A Novel Algorithm for Detection of Convective Initiation Using Multi-Source Satellite Images

Abstract

Severe convective weather is a type of extreme weather characterized by sudden and intense conditions, often featuring heavy short-term rainfall, lightning, strong winds, hail, tornadoes, and other related disasters. The initiation of convection indicates that severe weather is imminent and is essential for short-term forecasting. However, the horizontal movement of clouds can hinder the precision of algorithms designed to detect convection initiation. To address this issue, a novel algorithm utilizing cloud-top rapid cooling rates from multi-source satellite images has been developed. This algorithm leverages the high temporal resolution of the 6-minute fast scan data from the FY-2F satellite and includes a filter with three testing conditions to enhance the accuracy of detecting convective initiation. The algorithm significantly improves detection accuracy by addressing the challenge of horizontal cloud movement, which has been a persistent issue in previous detection methods. By integrating data from infrared, water vapor, and visible light channels, the algorithm provides a comprehensive approach to identifying CI signals in their early stages. This advancement is crucial for enhancing the timeliness and accuracy of short-term severe weather warnings, thereby contributing to more effective disaster prevention and mitigation efforts.