Ensemble Simulation Of Tropical Cyclone Activity Near China With CWRF And Assessment

Tropical cyclone(TC)is one of the most disastrous phenomena occurring in China's offshore waters and adjacent sea area,especially during summer.Therefore,the prediction of TCs is important for China's climate forecast.Regional Climate-Weather Research and Forecasting model(CWRF)as a new version of regional climate model is becoming popularized in China due to its excellent climate simulation skill.This study attempts to use CWRF to conduct an ensemble simulation on the seasonal and interannual variation of TCs in China's offshore waters and adjacent sea area,with a view to using CWRF for tropical meteorology seasonal forecast.This study used CWRF at 30-km grid to perform an ensemble simulation of TCs that occurred in China's offshore waters and adjacent sea area from 1982 to 2016.The ensemble average is compared with European Centre for Medium-Range Weather Forecasts-Interim(ERI)analysis and model's control run.The effects of different physical parameterization schemes on the ensemble results are analyzed with the environmental fields.The simulations were driven by ERI and consisted of 28 runs of varying CWRF physics configurations.The Joint Typhoon Warning Center(JTWC)best track data were used as the reference for observations,The NCEP/NCAR CFSR data were used as the environmental field observation and the Geophysical Fluid Dynamics Laboratory(GFDL)Vortex Tracker were adopted to track back the TCs to their origin positions and times in sea level pressure and horizontal wind fields.The main conclusions are as follows:(1)CWRF performances: the dynamic downscaling skill and lateral boundary condition treatment in CWRF model can significantly improve the performance of ERI in TC simulation,and the comprehensive performance of ensemble members indicating that CWRF can carry out ensemble simulation of TC frequency.(2)Ensemble performance: comparing to control run,it is better in interannual variation simulation and multi-year seasonal average simulation is similar to control run.Overall,ensemble average can improve model's TC frequency simulation and reduce single experiment's uncertainty.(3)The effects of different parameterization schemes to the ensemble results: radiation scheme CAM is better in simulating multi-year seasonal average and interannual variation of the TC frequency.radiation scheme RRTMG is the best in simulating interannual variation.They are helpful in improving the ensemble average.Cumulus schemes BMJ and NSAS,Microphysical scheme Lin and radiation scheme RRTMG are all good at simulating TC in August,so they may can be used as an ensemble to simulate TCs in this month.(4)Impacts of environmental field: In July,the West Pacific Subtropical High(WPSH)was westward and 850 h Pa cyclonic circulation anomaly was weak which restrained tropical cyclone activity.In August,eastward(westward)WPSH and strong(weak)850h Pa anticyclonic circulation anomaly were conducive to TC activity.In September,850 h Pa cyclonic circulation anomaly was abnormally southern and small,which decreased TC activities.The environmental deviation in October was not the main cause of the TC frequency error.