A Neural Network Short-Term Forecast of Significant Thunderstorms

Abstract

Case studies are the typical means by which meteorologists pass on their knowledge of how to solve a particular weather-forecasting problem to other forecasters. A case study helps others recognize an important pattern and enhances the meteorologist in the meteorologist–machine mix. A neural network is an artificial-intelligence tool that excels in pattern recognition. This tool can become another means of enhancing a forecaster's pattern-recognition ability. Since neural networks are a relatively new tool to meteorologists, some basics are given before discussing a 3–7-h significant thunderstorm forecast developed with this technique. Two neural networks learned to forecast significant thunderstorms from fields of surface-based lifted index and surface moisture convergence. These networks are sensitive to the patterns that skilled forecasters recognize as occurring prior to strong thunderstorms. The two neural networks are combined operationally at the National Severe Storms Forecast Center into a single hourly product that enhances pattern-recognition skills. Examples of neural network products are shown, and their potential impact on significant thunderstorm forecasting is demonstrated.