 |
APL-UW is adapting a technique developed by Hoffman, et. al., 1995 and Ebert and McBride, 2000 to rapidly assess mesoscale numerical forecasts. This technique is being expanded to assess amplitude, displacement, distortion and rotation errors. We are also adapting rapid image motion processing algorithms from the motion picture industry to develop a highly efficient, automated verification system. This verification method is then being applied to the University of Washington mesoscale ensemble prediction system, and possibly, the US Navy or NPS COAMPS prediction system where it can be used on a large sample of model predictions.
|
|
It is also being used on a variety of model forecast fields, and should yield a more accurate assessment of mesoscale ensemble prediction systems as well as a technique for identifying model systematic errors in fields other than the traditional 500mb height or precipitation fields. The research results are being incorporated into a prototype of an automated verification system that can be used as a tool for weather forecasters to evaluate mesoscale model predictions.
|
 |