Statistical Verifications for an Ensemble-Based Icing Forecasting Algorithm

Atmospheric icing continues to provide a significant threat to life and property in the aviation community. A microphysics-based approach to forecasting icing, utilizing a 10 member, 4-km Weather Research and Forecasting (WRF) mesoscale ensemble (MEPS), has been developed within the 16th Weather Squadron at the Air Force Weather Agency (AFWA). This algorithm uses supercooled water droplet size information within WRF-ARW microphysical parameterizations to forecast potential icing intensity and type, with specific interest in identification of glaze icing. This algorithm is currently is use in both AFWA's operational mesoscale ensemble (MEPS) and the developmental mesoscale ensemble. The development mesoscale ensemble is also run with 10 4-KM WRF-ARW configurations, with different configurations from the operational ensemble.

Statistics will be presented from both the operational version of MEPS, a domain that covers the continental United States, and the development version, a domain located in the eastern United States centered along the Ohio River Valley. Verification of icing type and intensity forecasts from the ensemble product and individual ensemble members is performed by comparing algorithm output to ADDS pilot report (PIREP) data for cases between November 2012 and July 2013. Verification is performed for all icing severity thresholds (light, moderate, severe, and extreme) as well as for each icing type (clear, mixed, and rime) by studying algorithm forecast icing values at grid points near PIREPs. Success and failure of individual members is studied as well, with the successes of specific parameterization schemes of particular interest. In addition to providing a statistical verification for the product, this verification will allow for improvements in the icing algorithm for both the intensity and type of icing.