The Wind Field Improvement And Storm Surges Simulation Of Typical Typhoons In Haikou Bay With WRF-ADCIRC Model

China is one of the countries that suffer from typhoon-induced disasters most.Strong typhoons can cause a wide range of storm surge disasters.The precise prediction and simulation of typhoons and simulation of storm surges can help predict extreme high water levels and minish the typhoon-induced damages,with lots of practical significances.Simulation of storm surges is affected a lot by the simulated wind field input data.As a result,to acquire simulated wind field input data with higher precision has become the key of storm surge simulation.This research used the advanced mesoscale meteorology modeling system WRF(Weather Research and Forecasting)to simulate typhoons,coupling with SWAN wave model and ADCIRC hydrodynamic model to similate storm surges.To acquire simulated wind field data with higher precision,researches of two main aspects are carried out,which are wind simulation parameters and underlying surface of wind field.Then the coupling model with wind simulation improved was used to simulate storm surges.The main results are as follows:(1)After some tentative researches about the WRF wind model in Haikou Bay area,several conclusions were obtained: when a nest model is used,the finest model mesh should cover the typhoon track for the whole simulation time or the simulation results would be affected.Using smaller model top pressure and more vertical divisions does not affect the simulated typhoon track a lot,but can help acquire better wind speed simulation results.(2)A lot of WRF wind simulations of typhoon Kalmaegi and Rammasun were conducted with 8 microphysics parameterizations and 6 cumulus parameterizations.By the simulation results of contrastive tests,it can be found that cumulus parameterization affects the simulation of typhoon track and central pressure more than the microphysics parameterization.For typhoon Kalmaegi and typhoon Rammasun,the best culumus parameterization is both KF paraterization.As a result,for another typhoon with similar track at the same area,the KF culumus parameterization is recommended.(3)Since that the land use data built-in the WRF model is of low original accuracy and lacks updates,to have a typhoon underlying surface condition that is closer to the actual condition during the simulation time,the built-in land use data was replaced by GLC2009 data in this study.By the comparision of two simulations of typhoon Rammasun with MODIS2001 land use data and GLC2009 land use data,it was found that there was not much difference between the simulation results of typhoon track and central pressure with different land use data.However,the simulated 10 m wind speed decreased and approached to the measured data a lot with the new GLC2009 land use data set.At a verification point,The simulation error of maximum wind speed decreased 74% than before the land use data replace,and the simulation error of average wind speed during the whole time decreased 50% than before.The reason is that the underlying surface roughness of wind field changed with the replace of model land use data set,high surface roughness causes more consumption of wind kinetic energy near ground.Futher researches showed that the change of wind speed near ground at one point was not completely determined by the land use type change at this point,which showed that the change of wind speed field is holistic.This change gets weaker where farther from the shorelines.(4)The simulated wind field input data with MODIS and GLC2009 land use data was imported to the SWAN + ADCIRC coupling model.Compared with the MODIS land use data,the simulated hydrodynamic result with GLC2009 data was obviously improved,especially the lowest tide level.It showed that the replace of the land use data built-in the WRF model can help improve the typhoon surge simulation in this area.It also showed that the WRF-SWANADCIRC coupling model used in this study is capable of well simulating the typhoon surges in Haikou Bay area.