09:45 〜 10:00
[AAS04-04] Real-time Average Recurrence Interval Map of Accumulative Rainfall for Japan
キーワード：再現期間、積算雨量、レーダー、実時間
Climate change has spurred more heavy rainfall events, triggering deadly flooding and landslides in Japan in recent years, and is predicted to intensity in the coming decades. Although the relationship between rainfall and the risk of disasters is not simple, communicating flood risk in terms of rareness can help us be aware of the threat posed by extreme rainfall. An average recurrence interval (ARI), also known as the return period, has been used for hydrologic infrastructure design and to describe extreme events for decades. Hence, describing the real-time heavy rainfall in a similar manner can provide an indicator of hydrological extremes and help the public understand the potential of disasters. This presentation will introduce a new system for conveying and expressing the magnitude of accumulative rainfall using the ARI map. This system includes eight types of accumulative rainfall datasets, 3-hour, 6-hour, 12-hour, 24-hour, 48-hour precipitations, and 90-minutes, 72-hour effective rainfalls, which describe the geographical variation. Firstly, the rainfall data is generated from the 1-minute rainfall intensity observed by the radar network (XRAIN). And then, the rainfall maps are converted into the ARI maps using pre-determined ARI statics given at each 5-km grid covering the whole of Japan. The pre-determined ARI statics is selected among five probability distribution functions (PDF), Gumbel, GEV, SQRT-ET, three-parameter log-normal, and Pearson type III distribution. The parameters for each PDF are fitted using the Radar/Rain gauge-Analyzed Precipitation since 1989 published by MLIT. The accumulative rainfall and ARI maps are processed within 10 minutes, and several maps are available online.