Concept of a Foodshed and Foodshed Analysis

Analogous to a watershed, the concept of a foodshed has been presented both as a tool for understanding the flow of food in the food system and as a framework for envisioning alternative food systems. In what may be the original use of the term, Walter Hedden described a ‘foodshed’ in 1929 as the ‘dikes and dams’ guiding the flow of food from producer to consumerReference Hedden³. Contemporary authors have used the term to refer to this geographic idea of a foodshedReference Kloppenburg, Hendrickson and Stevenson^4, Reference Stagl⁵. However, the term has also been used to describe the components of an alternative food system that connects local producers and consumersReference Kloppenburg, Hendrickson and Stevenson^4, Reference Kloppenburg, Hendrickson and Stevenson^6–Reference Feagan⁸. Since the term lacks consistent definition, we provide some history of its prior use and clarify its meaning in the context of foodshed analysis.

Hedden's book, How Great Cities Are Fed, describes the economic forces that influence where foods are produced and how they are transported to the cities in which they are consumed. His work focused on New York City and was prompted by the threat of a halt in nationwide railroad transportation in October 1921 when ‘it immediately became apparent that there was a dire lack of dependable information regarding the city's food needs, the sources from which they were supplied, and the manner in which these supplies were transported and handled’Reference Hedden³. The fact that 8 million people were fed without anyone understanding the whole system highlights both the power of the marketplace to meet human demands and the peril of taking its function for granted until it is in danger. Hedden does not identify the specific threat, but he likely refers to a scheduled 30 October 1921 strike of the nation's five largest railroad worker unions (railroad conductors, locomotive engineers, switchmen, railway trainmen and locomotive firemen) that would have affected 42 states had it not been called off 3 days before the strike deadline^{9, 10}. Just as this transportation crisis passed, the term ‘foodshed’ dropped into obscurity.

Sixty years later, Arthur Getz reintroduced the concept of a foodshed to provide an image that helps people to understand how food systems work and that suggests food comes from a source that must be protectedReference Getz¹¹. Getz found this image useful for envisioning how agriculture could thrive in low-density suburban and ex-urban areas by targeting consumers in metropolitan areas. Building on this local theme, Kloppenburg et al. used the term ‘foodshed’ to represent a more locally reliant, alternative food system that reduces the negative social and environmental impacts of agricultureReference Kloppenburg, Hendrickson and Stevenson⁴. Thus, the term often connotes a connection to local food systems even though its original use refers to the food system in general.

Though the literature lacks a precise definition, we adopt the term ‘foodshed’ here because its parent concept, the watershed, is so widely understood. However, we wish to clarify our use of the term. For the purposes of this paper, a ‘foodshed’ is the geographic area from which a population derives its food supply. As such, ‘foodshed analysis’ refers to study of the actual or potential sources of food for a population, particularly those factors influencing the movement of food from its origin as agricultural commodities on a farm to its destination as food wherever it is consumed. Such an analysis would be immediately relevant to the on-going debate about local food but could ultimately be applied to larger questions of food system sustainability.

Local Food and Sustainability

A growing body of literature connects local food to the larger goal of sustainable development. Advocates claim local food systems offer an array of economic, environmental and social benefits, but the evidence underlying these assertions is being challenged. Vigorous debate of these issues has entered the public discourse on food, as evidenced by recent articles in major popular press publications in the USReference Cloud¹² and the UK¹³ and the selection of ‘locavore’ as the Oxford word of the year for 2007¹⁴. To understand the significance of local food to sustainability, the subject and the surrounding debate must be examined more closely.

The term ‘local food’ evades easy definition. In part, it is a geographical concept referring to the distance between food producers and consumers. For example, in a recent survey of US consumers, most respondents defined ‘local’ as produced within 100 miles or within their home state¹⁵. While several authors claim that no consistent definition of ‘local’ existsReference Bellows and Hamm^16, Reference Edwards-Jones, Milá i Canals, Hounsome, Truniger, Koerber, Hounsome, Cross, York, Hospido, Plassman, Harris, Edwards, Day, Tomos, Cowell and Jones¹⁷, terms like ‘local food’, ‘local food system’ and ‘(re)localization’ are used almost interchangeably to refer to the concept of increasing reliance on foods produced near their point of consumption relative to the modern food system. In addition to this geographical meaning, ‘local food’ is also a political concept. This second construction refers to an alternative system of food production that addresses the perceived ills of the modern food system. It has been described as ‘a banner under which people attempt to counteract trends of economic concentration, social disempowerment, and environmental degradation in the food and agricultural landscape’Reference Hinrichs¹⁸. This connection between local and the creation of a more sustainable and just food system has been traced to seminal writings of Wendell Berry, Joan Gussow, Jim Hightower, and Frances Morre Lappé from the 1970sReference Feenstra¹⁹.

