University of Minnesota to develop techniques for monitoring global change

$1.43 million NSF grant will address impact of land use and needs of growing population

MINNEAPOLIS / ST. PAUL (09/11/2018) — The University of Minnesota announced today that it has received a three-year, $1.43 million grant from the National Science Foundation to advance machine learning techniques to better monitor global agricultural and environmental change—a practice that can help society address the challenges of adapting to a changing climate, managing land use and natural resources, and sustainably feeding a growing population.

Machine learning, where computers “learn” from the data they collect without additional manual programming, is an effective tool for analyzing information about earth systems from multiple sources across time and space to study how natural processes and human activities affect the planet’s physical landscape and environment.

The NSF grant funds a team of researchers at the University’s College of Science and Engineering (CSE); College of Food, Agricultural and Natural Resource Sciences (CFANS); and Minnesota Supercomputing Institute (MSI) to advance the state-of-the-art in machine learning for analyzing data from earth-observing satellites and generating actionable information on a global scale.

“Growth in the world's population and the acceleration of industrialization and urbanization are straining already scarce natural resources and food supplies, which must scale up to keep pace with growing demand,” said Vipin Kumar, Ph.D., Regents Professor and William Norris Endowed Chair of computer science and engineering in CSE and principal investigator (PI) on the project. “Machine learning methods, which have already transformed many aspects of our lives including transportation and commerce, can also play a major role in addressing some of the greatest challenges facing humanity.”

Addressing challenges related to global agricultural and environmental change requires timely information on the varying productivity performance of agricultural land, the changing location of crop production, conversion of forest to farmland or plantations, the loss of productive farmland to urbanization, and soil and water degradation.

The project’s main focus will be to advance the state-of-art in machine learning techniques for analyzing spatial and temporal agricultural cropping data and urban landscapes. Such analyses can produce critical information needed for developing sustainable practices for increasing crop yields and for managing water run-off flows and quality. In particular, the team aims to develop and advance deep learning (a subfield of machine learning) techniques, to monitor these global changes by analyzing remote-sensing data obtained from satellites. Collaborators at the Nature Conservancy and DC Water will help evaluate the effectiveness of the machine learning techniques developed in this project.

“While deep learning techniques have already shown success in other areas with complex data sets, such as with computer image recognition, these techniques have been of limited use for agricultural and environmental applications,” said James Wilgenbusch, Ph.D., associate director of MSI and a project co-PI. “This project aims to take parts of the existing computing techniques and develop them into an approach that can better monitor global change.”

The project is one example of the University’s larger capabilities for using big data analytics to drive innovation in the areas of food and agriculture, which are based in the University’s GEMS agroinformatics platform. GEMS is developed through a partnership led by CFANS and MSI and now serves as a nexus for large sets of genetics, environmental, management and socioeconomic data. The platform does the hard work of making these different data types interoperable and provides the framework for complex analysis that guides decision-making in areas such as responding to emerging diseases, developing sustainable farming practices and increasing resource efficiency.

“GEMS is leading the way in developing and deploying data sharing and data analytic tools to forge innovative partnerships in food and agriculture, spanning the public and private sectors,” said Philip Pardey, Ph.D., professor of applied economics and director of global research strategy in CFANS and the other project co-PI. “This project will contribute to the ever-closer, transdisciplinary partnership between CFANS, MSI and CSE with its laser-like focus on spurring economically and ecologically sustainable growth in the Minnesota, US and global food and agriculture sectors through big data analytics.”

About G.E.M.S™

G.E.M.S™ is an international agroinformatics initiative jointly led by the College of Food, Agricultural and Natural Resource Sciences (CFANS) and the Minnesota Supercomputing Institute (MSI) at the University of Minnesota. G.E.M.S™, GEMSOpen™, GEMShare™ and GEMSTools™ are registered trademarks of the University of Minnesota. G.E.M.S™ is the first and the only system designed from the very start to support and functionally integrate spatially and temporally distributed genomic, environmental, management, and socioeconomic data in a single integrated platform.

For more information on GEMS, visit the GEMS website.