9:15 AM - 9:30 AM
[HDS10-02] VR simulation for Tsunami evacuation in snowy and cold cities
Keywords:Tsunami Evacuation Training, VR, GIS, Education for disaster prevention
1. Introduction
Since the Great East Japan Earthquake, there has been a heightened awareness of the danger of tsunami disasters in Japan, and tsunami countermeasures have been discussed on various scales, from the national level to the citizen's level. In Hokkaido, an 7-40% probability has been predicted for a large earthquake off the Chishima Trench within 30 years, making tsunami countermeasures in coastal areas an urgent necessity (Headquarters for Earthquake Research Promotion, 2018). Furthermore, the Hokkaido earthquake in September 2018 caused a simultaneous power outage (blackout) in Hokkaido due to the shutdown of a thermal power plant, highlighting the need for citizens to anticipate unforeseen circumstances. Additionally, evacuation drills are often conducted during the day for safety reasons, and it is difficult to conduct them at night or during snowfall, so it is necessary to consider evacuation under these conditions.
1.Purpose and Methodology
The purpose of this research is to develop a virtual tsunami evacuation training system using WebVR and conduct the training under nighttime conditions during the snowy season to reproduce the behavior of evacuees during a tsunami disaster in virtual space. A virtual evacuation training system was developed to reproduce nighttime conditions during the snow season near the Kanemori warehouse, one of Hakodate's tourist attractions. Subsequently, a virtual evacuation drill was carried out. Finally, the study estimated and discussed the evacuation behavior of the participants in the drill, in the context of a tsunami evacuation. The evacuation training system developed in this study employs WebVR and is compatible with a wide array of devices, including multi-functional terminals and VR-specific head-mounted displays.
3.Result
The experimental results revealed that during nighttime, when visibility was limited, participants tended to aim for high buildings and urban centers rather than elevated locations like Mt.Hakodate. Moreover, when participants were unfamiliar with the geography of Mount Hakodate, they tended to move away from Hakodate Bay and closer to the Pacific Ocean coastline. At crossroads, respondents were more likely to continue straight, deviating from the shortest evacuation route. Such behavior might be assumed to be based on the principle of leaving the coastline immediately in the event of a tsunami disaster. However, this is not necessarily the correct course of action in the unique topography of the Mount Hakodate area, and therefore, appropriate countermeasures are needed.
However, the study also left some critical issues unaddressed. The system was constrained in tracing subsequent behaviors of participants who chose incorrect directions, limiting the understanding of their decision-making processes. Extended VR experiences might lead to physical symptoms such as 3D sickness, necessitating careful consideration in the system's deployment and the duration of the experiences.
4.Conclusion
Future work will also aim to simulate and analyze phenomena that may occur during an earthquake, such as building collapses, fires, and landslides, to understand how disaster victims' behavior changes in varying scenarios.
This research marks a crucial step toward a more nuanced understanding of human behavior in disaster scenarios and opens up possibilities for more immersive, realistic, and accessible training and planning, potentially saving lives and reducing damage in actual events.
Note: This study is part of the results of the Ministry of Education, Culture, Sports, Science and Technology's "Earthquake and Volcano Observation Research Project for Contributing to Disaster Mitigation (Phase 2 and 3)".
[1] Characteristics of seismic activity in eastern Hokkaido (Abashiri, Kushiro, Nemuro regions), Earthquake Research Institute of Japan (online), available at: https://www.jishin.go.jp/regional_seismicity/rs_hokkaido/p01_tobu/
Since the Great East Japan Earthquake, there has been a heightened awareness of the danger of tsunami disasters in Japan, and tsunami countermeasures have been discussed on various scales, from the national level to the citizen's level. In Hokkaido, an 7-40% probability has been predicted for a large earthquake off the Chishima Trench within 30 years, making tsunami countermeasures in coastal areas an urgent necessity (Headquarters for Earthquake Research Promotion, 2018). Furthermore, the Hokkaido earthquake in September 2018 caused a simultaneous power outage (blackout) in Hokkaido due to the shutdown of a thermal power plant, highlighting the need for citizens to anticipate unforeseen circumstances. Additionally, evacuation drills are often conducted during the day for safety reasons, and it is difficult to conduct them at night or during snowfall, so it is necessary to consider evacuation under these conditions.
1.Purpose and Methodology
The purpose of this research is to develop a virtual tsunami evacuation training system using WebVR and conduct the training under nighttime conditions during the snowy season to reproduce the behavior of evacuees during a tsunami disaster in virtual space. A virtual evacuation training system was developed to reproduce nighttime conditions during the snow season near the Kanemori warehouse, one of Hakodate's tourist attractions. Subsequently, a virtual evacuation drill was carried out. Finally, the study estimated and discussed the evacuation behavior of the participants in the drill, in the context of a tsunami evacuation. The evacuation training system developed in this study employs WebVR and is compatible with a wide array of devices, including multi-functional terminals and VR-specific head-mounted displays.
3.Result
The experimental results revealed that during nighttime, when visibility was limited, participants tended to aim for high buildings and urban centers rather than elevated locations like Mt.Hakodate. Moreover, when participants were unfamiliar with the geography of Mount Hakodate, they tended to move away from Hakodate Bay and closer to the Pacific Ocean coastline. At crossroads, respondents were more likely to continue straight, deviating from the shortest evacuation route. Such behavior might be assumed to be based on the principle of leaving the coastline immediately in the event of a tsunami disaster. However, this is not necessarily the correct course of action in the unique topography of the Mount Hakodate area, and therefore, appropriate countermeasures are needed.
However, the study also left some critical issues unaddressed. The system was constrained in tracing subsequent behaviors of participants who chose incorrect directions, limiting the understanding of their decision-making processes. Extended VR experiences might lead to physical symptoms such as 3D sickness, necessitating careful consideration in the system's deployment and the duration of the experiences.
4.Conclusion
Future work will also aim to simulate and analyze phenomena that may occur during an earthquake, such as building collapses, fires, and landslides, to understand how disaster victims' behavior changes in varying scenarios.
This research marks a crucial step toward a more nuanced understanding of human behavior in disaster scenarios and opens up possibilities for more immersive, realistic, and accessible training and planning, potentially saving lives and reducing damage in actual events.
Note: This study is part of the results of the Ministry of Education, Culture, Sports, Science and Technology's "Earthquake and Volcano Observation Research Project for Contributing to Disaster Mitigation (Phase 2 and 3)".
[1] Characteristics of seismic activity in eastern Hokkaido (Abashiri, Kushiro, Nemuro regions), Earthquake Research Institute of Japan (online), available at: https://www.jishin.go.jp/regional_seismicity/rs_hokkaido/p01_tobu/