Energy input imbalance on opposing slopes indicates the thermal regime of high-altitude permafrost. The resulting thermal asymmetry induces differential settlement, longitudinal cracks and slope instability, particularly in high embankment structures. While previous studies primarily focused on single orientations, the lack of systematic multi-orientation analysis hinders the development of comprehensive mitigation strategies. To address this, this study investigates the long-term effect of the hydrothermal process of the embankment under varying orientations to quantify internal thermal regimes and permafrost degradation. Numerical simulations yield the following quantitative findings: (1) For West-East oriented embankments, exhibit the most severe thermal asymmetry, which intensifies with service time and depth. The temperature discrepancy between the North and South slope toes at 1m depth consistently exceeds 1 °C in the 5th, 10th, 15th and 25th operation years; at the 5 m depth temperature differences rise sequentially to 0.27 °C, 0.36 °C, 0.40 °C and 0.44 °C. (2) Conversely, North-South oriented embankments present the minimum thermal asymmetry. At 1 m below the slope toe, their temperature difference stabilizes at approximately 0.1 °C across all four monitoring years; at 5 m depth, the values increase moderately to 0.11 °C, 0.15 °C, 0.16 °C and 0.18 °C, with all discrepancies confined within 0.2 °C. (3) The permafrost table elevation asymmetry is most prominent for West-East oriented embankments. Over the year 5-25, the spatial disparity expands continuously from 0.43 m, 0.69 m, 0.79 m to 1.14 m. Under North-South oriented embankments show the minimal permafrost table difference, which still grows gradually to 0.18 m, 0.24 m, 0.27 m and 0.40 m, at identical time nodes. These findings provide a critical theoretical basis for the design, construction, and maintenance of high-altitude embankment engineering.

In permafrost regions, the ground-air temperature serves as a crucial boundary condition for simulating the spatial distribution and predicting the changing trends of permafrost, as well as a primary parameter for assessing the surface energy budget in alpine regions. However, current research on the ground-air temperature relationship at the local scale remains insufficient, particularly in the context of the complex surface conditions of the Tibetan Plateau. This study observed and analyzed nearly seven years of air and ground surface temperature from eight sites with different microenvironments in the warm permafrost region of the Tibetan Plateau hinterland, investigating the quantitative impact of microenvironmental differences on the ground-air temperature relationship at the local scale. Results indicated that while the mean annual air temperature was relatively uniform, the ground surface temperature (5 cm depth) varied markedly, driven by shallow soil moisture, vegetation cover, and slope aspect. Sites with greater shallow soil moisture, higher vegetation cover, or north-facing (shady) slopes exhibited larger thermal offsets and longer lag times, particularly in the cold season. The ground-air temperature relationship at all eight sites was linear (R-2 > 0.90), with the slope (k) and intercept (b) values exhibiting significant spatial and temporal heterogeneity. Specifically, the k value decreased with increasing vegetation cover, being the smallest in alpine grassland (k = 0.76) and the largest in sunny slope (k = 0.97); the b value increased with increasing shallow soil moisture, and sunny slopes significantly promoted an increase in b value, being the smallest in shady slope (b = 1.18) and the largest in swamp meadow (b = 3.87). These microenvironmental differences further influenced permafrost stability, with high shallow soil moisture and dense vegetation (>30 % cover) reducing stability, while north-facing slopes provided more favorable thermal conditions. These findings have important implications for optimizing boundary conditions in permafrost model simulations.

