Predictability of the July 2020 Heavy Rainfall with the SCALE-LETKF

Abstract

The predictability of the July 2020 heavy rainfall event that saw record-breaking rainfall over Western Japan in July 2020 is examined with the near real-time SCALE-LETKF numerical modelling system in a low resolution 18-km configuration setting. Ensemble-mean 5-day rainfall total forecasts showed close agreement with Japanese Meteorological Agency 1-km precipitation analyses in relation to the large-scale distribution of rainfall and to location of heaviest rainfall over Kyushu. Onset and duration of rainfall at specific sites across Kyushu were also well predicted by the forecasts. However, the precise prediction of heavy rainfall, including over the worst-hit Kumamoto and Kagoshima prefectures, was severely underestimated. Examination of the atmospheric conditions at the time of the heavy rainfall from reanalysis datasets and ensemble member forecasts showed very high humidity over central Kyushu with strong transport of moisture from the southwest to central regions. In addition, strong low-level convergence was observed to the west of Kyushu in both reanalysis and best performing member forecasts during the time of heavy rainfall, suggesting a potential contributing factor to the record-breaking rainfall. (Citation: Taylor, J., A. Amemiya, T. Honda, Y. Maejima, and T. Miyoshi, 2021: Predictability of the July 2020 heavy rainfall with the SCALE-LETKF. SOLA, 17, 48-56, doi:10.2151/sola. 2021-008.)