Introduction
Concerns about environmental, economic and social impacts of the modern, global food system, and the implications of global changes on food security and sustainability have promulgated considerable investigation of, and action toward the localization or regionalization of food systems (e.g., Peters et al., Reference Peters, Bills, Lembo, Wilkins and Fick2009; Edward Jones Reference Edwards-Jones2010; Harris et al., Reference Harris, Nixon, Newman and Mullinix2016; Mullinix et al., Reference Mullinix, Dorward, Sussmann, Polasub, Smukler, Chiu, Rallings, Feeney and Kissinger2016). A significant motivation for the localization of food systems is a potential environmental benefit and in particular the notion that production at the local scale is more sustainable from an environmental stewardship perspective.
Advocates of food system localization argue that reducing ‘food miles’ can minimize energy use, pollution and greenhouse gas emissions (GHG). Food systems researchers also have posited such outcomes as a potential benefit of localization (Lea Reference Lea2005; Selfa and Qazi, Reference Selfa and Qazi2005; Vogt and Kaiser, Reference Vogt and Kaiser2008; Brown, Reference Brown2012). However, beyond energy use and GHG emissions, no other biophysical impacts of local and globalized food systems have been compared; water and land area, for example. Furthermore, even within the food-related GHG emissions studies, the advantage of food system localization is not clear. Some studies emphasize that while local food systems may contribute to reducing the distance food travels, and achieve socio-economic and other environmental outcomes they do not necessarily reduce GHG or otherwise contribute to mitigation of global warming (Milài Canals et al., Reference Milài Canals, Cowell, Sim and Basson2007; Weber and Matthews, Reference Weber and Matthews2008; Coley et al., Reference Coley, Howard and Winter2009; Pelletier et al., Reference Pelletier, Audsley, Brodt, Garnett, Henriksson, Kendall, Kramer, Murphy, Nemecek and Troell2011; Web et al., Reference Webb, Williams, Hope, Evans and Moorhouse2013).
It follows that the local versus global debate must be informed by more comprehensive analysis and comparison. Understanding positive and negative attributes of each and trade-offs resulting from pursuit of either will support meaningful discussion regarding the desired structure and sustainability of the food system. The discourse regarding local food systems is robust in British Columbia, Canada (BC), which is the broad geographic focus of this manuscript. BC's agriculture sector produces a wide range of primary agricultural products and generates over US$2.6 billion yr−1 in farm cash receipts on <3% of the province's land base (BC Ministry of Agriculture Statistics and Research, 2013). This province is one of Canada's leading tree and small fruit, dairy, poultry and egg producers (BC Ministry of Agriculture Statistics and Research, 2013).
In a way, BC has provided the world with two concepts relevant to this study. One is the ‘100 mile diet’ (Smith and MacKinnon, Reference Smith and MacKinnon2007), the idea of living exclusively on food that originates within one's local region The book by that name was written by two BC residents who experienced a year of local diet and inspired others in different parts of the world to do the same, and champion local food systems. The second concept is Ecological Footprint Analysis (EFA), which was initiated and developed by the University of BC researchers William Rees (1992) and Mathis Wackernagel (1996). EFA illuminates humanity's dependence on the earth's limited bioproductive land base. Rees (Reference Rees1997) first calculations of the footprint focused on the Southwest British Columbia (SWBC) lower mainland estimating the area of land required to support the food demands of that region's population. The study showed that the area of land required to support that demand was significantly greater than the available productive land of the region. Since then several studies examined the land used by different populated, urban areas in different parts of the world (Wackernagel, Reference Wackernagel1998; Barrett et al., Reference Barrett, Vallack, Jones and Haq2002; Moore et al., Reference Moore, Kissinger and Rees2013). While all concluded that the area of land required to supply consumption demands (of food and other materials) exceeded the area available locally only a few have attempted to consider the share of local resources used versus imported ones (Kissinger and Haim, Reference Kissinger and Haim2008; Stossel et al., Reference Stossel, Kissinger and Meir2014).
