AUTHOR=Gao Li , Zhao Zuosen , Qin Jun , Chen Quanliang , Cai Hongke 

TITLE=Stepwise correction of ECMWF ensemble forecasts of severe rainfall in China based on segmented hierarchical clustering

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 10 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2022.1079225

DOI=10.3389/feart.2022.1079225

ISSN=2296-6463

ABSTRACT=Ensemble forecast plays a vital role in numerical weather prediction. Hence, effectively extracting useful information from ensemble members to improve precipitation forecasting skills has always been an important issue. By using the ensemble forecast data of precipitation from the ECMWF-GEPS (Global Ensemble Prediction System), a stepwise correction method based on the segmented hierarchical clustering (SHC) for daily precipitation ensemble forecast is proposed in this study. This method obtains more probabilistic forecast information to improve forecasts by employing the segmented correction scheme. Validations of the SHC method have been performed by comparing with other two methods, namely the ensemble-mean (EM) method and the direct hierarchical clustering (HC) method. Results show that the deterministic forecast of SHC can elevate the forecast ability of heavy precipitation in the short- and medium-range weather forecast, and thus SHC performs better than EM and HC, by effectively extending lead time of skillful severe rainfall forecasts for 2–3 days relative to the other two. Meanwhile, SHC shows better performance through a continuous forecast verification in summer 2021 and even has better forecast effects on multiple heavy-precipitation cases, including the Zhengzhou extreme rainfall on July 20, 2021. Overall, the SHC method has large potential in improving ensemble rainfall forecasts in the current operational system.