Agriculture worldwide faces daunting challenges because of increasing population growth and changing food consumption patterns, natural resource scarcity, environmental degradation, climate change, and global economic restructuring. Yet, at the same time, there are unprecedented hopeful changes and opportunities for the future, including a remarkable emergence of innovations in farming practices and systems and technological advances that have generated promising results for improving agricultural sustainability and an increase in consciousness and concern by consumers about the sources of their food and how it is produced. One of the first comprehensive reports on the scope and importance of systems approaches to improve sustainability of agriculture was documented in the National Research Council report Alternative Agriculture (1989b). The report analyzed and described the economic and environmental results of agricultural practices that could improve sustainability being developed and used by a small subset of U.S. farmers in the second half of the 20th century, and it helped to legitimize an approach to agricultural-systems research that had previously been considered nonscientific. Many so-called alternative practices at that time (integrated pest management, no-till farming, and cover crop planting) are now used by some farmers in mainstream agriculture.

Despite the potential benefits of farming practices and systems that improve sustainability, their adoption is far less widespread than society might want. One reason for the low rate of adoption is because of social, economic, and policy incentives that discourage fundamental changes in farming systems. Another reason is that some of those practices have tradeoffs so that they might provide benefits in one aspect and negative consequences in another. The movement toward improved sustainability could be hampered by society’s lack of common agreement on which objectives are the highest priority and how tradeoffs should be managed.

Since Alternative Agriculture was published, many changes have been made in how farmers farm. While incremental approaches offer movement toward sustainability, such approaches might not tackle fundamental problems. Systemic changes and multiple paths will also need to be pursued. Indeed it can be argued that sustainability is the process of

constantly adapting farming or food systems to meet clearly articulated and desired outcomes in a robust and resource-efficient manner that also reflects social responsibility.

With the support of the Bill and Melinda Gates Foundation and the W.K. Kellogg Foundation, the National Research Council convened a committee to study the science and policies that influence the adoption of farming practices and management systems designed to reduce the costs and unintended consequences of agricultural production. (See Box 1-1 for the statement of task.) To address the statement of task, the committee solicited input from many experts in academia and federal agencies in a series of open meetings and workshops, in addition to drawing on members’ expertise. Two sets of case studies were used to examine farming systems that address those concerns and to explore the factors that affect their implementation, economic viability, and success in meeting environmental and other goals of sustainability.

This report reviews the state of knowledge on farming practices, technologies, and management systems that have the potential to improve the environmental, social, and economic sustainability of agriculture and discusses the tradeoffs and risks that might present themselves if more farms were to adopt those practices, technologies, and systems. The report also identifies knowledge gaps in improving agricultural sustainability and makes recommendations for future actions aimed at improving agricultural sustainability.

Sustainability in agriculture is a complex and dynamic concept that includes a wide range of environmental, resource-based, economic, and social issues. The committee’s definition of sustainable farming does not accept a sharp dichotomy between unsustainable or sustainable farming systems because all types of farming can potentially contribute to achieving different sustainability goals and objectives. Ultimately, the committee believes that sustainability is best evaluated against a range of environmental, economic, and social goals that reflect the views of diverse groups in society. The most intense controversies about the relative sustainability of different farming practices or farming systems necessarily take place within the domain of politics because preferences for particular farming systems reflect the priorities of various stakeholders with different working definitions of sustainability.

Although a final assessment of the sustainability of any particular farming practice or system is a social and political act, the committee believes that public debates about improving the sustainability of U.S. agriculture need to be based within a good understanding of the existing scientific research. Science documents the performance and impacts of different agricultural practices and systems, predicts outcomes likely to result from the use of different systems, develops indicators to measure progress toward sustainability goals, and expands the range of technological tools and farming management approaches available. The issue of water quality illustrates the critical role science plays. Societal concerns about water quality led to passage of the Clean Water Act, which has resulted in various local, state, and federal guidelines that require use of “best management practices.” Guidelines are provided for acceptable soil phosphorus upper limits, soil erosion rates, water conservation, tillage practices, pesticide use, and a wide array of other process-type practices, all of which were determined through extensive scientific research.

This chapter first defines key terms used in later chapters and then discusses concepts of agricultural sustainability. It identifies the boundaries, or scope, for the overall science assessment of sustainability. Ultimately, this report focuses on current scientific evidence about the performance of farming practices and systems that can contribute to moving U.S. farming systems along a trajectory toward meeting various sustainability goals. Indicators that can be used to provide quantitative assessment of progress toward sustainability are also discussed. The structure of the report is outlined at the end of this chapter.

The terms “farming,” “agriculture,” and “farming systems” can comprise a diverse range of activities. The definitions of agriculture used in the National Research Council (NRC) report Investing in Research: A Proposal to Strengthen the Agricultural, Food, and Environmental System (NRC, 1989a) provide some useful definitions and underscore the potential breadth of farming-related activities and goals (Box 1-2).

Various crop and livestock enterprises on individual farms can interact in complex ways with one another and with their surrounding ecosystems (USDA-CSREES, 2007). Combinations of different activities can generate properties of system behavior that might not be understood or predicted by looking at each practice individually. That behavior is manifested at various scales ranging from the field, whole farm, landscape, watershed, and region. Moreover, every farm is embedded in a particular biophysical environment that provides opportunities and constraints for using different practices or management strategies, which shapes the impacts of farming activities on environmental, economic,

and social outcomes. This committee also recognizes that farms operate within a complex market, policy, and community context, and interactions with those institutions and social systems affect the performance of every type of agricultural system.

Because of the complex interactions at various scales, this committee has chosen to cast a relatively wide net in its assessment of the performance of different farming practices and

systems. The ensuing chapters discuss the impact of individual farming practices not only as individual components of a particular farm enterprise, but also in the context of whole farming systems and as pieces of a larger farming landscape at the watershed, regional, or national scale. Although this report discusses how farmers might be affected by opportunities and constraints created by the larger agrifood system (see Box 1-2), a detailed assessment of the overall social, economic, and environmental performance of the food processing, distribution, and retailing sectors is beyond the scope of this report.

At the time that the report Alternative Agriculture (NRC, 1989b) was released, the term “alternative agriculture” was commonly used to refer to farming approaches (most notably, organic farming) that appeared to be dramatically different from the dominant or “conventional” farming systems that characterized contemporary crop and livestock production in the United States. Today such alternatives are more likely to be referred to as “sustainable” farming, but the phrase is unfortunately ambiguous. Not only do farming enterprises reflect many combinations of farming practices, organization forms, and management strategies, but also all types of systems can potentially contribute to achieving various sustainability goals (for individual farmers and for resource use and environmental sustainability at the landscape scale).

At some level, the distinctions between what some call “conventional” agricultural systems and a range of alternative systems have some basis in empirical reality in the U.S. context. The characteristics and examples of practices associated with each system are summarized briefly in the following sections to clearly define those terms and how they are used in this report.

Conventional agricultural systems draw from a set of predominant farming practices in the United States, although conventional farms are diverse in what they produce and in the specific combination of practices they use. The size of farms using conventional production methods ranges from small to large. Most agricultural commodities in the United States are produced in conventional agricultural systems.

• Conventional crop production makes use of synthetic pesticides and herbicides, and supplements nutrients generated on the farm (manure) with synthetic fertilizer to maintain soil fertility. Fields are more frequently planted in few rotations of marketable crops than left fallow or planted with cover crops. Conventional corn, soybean, and cotton farms are increasingly planted with seeds that are genetically engineered to facilitate weed control or to reduce pest losses (and pesticide use).

