Trends in Comparative Verification Scores for Guidance and Local Aviation/Public Weather Forecasts

Abstract

In this paper, we present trends in the verification scores of the Techniques Development Laboratory's (TDL's) operational guidance forecasts and the National Weather Service local forecasts made at Weather Service Forecast Offices (WSFO's). Verification statistics for objective guidance and subjective local forecasts of probability of precipitation (PoP), precipitation type, surface wind, opaque sky cover, ceiling height and maximum/minimum (max/min) temperature are shown for the cool and warm seasons as these guidance products became operational during the period 1970?77. TDL's forecasts are based on the Model Output Statistics (MOS) technique. The primary input to the MOS prediction equations comes from forecast fields from the Limited-area Fine-Mesh, Trajectory, and/or Primitive Equation models. For some equations, surface observations and other variables such as station elevation are also used as predictors. Verification shows that through the years TDL's aviation/public weather guidance forecasts have compared very favorably with local forecasts produced at WSFO'S. The guidance forecasts are usually better than the local forecasts for precipitation type, surface wind and opaque sky cover; the opposite is true for PoP, max/min temperature and ceiling height. The verifications also show that local and guidance forecasts have generally improved for all elements over the years.