New Model Said To Predict National Yields Better Than USDA

06:49PM Oct 01, 2018
Chrysanthemum
( )

Market-moving USDA reports are being rivaled by a new formula from the University of Illinois. The university shows they’re closer to actual yield in early- and late-season reports.

“Using seasonal forecasts and satellite data, we developed a very advanced yield prediction system for both the national and county levels,” said Kaiyu Guan. “Our research demonstrates that we can do better than the USDA’s real-time estimation.”

Guan is a principal investigator on the Geophysical Research Letters study, and an assistant professor in the Department of Natural Resources and Environmental Sciences (NRES) at University of Illinois, and a Blue Waters professor at the National Center for Supercomputing Applications (NCSA).

The University of Illinois team is combining satellite data to predict yield combined with seasonal climate prediction and crop growth information from satellite imagery. In research they evaluated end-of-season accuracy of individual and combined data sources compared to national corn yield forecasts in monthly USDA-WASDE reports.

This approach leads to earlier, more accurate end-of-season yield predictions. For example, between 2010 and 2016 the WASDE report for June was off by an average of 17.66 bu. per acre. In that same time, this new system was only off by 12.75 bu. per acre. In August, WASDE was off by 5.63 bu. per acre and the new system was only off by 4.36 bu. per acre.

"Compared with using historical climate information for the unknown future, which is what most previous research is based on, using seasonal climate prediction from the NOAA's National Centers for Environmental Prediction gave better forecasting performance, especially in reducing the uncertainties," said Bin Peng, the lead author of this study and a post-doctoral research associate in NRES and NCSA.

 

It seems satellite imagery is the key to beating USDA estimates.

 

“If we only use seasonal climate prediction data—temperature, rainfall, and vapor pressure deficit—our predictions were no better than the USDA's,” Guan said. “It was only when we added the satellite data that we started to see the improvement. That's a clear indication that satellite data is extremely useful in this case."

Knowing better predictions of yield is important to farmers as they create their marketing plans. It could also benefit grain companies and local elevators to give them a better idea of what kind of storage and sales opportunities they might have at the end of the season.