The stability of the compressed air energy storage (CAES) cavern is influenced by ground stresses during the construction phase and by cyclic high internal pressure (≥10 MPa) during the operational phase, resulting in different mechanical responses of the surrounding rock during the unloading and loading phases. This study investigates the stress paths of the CAES cavern from the perspectives of stress evolution and damage modes throughout both construction and operational phases. The findings indicate that during the excavation phase, active support technology significantly reduces deformation by adjusting the principal stress field of the surrounding rock, particularly in low-strength rock formations. Additionally, the load transfer effect decreases the burden on initial supports, such as steel arches. In the operational phase, the high internal pressure causes the sequence of principal rock stresses to shift, with radial stress increasing to the maximum principal stress and circumferential stress transitioning to the minimum principal stress. When exposed to hydrostatic pressure, characterized by low local stress and high rock strength, the surrounding rock at the tunnel wall may experience tensile failure during the high-pressure gas storage stage; otherwise, shear failure will occur at the tunnel wall. When the lateral pressure coefficient ( ) is less than 1, shear failure occurs in the surrounding rock at the side walls. In cases where the lateral pressure coefficient is particularly low, tensile failure may still happen at the arch crown. Conversely, when exceeds 1, shear failure is observed in the surrounding rock at the arch. If the lateral pressure coefficient is notably high, tensile failure may still occur at the side walls. As pressure increases, the area of tensile fractures may subsequently undergo shear failure. Further analysis reveals that conventional grouting and radial bolt systems significantly enhance stability only during the excavation phase. During operation, radial bolts not only lose their reinforcing capabilities but also exacerbate stress concentration in the surrounding rock under compression. In contrast, enhanced circumferential restraint effectively improves the stress state of the surrounding rock. For surrounding rock of good quality and high strength, grouting has limited efficacy in filling fractures. Conversely, grouting measures have been shown to increase the shear strength of formations characterized by low strength. Based on these insights, this study proposes an innovative solution for the stability control of CAES caverns subjected to complex stress paths by forming an anchor network system using crossed diagonal anchors and optimizing the inclination angle of the anchors to achieve a full-cycle tensile state.

Shale gas is an important unconventional natural gas resource in China. The Sichuan Basin serves as a major production area and is prone to frequent earthquakes, underscoring the urgent need for seismic suppression. This study aims to enhance seismic prevention and control in key production zones by first analyzing the relationship between shale gas development and seismicity. It then reviews the mechanisms of shale-gas-induced earthquakes, summarizes risk assessment and mitigation measures, and outlines future research directions. Statistical data indicate that large-scale shale gas development in the Sichuan Basin coincides with high-frequency, low-magnitude, shallow-depth seismicity, creating a seismic hotspot in southern Sichuan. Initially, seismic frequency increased before declining, while the average magnitude remained around＜4.2. During the early stages of development, numerous low-energy microseismic events released tectonic stress and elastic strain energy. Concurrently, soft rocks surrounding hydraulic fractures attenuated seismic waves, thereby reducing the likelihood of larger earthquakes. We also investigate potential links between staged horizontal-well fracturing and seismicity. Shale gas extraction influences stress and energy transfer within the reservoir, which can reactivate weak-plane faults and trigger earthquakes. The weak-fault-slip mechanism is described using a failure criterion that incorporates dynamic pore pressure evolution, stress perturbation, and fault strength weakening. The dense and active fault network of the Sichuan Basin renders induced seismic risk unavoidable. Its induction mechanisms are more complex than the “injection-driven” or “fracturing-driven” models observed in North America. Consequently, North American seismic control theories have limited applicability in this context. We propose the concept of “earthquakes without disasters” as the goal for safe shale gas extraction. Coordinated multi-well operations and superimposed multi-fracture perturbations for seismic suppression provide key pathways for optimizing safe and efficient development.