Given the breadth of the second definition, it is not surprising that a wide range of benefits have been attributed to ‘local food’. These purported benefits encompass all three dimensions of sustainability: ecological, economic and social. Some of the advantages of local food arise from the physical proximity of producers and consumers, such as reducing the amount of energy used in the transport of foodsReference Kloppenburg, Hendrickson and Stevenson^4, Reference Halweil^7, Reference Gussow and Clancy^20, Reference Gussow²¹ and the associated greenhouse gas emissionsReference Pirog, Van Pelt, Enshayan and Cook²². Similarly, local foods are purported to be better tasting and perhaps more nutritious than foods bred and picked for their ability to endure long-distance shippingReference Kloppenburg, Hendrickson and Stevenson^4, Reference Lapping²³. Others benefits are attributed to a combination of shorter supply chains and the relationships forged between producers and consumers, such as improving the economic viability of local farms and their communitiesReference Halweil^7, Reference Gussow^21, Reference Lyson and Green²⁴, increasing public awareness of issues related to the food systemReference Gussow²¹, improved environmental stewardship by producersReference Lyson and Green²⁴ and greater public control over the food systemReference Halweil^7, Reference Lyson and Green²⁴. Finally, it has been posited that a more local food system would decrease food safety risks by decentralizing food productionReference Halweil^7, Reference Gussow²¹. Taken together, these claims suggest that ‘localization’ is a vital component of a transition to a more sustainable and more just food system.

Of course, the validity of these claims has become a matter of debate. Within the scholarly literature, authors caution that there are potential risks (along with benefits) to localizing food systemsReference Bellows and Hamm¹⁶ and that local food systems are not inherently more environmentally sustainable or socially just than the global food systemReference Hinrichs^18, Reference Born and Purcell²⁵. Nonetheless, the weight of the evidence suggests to us that the local food movement is generally viewed in a positive light. Articles in the popular press raise broader issues such as the efficacy of consumer's attempts to promote change through their supermarket purchases¹³ and the merits of organic relative to local foodReference Cloud¹². In response to growing public interest, Edwards-Jones et al. examined evidence to see if local food is best, principally in terms of its greenhouse gas emissionsReference Edwards-Jones, Milá i Canals, Hounsome, Truniger, Koerber, Hounsome, Cross, York, Hospido, Plassman, Harris, Edwards, Day, Tomos, Cowell and Jones¹⁷. They concluded that because of the dearth of studies which examine greenhouse gas emissions across the entire food system, it is not possible to answer the question conclusively.

Given the contradictory nature of the available evidence, the prevalent claims from local food system proponents should not simply be dismissed. Rather, closer scrutiny is probably in order. More importantly, debate over whether or not ‘localization’ is desirable or if it is a first principle of sustainability misses a broader point. An interesting question is how might pressing issues of sustainability force the food system to become more local? To this end, several issues stand out because of their urgency and their potential to influence society's continued dependence on long-distance transport of food: namely, the challenges of climate change, petroleum depletion, and bio-energy production beg that this question be answered.

The Climate–energy Puzzle

There is growing consensus within the scientific community that climate change is a threat to sustainability that requires action in the near term. The warming of the planet's climate has been deemed ‘unequivocal’ by the most recent synthesis report of the Intergovernmental Panel on Climate Change (IPCC), and the available evidence strongly supports the assertion that this change in climate has anthropogenic origins²⁶. In addition, while there is no clear agreement on how much warming could be tolerated by the Earth's ecosystems or human society, immediate action seems warranted. The IPCC clearly states that mitigation of emissions can help reduce, delay or avoid many of the negative impacts of climate change and that the risks of severe negative consequences increase the longer society waits to take action²⁶. Meanwhile, the Scientific Expert Group on Climate Change and Sustainable Development (SEG) boldly asserts that collective action to address climate change is needed now²⁷, echoing a sense of urgency which appears to be shared by many scientistsReference Kerr²⁸.