Our analysis aims to contribute to the local food system discourse by using the footprint family set of indicators—land, water, carbon and ecological footprints, to undertake a biophysical comparison of local and global food systems as they operate in the SWBC bioregion. It is one of the first studies to measure multiple biophysical impacts of a local food system, and the first footprint family analysis to focus on the bioregional scale. Such analysis can stimulate a discussion on the extent to which local production can be more sustainable, and it can be the base for future modeling to optimize local supply (i.e., maximizing local food throughput while minimizing its biophysical impacts). We utilized data from a recent food system design project which focused on the SWBC bioregion (Mullinix et al., Reference Mullinix, Dorward, Sussmann, Polasub, Smukler, Chiu, Rallings, Feeney and Kissinger2016). This bioregion produces the vast majority of BC's agriculture output and is home to more than half of its population. First, we present an analysis of the bioregion's overall footprint (by family). Our analysis focuses on a single year (2011, the most current agriculture census year at the time of analysis) and reveals gaps between local food demand and local food supply. In an effort to identify potential biophysical benefits of local production, we then present and compare the footprint of specific agricultural commodities produced locally to that of the same commodity imported to the region.
Background
Local and regional food systems
To build resilient, sustainable food systems, and communities that can navigate the uncertainties of climate change and post-carbon economies, it may be most practical and prudent to develop food security strategies linked to localized food systems (Heinberg, Reference Heinberg2003; Greer, Reference Greer2009; Moreau et al., Reference Moreau, Moore and Mullinix2012; Ackerman-Leist, Reference Ackerman-Leist2013). There is emerging recognition that sustainable food system planning may appropriately be approached locally or regionally (Eaton, et al., Reference Eaton, Hammond and Laurie2007). Around the globe, governments, at all levels and communities are investigating and investing in local–regional food system strategies and action to directly address identified sustainability issues and food security concern (Getz, Reference Getz1991; Peters et al., Reference Peters, Bills, Lembo, Wilkins and Fick2009a; Colasanti and Hamm, Reference Colasanti and Hamm2010; Metcalf and Widener, Reference Metcalf and Widener2011; Horst and Gaolach, Reference Horst and Gaolach2015). BC, Canada is no exception (British Columbia Ministry of Agriculture and Lands, 2006; Smith and MacKinnon, Reference Smith and MacKinnon2007).
The local–regional food system scale is arbitrary, commonly adhering to some geopolitical demarcation. Kloppenburg et al. (Reference Kloppenburg, Hendrickson and Stevenson1996) for example suggested the ‘foodshed’, defined by the extent of their associated region, by political boundaries, or by a predetermined radial distance around a metropolitan area, as an appropriate unit of food system study and planning. This conceptual framework, resulting in variable scales, has been used extensively in food system research and planning (Getz, Reference Getz1991; Kloppenburg et al., Reference Kloppenburg, Lezberg, De Master, Stevenson and Hendrickson2000; Peters et al., Reference Peters, Bills, Lembo, Wilkins and Fick2009a, Reference Peters, Bills, Wilkins and Fickb; Peters et al., Reference Peters, Bills, Lembo, Wilkins and Fick2012) even while making comparative analysis difficult.
The SWBC bioregion
Harris et al. (Reference Harris, Nixon, Newman and Mullinix2016) used a method that incorporated population centers and regional district boundaries, terrestrial and marine ecoregions, and regional watershed boundaries to delineate the SWBC bioregion. It is a 41,380 km2 area in the southwest mainland corner of the province of BC, Canada. The area is both a major Canadian agricultural as well as urban and suburban center. Metro Vancouver alone is home to more than half of BC's total population (almost 2.7 million in 2011), and is one of the fastest growing metropolitan areas in Canada (Statistics Canada, 2014). The majority of agricultural land in SWBC is protected by the Agricultural Land Reserve (ALR), a provincially legislated zone in which agriculture is recognized as the priority use, farming is encouraged and non-agricultural uses are controlled Government of British Columbia – Ministry of Agriculture, 2014). In 2011, SWBC had almost 1500 km2 of ALR land (Government of British Columbia – Ministry of Agriculture, 2011).
