Japan Geoscience Union Meeting 2023

Presentation information

[J] Oral

A (Atmospheric and Hydrospheric Sciences ) » A-AS Atmospheric Sciences, Meteorology & Atmospheric Environment

[A-AS10] Extreme weathers and disasters in urban environments in East Asia, related to climate c hange

Sun. May 21, 2023 3:30 PM - 4:45 PM 103 (International Conference Hall, Makuhari Messe)

convener:Masaru Inatsu(Faculty of Science, Hokkaido University), Hiroyuki Kusaka(University of Tsukuba), Tetsuya Takemi(Disaster Prevention Research Institute, Kyoto University), Yukari Takayabu(Atmosphere and Ocean Research Institute, the University of Tokyo)

4:30 PM - 4:45 PM

[AAS10-05] Evaluation of extreme precipitation represented by CMIP6 models in terms of dependencies on humidity and stability.

*Minoru Chikira1, Yukari Takayabu1, Chie Yokoyama1 (1.The University of Tokyo, Atmosphere and Ocean Research Institute)

Keywords:Extreme precipitation, CMIP6

Our study investigated the relationship between extreme precipitation, humidity, and atmospheric stability using satellite-based observed precipitation data (GSMaP) and atmospheric profiles obtained from ERA5 reanalysis. We then evaluated the performance of the CMIP6 models with the aim of reducing uncertainties in future predictions of extreme precipitation.

To maximize the number of target models available, we utilized daily mean variables, as opposed to 6-hourly snapshots. We defined two indices for stability and humidity, the Surface Air Moist static energy Surplus (SAMS) and the Free atmospheric precipitable water (FPW), respectively. These indices can be calculated from daily mean variables, even with the coarse vertical resolution typical in CMIP6 model outputs.

We created two-dimensional precipitation bins for FPW and SAMS. Our analysis confirmed that the plots using the daily mean variables of GSMaP and ERA5 were qualitatively similar to those using the 3-hourly counterparts. We found that humidity was highest when the stratification instability was weak, creating a beak-like structure over which extreme precipitation occurred. Our analysis also revealed a clear difference in the precipitation patterns between the ocean and land. The peak precipitation occurred over the upper half of the beak-like structure over the ocean, while it occurred over the lower half over land.

Our evaluation of the CMIP6 models revealed that while many models were able to represent the overall structure of the features observed in the GSMaP and ERA5 data, some models had a defect in representing the beak-like structure and underestimated the extreme precipitation, especially over the ocean. Furthermore, only a limited number of models were successful in reproducing the location of the peak precipitation over the beak-like structure. Models that represented a reasonable amount of extreme precipitation over the ocean tended to overestimate it over land, while those over land tended to underestimate it over the ocean. We did not find any model that successfully represented the amount of extreme precipitation over both the ocean and land.