ACCESS Resources Provide New Forecasting Tool for Farmers

By Kimberly Mann Bruch, SDSC

Researchers at the University of Maryland (UMD) utilized two U.S. National Science Foundation (NSF) ACCESS resources to develop a unique decision-support dashboard by combining coupled climate-crop system modeling with emerging artificial intelligence (AI) technology that provides the agriculture community with multi-disciplinary insight on their specific farming area’s climate and crop prediction at a six-month lead time. Rockfish at Johns Hopkins University and Bridges-2 at Pittsburgh Supercomputing Center provided the platforms for the development of the team’s Dashboard for Agricultural Water Use and Nutrient Management – or DAWN – which enhances NOAA’s operational seasonal climate forecasts and was recently published in the Bulletin of the American Meteorological Society

“The complex algorithms and large memory required to run DAWN wouldn’t have been possible on our local machines; hence, we reached out to the NSF for ACCESS allocations,” said Chao Sun, an assistant research scientist at UMD. “Our main goal with the tool is to provide farmers with comprehensive predictions at six-months’ lead time alongside historical data so that they can make solid decisions regarding their field operation and crop production – Bridges-2 and Rockfish provided us with the power needed for our work.”

By providing the nation’s agriculture community with better forecasting tools, the NSF ACCESS allocations are helping us bridge the gap between farmers and agriculture research.

Chao Sun, University of Maryland Assistant Research Scientist

Why is so much “power” needed for this type of tool? Sun explained that DAWN uses regional high-resolution, climate-crop-coupled numerical model predictions combined with complex AI algorithms to contrast the operational climate-only forecasts with historical data. “We are simulating what is going to happen at six-months’ lead time in the atmosphere and cropland over the entire continental U.S. area, as well as the Gulf of Mexico, with a high-resolution grid,” Sun said. “With that information, our algorithm looks at past crop progress and compares that with predictive future situations based on precipitation, temperature, irrigation, fertilizer and more. By providing the nation’s agriculture community with better forecasting tools, the NSF ACCESS allocations are helping us bridge the gap between farmers and agriculture research.”

The ACCESS-enabled DAWN dashboard, which was developed at the University of Maryland, is an agriculture tool that encompasses growing degree days, crop progress, field identification, and a data viewer.
The ACCESS-enabled DAWN dashboard, which was developed at the University of Maryland, is an agriculture tool that encompasses growing degree days, crop progress, field identification, and a data viewer. Credit: University of Maryland

UMD Professor of Atmospheric Science Xin-Zhong Liang was the first author on the paper and principal investigator for the DAWN work. He also attributed much of the success of the project to the supercomputing resources Bridges-2 and Rockfish. “Without access to Bridges-2 and Rockfish, we would not have been able to conduct this work,” he said. “In addition to the facets mentioned by my colleague, our dashboard tools are co-produced with farmers so that they are tailored to the need and effective use for agricultural decision making.”

He explained that the team ran experiments on ACCESS-allocated supercomputers every five days and that wouldn’t have been possible on the computers they have in their lab at UMD. Now, they’re working with the farming community to perfect the dashboard tools and have plans to roll out the system for public use by 2025. He said the initial tools have been created specifically for corn crops, but will soon include soybean growers.


Resource Provider Institution(s): Pittsburgh Supercomputing Center (PSC), Johns Hopkins University (JHU)
Affiliations: University of Maryland
Funding Agency: DAWN is funded by the U.S. Department of Agriculture National Institute of Food and Agriculture (grant no. 2020-68012-31674). Computational work was funded by the NSF ACCESS program (grant no. EES210017). Additional computation support was provided by the Department of Energy Oak Ridge Leadership Computing Facility.
Grant or Allocation Number(s): EES210017, 2020-68012-31674

The science story featured here was enabled by the U.S. National Science Foundation’s ACCESS program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296.

Sign up for ACCESS news and updates.

Receive our monthly newsletter with ACCESS program news in your inbox. Read past issues.