Eutrophication Potential of Food Consumption ... - ACS Publications

Jul 22, 2010 - exhibit a highly variable nitrogen-intensity; on average, red meat and dairy products require much more nitrogen than cereals/ carbohyd...
0 downloads 0 Views 1MB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 6450–6456

Eutrophication Potential of Food Consumption Patterns XIAOBO XUE* AND AMY E. LANDIS Department of Civil and Environmental Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, 15261

Received November 12, 2009. Revised manuscript received May 25, 2010. Accepted June 11, 2010.

Although the environmental impacts and carbon footprints of foods are gaining more public attention and scientific debate, few studies have systematically evaluated the life cycle nitrogen and phosphorus flows among different food types. Disruption of natural nitrogen and phosphorus cycles already result in serious environmental quality degradation and economic losses, such as loss of fisheries due to hypoxia in the Gulf of Mexico. This study characterizes the nutrient flows during food production, processing, packaging, and distribution stages for eight food types; compares carbon footprints and nitrogen equivalent footprints of food groups; evaluates solutions to reduce excessive nitrogen outputs; and estimates effectiveness and efficiency of possible solutions. Different food groups exhibit a highly variable nitrogen-intensity; on average, red meat and dairy products require much more nitrogen than cereals/ carbohydrates. The ranking of foods’ nitrogen footprints is not consistent with their carbon footprints. For example, dairy products and chicken/eggs have relatively high nitrogen footprint and low carbon footprints. Finally, the study evaluates shifting food consumption patterns. Dietary shifts from dairy products and red meat to cereals can be an effective approach for lowering the personal nitrogen footprint.

Introduction Food production, processing, distribution and consumption activities have significant social, economic, and environmental impacts (1-8). While vast quantities of foods are required to satisfy a basic human need every day, food production also results in natural resources depletion, water quality degradation, and climate change. Nutrient fluxes from food supply chains have resulted in water quality degradation in the form of hypoxia and eutrophication causing loss of ecosystem services and species extinctions (9-11). The hypoxic zone in Gulf of Mexico (GOM) mainly caused by excess nutrients exported from agriculture production in Mississippi River Basin (MRB) resulted in reduced commercial and recreational fisheries (12). To minimize and mitigate this hypoxic zone, a task force of federal, state, and tribal representatives established a goal of reducing the hypoxic zone size to 5000 km2. Although this task force recognized the significant role of agriculture in MRB to achieve this goal, there is concern that increased agricultural production may further hinder achievement of hypoxic zone production (13). Therefore, it is important to identify eutrophication potential of different foods and mitigate their environmental impacts such as hypoxia in the GOM. Global warming potential of food production and transportation systems is reported widely for assorted food types * Corresponding author e-mail: [email protected]. 6450

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

and different food supply systems (1, 5-8, 14). Previous studies show that livestock production systems have higher carbon footprints than crop and vegetable production systems (8, 15). “Food miles” research focusing on carbon emissions during food delivery stage advocates localization of global supply network (5-7, 14). Weber et al. reported the carbon emissions of food choices, and discovered eating less red meat and dairy could be a more effective way to lower an average U.S. household’s food-related climate footprint than buying local food (8). Some research has addressed the eutrophication potential of food production (15-22). For example, Tilman et al. forecasted that 109 ha of natural ecosystems would be converted to agriculture by 2050 to meet food demand of increasing population and consumption. This would be accompanied by 2.4- to 2.7-fold increases in nitrogen and phosphorus-driven eutrophication of terrestrial, freshwater, and near-shore marine ecosystems, and comparable increase in pesticide use (16). Although advances in agricultural system management may reduce nutrient losses resulting from agricultural production, the risk of worsening eutrophication issues still exists due to continuously increasing food and biofuel demands in future. In addition to forecasting, the eutrophication potential from a single food type or a single sector of food supply chains has been assessed previously (17-21). However, systematic evaluations of food supply chains and comparisons of eutrophication footprints among food types are still lacking. Recently, life cycle assessments (LCAs) have been utilized to evaluate and improve the environmental performance of food production systems. LCA results have been used in the development of eco-labeling criteria with the aim of informing consumers of the environmental characteristics of products. However, most analyses are limited to case studies of either a single food or a limited set of items (22-27). With large groups of products, the resources and data availability do not often allow for detailed analyses. A few studies researched overall diet but these studies focused on the environmental relevance of carbon footprint and food consumption patterns (8). The study of nitrogen and phosphorus inventories over all food categories has not yet been performed. Additionally, the effects of reducing eutrophication potential through shifting food consumption pattern have not been addressed. Food supply activities should be evaluated from a life cycle perspective and are simplified into four phases: farming, processing, packaging, and transportation. Every phase emits considerable amounts of eutrophication species into the environment. Farming systems, as a primary stage of food production, are widely recognized as an important contributor for water quality degradation (15, 17, 18, 28, 29). Aqueous nitrate (NO3-) and phosphate (PO4-3) emissions from agricultural nonpoint source emissions such as fertilizer runoff and manure storage systems contribute to eutrophication. Air emissions such as NH3, N2O, NO and NOx generated from volatilization and denitrification processes can also contribute to eutrophication potential resulting from agriculture. Additionally, nitrogen compounds (NOx, N2O) are also emitted from combustion processes during agricultural operations and processing of food (17, 18). Food processing industries generate large amounts of organic materials such as protein and lipids, high biochemical and chemical oxygen demands (BOD and COD), and considerable amounts of dissolved nutrient concentrations (NH4-, NO3-, PO4-3) (2, 4). Similarly, food packaging can produce both air and water emissions to the environment (30-32). Transportation results in nitrogen air emissions, mainly including NH3, N2O, NO, 10.1021/es9034478