Expert bodies have concluded that since some warming is inevitable, both adaptation to and mitigation of climate change are necessary^{27, 29}. In addition, since no one sector alone can achieve the level of mitigation required³⁰, all emissions sectors (agriculture, buildings, energy, forestry and land use change, industry, transport and waste) must contribute to the effort. Since the food system crosses multiple emissions sectors, it is not easy to estimate its total impact. Agricultural production alone contributes 14% of anthropogenic greenhouse gas emissions²⁶. However, according to a broader, multi-sector analysis of livestock production, the entire livestock production cycle (from emissions associated with clearing land for crop production or pasture to emissions from processing and transporting livestock products) accounts for 18% of total emissionsReference Steinfeld, Gerber, Wassenaar, Castel, Rosales and de Haan³¹. Thus, the food system is an important source of emissions, and it is reasonable to evaluate how its impact will be mitigated.

Intimately related to climate change, a second global challenge to sustainability is the depletion of fossil fuel resources amidst rising energy demand. Global hydrocarbon (oil and natural gas) production has been forecast to peak and decline in the 21st centuryReference Edwards^32, Reference Bentley³³. Since oil currently supplies about 35% of the world's energy³⁴, its depletion poses particular concern. According to a recent report by the US Government Accountability Office, most studies estimate that the peak in petroleum production has already occurred or that it will occur by 2040³⁵. This is significant because the peak marks the transition from oil being a plentiful, relatively cheap resource to being an increasingly scarce and expensive resourceReference Duncan and Youngquist³⁶. Indeed, petroleum production capacity has just been able to keep pace with demand in recent years³⁵, and rising crude oil prices have surpassed records in both nominal³⁵ and real (inflation adjusted) termsReference Mouawad³⁷. Whether or not these price increases indicate that society is near the peak remains to be seen, but they have forced energy back into the public consciousness.

How important is this transition? Several authors believe the consequences are serious enough to warrant close attention, concluding that failure to adequately prepare for the peak will lead to serious economic disruptionReference Bentley^{33, 35,} Reference Hallock, Tharakan, Hall, Jefferson and Wu³⁸ or that reductions in population may be necessary to support people at an acceptable standard of livingReference Youngquist³⁹. Others argue that the world can extend the use of petroleum by tapping unconventional sources such as tar sands and oil shaleReference Greene, Hopson and Li^40, Reference Brandt and Farrell⁴¹. However, such resources are more energy intensive to extract and could significantly increase greenhouse gas emissionsReference Brandt and Farrell⁴¹. In either case, the energy constraints or potential climate impacts present a clear challenge to sustainability. Moreover, uncertainty about the quantity of reserves remaining and political instability in major oil-producing countries mean production is vulnerable to sudden disruptions³⁵. Since the transportation sector is almost completely reliant on oil³⁵, it is reasonable to question how movement of foodstuffs will be affected by long-term depletion or sudden disruptions in oil production.

In response to growing concerns about climate change and energy security, biofuels are being sought as a partial solution to both problems. However, while biofuels are a renewable energy source they face several major constraints: the scale of potential production relative to current energy demand, competition with food production and the carbon emissions of land conversion. With regard to scale, it is illustrative to examine the production of ethanol in Brazil and the US, who together produce 70% of the world's ethanolReference Seelke and Yacobucci⁴². For example, were the entire US corn (Zea mays L.) and soybean (Glycine max (L.) Merr.) crops used to produce ethanol and biodiesel, just 12% of gasoline demand and 6% of diesel demand could be satisfiedReference Hill, Nelson, Tilman, Polasky and Tiffany⁴³. In contrast, the sugar cane (Saccharum officinarum L.)-based ethanol industry in Brazil supplied 39% of the country's gasoline demand (based on energy content) with ethanol in 2004Reference Seelke and Yacobucci⁴² even though just 10% of its cropland is devoted to growing sugar caneReference Goldemberg⁴⁴. However, this comparison is unbalanced because Brazil consumed just 8.4 billion gallons of gasoline in 2004 relative to the 142 billion gallons consumed in the USReference Seelke and Yacobucci⁴². Moving beyond these two countries, Giampietro et al. estimated the land needed for 21 countries (including Brazil and the US) to supply both food and energy from agricultural land, assuming that energy needs in temperate regions were supplied from corn and sorghum (Sorghum bicolor (L.) Moench) ethanol and in tropical regions were supplied from sugar-cane ethanolReference Giampietro, Ulgiati and Pimentel⁴⁵. None of the 21 countries included in the analysis had enough arable land to meet both food and energy needs, though Brazil had a more favorable ratio of land needed to land available than the US (3.0 versus 14.6) because of its lower per capita energy consumption and reliance on the more efficient sugar cane-based ethanol systemReference Giampietro, Ulgiati and Pimentel⁴⁵. On balance, ‘first generation’ biofuels (fuels produced from grains, oilseeds and sugar crops) appear to have limited potential to replace society's current use of fossil fuels.