• Conventional animal production varies depending on the species produced, the size of livestock inventories, and the amount of cropland or pasture available per animal unit. Animals might be housed in partial to full confinement structures, with beef cattle, sheep, and goats being less confined (for at least part of their life cycle), dairy cows being more confined than sheep and beef cattle, and conventional hogs and poultry most confined among livestock. Although beef cattle, sheep, and goat farms rely heavily on pastures, most conventionally raised dairy cows, hogs, and poultry receive virtually all their feed from harvested forage and grain crops raised on the farm or purchased from other farmers. Depending on the number of

animals and the resources of the producers, vaccines, antibiotics, medicated feeds, and growth hormones could be used in production. Animal manure is spread on fields or managed in lagoons, with excess liquid sprayed on fields as in the case of some hog farms.

• Industrial crop and animal production is a term that has come to be associated with operations that, generally speaking, are characterized by large size combined in some cases with a high degree of specialization. Producers of industrial scale are more likely to produce under contract with food processors and handlers. Animals are more likely to be grown in confined housing, with no pasture in the case of swine and poultry, or with portions of the animal life cycle on pasture or low confinement for dairy and beef animals. Feed is more likely to be a purchased input, rather than produced on the farm, than operations of smaller scale. Many of the larger operations likely have cropped land for liquid manure application, but loading limits over time rarely allow long-term stability for adequate disposal or low-rate application for efficient use of water and nutrients. Contract arrangements often are used for distant application of manure. Industrial farms typically operate at a scale that allows for more extensive division of labor and the use of capital intensive machinery and buildings. They more often rely on a hired workforce than do their smaller-scale counterparts.

Over the last few decades there has been a major effort to develop new management approaches and farming practices that not only improve the economic performance and productivity of conventional and industrial farming systems, but also prevent and mitigate their potentially negative effects on soil erosion and water quality, some of which also can improve the economic performance and productivity of conventional farming. Examples of practices used in these strategies are summarized in Box 1-3 and will be discussed in depth in Chapter 3.

Some farming approaches have been developed, at least in part, to respond to perceived problems associated with conventional farming. They represent a concerted departure from some of the key features of conventional farming such as the reliance on off-farm and synthetic inputs. Such approaches emphasize the use of natural processes within the farming system, often called “ecological” or “ecosystem” strategies, which build efficiency (and ideally resilience) through complementarities and synergies within the field, the farm, and at larger scales across the landscape and community. Examples of such systems include organic and biodynamic farming (Box 1-4), although “pure” forms of each system are difficult to identify. Like conventional farms, the specific practices used could vary widely from farm to farm, and often include combinations of the practices discussed in Box 1-3.

A dichotomy between “ecologically based” and “conventional” farming systems has been a component of public debates over the performance of the U.S. farming sector for several decades. The dichotomy has been a heuristic device that scientists use to explore the comparative performance of different farming systems. However, in reality, these terms are to be used with great caution because farming enterprises found in the United States clearly reflect a potentially infinite set of combinations of particular farming practices, organizational forms, and management strategies. (See the case studies in Chapter 7.)

Moreover, the definitions that once clearly demarcated conventional from ecologically based farming systems have become muddled. One example is the “conventionalization” of the organic farming industry in the United States (Guthman, 2004), characterized by the entry of large-scale farms in the market for USDA-certified organic products. Most farms present examples of hybrid or intermediate stages on a continuum between the extremes of agricultural practices, and their adoption of various new practices has produced apparent gains in environmental performance (Keystone Center, 2009). Some of the “mixed farming systems” also have been given names (Box 1-5).

The committee concludes that no simple typology or set of categories can capture the complexity of the farming practices and systems used on diverse U.S. farms. The lack of a single accepted typology complicated the writing of this report. Because so much of the research literature is based on comparisons of particular farming practices, or of one or more of those stylized “farming systems,” research findings are cited throughout the report using the categories described by the scientists who conducted the research. For this reason, this report cites organic farming systems more frequently than other ecologically based systems. The illustrative use of organic systems is not intended to imply that organic

farming systems are more sustainable than other farming systems. Indeed, all farming system types have opportunities to improve in sustainability, and all farming system types could be unsustainable depending on their management and on environmental, social, and economic changes over time.

In its broadest sense, sustainability has been described as the ability to provide for core societal needs in a manner that can be readily continued into the indefinite future without unwanted negative effects. Most definitions of sustainability are framed in terms of three broad social goals: environmental, economic, and social health or well-being.¹ For example, a sustainable farming system might be one that provides food, feed, fiber, biofuel, and other commodities for society, as well as allows for reasonable economic returns to producers and laborers, cruelty-free practices for farm animals, and safe, healthy, and affordable food for consumers, while at the same time maintains or enhances the natural resource base upon which agriculture depends (USDA-NAL, 2007).

The legal definition of sustainable farming systems as defined in the Food, Agriculture, Conservation, and Trade Act (1990 Farm Bill and revised in 2007) is a useful starting point for identifying sustainability goals for the purposes of this report:

This legal definition mingles a description of societal sustainability goals with strategies that can be used to achieve those goals (for example, “an integrated system which will use, where appropriate natural biological cycles and controls”; or “make the most efficient use of nonrenewable and on-farm resources”). This report makes a clear distinction between societal sustainability goals and the management systems used to pursue these goals. That distinction recognizes that the same goals can potentially be achieved through a range of different management and organizational approaches.

Modifying the Farm Bill definition slightly, the committee isolated four key societal sustainability goals that serve as the organizing principles for the remainder of this report (Figure 1-1):

• Satisfy human food, feed, and fiber needs, and contribute to biofuel needs.

• Enhance environmental quality and the resource base.

• Sustain the economic viability of agriculture.

• Enhance the quality of life for farmers, farm workers, and society as a whole.

The sustainability of a farming practice or system could be evaluated on the basis of how well it meets various societal goals or objectives. To be sustainable, a farming system needs

FIGURE 1-1 Sustainability goals used in this report. The area where the four goals overlap represents the highest sustainability in the continuum.

to be robust (that is, be able to continue to meet the goals in the face of stresses and fluctuating conditions; to adapt and evolve), be sufficiently productive, use resources efficiently, and balance the four goals. There are, however, often tradeoffs or synergies among the various goals and their related objectives, toward which sustainability is directed.

In the discussions that follow, the scientific evidence surrounding different farming practices or farming systems that illustrate their ability to further each of these four societal goals is discussed. The committee is not suggesting or implying that a farming practice would have to simultaneously accomplish each of these goals to be considered “sustainable.” Rather, it recognizes and expects that combinations of practices used in a system will affect each of the four goals in different and often complicated ways. A sustainable system would balance and meet each of the four goals to a large extent.

Each of the four sustainability goals consists of a large number of more specific objectives that represent different paths toward achievement of the goal. For example, the goal to “satisfy human food, feed, and fiber needs” requires managing farming systems in the aggregate so that there will be enough affordable food and fiber (including for energy production) in the future for all on the globe, although U.S. production would only play a part in overall global production. That goal has a long history and is a fundamental concern of all societies through time and can be summarized with the crucial question: Will there be sufficient agricultural resources in the future?

Achieving that goal will, at a minimum, require sufficient productivity (for example, the sheer volume of outputs produced from a given agricultural activity), farming practices that produce the outputs at a price that consumers can afford, and marketing and distribution systems to ensure that people have ready access to farm products. The concepts of productivity, affordability, and access represent specific objectives that are required to meet the overall goal.

Even relatively simple objectives can quickly become more complex. For example, agricultural productivity over time is influenced by the technologies that are available.

