8:00am - 8:20am
Food Choices and the Food-Energy-Water Nexus: Evaluating Energy Demand, Water Scarcity and Carbon Footprints of Self-Selected Diets in the U.S.
1Center for Sustainable Systems, University of Michigan, United States of America; 2Department of Global Community Health, Tulane University, United States of America
Food choice has been implicated as an important driver of the environmental impacts of food systems. Assessment of diet-level environmental impact across multiple impact categories can inform not only consumption-oriented abatement strategies, such as dietary modifications, but also identify aspects of food production systems that warrant further attention.
In previous work, we developed dataFIELD (database of Food Impacts on the Environment for Linking to Diets) based on an exhaustive review of the food life cycle assessment literature. DataFIELD contains greenhouse gas emissions (GHGE) and cumulative energy demand (CED) associated with production of 332 food commodities, and was linked to dietary recall data from the 2005-2010 National Health and Nutrition Examination Survey (NHANES), a representative survey of U.S. dietary intake. In this work, we add a spatially explicit assessment of the blue water use (surface and ground water used for irrigation) associated with food production, including a characterization of water scarcity, and link this to self-selected diets from NHANES.
Blue water consumption per ton of crop at the watershed level across the U.S. was obtained from a dataset compiled by Pfister and Bayer (DOI: 10.17632/brn4xm47jk.2). To characterize water scarcity, we used the Available Water Remaining (AWARE) method, with characterization factors also at the watershed level. Water consumption and water scarcity (consumption x characterization) per crop were aggregated to the national level using production-based weights, also available in the Pfister and Bayer dataset. Since not all food consumed in the U.S. is domestically sourced, water use and water scarcity footprints (WSF) per crop were adjusted for imports using FAO detailed trade matrix data and country-of-origin national average water consumption and AWARE factors. Water use for animal based foods were based on simplified feed rations from Peters et al. (DOI: 10.12952/journal.elementa.000116). Only water use associated with fee requirements of farmed fish and seafood were included.
At the population average, meats, dairy, beverages and fish & seafood were the top contributing food groups in the U.S. diet for both GHGE and CED. For WSF, meat was also the highest contributor, followed by fruits, beverages, dairy and vegetables. On average, diets higher in GHGE were also higher in CED and WSF. Such correlation of impacts across individual diets was the strongest between GHGE and CED (r=0.68, p < 0.01) and next strongest between GHGE and WSF (r=0.45, p < 0.01). Rankings based on GHGE demonstrated that the top 20% of U.S. diets are responsible for 46% of emissions. Impacts for the top quintile of diets when ranked on energy demand or water scarcity footprint were 43% or 42%, respectively. 7% of diets evaluated were in the 5th quintile (highest impact) of all three metrics, whereas 9% were consistently in the 1st quintile.
Consideration of the food-energy-water nexus from a diet perspective offers valuable insight into ways that consumption behaviors influence environmental impacts of production systems. Understanding such relationships between food choices and the multiple dimensions of environmental impact allows for targeting policy work to address dietary hot-spots.
8:20am - 8:40am
Optimizing the environmental performance of food diets in Peru using Linear Programming and Life Cycle Assessment
Pontificia Universidad Católica del Perú, Peru
Food production and security has been highlighted as one of the most threatened sectors by the consequences of climate change. However, it is also true that food production itself is responsible for an important fraction of greenhouse gas (GHG) emissions. Hence, GHG emissions derived from food production and dietary patterns have been analyzed in many areas of the world from a life-cycle perspective, mainly in North America and Europe, but remain unexplored in many developing nations. A recent study by Vázquez-Rowe et al. (2017) applied a life cycle approach to identify the GHG emissions linked to dietary patterns in Peru. Results show that an average Peruvian generates 1.08 t CO2eq per year due to food expenditure, a value that is notably lower than those reported in other nations. However, a series of worrying tendencies are visible behind these numbers. For instance, in cities in the Amazon basin, consumption of fruit and vegetables is up to 85% lower than recommended by national authorities. Several studies have delved into ways in which diets in developing nations should be improved in terms of nutrition and health. These studies are then used for policy support to generate impact in communities were food education and access to healthy food products may be limited. Unfortunately, however, fewer are the studies that combine nutritional and environmental benefits of diets. Therefore, the main objective was to propose a methodology in which LCA-related results linked to dietary patterns in Peru were combined with nutritional and economic data to optimize diets. For this, a linear programming model was built in which the environmental, nutritional and economic information on a set of 25 dietary patterns in Peru were optimized in order to achieve the environmentally best-performing diet that would not imply an increase in the household’s food basket and would improve the nutritional balance of the diet. One of the scenarios represented the Peruvian average, whereas the remaining scenarios represented the average diet in selected cities across the country. The result of the proposed linear program allowed understanding the amount of each individual food product that should be consumed in each city that satisfy all the restrictions included in the model in order to attain the lowest GHG emissions possible. Results demonstrated that GHG reductions can be attained through optimization. In fact, if results obtained by Vázquez-Rowe et al. (2017) presented a range between 792 and 1,350 kg CO2eq per person and year, the optimization model applied would allow a reduction to 582-961 kg CO2eq. For instance, in the city of Lima the reduction would be of 200 kg CO2eq per year (22% less). However, it should be noted that this 22% decrease in environmental impacts would be counterbalanced with a 12% increase in the economic cost of the food basket. Considering that in most areas of the country food purchase accounts for approximately 50% of household expenditure, it is plausible to assume that food choice is a main carrier to achieve GHG emission mitigations. In this context, the method constitutes a useful tool for policy-makers to push forward joint regulations to improve health-related issues linked to the food diet and food choice together with recommendations to lower the climatic impact of diets.