Retrogressive thaw slumps (RTSs) are dynamic geomorphic responses to rapid degradation of permafrost, involving coupled thermal weakening, headwall erosion, sediment transport, and ground deformation. Conventional remote-sensing change detection can identify surface change, but it cannot fully resolve deformation timing, subsurface thermal conditions, or process attribution. Here, we propose a hybrid digital twin (HDT) framework that integrates multi-temporal UAV-LiDAR topography, DoD/LoD95 change detection, Siamese change modelling, meteorological forcing, borehole-constrained physics-informed neural network (PINN) thermal reconstruction and displacement monitoring into a unified evidence chain. The framework was tested on an RTS adjacent to the Qinghai-Tibet Railway near Yanshiping Town. Results show event-scale staged evolution rather than monotonic surface changes. The strongest erosion-dominated responses occurred during full 2023.10-2025.10 comparison with LoD95-filtered net volume changes of -12,427 m³; the warm-season interval in 2025 showed limited and nearly balanced change. Thermal reconstruction revealed widespread positive-temperature subsurface conditions, indicating thaw-induced weakening potential. The crack-meter event preceded the GNSS event by approximately 7.8 days, indicating a progression from localized headwall cracking to broader near-headwall settlement. The strongest HDT attribution occurred for the full 2023-2025 comparison, with an attribution score of 0.886 and confidence of 0.929, jointly supported by geomorphic change, SCM activity, thermal weakening, deformation events, and spatial proximity. These results demonstrate that the HDT framework advances RTS analysis from detecting where terrain changed to when deformation occurred and why RTS evolved, providing a practical and transferable pathway for mechanism diagnosis, risk identification, and future digital-twin monitoring of permafrost geohazards.

The cold and high-altitude regions are characterized by low air pressure, intense solar radiation, pronounced freeze-thaw cycles, and salt corrosion. For instance, the damage to concrete roads, culverts, and piers in the Tanggula area of the Qinghai-Tibet Plateau highlights the susceptibility of concrete structures in such regions to premature failure, thereby threatening the long-term safe operation of engineering projects. To enhance the service performance of concrete structures in high-altitude cold environments, this study proposes a novel approach involving the surface modification of basalt fibers using 3-glycidoxypropyltrimethoxysilane and nanosilica. The degradation characteristics of the material's mechanical properties, pore structure, and durability were systematically investigated through various characterization techniques, including apparent morphology analysis, compressive strength testing, relative dynamic elastic modulus measurement, scanning electron microscopy, and mercury intrusion porosimetry. Furthermore, the fractal dimension and KHSi-BF contribution rate were introduced to establish a damage prediction model, thereby improving the accuracy of damage assessment. The results demonstrate that KHSi-BF-N2B2 exhibits superior mechanical properties and durability compared to ordinary Portland cement (OPC) under low-pressure curing conditions. Upon incorporation of KHSi-BF, the particle size distribution of nano-silica becomes more uniform, dispersion is enhanced, and agglomeration is significantly reduced. This improves the interfacial bonding between KHSi-BF and the matrix, leading to increased compressive and tensile strengths, as well as enhanced durability. Compared with other specimens, KHSi-BF-N2B2 features fewer harmful pores and a denser microstructure. The KHSi-BF-SiO2 cementitious system demonstrates enhanced resistance to solar radiation, freeze-thaw cycles, and salt corrosion. This research provides valuable insights and serves as an important reference for addressing the deterioration challenges faced by concrete structures in alpine and high-altitude regions.

The thermal stability of permafrost, a foundation for engineering infrastructure in cold regions, is increasingly threatened by the dual stressors of climate change and anthropogenic disturbance. This study investigates the dynamics of the crushed rock revetted embankment at the Kunlun Mountain Section of the Qinghai-Tibet Railway, systematically investigating the coupled impacts of climate warming and engineering activities on permafrost thermal stability using borehole temperature monitoring data (2008-2024) and climatic parameter analysis. Results show that under climate-driven effects, the study area experienced an air temperature increase of 0.2 degrees C per decade over the 2015-2024. Concurrently, the mean annual air thawing degree-days (TDD) rose by 13.8 degrees C center dot d/a, leading to active-layer thickening at a rate of 3.8 cm center dot a- 1at natural ground sites. From 2008 to 2024, the active layer had thickened by 0.7-0.8 m. At the embankment toe (BH 5), the active-layer thickening rate (3.3 cm center dot a- 1) was 25 % lower than that at the natural ground borehole (3.8 cm center dot a- 1); correspondingly, the underlying permafrost temperature increase rate at the toe (0.3 degrees C per decade) was lower than that at the natural borehole (0.5-0.6 degrees C per decade). Permafrost warming rates decreased with depth. Shallow layers (above -2 m) were significantly influenced by climate, with warming rates of 0.3-0.6 degrees C per decade. In contrast, deep layers (below -10 m) showed warming rates converging with the background atmospheric temperature trend (0.2 degrees C per decade). Thermal regime disturbance was most pronounced at horizontal distances of 3.0-5.0 m from the embankment. Nevertheless, the crushed-rock revetment maintained a permafrost table 0.6 m shallower than that of natural ground, confirming its "thermal diode" effect (facilitating convective cooling in winter), which partially offset climate warming impacts. This study provides critical empirical data and validates the cooling mechanism of crushed-rock revetment, which is essential for predicting the long-term thermal stability and informing adaptive maintenance strategies for railway infrastructure in warming permafrost regions.