Sometimes built capital, such as machinery or chemicals, can substitute for natural capital, such as natural soil fertility. Food productivity has risen over time in the United States, in part, because of the substitution of fertilizers for natural soil fertility. If the substitution is viewed as socially acceptable, if fertilizer is affordable and effective, and if its use is not accompanied by unwanted or detrimental side effects, then the loss of natural soil fertility as a result of a farming practice might be viewed as sustainable. If, however, fertilizer is viewed as a poor substitute for natural fertility, as having important unwanted side effects, or is thought to be unaffordable or ineffective in the future, then a farming system that results in losses of natural fertility of the soil will be viewed as unsustainable (Batie, 2008b).

Likewise, each of the other three goals comprise a number of specific objectives. The objectives listed in Box 1-6 are representative of various objectives associated with sustainability goals, and they are used to organize the review of the scientific literature in Chapters 3, 4, and 5. The specific objective of access to food is not discussed in this report because it is covered by another report, The Public Health Effects of Food Deserts: Workshop Summary (IOM and NRC, 2009). Quality and safety of food output is discussed in the context of its quality and safety at the farm gate. Food processing is beyond the scope of this report.

As the representative objectives listed in Box 1-6 illustrate, the question of the sufficient quantity of agricultural resources is not the only societal concern underlying calls for more sustainability. For example, the objectives listed with respect to environmental quality reflect a societal concern about the impacts of agriculture on the functional integrity of envi-

ronmental resources and not just whether the environmental resources will be sufficient in quantity and quality for agricultural production. Similarly, the objectives about the welfare of farmers and animals are not always driven by concerns with respect to the sufficiency of agricultural resources; rather, they are ethical statements about how humans should treat each other and animals.

Agricultural systems that move toward greater sustainability generally strive for several fundamental qualities. One of those qualities is to work with natural ecological and biogeochemical processes and cycles to maximize synergistic interactions and the beneficial use of internal resources, and to minimize dependence on external inputs. Another quality is to close nutrient, energy, and other resource cycles to the maximum extent feasible to reduce undesirable losses to the environment and additional waste disposal activities. Third, farmers, conventional or alternative, who work toward improved sustainability tend to understand and work with the social, cultural, and economic goals of people and institutions throughout the farm and food chain, which encourages synergistic relationships in the social and economic realm and increases the likelihood of desired outcomes emerging from investment of time and resources. The following chapters will illustrate different approaches being used to enhance these qualities in both the agronomic and socioeconomic aspects of U.S. farming systems.

Farming is inherently a risky enterprise that requires constant adaptation to changes in environmental (for example, temperature, rainfall, wind), biotic (for example, prevalence of pests and diseases), as well as market (for example, commodity and input prices, consumer demand) and social conditions (for example, labor availability, policies). Farming practices or systems will differ in the extent to which they are vulnerable to different kinds of risks. With the advent of climate change and the corresponding increase in fluctuations and uncertainty in weather conditions, creating less vulnerable farming systems is of increased importance, because the capacity of crop insurance to cover greater crop losses will be limited and the potential for farming systems to fail might increase under climate change if appropriate adaptations are not made (Walker and Salt, 2006).

When thinking about vulnerability, two helpful concepts from ecosystem ecology are resistance and resilience. Resistance is the ability of a system to resist being dislodged from a stable condition by a disturbance such as some sort of system stressors and fluctuating conditions. In other words, resistance is the ability to resist change in functioning. Resilience has traditionally been regarded as the speed and extent at which stability returns to a system that is dislodged from a stable condition. New schools of thought emphasize that systems do not oscillate around a single stable state, but are highly dynamic and can shift between states depending on the extent of stresses affecting the system (Holling, 1996; Folke et al., 2002; Walker and Salt, 2006). Resilience, thus, is defined as “the capacity of the system to absorb a spectrum of shocks or perturbations and still retain and further develop the same fundamental structure, functioning and feedbacks” (Chapin et al., 2009). This report uses the new definition. Both resistance and resilience refer to the ability of a system, such as a farming system, to be able to function in the face of disturbances. At the landscape and community scales, resilience depends heavily on the diversity and types of farms and of their markets, as well as biodiversity (notably presence of perennial habitats). A related concept is adaptability (that is, the opposite of vulnerability), which reflects the ability of a system (biophysical or human) to evolve and change in response to long-term changes in

the surrounding environment. The concepts of resistance, resilience, and adaptability apply equally to the properties of natural ecosystems and to social and institutional systems. The overall robustness of a farming system—that is, the ability of farming systems to withstand stresses, pressures, and changes in circumstances—will result from some mix of resistance, resilience, and adaptability.

Concepts of resistance and resilience have long been a cornerstone of ecological theory and other fields that deal with uncertainty and risk management (Holling, 1996, 2001; Walker and Salt, 2006; Nelson et al., 2007; Chapin et al., 2009). Although those concepts have been discussed in general terms as a desirable attribute of sustainable agricultural systems for over 20 years (Conway, 1987), little empirical research on the subject has been done for agricultural systems. Nonetheless, system resilience, resistance, and the ability to adapt in the face of climate change has garnered considerable attention in such fields as economics (for example, Brown and Lall, 2006; Goldstein, 2009), planning and engineering (for example, Fowler et al., 2003), and social science (for example, Adger, 2006; Ebi et al., 2006; Janssen et al., 2006). That kind of work has resulted in the development of a number of “resilience frameworks,” modeling approaches, and other methods for determining resilience and resistance as well as mechanisms of adaptation (Turner et al., 2003; Eakin and Luers, 2006; Folke, 2006; Walker and Salt, 2006; Chapin et al., 2009). Furthermore, there is widespread agreement that an effective assessment of the long-term adaptive capacity of a given system requires linkages across multiple spatial scales and integration of biophysical, economic, and social considerations (Holling, 2001).

The ability to adapt to changing conditions is determined not only by the resilience and resistance of the biophysical system, but also by the capacity for self-organization and learning. The strength of social and institutional networks that support agriculture will play a pivotal role in the ability to adapt to climate change, increased variability in weather, and changing conditions. Considerable work, termed vulnerability science, is examining the relative vulnerability of different communities to climate change, and conceptualizing approaches to increase a community’s ability to adapt to change that involves building social networks, appropriate institutional arrangements, and infrastructural capacity (Turner et al., 2003; Nelson et al., 2007). Agroecological and social systems are linked in current frameworks being used to address resilience, resistance, and adaptation (Turner et al., 2003; Eakin and Luers, 2006; Folke, 2006), and the linkage had been demonstrated by individual case studies (Milestad and Darnhofer, 2003; Robledo et al., 2004). Naylor (2008) provided a synthesis of resilience issues facing agricultural production systems globally and emphasized the important role that policies can play in supporting or detracting from creation of resilient systems. (See also Chapter 6.) A given farming system can therefore be robust when managed by farmers with access to adequate resources (for example, capital and labor) and where strong social networks and institutions are in place, but the system can be vulnerable when attempted by resource-poor farmers with a fragmented social system.

This report discusses how farming practices, management systems, and social organization can further various social objectives and goals. When possible, it examines evidence that those approaches help increase system robustness—that is, enhance the ability of a farm, farm household, or community to resist shocks (for example, market volatility, weather events, pest outbreaks), adapt or evolve in the face of changing conditions, and be resilient over the long term.

Assessing the sustainability of a farming system can become complex because of the importance of spatial and temporal scales. Initially, a farming system’s sustainability might be evaluated differently at different spatial scales (such as at the farm, community, wa-

tershed, nation, or global level) and across different interconnected economic systems (including connections between food production, processing, and consumption). A good example is an assessment of the nutrient dynamics of farming systems. Annual cropping patterns might temporarily result in the deposition, uptake, or loss of various crop nutrients in a single season. Across a series of crop rotations, however, a field might be managed in a manner to present a relatively efficient and balanced nutrient budget over a whole rotation. Analysis of whole farm nutrient budgets on different types of farms can illustrate the ways that combinations of land type and availability, cropping patterns, and livestock feeding and manure management practices interact to create dramatically different nutrient outcomes depending on how producers manage their set of resources. At a larger scale, individual farms that are not able to use all their livestock manure nutrients efficiently can conceivably be organized under certain institutional settings to provide that resource to neighboring crop producers who can use them and effectively recycle the excess nutrients. Diversification of production systems and nonfarm habitat at a landscape or watershed level can greatly enhance the robustness of the system and reduce negative environmental impacts even if individual farms are still specialized and have limited diversification in land use (Santelmann et al., 2004; Boody et al., 2005).