 2010 American Chemical Society

Published on Web 07/22/2010

FIGURE 1. System boundary of foods for identifying their eutrophication potential. Farming production, food processing, food packaging, and food delivering stages are included in the system boundary. NOx, which can also contribute to eutrophication (33, 34). Thus, it is important to consider all these species during each stage to evaluate eutrophication potential of food supply chains. Food choices and changes in food production may offer a unique opportunity for consumers to lower their personal environmental footprints and may offer national improvements in water quality. A systematic evaluation of multiple food products and their environmental impacts such as eutrophication potential and carbon footprint is needed to make informed policy decisions. This paper presents such an analysis and both combines and compares the results with food products’ carbon-footprints. Nitrogen and phosphorus emissions and environmental impacts throughout the life cycle of the food production system including agricultural production, processing, packaging and distribution stages are quantified using a life cycle approach. This paper also evaluates strategies that may reduce eutrophication and hypoxia and estimates the reduction of nutrient outputs due to food consumption pattern shifts.

Materials and Methods As explained above, LCAs have been employed to investigate environmental profiles of different foods. Guidelines for performing LCAs are presented by the International Organization for Standardization’s (ISO) 14040 series (35). The main phases of an LCA include goal and scope definition, inventory analysis, impact assessment, and interpretation. This study utilizes the LCA framework to quantify eutrophication potential over the life cycle of the food production system. The life cycle inventory evaluates N, P compounds and BOD/COD contents that contribute to eutrophication. Eutrophication potential was estimated based on traditional life cycle impact analysis methods, as described below. System Boundary. The life cycle stages vary with different food products. Generally, food LCA stages include farming production, food processing, packaging, and delivery. Figure 1 shows the boundaries for the researched food groups.

