Soil water content is one of the most important physical indicators of landslide hazards. Therefore, quickly and non-destructively classifying soils and determining or predicting water content are essential tasks for the detection of landslide hazards. We investigated hyperspectral information in the visible and near-infrared regions (400–1000 nm) of 162 granite soil samples collected from Seoul (Republic of Korea). First, effective wavelengths were extracted from pre-processed spectral data using the successive projection algorithm to develop a classification model. A gray-level co-occurrence matrix was employed to extract textural variables, and a support vector machine was used to establish calibration models and the prediction model. The results show that an optimal correct classification rate of 89.8% could be achieved by combining data sets of effective wavelengths and texture features for modeling. Using the developed classification model, an artificial neural network (ANN) model for the prediction of soil water content was constructed. The input parameter was composed of Munsell soil color, area of reflectance (near-infrared), and dry unit weight. The accuracy in water content prediction of the developed ANN model was verified by a coefficient of determination and mean absolute percentage error of 0.91 and 10.1%, respectively.

If a slope located near a densely populated region is susceptible to debris-flow hazards, barriers are used as a mitigation method by placing them in flow channels; i.e., flowpaths. Selecting the location and the design of a barrier requires hazard assessment to determine the width, volume, and impact pressure of debris-flow at the moment of collision. DAN3D (Three-Dimensional Dynamic Analysis), a 3D numerical model for simulating debris-flow, has been widely used to perform hazard assessment; however, solely using DAN3D would be both insufficient and inefficient in finding the optimal barrier location. Therefore, the present study developed a framework that interprets the results from DAN3D simulation without considering any barriers. Then, the framework generates hazard assessment maps showing the impact parameters of debris-flow along the flowpath by various algorithms and machine learning methods, such as the k-means clustering algorithm, and also computes the width of the debris-flow, which is not explicitly calculated in DAN3D. A case study of the debris-flow at Umyeon mountain, Korea, in 2011, was used to generate hazard assessment maps. The maps were demonstrated to be a tool to quickly compute the impact parameters for conceptual barrier design with the aim of finding potential barrier locations.

Soil water content is one of the most common physical parameters that cause landslides or debris flow. Therefore, it is of very importance to determine or predict the water content variation due to infiltration of rainfall quickly and non-destructively. This study investigates the hyperspectral informations in the visible near-infrared regions (VNIR, 400nm~1000nm) of different samples of granite soils possessing varying water contents. Totally 162 granite samples were taken from 3 mountain areas. A Partial Least Squares Regression (PLSR) analysis was applied to develop calibration models and prediction models.  In the water content variation prediction model, the Area of ​​Reflectance(Near-infrared, NIR) parameter was the most suitable parameter to determine the water content. The results demonstrate that the hyperspectral camera combined with the PLSR model can be a useful and non-destructive tool for the determination of soil water content variation in the weathered granite soils that could be applied to the evaluation of possible instable area in a mountain site.

In order to mitigate the potential damage from debris-flow hazards, the debris-flow hazard assessment is conducted to quantify the risk of landslide occurrence and assign potential levels of damage to surrounding buildings. DAN3D (Three-Dimensional Dynamic Analysis), a 3D numerical model for debris-flow simulation, has been widely used to conduct the debris-flow hazard assessment by computing the magnitudes of debris-flow characteristics, such as the velocity, depth, and volume. DAN3D software presents the results on a map plot that shows the magnitude of debris-flow characteristics over the inundated areas at a particular simulation time. These plots neither provide the exact values of debris-flow characteristics at a specific location nor compute the area and the width of debris-flow. Furthermore, a static image can be inconvenient for visualizing the changes in the debris-flow characteristics through time. Therefore, the present study created a program to interprets the DAN3D simulation results and to generate interactive plots using web programming. The interactive plots represent the position of a debris-flow cluster with a centroid location and show debris-flow characteristics, including the area and width of debris-flow, with a color scale. Additionally, the generated plots provide a graphical-user-interface to extract more details or change the plot. The accessibility and customization provided by the generated plots can be very useful for the design of the protection measures and evaluation of the effectiveness of them. A case study of the debris-flow at Sindonga watersheds, Mt. Umyeon, Korea, in 2011, was used to generate several interactive plots and their usefulness in designing barriers as mitigation against debris-flow.