The SWBC bioregional food system project
Predicated upon the idea that a bioregional framework may facilitate the achievement of major food system sustainability goals, a multi-disciplinary food system design and planning project was initiated to explore and elucidate the economic, environmental stewardship and food self-reliance potentials of a SWBC bioregional food system, by the Institute for Sustainable Food Systems, Kwantlen Polytechnic University (Mullinix et al., Reference Mullinix, Dorward, Sussmann, Polasub, Smukler, Chiu, Rallings, Feeney and Kissinger2016). The bioregion is a highly productive and important Canadian agriculture area, and a place similar to other North American jurisdictions where agricultural and food system capacity is severely threatened by urban and industrial-neoliberal economic interests. Specific objectives of the project included: (1) an estimate of bioregional food self-reliance potential; (2) estimates of income generation, job creation and other economic outcomes; (3) determining the potential to reduce GHG; (4) determining the ecological footprint (EF) of the bioregion's food system; (5) calculating the impact of balancing nitrogen and phosphorous generation from manure with crop need; and (6) calculating the impact of integrating ecologically beneficial farmscape features—hedgerows and riparian buffers throughout the bioregion. Some initial results of the project from which this paper draws have been published in Dorward et al. (Reference Dorward, Smukler and Mullinix2016); Harris et al. (Reference Harris, Nixon, Newman and Mullinix2016); and Mullinix et al. (Reference Mullinix, Dorward, Sussmann, Polasub, Smukler, Chiu, Rallings, Feeney and Kissinger2016).
Measuring biophysical resources using the footprint family methodology
The footprint family has been described as: ‘a set of indicators, characterized by a consumption-based perspective, able to track human pressures on the surrounding environment, where pressure is defined as appropriation of biological natural resources and carbon dioxide (CO2) uptake, emission of GHG's, and consumption and pollution of global freshwater resources’ (Galli et al., Reference Galli, Wiedmann, Ercin, Knoblauch, Ewing and Giljum2012; 103). It can be used to identify and assess environmental loads associated with a process, product or system, and allows for examination of potential biophysical tradeoffs from proposed policy or other measures (Giljum et al., Reference Giljum, Burger, Hinterberger, Lutter and Bruckner2011; Galli et al., Reference Galli, Wiedmann, Ercin, Knoblauch, Ewing and Giljum2012; Steen-Olsen et al., Reference Steen-Olsen, Weinzettel, Cranston, Ercin and Hertwich2012).
Despite recent acknowledgment of the advantages of using the footprint family, most empirical studies have used a single type of footprint accounting. Only a few have attempted to integrate more than a single indicator (for a comprehensive list of studies see Fang et al., Reference Fang, Heijungs and de Snoo2014). Further, while all footprint studies acknowledge any entity's dependence and impact on both local and global environments, only a few studies have separated the footprint into domestic and global components (e.g., Steen-Olsen et al., Reference Steen-Olsen, Weinzettel, Cranston, Ercin and Hertwich2012; Kastner et al., Reference Kastner, Erb and Haberl2014; Stossel et al., Reference Stossel, Kissinger and Meir2014; Kissinger and Dickler Reference Kissinger and Dickler2016). Similarly, a few have traced the footprint to specific external geographic locations (e.g., Kissinger and Rees Reference Kissinger and Rees2009; Steen-Olsen et al., Reference Steen-Olsen, Weinzettel, Cranston, Ercin and Hertwich2012; Kissinger and Dickler Reference Kissinger and Dickler2016), or quantified the footprint of individual life cycle stages along the commodity chain of a product or entity (e.g., production and shipping). In the current footprint literature, the indicators ranked as most important include ecological, carbon and water footprints (Galli et al., Reference Galli, Wiedmann, Ercin, Knoblauch, Ewing and Giljum2012; Fang et al., Reference Fang, Heijungs and de Snoo2014). Other suggestions for a suite of footprint indicators have identified a set of four: material, land, water and carbon footprints (CF) (European Commission, 2011; Tukker et al., Reference Tukker, Bulavskaya, Giljum, de Koning, Lutter, Simas, Stadler and Wood2014).