Functional Units. The functional unit is the reference unit that forms the basis for comparison between different systems. Most published food LCA research uses mass or volume based functional units (36). A more sophisticated way of defining the functional unit is to include quality aspects, for example, nutrient content of food. The nutrient content is described by factors like the amount of carbohydrate, fiber, vitamins, minerals, essential amino acids, energy, fat and protein, or others. In this article, g nitrogen/ kg food is defined as the functional unit to compare nitrogen profiles of different food groups. Normalized units (g nitrogen/kcal in food; g nitrogen/$ food) are also employed to reflect the influences of economic value and energy content. Life Cycle inventory (LCI) and Life Cycle Impact Analysis (LCIA). Several existing tools were used to compile the LCI, including SimaPro and GREET. SimaPro software developed by Pre´ consultants is expandable and transparent software that integrates inventory data for a broad spectrum of industrial and economic sectors (37). The greenhouse gases, regulated emissions, and energy use in transportation (GREET) model created by Argonne National Laboratories delineates life cycle energy use and emissions of criteria air pollutants based on EPA emission factors of transportation stages (34). SimaPro software and associated databases including ecoinvent V2 (38), LCA food (39), Industry data 2.0 (40), BUWAL250 (41), IDEMAT 2001 (42) were used to compile nutrient inventory for the food processing and packaging stages. GREET was employed to account for inventory in the transportation stage. Additionally, the LCI for the agricultural stage was created using a variety of data collected from published articles and SimaPro databases. The LCI data sources are outlined in Table 1, while detailed data sources are illustrated in SI. Total nutrient output is equal to the sum of the nutrient flows from every stage of the food production system including agriculture, processing, packaging, and transportation. Packaging and packing containers can be divided into two groups: commercial packages and transportation containers. This article only considers commercial packages (43, 44). Commercial packages protect the product and guarantee its quantities and composition for direct consumer purchase. Packages are produced from different materials in a variety of types, for example, bags, cartons, glassware, cans, etc. Detailed assumptions of packaging materials for foods are given in the SI. Distances and transportation modes for delivering food subgroups are obtained from literature (8). All GREET default assumptions are followed to calculate air emissions during transportation stage (34). The LCIA was conducted utilizing TRACI (Tool for Reduction and Assessment of Chemical and other environmental Impacts), which was developed by the U.S. Environmental Protection Agency (33). TRACI is used to calculate eutrophication potential for the system. TRACI defines characterization factors (CFs) relating N and P species to eutrophication potential, thus allowing the LCI data to be expressed in terms of the TRACI defined reference compound, N-equivalents. Monte Carlo analysis (MCA) is used to quantify variability and uncertainty of the LCI (17, 28, 29). Any independent variable with a range of estimates or possible values were assigned a probability distribution. Best-fit probability distributions were determined using Anderson-Darling tests. Independent nitrogen equivalent values were collected or calculated for each LCA stage of every food group. Crystal Ball 7 software was used to define probability distributions of nitrogen equivalent values for every stage and to conduct the MCA (29). The distribution of output variables, as a VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6451

TABLE 1. Stages of Food LCA and Associated Emissions with Eutrophication Potentials stages farming food processing food packaging transportation a

emissions of concern NH3, NO, N2O, NOx, NO3-, PO4-3, NH4-,BOD, COD NH3, NO, N2O, NOx, NO3-, PO4-3, NH4-, BOD, COD NH3,NO, N2O,NOx, NO3-,PO4-3, NH4-,BOD, COD NO,N2O,NOx

databasea peer reviewed articles, ecoinvent V2 peer reviewed articles, ecoinvent V2, LCA food, industry data 2.0, BUWAL250, IDEMAT 2001 ecoinvent V2, LCA food, Franklin US 98, Industry data 2.0, BUWAL250, IDEMAT 2001 GREET1.8

Peer reviewed articles and use of database is explained and referenced in the Supporting Information.

FIGURE 2. Eutrophication potential of researched food groups by life cycle stage. Stages include the agricultural production, food processing, food packaging and transportation. Median values within this study are presented in the bar graph. Certainty bars represent the 10 and 90% confidence intervals. function of independent values, was generated through MCA, which repeatedly and randomly samples values from the probability distributions of independent values. The distribution ranges of total nitrogen equivalent were determined from distributions of each stage’s nitrogen equivalent value through the MCA method.

Results Contribution to Eutrophication Potential at Each Life Cycle Stage. Figure 2 shows eutrophication potentials for the different food groups. Red meat has the highest eutrophication potential, followed by dairy products, chicken/eggs and fish. The cereal and carbohydrate (cereal/carbs) subgroup is identified to have the lowest nutrient footprint among all food subgroups. While producing, processing, transporting, and packaging 1 kg of red meat generates on average 150 g nitrogen-equivalent emissions, around 2.6 g nitrogen equivalent emissions are released to supply 1 kg cereal/carbs. The agricultural stage is the largest eutrophication emission sector, which shares more than 70% of total eutrophication potential for each food group. Both plant production and animal raising systems are reported to be responsible for eutrophication-related emissions to surrounding water bodies. Corn and soybean farming systems, providing feedstock for human diet and animal feed, emit large amounts of NO3- and PO4-3 into groundwater and surface water (17, 28, 29). Manures from animal raising systems contain high nutrient contents. Atmospheric NH3 and N2O, (generated from nitrification/denitrification processes) and aqueous N and P species (transformed or dissolved from manures) can significantly influence the 6452