The bearing capacity and stability of permafrost engineering are fundamentally governed by the ground thermal regime. In high-altitude regions, the coupling of intense solar radiation and the high heat absorption of wide asphalt pavements exposes embankments to severe energy input, exacerbating permafrost degradation. Consequently, two-phase closed thermosyphons (TPCTs) have been adopted as efficient cooling measures. However, a systematic understanding of TPCT energy regulation mechanisms, cross-seasonal cold transport, and long-term cumulative effects in wide, high-heat-absorbing embankments remains lacking. To address this gap, this study analyzes the energy regulation and cold migration characteristics of TPCTs based on in-situ monitoring. Results indicate that TPCTs effectively mitigate warming, maintaining the permafrost table nearly identical to the natural state, with a difference of only 0.1-0.2 m. The effective cooling period spans mid-October to mid-April, creating a significant cooling zone at 4-8 m depth with an influence radius of 2.5 m and reducing ground temperatures by 2-3 degrees C. Although TPCTs cease operation during the warm season, the embankment retains a low-temperature advantage, keeping soil temperatures at 3-9 m more than 0.7 degrees C lower than natural ground by the season's end. Furthermore, TPCTs exhibit a continuous inter-annual cumulative cooling effect, characterized by a decrease in deep permafrost temperature of 0.01-0.02 degrees C/a and a permafrost table rise of 0.02 m/a. These findings confirm that TPCTs effectively inhibit permafrost degradation and enhance embankment thermal stability, providing a scientific basis for the design and construction of wide highways in high-altitude permafrost regions.

Permafrost ground ice represents a critical solid water resource and engineering medium, with high ice-content permafrost in the Qinghai-Xizang Engineering Corridor (QTEC) exhibiting heightened sensitivity to degradation, which in turn limiting our ability to accurately simulate permafrost degradation in climate models and to develop effective hazard mitigation strategies for engineering infrastructure in vulnerable cryospheric regions. Using 1158 subsurface borehole datasets, we employed Random Forest and Decision Tree modeling to simulate spatial ice-content distributions across the QTEC. Our results show that along a 560-km transect from Xidatan to Amdo, geological drilling data reveal 220 km of high ice-content permafrost and 180 km of seasonal frozen soil or ice-poor permafrost. High ice-content layers predominantly occur at the permafrost table (3-6 m depth). Vertical zonation analysis indicates that high ice-content permafrost forms continuous bands with increasing altitude. 200-m resolution simulations indicate 50% of the QTEC comprises high ice-content permafrost, with mean annual air temperature, ground temperature, active layer thickness, elevation, and precipitation as primary controlling factors. This study provides critical baselines for quantifying complex ground ice reserves and informs ecological conservation and engineering design in permafrost regions.

