2015年5月24日(日) 16:15 〜 18:00 102B (1F)

コンビーナ:*鶴岡 弘(東京大学地震研究所)、Danijel Schorlemmer(GFZ German Research Centre for Geosciences)、平田 直(東京大学地震研究所)、座長:庄 建倉(統計数理研究所)、石辺 岳男(東京大学地震研究所)

Modern earthquake predictability studies emphasize the importance of testing earthquake forecast models. Japan, California, New Zealand, and other regions are currently operating or developing regional centers for such testing experiments under the auspices of the Collaboratory for the Study of Earthquake Predictability (CSEP). The experiments in Japan were started on 1st November 2009 and various models are currently under test in three testing regions (so-called 'AllJapan', 'Mainland' and 'Kanto'), each covering different forecasting periods (1 day, 3 months, 1 year, and 3 years). The 2011 Tohoku earthquake and its aftershocks has provided a great opportunity for these testing experiments . This session aims at discussing their results and the lessons learned from this earthquake leading to improvements of earthquake forecast models. Current testing in Japan is limited to earthquake rates. We solicit presentations that discuss extending CSEP activities in Japan to the fields of external forecasts, hazard model testing, ground-motion prediction testing, risk model testing, and the development of new testing metrics. All presentations exploring the limits of understanding the earthquake phenomenon and its impact on society are welcome.