Current practice to model the occurrence of submarine landslides is based on methods that assess the potential of site-specific failures, all with the objective of providing elements to identify and quantify regional features associated to geohazards, before a project development takes place. Also, survey data to estimate parameters required to model submarine landslides show typically limited availability, mainly because of the cost associated to offshore surveying campaigns. In this paper, a probabilistic calibration approach is introduced using Bayesian statistical inference to maximize the use of available site investigation data, and to best estimate the occurrence of a marine landslide. For this purpose, a landslide model thought for its simplicity is used to illustrate the applicability and potential of the calibration methodology. The aim is to introduce a systematic approach to produce prior probability distributions of the model parameters, based on an actual integrated marine site investigation including geological, geophysical, and geomatics data, to then compare it with a posterior probability distribution of the same model parameters, but estimated after collecting in situ soil samples and testing them in the laboratory to produce the corresponding soil strength properties. This comparison allows to explore (a) the influence of the number of in situ samples, (b) the influence of a landslide factor of safety, and (c) the influence of the soil heterogeneity, into the likelihood of the occurrence of a marine landslide. The model parameters that are considered for calibration include the initial state of the submerged and saturated soil unit weight, the thickness of the soils' unit layers, the pseudo-static seismic coefficient, and the slope angle, while the soil undrained shear strength is considered as the reference parameter to conduct the calibration (i.e., to compare model predictions vs. actual observations). Results show the potential of the proposed methodology to produce landslide geohazard maps, which are needed for the planning and design of marine infrastructure.

Geohazards such as slope failures and retaining wall collapses have been observed during thawing season, typically in early spring. These geohazards are often attributed to changes in the engineering properties of soil through changes in soil phase with moisture condition. This study investigates the impact of freezing and thawing on soil stiffness by addressing shear wave velocity (Vs) and compressional wave velocity (Vp). An experimental testing program with a temperature control system for freezing and thawing was prepared, and a series of bender and piezo disk element tests were conducted. The changes in Vs and Vp were evaluated across different phases: unfrozen to frozen; frozen to thawed; and unfrozen to thawed. Results indicated different patterns of changes in Vs and Vp during these transitions. Vs showed an 8% to 19% decrease for fully saturated soil after thawing, suggesting higher vulnerability to shear failure-related geohazards in thawing condition. Vp showed no notable change after thawing compared to initial unfrozen condition. Based on the test results in this study, correlation models for Vs and Vp with changes in soil phase of unfrozen, frozen, and thawed conditions were established. From computed tomography (CT) image analysis, it was shown that the decrease in Vs was attributed to changes in bulk volume and microscopic soil structure.

The current study focuses on the long term strength reduction in lime stabilised Cochin marine clays with sulphate content. By introducing 6% lime and 4% sulphates to untreated Cochin marine clay, the research aims to investigate the effect of sulphates in these clays. Unconfined compression tests were conducted on lime treated clay both with and without additives, immediately after preparation and over 1 week, 1 month, 3 months, 6 months, 1 year and 2 years of curing. Test results indicated that both sodium sulphate and lithium sulphate has a negative impact on the strength gain of lime stabilised clay. To address this issue, Barium hydroxide, in both its pure laboratory form and the commercial product known as baryta, was incorporated into the lime stabilised soil. The study showed a consistent increase in shear strength with the addition of both barium hydroxide and baryta. When twice the predetermined quantity of baryta was added to lime stabilised clay, it outperformed pure barium hydroxide in terms of strength enhancement. Results of SEM and XRD analysis align with the strength characteristics. The cost-effective use of baryta offers a practical solution to counteract strength loss in lime stabilised, sulphate bearing Cochin marine clays.

The investigation of river levees holds significant implications for mitigating flood damage. Sand boiling, backward erosion piping, and phenomena manifesting along the riverside of levees directly imperil the integrity of these structures. It is imperative to address these phenomena comprehensively to safeguard both lives and property amidst flood events. The principal aim of this research is to delineate the variances in geotechnical conditions between sand boils observed at slope toes on the landside and those occurring at a distance from this region along the levee. Therefore, this study conducted extensive boring investigations at sites where sand boils occurred. The soil samples sampled from the boring investigations were analysed for grain size. The results of a series of geotechnical investigations showed that in the cases where sand boils occurred near the toe of the slope, a series of sandy soils with grain size characteristics similar to those of the sand boils were deposited in the foundation of the levee. On the other hand, in the case where the sand boil occurred far from the toe of the slope, sandy soil with grain size characteristics similar to that of the sand boil was deposited only on the landside.

