It is widely accepted that land use and land cover (LULC) is an important conditioning factor for landslide occurrence, especially when considering the role of tree roots in stabilizing slopes and consolidating the soil. However, it is still difficult to assess the impacts of a specific LULC type on landslide distribution. The objective of the present study is to reveal the relationship between bamboo and landslide distribution at the regional scale. We aim to answer the following question: do the areas covered by bamboo have a higher susceptibility to landslides? Wenzhou City in SE China was taken as the study area, and a landslide inventory containing 1725 shallow landslides was constructed. The generalized additive model (GAM) was employed to assess the significance of LULC and nine additional factors, all of which were generated using the GIS platform. The frequency ratio (FR) method was used to analyze and compare the landslide density in each LULC category. Machine learning models were applied to perform landslide susceptibility mapping of the region. The results show that in the Wenzhou region, LULC is the second most important factor for landslide occurrences after the slope factor, whereas bamboo has a relatively higher FR value than most other LULC categories. The accuracies of the landslide susceptibility maps obtained from the random forest and XGBoost models were 79.6% and 85.3%, respectively. Moreover, 23.8% and 25.5% of the bamboos were distributed in very-high- and high-susceptibility-level areas. The incidents and density of landslides in bamboo areas were significantly higher than those with debris flow and rock collapses, indicating a promotional effect of bamboo on slope failure in the study area. This work will improve our understanding regarding the role of geological and ecological conditions that affect slope stability, which may provide useful guidance for land use planning and landslide risk assessment and mitigation at the regional scale.
