An object‐oriented technique for nowcasting heavy showers and thunderstorms

Abstract Accurate forecasting of heavy showers and thunderstorms with associated hazards is vitally important for many business sectors and national utilities. In the UK a fully automated procedure is being developed in the Met. Office for the National Rivers Authority. The GANDOLF (Generating Advanced Nowcasts for Deployment in Operational Land surface Flood forecasting) system seeks to provide warnings of heavy rain and forecast accumulations in sensitive river catchments to flood hydrologists. GANDOLF will automatically choose the most appropriate nowcasting technique depending upon synoptic conditions. In a convective situation an important method available to GANDOLF is an object‐oriented nowcasting procedure. Multi‐beam, high resolution radar data and Meteosat IR satellite data are used to analyse convective cells in all stages of growth; subsequent movement and development up to 3 hours ahead is then predicted using a conceptual life‐cycle model combined with mesoscale NWP data. This paper describes the object‐oriented technique and demonstrates its usefulness in a severe convective situation with a case study.