Ancient landslides with platform geomorphology occasionally reactivate, posing serious geohazards. On September 9, 2021, persistent heavy rainfall triggered the reactivation of the Dahekou ancient landslide within a gently sloping geomorph0logy at the core of Zhangjiantan syncline in China's western Qinling-Daba Mountains. This event caused one death, damaged 80 houses, and blocked the Yushui River. This study reconstructs the sliding process of the Dahekou landslide and deciphers the complex landslide initiation mechanisms through field surveys, unmanned aerial vehicle (UAV) imagery analysis, drilling, electrical resistivity tomography (ERT) and small baseline subset-interferometric synthetic aperture radar (SBAS-InSAR) monitoring. We divide the sliding process of the Dahekou landslide into three stages. Two new landslides (#1 and #2) occurred at 18:30 on September 9, 2021. Subsequently, the ancient landslide (#3) slid in the 230 degrees direction at approximately 20:30 on September 9, 2021, then changed the direction to 170 degrees-240 degrees at 22:30 on the same day, and moved in the direction of 300 degrees at 10:00 the next day. Finally, the reactivated ancient landslide (#3) formed two partially sliding masses, with volumes of approximately 158x10(4) m(3) and 160x10(4) m(3), along the directions of 170 degrees-240 degrees and 300 degrees, respectively, damaging 80 houses and blocking the Yushui River. Field surveys suggest that new landslides #1 and #2 are rock landslides and soil landslides, respectively, with volumes of approximately 230x10(4) m(3) and 7.49x10(4) m(3). Compared with the InSAR data, the new landslide #1 thrust the ancient landslide #3, with an uplift velocity rate of 22.68 mm/a at the rear edge, from September 2020-September 2021. An analysis of drill hole data reveals that the bedding in the landslide area has complex geological conditions, comprising mudstone prone to slipping with different degrees of weathering. Notably, the core of the Zhangjiatan syncline sits on the sliding bedding of the ancient landslide, contributing to a change in the sliding direction. This comprehensive study reveals that the landslide #1 loading and thrusting, the persistent and heavy rainfall, and the complex geological conditions influenced the reactivated ancient landslide. Considering the intricacies of landslide failure mechanisms, we advocate for giving more attention in the future to the zone of potentially slip-prone strata located at the edge of ancient landslides.

Different slope geohazards have different causal mechanisms. This study aims to propose a method to investigate the decision-making mechanisms for the susceptibility of different slope geohazards. The study includes a geospatial dataset consisting of 1203 historical slope geohazard units, including slope creeps, shallow slides, rockfalls and debris flows, and 584 non-geohazard units, and 22 initial condition factors. Following a 7:3 ratio, the data were randomly divided into a test set and a training set, and an ensemble SMOTE-RF-SHAP model was constructed. The performance and generalization ability of the model were evaluated by confusion matrix and the receiver operating characteristic (ROC) for the four types of geohazards. The decision-making mechanism of different geohazards was then identified and investigated using the Shapley additive explanations (SHAP) model. The results show that the hybrid optimization improves the overall accuracy of the model from 0.486 to 0.831, with significant improvements in the prediction accuracy for all four types of slope geohazards, as well as reductions in misclassification and omission rates. Furthermore, this study reveals that the main influencing factors and spatiotemporal distribution of different slope geohazards exhibit high similarity, while the impacts of individual factors and different factor values on different slope geohazards demonstrate significant differences. For example, prolonged continuous rainfall can erode rock masses and lead to slope creep, increased rainfall may trigger shallow mountain landslides, and sudden surface runoff can even cause debris flows. These findings have important practical implications for slope geohazards risk management. (c) 2025 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).

