An Automated Mesoscale Forecast Verification System

Scott Sandgathe

Principal Meteorologist, APL/UW

Traditional verification schemes tend to penalize mesoscale numerical weather prediction (NWP) systems, which realistically portray high amplitude, short duration mesoscale phenomena. Small phase, timing or location errors for high amplitude features result in apparent poor performance when traditional verification schemes based on synoptic observations or grid point analyses are employed. Yet these same NWP systems provide much more realistic and often more operationally useful depictions of weather events than smoother global NWP or ensemble-mean systems. Unfortunately, taking into consideration small phase or timing errors generally requires labor-intensive case studies which are unable to address the large number of cases required for reliable mesoscale NWP or mesoscale ensemble verification. NWP centers and forecasters need automated, rapid and more realistic evaluation of mesoscale NWP and ensemble performance.

The University of Washington, as part of a DoD-funded interdisciplinary university research project, is developing an automated mesoscale forecast verification system. This system follows the ideas of Hoffman, et. al. (1995) and incorporates rapid search techniques developed by the motion picture industry. The system will be used to assess the performance of the UW MM5 ensemble forecast system and provide guidance to Navy and NOAA forecasters in the NW region. Besides addressing issues of rapid evaluation, the research will also address methods of evaluating and displaying the relative “goodness” of phase, amplitude, and distortion errors for customer use.

Recent efforts by Du and Mullen (2000) and Ebert and McBride (2000) have adapted the Hoffman technique for verification of precipitation forecasts. In addition to precipitation verification which is significantly complicated by terrain in the NW region, the UW group hopes to address overall mesoscale NWP performance on continuous state variables.