Methods
Analysis of the footprint family indicators: Land, water, carbon and ecological footprints per unit of local and of imported agricultural products required integration of several kinds of data from different sources. In the following, we describe key data sources and present the calculation procedure for each biophysical indicator. A supplementary materials file with further data used is also attached to this paper.
Data on food demand are from Dorward et al. (Reference Dorward, Smukler and Mullinix2016) who calculated the annual food demand of the bioregion's residents. Food demand in the bioregion comprised 59 agricultural commodities (see Table 1 in the supplementary materials file for the full list of agricultural commodities). For each commodity, we ascribed a portion to local and to imported products. The study compared the biophysical demands of each commodity grown in the SWBC bioregion to demands of the same commodities imported to the bioregion in the study year. Of the commodities examined, 12 were not grown locally for climatic, (e.g., tropical and subtropical fruits) or resource limitation (e.g., available land area for grain) reasons. This research did not intend or attempt to evaluate the capacity of the region to supply local food demands by local sources but rather present a biophysical snapshot of the current system and compare the biophysical ‘performances’ of the same commodities grown locally to those imported.
Bioregional food crop production: Tons produced per crop and yield per hectare were calculated using data from the Canadian Census of Agriculture (2011). Quantities of livestock products were calculated from the number of livestock present multiplied by the quantity of product produced per animal (fluid milk, egg, meat) following Statistics Canada (2014). Data on nitrogen fertilizer application per crop type are from BC Crop Production Guides (nitrogen fertilizer application per crop type) (BC Ministry of Agriculture, 2014). In the absence of any data on local consumption of locally produced food products, we assumed that any local production first met local food demand and the remaining demand was fulfilled by import from other parts of the country and from the rest of the world.
Bioregional food crop import: As data on the specific import sources to the studied bioregion do not exist, we used Canadian and international sources (FAOSTAT, 2016; Statistics Canada 2016) followed the method presented by Kissinger (Reference Kissinger2012, Reference Kissinger2013) (i.e., if a certain percentage of specific commodity was imported to Canada from a specific source, we assumed the same proportion for the import commodities consumed in the studied bioregion). For detailed figures used here see Table 2 of the supplementary materials file. Differing from previous biophysical flow accounting studies that determined sources of supply and yields at the national scale (i.e., import from the USA using national average yield factors), for import from the USA(which comprise the main source of supply) we utilized specific state-level source supply and yield data (i.e., the amount of commodities imported from California versus Florida, etc.).
Biophysical calculations
Each biophysical indicator—carbon, land, water or the EF was calculated for both locally produced and imported commodities as described in the following paragraphs, which also highlight main data sources and limitations.
Land footprint (LF)
The LF included the agricultural land (hectares) required for growing one unit (a ton) of a commodity in the SWBC bioregion and in the main regions from which the commodity was imported. For SWBC land input calculation, we followed Dorward et al. (Reference Dorward, Smukler and Mullinix2016) published in this journal. Data are from Statistics Canada (2014). Land required for growing feed crops for livestock used to produce food commodities, dairy, poultry and meat were also included. Livestock feed requirements are from Statistics Canada (2003a, b). See the supplementary materials file for further details.
For calculating the area of land in different parts of the world used to produce agricultural food commodities for consumption by SWBC residents we traced the main sources of all studied commodities. While imported food originated from all over the world the vast majority of that supply originates from other parts of Canada and from the USA. We used detailed yield factors from Statistics Canada (2011), the USDA (2011), UCDavis (2017) and the FAOSTAT (2016). Details on yield factors from major sources of supply are presented in the supplementary materials file.