Thermokarst lakes (TLs) are major hotspots of carbon emissions in permafrost regions, where the sources and transformation of dissolved organic matter (DOM) strongly regulate greenhouse gas production. On the Qinghai-Tibet Plateau (QTP), however, the coupling between DOM sources and CO2 and CH4 dynamics remains poorly understood. Ultraviolet-visible (UV-Vis) absorption and three-dimensional excitation-emission matrix (3D-EEM) fluorescence spectroscopy were applied to characterize DOM in TLs and suprapermafrost groundwater (SPGW) across alpine wet meadow, meadow, and steppe ecosystems. The results showed that average dissolved organic carbon (DOC) concentrations were 10.1 f 7.6 mg L-1 in thermokarst lake water (TL water) and 9.1 f 4.7 mg L-1 in SPGW, contributing 19.6% and 7.8% of total carbon (TC), respectively. DOM in TLs exhibited obvious spatial differences in optical properties across alpine ecosystems, with specific ultraviolet absorbance at 254 nm (SUVA254) values ranging from 3.1 to 4.2 L m-1 mg-1 and spectral slope ratios (SR) from 1.4 to 2.2, indicating variations in aromaticity and molecular weight. Fluorescence-PARAFAC analysis showed that the DOM in TLs consisted of both terrestrially derived humic substances (46-52%) and microbially processed components (48-54%), whereas DOM in SPGW was dominated by humic-and fulvic-like material from soils and vegetation. Importantly, DOM composition and optical indices were significantly correlated with dissolved CO2 and CH4 concentrations in TLs, underscoring the regulatory role of DOM quality in greenhouse gas emissions. Elevated dissolved CO2 and CH4 content in SPGW further suggest that lateral inflows are a key pathway fueling TLs carbon emissions. Overall, this study highlights how DOM sources and transformations regulate the greenhouse gas potential of QTP TLs, advancing our understanding of permafrost carbon-climate feedbacks.

Under the complex terrain of the Himalayan region, permafrost degradation easily triggers severe engineering distress and geohazards, yet quantitative descriptions linking macroscopic climate to field hazard occurrence remain limited. Using historical observations and multi-scenario projected gridded air temperature datasets, this study analyzes the spatiotemporal patterns of freezing‒thawing indices in the Himalayan region (1960–2020). The thermal state parameter β is introduced (the ratio of the thawing index to the freezing index). Combined with field survey data along the engineering route (2020–2021), permafrost degradation and hazard mechanisms under complex terrain are investigated. Historically, the freezing index (−11.96 °C d per year) and thawing index (11.25 °C d per year) exhibited inverse trends at similar rates (p < 0.001). From a climatological perspective, the thermal evolution of permafrost exhibits a positive feedback mechanism characterized by nonlinear acceleration. β-index analysis reveals that within the primary degradation interval (0.4 ≤ β ≤ 4.8), the long-term rate of change follows an upward-opening parabolic trajectory relative to initial β values. This quantitatively outlines the accelerated degradation potential of permafrost as the regional heat surplus intensifies. Furthermore, CMIP6 multi-model projections indicate that under medium and high emission scenarios (SSP2-4.5 and SSP5-8.5), the regional freeze‒thaw environment will face irreversible and continuous degradation. Spatially, the expanding extent of the heat surplus prominently manifests as a contraction of high-freezing-index zones and an expansion of thawing indices, ensuring the persistence of spatial thermal differentiation. Ultimately, this study elucidates the evolutionary patterns of the freeze‒thaw environment in the Himalayan region, establishing a robust scientific foundation for long-term decision-making and risk assessment of major engineering projects in cold regions.

Climate warming-induced permafrost degradation significantly impacts the safe operation of road infrastructure, leading to the increased prevalence of embankment diseases (ED). To compare the differences in ED under different permafrost distributions and local factors, this study examines two important highways located in different regions: the Qinghai-Tibet Highway (QTH) and the Gonghe-Yushu Expressway (GYE). Statistical analysis and machine learning techniques were used to assess the types and characteristics of ED and to investigate their relationship with local factors. The analysis also explores how permafrost degradation, crack propagation, and moisture infiltration contribute to ED development. Furthermore, the study explores the causes of significant variations in ED across different highways (ordinary national road and expressway). Results indicate that ED in Qinghai Province are primarily characterized by cracks. In the southern GYE region, block, transverse, longitudinal, and alligator cracking make up 93.13 % of total ED, while 73.36 % of ED in the western QTH region are crack-related. The overall occurrence rate of ED on the QTH is much higher than that on the GYE. ED are strongly correlated with factors such as volumetric ice content (VIC), aspect, mean annual ground temperature (MAGT), and mean annual precipitation (MAP) (p <= 0.0001, r > 0). MAGT, VIC, and MAP significantly influence cumulative disease areas, pavement damage rates, and the International Roughness Index. These findings provide important insights into the challenges faced by road infrastructure in permafrost regions of Qinghai Province, offering guidance for future construction and maintenance strategies.