8:40am - 9:00am
Carbon sequestration, greenhouse gas mitigation, and hydrologic benefit potential of land applying bioenergy byproducts derived from food waste
Berkeley Laboratory, United States of America
Large scale improvements to soil carbon (C) in grasslands could play an important role in lowering greenhouse gas emissions and improving vital soil properties such as water holding capacity and net primary productivity. Practices which enhance soil C may also impact the bioenergy industry’s future, as the gasification and anaerobic digestion of lingocellulosic materials, biomass residues such as food waste, manure, and wastewater produce biochar and digestate which require large-scale, low-carbon disposal pathways. Here, we present a novel approach for conducting an attributional life-cycle assessment of biochar and raw or composted digestate disposal on grasslands, and demonstrate our methodology to evaluate the implications of large-scale bioenergy infrastructure and byproduct generation in California. Developed in R software, our code can be used to approximate life-cycle emissions and C accumulation from raw or composted digestate, or biochar disposal on marginal land for any region provided the user has details on biomass production and transportation distances. We find that while soil C accumulation and net negative life-cycle greenhouse gas emissions are possible for raw or composted digestate, and biochar application on critical rangelands in California, varying key parameters for byproduct quality and soil response can lead to net positive life-cycle greenhouse gas emissions. Furthermore, although we can estimate significant carbon sequestration and hydrological benefits by using an active land management strategy for 6% of California’s rangeland, such an estimate would require C application rates that exhaust the digestate and biochar that could potentially be generated from the State's biomass residues such as food waste. However, by using technically available biochar and digestate for land disposal generated each year with a more moderate C application rate, we could treat 12% of the most critical rangelands in California over a period of 9 years, generating a cumulative benefit of negative 120 MMTCO2eq. Improving our understanding of the sustainable use and disposal of byproducts such as digestate, compost, and biochar generated during the conversion of food waste to energy is critical. This is especially true in areas such as California, where emerging actors in the bioeconomy rely on accurate estimation of carbon offsets to qualify for low-carbon fuel and renewable fuel credits.
9:00am - 9:20am
Spatial optimization of food waste recycling infrastructure in California
1Department of Food Science and Technology, University of California, Davis, United States of America; 2Geography Graduate Group, University of California, Davis, United States of America
California Senate Bill (SB) 1383 calls for diverting 75% organic waste from landfills by 2030 as part of a larger mandate to reduce greenhouse gas (GHG) emissions. Current alternative treatment facilities do not have enough capacity to treat this food waste (FW), so an expansion of treatment infrastructure will be required. Most often, waste treatment facilities are located away from urban areas and require increased trucking mileage to transport FW, which translates into increased transportation GHG emissions and energy use. This linkage is exacerbated by ongoing and rapid urbanization. To develop solutions for the mitigation of GHG emissions from FW, the amount of FW currently generated, its current disposal pathways, and its spatial distribution need to be assessed. In this study, we attempt to understand how the spatial distribution of FW generated throughout California affects its ability to be treated and reduce GHG emissions and energy use compared to current disposal practices.
Using the spatial distribution of FW generated, this study will develop a strategy for developing new treatment infrastructures. The strategy will include an assessment of the optimal location and scale of alternative treatment methods, with a focus on anaerobic digestion (AD). AD technology converts organic waste into biogas, which can be transformed into useable energy forms, and nutrient-rich digestate and is a well-established alternative to landfilling. California has existing AD facilities, but more will be needed to handle the substantial increase in FW diversion mandated by SB 1383.
Small-scale, containerized digesters are a new technology option that can be rapidly deployed to treat the expected diverted waste from more localized FW catchments, thereby reducing transportation energy costs and associated GHG emissions. These systems can produce energy on-site to offset energy consumption and reduce transmission losses to the grid. In addition to siting and establishing new treatment facilities of various scales to treat FW streams, markets for food waste-derived digestate (FWDD) need to be established to absorb this nutrient-rich outflow from the AD process. FWDD has higher moisture and organic matter content and when land applied, it can reduce water inputs and increase the water holding capacity of soils. Understanding this spatial distribution of FW generation and potential FWDD markets will inform strategies for designing an effective infrastructure network for FW treatment in California.
Objectives of this study are to determine the quantity and spatial variation in FW generation and use this data to model potential regional waste management infrastructure that minimizes GHG emissions. Waste characterization studies produced by the California Department of Resources Recycling and Recovery (CalRecycle) will provide the baseline FW generation data for both California counties and cities, differentiated by the commercial business groups and the residential sector. We will map the spatial distribution of these data and develop strategies to identify hot-spot locations, generators with the largest contribution, and the homogeneity of FW generation rates across regions and generators. This analysis will inform the identification of the optimal scale and locations for an expanded network of AD treatment facilities in California.