Research On Thunderstorm Identification And Tracking Method For Dual Polarization Doppler Weather Radar

As one of the most severe weather systems in the world,thunderstorms have caused great harm to human production and life due to its strong discharge process between the atmosphere and the ground.Therefore,improving the recognition accuracy and tracking efficiency of thunderstorms has very important practical significance.This paper studies the method of thunderstorm identification and tracking based on S-band dual-polarization doppler weather radar,aiming to improve radar detection of thunderstorms.The main research contents and results of the thesis are as follows:(1)The radar data is filtered and speed deblurred.The“denoising”process of K_DP.is performed by a 9-point and 25-point dual-window sliding filtering method,and the system noise of Z_DR.is corrected by the natural target method.Therefore,the reliability of the radar data is improved.At the same time,a large number of strong monomers were selected as the sample set for current research.(2)The original data was pre-processed to obtain radar CAPPI(Constant Altitude Plan Position Indicator),CR(Combined reflectivity)and VCS(Vertical cross sections)data.After a statistical analysis of spatio-temporal characteristics of radar parametersZ_H,Z_DR andK_DP with the development and evolution of thunderstorms,an optimal index of thunderstorms identification based on each of three parameters is proposed.(3)The radar data and sounding data are used to the identification of hydrometeors in the cloud through fuzzy logic algorithm,and the recognition results are verified.After a correlation analysis between the hydrometeors in the cloud and the development and evolution of thunderstorms,the thunderstorm identification index based on hydrometeors distribution is proposed.(4)By using the polarization parameters and hydrometeors distribution characteristics,the traditional storm identification and tracking algorithm is improved to a new algorithm for thunderstorm recognition and tracking.The effect of recognition tracking method was evaluated and an identification model of the initial thunderstorm was initially established.