Regional Differences in Water Quality Impacts from the Bioenergy Mandate: A Scenario-Based Approach to Quantifying the Impacts from RFS2

Abstract

The Renewable Fuels Standard (RFS2) under the Energy Independence and Security Act (EISA 2007) mandates the production of 36 billion gallons of biofuel by 2022 for use in the transportation sector. Although the mandate seeks to reduce greenhouse gas emissions by reducing our dependence on oil, there is concern that consequent land-use/cover (LULC) changes could result in significant unintended environmental impacts. The evaluation of water quality impacts from the mandate is challenging, especially for cellulosic and other advanced biofuels because the bioenergy supply chain is an emerging system with inherent uncertainties.

This research addresses several gaps in the literature on the water quality impacts of the mandate and is organized into three chapters. The first chapter addresses the regional differences in impacts to water quality from a similar land-use change for two watersheds in the Midwest (Upper Cedar) and Southeast (Lumber). The Soil and Water Assessment Tool, a hydrological model that is able to simulate crop growth and water and nutrient outputs is set up, calibrated and validated. The potential reduction in nitrogen loading per potential gallon of ethanol to surface waters, for a change from baseline corn/soy to switchgrass, is about 40% in the Midwest and around 80% in the Southeast. Although, the trend in reduction is similar in both watersheds, this study shows that results extrapolated from the Midwest, where a lot of the bioenergy literature is based, may not be representative of other bioenergy producing regions.

The second chapter investigates the impact of uncertainties in production costs of perennials on three objectives incentivizing different aspects of the biofuel industry – maximizing farmer profitability, surplus from feedstock production and ethanol production and land-use efficiency, using a Monte-Carlo Analysis. The study also investigates the impact of current farmer safety net for corn-soy production and bioenergy subsidies on the feedstock choice in each region. The analysis indicates that the three objectives result in different feedstock options in the two regions. Further, results indicate that the current incentive structure for perennial biomass is insufficient to encourage production on cropland, especially because the commodity crop alternative has better risk management.

The final chapter of the dissertation links the feedstock production and ethanol production stages of the bioenergy supply chain. A Mixed Integer Linear Programming (MILP) optimization is used to drive land-use change at the Hydrologic Response Unit (HRU) level for the three objectives investigated in the second chapter. Results indicate there are tradeoffs between profitability and water quality for the three objectives. When the location of the biorefinery was considered, the supply of biomass and changes to water quality were localized around it, and these changes were significant at the sub-watershed scale. Optimization of biorefineries is usually done at the county or other large administrative scales. Our results indicate that such a scale would miss the localization of impacts which could be especially sub-optimal for sensitive watersheds. Any optimization of the biofuel supply chain system for water quality will therefore have to be at a watershed or sub-watershed resolution. A limitation of this work is the use of a single water quality indicator to quantify water quality impacts. Optimizing the supply chain must also involve development of appropriate multi-metric indicators of water quality for the region under consideration.