Scholarlink

<p>In this paper, we analyze the spatio-temporal pattern of lightning strokes in Korea in 2022 for hazard characterization using statistical methods. Descriptive statistics show that lightning strokes are highly concentrated within short time intervals. To describe the spatio-temporal concentration of lightning activity, we examine periods of elevated lightning occurrence in South Korea during 2022. Using an event-based clustering procedure, we identify 34 lightning clusters. Although these clusters occupy only a small fraction of the total time period (approximately 5<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\%\)</EquationSource> </InlineEquation>), they contain a substantial proportion of the recorded lightning strokes (about 70<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\%\)</EquationSource> </InlineEquation>), illustrating the highly intermittent nature of lightning activity. Furthermore, we model the global spatio-temporal behavior of lightning stroke clusters using principal curves, revealing that most lightning activity in Korea tends to extend from the Northwest to the Southeast. In contrast, to capture the local spatio-temporal dynamics of individual accumulated lightning events, we apply a nonparametric Hawkes process, which flexibly characterizes their clustered structures, including the branching ratio and characteristic space-time triggering scales. The identification of lightning cluster structures provides quantitative summaries useful for lightning risk assessment and situational awareness.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Characterizing local hazard scales of lightning strokes via nonparametric spatio-temporal hawkes processes

  • 期刊论文

  • 发表日期：

    2026年5月22日

查看全文
  • Junhyeon Kwon,
  • Seoncheol Park

摘要

In this paper, we analyze the spatio-temporal pattern of lightning strokes in Korea in 2022 for hazard characterization using statistical methods. Descriptive statistics show that lightning strokes are highly concentrated within short time intervals. To describe the spatio-temporal concentration of lightning activity, we examine periods of elevated lightning occurrence in South Korea during 2022. Using an event-based clustering procedure, we identify 34 lightning clusters. Although these clusters occupy only a small fraction of the total time period (approximately 5 \(\%\) ), they contain a substantial proportion of the recorded lightning strokes (about 70 \(\%\) ), illustrating the highly intermittent nature of lightning activity. Furthermore, we model the global spatio-temporal behavior of lightning stroke clusters using principal curves, revealing that most lightning activity in Korea tends to extend from the Northwest to the Southeast. In contrast, to capture the local spatio-temporal dynamics of individual accumulated lightning events, we apply a nonparametric Hawkes process, which flexibly characterizes their clustered structures, including the branching ratio and characteristic space-time triggering scales. The identification of lightning cluster structures provides quantitative summaries useful for lightning risk assessment and situational awareness.