Bayesian inference of forest road failure frequency depending on rainfall intensity for every prefecture in Japan

Autor: Muneoka, Hiroko, Shirasawa, Hiroaki
Zdroj: International Journal of Forest Engineering; January 2025, Vol. 36 Issue: 1 p89-102, 14p
Abstrakt: ABSTRACTTo make a quantitative prediction on forest road failures in the future under the influence of climate change, forest road failure frequency depending on rainfall intensity needs to be clarified. The forest road failure frequency at a certain rainfall intensity should vary regionally because of the variability in susceptibility. In the current situation of a lack of nationwide detailed forest road failure inventory data and GIS-based forest road data, this study analyzed the relationship between rainfall intensity and forest road failure frequency based on Bayesian inference using data available for all the 47 prefectures in Japan; the annual number of forest road failures, forest road length, analyzed precipitation data and polygon data of forest areas. The predicted forest road failure frequency varied broadly by prefecture; 10−4–10−2km−1at a rainfall event of 100 mm/24 h, while 10−2–100km−1at that of 400 mm/24 h. The frequency at the same rainfall intensity tended to be lower in conventionally rainy prefectures. The results imply that the number of forest road failures in currently less rainy regions may show a larger increase in the future as intense rainfall events are predicted all over Japan due to climate change.
Databáze: Supplemental Index