In the process of deep mining, the dynamic waves generated by mining operations and other activities are predominantly influenced by low-frequency disturbances during the propagation of surrounding rock. These disturbances can easily trigger dynamic disasters such as rock bursts and roof collapses, thereby compromising the stability of the roadway surrounding rock. To investigate the damage characteristics of low-frequency disturbances on creep rock masses, this study focuses on red sandstone and employs the RRTS-IV rock rheological disturbance effect test system to conduct rheological disturbance tests under low-frequency conditions. The results indicate that: (1) A disturbance-sensitive point exists in the rheological rock mass under low-frequency disturbance conditions. As the disturbance frequency increases, the intensity of this sensitivity point decreases, making the rock mass more susceptible to entering the disturbance-sensitive area. (2) Before the axial static load pressure reaches the strength of the disturbance-sensitive point, the axial strain changes gradually decrease with increasing axial pressure, demonstrating nonlinear elastic behavior. Once disturbance-sensitive point is exceeded, the rock mass exhibits heightened sensitivity to external disturbances, leading to intensified damage accumulation and a higher likelihood of instability failure. (3) Analyzing the T2 spectrum and spectral peak area, the ratio of the large pore peak area to the small pore peak area ( ) is utilized to characterize the damage degree of the rock mass under varying frequency conditions. It is observed that the damage degree of the rock mass is positively correlated with the disturbance frequency. (4) From the perspective of the peak area of the T2 spectrum, a microscopic damage evolution equation for the rock under disturbance frequencies of 0.2, 0.5, and 1.0 Hz is established, and the rationality and applicability of this equation are validated by employing the constitutive model of rock mechanics. The research findings provide theoretical support for predicting the failure process of surrounding rock and evaluating the long-term stability of roadway surrounding rock.

In situ CT testing was conducted on two typical types of granite from the Three Gorges region, specifically medium coarse-grained and medium fine-grained granite. Non-invasive measurement of the internal deformation field of rock samples was achieved using digital volume correlation (DVC) technology. This study examined the influence of the microstructural characteristics of granite on its macroscopic mechanical properties and failure modes. The results indicate that the difference in microstructure between medium coarse-grained and medium fine-grained granites is primarily manifested in the directional arrangement of mica. Mica minerals generally exhibit a flat ellipsoidal structure, and the mica particles in medium coarse-grained granites are flatter in shape compared to those in medium fine-grained granites, suggesting a more pronounced spatial directional arrangement of mica in the former. The fracture network of medium and fine-grained granite displays an isotropic distribution, with the fracture mode being stress-controlled. The orientation of the fracture surface in medium coarse-grained granite closely aligns with the orientation and arrangement of mica, demonstrating typical structural control failure. Areas exhibiting larger equivalent strain changes correlate well with the post-peak fracture morphology of the rocks, indicating that equivalent strain can effectively predict the actual fracture morphology of post-peak rocks. In medium coarse-grained granite, regions with equivalent strain changes exceeding 0.4% are identified as potential failure areas, with their distribution patterns aligning closely with the direction of mica arrangement, revealing the dominant role of the mesostructure of granite in its fracture process.

The real contact area of red sandstone, green sandstone, limestone, and granite was measured under various offset displacements using pressure-sensitive films. The structural surfaces of all specimens were standardized through high-precision sculpting technology to ensure morphological consistency. This study systematically investigated the effects of lithology, offset displacement, and normal stress on the contact area of the structural surfaces. The experimental results indicate that lithology significantly impacts the contact area, and this effect becomes more pronounced as offset displacement increases. Rocks with higher uniaxial compressive strength are more influenced by offset displacement, exhibiting a more pronounced reduction in contact area in response to offset displacement. A distinct exponential relationship was observed between offset displacement and structural surface contact area: as offset displacement increases, the rate of contact area reduction gradually diminishes. Furthermore, larger normal stress results in greater contact areas, with the influence of normal stress being more pronounced at lower offset displacements. The structural surfaces were divided into nine regions to analyze the characteristics of contact area distribution after offset displacement. The results demonstrate that, following offset displacement, the contact area tends to concentrate in regions with higher joint roughness coefficient (JRC), and this tendency becomes more evident as offset displacement increases. Analysis of the normal stress applied to the specimen surface and the actual stress within the specimen revealed that the actual stress is typically several times greater than the normal stress. During offset displacement, the actual stress can exceed the normal stress by hundreds of times in some instances. This study provides critical insights into the evolution of contact area under dynamic offset displacement in multi-lithology rocks.

