An Automatic Detection Algorithm for Sea Breeze Fronts: A Case Study over the Gulf of Guinea in West Africa

Abstract

In this paper, we present a new approach to the detection of Sea Breeze Fronts (SBF) in the Gulf of Guinea using automated methods. The study focuses on southern West Africa, where SBFs play a crucial role in local weather. The research demonstrates that the dynamic of SBFs exerts a significant influence on local weather conditions and acts as a favourable mechanism for convection. The aim of this study is to improve the effectiveness of conventional SBF detection techniques by applying an automated methodology through the analysis of images obtained by the second generation Meteosat (MSG) satellite. Our method, based on an active contour technique called morphological snake, is capable of automatically detecting the cumulus lines that are associated with SBF in a relatively short period of time using a substantial number of MSG images taken every 15 min. To delineate the SBFs and to model their inland propagation by isochrones, several regression methods were employed. Among these, the kernel-weighted local polynomial regression (kwLPR) provided the greatest accuracy in modeling the SBF propagation, with an average spatial root mean square error (RMSE) of only 0.0034˚. The SBF penetrated as far as 100 to 146.3 km inland at certain longitudes. Its average penetration along the coast is 103.17 km. The algorithm is highly robust and has a wide range of practical applications, including automatic pattern recognition and dynamic imaging. Furthermore, it has significant potential for future research into other complex phenomena, such as the propagation of pollutants and other atmospheric particles.