Cultivating Innovations at the Nexus of Food, Energy, and Water Systems (INFEWS): Winter Annual Oil Seed Crops for Water Quality and Economic Productivity
“Innovations for Sustainable Food, Energy, And Water Supplies In Intensively Cultivated Regions: Integrating Technologies, Data, And Human Behavior”
- NSF Project description
- NSF Video - “Using technology, data and human behavior to drive change in cultivated regions”
- CFANS Video/”Cover” story
To keep pace with the demands of a growing global population, innovations are needed to meet the unprecedented challenge of producing more food in intensively cultivated regions with less energy and lower environmental impacts. In this project, researchers from the biophysical, socioeconomic, and computational sciences investigate two types of innovations using data from the northern U.S. Corn Belt. First, a novel oilseed crop, winter camelina, is being studied for its potential incorporation into existing corn-soybean rotations to produce a new supply of biodiesel energy while lowering water resource impacts and creating positive ecological benefits. Second, emerging systems of sustainability certification are being studied for their potential to lead to broad-scale adoption of this new cropping system. Detailed computational models are being evaluated and applied for systems-level assessments of two innovations: developing novel approaches to influence beneficial land use, and accounting for energy and environmental impacts within food supply chains. Because of the importance of the project results on the local economy, outreach activities are targeted towards the rural community, policymakers, the general public, and local watershed planners. Although this project focuses on the Northern Corn Belt, the approaches used in the research could be adopted to achieve beneficial outcomes for food, energy, and water systems elsewhere.
This research project is comprised of four overlapping and interdependent research teams. The biophysical research team conducts cropping systems studies of corn-soybean in rotation with winter camelina at two Minnesota research stations. Experimental treatments vary by winter timing of planting and harvest and fertilization rates. Meteorological data along with soil, water, and crops data is being collected to develop management strategies for producers and to calibrate and evaluate the crop models that will be used. The socioeconomic research team collects data from surveys and randomized control trials to study the forces determining whether, and by whom, new cropping systems are likely to be adopted under different policy and market conditions. Of particular interest is the role of incentives from certification programs, including the feedback effects of using producer data for peer benchmarking. The data science team applies novel deep learning computational approaches to identify crops, including winter cover crops, from satellite imagery. Study plots for the biophysical plots provides training data for crop identification and the final statewide datasets are being incorporated in the socioeconomic analysis. Finally, the integrated modeling team develops a suite of connected modeling tools to quantify systems-level outcomes. These simulations shed light on the feasibility and impacts of innovations in the food-energy-water system under different scenarios, including spatial patterns and the role of socio-economic drivers.
This project is investigating innovative approaches to support sustainable supplies of food, energy, and water in intensively cultivated regions. We are focused on specific, high-impact innovations along two related frontiers: using new crops and technologies and inducing behavioral change. Our work is divided into four areas, each with a specific goal. The first is Biophysical Research, where our goal is to understand the biophysical processes affecting innovations in corn and soybean cropping systems in a humid and temperate environment. The second is Data Science, where our goal is to distinguish different crop types at varying spatial and temporal resolutions using machine learning and data from earth-observing satellites. The third is Socioeconomic Research, where our goal is to quantify the factors influencing the adoption of cropping systems and to develop spatially distributed predictive models of landholder behavior. The final area is the intersection of the previous three -- Integrative Modeling, where our goal to quantify systems-level behavior, accounting for the causal relationships and feedbacks within and between socioeconomic and biophysical systems.
Findings from field research are expected to advance the understanding of the potential of diversified cropping practices in northern climates. Data science approaches developed in this project will find use in many other applications that have data available at multiple spatial and temporal resolutions. Survey research will advance understanding of the drivers of and constraints to the adoption of cover crops. Qualitative findings will document farmer perspectives about cover crop adoption. Finally, the modeling work in this project will develop techniques for spatial scaling in economics, watershed modeling, and life cycle analysis.
Jia, Xiaowei and Wang, Mengdie and Khandelwal, Ankush and Karpatne, Anuj and Kumar, Vipin "Recurrent Generative Networks for Multi-Resolution Satellite Data: An Application in Cropland Monitoring" Recurrent Generative Networks for Multi-Resolution Satellite Data: An Application in Cropland Monitoring, 2019
Jia, Xiaowei, Ankush Khandelwal, David J Mulla, Philip G Pardey, Vipin Kumar. “Bringing automated, remote‐sensed, machine learning methods to monitoring crop landscapes at scale”. Agricultural Economics, Volume 50, Issue S1, October 2019.
Peterson, Jeffrey M. "Innovation as a policy strategy for natural resource protection" Natural Resource Modelin, 2019
Support for this work is provided by the National Science Foundation Innovations at the Nexus of Food, Energy and Water Systems (INFEWS) Program, Award #1739191