H020-05
Spatial Variability and Linkage Between Extreme Convections and Extreme Precipitation

Monday, 7 December 2020: 17:46
Virtual
Tsechun Wang, Peking University, Institute of Remote Sensing and GIS, School of Earth and Space Sciences, Beijing, China and Guoqiang Tang, University of Saskatchewan Coldwater Laboratory, Canmore, AB, Canada
Abstract:
Extreme precipitation is one of the key factors triggering natural hazards such as floods and landslides. Extreme convections also have remarkable influences on global water and energy cycle. Space-borne radars provide the most direct precipitation observations from space, which are valuable for studying global extreme precipitation. Based on the unprecedented 22-year records from the Tropical Rainfall Measuring Mission (TRMM) precipitation radar (PR) and the Global Precipitation Measurement (GPM) dual-frequency precipitation radar (DPR), we reveal the spatiotemporal variability of extreme precipitation and convection events and their relationship in the globe. Generally, there are two commonly used methods to define extreme precipitation thresholds in multiple time scales, including the fixed threshold and percentile‐based method. In this study, variable fixed-thresholds methods are used to ensure our results are comparable in different regions. It is found that extreme precipitation is dominant in tropical ocean while extreme convections are exclusive in land from tropics to higher latitude. Both extremes exhibit significant seasonal variations and occur more frequently in the rainy seasons for both hemispheres. Furthermore, extreme events exhibit notable diurnal characteristics over global land with most extreme precipitation occurring in afternoon, while diurnal cycles are more homogeneous over ocean. Finally, we find that extreme convections and precipitation show significant positive correlation over land but relatively weak correlation over oceans using the threshold-based definition of extreme events, whereas percentile-based definition acquires inverse conclusions. This discrepancy indicates that researchers should be cautious with the selection of extreme event definitions for large-scale studies.