Even ‘second generation’ biofuels, which would produce fuel from cellulose, beg the same question about scale. Studies of the potential energy contribution from global biomass production vary widely, but generally suggest that biomass could provide between 100 and 400 exajoules (EJ) of energy by 2050Reference Berndes, Hoogwijk and van den Broek⁴⁶. For comparison, global energy consumption in 2004 was 472 EJ⁴⁷. On the surface, this seems promising. However, the total amount of biomass needed to achieve this level of energy output is comparable with the levels already harvested in the agricultural systemReference Haberl, Erb, Krausmann, Gaube, Bondeau, Plutzar, Gingrich, Lucht and Fischer-Kowalski^48, Reference Field, Campbell and Lobell⁴⁹. Thus, both the ecological and socio-economic consequences of such large-scale production are causes for concernReference Field, Campbell and Lobell⁴⁹.

One such ecological concern is the emission of carbon dioxide from conversion of land to agricultural production. Recent studies have shown that although production of biofuel feedstock removes carbon dioxide from the atmosphere and thus displaces fossil fuel emissions, the net climate impact is greatly influenced by land use changeReference Fargione, Hill, Tilman, Polasky and Hawthorne^50, Reference Searchinger, Heimlich, Houghton, Dong, Elobeid, Fabiosa, Tokgoz, Hayes and Yu⁵¹. The term ‘carbon debt’ has been coined to describe the period of time that would be required for a biofuel system to mitigate the emissions caused by converting land to active agricultural useReference Fargione, Hill, Tilman, Polasky and Hawthorne⁵⁰. The size of the carbon debt varies depending on the previous use of the land converted to crop production and the system of biofuel in question. For example, the ‘payback’ period ranges from 17 years for ethanol derived from sugar cane (S. officinarum L.) planted on former savannah to 420 years for biodiesel from oil palm (Elaeis guineensis Jacq.) planted on peatland rainforestReference Fargione, Hill, Tilman, Polasky and Hawthorne⁵⁰. Moreover, even if biofuel production itself occurs on active agricultural land, it may push food production onto new land or land that had been abandoned for agricultural use. As a result, even cellulosic ethanol production from perennial crops such as switchgrass (Panicum virgatum L.) can indirectly cause land conversion that result in net carbon emissions for decadesReference Searchinger, Heimlich, Houghton, Dong, Elobeid, Fabiosa, Tokgoz, Hayes and Yu⁵¹. Thus, it has been argued that from the standpoint of climate emissions, biofuel production should be derived from waste productsReference Fargione, Hill, Tilman, Polasky and Hawthorne^50, Reference Searchinger, Heimlich, Houghton, Dong, Elobeid, Fabiosa, Tokgoz, Hayes and Yu⁵¹ or from feedstocks produced on abandoned or marginal landsReference Field, Campbell and Lobell^49, Reference Fargione, Hill, Tilman, Polasky and Hawthorne⁵⁰.

In light of these findings, it seems fair to conclude that society cannot simply ‘grow’ its way out of the climate change and energy security problems. Nonconventional sources of petroleum can be used as substitutes for conventional oil, but are likely to exacerbate climate emissions. Biofuels are more promising, but face limits in their potential scale and will have to be developed carefully to avoid unintentionally increasing emissions. Thus, more attention must be paid to possibilities for reducing emissions and demand for energy through more efficient transport and different patterns of consumption. To this end, evaluating the potential for strategies, such as increased reliance on local and regional food production, to reduce energy use and emissions is worthwhile.

