Landslides triggered in mountainous areas can have catastrophic consequences, threaten human life, and cause billions of dollars in economic losses. Hence, it is imperative to map the areas susceptible to landslides to minimize their risk. Around Abbottabad, a large city in northern Pakistan, a large number of landslides can be found. This study aimed to map the landslide susceptibility over these regions in Pakistan by using three Machine Learning (ML) techniques, specifically Linear Regression (LiR), Logistic Regression (LoR), and Support Vector Machine (SVM). Several influencing factors were used to identify the potential landslide areas, including elevation, slope degree, slope aspect, general curvature, plan curvature, profile curvature, landcover classification system, Normalized Difference Water Index (NDWI), Normalized Difference Vegetation Index (NDVI), soil, lithology, fault density, topographic roughness index, and road density. The weights of these factors were calculated using ML techniques. The weightage overlay tool is adopted to map the final output. According to three ML models, lithology, NDWI, slope, and LCCS significantly impact landslide occurrence. The area under the ROC curve (AUC) is applied to validate the performance of models, and the results show the AUC value of LiR (88%) is better than SVM (86%) and LoR (85%) models. ML models and final susceptibility map gives good accuracy, which can be reliable for the results. The study’s outcome provides baselines for policymakers to propose adequate protection and mitigation measures against the landslides in the region, and any other researcher can adopt this methodology to map the landslide susceptibility in another area having similar characteristics.

Landslides (LS) represent geomorphological processes that can induce changes over time in the physical, hydrogeological, and mechanical properties of the involved materials. For geohazard assessment, the variations of these properties might be detected by a wide range of non-intrusive techniques, which can sometimes be confusing due to their significant variation in accuracy, suitability, coverage area, logistics, timescale, cost, and integration potential; this paper reviews common geophysical methods (GM) categorized as Emitted Seismic and Ambient Noise based and proposes an integrated approach between them for improving landslide studies; this level of integration (among themselves) is an important step ahead of integrating geophysical data with remote sensing data. The aforementioned GMs help to construct a framework based on physical properties that may be linked with site characterization (e.g., a landslide and its subsurface channel geometry, recharge pathways, rock fragments, mass flow rate, etc.) and dynamics (e.g., quantification of the rheology, saturation, fracture process, toe erosion, mass flow rate, deformation marks and spatiotemporally dependent geogenic pore-water pressure feedback through a joint analysis of geophysical time series, displacement and hydrometeorological measurements from the ground, air and space). A review of the use of unmanned aerial vehicles (UAV) based photogrammetry for the investigation of landslides was also conducted to highlight the latest advancement and discuss the synergy between UAV and geophysical in four possible broader areas: (i) survey planning, (ii) LS investigation, (iii) LS dynamics and (iv) presentation of results in GIS environment. Additionally, endogenous source mechanisms lead to the appearance of deformation marks on the surface and provide ground for the integrated use of UAV and geophysical monitoring for landslide early warning systems. Further development in this area requires UAVs to adopt more multispectral and other advanced sensors where their data are integrated with the geophysical one as well as the climatic data to enable Artificial Intelligent based prediction of LS.