Field Evaluation of NOAA's National Water Model

The National Weather Service (NWS) implemented the initial version (V1.0) of the National Water Model (NWM) in August of 2016. Since that time, one cycle of model improvement (V1.1) has been evaluated and implemented operationally and a second development cycle (V1.2) of the NWM is currently undergoing a scientific evaluation. Because the NWM is a gridded, physically based-model that runs in a high performance computing environment, the output from the model represents a significant paradigm shift from the current modeling approach used at the NWS River Forecast Centers (RFCs). Although the RFCs have developed considerable skill and knowledge base with respect to forecasting river stage and flow at defined locations, the NWM provides orders of magnitude more information, including flow at ungaged locations as well as gridded datasets that define the entire water budget. Because of their focused, regional expertise, RFC forecasters are a primary resource for the evaluation of the NWM output across their respective service areas. Beginning in early 2016, the RFCs have engaged in a significant effort to examine the different versions and cycles of the NWM output at current RFC forecast locations and compare them to RFC model forecasts. This talk will examine the approach that has been used in this important activity and present some of the results from the RFC evaluation.