Abstract
A heavy rainstorm occurred in Henan Province, China, between 19 and 21 July, 2021, with a record-breaking 201.9 mm of precipitation in 1 h. To explore the key factors that led to forecasting errors for this extreme rainstorm, as well as the dominant contributor affecting its predictability, we employed the Global/ Regional Assimilation and Prediction System-Regional Ensemble Prediction System (GRAPES-REPS) to investigate the impact of the upper tropospheric cold vortex, middle-low vortex, and low-level jet on predictability and forecasting errors. The results showed that heavy rainfall was influenced by the following stable atmospheric circulation systems: subtropical highs, continental highs, and Typhoon In-Fa. Severe convection was caused by abundant water vapor, orographic uplift, and mesoscale vortices. Multiscale weather systems contributed to maintaining extreme rainfall in Henan for a long duration. The prediction ability of the optimal member of GRAPES-REPS was attributed to effective prediction of the intensity and evolution characteristics of the upper tropospheric cold vortex, middle-low vortex, and low-level jet. Conversely, the prediction deviation of unstable and dynamic conditions in the lower level of the worst member led to a decline in the forecast quality of rainfall intensity and its rainfall area. This indicates that heavy rainfall was strongly related to the short-wave throughput, upper tropospheric cold vortex, vortex, and boundary layer jet. Moreover, we observed severe uncertainty in GRAPES-REPS forecasts for rainfall caused by strong convection, whereas the predictability of rainfall caused by topography was high. Compared with the European Centre for Medium-Range Weather Forecasts Ensemble Prediction System, GRAPES-REPS exhibits a better forecast ability for heavy rainfall, with some ensemble members able to better predict extreme precipitation.
NETL
NSTL
Springer Nature