Similar examples can also be found in the assessment of social and economic outcomes. A farm’s separate crop or livestock enterprises might each produce positive or negative economic returns in a given year, yet the synergistic effects of the farm’s combined enterprises might produce a different overall level of farm performance. It is common, for example, for dairy farmers to focus their management efforts on dairy herd performance, while using merely adequate (but perhaps not the best) management practices on their field crops.

Sustainability can also be assessed across different time scales, with potentially different results emerging across short-term and long-term time horizons. Those differences become important when implementing environmental policies based on performance and outcome-based indicators (discussed below). For example, the characteristics of perched water tables², and soils or stream sediments with high storage and release capacity for nutrients, will reflect the accumulation of chemicals or nutrients over many years. In those cases, effects of changed farming practices will not likely be detected until many years later. In those situations, means-based indicators based on the extent of use of different BMPs might be more appropriate as a policy tool than outcome measures such as nutrient concentrations in the water, at least over the short term. Time scale is an important factor because sustainability is a dynamic process moving along a trajectory toward meeting societal goals (that are also dynamic), as opposed to achieving some well-defined end point.

Recognizing that there are multiple goals toward which sustainability can be examined (for example, generating food, feed, fiber, biofuel, environmental, economic, and social outputs), the complexity of the concept becomes readily apparent, and it is obvious why the topic has engendered much debate and contention. Each of the goals is made up of multiple aspects, and different goals can be mutually reinforcing (or synergistic) or present difficult tradeoffs among competing, socially desired outcomes. Synergies might create opportunities for potential win-win situations where pursuit of one outcome generates corollary benefits in other categories. Conflicts might result in tradeoffs, and the relative priority given to each goal will depend on the values of the stakeholders who are part of the decision-making process.

Tradeoffs can occur among different types of environmental impacts. For example,

some practices designed to minimize negative impacts of farming practices on water quality can worsen problems with air quality. That point can be illustrated by the use of riparian zones and treatment wetlands. They can reduce nitrogen fluxes into surface waters in part by increasing rates of denitrification. However, the process of denitrification does not always result in the complete conversion of nitrate to nitrogen gas, in which case various potent greenhouse gases, nitrogen oxides, are produced; thus, a tradeoff exists between improving water quality and air quality (Crumpton et al., 2008). While riparian buffer strips are designed to reduce negative effects of crop farming on nearby water bodies and are beneficial in most extensive cropping systems, there are concerns that they provide habitats for wildlife that might defecate in the crop fields and contaminate vegetables and fruits that are consumed fresh (Atwill, 2008; Doyle and Erickson, 2008).

Contentious tradeoffs can also occur between environmental, social, and economic goals. Examples include production of food to feed a growing world population versus a desire to use production practices that protect soil, air, water, and biological resources and preserve some resources for nonfood production uses such as wildlife habitat. Efforts to use environmentally friendly practices or to improve the economic conditions of farmers or farm workers can sometimes increase production costs and possibly hinder access to affordable healthful food among low-income consumers. Opinions differ widely as to whether those goals necessarily are in direct conflict, or the extent of tradeoffs involved, but nonetheless balancing the different goals clearly has to be addressed.

Another potential tradeoff could be between the ability of a system to produce the outputs desired by society (for example, food, fiber, and fuel) and the resilience and resistance of that system. For example, diverse farming systems with multiple crops or integrated systems with livestock might be more able to sustain reasonable production and profit in the face of climatic or market volatility, but they might be less productive when measured by volume of production or by profits in “normal” or optimal years. However, the more variable and unpredictable conditions become, then the argument for trading some degree of maximum productivity, or efficiency, for greater stability becomes stronger (Walker and Salt, 2006). In the case of agriculture, that tradeoff may mean sacrificing the ability to achieve maximum yields and income in good years in return for a system that performs well over a wide range of conditions and is less likely to fail in bad years. Managing a system to achieve high yields clearly is an important component of sustainability, but maximizing one component can come at the expense of overall system resilience, which in turn reduces overall sustainability. (See Walker and Salt, 2006, for illustrative examples.) As used above, the term “efficiency” reflects efforts to maximize input use efficiency per unit production. In the rest of this report, the term “efficiency” is sometimes used with a similarly narrow definition (as in the case of the discussion on water use efficiency in Chapter 3). Other times, the term is used in the broad context of “systems efficiency” to reflect the notion of minimizing undesired outcomes (such as pollution and waste) from resource use while maximizing a wide group of desired outcomes (which could include production and support for important ecosystem services) and reducing the need for external inputs (which could be achieved by increasing nutrient cycling between animal and crop production).

Any single farming system is unlikely to meet fully all of society’s production, environmental, economic, and social goals and objectives. Indeed, it is most probable that meeting many of society’s goals will require a mixture of many farming types and systems rather

than the adoption of any one type. The societal choice about agricultural systems is a social choice about what types of agriculture are desirable and therefore what the future of agriculture ought to be. The debate over the wisdom of the “alternative” futures is made difficult by underlying value systems or philosophies of agriculture that produce competing opinions about that future. (See Box 1-7 for a discussion of contending philosophies of agriculture that underlie many of the societal disagreements about these goals.) The competing

opinions about the future of agriculture then leads to the questions of how to prioritize the sustainability goal(s), what are the appropriate tradeoffs to be made, and, importantly, who gets to decide. That is, whose values matter, who benefits, and who bears the costs?

Currently in the United States, and unlike the European Union (EU) (Dobbs, 2008), no comprehensive policies promote broadly and coherently defined sustainable agricultural trajectories. Instead, sustainability goals tend to be identified and addressed separately across a mix of settings, including farm commodity programs, farm and nonfarm environmental regulation, agricultural research and technology development, land use policies, grassroots activism, and public and private efforts to develop markets that reflect emerging consumer preferences for food products raised under certain production conditions (for example, organic, natural, fair trade, or cruelty-free livestock practices). Sometimes other policies work at cross purposes with those policies pursuing sustainability goals. For example, policies to mandate the use of biofuels for the fueling of automobiles could result in increased food prices as some commodity crops are used for fuel production (Collins, 2008; Tokgoz et al., 2008), or they could encourage extension of agriculture into previously nonfarmed lands with attendant losses of important habitat or unwanted contributions to climate change (Searchinger et al., 2008).

From the preceding discussion, it is apparent that the sustainability of agriculture is not simply a question of science. Decisions about selecting among various alternative futures for agriculture and their attendant environmental, economic, and social goals emerge from an articulation of social aspirations, which falls within the realm of politics. It is through deliberative, democratic processes that the expression, discovery, transformation, and creation of social beliefs and policy preferences can occur.

