Study On Interdecadal Prediction Of Extreme Precipitation And Extreme High Temperature In Summer In South Chin

This paper used the CN05.1 gridded daily dataset based on 2416 station observations over China,the ERA5 reanalysis dataset of the European Centre for Medium-Range Weather Forecasts,the National Oceanic and Atmospheric Administration(NOAA)dataset and the UK Met Office Hadley Centre dataset to analyze the decadal variability of the summer extreme precipitation(EPSC)and the extreme heat events over South China(EHSC)based on the the recommendation of the Expert Team on Climate Change Detection and Indices(ETCCDI).This paper further reveals the potential influencing factors and its possible mechanisms for the decadal summer extreme precipitation and the extreme heat over South China.Additionally,the statistical decadal model is constructed for the summer extreme precipitation and extreme heat events based on the decadal increment method.The main conclusions are summarized as follows:(1)The decadal variability of the summer extreme precipitation over South China is remarkable,especially for the significant decadal enhancement after the 1990 s.The study documented that the summer dipolar sea surface temperature(SST)over the North Atlantic and spring sea ice concentration(SIC)over the East Siberian Sea can significantly affect the EPSC.The summer SST over the North Atlantic influences the low-pressure cyclone and easterly anomalies in the western Pacific by modulating the SST over the tropical Pacific,thus affecting EPSC.A decrease in the SIC of the East Siberian Sea induces a negative Arctic Oscillation,which induces the increased SST over northwest Pacific,northerly anomalies and water vapor flux divergence anomalies over there,in turn,affecting EPSC.Both predictors have a quasiperiod of 10-14 years,which provides useful predictive signals for EPSC.The leading 7-year SST and the leading 5-year SIC are therefore chosen to establish the prediction model based on the decadal increment method.Both the cross-validation and independent hindcast results illustrate that the decadal increment method can predict the EPSC well,especially the shift in the early 1990 s.Noteworthily,the real-time prediction results show that the EPSC would increase during 2020-2022,which is consistent with the actual situation of more EPSC.(2)The summer extreme heat events over South China have been increased in the past 20 years,which also experienced obvious decadal shifts in the early 1990 s and early 21 st century.The main predictors from the external forcing factors such as sea temperature,sea ice,soil humidity and soil temperature,and the atmospheric internal variability such as sea level pressure and geopotential height field are determined by the stepwise regression with the F-test at the 95% confidence level.The leading 6-year summer dipolar sea surface temperature(SST)over the Indian and leading 5-year spring sea ice concentration(SIC)from the East Siberian Sea to the Beaufort Sea are indicated to affect the EHSC,which can explain 28% and 29% of the variance of the EHSC respectively.Two predictors are well correlated with the predictand and independent of each other.Both predictors have a quasi-period of 10-14 years with a significant negative autocorrelation of 5-7 years,which can provide useful predictive signals for EHSC.By constructing a statistical model,the prediction effects of the model shows that the incremental method can well predict the EHSC.At the same time,the real-time prediction results show that EHSC would increase in 2020-2022,which is also consistent with the actual results.(3)The increment method shows its advantages in the decadal prediction of extreme climate.The decadal increment method not only adds leading observational data to obtain useful decadal information,but also adds valid samples in the correlation analysis,which predicts the summer EPSC and EHSC well.This statistical method provides a possible way for the current difficult decadal prediction of extreme climate events.