Oct 24, 2014
Prediction Research to Prevent and Mitigate Damage from Natural Disasters
Dr. Yoshiyuki Kaneda serves as manager of R&D projects for the HPCI strategic program field 3 -Advanced Prediction Research for Natural Disaster Prevention and Reduction-conducting research for improvement of prediction accuracy of earthquakes and tsunamis. This research has three main themes: earthquake, tsunami, and evaluation of damage to cities caused by earthquakes and tsunamis. This research is conducted by using the K computer. Research for disaster prevention and mitigation measures is being continued with the aim of the construction of a highly detailed and accurate integrated simulator.
"We witnessed tremendous tsunami damage from the Great East Japan Earthquake in 2011. Either a megathrust earthquake in the Nankai Trough or an earthquake with an epicenter in the Tokyo metropolitan area that will likely occur at some point is expected to cause great damage. When considering a complex disaster from an earthquake and tsunami that covers a wide area, we are able to perform various extremely sophisticated simulations of the Japanese archipelago, southwestern Japan, and the Tokyo metropolitan area. This is a really wonderful thing. By using a supercomputer like the K computer that allows use of a greater number of more complex parameters, we can start to see a part of phenomena that we have not been able to understand before. I believe this is something to look forward to on the research front."
Based on data obtained from a survey conducted after the Great East Japan Earthquake, a tsunami model and a town model are built on the K computer. The degree of accuracy and reliability of the simulation is being verified now by comparing the simulated damage to actual damage. Currently, simulations can use a mesh resolution setting of 5 meters.
"In fact, we are studying to what extent we can make various evaluations in real time when using next-generation supercomputers including the current K computer. Of course it is not enough for only the computer to simply be fast. In other words, it requires acquisition of various measurement data to be as immediate as possible and to be able to use the data in simulations. To do this, it is important to properly acquire various data particularly on earthquakes and tsunamis. For tsunamis this means offshore data. This requires improvement of observation systems and establishment of a framework allowing immediate use of the data."
In this project, archiving acquired data into a database will make various detailed simulations at the city and town level possible. Secured access to the database will be made possible from a location that is safe even if an earthquake or a tsunami were to occur in southwestern Japan or in the Tokyo metropolitan area.By utilizing this database, the goal is to envisage and evaluate the scale of damage in real time or semi-real time.
- This Week
- This Month
- All Time