Lithium slag, a novel type of solid waste with potential cementitious activity, can be efficiently utilized when co-processed with tailings to produce backfill. To investigate the mechanical properties and damage evolution behavior of lithium slag-cement-tailings cemented backfill under dynamic impact, backfill with varying ratios of lithium slag, cement, and tailings were prepared. The mechanical properties and damage evolution were analyzed using a separated Hopkinson pressure bar apparatus. A new modified damage constitutive model for lithium slag-cement-tailings cemented backfill was established based on the Weibull model, incorporating a correction coefficient (k). The results indicate that the incorporation of lithium slag enhances the dynamic compressive strength of the cemented backfill. Notably, as the lithium slag content increases, the dynamic compressive strength of the backfill initially rises, reaching a peak at 20% lithium slag content, before subsequently decreasing. The damage evolution of the lithium slag-cement-tailings cemented backfill can be categorized into three stages: the slow growth stage, the accelerated growth stage, and the rapid growth stage of damage. The primary role of lithium slag is to increase the strain during the slow growth stage. The newly established modified damage constitutive model significantly reduces the prediction error by over 50% compared to the previous model and accurately represents the stress-strain behavior of the lithium slag-cement-tailings cemented backfill during the failure stage. These findings provide a theoretical foundation for the application of lithium slag in filling mining operations.

The wetting deformation of rockfill materials exacerbates the uncoordinated deformation of rockfill dams during the water storage period and may even lead to cracks in the dam structure. Investigating its characteristics is crucial for controlling deformation in rockfill dams and ensuring their safety. Wetting deformation is not an instantaneous variable; rather, it exhibits a coupling relationship with creep. Therefore, studying the time effect of wetting deformation in rockfill materials is essential. While numerous experimental studies have explored the wetting deformation and development processes of rockfill materials at the macro level, systematic research on the variation of crushing strength in rockfill particles throughout the complete wetting process remains insufficient. In this study, indoor compression tests were conducted on rockfill particles from both soft and hard lithologies after different wetting durations to assess the influence of wetting time on particle breakage and strength characteristics. The results revealed that particles of varying sizes exhibited significant size and wetting time effects upon immersion in water. Specifically, the average breaking strength of particles within each size group decreased as particle size and wetting duration increased. The particle breaking strength, in both dry and various wetting states, follows a Weibull distribution, with the size effect diminishing as wetting time extends. Based on the Weibull size effect model, this paper introduces a deterioration coefficient and establishes a predictive model for the characteristic crushing strength of rockfill particles, incorporating both wetting time and size effect parameters. This model provides a theoretical reference for further research on the wetting deformation of rockfill materials.

On February 8, 2025, a significant landslide occurred in Jinping Village, Mu′ai Town, Junlian County, Yibin City, Sichuan Province, under winter conditions without heavy rainfall or seismic activity. This disaster resulted in 10 fatalities and 19 missing persons, igniting widespread public discourse regarding the formation mechanisms of this landslide and the potential for winter to be a high-incidence period for landslides in the Wumeng Mountainous Region. In response, a multi-source data fusion analysis method was employed, integrating field geological surveys, historical disaster investigations, airborne LiDAR terrain interpretation, time series InSAR surface deformation monitoring, and discrete element numerical simulation techniques. The study comprehensively examined the landslide′s causes from various aspects, including terrain and geomorphological conditions, geological and lithological characteristics, rock structure features, historical disaster context, rainfall, earthquakes, and the “chimney effect”. Furthermore, the post-instability movement process of the landslide was simulated. Through comparative analysis of three typical landslide cases—Jinping Village in Junlian County, Liangshui Village in Zhenxiong County, and Zhaojiagou in Zhenxiong County—the research discussed the mechanisms of winter landslides induced by the “butterfly swarm effect” in the Wumeng Mountainous Region and proposed preventive recommendations. The findings indicate that continuous rainfall is the direct trigger of the Jinping Village landslide, with the landslide process categorized into five distinct stages: initiation, shearing and falling, landing and scraping, oblique throwing, and scraping and accumulation. Continuous rainfall or rain-snow events are critical triggering factors for winter landslides in the Wumeng Mountainous Region. Under the synergistic effects of steep terrain, landslide-prone strata, and unfavorable rock mass structures, a“butterfly swarm effect”emerged, where multiple factors interacted through nonlinear superposition and cascading reactions, ultimately leading to macro-events (landslide disasters) that far exceeded the impacts of individual factors (precipitation). This study serves as a significant reference for further investigation into the Jinping Village landslide in Junlian County and winter landslides in the Wumeng Mountainous Region.

