Ongoing development of an experimental, real-time prediction system for high-impact convective weather events

Using model forecast output from the Purdue WRF-ARW model, a prediction tool is being developed for the identification of potential high-impact convective weather events. These events are identified through the use of an object identification method discussed in Baldwin et al. (2005). Once the high-impact events have been identified, they are tracked in time using model output and are assigned various distinct attributes, such as; shape, identification threshold, location, and eccentricity. In addition to these characteristics, users of the prediction system are able to generate forecast soundings and cross sections for each identified object of forecaster interest.

Verification data are also an important factor in assessing the quality of this high-impact convective weather prediction system, and therefore an object oriented verification approach is taken. Predicted events are compared to severe weather reports as well as standard meteorological observations.

This tool aims to aid forecasters in the identification of potentially hazardous weather, as well as severe weather mode, by drawing attention to significant model-based information. This prediction system will help maximize the valuable time involved in the severe weather forecasting process.