Carbon footprint
The CF of each commodity was calculated by incorporating CO2, nitrous oxide (N2O) and methane (CH4) emissions along the commodity chain for 1 ton of produced and imported commodity consumed in the SWBC bioregion. Results are presented in CO2 equivalent, using a factor of 1 kg CO2/kg CO2, 310 kg N2O/kg CO2 and 21 kg CH4/kg CO2 (IPCC 2014).
Energy-related emissions associated with food production include CO2 from farm machinery fuel use and emissions embodied in the production of fertilizers used. Data on fertilizers use for the local agricultural system were based mostly on BC farm Production Guides (BC Ministry of Agriculture, 2014); and other sources for specific agricultural commodities such as BC MAL (2010). Data for imported commodities divided into commodities that are mainly imported from other parts of Canada and the USA are based on USDA-NASS (2000; 2005a, b; 2007); and UCDavis (2017), while commodities from other parts of the world are based on IFA (2002). The fertilizer(s) GHG conversion factor used followed Wood and Cowie (Reference Wood and Cowie2004). Data on farm machinery energy use relied on several North American sources, including Pelletier et al. (Reference Pelletier, Arsenault and Tyedmers2008); USDA-ERS (2008a, b); Pimental (2008); Agriculture Agri-Food Canada (1999). For detailed input data used and related GHG emissions for each studied commodity see Tables 3 and 4 in the supplementary materials file. Other GHG emissions (N2O and CH4) were calculated for each livestock commodity following data from Statistics Canada (2003a, b); Hofmann & Beaulieu (Reference Hofmann and Beaulieu2001); Ominski et al Reference Ominski, Boadi, Wittenberg, Fulawka and Basarab2007. To validate that the CF of imported commodities we calculated were within a reasonable range of emissions we compared our figures to those presented in the food-related GHG emissions review by Heller et al. (Reference Heller and Keoleian2015). The comparison can be viewed in file 4 of the supplementary materials file. Emissions related to shipping the imported food commodities relied on data from the Kissinger (Reference Kissinger2012) study which analyzed and reported Canada's food miles related emissions. The research included the average shipping-related emissions of different agricultural groups (e.g., fruits and vegetables) from the growing regions to the studied bioregion.
Ecological footprint (EF)
The EF indicator integrates the LF with the CO2 component of the CF. The land types included in the EF calculation are croplands required to grow studied agricultural commodities, grazing land for livestock and forest land needed to sequester the CO2 emitted along the commodity lifecycle (Borucke et al., Reference Borucke, Moore, Cranston, Gracey, Iha, Larson and Galli2013). Conversion factors, to global hectares (GH) units used in this study, are 0.48 for grazing land; 2.39 for cropland; and 1.3 for carbon land following Ewing et al. (Reference Ewing, Moore, Goldfinger, Oursler, Reed and Wackernagel2010). The use of the GH approach captures the different land types based on their inherent capacity to produce human useful biological resources in relation to the global average productivity across all land types (Galli et al., Reference Galli, Wiedmann, Ercin, Knoblauch, Ewing and Giljum2012). To calculate the EF we integrated data from other indicators (i.e., land use and CO2 emissions) as presented above with the EF conversion factors.
Water footprint (WF)
Differing from the other footprint indicators which required integrating data from various sources, to calculate the WF we utilized the Mekonnen and Hoekstra (Reference Mekonnen and Hoekstra2012) database on the virtual water—green (i.e., precipitation) and blue (i.e., irrigation) water (m3 ton−1) in different regions of the world. For the water footprint of local production, data were directly extracted from the above source, which included specific data for BC. For the imported food-related WF we followed the same method as for imported land, adjusting trade data with water footprint conversion factors for each source of supply.