Salt precipitation in fractured rock is a significant area of research within underground energy engineering. The injection of CO2 into deep saline aquifers leads to the evaporation of formation water, which causes salt crystals to precipitate in fractures, thereby significantly reducing storage efficiency. To elucidate the impact of this process on CO2 injection, this study integrates laboratory visualization with numerical simulation. A self-developed visualization setup was utilized, employing rough glass to construct transparent fracture models and simulating salt precipitation induced by gas injection into brine-filled fractures. Various injection rates (Q = 10–1 000 mL/min) were tested to evaluate spatial precipitation patterns. At low flow rates, localized ex situ precipitation was observed, while higher flow rates resulted in homogeneous precipitation. Based on these observations, the mechanisms governing the evolution of fracture permeability for each precipitation pattern were identified, and a power-law model for permeability changes was developed. This permeability evolution model was incorporated into the TOUGH3 simulator and applied to the Illinois Basin CO2 storage project. The classic tubes-in-series model was found to overestimate clogging risk, whereas the power-law evolution model demonstrated a closer alignment with field measurements. These findings establish a direct relationship between laboratory-scale precipitation patterns and field-scale spatial variability, providing scientific support for optimizing CO2 injection conditions. This work systematically reveals the impact mechanism of salt precipitation on reservoir permeability from three aspects: visualization experiments, permeability evolution models, and numerical simulations, to provide scientific support for the optimization of CO2 geological sequestration injection conditions.

The locked segment is a critical seismogenic tectonic unit controlling the transition of a fault from stable slipping to unstable rupture. To investigate precursory characteristics of fault instability (e.g., stress disturbance, rupture damage near the locked segment), a 2 m × 1 m × 0.6 m locked strike-slip fault model was fabricated using river sand, cement, barite powder, and gypsum. Embedded strain cubes and acoustic emission (AE) technology were used to monitor the locked segment?s local strain and model rupture nucleation process under horizontal uniaxial loading. The study focused on local principal stress deflection near the locked segment, temporal variations of AE events, and accelerated strain release phenomenon before model instability. The results indicate that the instability process of the model can be divided into a stable, sub-stable, sub-unstable and unstable stages, based on the stress state. Abrupt strain rate increase and abnormal high value of the loading-unloading response ratio (LURR) at the end of the sub-unstable stage predict fault instability. Shear stress at locked segment’s ends undergoes an abrupt change before model instability. The surge in AE events acts as an early precursor for the fault entering the sub-unstable stage, and the b-value maintains a low level until fault instability manifests. Cumulative Benioff strain (CBS) release of AE events adheres to the accelerating strain release (ASR) model, suggesting that the locked strike-slip fault model features sudden instability. The strong negative correlation between the deflection angle in dilatation stress quadrants and CBS suggests sensitive decay of this deflection angle may predict the fault system instability. The above results provide experimental evidence for understanding the local stress evolution, strain release, and activity characteristics of locked strike-slip faults.

