Author(s): Tomoya Kataoka, Yasuo Nihei
Linked Author(s): Tomoya Kataoka
Keywords: Water level, precipitation, annual variation, flood risk, big data analysis.
Abstract: We analyzed the water level and precipitation observed at many observation stations during the fourteen years from 2002 to 2015 in the 109 primary water systems in Japan. Based on this analysis, we evaluated the flood risks from the hourly water level (Risk 1) and the daily precipitation (Risk 2) using the data set of water levels at 1, 003 stations and precipitation at 2, 095 stations from the database of the Water Information System, Ministry of Land, Infrastructure, Transport and Tourism. The annual variations of both Risk 1 and Risk 2 did not increase linearly to a large extent. The number of Risk 1 among the water-level observation stations and water systems was the highest in 2004 and the lowest in 2008. The highest number of Risk 1 in 2004 was resulted from the landfall of 10 typhoons, which were the most frequent in recorded history. On the other hand, the number for Risk 2 was the highest in 2011 and the lowest in 2008. The highest number for Risk 2 was caused, again, by the landfall of a typhoon. Therefore, the annual variation of Risk 1 and Risk 2 depends on the frequency and magnitude of typhoon landfalls. In addition, five water systems with high flood risks (Tone, Shinano, Kiso, Shingu and Watari Rivers) were identified by comparing the horizontal distributions of Risk 1 and Risk 2. In the future, we will analyze the long-term water level and precipitation data observed at additional stations, and then investigate the trend of flood risks in Japan
Year: 2017