The World Food Situation

The urgency of these converging problems becomes most apparent when one examines their effect on what has traditionally been the principal goal of agriculture, food production. In the late 1990s, rising incomes and associated dietary shifts were projected to boost consumption of livestock products in the developing world by 2020 and necessitate a significant increase in the intensity of agricultural productionReference Delgado, Rosegrant, Steinfeld, Ehui and Courbois¹. Nonetheless, it was expected that supply would be able to keep pace with demand over this intervalReference Rosegrant and Ringler^52–Reference Johnson⁵⁴. While the environmental implications of increased livestock production were a concern, this transformation held promise for improving the incomes of small farmers and adding diversity to the diets of people in the developing worldReference Delgado, Rosegrant, Steinfeld, Ehui and Courbois¹.

Recently, assessments of the world food situation have become less sanguine. Global food prices, in real terms, have increased by an average of 15% annually between 2006 and 2008, relative to a modest rate of 1.3% between 2000 and 2005⁵⁵. Poor grain harvests due to droughts, rising oil prices and increasing demand for biofuels, in addition to the more familiar drivers of rising incomes and population growth, have all been implicated in the rising commodity prices that have caused these increases in world food pricesReference von Braun^{2, 56}. Prior to the recent spike in food prices, food security was already a long-standing concern. Approximately 840 million people suffer from chronic hunger and more than 2 billion suffer from micronutrient deficiencies or ‘hidden hunger’Reference Kennedy, Nantel and Shetty⁵⁷. Increases in food prices threaten to reduce the purchasing power of household incomes, pushing more people into deprivation. If this is an indication of how responsive our food system is to climate perturbations and competition between food and energy, then society has cause for alarm.

In this context, a major challenge facing agriculture and the food system in this century will be trying to improve food security and human nutrition while using less fossil energy and reducing its greenhouse gas emissions. Given the tension between agriculture as a source of food and fuel, changing consumption patterns may be essential to achieving these goals. Though lifestyle changes may be difficult to initiate, it is possible to reduce demand for food and feed commodities without sacrificing dietary qualityReference Bender⁵⁸. The most obvious example is reducing excess consumption of calories, which could address both environmental and resource issues as well as health problems associated with obesity. Another example would be substituting plant protein sources for livestock sources, which can reduce the land requirements for growing feed crops while still supplying adequate nutrition. Thus, it is worth investigating how efforts to modify diets might contribute to solving the food–climate–energy puzzle by reducing the demand for the foods that occupy the most extensive areas of land, require the greatest energy inputs or cause the largest emissions of greenhouse gases.

To this end, would shifts to diets based on more local foods reduce energy use or climate forcing emissions? A recent review of the assertion that local food systems emit less greenhouse gas emissions concluded that too few life cycle assessments of food system emissions exist at present to answer the questionReference Edwards-Jones, Milá i Canals, Hounsome, Truniger, Koerber, Hounsome, Cross, York, Hospido, Plassman, Harris, Edwards, Day, Tomos, Cowell and Jones¹⁷. Thus, it would appear further analysis is warranted. Moreover, the concept of carbon debt shows that where something is grown is as important as how it is grown. Thus, facing these problems will require us to reconsider not only how food is produced but where it is produced.

Foodshed analysis may provide valuable insights into such questions. While attempts have been made to quantify the energy use or greenhouse gas emissions associated with each stage of the food system, the framework of foodshed analysis is unique because it considers geography in two distinct ways. First, a foodshed analysis would entail tracing the flow of food from its origin as an agricultural commodity on a farm to its ultimate point of consumption. Secondly, it would also measure different ‘costs’ of producing and transporting the products through the system, such as energy consumed, greenhouse gases emitted, or prices paid, not only at each stage in the food system, but for different locations. Such a framework would be valuable for evaluating how the geography of the food system influences its impact on the environment and the vulnerability of populations to disruptions in their food supplies. Moreover, foodshed analysis would help to plan how the geography of food systems should change to enhance sustainability. Many variations on this theme are possible. One example, which is explored in an upcoming paper, is estimating the capacity for population centers to supply more of their food from local sourcesReference Peters, Bills, Lembo, Wilkins and Fick⁵⁹.