The diffusion of quick-setting grout in high-temperature fissured rock mass involves the process of grout filling rock fissures, accompanied by the fluid-solid phase transition of the grout and heat transfer between the grout and the surrounding rock. Based on this understanding, a breakthrough diffusion model for grouting has been developed. A modified calculation method for the rheological parameters of quick-setting grout, considering the combined effects of heat transfer and fluid-solid phase transition, has been proposed. Additionally, stepwise algorithms for the initial circular diffusion stage and the breakthrough diffusion stage have been formulated. To validate the reliability of the breakthrough diffusion model, simulation tests were conducted on the grouting diffusion process within a single horizontal fissure under high-temperature and water-saturated conditions. The analysis of the breakthrough diffusion mechanism of quick-setting grout in high-temperature rock fissures was performed, focusing on the spatial-temporal distributions of grout temperature, rheological parameters, and grout pressure, as well as the impact of surrounding rock temperature on the fissure grouting diffusion process. Results indicate that the stage characteristics observed in the grouting simulation tests were consistent with theoretical predictions. The experimental end times for the circular and breakthrough diffusion stages deviated from theoretical values by less than 15%, while the theoretical errors in grout pressure and temperature remained within 10%, thus confirming the reliability of the breakthrough diffusion model. Theoretical findings revealed that as the temperature of the surrounding rock increased from 20 ℃ to 80 ℃, the end time of the fluid-solid phase transition of the grout was reduced by over 50%. Furthermore, the peak yield stress of the grout decreased by more than 20%, and the peak viscosity diminished by over 50%. During the circular diffusion stage and each breakthrough diffusion stage, the spatial distributions of grout temperature, rheological parameters, and grout pressure along the flow path exhibited periodic characteristics. Specifically, the grout temperature, rheological parameters, and grout pressure corresponding to each diffusion stage displayed approximately linear, downward-convex, and upward-convex spatial distributions, respectively. The length of the flow path demonstrated a positive correlation with the surrounding rock temperature. However, the relationship between grout pressure and temperature exhibited a certain degree of complexity, as an increase in the surrounding rock temperature did not result in a significant increase in grout pressure.

Existing static sealing evaluation methods exhibit limited applicability to heterogeneous shale formations. To enhance the representativeness and objectivity of evaluation results for the static sealing of CO2 geological storage reservoirs, this study proposes and implements innovative approaches: introducing a three-dimensional heterogeneous model into the traditional evaluation framework; and combining the analytic hierarchy process (AHP) with the entropy weight method (EWM) to establish a comprehensive weighting strategy that minimizes the impact of subjective bias. The research findings indicate that: (1) the evaluation results based on the heterogeneous model (5.84–9.95 points) demonstrate that the sealing performance in most areas of the reservoir meets storage requirements, although there exists a localized leakage risk, effectively addressing the issues of inadequate characterization and overly optimistic outcomes associated with homogeneous models; and (2) the average consistency of the weight coefficients for the primary influencing factors, calculated using AHP and EWM, reaches 83.3%, confirming the effectiveness of the comprehensive weighting strategy in mitigating subjective interference. This method significantly enhances the credibility of static sealing evaluations, providing an efficient and theoretically robust evaluation framework for CO2 geological storage projects.

The frictional slip behavior of faults is influenced by various factors, with the stiffness of the fault and the surrounding rock mass playing a pivotal role in determining the mechanical behavior patterns of fault slip. In laboratory experiments, the controlled simulation of fault frictional behavior through the adjustment of shear loading stiffness is an essential approach to investigate the associated mechanical characteristics and underlying mechanisms. However, effective methods for regulating shear loading stiffness during experiments are currently lacking. To address this challenge, the study systematically analyzes the variation of shear loading stiffness in experimental apparatus under various external conditions employing both direct and indirect testing methods. A regulation method is proposed, which involves connecting elastic blocks in series with the apparatus to adjust its shear loading stiffness. The experimental investigation of fault frictional behavior under varying shear loading stiffness conditions yields the following findings: (1) the shear loading stiffness of the apparatus is not constant; it generally increases with rising shear force and exhibits significant dependence on the apparatus deformation rate—it is weakly dependent during the initial and final loading stages but strongly dependent during the intermediate stage; (2) Materials PA, POM, and ABS demonstrate favorable linear elastic properties within a certain loading range and can be used as elastic blocks in series to regulate the apparatus shear loading stiffness; (3) as shear loading stiffness decreases, the stick-slip period, unstable slip displacement increment, unstable slip velocity, shear force drop, and fault stiffness all show increasing trends, while the viscous slip velocity gradually decreases; (4) the sensitivity of different stick-slip parameters to changes in shear loading stiffness varies—specifically, the stick-slip period, unstable slip displacement increment, and unstable slip velocity exhibit higher sensitivity, whereas the shear force drop, fault stiffness, and viscous slip velocity are relatively less sensitive. These findings offer important insights into the mechanical behavior of fault friction and the variation characteristics of seismic source parameters.