Because societal sustainability goals do not emerge from science (although they can be informed by scientific knowledge), there are implications for what science can and cannot tell us. For example, science cannot with validity tell us what ought to be (for example, provide societal objectives or decide what course of action should be taken), but it can provide an analysis of alternatives and options and make predictions about potential outcomes from the use of different approaches. In essence, a major role for scientists is to serve as honest brokers in terms of involvement in policy formulation—adding knowledge to the “what is,” “what if,” and “if, then” types of questions, but leaving the “what ought to be” questions to nonscientific forums (NRC, 1996; Pielke, 2007). Thus, science is needed to help identify and clarify issues, and to seek to expand the choices available for whoever is making decisions about management of agricultural systems, be they policy makers, commodity organizations, farmer groups, or individual farmers. Science can also supply the knowledge necessary for the development of new agricultural technology (for example, technology for controlling water pollution), but scientists can only validly advocate the adoption of such technology when there is general agreement on the overall social objective to be accomplished (for example, water quality protection). The more contentious the debate over desirable objectives and the more uncertain the related science as to causes and effects, then the more important it is for scientists to adopt an “honest broker” strategy.³ It is

in the “honest broker” spirit that this committee undertook to review the various scientific (social and biophysical) literature on sustainable farming systems.

In 1999, the National Research Council’s Board on Sustainable Development observed that the “quest for sustainability will necessarily be a collective, uncertain and adaptive endeavor in which society’s discovering of where it wants to go and how it might try to get there will be inextricably intertwined” (NRC, 1999, p. 17). That is, sustainability will be discovered and more clearly defined for agriculture as society experiments with different farming systems and observes their consequences relative to sustainability goals and makes legal, institutional, and management adjustments in response. In that sense, sustainable management of agricultural resources is a journey of discovery and adaptive management, more than a specific destination.

This committee, thus, regards sustainability as more of a process that moves farming systems along a trajectory toward meeting various socially determined sustainability goals (that is, desired outcomes) as opposed to achieving any particular end state. Sustainability goals and strategies used to achieve them are expected to change over time as more is learned about tradeoffs and societal objectives, and as biophysical or socioeconomic conditions change. For the remainder of this report, those assumptions are used to guide the scientific review to identify what attributes a farming systems needs to exhibit to be sustainable (that is, to put agriculture on a sustainable trajectory overtime).

Evaluation of progress toward sustainability in agricultural systems inevitably requires monitoring of some set of measurements or indicators. The National Research Council’s report, Our Common Journey; a Transition Toward Sustainability, noted the importance of appropriate monitoring and indicator systems, with indicators being “repeated observations of natural and social phenomena that represent systematic feedback. They generally provide quantitative measures of the economy, human well-being, and impacts of human activities on the natural world” (NRC, 1999, pp. 233–234). Indicators are used to represent or serve as proxies for the impacts or outcomes of concern. In the policy arena, indicators can be used to inform the design and implementation of programs and policies.

Figure 1-2 provides an example of the use of indicators in an adaptive management framework to pursue the broad goal to “enhance environmental quality and the quality

FIGURE 1-2 Adaptive management for sustainability of agricultural farming systems.

of resource base.” An example of a specific objective within that goal could be to improve water quality, which might then be narrowed to one particular water quality concern, such as nitrogen runoff from corn production that is thought to contribute to hypoxia conditions in the Gulf of Mexico. Decision makers could use information from indicators that directly or indirectly approximates nitrogen runoff by assessing the area where particular agricultural practices or farming systems are present. Policy actions could then be based upon an evaluation of the effectiveness of behavioral changes on nitrogen loading into the Mississippi River and a multilevel spatial assessment to target efforts in what are found to be the most critical areas.

Indicators of sustainability presume the existence of goals and objectives, and yet there is no guarantee that all parties will agree on which sustainability objectives and goals are desirable or most important, particularly if tradeoffs are involved. As argued previously, the choice and priority of different goals, and hence use of different indicators, require a political process involving multiple stakeholders to best serve society. Indeed the authors of the 1999 NRC report concluded “that there is no consensus on the appropriateness of [a single] … set of indicators or the scientific basis for choosing among them” (NRC, 1999, p. 243).

Once sustainability objectives are clearly prioritized, indicators are useful for measuring progress toward the desired state as a result of changes to management. There is a large body of literature devoted to the development, validation, and use of indicators designed to assist in policy implementation and for conducting research on the relative sustainability of different farming systems (Rigby et al., 2001; Zhen and Routray, 2003; Bell and Morse, 2008). Conclusions made are predicated on the assumption that the selected indicators accurately reflect how the system is performing, and hence the choice of indicators needs to be transparent. Each indicator needs to be validated to show that it is a reasonable proxy for the processes and functions it is meant to represent (Bockstaller and Girardin, 2003).

Because sustainability involves multiple goals and objectives, it is also important to use a mix of environmental, agronomic, economic, and social well-being indicators when evaluating the sustainability of whole system performance (Zhen and Routray, 2003; Van Cauwenbergh et al., 2007). The use of multiple indicators also enables the identification of tradeoffs and synergies. At present, the bulk of the work on sustainability indicators for agricultural systems has focused on the first three goals (to satisfy human food and fiber needs, to enhance environmental quality and the resource base, and to sustain the economic viability of agriculture), with social indicators for the fourth goal (to enhance the quality of life for farmers and society as a whole) less developed and researched.

Indicators selected need to be appropriate to the sustainability objectives and need to have the following characteristics:

• Accurately reflective of the process or function it represents.

• Sufficiently sensitive to pick up changes over time and among different farming systems.

• Feasible to measure in terms of time, expense, and level of skill required.

• Understandable and relevant to end-users.

In reality, finding ideal indicators is difficult and compromises have to be made. For example, an accurate representation of many system functions (such as nutrient cycling and retention) requires intensive data collection, monitoring from multiple locations, and

repetition at appropriate time intervals. Those tasks are expensive, time-consuming, and require specialized skills. Thus, a spectrum of indicators has emerged that range from simple, low-cost, means-based indicators to more costly outcome- or effect-based indicators. Means-based and outcome-based or effect-based indicators fall in different stages along the cause–effect chain of environmental impacts as a result of farming production practices (Figure 1-3). Means-based indicators are restricted to description and quantification of practices, whereas outcome-based (or effects-based, as in Figure 1-3) indicators can be assessed at a number of steps along the chain, from emissions (for example, what pollutants are released from the farming operation) to direct measurements of resulting environmental impacts. Measuring impacts directly seems to be ideal, but it can be difficult and expensive to accomplish, with considerable uncertainty in assigning cause-and-effect relationships. Furthermore, some effects might not be detectable for many years, by which time they might be difficult to reverse. Thus, having good indicators from earlier steps in the cause–effect chain is essential for taking timely preventive actions. (See Figure 1-4, Payraudeau and van der Werf [2005] for more detailed discussion of indicator classification.)

An example on nonpoint source pollution helps illustrate the difference between means-based and outcome-based indicators. Means-based indicators record what practices and technologies are being used and then rely on previous scientific models to infer the likely effects of using the practices. For example, it is common to equate the use of recommended BMPs (such as presence of buffer vegetation or use of nutrient budgets, slow release fertilizers, cover crops, or sufficiency tests) with a standardized coefficient to reflect

FIGURE 1-3 Cause–effect chain from farmer production practices to environmental impacts for a farming region.

NOTE: Pressure indicator is an indicator that provides information about anthropogenic stresses acting on an ecosystem. State indicator is an indicator that provides information about the state of an ecosystem or specific biota within ecosystems (Payraudeau and van der Werf, 2005).

SOURCE: Payraudeau and van der Werf, 2005. Reprinted with permission from Elsevier.

expected reductions in nonpoint source pollution. On the other hand, outcome-based or effects-based indicators require some combination of direct measurement of reductions in runoff, leaching, levels of nutrients in surface and ground water, or outputs from simulation models.

In some situations, only means-based indicators are feasible and accessible to end users, particularly for large-scale monitoring for policy and regulatory contexts. Nonetheless, caution is to be used when relying on means-based indicators, particularly if changes measured in such means-based indicators (for example, BMP use) have a more complex relationship to landscape-scale outcomes than was present in the experimental conditions used to estimate their original impact coefficients. The most robust means-based indicators are those with simple causal links derived from a substantial body of evidence compiled from real-world conditions. A good example is the use of percentage of ground cover as an indicator of soil erosion potential, a relationship that has been demonstrated many times across different soil types, slopes, and climates.