The durability evaluation of permafrost infrastructure heavily relies on interfacial strength characterization. However, existing constitutive models systematically underestimate the damage accumulation rate during freeze-thaw cycling while overestimating residual strength, leading to significantly increased structural safety risks and severely shortened engineering service life. This study investigates shear behavior at concrete-crushed rock soil interfaces under freeze-thaw cycles through laboratory direct shear tests. The shear stress-displacement relationship is analyzed as a function of cycle number, with nuclear magnetic resonance quantifying interfacial pore structure evolution. A four-parameter modified Duncan-Chang model is developed to establish a higher-order nonlinear constitutive framework that integrates freeze-thaw damage effects. Unlike traditional one- or two-parameter hyperbolic models, the proposed model captures complex deformation phases including plastic hardening and incipient strain softening, which are empirically observed in freeze-thaw-damaged interfaces. A two-stage energy decoupling mechanism was proposed to separately describe interfacial debonding energetics and particulate friction thermodynamics, establishing a direct correlation that correlates microstructural ice cementation rupture patterns with continuum-scale elastoplastic deformation characteristics under freeze-thaw cycles. Interfacial shear strength exhibits dual dependence on normal stress magnitude and freeze-thaw history, showing a 40.5% increase in strength at 300 versus 100 kPa normal stress after 15 cycles, followed by stabilized degradation rates attributed to self-organized ice recrystallization patterns. Porosity progressively expands by 0.9%-2.3% with cycling, driven by phase transition-induced microcrack bifurcation and bidirectional pore restructuring (micropore coalescence/macropore fragmentation), which inversely correlates with cohesion reduction. The four-stage constitutive model with cubic-hyperbolic cyclic damage corrections achieves R-2 > 0.95 via nonlinear least-squares validation. The model can explain the stress-displacement process of the interface and the strain softening phenomenon in detail, which can provide a basis for the numerical simulation and theoretical calculation of the structure in the frozen soil ground.

Permafrost proximate to the boundary is extremely vulnerable to climate change. The Xidatan region, located at the northern boundary of permafrost on the Qinghai-Xizang Plateau (QXP), has been experiencing accelerated degradation, which introduces considerable uncertainties to future land use, water resource management, and infrastructure maintenance on the QXP. In this study, we obtained surface temporal deformation for the Xidatan region from 2017 to 2025 using Sentinel-1 data. We investigated permafrost deformation characteristics in the Xidatan region using long-term deformation rates from five profiles and ground temperature monitoring data. Ground temperature monitoring data from the Xidatan region indicate that over the past decade, mean annual ground temperature warming rates at depths of 6 m (WR_6) and 15 m (WR_15) were 0.012 degrees C/a and 0.015 degrees C/a, respectively. This indicates a significant warming process in the permafrost layer. InSAR analysis reveals a pronounced long-term subsidence trend in the Xidatan region, with substantial spatial differences in surface deformation. Subsidence is particularly evident in higher-elevation areas, while lower-elevation areas exhibit slight surface uplift. Analysis of surface deformation characteristics and borehole validation indicate that the lowest elevation of permafrost occurrence in the Xidatan region was approximately 4406 m by 2025. Compared to previous research, the permafrost limit has risen by a maximum of 37 m from 2012 to 2025. Permafrost degradation in the Xidatan region is primarily attributed to climate warming and increased anthropogenic disturbances. Should air temperatures continue rising, the northern boundary of permafrost distribution on the QXP will continue to ascend to higher elevations.

