Improvement of Anomaly Forecast for Extreme Weather Events

In 2006, the post-processing of NCEP Global Ensemble Forecast System (GEFS) and North American Ensemble Forecast System (NAEFS) has been implemented to enhance probabilistic forecast through anomaly forecast of various weather elements. Anomaly forecast is one of NAEFS products from bias corrected forecast and reanalysis based climatology. It is measuring the forecast departure (bias–free) from climatology (observation). Based on NCEP/NCAR 40-year reanalysis, daily climatological distribution (PDF) has been built up for 19 atmospheric variables, such as height, temperature, winds and etc. The uncertainty information for anomaly by comparing forecast PDF to climatological PDF allows users to identify the extreme weather event easily. There are many applications in past years for extreme heat waves, winter storms and etc.

Later, a new daily climatology has been generated from latest Climate Forecast System (CFSR) reanalysis. There is also another way to build up anomaly forecast (or Extreme Forecast Index (EFI)) in the user communities, that is based on raw ensemble forecast and model climatology. In this presentation, a verification methodology for extreme weather events will be presented. A multi statistics comparison of anomaly forecast for extreme weather/climate events between different methods, different analysis datasets, and different forecast versions will be performed. The considered extreme weather elements include temperature, precipitation, and 10-m wind.