Failure of cutting slopes in transportation engineering pose great threat to the construction and use of such infrastructure. Based on site investigation, deep displacement monitoring, and numerical simulation analysis, the formation and failure mechanism of Tangjiashui landslide in Lin-Wu motorway, Hunan Province, China, were analyzed, aiming at providing references for slopes with similar conditions. The results indicated that poor engineering geological characteristics of the sliding body, inadequate survey and unreasonable construction sequence of the initial reinforcement were main triggering factors. The slope body composed of ancient landslide deposits was prone to form a potential sliding surface. The excavation at the toe of the slope triggered the landslide. During the initial landslide treatment, the anti-slide piles were insufficiently embedded, and the toe counterweight fill and the soil behind the piles were excavated in a single operation, causing the slope sliding again. The numerical simulation results indicated that the recommended prevention and control measures of a combination of piles and cables could effectively prevent the landslide from sliding again. Thus, during construction in sections with similar geological conditions, special attention should be paid to the slope structure, the properties of the geo-technical materials, and the sequence of protective structure construction. Copyright © 2025. Published by Elsevier Inc.

Cement-based vegetation substrate (CBVS) provides dual benefits for ecological restoration and slope engineering stabilization, yet its alkaline environment poses a major challenge to plant survival. This study investigates the evolution of pH in a vegetation-compatible cementitious system incorporating silica fume and a composite ecological regulator (CER). Through SEM and vegetation experiments, we reveal multi-component alkaline-regulation mechanisms and demonstrate bidirectional feedback between plant growth and substrate pH. Cement content was identified as the primary factor controlling initial alkalinity. The CER effectively reduced alkalinity via acidic ion neutralization, humate complexation, and microbially induced precipitation, achieving a maximum pH reduction of 11.2% at a 0.3% dosage. Concurrently, a 2% silica fume content consumed portlandite to form low-alkalinity C-S-H gels, decreasing pH by 5.2% while improving matrix strength. Notably, Tall fescue exhibited the strongest adaptability and further reduced the system pH by 15.7% after 100 days, confirming a significant plant-material feedback loop.

A model of a steep bedding rock slope is designed and produced, and large-scale shaking table tests are conducted to analyze the dynamic response and deformation failure mode of the steep bedding rock slope under earthquake action. The results show that the natural vibration frequency of the steep bedding rock slope model decreases gradually with increasing vibration number and that the damping ratio increases gradually with increasing vibration frequency. The horizontal acceleration amplification coefficient of the model slope shows an obvious elevation amplification effect and surface trend effect. There are obvious differences in the slope acceleration response under the action of different types of input seismic waves. The acoustic emission parameters increase nonlinearly with increasing input seismic wave amplitude, and the acoustic emission characteristics during the failure process of the steep bedding rock slope can be divided into two stages: a slow rise period and a sharp rise period. The deformation evolution process of the steep bedding rock slope under the action of seismic dynamics can be divided into three stages: the formation of a tensile crack; the downward expansion of the tensile crack; and the sudden shear of the locked segment and the sudden instability of the slope. According to the abrupt cusp catastrophe theory, a correlation instability criterion is established based on the test results; the critical acceleration amplitude of the slope dynamic instability is quantitatively determined to be 0.4 g.

Altered rock, often encountered in major engineering projects, can seriously affect the stability of rocks and slopes surrounding deeply buried tunnels. This study addressed the alteration mechanism, alteration degree classification, and mechanical parameters of altered rock in engineering project areas using field testing, thin slice identification, X-ray diffraction, and major element testing. Results showed that the altered rock types in the areas of the Pingjiang Pumped Storage Power Station, Hunan Province, and a diversion tunnel in northern Xinjiang Uygur Autonomous Region include biotite granodiorite, biotite monzogranite, and cataclastic granite, and that the main alteration mechanisms are chloritization of biotite and clayization of feldspar minerals. The altered rocks were classified as slightly, moderately, or strongly altered according to their apparent characteristics, rebound value, longitudinal wave velocity, metamorphic mineral content, and porosity. The bulk density, elastic modulus, cohesion, and internal friction angle (Poisson’s ratio) of the altered rocks decreased (increased) with increase in the degree of alteration. Numerical simulations showed that in altered rock slope areas, the zone of strong rock alteration and the moderate-strong alteration contact zone exhibit locally large deformations that represent a certain hazard to engineering projects. These results provide valuable guidance and support for major projects in altered rock areas. © China University of Geosciences (Wuhan) and Springer-Verlag GmbH Germany, Part of Springer Nature 2025

The landslide susceptibility assessment (LSA), which presents the probability of a landslide occurrence under site-specific geological and environmental conditions, is essential to prevent and mitigate the landslide risk. The machine learning is one of important techniques for the study on LSA, and various models about the machine learning differ from each other in the LSA. In this paper, the Meishan township, Zhejiang Province, China was selected as a case study, and the Pearson correlation assessment and multicollinearity test were conducted to measure the relationship among 12 factors such as elevation, slope, and slope aspect. The training models were separately established with the decision tree (DT), random forest (RF), gradient-boosting decision tree (GBDT), extreme gradient boosting (XGBoost), and convolutional neural network (CNN), and the results were assessed by means of the area under the curve (AUC) of receiver operating characteristic plot as well as the frequency ratio (FR). The results demonstrated that the CNN achieved the best performance of AUC=0.874, followed by XGB (AUC=0.806), GBDT (AUC=0.801), FR (AUC=0.794), and DT (AUC=0.735), indicating that the DT-based ensemble models predict better than the single DT models. Furthermore, the CNN model, due to its complex structures, performs better than conventional machine learning model in dealing with the nonlinear problems.

