An important characteristic of hydrological is the need for accurate forcing data, such as precipitation and temperature. Acquiring precipitation and temperature gauge data poses a variety of chal¬lenges, not least the fact that gauges are often located outside of target watersheds and may not accurately represent local conditions. Over the last decade, there has been a drive to archive global atmospheric data from which our daily and hourly weather forecasts originate, primarily for the purpose of weather forecast improvement. We are investigating ways to utilize these products for hydrological modeling purposes and to address some of the inherent problems associated with the use conventional gauge data. In this study, we compare calibrations of a watershed model using derived statistical representations of precipitation forecasts from a “poor-man’s” ensemble of raw gridded atmospheric models interpolated to the center of the model subbasin, versus, calibration to the closest precipitation gauge measurement. In addition, we investigate at what scale and radii the use of direct gridded model outputs may introduce less or equal error to watershed modeling projects using the closest gauge station.