The influence of orchard and forest management on rainfall-induced landslides: A study of Hiraenoki Community in southwestern Japan

Landslides triggered by heavy rainfall pose significant risks to mountainous communities worldwide, and their frequency is expected to increase due to climate change. Vegetation and land management are known to influence slope stability, but their effects in rural landscapes remain understudied. This study evaluates the influence of forest and orchard vegetation, along with topography, rainfall, and geology, on shallow landslide occurrence in the Hiraenoki Community, Asakura City, southwestern Japan, which experienced severe rainfall-induced disasters in July 2017. Landslide areas were identified using pre- and post-disaster digital elevation models, and vegetation and orchard data were compiled from forest registers, aerial photographs, and local interviews. Binary logistic regression analysis was conducted with landslide presence as the dependent variable and explanatory variables including vegetation type, orchard type, slope angle, slope aspect, elevation, rainfall, surface geology, and topographic position index (TPI). The final model revealed that mixed secondary forests and 50–59-year-old sugi cedar (Cryptomeria japonica) plantations exhibited higher landslide probabilities. Rainfall, slope angle, elevation, and TPI also significantly influenced landslide occurrence, with valleys and lower elevations being particularly vulnerable. The results suggest that overtopped arakashi (Quercus glauca) and sudajii (Castanopsis sieboldii) in mixed secondary forests and old sugi cedar are considered to have acted as triggers that increased the probability of landslides. Periodic thinning of canopy trees is proposed to enhance root anchorage and promote natural regeneration. These findings provide guidance for the conservation of satoyama landscapes with consideration of disaster risk.