Freeze-thaw cycles (FTCs) induce structural deterioration and mechanical degradation, further triggering the instability of loess subgrade in seasonally frozen regions of Northwest China. Coal gasification coarse slag (CGS), a Si-Al-Ca-rich industrial solid waste, offers a low-carbon and sustainable solution for loess improvement. However, existing studies focus mainly on macroscopic mechanical enhancement of CGS-modified loess, while the multi-scale coupling evolution of mechanical properties, pore fractal characteristics, and microstructure under FTC remains unclear, limiting long-term performance evaluation and application. To address this gap, unconfined compressive strength, direct shear, consolidation compression, and scanning electron microscopy (SEM) tests were conducted on CGS-modified loess with varying dosages before and after FTCs. The FTC-induced evolution of mechanical behaviors and pore fractal features was quantitatively analyzed from multi-scale perspectives. Results show that CGS significantly improves loess mechanical performance, with 6% as the optimal dosage. Compared with untreated loess, 6% CGS-modified loess exhibits increases of 58.8%, 18.3%, and 99.3% in unconfined compressive strength (UCS), shear strength (SS), and cohesion strength (CS), respectively. FTC causes irreversible mechanical degradation and pore damage in natural loess, but CGS incorporation effectively mitigates such deterioration. Microscopic analysis reveals that CGS promotes particle flocculation and aggregation, refines micropores, and suppresses pore expansion and structural loosening during FTC. Specifically, the average pore area ratio of 6% CGS-modified loess is reduced by 25.4% relative to pure loess. Fractal characterization further confirms that coordinated evolution of macroscopic mechanics and microstructure governs freeze-thaw durability. This study establishes a macro-mechanics-fractal-microstructure multi-scale correlation mechanism for FTC damage in CGS-modified loess. The findings provide theoretical and technical support for long-term stability evaluation of solid waste-modified loess subgrade in seasonal frozen regions and large-scale utilization of CGS.

Permafrost degradation along the vital Qinghai‒Tibet Engineering Corridor (QTEC) poses a severe, escalating threat to critical infrastructure, yet high-resolution assessments of future risk and economic costs are urgently required to inform resilient infrastructure planning and cost-benefit optimization. Here, we integrate 90-m resolution multi-source data using machine learning and a multi-indicator framework to project the evolution of permafrost stability, infrastructure risk, and associated economic losses through the year 2090 under different climate scenarios. Our projections reveal a severe degradation trajectory. By 2090 under a high-emission scenario (SSP5-8.5), permafrost coverage will shrink by over 80%, with mean ground temperatures rising by 5.2 °C and active layer thickening by 1.34 m. Consequently, high-risk zones will expand to cover 84% of the corridor. Our risk zonation is robustly validated by InSAR observations, which show significantly (p < 0.01) higher deformation rates in the predicted high-risk zones, and by thermokarst inventories, with over 70% of hazards occurring in these zones. Economically, this translates to additional infrastructure replacement costs reaching 2.55 billion USD by 2090 (SSP5-8.5), with road and rail systems accounting for 91.3% of costs. This study provides the first 90-m resolution quantification of coupled permafrost‒risk‒cost dynamics, offering a critical scientific foundation for climate-adaptive engineering and proactive infrastructure investment.

Near-surface temperature and moisture are key boundary conditions for simulating permafrost distribution, projecting its response to climate change, and evaluating the surface energy balance in alpine regions. However, in desertified permafrost zones of the Qinghai-Tibet Plateau (QTP), the observations remain sparse, and reported trends vary considerably among sites. This lack of consistent evidence limits the ability to represent microenvironmental processes in models and to predict their influence on permafrost stability. From September 2021 to August 2024, we conducted continuous observations at a desertified permafrost site on the central QTP, covering the vertical range from 150 cm above to 100 cm below the ground surface (boundary layer). Measurements included air and ground temperature, air humidity, soil moisture, wind speed, and net radiation. Results showed that the mean annual air temperature increased with decreasing height at a gradient of approximately 0.42 degrees C/m, while mean annual air humidity remained nearly constant at 56.8 +/- 1.1 % (150-0 cm). In the near-surface soil layer (0 similar to -10 cm), temperature rose by 3.6 +/- 0.1 degrees C and moisture decreased by 34.0 +/- 2.7 %. The mean annual ground temperature increased with depth at a rate of about 0.55 degrees C/m, whereas soil moisture decreased between -20 and -60 cm (52.86 %/m) and increased between -60 and -100 cm (56.30 %/m). Seasonal patterns showed marked difference: in the freezing season, the calculated total temperature increment within the boundary layer (1.91 degrees C) was 61 % lower than the observed value (4.88 degrees C), while in the thawing season, it was 58 % higher (4.38 degrees C > 2.77 degrees C). These results reveal strong vertical gradients and seasonal contrasts in thermal and moisture regimes, emphasizing the need to integrate coupled temperature-moisture processes into boundary layer parameterizations for cold-region environments. Improved representations can enhance permafrost modeling and inform infrastructure design in regions experiencing both warming and desertification.