Exploring rates of local biophysical advantages
After calculating the footprints for each food commodity the subsequent footprints were explored for the entire bioregion. We then analyzed the local and imported share of footprints for all studied commodities, and categorized each, based on the outcomes for the four biophysical indicators (i.e., four local footprints and four imported ones), all according to local or imported biophysical advantage. Differing from most studies for our research differentiated between local and imported commodities and calculated four biophysical indicators for given commodities of each type. Doing so allowed us to compare each commodity by each footprint category to ascertain whether it being locally produced or imported offered a possible biophysical advantage or disadvantage. Commodities either had a local advantage (+) for a specific footprint indicator, meaning that the footprint per unit local production was smaller than the footprint per unit of the imported commodity, or they had an import advantage (−). If an indicator had <15% difference in footprint per ton, between an imported and local commodity then we deemed them to have no biophysical advantage (=).
We then summarized the local biophysical performances and reported it for each commodity in one of five categories:(i) absolute local advantage—suggests smaller footprints for all biophysical categories; (ii) significant advantage—suggests smaller footprints for three out of four biophysical categories; (iii) partial advantage—suggests smaller footprints for only two categories; (iv) single footprint indicator advantage—suggests that in all categories other than one the imported product footprint was smaller; and (v) no local advantage—suggests that none of the local commodity footprints are smaller than the imported one. This analysis uses the footprint family indicators to highlights the biophysical relative strength and weaknesses of local production in the study bioregion compared with imports of the same commodities. We had no intention to and did not include weighting of the relative importance of one footprint indicator over another. Similarly, we made no attempt to assess or analyze the biophysical footprint implications for varying production methods (e.g., organic). Such analysis may be relevant for future research.
Results
Total food commodities consumed (inclusive of food waste) in the SWBC bioregion in 2011 was approximately 2.26 million tons (or the equivalent of 820 kg per bioregion resident). Of that, approximately 35% was produced in the bioregion. The rest originated from other regions in Canada or abroad. While some of the imported foods are not produced at all in the bioregion (e.g., rice and tropical fruits), about 86% of total food consumption by weight is comprised commodities that can be.
In what follows, we first present a snapshot of the bioregion's food consumption, identifying the demand for different commodities by food category and the share of demand met by the local and global supply. We then report the ‘biophysical performance’ of the bioregional food system emphasizing the differences between different biophysical footprint indicators. This is followed by a detailed analysis for each indicator which allows exploring the contribution of specific commodities to the overall footprint and the share of local versus imported footprint. Lastly, we present an analysis of the system in regard to commodities with local or import biophysical advantage, highlighting a potential to consider optimizing a local supply system from a biophysical perspective. Annual food demand is categorized by major food groups. Three food categories comprise more than 80% of the overall regional food demand—livestock products, fruits and vegetables (Fig. 1a). For each the share of local versus imported sources varies (Fig. 1b). Of the above-mentioned three categories, only livestock products display a substantial share of local production (mostly chicken, egg and dairy). However, that local production relies on imported feed grain. Also, beef which requires significant pasture land and feed grain is mostly imported from other parts of Canada and abroad.
Fig. 1. (a) SWBC annual food requirement; (metric tons). (b) The share of local versus imported food.
Local and global footprints, land, water, carbon and ecological
Supplying food demands requires biophysical inputs (i.e., land, water and energy) domestically and in the other parts of the world from which SWBC imports food. It generates emissions along each commodity's lifecycle and has footprints on different terrestrial ecosystems (Fig. 2). The quantities of biophysical inputs required and emissions produced are influenced by the types of food consumed (e.g., beef versus fruit) and the sources of the food (i.e., local or imported). Overall our analysis revealed the need for a total land area of just over 2 million hectares and 3 billion m3 of water. Producing and supplying food from both local and global sources generated 2.8 million tons of CO2e conferring an EF of approximately 2.5 million GH.
Fig. 2. The local share of the region's food demand footprint by footprint indictor.
Figures 3a and 3b disaggregate each footprint indicator into agricultural commodity groups and to local and imported biophysical implications. In comparing commodity groups, some findings are very clear. For example, livestock products make up a significant share of all footprint indicators. A closer look at local food production reveals that its impact varies between the different footprint indicators. For example, the LF of local livestock production as a percent of total livestock LF is relatively small compared with the local livestock share of the water, carbon and even ecological footprints. This is probably the outcome of the types of local livestock products such as chicken, egg and dairy, their feed system and the smaller weight of pasture lands in EF compared with other mostly imported livestock products such as beef. Comparing findings in Figure 3 to the share of consumed products presented in Figure 1, reveals the difference between the significant share of fruits and vegetables consumed by weight (47% of overall food consumption) and their share in each of the footprint indicators (together representing 2% of land, 11% of water, 16% of carbon and 8% of ecological footprints).
