Yield Forecast Center provides producers with real-time, reliable yield data

September 27, 2016

Lincoln, Neb. — In recent years, producers in the Corn Belt have experienced the extremes in weather and crop yields. Following the devastating 2012 drought,  corn yields were the worst in 30 years. Two years later, yields reached all-time record highs in Nebraska. Facing such variability, corn producers and agribusiness have looked to the Yield Forecast Center developed at the University of Nebraska–Lincoln for accurate corn yield predictions to help guide in-season decision making.

The biweekly model forecasts yield for 41 sites in 10 states across the Corn Belt, including Illinois, Indiana, Iowa, Kansas, Michigan, Missouri, Minnesota, Nebraska, North Dakota and Ohio. For each site the forecast considers over 20 years of historical yield data on yield, soil property, water regime and crop maturity as well as current data on solar radiation, minimum and maximum air temperatures, rainfall and plant evapotranspiration.

Patricio Grassini, assistant professor in the Department of Agronomy and Horticulture, started releasing corn yield forecasts in 2011. He initially focused on four sites in Nebraska before taking his model to sites in Iowa and Illinois. As the in-season forecasts proved accurate when compared to end-of-season actual harvest yields and demand for the forecasts grew, Grassini formalized the Yield Forecast Center, and built a team of faculty and extension educators from across 10 universities. The approach to forecast corn yield consists of robust simulations that capture the influence of weather, soil, and management on crop growth and yield, together with a spatial framework to upscale results from location to state and region.

“The approach is transparent. It has a strong agronomic foundation and it has been validated on its ability to reproduce observed yields,” said Grassini. A journal publication describing the Yield Forecast Center approach has been recently published in the international, peer-reviewed Field Crops Research journal.

Beyond yield, UNL’s model predicts real-time crop stage and also forecasts the date of maturity. It’s helpful for producers, grain elevators, trucking companies and others to know when the crop will mature so they can prepare for harvest.

“Free, open-access information is in high demand, and we believe that the Yield Forecast Center helps level the playing field for everybody who trades grain in the open market,” said Grassini. “Farmers, crop consultants and others in the industry are using our model to inform management, logistical and marketing decisions.” 

Another yield forecast is distributed by the U.S. Department of Agriculture’s National Agricultural Statistics Service (USDA-NASS). USDA-NASS predicts crop yields by state on a monthly basis. According to that model, corn producers are expected to harvest a record crop this year, which slightly differs from what the UNL Yield Forecast Center projects for 2016. 

According to Grassini, the UNL Yield Forecast Center is not seeing solid evidence of a record corn yield this year. The model currently projects a yield slightly above average, but not a definitive record crop. Despite the differences in the forecast, Grassini views the UNL model as complementary to the USDA-NASS forecast model.

“Our approach can help develop forecasts at a higher level of spatial and temporal resolution and provide a probability associated with each forecast instead of a single figure for each state once a month as currently provided by USDA-NASS,” said Grassini. 

As of Sept. 21 the UNL hybrid-maize model is forecasting 200 bushels of corn per acre for irrigated cropland in Nebraska, a yield slightly higher than average. For rainfed cropland, the model is forecasting 151 bushels of corn per acre, also higher than average. Both of these forecasts are lower than the historical record yield.

A significant amount of data is collected to develop the yield forecasts. For each forecast, faculty and extension educators from all partner universities provide input on their crop stage. Assisting Grassini to run the forecasts at UNL are Gonzalo Rizzo, a visiting scholar from the University of the Republica (Uruguay), and Juan Ignacio Rattalino Edreira, a UNL post-doctoral research associate. Others involved with the project in Nebraska include Ken Cassman, Emeritus Robert B. Daugherty professor in the Department of Agronomy and Horticulture;  Roger Elmore, extension cropping systems agronomist and professor in the Department of Agronomy and Horticulture; Keith Glewen, extension educator; Jenny Rees, extension educator; Charles Shapiro, professor in the Department of Agronomy and Horticulture; and Haishun Yang, associate professor in the Department of Agronomy and Horticulture.

Yield forecasts are produced every two weeks during the corn crop season, starting by mid-July, and are published in CropWatch, a central resource for information on crop production and pest management from Nebraska Extension. To view the latest corn yield forecast, visit http://cropwatch.unl.edu/tags/corn-yield-forecasts.

Author: Haley Steinkuhler