Capturing the spatial and temporal distribution of rainfall is important for accurate hydrologic prediction. One of the most important sources of spatially distributed rainfall data is radar. Beginning in the early 1990's, the WSR-88D (popularly known as NEXRAD) radar network was deployed by the US National Weather Service (NWS) for surveillance and detection of severe weather. Producing accurate rainfall estimates from this system requires post processing within a statistical framework that recognizes systematic and random errors as components of uncertainty. Statistical control of this input data is shown to have important consequences on distributed hydrologic modeling. Hydrologic evaluation of radar rainfall using archived radar rainfall provides the opportunity to evaluate the predictability of a fully distributed physics-based model through event reconstruction.