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

nutrient inventory of food supply chains (19, 45-48). Nutrient footprints of red meat and dairy products include direct nutrient emissions from animal raising systems and upstream nutrient emissions from plant production; this explains the relative intensity of red meat and dairy products which have the highest eutrophication potential from a life cycle perspective. Eutrophication potentials of food processing stages vary with processing techniques, while transportation distances contribute minimally to eutrophication impacts. Processing dairy products and meat products has important influence on food’s eutrophication potential. Industrial milk processing (including liquid milk, milk powder, cheese, butter, etc.) generates distinct nutrient waste. N and P species in dairy processing effluents originate from cleaning compounds and from milk or product spillage during dairy product processing. Significant amounts of phosphate based cleaners and nitric acid based cleaners were used during washing procedures at the end of production cycle, consequently resulting in high levels of N and P in most dairy wastewater (49). Eide et.al reported that eutrophication potential of processing milk ranged from 6.2 to 8.0 g O2/L milk which corresponds to 0.31 g N/L milk and 0.4 g N/L milk (21, 25, 50). Slaughtering animals also influences the nutrient inventory significantly. The major source of nitrogen and phosphorus is from the protein in the meat particles and blood in the wastewater from slaughter plants (2). Other sources of nitrogen are the manure and partially digested feeds from stomachs and gizzards and intestines, as well as urine (2). The packaging stages and transportation stages have negligible impact on eutrophication profiles of food groups. The usage of packing

FIGURE 3. Comparison of normalization factors for eutrophication potential of food groups. From left to right: by expenditure (year 2007) and energy content. All values are shown relative to the value of cereals/carbohydrates. materials for supplying foods contributes less than 15% of food groups’ eutrophication profiles. Although transportation distances can be long, eutrophication potential resulting from atmospheric NH3 and NOx deposition are relatively small, representing less than 2% of the total eutrophication potential for most of good categories. Comparative results of eutrophication potential among different food groups inform consumers of the relevance between lowering eutrophication footprints and food consumption pattern. However, different foods groups have different prices, different nutrients, and of course are more or less desirable depending on consumers’ preference. Figure 3 shows a comparison of total impacts with impacts normalized by expenditure and caloric content. Prices and calories of different food groups are published by the U.S. Department of Agriculture (USDA) (51, 52). Results show that when consumers spend 1 dollar on foods, the supply chains of those foods emit approximately 0.7-16 g N equivalent. Similarly, the supply chains produce 0.5-41 g N equivalent, when 1 kcal of energy is delivered to consumers’ baskets. Cereals/carbs group has the lowest N emissions normalized by food price and calories. Compared with other food groups, cereals/carbs group is the most environmentally friendly choice for reducing nutrient emissions, when the same amount of expenditure or the same energy content is considered.

Discussion Uncertainty in Results. Uncertainty of nutrient inventory among food groups was assessed using MCA. Results show that the agricultural systems exhibit considerable variability and uncertainty in emission profiles due to differences in geography, climate, and agricultural practices. Uncertainty in the meat production stage stems from different feed choices, animal raising practices, farms’ locations and other factors. The choices of feed intake influence the amount of nitrogen excreted by animals and nitrogen emitted from feed production processes (53). Besides feed choices, production modes also have an impact on eutrophication potential (19). The uncertainty of eutrophication potential resulting from pork production was investigated through differentiating the production modes. The analysis shows that the environmental impacts of current intensive pig production systems are significantly different from alternative production systems in France (19). Additionally, the impact of farms’ locations on eutrophication potential has been quantified by researchers. Kumm et.al found that the nitrate leaching potential is related to the spatial location of farms in Sweden (54). Farms in central Sweden (lower precipitation, clay soils) had only

