Initial work implemented a real-time atmospheric prediction system using the Penn State/NCAR atmospheric Mesoscale Model Version 5 (MM5) system in a triply-nested configuration centered over the southern half of the Gulf of Maine. The Advanced Research - Weather Research and Forecasting Model (ARW-WRF) has since replaced the MM5. A primary project goal is to collect and optimally assimilate in MM5/WRF 3d-VAR a diverse set of observations - currently upper-air; land and ocean surface reports; and QSCAT surface winds. Current real-time users include the US National Weather Service Forecast Offices in Gray and Caribou, ME, and the UNH terrestrial water budget modeling community. One recent example of this collaborative effort occurred in December 2006. During a mesoscale snow event, NWS Gray forecasters were able to improve their forecast by taking advantage of the mesoscale detail and skill provided by the MM5 model.
In this poster we will report both subjective initial evaluations of the atmospheric forecasting systems from the perspective of the NWS users in Gray, ME, and objective verification statistics from the MM5 and WRF forecast domains positioned over New England and the Gulf of Maine.