There are dangers associated with the use of outcome-based indicators, notably when effects can be measured, but the causal links to what is actually responsible for causing the effects are not clear. That particular concern occurs when outcome-based indicators are used in a regulatory context or for awarding incentives. Using water quality as an example, changes in nutrient loadings in streams or lakes can be measured directly, but identifying the cause of the changes can be extremely difficult in the absence of detailed and comprehensive knowledge of surface or ground water hydrology, pollutant transport, and in-stream processing of pollutants in the region. Nutrient loadings in streams and lakes might reflect more of the effects of historical management than of the current one. An increase in nutrient levels in a section of a stream might not relate to the practices used on farms immediately adjacent to the stream, but rather to some combination of local input and loading from subsurface water that contain nutrients derived from a much larger area within the watershed. In that case, stream nutrient levels would not be a good indicator for forming the basis of farm-level regulatory decisions.

Means-based and outcome-based indicators can also be developed for measuring the impacts of farming systems on economic and social sustainability goals. However, there is typically a small scientific knowledge base to allow an analyst to link the use of particular production practices or agricultural system with a prediction for economic or social outcomes. For example, whether adoption of a particular farming practice or system is likely to increase the economic well-being of farmers and their households might be estimated by a means-based indicator such as the degree of adoption of a particular practice. However, to link well-being to that practice might require linking adoption to such factors as reduced costs of production per unit output. An outcome-based indicator might be to examine levels of net farm or household income among farms using the practice. Regardless of the indicator selected, complexities in the relationship between use of the practice and accomplishment of the ultimate outcome of improved economic well-being (such as variation in the management abilities or approaches of individuals, interactions among different cropping or livestock production activities on a farm, or different marketing outlets) make interpretation of each type of indicator challenging.

The usefulness of indicators can be improved if they are designed to reflect performance in terms of a more complete set of systems attributes, including measures of productivity, resource-use efficiency, and robustness. For example, a commonly used indicator for the objective of high productivity is crop yield, typically average yield per acre. However

a more complete indicator would also include assessment of system robustness (such as a measure of variability in yields over space and time, or the probability of yields falling below a certain threshold) and of resource-use efficiency of the system as measured by nutrient, fertility, water, and energy use expressed per unit crop yield. Similarly, measures of biodiversity could be improved by incorporating considerations of productivity (number of species supported and their population sizes), robustness (fluctuations in population sizes, and the number of extinctions and invasions over space and time), and system efficiency (for example, the amount of land and water used to support a given number of species and populations).

In the case of social goals, such as increasing community social and economic wellbeing, it would be valuable not only to examine productivity in terms of the ability of an agricultural system to produce desired outcomes (for example, increased net farm income, improved availability of affordable quality food, and decreases in income inequality), but also to pay attention to the robustness (stability of outcomes in the face of changing biophysical, market, and policy circumstances).

Indicator values have to be interpreted to be meaningful: that is, significance needs to be assigned to an indicator ’s numerical (or qualitative) value. Threshold values have not been established for many environmental indicators. Conclusions are typically drawn on the basis of whether numerical values are higher or lower than before, and rarely in terms of whether the differences are likely to be functionally significant. For example, measures of soil organic matter are a cornerstone of most sustainability and soil quality assessments, being seen as an integrative indicator for soil properties such as moisture-holding capacity, physical structure, and nutrient supply capacity. However, the numerical level that would be considered good, or what change in soil organic matter levels constitutes a significant functional change, has not been established (Loveland and Webb, 2003). In contrast, nutrient concentration thresholds for ecological integrity have been established for some freshwater lakes (Carpenter and Lathrop, 2008), but no such thresholds for ecological integrity currently exist for nitrate or phosphorus concentrations in freshwater streams. Clearly, improving the understanding of the relationships between sustainability indicators and their functional significance is a priority for future work.

Using even well-designed indicators still begs the question of how to make a holistic assessment of the relative sustainability of different systems given the multiple indicators that represent various sustainability goals and objectives. Even a single sustainability goal, such as enhancing environmental quality, contains many subobjectives, such as water quality, air quality, water use, and biodiversity conservation, each of which may be measured by multiple indicators. One practice is to combine individual indicators into an index, based on some additive (often weighted) procedure. A single index, however, obscures the values inherent in its calculation—which attributes are weighted more than others—and can be particularly problematic if the direction of change is positive for some measures but negative for others (Suter, 1993; Fisher et al., 2001). Despite that issue, reducing multiple indicators into a single indicator is done in some policy contexts, such as the use of an environmental benefits index to implement the Conservation Reserve Program of the Farm Bill.

An alternative is to evaluate system performance without creating a single number, in which case any tradeoffs or synergies might be identified. For example, the same suite of indicators is monitored over time across all countries and regions in EU, making it possible to look at the spatial distribution of relative importance of different sustainability issues and where trends are positive or negative (European Environment Agency, 2006). Data show that soil erosion and water overdrafts are most serious in the Mediterranean countries, whereas nitrate pollution is greatest in northern Europe. However, positive nitrogen balances are declining in most countries where the problems are most severe (for example, the Netherlands) and have decreased by 16 percent across the EU as a whole between 1990 and 2000. Other indicators used include greenhouse-gas emissions, changes in biodiversity, and landscape patterns. (See European Environment Agency [2006] for the complete list.) Those findings are then integrated into the various policy directives and used to evaluate program effectiveness.

Programs on indicators have been developed by other institutions, including the FAO (Tschirley, 1997), United Nations (United Nations, 2007), World Bank (Dumanski et al., 1998), USDA (USDA-ERS, 2003), various university programs in the United States (Aistars, 1999), and such nonprofit organizations as the International Institute of Sustainable Development (Pintér et al., 2005) and the Land Stewardship Project (Keeney and Boody, 2005). Although the proliferation of programs to develop indicators reflects the growing interest in quantification and evaluation of the sustainability of agriculture, there is not an alignment or consensus among the different organizations and scientists involved about the most appropriate or useful set of indicators. In other words, there are different interpretations, methods, and approaches for developing and using indicators. Collaborative efforts among different organizations to develop agreement about key indicators needed for measuring sustainability—and particularly performance outcomes—have emerged in the U.S. agriculture sector, such as Field to Market: The Keystone Alliance for Sustainable Agriculture (Keystone Center, 2009) and the Stewardship Index Initiative for Specialty Crops (Stewardship Index for Specialty Crops, 2009). Some university programs in the United States have also convened and attempted to collaborate in the development of indicators (for example, the Sustainable Agriculture Research and Education Program at University of California).

It should be noted, however, that the conclusions reached by the different methods can vary substantially even when applied to the same system (for example, van der Werf et al., 2007). Those differences emphasize the need for careful assessment of the assumptions behind each method and for more method comparison studies to be done.

• Sustainable agriculture can involve a diverse number of possible farming practices or farming systems. The committee’s definition of sustainable farming does not accept a sharp dichotomy between conventional and sustainable farming systems, not only because farming enterprises reflect many combinations of farming practices, organization forms, and management strategies, but also because all types of systems can potentially contribute to achieving various sustainability goals and objectives.

• Sustainability is a process that moves farming systems along a trajectory toward meeting societal defined goals, as opposed to any particular end state. As such, management systems will inevitably need to be adjusted to meet sustainability goals and objectives when circumstances and societal desires change.