FIGURE 4. Comparison of carbon footprints and nitrogen equivalent footprints. Carbon footprints are obtained from Weber et al. (8). one-third of the leaching level of farms in southwestern Sweden (higher precipitation, sandy soils). Temperature, humidity, and soil compositions can greatly influence denitrifaction/nitrification rates, and P adsorption capacity of soil particles, consequently influencing the amounts of nutrients transported into water bodies. The use of average data to characterize agricultural system may not represent emissions occurring during “extreme” years (such as rainy or drought years), and the subsequent environmental impacts (29). Since the agricultural sector is a dominant contributor for eutrophication potentials of foods, identifying and characterizing the uncertainty of nutrient flows in agricultural systems is important for future research. When aggregate food groups are researched rather than specific food types (such as the difference between grass-fed versus grain-fed meat, organic farming vs conventional farming, etc), uncertainty and variability are enhanced. The constrain of data availability limited the number of researched food types. However, the average values and variability evaluation of the nutrient inventory presented within this research among food groups are still meaningful in investigating eutrophication footprints of dietary choices. Comparing Carbon Footprints and Nitrogen Footprints. Figure 4 compares nutrient footprints and carbon footprints of different food groups. Food groups close to the origin have both low carbon footprint and low nitrogen footprint. For example, cereals/carbohydrates are the most environmentally preferred food types from a carbon and nitrogen life cycle perspective. By contrast, red meat has the highest carbon footprint and the highest nitrogen footprint. Dairy products, fish, and chicken/eggs have relatively higher nitrogen footprint and lower carbon footprints. In contrast, sweets, oils, fruits, and vegetables have relatively lower nitrogen footprints and higher carbon footprints. The inconsistency between carbon footprints and nitrogen footprints indicates trade-offs of shifting food consumption habits and inherent environmental complexities of food policy decisions. For example, solely minimizing consumers’ C-footprint would suggest that one consume cereals/carbs, dairy, chicken/eggs, and fish. However, if the N-footprint is also considered, as shown in Figure 4, dairy products are not necessarily ideal since they have the second highest eutrophication potential, while fruits and vegetables might be reconsidered since they have a minimal N-footprint. Nitrogen Output Reduction Due to Consumption Pattern Shifts. The effect of food consumption pattern shifts on nitrogen equivalent emissions reduction is estimated assuming (1) eutrophication profiles of foods are unchanged during consumption shifts, (2) a linear relationship exists between nutrient output and mass of consumed food for VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6453

FIGURE 5. Eutrophication potential reductions due to consumption shifts from high nitrogen profile foods to low nitrogen profile foods. The same caloric contents are maintained. Baseline is average nitrogen equivalent output resulting from annual U.S. food consumption per capita (56). Other lines describe eutrophication potentials of other dietary consumption scenarios. The legend depicts the ranking of the food consumption shifts from low N equivalent reduction (e.g., the lowest is USDA baseline) to high N equivalent reduction (e.g., the highest is shifting dairy products to cereal products).

FIGURE 6. Cost reductions due to consumption shifts from high nitrogen profile foods to low nitrogen profile foods. The same caloric contents are maintained. Baseline is annual U.S. food cost per capita in 2007 (51). Other lines describe cost of other dietary consumption scenarios. The legend depicts the ranking of the food consumption shifts from high food cost (e.g., the highest is shifting red meat to fruits/vegetables) to low food cost (e.g., the lowest is shifting dairy products to cereal products).

every researched food group, and (3) consumers maintain constant calorie consumption during food pattern shifts. In reality, the shift of dietary habits is a relatively slow process (55, 56). Although technology improvements and policy incentives have the potential to reduce environmental footprints of food production and processing on a large scale, shifting red meat and dairy products to other low nitrogen intensive food groups may significantly reduce personal eutrophication potential. Estimated results (Figure 5) show that food supply chains generate 40 kg nitrogen equivalent for meeting one person’s food needs annually. Among possible consumers’ behavioral changes, shifting dairy products to cereals products is the most effective way to mitigate personal eutrophication potential from both cost and nutrient emission perspectives (Figures 5 and 6). A change in milk has larger eutrophication effect than a change in red meat consumption because of meat’s higher caloric density. Shifting 5% of dairy product consumption to cereals groups and maintaining the same caloric intake can prevent 380 g per capita annual nitrogen equivalent emissions to the environment. On the extreme, 7630 g of nitrogen equivalent/ year could theoretically be avoided if 100% of dairy products were replaced by cereals/carbs products. The fluctuation of food cost as a result of food consumption pattern shifts is estimated based on the same set of assumptions as discussed previously in addition to the assumption that (4) the price of food groups remains the same as the 2007 baseline and (5) a linear relationship exists between food cost and mass of bought food for every researched food group. Estimated results (Figure 6) are calculated in comparison to a baseline of $4840 for meeting one person’s food needs annually. This study does not account for cost of dining out. The most economically effective choice is to shift dairy product consumption to cereals groups, while shifting red meat to vegetables may increase cost. Shifting 5% of dairy products to cereals groups and maintaining the same calories can save $40 annually for each person. In the extreme case, replacing 100% of dairy product with cereals/carbohydrate products could save $810. Outlook. Because of demand for food to support the expanding world population and changing dietary preferences with development, the application of fertilizers is

