Using field scale models to predict peak flows on agricultural watersheds

The goal of this study was to develop a methodology for generating storm hydrographs at a watershed scale based on daily runoff estimates from a field scale model. The methodology was evaluated on a small agricultural watershed using the ADAPT field scale process model. A comparison of observed and predicted peak flows for 11 of the largest events that occurred in a three year period gave r(2) values of 0.84, 0.82, and 0.81 when the watershed was subdivided into 1, 5, and 10 subwatersheds. However, all other statistical measures improved when the watershed was subdivided into at least five subwatersheds. Guidelines need to be developed on the use of the procedure but it first needs to be evaluated on several watersheds that exhibit a range in sizes, land uses, slopes, and soil properties.