Research On The Influence Of Assimilating X-band Phased-Array Radar Data In The Pearl River Region By EnKF For The Analysis And Nowcast Of Tornadoes

Accurate analysis and prediction of the structure,intensity and the development process of tornado storm remained a difficult problem in the field of numerical forecast research.The Doppler weather radar is the only tool to monitor the three-dimensional structure of tornado.The data obtained by Doppler weather radar is able to improve the initial field precision of numerical model.However,the traditional 88D S-band operational radar takes 6 minutes to complete a volume scan and its spatial resolution is 250 m,which cannot capture the development process of small-scale disasters.The newly developed X-band phased-array radar in the Pearl River region only needs 2 minutes to complete a volume scan with a small spatial resolution of 30 m,which can help to capture the development process of small-scale disasters and observe the internal fine structure.However,research about the assimilation of the high scanning frequency phased-array radar data is still rare.In this paper,the ensemble Kalman filter(EnKF)assimilation system developed by the Pennsylvania State University(PSU)is improved and Weather Research and Forecasting Model(WRF)is used.Two tornadoes happened in Foshan in 2018 are chosen as cases.This paper examines the influence of the cyclic assimilation of X-band phased-array radar data on the initial and forecast fields at high resolution scale.The first case is an EF1 tornado which occurs in the rainband around the typhoon"Ewiniar".Because of its weak intensity and short duration,we mainly study the intensity analysis and predictability of this tornado.It is found that the assimilation of both X-band phased-array radar data and S-band operational radar data can generate tornado storm vortex at the low level of model.The main body of vortex is vertical and could contact the ground.After increasing the frequency of assimilation of X-band phased-array radar data,the tornado vortex can be predicted 8 minutes in advance.The second case is an EF2 tornado which occurs in the rainband around the typhoon "Mangkhut".Because of its great strength and long duration,this study mainly focused the evolution process,structure analysis and path prediction of this tornado.It is found that after assimilating the X-band phased-array radar data,the analysis field can produce typical supercell storm characteristics like hook echo,low-level vortex et al.,which match well with the observation and classical storm model.And the tornado storm path can be well predicted by assimilating X-band phased-array radar data.