The Formation Mechanism And Prediction Application Of Tropical Cyclone Accumulated Rainfall Structure Over China

In this study,the accumulated precipitation structure and variation characteristics of tropical cyclones(TCs)in different regions of China are analyzed.The precipitation data are from the CLDAS-Prcp dataset between 1998 and 2019.The results show that there are significant differences in precipitation structure characteristics of samples in different regions.Then,based on the observation data and ERA5 reanalysis data,the formation mechanism of the structure and variation characteristic of TC precipitation in Northern China is analyzed.Finally,the TC intensity factor,which has obvious influence on TC precipitation in all regions,is introduced into the DSAEF?LTP model for prediction and verification.The main conclusions are shown as below.(1)All samples have asymmetrical daily precipitation structures,but the characteristics of TCs in different regions are obvious different.Among the three groups of samples,the precipitation asymmetry of TCs in Northern China is the most obvious.The relative standard deviation(RSD),which is used to represent asymmetry,is the highest(between 0.2 and 0.6)and increases after landfall.The heavy precipitation area shifts from the southeast quadrant to the northeast quadrant and gradually moves away from TC center.The RSD of TCs in South China ranges from 0.2 to 0.5,and gradually decreases after landfall.The heavy precipitation center shrinks from southwest quadrant to the area near TC center.The RSD of TCs in Southeast China is the smallest(between 0.2 and 0.4)and relatively stable.The heavy precipitation center gradually shifts from the southeast quadrant to the northeast quadrant.After the TCs from different regions are divided into strong and weak intensity sample groups,it can be found that the precipitation of strong TC is generally stronger,and the asymmetry on the day before landfall is usually weaker.(2)The study of northward TCs shows that the precipitation is caused by the multi-scale interaction between TC-self factors and environmental factors,and the terrain enhances the precipitation.The intensity of TC is an important factor affecting the overall strength of precipitation and the distance of extreme rainfall point,while other factors mainly affect the location of extreme rainfall center.Typical samples in northward TCs usually have stronger intensity(typhoon level or above)and slower motion speed(between 14km/h and 20km/h)at the landfall time,and the landing sites are more southerly(south of 30°N,which is favorable for water vapor supply,and land in mountainous areas).In combination with favorable weather systems(such as South Asia high,subtropical high,westerly trough,etc.)and topography,the precipitations are always stronger,and the locations of heavy rainfalls are more consistent with the area conducive to the strengthening of convection.For atypical TCs,the westerly trough is relatively stronger and more southerly.When the westerly trough is combined with TC,it is easy to cause heavy rainfall in front of the trough.(3)After introducing the intensity factor,which has great influence on TC precipitation,into DSAEF?LTP model,the prediction experiments of TC accumulated rainfall are carried out in China.The results show that the prediction performance of the model has been improved and is comparable to that of the NWP models.After introducing the intensity factor,the prediction skill of training and independent samples is improved by 8.64% and 9.15%respectively.Then,according to the regional differences in the prediction performance of the model,a regional forecast experiment is carried out in China.The result shows that the TS of the independent samples increased by 19.8% overall.Comparing to the four NWP modes,it is second only to SMS-WARM.