The SMART-based Dynamical-Statistical Combined Model For Seasonal Climate Prediction And Its Application

Extreme flood and drought disasters occur frequently in China due to the climatic variation of East Asian Monsoon,which seriously endanger the economic and social development of China and the lives and property of the people.It is an urgent need for disaster prevention and reduction to improve the accuracy of seasonal prediction for floods and droughts.However,due to the impacts of multiscale,multifactor,and the Chaos variation,it is difficult to improve the accuracy of drought and flood prediction.The current dynamical model for seasonal prediction could not give objective and quantitative predictions for summertime rainfall accurately.In this study,a new SMART-based dynamical-statistical combined model for seasonal climate prediction,whose hindcasts improved the prediction skills of dynamical model significantly,was established combined with dynamical model BCC?CSM1.1(m).The main conclusions are as follows:BCC?CSM1.1(m)dynamical model has a certain ability to predict the tropical outgoing longwave radiation(OLR)anomaly and 500 hPa geopotential height anomaly in the northern hemisphere,but its prediction skills of summertime rainfall in China is extremely limited,which is significantly lower than those in the tropical region.The annual average ACC of the summertime rainfall predicted by the BCC?CSM1.1(m)in China 160 Stations is 0,which is lower than the results of the Interannual Tendency Method and the Persistence Prediction Method.Therefore,the BCC?CSM1.1(m)cannot meet the actual demand of seasonal climate prediction in China.The SMART climate prediction principle,including the selected predictable climate modes(SM)and the anomalous relative tendency(ART),can effectively solve the difficulty of seasonal climate prediction.ART has a significant filter effect on climate factors,which can amplify the interannual signals with 2?6 years period and filter out the decadal signals with more than 6 years period.Several large-scale atmospheric circulation modes,i.e.,SMs,could be extracted from observations by SVD method,which have physical significance and determine the ARTs of summer rainfall in China.In this study,these models are used to reconstruct the ARTs of summer rainfall in China effectively,and the independent sample reconstruction test selected optimal combination of predictors and established the optimal statistical prediction model.Combining the optimal statistical prediction model and the predictions by BCC?CSM1.1(m),the SMART-based dynamical-statistical combined model for seasonal climate prediction of summertime rainfall in China can be established finally.In this model,the predictable climate modes in the ARTs of summertime tropical OLR and mid and high latitude 500 hPa geopotential height in the Northern Hemisphere that determine the simultaneous ART of rainfall in China are firstly extracted with historical observed data and SVD analysis,and a statistical prediction model for the simultaneous relationship between those modes and ARTs of rainfall in China is constructed with multiple linear regression method and the optimal predictable climate modes are selected by independent sample reconstruction tests.Then,with the optimal statistical prediction model,the ARTs of rainfall in China are predicted with SMs predicted by the global dynamical model BCC?CSM1.1(m).Finally,the total anomalies of summertime rainfall in China are obtained by adding the recent background anomalies to the predicted ARTs.The hindcasts for summertime rainfall during 1991?2019 show that the average ACC and PS score predicted by the SMART-DSM are 0.1 and 8 points higher than BCC?CSM1.1(m)respectively,the average PS score is 18 points higher in the last five years especially.In the prediction test of summer rainfall in 2020,the SMART-DSM also accurately predicted the abnormal tendency of the floods in the middle and lower reaches of the Yangtze River.Therefore,the SMART-DSM developed in this study can significantly improve the prediction skills of the BCC?CSM1.1(m)for summertime rainfall,which provides an effective solution for seasonal climate prediction in China.