Bulletin of Engineering Geology and the Environment最新文献

The disintegration of expansive stiff clay will cause irreversible damage and deterioration of mechanical properties of the soil. The latest studies show that the disintegration is related to the swelling capacity of soil. In this study, a series of hydration disintegration tests and swelling pressure tests were performed on compacted Nanning expansive stiff clay samples with different initial water contents and dry densities. The observed disintegration process of all samples could be divided into initial, rapid and residual disintegration stages, among which the rapid stage dominated the whole process. By introducing relevant indicators to quantify the disintegration process, it was found that at a given dry density, the average disintegration rate of the sample decreased with increasing initial water content; while at a given water content, it decreased with increasing initial dry density. Such phenomena coincided well with the obtained evolution of swelling pressure at different initial water contents and dry densities. Based on these findings, the expansion-disintegration interaction mechanism of expansive stiff clay was finally analyzed from the perspectives of microstructure and hydration cracking. The initial conditions of the compacted samples determine the volume of inter-aggregates pores and thus the water transfer rate in soils, which affects the formation of hydration cracks. The cracking is induced by tension failure due to the expansion gradient formed during the hydration of sample, destructing the soil integrity to facilitate the disintegration. The disintegration, in turn provides preferential water infiltration channels to accelerate further soil expansion and hydration cracking. Such interactions proceeded until the completion of sample disintegration.

Two disastrous earthquakes, named Pazarcık (M_w7.8) and Ekinözü (M_w7.6), occurred on February 6, 2023 in the southeast part of Türkiye and were collectively named “Kahramanmaraş earthquakes”. These seismic events were caused by a left lateral strike-slip faults, and resulted in significant loss of life, severe damage to infrastructures and buildings, and geotechnical damages such as mainly large-scale slope failures, rockfalls, and ground liquefaction. The main goal of this study is to assess the extend and impact of widespread ground liquefaction, particularly on built environment. Additionally, the ranges of amount of settlement and tilting of buildings due to ground liquefaction were briefly discussed and liquefaction caused by Kahramanmaraş earthquakes were compared with those others occurred in Türkiye. The site observations indicated that except a village, a short section of a highway, a few bridges and two settlements, widespread liquefaction was mainly observed in agricultural non-urbanized fields. The maximum amount of settlement at some liquefaction locations reached up to 2 m and high-raise buildings tilted 7–8° from the vertical reaching up about 20°. Observations indicated that single-storey and two-storeys buildings with a basement to a certain depth, a lower center of gravity and raft foundation should be considered suitable on soils susceptible to liquefaction in earthquake-prone regions without taking any counter-measures against ground liquefaction. Mass movements along the shoreline of the Gölbaşı Lake were unlikely to be caused by lateral spreading resulting from ground liquefaction and they were rather due to planar sliding along a weak layer dipping towards the lake with progressive failure.

To address the reduced bearing capacity of rubber-reinforced calcareous sand while also alleviating pollution from discarded tires, a combined reinforcement method using rubber and geogrid is studied. This approach utilizes the geogrid’s capability to enhance material bearing capacity through the netting effect, complementing the rubber’s reinforcement of calcareous sand. By cyclic direct shear tests, this study investigates the dynamic shear characteristics and particle fragmentation mechanism of calcareous sand reinforced with rubber and geogrid. The analysis focuses on the shear characteristics of reinforced calcareous sand and the corresponding patterns of particle crushing. The findings indicate that: (I) Geogrid reinforcement effectively restrains volume deformation in the soil induced by rubber particles. (II) Reinforcement with rubber and geogrid compensates to a certain extent for the reduction in shear stiffness attributed to the rubber mixture. (III) The relative crushing rate of unreinforced calcareous sand post-test is 25.9%, which increases to 29.8% with geogrid reinforcement, decreases to 19.1% with rubber reinforcement, and reaches 22.6% with combined rubber and geogrid reinforcement. The geogrid’s particle crushing rate disadvantage is offset by the advantage provided by rubber reinforcement. The collaboration of rubber and geogrid reinforcement effectively addresses the limitations of rubber sand as a foundation material, rendering it more suitable for practical applications.

