind with soil particles, soil loss through erosion is also an important factor in water quality. According to EPA, corn produced on the same fields year after year results in higher infestation of corn pests, which can lead to greater pesticide use and pesticide leaching to waters. Similar concerns about soil erosion, sedimentation, and nitrogen use have been raised about sugarcane, another major source of ethanol, particularly in Brazil.11 Additional studies of the Table 2. Water use of various biofuel feedstocks Feedstock type Water use (volume used per volume fuel produced) Ethanol feedstocks Corn (Maize) 5-2,570a,b,c (depending in part on irrigation) Switchgrass 2.9 – 423 (depending in part on irrigation)b Sugar beet 1,388a Potato 2,399a Sugar cane 2,516a Cassava 2,926a Barley 3,727a Rye 3,990a Paddy rice 4,476a Wheat 4,946a Sorghum 9,812a Biodiesel feedstocks Soybean 14-321(depending in part on irrigation)b Algae (pond production) 25-1,421b,d,e Algae (enclosed production) 30-63b Sources: a Gerbens-Leenes, W., A. Y. Hoekstra, and T. H. Van der Meer. 2009. Proceedings of the National Academy of Sciences 106(25): 10219-10223. b Harto, C., R. Meyers, and E. Williams. 2010. Energy Policy 38: 4933–4944. c Chiu, Y.W., B. Walseth, and S. Suh. 2009. Environmental Science and Technology 43: 2688-2692. d EPA. 2010. Renewable Fuel Standard Program (RFS2) regulatory impact analysis. EPA-420-R-10-006. e Wigmosta, M. S., A. M. Coleman, R. J. Skaggs, M. H. Huesemann, and L. J. Lane. 2011. Water Resources Research 47: W00H04, doi:10.1029/2010WR009966