Concrete infrastructure in high-altitude cold regions suffers severe durability degradation due to complex multi-field environmental coupling. Furthermore, traditional structural health monitoring parameters lack stability in severely damaged materials. To address these challenges, this study aims to develop a high-durability modified concrete and establish a highly reliable acoustic emission (AE) early-warning method. A novel KHSi-BF-SiO2 system was prepared using basalt fibers synergistically modified by gamma-glycidoxypropyltrimethoxysilane and nano-silica. Specimens were subjected to simulated plateau environments (coupled low air pressure, solar radiation, freeze-thaw cycles, and salt erosion). The macroscopic mechanical degradation and damage evolution were evaluated using uniaxial compression, Digital Image Correlation (DIC), and AE monitoring, while microstructural deterioration was analyzed via Scanning Electron Microscopy (SEM) and Mercury Intrusion Porosimetry (MIP). Furthermore, Mel-Frequency Cepstral Coefficients (MFCC) were extracted from full-waveform AE signals to identify failure precursors. The results indicate that: (1) Although low-pressure curing impedes cement hydration, the KHSi-BF-SiO2 modification significantly enhances structural durability through "micro-densification" and "mesoscopic toughening". (2) Chemical freeze-thaw degradation exhibits strong medium-dependence (NaCl > mixed solution > Na2SO4). Chloride-induced crystal expansion and C-S-H gel disintegration cause severe pore coarsening and macroscopic structural collapse. (3) Compared to conventional AE parameters, the proposed MFCC index demonstrates superior spatial channel consistency (deviation < 15%) and anti-interference capabilities. The distinct fluctuations of low-order MFCC accurately capture the precursor transition from stable deformation to macroscopic instability. This research provides a robust theoretical foundation for high-durability material design and reliable structural health monitoring in extreme environments.

The thermal coupling between the atmosphere and the subsurface on the Qinghai-Tibetan Plateau (QTP) governs permafrost stability, surface energy balance, and ecosystem processes, yet its spatiotemporal dynamics under accelerated warming are poorly understood. This study quantifies soil-atmosphere thermal coupling ((3) at the critical 0.1 m root-zone depth using in-situ data from 99 sites (1980-2020) and a machine learning framework. Results show significantly weaker coupling in permafrost (PF) zones (mean (3 = 0.42) than in seasonal frost (SF) zones (mean (3 = 0.50), confirming the powerful thermal buffering of permafrost. Critically, we find a widespread trend of weakening coupling (decreasing (3) at 66.7 % of sites, a phenomenon most pronounced in SF zones. Our driver analysis reveals that the spatial patterns of (3 are primarily controlled by surface insulation from summer rainfall and soil moisture. The temporal trends, however, are driven by a complex and counter-intuitive interplay. Paradoxically, rapid atmospheric warming is the strongest driver of a strengthening of coupling, likely due to the loss of insulative snow cover, while trends toward wetter conditions drive a weakening of coupling by enhancing surface insulation. Spatially explicit maps derived from our models pinpoint hotspots of accelerated decoupling in the eastern and southern QTP, while also identifying high-elevation PF regions where coupling is strengthening, signaling a loss of protective insulation and increased vulnerability to degradation. These findings highlight a dynamic and non-uniform response of land-atmosphere interactions to climate change, with profound implications for the QTP's cryosphere, hydrology, and ecosystems.

