Numerical Simulation Of Seasonal Activity Of Typhoon Over North Western Pacific

North Western Pacific(WNP)is the region with the highest frequency of typhoon in the world,with an average of 28.2 typhoons generated each year.China is seriously affected by typhoons for its eastern part is adjacent to the WNP,with an average of 7 typhoons landing in China every year.Torrential winds,rainstorms,storm surges and secondary disasters caused by typhoons can cause significant property losses and casualties in coastal areas.Typhoon activity has significant seasonal,interannual and interdecadal variations.From 1977 to 2016,the highest annual frequency of typhoon is 37,while the lowest is only 14.Therefore,accurate prediction of typhoon seasonal activity before the onset of typhoon season can provide sufficient time for the government and the public to prepare for typhoon disasters and minimize risks and losses.In this paper,the typhoon activity in the WNP from May to October is studied and forecasted by using the regional coupled model COAWST.Large-scale field plays an important role in the generation and movement of typhoon.Large-scale field drifting always occurs in the long-term integration process of regional model.In order to improve the seasonal prediction skills of the model,the influence of spectral nudging assimilation scheme on Typhoon seasonal activity simulation is studied.The results show that the simulation of typhoon frequency,intensity,ACE distribution and path density distribution can be significantly improved by the spectral nudging assimilation experiment compared with the control experiment without the spectral nudging assimilation scheme and the simulated typhoon seasonal activity is more consistent with the observation.The improvement of typhoon seasonal activity simulation with spectral nudging assimilation scheme is mainly realized by improving the large-scale field of the model.The spectral nudging assimilation scheme can make the large-scale field(height field,temperature field,horizontal velocity field and humidity field)in the upper and middle levels of the model close to the reanalysis field to improve the long-term integration for large-scale simulation,and then improve the process of air-sea interaction and meso-and small-scale processes in the model,so as to improve the simulation of typhoon seasonal activity.On this basis,the effects of different cumulus convection parameterization Schemes in the coupled model on Typhoon seasonal activity simulation are further studied.Cumulus convection process is very important to the generation and development of typhoons.Whether the cumulus convection parameterization scheme introduced in the model is reasonable or not is the key to the successful simulation of typhoons.The differences of four common cumulus convection parameterization schemes(KainFritsch(new Eta),Betts-Miller-Janjic,Grell-Freitas and Grell-3)in COAWST model are compared,and the sensitivity of numerical simulation of typhoon seasonal activitity to the selection of different cumulus convection parameterization schemes is investigated.The simulation results of different cumulus convection parameterization schemes show that in the COAWST model,using the Kain-Fritsch(new Eta)scheme to simulate the typhoon seasonal activity has the best effect on the typhoon frequency and typhoon path density simulation.The numerical simulation of the typhoon season activity using the Grell-Freitas scheme has the best effect on the typhoon intensity simulation.Compared with the other three cumulus convection parameterization schemes,the typhoon frequency and intensity simulated by the Betts-Miller-Janjic scheme has the largest difference from the observation.Due to the current typhoon seasonal prediction pays more attention to the information of typhoon frequency and its influence range,Kain-Fritsch scheme is a better scheme for typhoon seasonal prediction.Based on the spectral nudging assimilation scheme and optimization of cumulus convective parameterization options,a coupled typhoon seasonal forecasting system is further established.The CFSv2 global seasonal forecast data was used as the initial and boundary field,and the hindcast experiment of the typhoon activity in the typhoon season of the WNP in 2011-2018 was carried out to test the performance of the forecasting system.The results of the hindcast experiment show that the typhoon seasonal activity is less frequent and weaker in the CFSv2 forecast.Compared to the CFSv2 forecast,the new system can effectively improve the typhoon frequency and intensity forecast,especially the typhoon frequency forecast of the typhoon exceeding the TY level,but there is still a certain gap between the results in the new system and the observed typhoon frequency and intensity,which is mainly reflected in the fact that the forecasting season has a higher frequency of typhoons and the peak of strong typhoons is relatively weaker.By analyzing the factors that are important for the formation,development and movement of typhoon,we found that the forecasting effect of typhoon frequency is related to the forecasting effect of monsoon trough,850 hPa relative vorticity,low-altitude convergence and high-altitude divergence,sea surface latent heat release and sea surface temperature.When the forecasting monsoon trough is eastward,the relative vorticity in the low latitude sea area is relatively large,the lower layer convergence is strong,the high-altitude divergence is weak,the sea surface latent heat release and the sea surface temperature are positively deviated,typhoon is more likely to generate and develop,which leading to the forecasting total frequency of typhoons exceeds the observed typhoon frequency.