predicted to continually increase. This will likely worsen coastal eutrophication and hypoxia. Effective and efficient solutions should be employed to reduce nitrogen needs and environmental nitrogen output to eventually minimize eutrophication. Changing food purchase behaviors may be an effective mitigation strategy. If people consume less red meat and dairy products, the nitrogen usage for food production will decrease. And surprisingly, shifting away from dairy products to cereal products achieves larger eutrophication reduction than shifting away from red meat to cereal products when the same energy content is maintained. However, complicated environmental impacts of foods including carbon and nitrogen profiles require overall environmental evaluation of food choices. Policy decisions based solely on one aspect of environmental impacts, either carbon footprint or nitrogen footprint, are likely to result in trade-offs such as environmental burdens shifting from global warming to eutrophication impacts. Additionally, this analysis focused on environmental impacts of food choice, which is one factor related to food choices. A variety of factors, such as taste, safety, nutrition contents, affordability, availability, and environmental concerns may also influence food choices. Food consumption shifts based only on one factor is unlikely to happen. The reliability of estimating eutrophication potential reduction due to consumer behaviors’ changes is impaired when the importance of other factors have not been adequately addressed. Although an ideal tool which can quantify the influences of all factors is not available, simplified approaches still aid in understanding of the complex nature of food choices. Other solutions also exist to reduce eutrophication potential of foods, for example, optimizing farming practices during the production stage; improving food processing techniques; and implementing prevention strategies such as installing buffer strips, constructing wetlands, or using other water treatment facilities to remediate nutrient runoff. Trade-offs exist for every possible solution; a portfolio of solutions should be suggested to meet the requirements of abundant food supply, acceptable environmental impacts, and sustainable social and economic development.

6454

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

Acknowledgments We thank Dr. Melissa Bilec and Dr. Joe Marriot at University of Pittsburgh for their guidance and helpful suggestions.

Supporting Information Available Detailed discussion of assumptions and datasources and additional figures and tables. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Eshel, G.; Martin, P. A. Diet, energy, and global warming. Earth Interact. 2006, 10 (9), 1–17. (2) Marie-He´le`ne. Tusseau-Vuillemin, Do food processing industries contribute to the eutrophication of aquatic systems? Ecotox. Environ. Safe. 2001, 50 (2), 143–152. (3) Hertwich, E. G. Life-cycle approaches to sustainable consumption: A critical review. Environ. Sci. Technol. 2005, 39 (13), 4673– 4684. (4) Kroyer, G. T. Impact of food processing on the environmentsAn overview. Lebensm.-Wiss. Technol. 1995, 28 (6), 547–552. (5) Paxton, A. The Food Miles Report: the Dangers of Long Distance Food Transport; SAFE Alliance: London1994. (6) Pirog, R.; Van Pelt, T.; Enshayan, K.; Cook, E., Food, Fuel, and Freeways: An Iowa Perspective on How Far Food Travels, Fuel Usage, and Greenhouse Gas Emissions; Leopold Center for Sustainable Agriculture: Ames, IA, 2001; p 1-37. (7) Smith, A., The Validity of Food Miles as an Indicator of Sustainable Development.; UK DEFRA: London, 2005; 1-117. (8) Weber, C. L.; Matthews, H. S. Food-miles and the relative climate impacts of food choices in the united states. Environ. Sci. Technol. 2008, 42 (10), 3508–3513. (9) Anderson, D. M.; Gilbert, P. M.; Burkholder, J. M. Harmful algal blooms and eutrophication:nutrient sources, composition, and consequences. Estuar. Coast. 2002, 25 (4), 704–726. (10) Bricker, S. B.; Longstaff, B.; Dennesion, W.; Jones, A.; Boicourt, K.; Wicks, C.; Woerner, J. Effects of nutrient enrichment in the nation’s estuaries: a decade of change. Harmful Algae 2008, 8 (1), 21–32. (11) Smith, V. H.; G.D., T.; Nekola, J. C. Eutrophication: Impacts of excess nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environ. Pollut. 1999, 100 (3), 179–196. (12) Board, E. S. A. Hypoxia in Northern Gulf of Mexico. http:// yosemite.epa.gov/sab/sabproduct.nsf (accessed July 2009). (13) EPA Mississippi River/Gulf of Mexico watershed nutrient task force. Available at http://www.epa.gov/msbasin/taskforce/pdf/ 2008draft (accessed July 2009). (14) Sim, S.; Barry, M.; Clift, R.; Cowell, S. J. The relative importance of transport in determing and appropriate sustainability strategy for food sourcing. Int. J. Life-Cycle Assess. 2007, 12 (6), 422–431. (15) Steinfeld, H.; Gerber, P.; Wassenaar, T.; Castel, V.; Rosales, M.; de Haan, C., Livestock’s Long Shadow: Environmental Issues and Options; United Nations Food and Agricultural Organization: Rome, 2006; Vol. 1, p 408. (16) Tilman, D.; Fargione, J.; Wolff, B.; D’Antonio, C.; Dobson, A.; Howarth, R.; Schindler, D.; Schlesinger, W. H.; Simberloff, D.; Swackhamer, D. Forecasting agriculturally driven global environmental change. Science 2001, 292 (5515), 281–284. (17) Landis, A. E.; Miller, S. A.; Theis, T. L. Life cycle of the cornsoybean agroecosystem for biobased production. Environ. Sci. Technol. 2007, 41 (4), 1457–1464. (18) Miller, S. A.; Landis, A. E.; Theis, T. L. Feature: Environmental trade-offs of biobased production. Environ. Sci. Technol. 2007, 41 (15), 5176–5182. (19) Basset-Mens, C.; van der Werf, H. M. G. Scenario-based environmental assessment of farming systems: the case of pig production in France. Agr. Ecosyst. Environ. 2005, 105 (1-2), 127–144. (20) Berlin, J. Environmental life cycle assessment (LCA) of Swedish semi-hard cheese. Int. Dairy J. 2002, 12 (11), 939–953. (21) Cederberg, C.; Stadig, M. System expansion and allocation in life cycle assessment of milk and beef production. Int. J. LifeCycle Assess. 2003, 8 (6), 350–356. (22) Andersson, K. LCA of food products and production systems. Int. J. Life-Cycle Assess. 2000, 5 (4), 239–248. (23) Brentrup, F.; Ku ¨ ters, J.; Kuhlmann, H.; Lammel, J. Application of the life cycle assessment methodology to agricultural production: an example of sugar beet production with different forms of nitrogen fertilisers. Eur. J. Agron. 2001, 14 (3), 221–233.

