Research On Lightning Activity Characteristics And Approaching Warning Methods Based On Multi-source Data

With the development of lightning positioning technology and observation methods,single observation data can no longer meet the current business requirements for lightning monitoring and early warning.According to the different characteristics of various data,the use of different observation data for comprehensive lightning monitoring and early warning has developed into a trend.Based on the VLF/LF magnetic field lightning detection and positioning network independently built by Nanjing University of Information Science and Technology in Nanjing and Kunming,the lightning positioning data,radar reflectivity data and satellite cloud top temperature data are combined to extract characteristic parameters of thunderstorms in different terrains.Then the TITAN algorithm is used to identify,track,and extrapolate thunderstorm areas to achieve the purpose of early warning of thunder and lightning.The main results are as following.The Monte Carlo method was used to simulate the theoretical horizontal positioning errors of the Nanjing and Kunming lightning location networks.The results show that the horizontal errors in the Nanjing and Kunming lightning location networks are both less than250m.Considering the rugged and complex terrain in Kunming,the effect of real terrain on the horizontal error in Kunming is further simulated.The results show that the mountain topography in Kunming and surrounding areas does reduce the positioning accuracy,and the error increases by about 2km within a range of tens of kilometers.Based on the synchronization data of the VLF/LF magnetic field lightning detection system,the three-dimensional positioning result of a ground-to-ground lightning was analyzed.The lightning started at a height of about 5.7km in the cloud,and then developed upward and downward.The average upward development speed was about 4.27×10⁵ m·s^-1,and the downward average development speed was about 4.74×10⁵ m·s^-1.The rising edge time of the first return waveform is about 8ms,the falling edge time is about 33ms,and the half-peak width is about 14ms.The average speed of the first return attack was about 3.5×10⁵m·s^-1,and the average speed of subsequent return attacks was an order of magnitude larger than that of the first return attack,about 3.875×10⁶ m·s^-1.It shows that the lightning detection system has better monitoring and resolution capabilities for the three-dimensional lightning discharge channel.Lightning positioning data in Nanjing and Kunming,combined with radar reflectivity and cloud top temperature,are used to extract characteristic parameters of lightning occurrence.For Nanjing,30dBz radar echo tops up to 7km or cloud top temperature gradients up to 0.8°C/km and cloud top temperature drop rate up to 12°C/h can be used as reference indicators for lightning occurrence.And the area where the radar reflectivity is greater than25dBz or the cloud top temperature is less than 230K can be regarded as the lightning-intensive area;For Kunming,30dBz radar echo tops up to 8km or cloud top temperature gradients up to 1°C/km and cloud top temperature drop rate up to 12°C/h can be used as reference indicators for lightning occurrence.And the area where the radar reflectivity is greater than 30dBz or the cloud top temperature is less than 220K can be regarded as the lightning-intensive area.For the typical multi-cell thunderstorm in Nanjing,a good extrapolation warning effect is provided by radar reflectivity and cloud top temperature,and the effect of cloud top temperature is better.The lightning data cannot be used for warning because of the randomness.For the typical local mountain thunderstorms in Kunming,a good and refined extrapolation warning effect is provided by the lightning location data.The cloud top temperature is suitable for large-scale weather situation judgments.The single radar data is not suitable for extrapolating early warning of thunderstorm areas because of many limitations.