Research On Lightning Nowcasting Based On Radar And Lightning Observation

Lightning is one of common natural disasters,posing major threats to human life and industrial infrastructure.Lightning nowcasting under fine spatiotemporal resolution is of great significance for relieving the losses caused by this disaster.However,the nowcasting task is still facing serious challenges due to the instability of thunderstorm development.Considering the temporal and spatial characteristics of lightning,this paper classifies the problem of lightning nowcasting as a spatiotemporal series prediction problem,and proposes two lightning nowcasting methods.Firstly,in this paper,we propose a Master-slave Spatio-Temporal prediction Network(MSTNet)for lightning nowcasting.The model inputs consist of two spatiotemporal sequences: lightning observation and radar reflectivity in recent period.MSTNet employs a master encoder and a slaver encoder to mine the complementary information underling the two kinds of data,respectively.The master encoder aims at modeling the strong temporal dependency between historical lightning and future lightning;the slaver encoder intends to simulate the development of radar echoes,providing potential knowledge for lighting nowcasting.The features extracted by the two encoders are then integrated via a fusion module,based on which the final nowcasting results are generated.Second,this paper proposes an end-to-end lightning nowcasting model ARLNet in the scene of missing lightning data.ARLNet only uses historical radar reflectivity to forecast the near early warning of future lightning in the scene of high spatiotemporal resolution.This spatiotemporal sequence prediction method using heterogeneous data has a great test on the ability of feature extraction of the model.ARLNet uses two different attention mechanisms(attention mechanism based on geographical features and spatial domain)to enhance the feature extraction ability of the model,promote the model to learn the internal relationship between radar data and lightning data,and realize lightning nowcasting in the scene of high spatiotemporal resolution only using radar reflectivity.Experiments are conducted on a real-world dataset(released with our paper)collected in South China.The experimental results demonstrate that MSTNet and ARLNet achieve state-of-the-art performance compared with several advanced baselines utilized for spatiotemporal series weather nowcasting.