(24) Brentrup, F.; Ku ¨ters, J.; Kuhlmann, H.; Lammel, J. Environmental impact assessment of agricultural production systems using the life cycle assessment methodology: I. Theoretical concept of a LCA method tailored to crop production. Eur. J. Agron. 2004, 20 (3), 247–264. (25) Eide, M. Life cycle assessment (LCA) of industrial milk production. Int. J. Life-Cycle Assess. 2002, 7 (2), 115–126. (26) Hospido, A.; Tyedmers, P. Life cycle environmental impacts of Spanish tuna fisheries. Fish. Res. 2005, 76 (2), 174–186. (27) Ziegler, F.; Nilsson, P.; Mattsson, B.; Walther, Y. Life cycle assessment of frozen cod fillets including fishery-specific environmental impacts. Int. J. Life-Cycle Assess. 2003, 8 (1), 39– 47. (28) Costello, C.; Griffin, W. M.; Landis, A. E.; Matthews, H. S., Impact of biofuel crop production on the formation of hypoxia in the Gulf of Mexico. Environ. Sci. Technol. 2009. (29) Miller, S. A.; Landis, A. E.; Theis, T. L. Use of Monte Carlo analysis to characterize nitrogen fluxes in agroecosystems. Environ. Sci. Technol. 2006, 40 (7), 2324–2332. (30) Busser, S.; Jungbluth, N. The role of flexible packaging in the life cycle of coffee and butter. Int. J. Life-Cycle Assess. 2009, 14 (1), 80–91. (31) Oki, Y.; Sasaki, H. Social and environmental impacts of packaging(LCA and assessment of packaging functions). Packag. Technol. Sci. 2000, 13 (2), 45–53. (32) Williams, H.; Wikstrom, F.; Lofgren, M. A life cycle perspective on environmental effects of customer focused packaging development. J. Cleaner. Prod. 2008, 16 (7), 853–859. (33) Bare, J. C.; Norris, G. A.; Pennington, D. W.; Makone, T. TRACIthe tool for the reduction and assessment of chemical and other environmental impacts. J. Ind. Ecol 2003, 6 (3), 49–78. (34) Burnham, A.; Wang, M.; Wu, Y., Development and applications of GREET 2.7sThe transportation vehicle-cycle model; Argonne National Laboratory: Argonne, IL, 2006; p 1-124. (35) ISO. Environmental Management-Life Cycle Assessments Principles and Framework, ISO 14040, 2006. (36) Schau, E.; Fet, A. LCA studies of food products as background for environmental product declarations. Int. J. Life-Cycle Assess. 2008, 13 (3), 255–264. (37) Pre´ consultants. SimaPro LCA software. Available at http:// www.pre.nl/simapro/ (accessed February 2010). (38) Frischknecht, R.; Jungbluth, N.; Althaus, H.-J.; Doka, G.; Dones, R.; Heck, T.; Hellweg, S.; Hischier, R.; Nemecek, T.; Rebitzer, G.; Spielmann, M. The ecoinvent database: Overview and methodological framework (7 pp). Int. J. Life-Cycle Assess. 