The significant swelling characteristics of red-bed mudstone often lead to uplift deformation in high-speed railway subgrades, highlighting the urgent need for in-depth research into the coupled mechanisms of water absorption and swelling. This study investigates the anisotropic characteristics and time-dependent evolution of swelling deformation in red-bed mudstone from Central Sichuan through hygroscopicity tests, triaxial swelling rate tests, and swelling pressure tests. The results indicate that key physical and mechanical properties of red-bed mudstone—such as saturated water content, swelling pressure, and clay mineral content—exceed the thresholds for identifying swelling rocks, thus demonstrating a notable swelling potential. The swelling deformation exhibits significant anisotropy, with the swelling rate in the direction perpendicular to the bedding plane being markedly higher than that in the parallel direction, which can be attributed to the oriented arrangement of microscopic clay minerals. Following water absorption, the swelling process of red-bed mudstone can be categorized into three stages: rapid growth, slow development, and stabilization. The swelling rate, swelling pressure, and water absorption rate all exhibit a negative exponential time-dependent evolution pattern. Lateral constraints significantly inhibit water absorption and delay crack propagation. Utilizing the humidity stress field theory and the creep constitutive framework, a saturation variable is introduced to develop a three-dimensional swelling constitutive model for red-bed mudstone, establishing a quantitative relationship between swelling strain and saturation state. The findings of this research provide theoretical support for multi-field coupling analysis and the prevention of subgrade deformation in red-bed regions.

The fracturing of high-position, thick, and hard roof strata overlying deep coal seams generates dynamic stress waves that can trigger rockbursts in stopes. The fractured zone serves as the primary pathway for wave transmission from the source to the stope. To investigate the propagation behavior of stress waves within fractured rock masses, this study employs a combination of split Hopkinson pressure bar (SHPB) testing and FLAC3D numerical modeling to analyze the attenuation characteristics of stress waves in fractured sandstone of varying sizes. The findings reveal the following: (1) Under dynamic loading, the failure mode of fractured specimens significantly differs from that of intact specimens. As the size of the fractured sandstone increases, the transmission coefficient decreases while the energy dissipation ratio increases. When the impact velocity rises from 7.5 m/s to 12.5 m/s, the transmission coefficient and energy dissipation ratio of fully fragmented specimens increase by 27.3% and 13.3%, respectively; for moderately fragmented specimens, the increases are 22.9% and 14.7%. (2) The propagation of dynamic stress waves from the overlying strata to the mining site can be categorized into three stages: stress wave initiation, propagation through intact rock, and propagation through plastic rock. In the plastic rock stage, stress wave attenuation is markedly more pronounced than in the other stages. The peak particle velocity of the surrounding rock decreases by 33.8%, and the attenuation rate of elastic strain energy reaches 47%. (3) Numerous discontinuities within the fractured sandstone induce multiple transmissions and reflections of dynamic stress waves, resulting in significant energy loss. The intensified fragmentation of the sandstone further dissipates wave energy, which constitutes the primary cause of the substantial attenuation of dynamic stress waves. (4) In designing rockburst prevention measures for mining stopes subjected to strong dynamic stress, it is essential to implement zoned pressure relief in the overlying strata. By regulating the degree of overburden fragmentation, the propagation paths of dynamic stress waves can be progressively disrupted, thereby reducing their impact on the stability of the surrounding rock at the mining site. The research findings may serve as a valuable reference for the prevention and control of rockburst disasters induced by dynamic stress waves in deep mines.

