A Statistical Sea-Breeze Prediction Algorithm for Charleston, South Carolina

Abstract

A simple and useful technique for prediction of sea breezes, based on readily available wind vector and air temperature predictions from synoptic models in conjunction with observed coastal sea surface temperatures, is presented for evaluation by coastal forecasters. A statistical prediction scheme using the sea-breeze index has been devised and was found to possess significant nowcasting skill, when used with observed synoptic wind vectors and temperatures and with observed coastal sea surface temperatures. The ready availability of data for these variables and the simplicity of the scheme give it the potential of being useful to many coastal forecast offices, once tuned to the area in which it is applied. The technique can be used on simple workstations or even by manual calculation and thus provides a simple method for local tuning of general-area forecasts to coastal areas. For the 1998 sea-breeze season tested, the algorithm, which was built on June data and tested by nowcasting on data from July through October, had a skill score of 31.3% over climatology.