• Sustainability of agriculture is a complex and dynamic concept. The committee rec-

ognizes that the concept is inherently subjective in that the goals will be viewed in different ways by various groups in society. Further, even with broad agreement for certain goals, the relative importance assigned to one goal over another will be highly contested. The definition of societal goals for sustainable agriculture emerges from the domain of politics, not science. Yet, science plays essential roles by generating knowledge to inform the political process, making predictions about outcomes likely to result from different management systems, answering specific questions when needed, and expanding the range of alternatives considered. Science also can supply the knowledge needed to develop new agricultural technologies.

• Finding ways to measure progress along a sustainability trajectory is an important part of an experimentation and adaptive management process. The rationale for selecting the indicators used to measure progress needs to be explicitly stated and justified since the choice is a political rather than a scientific question.

• It is important that indicators used are shown to be appropriate surrogates for the sustainability outcomes they are meant to represent, especially in the case of means-based indicators.

• Social indicators, in addition to environmental and economic measures, will help provide a more comprehensive assessment of movement toward sustainability goals; however, much more research is needed to develop appropriate social indicators because the development of such measures to date has been fledgling.

Using the terms and the boundaries defined in this chapter, this report provides an overview of how U.S. agriculture has evolved over the years to the current state, and farmers’ successes and the challenges they face (Chapter 2). The state of knowledge on farming practices and management systems that can increase the environmental and production (Chapter 3), social, and economic (Chapter 4) sustainability of agriculture are then discussed. However, individual farming practices are components of an agricultural system. Knowledge and understanding of the sum of the parts are important in designing, fine-tuning, and adapting the system to improve sustainability. Chapter 5 uses a few systems to illustrate how a collection of farming practices works in concert to improve sustainability and to illustrate some potential tradeoffs. In spite of the positive attributes of some farming practices and systems that can improve sustainability, they are not necessarily widely adopted by farmers. Chapter 6 highlights the importance of markets, policies, and research in shaping trajectories of farm change, and examines influences on farmers that make some of them more likely and able than others to change production practices or convert to new farming systems. The approaches that could improve environmental, social, and economic sustainability, their practical applications, and their ability to meet those goals are best illustrated in examples of real-life farms in Chapter 7. The committee used seven farms featured in the report Alternative Agriculture (NRC, 1989b) and nine additional farms as case studies to illustrate points made earlier in the report. Conclusions drawn in earlier chapters were used as the basis of the discussion on the lessons learned in U.S. agriculture that are relevant to agriculture in other regions in Chapter 8. A representative of the Gates Foundation specified that the foundation was most interested in the relevance of lessons learned to sub-Saharan Africa, which is the focus of Chapter 8. The committee’s findings and recommendations for how to advance a systems approach to farming and to strengthen federal policies and programs related to improving agricultural production in the United States are summarized in Chapter 9.

Adger, W.N. 2006. Vulnerability. Global Environmental Change 16:268–281.

Aistars, G.A. 1999. A life cycle approach to sustainable agriculture indicators. Available at http://css.snre.umich. edu/css_doc/Proceedings.PDF. Accessed on May 20, 2009.

Atwill, E.R. 2008. Implications of wildlife in E. coli outbreaks associated with leafy green produce. Paper read at 23rd Vertebrate Pest Conference, San Diego.

Batie, S. 2008a. Wicked problems and applied economics. American Journal of Agricultural Economics 9(5): 1176–1191.

Batie, S.S. 2008b. The sustainability of U.S. cropland soils. In Perspectives on Sustainable Resources in America, R.A. Sedjo, ed. Washington, D.C.: RFF Press.

Bell, S., and S. Morse. 2008. Sustainability Indicators: Measuring the Immeasurable. London: Earthscan.

Bockstaller, C., and P. Girardin. 2003. How to validate environmental indicators. Agricultural Systems 76(2): 639–653.

Boody, G., B. Vondracek, D.A. Andow, M. Krinke, J. Westra, J. Zimmerman, and P. Welle. 2005. Multifunctional agriculture in the United States. BioScience 55(1):27–38.

Brown, C., and U. Lall. 2006. Water and economic development: the role of variability and a framework for resilience. Natural Resources Forum 30 (4):306–317.

Carpenter, S.R., and R.C. Lathrop. 2008. Probabilistic estimate of a threshold for eutrophication. Ecosystems 11(4): 601–613.

Chapin, F.S., G.P. Kofinas, and C. Folke (eds.) 2009. Principles of Ecosystem Stewardship: Resilience-based Natural Resource Management in a Changing World. New York: Springer.

Collins, K. 2008. The role of biofuels and other factors in increasing farm and food prices—a review of recent developments with a focus on feed grain markets and market prospects. Available at http://www.foodbeforefuel.org/files/Role%20of%20Biofuels%206-19-08.pdf. Accessed on June 8, 2009.

Conway, G.R. 1987. The properties of agroecosystems. Agricultural Systems 24(2):95–117.

Crumpton, W.G., D.A. Kovacic, D.L. Hey, and J.A. Kostel. 2008. Potential of Restored and Constructed Wetlands to Reduce Nutrient Export from Agricultural Watersheds in the Corn Belt. St. Joseph, Mich.: ASABE.

Dobbs, T.L. 2008. Economic and policy conditions necessary to foster sustainable farming and food systems: U.S. policies and lessons from the European Union. Available at http://www.foodandsocietyfellows.org/publications.cfm?refID=104050. Accessed on August 6, 2009.

Doyle, M.P., and M.C. Erickson. 2008. Summer meeting 2007—the problems with fresh produce: an overview. Journal of Applied Microbiology 105(2):317–330.

Dumanski, J., E. Terry, D. Byerlee, and C. Pieri. 1998. Performance indicators for sustainable agriculture. Available at http://siteresources.worldbank.org/INTARD/864477-1112703179105/20434502/SustInd.pdf. Accessed on May 20, 2009.

Eakin, H., and A.L. Luers. 2006. Assessing the vulnerability of social-environmental systems. Annual Review of Environment and Resources 31:365–394.

Ebi, K.L., R.S. Kovats, and B. Menne. 2006. An approach for assessing human health vulnerability and public health interventions to adapt to climate change. Environmental Health Perspectives 114(12):1930–1934.

European Environment Agency. 2006. Integration of Environment into EU Agricultural Policy—the IRENA Indicator-Based Assessment Report. Copenhagen: European Environment Agency.

Fisher, W.S., L.E. Jackson, G.W. Suter, and P. Bertram. 2001. Indicators for human and ecological risk assessment: a US Environmental Protection Agency perspective. Human and Ecological Risk Assessment 7(5):961–970.

Folke, C. 2006. Resilience: the emergence of a perspective for social-ecological systems analyses. Global Environmental Change—Human and Policy Dimensions 16(3):253–267.

Folke, C., S. Carpenter, T. Elmqvist, L. Gunderson, C.S. Holling and B. Walker. 2002. Resilience and sustainable development: building adaptive capacity in a world of transformations. Ambio 31(5): 437–440

Fowler, H.J., C.G. Kilsby, and P.E. O’Connell. 2003. Modeling the impacts of climatic change and variability on the reliability, resilience, and vulnerability of a water resource system. Water Resources Research 39(8):1222.

Goldstein, B.E. 2009. Resilience to surprises through communicative planning. Ecology and Society 14(2):33. Available at http://www.ecologyandsociety.org/vol14/iss2/art33/. Accessed on March 3, 2010.

Guthman, J. 2004. Agrarian Dreams: The Paradox of Organic Farming in California. Berkeley: University of California Press.

Holling, C.S. 1996. Surprise for science, resilience for ecosystems, and incentives for people. Ecological Applications 6(3):733–735.

———. 2001. Understanding the complexity of economic, ecological, and social systems. Ecosystems 4:390–405.