采用自制的循环加载系统，通过模型试验探究了水平循环荷载作用下螺旋锚桩基础的承载性能以及土体变形的各因素影响规律。结合粒子图像测速（PIV）技术分析桩周土体的变形场，揭示了不同影响因素作用下桩周土体的变形机理。结果表明：①循环荷载幅值和频率对循环荷载作用下螺旋锚桩的承载特性有显著影响，且相比原始静荷载作用情况，密砂中循环后施加静载作用下锚桩的承载性能均呈现出降低趋势。②滞回曲线由初始阶段的未闭合状态逐渐转变为闭合状态，同时整体响应模式表现出明显的正向偏移趋势。在相同循环周期下，滞回曲线面积随幅值的增大而增大，随频率的增大而减小。③根据水平累积位移规律提出了锚桩水平循环累积位移预测模型，且预测效果较好。④埋深比、密实度、循环荷载幅值、频率和周期对桩周土体变形有重要影响，桩前被动区位移场受幅值影响最为明显，桩后主动区砂土位移受密实度影响较大。砂土内部剪切场通常关于锚桩轴线平行对称，且剪切强度与循环荷载幅值和次数呈正相关，与循环荷载频率呈负相关。

Using residual soil from the Shanghecun landslide in the western Henan Province, shear creep tests of residual soil samples with different rock contents (RCs) were performed to explore the creep characteristics, creep rate, and long-term strength of the residual soil. The test results indicate that the residual soil samples with different RCs display typical creep characteristics. With increasing RC, both the instantaneous deformation and the total creep deformation of the residual soil gradually decrease. The shear strain of the residual soil increases gradually with increasing shear stress for the different RCs. With increasing time, the slope of the isochronous stress-strain curves of the residual soil samples with different RCs increases gradually. The Burgers model can simulate the rheological process of the residual soil samples with different RCs. The RC has a significant effect on the shear strength and the long-term strength of the residual soil. With increasing RC, the shear strength and the long-term strength of the residual soil gradually increase, with the long-term strength being approximately 39%-63% of the shear strength.

The route of the South-to-North Water Diversion channel strides across part of the coal mine goaf in Yuzhou County, Henan Province, China, and long-term deformation due to coal seam recovery poses a threat to the safe operation of the main canal. Therefore, the study of the deformation mechanisms induced by coal seam recovery is of great significance to the canal's safe operation, as well as to deformation monitoring and to the development of early warnings. The geologic model was established based on the geological engineering conditions of the Yuzhou Gongmao mining area, spanning the main canal of the South-to-North Water Diversion Project; then, the physical model test was carried out according to similar theories. The deformation characteristics of the rock overlay and the channel above the goaf were analyzed, and failure criteria for overburdened rock and the channel were proposed. The results showed that horizontal fissures were gradually observed in the overlying rock as the coal mining progressed, extending and widening. When the goaf was excavated to 76 cm, the overlying rock body suddenly collapsed as a whole, and the channel collapsed and was destroyed. During the formation of the goaf, there was a critical span ratio (R): When the height-to-span ratio was greater than 0.039, the collapse of overlying rock occurred only within a certain range above the goaf. When the height-to-span ratio was less than 0.039, the overlying rock body collapsed in a wide area, and the soil on both sides of the channel collapsed to the center of the channel, presenting a "V" glyph collapse. The sediment in the center of the channel measured 22 mm, and there were multiple tensile cracks on both sides of the embankment, with a width of 5-10 mm. The vertical deformation of the channel went through three stages, namely, the initial deformation stage, the deceleration deformation stage, and the stability stage. This study can provide scientific guidance for early warnings of channel deformation and safe operation across the goaf.

In this study, the evolution process of a landslide model under continuous rainfall conditions with a rainfall intensity of 30 mm/h is studied in depth based on an outdoor rainfall model test of a colluvial slope as the research material. The response law of pore water pressure and settlement amount is also obtained, and the influence of bedrock inclination angle on the development and deformation failure of the colluvial landslide is discussed. When the dip angle of the bedrock is 40°, it is prone to sudden slip-type landslides, and the evolution process is as follows: tensile cracks appear at the trailing edge, and these cracks continue to increase, leading to overall sliding. When the bedrock dip angle is 30°, traction landslides are prone to occur, and the evolution process is as follows: there is sliding at the foot of the slope, tensile cracks appear in the middle, sliding occurs in the middle, and tensile cracks appear in the upper part, leading to overall sliding. Before the landslide starts, the pore water pressure rises significantly. In the process of landslide evolution, the fine particles move to the foot of the slope with the rainwater, and the larger the angle of the slope, the greater the number of fine particles that accumulate at the foot of the slope, and the higher the elevation and the larger the scale of the trailing edge of the sliding body during sliding. © 2025 by the authors.