Fig. 3. Biophysical disaggregation of SWBC food consumption.
A local biophysical advantage?
Biophysical advantages accruing to either local or imported products were revealed for each commodity (Table 1). Our summation shows that only a few agriculture commodities grown in the SWBC bioregion can be categorized as having an absolute local advantage and about half have a significant or partial advantage. Of the 40 agricultural food commodities that are grown in SWBC, we calculated an absolute local biophysical advantage for six crops, and an absolute import advantage for 13. The remaining 19 commodities presented partial or significant local advantages.
Table 1. Examining the potential local advantage
Table 2 indicates how much of total food consumed in SWBC by weight, fall into each of the five biophysical performance categories per above (plus a category of commodities that are not grown locally). It reveals that current food consumption predilections favor agricultural commodities without local production advantage. Out of the total tons consumed approximately 11% has absolute or significant local production advantage, while the vast majority of commodities consumed, 67% by weight, have no local advantages or cannot be grown locally. Table 2 also reveals that local production of commodities with absolute local biophysical advantage does not meet local consumption demand. Only 32% by weight of the total consumed is produced locally; the remainder is imported. Conversely of products showing no local advantage almost 53% consumed by weight are produced locally (the vast majority of this is related to local dairy production).
Table 2. A biophysical analysis of SWBC food consumption
Table 3 disaggregates the bioregion's footprint into the five categories. It highlights that the majority of all footprint categories is related to locally grown products with no local biophysical advantage or a single local advantage. Furthermore, commodities with a partial biophysical advantage such as most livestock products (Table 1) represent the largest import related footprint. While the overall footprint presented in this table is the same as presented in Figure 3 and Table 2, such presentation highlights the potential footprint pressure arising from different food categories and can be the basis for examining the biophysical contribution of changing commodities compositions.
Table 3. A detailed breakdown of the footprint indicators into the region's categories
Discussion and Conclusions
Localization is commonly perceived and proffered (by advocates and researchers) as one means to advance sustainable food systems. This study focused on a single region and therefore its findings and conclusions are related only to that particular region (based on a specific year and data sources). While indeed localizing food systems may convey substantial benefits, our analysis of four biophysical indicators for an SWBC bioregion food system suggests the need to reconsider the assumption that localization will automatically result in overall environmental stewardship and resource conservation advantage.
Like many other regions in which human settlements have historically established evolved around the world, the SWBC bioregion is replete with the highly fertile arable land. However, its growing population combined with various socio-economic factors means that the demand for food in general and for the diversity of food products in particular, is also growing. Analyzing current biophysical implications of the bioregion's food system, and comparing impacts of local and global sources as reported herein has several implications for the studied bioregion food system sustainability and for other regions that may be interested following similar examination: (a) highlights the current rate of reliance on local and external sources; (b) identifies specific agricultural commodities presenting local biophysical advantages; (c) indicates the individual footprint family advantages; and (d) generates a baseline for modeling future required and/or potential changes to advance more biophysically sustainable food systems.
SWBC dependence on external sources: Similar to Rees (Reference Rees1992), our study found that indeed the SWBC bioregion population relies mostly on imported biocapacity. The SWBC bioregion's current EF extends far beyond its boundaries. Approximately 35% of the bioregion's food supply is local. Further, out of the 53 commodities included in this study 12 were not grown locally at all during the studied year. The demand for those commodities can be related to the region's diverse socio-cultural composition, to the globalization of the food system, and to the near complete disassociation of consumption and its biophysical character and capacity.
