Evaluating warning verification statistics and methodologies for displaced real-time (DRT) scenarios

During the first few months of 2000, the WSR-88D Operational Support Facility (OSF) Operations Training Branch (OTB) had three Warning Decision Making workshops (WDM I) for intern forecasters of the National Weather Service (NWS). During these workshops, the participants issued simulated storm warnings for several scenarios using Warngen and AWIPS in a displaced real-time (DRT) mode. These workshops provided data on warning issuance that are unique: several groups of forecasters issuing warnings for the same severe weather events. These data allow for study of forecasting skill with a large group of forecasters all forecasting the same event.

Verification statistics are being computed for the workshop participants for multiple scenario exercises. Comparisons are planned of the traditional verification method (county-by-county) verses a new method that is possible with AWIPS (Warngen polygon). If time allows, further examination of other variables of the warning (such as warning temporal length and spatial size) will be conducted.

There is discussion in the NWS of switching to the new verification method for operational warnings. These results will highlight the significant differences that will result if that change is made. These results can also be used to examine the influence of warning parameters such as spatial size and temporal duration.