Improved Method for Rapid Identification and Warning of Sudden Severe Convection Weather

A fixed threshold discriminance method was used in the conventional Doppler weather radar for severe convection disaster weather monitoring to get the warning results of hail and lightning disaster in severe storms,which is not applicable to the identification and warning of severe convective weather in different latitude and longitude,seasons and complex terrain. In this paper,the TITAN algorithm was improved by the recursive regional growth method to quickly identify the three dimensional strong storm and its radar characteristic physical quantities. The historical observation data of Doppler weather radar and TRMM space-borne meteorological radar were used to retrievel the severe storm disaster weather Logistics multivariate linear regression probabilistic warning model under the conditions of complex season and topography in Shijiazhuang,Hebei province. Doppler radar identification and warning experiments of hail and lightning were carried out for a strong squall line in summer in Shijiazhuang and a super multi-cell storm near in spring in Shijiazhuang.The experimental results showed that the proposed method has higher identification accuracy based on the azimuth and distance than the traditional algorithm. Moreover,the false negative and false report rates of the proposed method are low,which helps to quickly identify and warn against severe weather.