The local advantage: The biophysical analysis of a regional food system advanced herein revealed that some food commodities have a local advantage and can contribute to a lower footprint. The analysis also highlighted that currently the local food production in the bioregion is producing many commodities that have no or very few biophysical advantages, perhaps even at the expense of those that do. Certainly, numerous socio-political, economic and ecological factors shape regional food system production and consumption patterns. Factors such as economic benefits (e.g., potential return on investment), consumers’ demands, governmental subsidies and policy, sector clustering, and personal preference influence farmers’ decisions about which agricultural commodities to produce.
Following the logic of conventional mainstream economics, biophysical factors, internalized into the costs of production, should be reflected in the final commodity price. However, it is widely acknowledged in the literature, that these factors rarely receive full consideration and their implications for environmental degradation are in fact externalized and not reflected in the market. Furthermore, while the global food system facilitates importing from regions that are more ‘efficient’ (e.g., with lower labor costs, less regulation and lower land costs), doing so does not reflect scarcity of resources (land, water or energy) necessary for production or the pressure on global ecosystems (carbon emissions, biodiversity, ecosystem health). Identifying those commodities with higher/better local biophysical performance has the potential to minimize that gap between the contemporary system and one more biophysically sustainable. The information from such analysis can inform people on consumption and diet composition predilection and choice. It can also signal to policymakers the potential contribution of promoting the production and consumption of specific commodities with greater absolute and significant local advantage to the region's sustainability.
The ‘footprint family’ indicators potential: To date most food systems biophysical analysis have focused on a single indicator (e.g., land, water and perhaps mostly GHG emissions) or used a very detailed lifecycle assessment (LCA) research method for specific commodities. While both approaches have several advantages the drawback of the first is that by focusing on a single footprint indicator it fails to reveal the biophysical complexity (e.g., a commodity may have large CF but a lower land requirement, etc.). In regard to the latter, while ideally we would have a full LCA study for all the commodities included in this research, in practice we are still far from that day. Integrating all indicators and embracing a life cycle assessment approach in a single study as advanced here reveals the biophysical complexity and allows generating a detailed analysis of each commodity's key biophysical interactions. Following the results of this study, a detailed LCA can be advanced for specific commodities.
Modeling the future: The idea of a ‘100 miles diet’ (Smith and MacKinnon, Reference Smith and MacKinnon2007), which originated in of the SWBC region, among other attempts in different parts of the world to emphasize advantages of local food systems is important, as it calls into question conventional and unchallenged acceptance of the desirability and inevitability of global food system hegemony, a system disconnected from its location/region. However, as our study revealed localization of the food system is not necessarily more sustainable biophysically. Therefore following the case analyzed in this research biophysical advantage may not be the most compelling rationale for localizing food systems. However, by identifying the current level and composition of dependence on local and imported food commodities, and the advantages and disadvantages of different local commodities, we can explore and contemplate a food system structure conferring maximum biophysical sustainability.
Further, while the analysis used the footprint family indicators to highlight the local biophysical relative strength and weaknesses it did not include or intended to weight the relative importance of one footprint indicator over the other or the contribution of different local production scenarios on the region's footprint. Such categorization may be relevant for future research.
Of course, biophysical disadvantage or advantage is not the sole determinant of food system sustainability and must be considered in relation to social and economic parameters as well. Likewise, global food systems are, at this juncture, entirely dependent on non-renewable energy supply. The analysis undertaken in this study can be used in the future to model and optimize food systems which seek a balance between food demand and food supply while considering biophysical constraints as well as social and economic factors. While this study accounted for the SWBC bioregion's current situation, the future analysis could explore the extent to which the region's food demands can be supplied from local sources and the extent to which such changes can minimize the region's footprint in the future.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1742170518000078
Acknowledgement
The work reported on here was part of the larger Southwest British Columbia Bioregion Food System Design Project (http://www.kpu.ca/isfs/swbcproject). Major funders of this project were Real Estate Foundation of British Columbia, Vancouver City Savings Credit Union, R. Howard Webster Foundation, and Vancouver Foundation.