IOM (Institute of Medicine) and NRC (National Research Council). 2009. The Public Health Effects of Food Deserts. Workshop Summary. Washington, D.C.: National Academies Press.

Janssen, M.A., M.L. Schoon, W.M. Ke, and K. Borner. 2006. Scholarly networks on resilience, vulnerability and adaptation within the human dimensions of global environmental change. Global Environmental Change— Human and Policy Dimensions 16(3):240–252.

Josling, T. 2002. Competing paradigms in the OECD and their impact on the WTO agricultural talks. In Agricultural Policy for the 21st Century, L. Tweeten and S.R. Thompson, eds. Ames: Iowa State Press.

Keeney, D., and G. Boody. 2005. Performance based approaches to agricultural conservation programs dealing with non-point source pollution, including utilization of the provisions of the conservation security program. Available at http://www.landstewardshipproject.org/mba/performance_based_policies.pdf. Accessed on August 10, 2009.

Keystone Center. 2009. Field to market: the Keystone Alliance for Sustainable Agriculture. Available at http://www.keystone.org/spp/environment/sustainability/field-to-market. Accessed on August 10, 2009.

Loveland, P., and J. Webb. 2003. Is there a critical level of organic matter in the agricultural soils of temperate regions: a review. Soil & Tillage Research 70(1):1–18.

Milestad, R., and I. Darnhofer. 2003. Building farm resilience: the prospects and challenges of organic farming. Journal of Sustainable Agriculture 22(3):81–97.

Naylor, R.L. 2008. Managing food production systems for resilience. In Principles of Natural Resource Stewardship: Resilience-Based Management in a Changing World, F.S. Chapin, G.P. Kofinas, and C. Folke, eds. New York: Springer.

Nelson, D.R., W.N. Adger, and K. Brown. 2007. Adaptation to environmental change: contributions of a resilience framework. Annual Review of Environment and Resources 32:395–419.

NRC (National Research Council). 1989. Investing in Research: A Proposal to Strengthen the Agricultural, Food, and Environmental System. Washington, D.C.: National Academy Press.

———. 1989b. Alternative Agriculture. Washington, D.C.: National Academy Press.

———. 1996. Understanding Risk: Informing Decisions in a Democratic Society. Washington, D.C.: National Academy Press.

———. 1999. Our Common Journey: A Transition Toward Sustainability. Washington, D.C.: National Academy Press.

Payraudeau, S., and H.M.G. van der Werf. 2005. Environmental impact assessment for a farming region: a review of methods. Agriculture Ecosystems & Environment 107(1):1–19.

Pielke, R.A., Jr. 2007. The Honest Broker: Making Sense of Science in Policy and Politics. New York: Cambridge University Press.

Pintér, L., P. Hardi, and P. Bartelmus. 2005. Sustainable Development Indicators: Proposals for a Way Forward. Winnipeg: International Institute for Sustainable Development.

Rigby, D., P. Woodhouse, T. Young, and M. Burton. 2001. Constructing a farm level indicator of sustainable agricultural practice. Ecological Economics 39(3):463–478.

Robledo, C., M. Fischler, and A. Patino. 2004. Increasing the resilience of hillside communities in Bolivia—has vulnerability to climate change been reduced as a result of previous sustainable development cooperation? Mountain Research and Development 24(1):14–18.

Santelmann, M.V., D. White, K. Freemark, J.I. Nassauer, J.M. Eilers, K.B. Vache, B.J. Danielson, R.C. Corry, M.E. Clark, S. Polasky, R.M. Cruse, J. Sifneos, H. Rustigian, C. Coiner, J. Wu, and D. Debinski. 2004. Assessing alternative futures for agriculture in Iowa, USA. Landscape Ecology 19(4):357–374.

Searchinger, T., R. Heimlich, R.A. Houghton, F. Dong, A. Elobeid, J. Fabiosa, S. Tokgoz, D. Hayes, and T-H. Yu. 2008. Use of U.S. croplands for biofuels increases greenhouse gases through emissions from land use change. Science 319:1238–1240.

Stewardship Index for Specialty Crops. 2009. Welcome to the homepage of the Stewardship Index for Specialty Crops. Available at http://www.stewardshipindex.org/. Accessed on August 10, 2009.

Suter, G.W. 1993. A critique of ecosystem health concepts and indexes. Environmental Toxicology and Chemistry 12(9):1533–1539.

Thompson, P.B. 2010. The Agrarian Vision: Sustainability and Environmental Ethics. Lexington: University of Kentucky Press.

Tokgoz, S., A. Elobeid, J. Fabiosa, D.J. Hayes, B.A. Babcock, T.H. Yu, F.X. Dong, and C.E. Hart. 2008. Bottlenecks, drought, and oil price spikes: impact on U.S. ethanol and agricultural sectors. Review of Agricultural Economics 30(4):604–622.

Tschirley, J.B. 1997. The use of indicators in sustainable agriculture and rural development: considerations for developing countries. In Sustainability Indicators, B. Moldan and S. Billharz, eds. New York: John Wiley and Sons.

Turner, B.L., R.E. Kasperson, P.A. Matson, J.J. McCarthy, R.W. Corell, L. Christensen, N. Eckley, J.X. Kasperson, A. Luers, M.L. Martello, C. Polsky, A. Pulsipher, and A. Schiller. 2003. A framework for vulnerability analysis in sustainability science. Proceedings of the National Academy of Sciences of the United States of America 100(14):8074–8079.

United Nations. 2007. Indicators of sustainable development: guidelines and methodologies. Available at http://www.un.org/esa/sustdev/natlinfo/indicators/guidelines.pdf. Accessed on May 20, 2009.

USDA-CSREES (U.S. Department of Agriculture Cooperative State Research, Extension, and Education Service). 2007. Agricultural systems overview. Available at http://www.csrees.usda.gov/nea/ag_systems/ag_systems.cfm. Accessed on October 29, 2008.

USDA-ERS (U.S. Department of Agriculture Economic Research Service). 2003. Agricultural resources and environmental indicators, 2003. Available at http://www.ers.usda.gov/publications/arei/ah722/dbgen.htm. Accessed on May 20, 2009.

———. 2008. Farm income: partitioning the definition of a farm. Available at http://www.ers.usda.gov/data/farmincome/Sizedefinition.htm. Accessed on September 24, 2009.

USDA-NAL (U.S. Department of Agriculture National Agricultural Library). 2007. Sustainable agriculture: definitions and terms. Available at http://www.nal.usda.gov/afsic/pubs/terms/srb9902.shtml. Accessed on August 10, 2009.

USDA-NRCS (U.S. Department of Agriculture National Resource Conservation Service). 2009. National conservation practice standards. Available at http://www.nrcs.usda.gov/technical/standards/nhcp.html. Accessed on September 10, 2009.

Van Cauwenbergh, N., K. Biala, C. Bielders, V. Brouckaert, L. Franchois, V.G. Cidad, M. Hermy, E. Mathijs, B. Muys, J. Reijnders, X. Sauvenier, J. Valckx, M. Vanclooster, B. Van der Veken, E. Wauters, and A. Peeters. 2007. SAFE—a hierarchical framework for assessing the sustainability of agricultural systems. Agriculture Ecosystems & Environment 120(2–4):229–242.

van der Werf, H.M.G., J. Tzilivakis, K. Lewis, and C. Basset-Mens. 2007. Environmental impacts of farm scenarios according to five assessment methods. Agriculture Ecosystems & Environment 118(1–4):327–338.

Walker, B., and D. Salt. 2006. Resilience Thinking. Washington, D.C.: Island Press.

Zhen, L., and J.K. Routray. 2003. Operational indicators for measuring agricultural sustainability in developing countries. Environmental Management 32(1):34–46.

This page intentionally left blank.