|KYUNG LEE, EUN - New York State University|
|ZHANG, WANG-JIAN - Global Change Research Institute|
|XUE, XIAOBO - New York State University|
|LIN, SHAO - New York State University|
|FEINGOLDA, BETH - New York State University|
|HAIDER, KHWAJA - New York State University|
|ROMEIKO, XIAOBO - New York State University|
Submitted to: Science of the Total Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/13/2020
Publication Date: 4/20/2020
Citation: Kyung Lee, E., Zhang, W., Adler, P.R., Xue, X., Lin, S., Feingolda, B.J., Haider, K.A., Romeiko, X.X. 2020. Projecting life-cycle environmental impacts of corn production in the U.S. Midwest under future climate scenarios using a machine learning approach. Science of the Total Environment. 714:1-11. https://doi.org/10.1016/j.scitotenv.2020.136697.
Interpretive Summary: Process based models are often used to describe the variation in environmental impacts of crop production across the landscape, but are very complicated, requiring expert knowledge and significant computing power for web-based applications. Using a computer machine learning approach, we developed a rapid predictive model capable of capturing current and future life cycle environmental impacts of corn production across the United States. This study demonstrates the potential to generate simplified models capable of being used by nonexperts in industry, to capture the spatial variability in environmental impacts of corn production for life cycle assessments used in carbon footprint as well as other environmental assessments.
Technical Abstract: Emission factors and process-based mechanism models are popular approaches used to estimate life-cycle environmental impacts. However, emission factors are often incapable of describing spatial and temporal heterogeneity of agricultural emissions, whereas process-based mechanistic models, capable of capturing the heterogeneity, tend to be very complicated and time-consuming. Efficient prediction of life-cycle environmental impacts from agricultural production is lacking. This study develops a rapid predictive model to quantify life-cycle global warming (GW), eutrophication (EU) and acidification (AD) impacts of corn production using a novel machine learning approach. We used the boosted regression tree (BRT) model to estimate current and future life-cycle environmental impacts of corn production in U.S. Midwest counties under four Representative Concentration Pathways (RCP) for years 2009-2100. Results show that future life-cycle GW, EU and AD impacts of corn production will increase in magnitude under all four scenarios, with the highest environmental impacts shown under the RCP8.5 scenario. Moreover, this study found significant spatial variation in all life-cycle impacts across Midwest counties and identified the key influencing factors explaining spatial heterogeneity of life-cycle impacts. Findings from this study demonstrate the importance of considering county-level life-cycle impacts and climate change in future years to aid in developing potential adaptation and mitigation programs/policies.