In Season Yield Forecasting for Early and Late Planted Soybean in Iowa.

Accurately forecasting crop yield in advance of harvest could greatly benefit farmers. We used the Weather Research and Forecasting Model (WRF) to predict in season weather variables, such as, high and low temperature, precipitation and radiation combined with the Agricultural Production Systems sIMulator (APSIM) to translate the weather forecast to soybean growth and final yield every second week during the 2015 season in Iowa. The Global Forecast System was used to provide boundary conditions for the WRF model. Nested domains have been used in the WRF model with the inner domain being focused over the Midwestern United States with a grid spacing of 17 km. This forecasted weather data along with the current and historic (previous 35 years) data from the Iowa Environmental Mesonet was combined to form the weather file to drive APSIM simulations of soybean production. APSIM predictions of crop variables were evaluated against bi-weekly field measurements of phenology, biomass, and leaf area index from plots located at the Agricultural Engineering and Agronomy Research Farm in central Iowa as well as the Northwest Research Farm in northwestern Iowa. At each farm there were six plots of soybean with early and late planting dates (n=3, May 1^st and June 1^st respectively). WRF model predictions (14-days) were evaluated against actual weather data collected at the experimental fields. Current results up to August 2015 from both models showed a good agreement with the in-season crop and weather measurements. An overall assessment of both modeling platforms and their interaction is currently being performed following the recent harvest.