Much is known about the ecophysiological processes underlying wheat yield and models exist which can integrate these processes to simulate yield well when management, cultivar, soil and weather conditions can be adequately characterized. Characterizing all of these factors accurately remains a challenge at a field scale, which has hitherto limited the scope of simulation analyses across geographic areas. However, the increasing availability of publicly available data (including satellite imagery, soil mapping, and automated weather network data) has greatly improved the feasibility of mesoscale simulation analysis. This presentation will review recent results from the Computational Agricultural Research Lab at Oklahoma State University which characterize spatiotemporal trends in Oklahoma wheat yields using an interface that links gridded weather and soil datasets directly with crop models. An overview of the simulation framework including the models and input data sets used, as well as an analysis of the drivers of yield and sources of uncertainty in simulations will be discussed. 

Presenter(s)

Phillip Alderman