Japan Geoscience Union Meeting 2021

Presentation information

Print

[J] Oral

S (Solid Earth Sciences ) » S-CG Complex & General

[S-CG53] Reducing risks from earthquakes, tsunamis & volcanoes: new applications of realtime geophysical data

Sun. Jun 6, 2021 1:45 PM - 3:15 PM Ch.18 (Zoom Room 18)

convener:Masashi Ogiso(Meteorological Research Institute, Japan Meteorological Agency), Masumi Yamada(Disaster Prevention Research Institute, Kyoto University), Yusaku Ohta(Research Center for Prediction of Earthquakes and Volcanic Eruptions, Graduate School of Science, Tohoku University), YAMAMOTO Naotaka CHIKASADA(National Research Institute for Earth Science and Disaster Resilience), Chairperson:Naotaka YAMAMOTO CHIKASADA(National Research Institute for Earth Science and Disaster Resilience), Yusaku Ohta(Research Center for Prediction of Earthquakes and Volcanic Eruptions, Graduate School of Science, Tohoku University)

1:45 PM - 2:00 PM

[SCG53-01] An advanced system for automatic strong gust detection and warning for railroads using deep learning

★Invited Papers

*Kusunoki Kenichi1, Toru Adachi1, Hanako Inoue1, Osamu Suzuki1, Naoki Ishitsu2, Ken-ichiro Arai2, Chusei Fujiwara3, Hiroto Suzuki3 (1.Meteorological Research Institute, 2.Alpha-denshi Co., Ltd., 3.East Japan Railway Company)

Keywords:tornado, Doppler radar

In 2007, the Meteorological Research Institute and the East Japan Railway company started a project to develop an automatic strong gust detection and warning system for railroads , which the decision to warn is based upon information from a single-Doppler radar at low elevation angles. Through over ten years of R&D, it had reached the stage of practical application with a reliable real-time performance and a comprehensive operational system has been implemented in some areas over the coast of the Sea of Japan (limited to Yamagata, Akita, and Niigata Prefectures) . In this presentation, we will introduce an overview of the project since the introduction of the operational system since 2017 and the recent progresses especially about the use of deep learning. Possibilities for further development of the sysytem and their application to other traffic systemss will be also discussed.

This study is partly supported by Cabinet Office, Government of Japan, Public/Private R&D Investment Strategic Expansion Program (PRISM).