Forecasting crop yields and water-nitrogen dynamics during the growing cycle of the crops can greatly advance in-season decision making processes. To date, forecasting approaches include the use of statistical or mechanistic simulation models, aerial images, or combinations of these to make the predictions. Different approaches and models have different capabilities, strengths, and limitations. System-level mechanistic simulation models (crop and soil models together) usually offer more prediction and explanatory power at the cost of extensive input data. In contrast, statistical approaches or aerial images can be more robust than mechanistic models but their applicability and prediction/explanatory power is limited. The combination of these technologies is viewed as a very promising tool to assist Midwestern agriculture, but in general, all of these technologies are in their initial stages of implementation and more time is needed to prove their potential. Here we present results from a pilot project that aimed to forecast weather, soil water-nitrogen status, crop water-nitrogen demand, and end-of-season crop yields in Iowa using two process-based mechanistic simulation models.