The safety of several large-scale reservoirs all over the world has been of concern due to dam foundation stress (DFS) that gradually changes following impoundment inducing the river valley contraction (RVC). Presently, there are limited approaches for the prediction of DFS and RVC based on complex hydro-geomechanics principles. However, these approaches require extensive information that is cumbersome and time-consuming to gather and hence expensive. In this paper, five machine learning models (MLMs) for DFS and RVC prediction were established by merging innovative analytical, BP neural networks and optimized algorithm approaches. Three key influencing factors; namely: seepage, temperature, and creep are used as input information in these models. The developed MLMs were validated using well-documented case study results over nine years for Xiluodu reservoir in China. The trend-fitting effect and statistical indicators of the proposed MLMs demonstrated strong predictive ability (R² > 0.9). Among the MLMs, Generic algorithm-BP and Sparrow search algorithm-BP methods were found to be comprehensive. The predicted RVC and DFS using MLMs are consistent with the coupled multi-field analytical method from the literature and provide reliable predictions using limited information. This study serves as a valuable reference for predicting DFS and RVC of large reservoirs for ensuring long-term safety.

The reactivation events of old landslides in the Three Gorges Reservoir area occur frequently, making it imperative to study the water softening characteristics and reactivation mechanism. An old clay landslide was selected as the focus of the research, and a segmented water injection permeable sliding surface was designed to simulate the formation and evolution of the old sliding zone during the process of groundwater rise. Volumetric water content sensors, pore water pressure gauges, high-speed camera devices, and Geopiv-RG digital image processing technology were used to obtain data on multiple physical fields. The analysis results indicated that the decrease in shear strength of the sliding zone soil and the sudden increase in pore water pressure on the sliding surface were important factors in the reactivation of old landslides. The surface deformation exhibited prominent zoning characteristics, primarily categorized into zones of strong deformation, weak deformation, and traction deformation. The failure mechanism involved shear sliding at the front edge, tensile cracking and failure at the trailing edge, and shear creep in the middle section. The development of multi-stage secondary sliding zones in old landslides can be categorized into three types: parallel to the original old sliding zone, partially overlapping with the original sliding zone to form a layered landslide, and completely overlapping with the original sliding zone, indicating overall reactivated deformation.

Occurrence of loess landslide has been more frequent due to the drastic global climate change, rapid expansion of human disturbances and continuous intensification of engineering activities. The activation and evolution mechanisms of the loess landslides under the rainfall are yet to be studied. In this paper, with reference to the Yangpoyao slope with seepage fissures under rainfall, an adjustable-angle landslide model test system is developed, integrating the rainfall simulation system, the measurement system and the data acquisition system, and the deformation development of the model, the rainfall infiltration, the change of water content and the destructive process of the model are monitored by the monitoring technology of multi-means and multi-methods throughout the course of the disaster. A distributed fibre-optic sensor system with the characteristics of continuity and high precision is used to monitor the temperature and strain within the slope model. The deformation evolution mechanism of fissured loess slopes under rainfall was elucidated through the observation of experimental phenomena and the analysis of the internal strain values of the soil, as measured by fibre optic sensors. The experimental results show that the collapse process of loess slopes can be categorised into three types, i.e. sinkhole collapse, block collapse and gully collapse, and that the deformation and damage patterns of the loess landslide model are mainly caused by shallow soil movement induced by erosion. Through the comparative analysis of the model test and the photographs of the field investigation, it is further demonstrated that the damage pattern shown in the physical model test is basically consistent with the slope condition of the real Yangpoyao slope, which provides a new theoretical reference for natural disaster prediction and management of loess slopes and landslides.

In order to explore the influence of water glass on the strength and microscopic mechanism of loess, water glass with different content and Baume degree was used to solidify loess. Unconfined compressive strenght test, particle analysis tests, mercury intrusion tests and SEM tests were carried out to qualitatively and quantitatively analyze the strength characteristics and microstructure changes of loess and water glass solidified loess under different conditions. The results showed that the unconfined compressive strength of water glass solidified loess increased with the increase of the water glass content, the Baume degree and the curing period. After 28 days of curing, the unconfined compressive strength reached the maximum value when the Baume degree of water glass was 52 Be’ and the content was about 20%. The incorporation of water glass gradually increased the cements attached to the loess skeleton particles and their surfaces. The particle diameter increased and the macropores were filled with cements. The macropores in the soil gradually decreased and the integrity was enhanced. The reason for the increase in the strength of water glass solidified loess was that the cementing material generated by the hydration reaction of water glass enhanced the degree of cementation between soil particles and increased the mutual friction between particles, which made the soil structure more stable and improved the soil strength. The research results provide a reference for the selection of relevant parameters in the design of improved loess engineering.