2005, 10 (1), 3–9. (39) Danish Institue of Agricultural Sciences. LCA food database. Available at http://www.lcafood.dk/ (accessed February 2010). (40) Pre´ consultants. Industry data v2.0. Available at http://www. pre.nl/simapro/inventory_databases.htm#INDUSTRY (accessed February 2010). (41) Pre´ consultants. SimaPro 7 database manual of the Buwal 250 library. Available at http://www.pre.nl/download/manuals/ DatabaseManualBUWAL250.pdf (accessed February 2010). (42) Pre´ consultants. SimaPro 7 database manual of the Franklin US LCI 98 library. Available at http://www.pre.nl/download/ manuals/DatabaseManualFranklinUS98.pdf (accessed February 2010). (43) Robertson, G. L., Food Packaging: Principles and Practice; CRC Press: Boca Raton, FL, 2006. (44) USDA Food packaging guidelines. Available at http://www. fsis.usda.gov/factsheets/Meat_Packaging_Materials/index.asp (accessed June 2009). (45) Amon, B.; Amon, T.; Boxberger, J.; Alt, C. Emissions of NH3, N2O and CH4 from dairy cows housed in a farmyard manure tying stall (housing, manure storage, manure spreading). Nutr. Cycling Agroecosyst. 2001, 60 (1), 103–113. (46) Berges, M. G. M.; Crutzen, P. J. Estimates of global N2O emissions from cattle, pig and chicken manure, including a discussion of CH4 emissions. J. Atoms. Chem. 1996, 24 (3), 241–269. (47) Phetteplace, H.; Johnson, D.; Seidl, A. Greenhouse gas emissions from simulated beef and dairy livestock systems in the United States. Nutr. Cycling Agroecosyst. 2001, 60 (1), 99–102. (48) Strid Eriksson, I.; Elmquist, H.; Stern, S.; Nybrant, T. Environmental Systems Analysis of Pig ProductionsThe Impact of Feed Choice Int. J. Life-Cycle Assess. 2005, 10 (2), 143–154. (49) Danalewich, J. R.; Papagiannis, T. G.; Belyea, R. L.; Tumbleson, M. E.; Raskin, L. Characterization of dairy waste streams, current treatment practices, and potential for biological nutrient removal. Water Res. 1998, 32 (12), 3555–3568. VOL. 44, NO. 16, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6455

(50) Cederberg, C.; Mattsson, B. Life cycle assessment of milk production -- a comparison of conventional and organic farming. J. Cleaner Prod. 2000, 8 (1), 49–60. (51) USDA 2007 food prices. Available at http://www.census.gov/ compendia/statab/cats/prices/food_cost_and_prices.html (accessed August 2009). (52) USDA food energy and protein content, data require entry. Available at http://www.ers.usda.gov/Data/FoodConsumption/ (accessed August 2009). (53) Strid Eriksson, I.; Elmquist, H.; Stern, S.; Nybrant, T. Environmental systems analysis of pig productionsThe impact of feed choice. Int. J. Life-Cycle Assess. 2005, 10 (2), 143–154.

6456

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 16, 2010

(54) Kumm, K.-I. Ways to reduce nitrogen pollution from Swedish pork production. Nutr. Cycling Agroecosyst. 2003, 66 (3), 285–293. (55) Gehlhar, M.; Coyle, W. Changing structure of global food consumption and trade/ WRS-01-1. Available at http://www. ers.usda.gov/publications/wrs011/wrs011c.pdf (accessed August 2009). (56) USDA Profiling food consumption in America. Available at http://www.usda.gov/factbook/chapter2.pdf (accessed August 2009).

ES9034478