To investigate the influence mechanism of high temperature on thermal damage and fracture of surrounding rock, tight sandstone samples from the second member of the Xujiahe Formation in southern Sichuan were selected. Uniaxial compression, Brazilian splitting, and thermogravimetric analysis experiments were conducted on the rock samples after heat treatment at temperatures ranging from 100 ℃ to 1 000 ℃. By integrating a micromechanical model based on thin section analysis and nanoindentation, we obtained insights into the deformation behavior, strength characteristics, failure modes, and micro-damage mechanisms of the heat-treated tight sandstone. The results indicate that: (1) the thermal weight loss of the rock samples exhibits three distinct stages: water loss (100 ℃–800 ℃), mineral structure transformation (around 400 ℃), and destruction (800 ℃–  1 000 ℃). Thermally induced microcracks within the grains begin to develop at 500 ℃ and become pronounced between 500 and 800 ℃, displaying characteristics of intergranular tension, accompanied by minor intragranular tension and intergranular shear. At temperatures between 800 ℃and 1 000 ℃, the number of intragranular microcracks increases rapidly, revealing a dominance of intragranular tension, secondary intergranular tension, and sporadic intergranular shear. (2) In uniaxial compression and Brazilian splitting tests on heat-treated rock samples, the deformation mechanism transitions from brittle to plastic as temperature rises; uniaxial compressive strength, tensile strength, and elastic modulus initially increase before decreasing. Between 300 ℃and 400 ℃, these parameters fluctuate locally due to the mesoscopic damage evolution of the rock samples, while the gradual increase in Poisson′s ratio confirms the synergistic effects of stiffness weakening and plastic strengthening. (3) The failure mode of the rock sample under Brazilian splitting is predominantly tensile, with pronounced central cracking characteristics as thermal damage intensifies. Under uniaxial compression, the failure mode of the rock sample changes with temperature in the following sequence: single cylindrical splitting→single inclined plane shear→V-shaped conjugate shear→multi-crack splitting exhibiting crushing features.

The Yellow River silt is widely utilized as subgrade fill material; however, it exhibits low strength, poor compactability, and significant variability in particle gradation, which undermines subgrade stability in highway construction. This study employed soybean urease induced calcium carbonate precipitation (SICP) to enhance the mechanical properties of the silt. Consolidated drained triaxial shear tests were conducted, along with measurements of calcium carbonate content and scanning electron microscopy (SEM) analyses, to investigate the influence of particle gradation, characterized by the coefficient of uniformity ( ) and the curvature coefficient ( ), as well as varying levels of cementation (uncemented, lightly cemented, and moderately cemented) on the shear characteristics of SICP-treated specimens. The following conclusions can be drawn: (1) The shear characteristics of Yellow River silt with different particle gradations is significantly improved by SICP treatment. Higher levels of cementation lead to increased shear strength and more pronounced dilatancy. (2) The reinforcement effect of SICP is notably affected by and . The peak deviator stress ( ) of the treated Yellow River silt increases with rising and initially increases before decreasing as increases. (3) The calcium carbonate content in the SICP-treated specimens is significantly influenced by and , with the highest proportion of effective calcium carbonate precipitation observed at = 5.71 and = 0.89. Considering the impact of particle gradation on the efficiency of SICP treatment, an empirical relationship has been established among the shear strength parameters, gradation parameters, and calcium carbonate content. This relationship serves as a reference for the application of SICP in reinforcing fine-grained subgrade soils.

