Machine Learning-Based and 3D Kinematic Models for Rockfall Hazard Assessment Using LiDAR Data and GIS

Abstract

Rockfall is one of the most hazardous phenomena in mountainous and hilly regions with high and steep terrain. Such incidents can cause massive damage to people, properties, and infrastructure. Therefore, proper rockfall hazard assessment methods are required to save lives and provide a guide for the development of an area. The aim of this research is to develop a method for rockfall hazard assessment at two different scales (regional and local). A high-resolution airborne laser scanning (ALS) technique was utilized to derive an accurate digital terrain model (DTM); next, a terrestrial laser scanner (TLS) was used to capture the topography of the two most critical areas within the study area. A staking machine-learning model based on different classifiers, namely logistic regression (LR), random forest (RF), artificial neural network (ANN), support vector machine (SVM), and k-nearest neighbor (KNN), was optimized and employed to determine rockfall probability by utilizing various rockfall conditioning factors. A developed 3D rockfall kinematic model was used to obtain rockfall trajectories, velocity, frequency, bouncing height, kinetic energy, and impact location. Next, a spatial model combined with a fuzzy analytical hierarchy process (fuzzy-AHP) integrated in the Geographic Information System (GIS) was developed to assess rockfall hazard in two different areas in Ipoh, Malaysia. Additionally, mitigation processes were suggested and assessed to provide a comprehensive information for urban planning management. The results show that, the stacking random forest–k-nearest neighbor (RF-KNN) model is the best hybrid model compared to other tested models with an accuracy of 89%, 86%, and 87% based on training, validation, and cross-validation datasets, respectively. The three-dimension rockfall kinematic model was calibrated with an accuracy of 93% and 95% for the two study areas and subsequently the rockfall trajectories and their characteristics were derived. The assessment of the suggested mitigation processes proves that the proposed methods can reduce or eliminate rockfall hazard in these areas. According to the results, the proposed method can be generalized and applied in other geographical places to provide decision-makers with a comprehensive rockfall hazard assessment.

An Integration of UAV-Based Photogrammetry and 3D Modelling for Rockfall Hazard Assessment: The Cárcavos Case in 2018 (Spain)

“…Among remote acquisition techniques, one of the most used is Light Detection and Ranging (LiDAR), which provides a 3DPC of the scanned surface. This type of technique is widely used, both for characterization of rockfalls and for related risk assessment, as well as for monitoring and modelling purposes [45,46,53]. In this regard, despite the LiDAR technique allowing for the obtaining of greater accuracy, UAV photogrammetry was used for detailed investigations of rockfalls in remote areas characterized by difficult access-as in the case of Cárcavos 80 m vertical scarp.…”

“…However, fragmentation was not considered in such studies and mitigation measures were not evaluated. Another study [53] optimized a hybrid machine learning model, based on various classifiers, to produce a rockfall probability map and thus identify rockfall sources at regional scale, based on results of a 3D rockfall kinematic model to assess rockfall trajectories and their characteristics at a local scale. The authors also developed a spatial model based on fuzzy analytical hierarchy process, integrated into a GIS, to generate the final rockfall hazard maps, aiming at suggesting and assessing mitigation actions.…”

An intelligent framework for end‐to‐end rockfall detection

“…Point clouds are heavily utilized by various researchers, mainly because they carry detailed, high quality information. Specifically, recent studies, including References [4,24-30] use point clouds as their main data source for their analysis. Moreover, advanced research studies use both images and point clouds portraying a speciality and differentiate themselves from the majority of the related studies 31‐33 .…”

Linking the Remote Sensing of Geodiversity and Traits Relevant to Biodiversity—Part II: Geomorphology, Terrain and Surfaces

“…Remote sensing (RS)-aided derived in monitoring examples in terrain and surfaces, aeolian geomorphology, fluvial geomorphology and coastal geomorphology landslides and their traits. Mountain types, relief types, relief classes IKONOS OSA 3/M , DHM25 3/R , GTOPO30-DEM 3/R , LiDAR 2/L [330][331][332] Volcano types (volcanic full forms),volcanoes, lava flow fields, hydrothermal alteration, geothermal explorations, heat fluxes, volcanoes hazard monitoring Doves-PlanetScop, Terra/Aqua MODIS 3/M , EO-1 ALI 3/M , Landsat-8 OLI 3/M/TIR , Terra ASTER 3/M/TIR , MSG SEVIRI 3/M/TIR , LiDAR 2/L [333][334][335][336][337] Mountain hazards, mass movement (rock fall probability, boulders, denudation, mass erosion, rock decelerations, rotation changes, slope stability, rock shapes, particle shapes, patterns, structures, faults and fractures, holes and depressions) InSAR 3/R , SAR 3/R , LiDAR 2/L , Digital Orthophoto 1/RGB [338][339][340][341][342][343][344][345][346][347] Landslide chances, landslide evolution Digital Orthophoto 1/RGB [348] Above ground-chances, disturbances Opencast mining, sand mining and extraction, tipping, dumps TanDEM-X 3/R , SRTM DEM 3/R , ALOS PALSAR 3/R , ERS-1 3/R , GeoEye GIS 3/M , WorldView-3 Imager 3/M , IKONOS OSA 3/M , Landsat-5 TM/-7 ETM+/-8 OLI 3/M/TIR , IRS-P6 LISS-III 3/M , High resolution satellite data of Google 3/M , LiDAR 2/L [349][350][351][352][353][354][355] Vegetation traits as proxy of the geochemical parameters HyMAP 2/H [356] Below ground-chances, disturbances Salt mines, fracking ERS-1/-2 3/R , ASAR 3/R , ALOS PALSAR 3/R , Landsat-5 TM/-7 ETM+/-8 OLI 3/M/TIR [113,357] Table 5. Cont.…”