Ensuring safety in geotechnical engineering has consistently posed challenges due to the inherent variability of soil. In the case of slope stability problems, performing on-site tests is both costly and time-intensive due to the need for sophisticated equipment (to acquire and move) and logistics. Consequently, the analysis of simulation models based on soft computing proves to be a practical and invaluable alternative. In this research work, learning abilities of the Class Noise Two (CN2), Stochastic Gradient Descent (SGD), Group Method of Data Handling (GMDH) and artificial neural network (ANN) have been investigated in the prediction of the factor of safety (FOS) of slopes. This has been successfully done through literature search, data curation and data sorting. A total of three hundred and forty-nine (349) data entries on the FOS of slopes were collected from literature and sorted to remove odd values and unlogic results, which had been used together in a previous research work. After the sorting process, the remainder of the realistic data entries was 296. The previous work which had included unrealistic data entries had unit weight, gamma (kN/m3), cohesion, C (kPa),angle of internal friction (Phi degrees), slope angle (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\Theta}$$\end{document}degrees), slope height H (m), and pore water pressure ratio, ru as the studied parameters, which formed the independent variables. After careful checks, the initial results showed poor correlation with the individual factors and the factors were collected into three non-dimensional parameters based on the understanding of the physics of flows, which are: C/gamma.h-Cohesion/unit weight x slope height, tan(phi)/tan(beta)-the tangent of internal friction angle/Tangent of slope angle, and rho/gamma.h-Water pressure/unit weight x slope height, which are deployed as inputs and FOS-the safety factor of the slope as the output. At the end of the exercise, the ANN outclassed the other techniques with SSE of 62%, MAE of 0.27, MSE of 0.21, RMSE of 0.46, average total error of 24%, and R2 of 0.946 thereby becoming the decisive intelligent model in this exercise. However, there is an advantage the deployment of GMDH, which comes second in order of superiority, has over the ANN. This is the development of a closed-form equation that allows its model to be applied manually in the design of slope stability problems. Overall, the present research models outperformed the eleven (11) models of the previous work due to sorting and elimination of unrealistic data entries deposited in the literature, the application of dimensionless combination of the studied slope stability parameters and the superiority of the selected machine learning techniques.

Hydrologically-induced landslides are ubiquitous natural hazards in the Himalayas, posing severe threat to human life and infrastructure. Yet, landslide assessment in the Himalayas is extremely challenging partly due to complex and drastically changing climate conditions. Here we establish a mechanistic hydromechanical landslide modeling framework that incorporates the impacts of key water fluxes and stocks on landslide triggering and risk evolution in mountain systems, accounting for potential climate change conditions for the period 1991-2100. In the drainage basin of the largest river in the northern Himalayas- the Yarlung Zangbo River Basin (YZRB), we estimate that rainfall, glacier/snow melt and permafrost thaw contribute similar to 38.4%, 28.8%, and 32.8% to landslides, respectively, for the period 1991-2019. Future climate change will likely exacerbate landslide triggering primarily due to increasing rainfall, whereas the contribution of glacier/snow melt decreases owing to deglaciation and snow cover loss. The total Gross Domestic Productivity risk is projected to increase continuously throughout the 21st century, while the risk to population shows a general declining trend. The results yield novel insights into the climatic controls on landslide evolution and provide useful guidance for disaster risk management and resilience building under future climate change in the Himalayas.

Forecasting landslide deformation is challenging due to influence of various internal and external factors on the occurrence of systemic and localized heterogeneities. Despite the potential to improve landslide predictability, deep learning has yet to be sufficiently explored for complex deformation patterns associated with landslides and is inherently opaque. Herein, we developed a holistic landslide deformation forecasting method that considers spatiotemporal correlations of landslide deformation by integrating domain knowledge into interpretable deep learning. By spatially capturing the interconnections between multiple deformations from different observation points, our method contributes to the understanding and forecasting of landslide systematic behavior. By integrating specific domain knowledge relevant to each observation point and merging internal properties with external variables, the local heterogeneity is considered in our method, identifying deformation temporal patterns in different landslide zones. Case studies involving reservoir-induced landslides and creeping landslides demonstrated that our approach (1) enhances the accuracy of landslide deformation forecasting, (2) identifies significant contributing factors and their influence on spatiotemporal deformation characteristics, and (3) demonstrates how identifying these factors and patterns facilitates landslide forecasting. Our research offers a promising and pragmatic pathway toward a deeper understanding and forecasting of complex landslide behaviors. (c) 2025 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/).

Soil-rubber mixtures have been proposed as cost-effective seismic and dynamic risk mitigation techniques. The granulated rubber used for these mixtures is obtained from end-of-life tires, allowing for stockpiles of waste rubber tires to be recycled. To date, most of the research has focused on the mechanical properties of sand-rubber mixtures, while limited studies have been performed on gravel-rubber mixtures (GRMs). In particular, GRMs with well-graded gravel (wgGRMs), which are of significant practical interest due to their availability, have only been poorly characterised. As part of a wider investigation aimed at facilitating the use of wgGRMs as geotechnical dynamic isolation systems, this paper presents bender element and small-strain cyclic triaxial test results performed on mixtures with 25%, 40%, and 55% volumetric rubber content. It is found that, thanks to their excellent energy absorption properties, wgGRMs can be efficiently adopted as geotechnical dynamic isolation to mitigate seismic risk of and anthropically induced vibrations on existing and new structures/infrastructures. Their easy implementation, low-cost, and widespread availability further facilitate their use.