A Fast and Fine‐Resolution Location Method for Lightning Channels Based on Time Series Segmented Feature of Low Frequency Signal

Abstract Most real‐time lightning location systems are based on feature matching to locate lightning, but they often lack the ability to locate lightning channels. To achieve lightning channel location based on feature matching, a new location algorithm is proposed by utilizing time series segmented...

Full description

Saved in:
Bibliographic Details
Main Authors: Jingxuan Wang, Yang Zhang, Yanfeng Fan, Yijun Zhang, Dong Zheng, Weitao Lyu
Format: Article
Language:English
Published: American Geophysical Union (AGU) 2025-03-01
Series:Earth and Space Science
Online Access:https://doi.org/10.1029/2024EA003896
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Most real‐time lightning location systems are based on feature matching to locate lightning, but they often lack the ability to locate lightning channels. To achieve lightning channel location based on feature matching, a new location algorithm is proposed by utilizing time series segmented feature to match lightning pulses. The features of waveform time series do not require complex signal processing, making it suitable for real‐time and fast location. Compared with the location results of the other three existing methods for a lightning event, the new method achieves the highest matching efficiency of 28.4% and demonstrates fine channel location capability. For a thunderstorm process, the new method also has the highest location efficiency, as well as the highest number of valid location points per second, and the lowest computation time of per valid location point, which are 26%, 39.6, and 0.025s, respectively. The new algorithm also provides better location results for irregular pulse clusters, which more realistically depict the development process of downward leader compared to the location methods based on encoding feature matching. This may be caused by the fact that the time series segmented feature can correctly represent the change trend of the signal under the condition of low signal‐to‐noise ratio.
ISSN:2333-5084