Evaluating the sustainability of emerging agricultural systems

Abstract

As the global community seeks to balance rising demands for food and energy with the planet's ecological limits, stakeholders must scrutinize production systems across sectors for opportunities to innovate and reduce impacts. Foremost among these challenges is the threat of climate change driven by greenhouse gas (GHG) emissions. In order to address these emissions, the agricultural sector will play a role as producers mitigate GHG's associated with essential food production while also rising to meet other decarbonization initiatives such as biofuel usage. To quantify these impacts, life cycle assessment (LCA) is used as a well-established method that captures the flows of resources and pollution across supply chains, translating them into impacts. These LCA results inform stakeholders of ecological tradeoffs and mitigation opportunities. By leveraging more advanced methods to differentiate between locations and to model future conditions, these tools can steer the consideration and development of emerging systems towards eco-efficient outcomes. In this dissertation, LCA is utilized in three phases to evaluate emerging agricultural systems, assessing their impacts and identifying sustainability strategies. In the first research phase, geographically-resolved LCA was applied to compare climate impacts and water usages of new local food production systems across the contiguous United States with a centralized supply chain. Using leaf lettuce as the study crop, hydroponic systems were modeled using building energy demand software to simulate indoor plant factories and greenhouses under different local climate and grid conditions. Additionally, crop modeling software was utilized to simulate seasonal lettuce production on farms at each location. Finally, these localized systems were compared to a modeled conventional system of California field cultivation and shipping. Results across all sites indicate that indoor production systems have substantially higher GHG emissions than the conventional supply chain owing to energy consumption from heating and dehumidification demand. Thus, consumers seeking low-impact options should eat from local farms when in-season and otherwise use conventional supply chains. However, local stakeholders may consider a food-climate-water tradeoff, since indoor hydroponic systems use far less water than outdoor systems. Technology adoption scenarios are also considered to evaluate how heating electrification and decarbonized electricity affect the GHG outcomes of indoor cultivation, providing insights for operators in this emerging sector. Continuing to leverage geographic resolution in LCA, the second research phase considers an emerging biofuel feedstock production supply chain at the field-by-field level. In partnership with a company building a plant to convert ethanol to jet fuel, corn fields across South Dakota and Minnesota were evaluated for the 2023 harvest. Utilizing upstream supply chain modeling and the Daycent biogeochemical soil model, localized cultivation practices were translated into farm-gate carbon intensities. Stochastic modeling was then applied to compare results across hundreds of sample sites to default modeling assumptions typically utilized in biofuel LCA. Results demonstrate that this supply chain, on average, achieves lower GHG emissions than the default model suggests, stemming from local variations in energy usage, agrochemical application, and soil conservation. Further analysis suggests that producers focus on nitrogen fertilizer efficiency and land management initiatives with a particular need to ensure accurate, field-level modeling of soil dynamics. Alongside downstream decarbonization efforts, such cultivation-stage initiatives can contribute to creating aviation fuel that meets clean energy standards. While geography-specific LCA insights are useful to understand agricultural emissions, developments across time are also poised to make significant changes in the sector. Thus, in the third research phase, dynamic LCA (DLCA) methods were applied to the outdoor lettuce cultivation model, incorporating modeled ecospheric and technospheric transformations from the present-day to 2050. Dynamic process modeling was used to evaluate how changing climate could affect crop growth. Meanwhile, background technospheric transformation and on-farm technology adoption were modeled to consider how decarbonizing supply chains and zero-emissions equipment could mitigate life cycle GHG emissions. DLCA baseline results show that 2050 emissions will be substantially reduced compared to present-day assumptions; in particular, the deployment of electrified irrigation, combined with decarbonizing generation, provides a near-term avenue for mitigation. In more optimistic technology change scenarios, emerging technologies like green-hydrogen-derived fertilizer and battery electric heavy machinery could provide further emissions reductions; stakeholders could take steps in the present to support the development and adoption of these systems, enabling ecoefficiency gains in food production. Throughout this work, standard and enhanced LCA methods were employed to evaluate the sustainability of emerging agricultural systems that could one day meet global food and fuel demands. The results of these assessments provide quantifiable estimates that support inter-system comparisons and process improvement strategies. Thus, these methods and results can support decision-making for stakeholders across the supply chain to invest in a more sustainable agricultural future.