To investigate the suffusion characteristics of soil with anisotropic initial fabric and the localized characterization of suffusion, two apparatuses were developed in this study. The first apparatus is designed for soil specimen preparation, taking into account initial fabric anisotropy and allowing for controllable sampling directions. The second apparatus is intended for suffusion testing and measures local pore water pressure in soil under triaxial stress conditions. The preparation apparatus enables precise control over the inclination angle of the compaction direction relative to the sampling direction. This inclination angle corresponds to the complementary angle of , where represents the angle of the principal axis direction of the initial fabric relative to the seepage direction. The suffusion apparatus captures various information, including local pore water pressure, deformation, and fine particle loss, under triaxial stress conditions from both global and local perspectives. Suffusion tests were conducted on the anisotropic soil specimens under conditions of = 0°,  = 30°, = 45°, and = 90°, thereby verifying the reliability of the developed apparatuses. The results indicate that, globally, the suffusion process can be divided into three stages: initial, development, and failure. Locally, the migration of fine particles exhibits significant localized characteristics and spatial heterogeneity. The local hydraulic gradients are uniformly distributed during the initial stage but become non-uniform during the development stage, characterizing the initiation of suffusion. Furthermore, a smaller results in greater heterogeneity of the soil pore structure induced by suffusion, leading to reduced internal stability of the soil. This weakening effect is directly reflected in the reduction of the initiation and failure hydraulic gradients as decreases (90°→45°→30°→0°). This phenomenon is closely related to the increased migration resistance of fine particles caused by the horizontal arrangement of soil particles induced by compaction. The apparatuses developed in this study provide reliable and effective tools for investigating suffusion anisotropy.

 To investigate the dynamic response characteristics of the free field in saturated frozen soil under the incidence of S1-waves with thermal effects, a porous thermo-elastic model for saturated frozen soil was proposed based on kinematic equations, constitutive equations, and generalized heat conduction equations. The analytical expressions for the dynamic responses of the free surface in saturated frozen soil under S1-wave incidence were derived using the Helmholtz decomposition method and the boundary conditions at the interface. Numerical calculations were conducted to examine the effects of phase lag of the heat flux, thermal conductivity, incident frequency, temperature, and contact parameters on reflector thermoelastic waves and the seismic response of the free field. The results indicate that the consideration of thermal effects significantly influences the magnitude of reflected waves and the outcomes of the free-field seismic response. When thermal efficiency is taken into account, the phase lag of the heat flux has a relatively minor effect on the reflectivity of reflected waves and the free-field seismic response. However, thermal conductivity has a significant impact on the reflectivity of reflected thermal waves, with both factors exhibiting a positive correlation. Additionally, temperature and contact coefficient parameters demonstrate complex interdependencies affecting wave reflection and the free-field seismic response. The incident frequency notably affects the reflected P2, P3, and S2 waves, showing a positive correlation.

Rammed-earth heritage sites in cold regions are often susceptible to top-edge erosional failure during rapid snowmelt, as the shallow surface layer is weakened and destabilized by the combined effects of thermo-hydro-saline-mechanical actions. This study examines the key hydrothermal processes and the evolution of erosion at the top edge during abrupt snowmelt events. A full-scale, in situ simulation field that replicates the original form and materials of the Beiting Ancient City site in Xinjiang was constructed. By integrating external meteorological monitoring with an internal macro-micro sensing system for heat, moisture, salt, and stress, we investigated the characteristics and controlling factors of top-edge erosional failure in a rammed-earth test wall. The results reveal distinct seasonal stages. During the freezing period, a cold and humid climate facilitates prolonged snow accumulation; rapid warming in early spring triggers swift snowmelt and infiltration, which serve as the critical environmental drivers of erosion. On the shaded slope segment of the wall top, the shallow layer experiences an erosion sequence of “particle migration→undercutting rills→coalesced rills→stripping and deepening”, following a staged pathway of “snowmelt infiltration-migration→freeze-induced expansion→melt-induced loosening →abrupt-melt erosion”. The wall exhibits a typical “top tension-bottom compression” shear failure mode that progresses incrementally each year, with a heightened risk of failure occurring when the mean air temperature exceeds 0? ℃ for two consecutive days and the diurnal temperature range surpasses 15?  ℃. These findings provide theoretical support and technical guidance for early warning and conservation strategies addressing erosion and sliding hazards in earthen sites located in cold regions.