Scholarlink

<p>Issuing a tsunami warning rapidly is crucial to maximize the lead time for evacuation and other response actions. However, in the first minutes after an earthquake, no tsunami observations are available, and forecasts rely on rapid but uncertain estimates of earthquake location and magnitude. As more data become available, forecasts can be updated for more distant coastlines or near-source immediate post-event assessments. Here we present a method for real-time updating of Probabilistic Tsunami Forecasting that continuously integrates new seismic and tsunami observations. We apply it to the 2010 Mw8.8 Maule event and to several synthetic and real Mw 6.5–8.5 events in the Mediterranean Sea, demonstrating that it consistently preserves accuracy and generally reduces uncertainty, especially when using local tsunami data rather than focal mechanisms. Based on reweighting the initial scenario ensemble, this evolutionary approach enables fast updates and synchronization of forecast uncertainty with evolving information.</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

Dynamic management of uncertainty in rapid tsunami forecasting

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月6日

查看全文
  • Louise Cordrie,
  • Jacopo Selva,
  • Fabrizio Bernardi,
  • Roberto Tonini,
  • Fabrizio Romano,
  • Manuela Volpe,
  • Stefano Lorito

摘要

Issuing a tsunami warning rapidly is crucial to maximize the lead time for evacuation and other response actions. However, in the first minutes after an earthquake, no tsunami observations are available, and forecasts rely on rapid but uncertain estimates of earthquake location and magnitude. As more data become available, forecasts can be updated for more distant coastlines or near-source immediate post-event assessments. Here we present a method for real-time updating of Probabilistic Tsunami Forecasting that continuously integrates new seismic and tsunami observations. We apply it to the 2010 Mw8.8 Maule event and to several synthetic and real Mw 6.5–8.5 events in the Mediterranean Sea, demonstrating that it consistently preserves accuracy and generally reduces uncertainty, especially when using local tsunami data rather than focal mechanisms. Based on reweighting the initial scenario ensemble, this evolutionary approach enables fast updates and synchronization of forecast uncertainty with evolving information.