The unfrozen water content is crucial to soil's physical and mechanical properties. Soils on the Qinghai-Tibet Plateau are frequently subjected to freeze-thaw (F-T) cycles. The quantitative relationship between F-T effects and the unfrozen water content of soils requires further investigation. This study employs a nuclear magnetic resonance (NMR) scanner with a temperature-control module to measure the unfrozen water content of silty clay during multiple F-T cycles. The soil freezing characteristic curves (SFCC) of silty clay are derived from the T₂ (transverse relaxation time) distribution curves based on NMR measurements. Two distinct T₂ cutoff values are used to classify three types of water in soils: bound water, capillary water, and bulk water. The impact of F-T cycles on the evolution of unfrozen water content as temperatures decrease has been analyzed. The testing results indicate that the SFCC of silty clay can be segmented into three stages: super-cooling, fast-declining, and stable. As the number of F-T cycles increases, capillary water content decreases while bulk water content increases during the super-cooling stage. The damage coefficient, derived from pore volume measurements, increases sharply during the first four F-T cycles before stabilizing gradually. Additionally, there is a negative linear correlation between the damage coefficient and the initial capillary water content, and a positive linear correlation with the initial bulk water content. This study offers valuable insights for the quantitative analysis of unfrozen water content in seasonally frozen regions and serves as an essential guide for geotechnical construction projects in cold areas.

Accurately predicting the compressive strength of rock (RCS) is crucial for the construction and maintenance of rock engineering. However, RCS prediction based on single machine learning (ML) algorithms often face issues such as parameter sensitivity and inadequate generalization. To address these challenges, a new (RCS) prediction model based on a stacking ensemble learning method was proposed. This method combines multiple ML algorithms to achieve more accurate and stable prediction results. Firstly, 442 sets of rock mechanics experimental data were collected to form the prediction dataset, and data preprocessing techniques, including missing value imputation and normalization, were applied for data cleaning and standardization. Secondly, nine classic ML algorithms were used to establish RCS prediction models, and the optimal configurations were determined using k-fold cross-validation and Bayesian optimization. The selected base learners were LightGBM, Random Forest, and XGBoost, and the meta-learners were Ridge, Lasso, and Linear Regression. Finally, the models were verified using the testset, and the comparison showed that the proposed stacking models were better than all single models. Notably, the Stacking-LR model exhibited the best predictive accuracy(R²=0.946, MAE=5.59, MAPE=9.94%). Furthermore, the Shapley Additive exPlanations (SHAP) method was introduced to analyze the impact and dependencies of input features on the prediction results. It was found that both Young’s modulus and confining pressure are the most critical parameters influencing RCS and exert a positive impact on the prediction results. This finding is consistent with domain expert knowledge, enhances the model’s interpretability, and provides robust support for the predicted results.

With the construction and operation of a large number of hydropower engineering projects in China, the deformation and failure behaviors of reservoir bank slopes pose increasing threats to the stability of hydraulic structures, as well as upstream and downstream residents and infrastructures. This study aims to offer a novel perspective for exploring the long-term deformation behavior of near-dam reservoir bank slopes through statistical analysis, and to put forward a quantitative method to determine velocity thresholds for early warning procedures. In this paper, the monitoring data of velocities recorded from four reservoir bank slopes are counted at first, and then generalized Pareto distribution and power law distribution are applied to fit the empirical cumulative distribution of statistical data. Further, two risk measures, value-at-risk and conditional value-at-risk, are used to analyze the characteristics of the distribution tails and evaluate the risk levels corresponding to different velocities. Combined with risk measures, a quantitative method for defining threshold levels based on velocities is proposed for early warning procedures. Statistical results show that generalized Pareto distribution and power-law distribution have the potential as unified models to describe the deformation behavior of the reservoir bank slopes. Case analyses verify the effectiveness of the method for defining the early-warning thresholds. The proposed methods could easily be transferred to slope movements with similar deformation behaviors in other hazardous areas.