The China–Pakistan Karakoram Highway is an important land route from China to South Asia and the Middle East via Pakistan. Due to the extremely hazardous geological environment around the highway, landslides, debris flows, collapses, and subsidence are frequent. Among them, debris flows are one of the most serious geological hazards on the Karakoram Highway, and they often cause interruptions to traffic and casualties. Therefore, the development of debris flow susceptibility mapping along the highway can potentially facilitate its safe operation. In this study, we used remote sensing, GIS, and machine learning techniques to map debris flow susceptibility along the Karakoram Highway in areas where observation data are scarce and difficult to obtain by field survey. First, the distribution of 544 catchments which are prone to debris flow were identified through visual interpretation of remote sensing images. The factors influencing debris flow susceptibility were then analyzed, and a total of 17 parameters related to geomorphology, soil materials, and triggering conditions were selected. Model training was based on multiple common machine learning methods, including Ensemble Methods, Gaussian Processes, Generalized Linear models, Navies Bayes, Nearest Neighbors, Support Vector Machines, Trees, Discriminant Analysis, and eXtreme Gradient Boosting. Support Vector Classification (SVC) was chosen as the final model after evaluation; its accuracy (ACC) was 0.91, and the area under the ROC curve (AUC) was 0.96. Among the factors involved in SVC, the Melton Ratio (MR) was the most important, followed by drainage density (DD), Hypsometric Integral (HI), and average slope (AS), indicating that geomorphic conditions play an important role in predicting debris flow susceptibility in the study area. SVC was used to map debris flow susceptibility in the study area, and the results will potentially facilitate the safe operation of the highway.

Groups of landslides induced by heavy rainfall are widely distributed on a global basis and they usually result in major losses of human life and economic damage. However, compared with landslides induced by earthquakes, inventories of landslides induced by heavy rainfall are much less common. In this study we used high-precision remote sensing images before and after continuous heavy rainfall in southern Tianshui, China, from 20 June to 25 July 2013, to produce an inventory of 14,397 shallow landslides. Based on the results of landslide inventory, we utilized machine learning and the geographic information system (GIS) to map landslide susceptibility in this area and evaluated the relative weight of various factors affecting landslide development. First, 18 variables related to geomorphic conditions, slope material, geological conditions, and human activities were selected through collinearity analysis; second, 21 selected machine learning models were trained and optimized in the Python environment to evaluate the susceptibility of landslides. The results showed that the ExtraTrees model was the most effective for landslide susceptibility assessment, with an accuracy of 0.91. This predictive ability means that our landslide susceptibility results can be used in the implementation of landslide prevention and mitigation measures in the region. Analysis of the importance of the factors showed that the contribution of slope aspect (SA) was significantly higher than that of the other factors, followed by planar curvature (PLC), distance to river (DR), distance to fault (DTF), normalized difference vehicle index (NDVI), distance to road (DTR), and other factors. We conclude that factors related to geomorphic conditions are principally responsible for controlling landslide susceptibility in the study area.

Debris flows represent major hazards in most mountainous regions of the world where they repeatedly result in disasters. In order to protect people and infrastructure against future debris flows, many debris flow catchments have been artificially intervened by employing various mitigation measures, including civil engineering works. However, the commonly adapted engineering measures, such as check dams, are not effective for every debris flow catchment, and the failure of such measures even causes more damage, e.g. the Sanyanyu debris flow catchment in Zhouqu, China, killed 1756 people. In order to research the effectiveness of engineering strategies and explore much more effective mitigation works for debris flows in the mountainous regions, we took the Bailong River catchment of Southern Gansu of China as study area, with special emphasis on Sanyanyu debris flow catchment (with civil engineering works) and Goulinping debris flow catchment (without civil engineering works), and comparatively analysed the two catchments. The comparative results show that both catchments have similar material source, geomorphological/environmental and climatic conditions, however, vegetation cover and rock hardness are poorer in Goulinping than in Sanyanyu, the catchment that underwent larger-scale debris flows, suggesting that the mitigation measures had been applied in Sanyanyu catchment were inappropriate. Subsequently, we simulated the effectiveness of controlling debris flow peak discharge with check dams at the lower part of Sanyanyu and Goulinping catchment using the Kanako simulator, and summarised argument based on the hypothesis and facts from positive and negative aspects. We draw the conclusion that it is not reasonable to build check dams in the two catchments and instead, drainage channels should be primarily considered for reducing debris flow hazards in such densely populated areas. Finally, we undertook detailed field investigations and experiments on the native plants in the region, and found that the ecological mitigation measure with planting Robinia Pseudoacacia on the debris flow deposits is an effective method to alleviate debris flow hazards. It is concluded that channel works combined with ecological measures are the preferable approaches to minimize the debris flow damage in debris flow catchments characterised with high mountains, concentrated rainfalls and active neotectonic movement.

Debris flows are a major geological hazard in mountainous regions. For improving mitigation, it is important to study the spatial distribution and factors controlling debris flows. In the Bailong River Basin, central China, landslides and debris flows are very well developed due to the large differences in terrain, the complex geological environment, and concentrated rainfall. For analysis, 52 influencing factors, statistical, machine learning, remote sensing and GIS methods were used to analyze the spatial distribution and controlling factors of 652 debris flow catchments with different frequencies. The spatial distribution of these catchments was divided into three zones according to their differences in debris flow frequencies. A comprehensive analysis of the relationship between various factors and debris flows was made. Through parameter optimization and feature selection, the Extra Trees classifier performed the best, with an accuracy of 95.6%. The results show that lithology was the most important factor controlling debris flows in the study area (with a contribution of 26%), followed by landslide density and factors affecting slope stability (road density, fault density and peak ground acceleration, with a total contribution of 30%). The average annual frequency of daily rainfall > 20 mm was the most important triggering factor (with a contribution of 7%). Forest area and vegetation cover were also important controlling factors (with a total contribution of 9%), and they should be regarded as an important component of debris flow mitigation measures. The results are helpful to improve the understanding of factors influencing debris flows and provide a reference for the formulation of mitigation measures.