Assessment Of Climate Change Effects On The Activity Of Tropical Cyclone In The Northwest Pacific Ocean

Tropical cyclones are a highly destructive natural disaster.The Northwest Pacific Ocean is the most active region of tropical cyclone activity in the world,and tropical cyclones cause significant economic losses and casualties to the region every year.With the progress of building technology and construction techniques,a large number of high-rise buildings with novel structural systems have been built,especially in the coastal areas of southeast China,and most of these structures have the characteristics of towering,light and complex,and thus are very sensitive to wind loads.Numerous studies have shown that climate change is closely related to tropical cyclone activity,which can further influence wind-climate characteristics.Therefore,it is necessary to conduct an in-depth study on the effect of climate change on tropical cyclone activity in order to further investigate the effect of climate change on typhoon extreme wind climate,and then determine the wind load/wind response of building structures to ensure the safety and applicability of buildings.Climate change will lead to changes in atmospheric and oceanic environmental parameters,which will affect tropical cyclone activities.But the large volume and dimensionality of data on atmospheric oceanic environmental parameters make it difficult to use traditional regression equations to establish links between key parameters of tropical cyclones and multiple atmospheric oceanic environmental parameters in multiple pressure layers.Based on the above situation,this paper explores the time-space variation characteristics of atmospheric and oceanic environmental parameters using data from upper air sounding balloons,the atmospheric and oceanic environmental reanalysis data and the atmospheric and oceanic environmental parameters output from four CMIP6 climate models.In addition,based on the above data,we use the statistical dynamics method and deep learning method to build a full path synthesis model for tropical cyclones,which can simulate and predict the whole life cycle of tropical cyclones from generation to extinction,respectively,without considering the influence of climate change,considering the influence of single/multiple meteorological elements in single pressure layer and considering the influence of multiple meteorological elements in multiple pressure layers.Finally,the impact of climate change on the key parameters of tropical cyclones in the northwest Pacific Ocean is investigated by comparing and analyzing the output of the two models.The final results show that climate change is not only manifested in SST,but also in other meteorological parameters at different heights,such as air temperature,relative humidity,relative vorticity and vertical wind shear.By comparing and analyzing the results of historical tropical cyclone generation simulations,compared to statistical dynamics generation model,the deep learning model is more consistent with the real distribution of tropical cyclone generation and location.When predicting the location of tropical cyclone generation in the next60 years,the statistical dynamics model indicates that the genesis of tropical cyclones will move to the east/southeast under the influence of multiple meteorological parameters,while the deep learning model predicts that the location of the genesis will not move significantly,but both models indicate that the number of tropical cyclones will increase in the future.Next,this paper further explores the simulation and prediction of tropical cyclone track and intensity,and the results show that the deep learning models have the advantage of non-linearly considering multiple meteorological parameters and are computationally efficient.When considering only the effects of multiple meteorological parameters in a single pressure layer,the intensity predictions of both models are consistent,indicating a significant increase in the number of low-intensity tropical cyclones and a decrease in the number of high-intensity ones in the future.However,when predicting the future track of tropical cyclones,the prediction results of the two models show significant differences.Finally,based on the deep learning model,this paper investigates the effects of multiple meteorological parameters in multiple pressure layers on the track and intensity of tropical cyclones,and the results show that climate change will lead to an eastward shift in the future tropical cyclone track and an increase in the number of less intense tropical cyclones in the future.The research content and related results of this master's paper can be used to assess the impact of climate change on the typhoon extreme wind climate in the coastal zone of southeast China,and as a basis to provide important references for future-oriented disaster assessment and disaster prevention and mitigation in various regions.