Identification Of Hail Storms In The Early Stage And Hail Forecast

Hail storm is a kind of meteorological disaster caused by severe convective,which has the characteristics of fast generation and short life cycle.These characteristics of hail bring obstacles to hail identification and hail forecast.At present,most weather forecasters use weather radar products to identify,track and forecast hail.The main work of this paper is as follows.Firstly,the three-dimensional grid data are obtained through a series of Doppler Weather Radar data processing.Secondly,ten features are constructed to represent the strength of updraft,the density,scale,and volume of hydrometeors of a hail storm.These ten features encompass three aspects:the height,intensity,and overhang.The two-dimensional comprehensive features(VMCF and HGCF)are obtained through feature analysis and combination.Through analysis,the time series exhibit a distinct increase in VMCF or HGCF values in the early stage of hail storms.Moreover,a decision tree model is trained based on the VMCF and HGCF increments for 49 hail cases and 35 heavy rainfall cases with S-band radar data.Finally,the detailed analysis of five cases of supercell storm which can produce hail is carried out.The forecast method of supercell storm which produce hail is realized according to their common characteristics.Based on the distinct increase of the two-dimensional comprehensive features,the early stage of supercell storm is identified.Then,we focus on the flow field information.When the marc,mesocyclone and overhang structure appear at the same time,we judge that the ordinary hail cell has evolved into supercell,and there will be hail falling after half an hour.In summary,this paper proposes a hail recognition method from a new perspective.This method can identify hail in the early stage,and achieve the forecast of supercell storm which can produce hail.The paper provides a new way for severe convective weather recognition based on radar.