ObjectiveThe influence of cyclic freeze-thaw cycles has been identified as a primary factor contributing to the susceptibility of subgrades in loess areas of Northwest China to freeze-thaw diseases. Additionally, frost heave and thaw subsidence have been recognised as significant engineering challenges in road construction in cold regions. The quality of the soil constituting the subgrade is a pivotal factor in determining its vulnerability to freeze-thaw disease. However, the current civil engineering industry in cold regions and road projects are confronted with a scarcity of sand and gravel and other raw materials. Consequently, in loess areas, improved loess as a filler is frequently the preferred material for road construction. However, the utilisation of cement, lime and other inorganic binding materials to enhance loess can result in environmental degradation, while the employment of microorganisms and other novel materials to improve loess is more costly. Consequently, the present study aims to address the pressing issues of coal gangue, coal gasification coarse slag resourceful use and ecological environmental protection in Ningxia and other regions, by exploring the potential of combined utilisation of these novel materials. The study focuses on the utilisation of coal gangue and coal gasification coarse slag in the enhancement of loess subgrades, particularly in the context of freezing and thawing, moisture changes, displacement, and the underlying mechanisms that facilitate these processes. The objective is to mitigate the freezing and thawing disease afflicting the seasonal permafrost area loess subgrades, thereby broadening the scope of the utilisation of coal-based solid waste and offering a novel approach as a reference.MethodsThe research findings on the mechanical properties of gangue and coal gasification coarse slag-improved loess, along with the pre-test results and extant research results, were taken into consideration. The study determined three coal-based solid waste improved loess conditions, with 30% or 50% gangue mixing and 15% coal gasification coarse slag mixing. At the same time, vegetal loess and lime-improved loess were set up as the control group. Taking the Lanqin Expressway in Gaolan County, Lanzhou as a prototype, an indoor half-span scaled subgrade model was set up in the cryogenic test chamber, and three control sections were set up in each model, with temperature and moisture sensors arranged in each control section, and a percentage meter was set up on the top surface of the subgrade as a method to monitor the change of displacement. The low-temperature test hall is capable of simulating the seasonal temperature changes in the study area. Through monitoring the temperature, moisture and displacement change rules inside the subgrade under various working conditions under freeze-thaw action, and combining the results of SEM electron microscope scanning test and CT scanning test, the improvement effect and improvement mechanism of the loess subgrade filler improved by the gangue and gasification slag at the macroscopic and microscopic levels is illustrated.Results and Discussions The main research content and results are as follows:(1)An analysis of the internal temperature change rule of the subgrade model indicates that, in comparison with the plain loess subgrade, the 30% and 50% coal gangue improved loess subgrade and the 15% coal gasification coarse slag improved loess subgrade, at a depth of 0.1m, exhibit a 15% increase in temperature. At a depth of 3m, the lowest recorded value of the internal subgrade temperature increased by more than 16%, and at a depth of 0.5m, the 15% coal gasification coarse slag improved loess subgrade and the 50% coal gangue improved loess subgrade did not have a completely frozen moment. This finding suggests that the 15% gasification slag improved loess subgrade and the 50% coal gangue improved loess subgrade exhibit excellent thermal insulation performance and temperature stability.(2)By analysing the internal moisture change rule of the subgrade model, it is evident that compared with the plain loess subgrade fill, the optimal moisture content of 30% and 50% gangue improved loess subgrade fill and 15% coal gasification coarse slag improved loess. The moisture content of the three kinds of improved loess subgrades This finding indicates that the presence of gangue and gasification slag within the soil of the improved loess subgrade significantly hinders the movement of water.(3)Following the freezing and thawing of the three groups of improved subgrades under optimal water content conditions, it was found that the maximum freezing and swelling of 15% coal gasification coarse slag improved loess subgrades, 30% and 50% coal gangue improved loess subgrades were reduced by 67% and 59%, respectively. Furthermore, the maximum thawing and sinking were reduced by 62%, 57% and 63%, respectively, indicating that the mixing of 15% coal gasification coarse slag or 30% or more coal gangue can significantly reduce the frost expansion and thawing deformation of the loess subgrades. The findings demonstrate that the incorporation of 15% coal gasification coarse slag or more than 30% coal gangue can effectively mitigate the freezing and thawing deformation of loess subgrades. Furthermore, the efficacy of controlling frost heave deformation of these three groups of improved subgrades is comparable to that of 4% lime, and the effectiveness in controlling thawing and sinking deformation is even superior to that of 4% lime. (4) Microstructure analysis indicates that the addition of coal gangue can reduce the porosity of the improved loess, thereby diminishing water migration within the soil body of the subgrade during periods of freezing and thawing. Similarly, the incorporation of coal gasification coarse slag can decrease the free water content within the improved soil body, thus reducing the freezing point of the improved loess. The reduction in heat propagation efficiency within the soil body is achieved via different mechanisms, which in turn reduces the freezing and thawing deformation of the subgrade. Concurrently, the findings of thermal conductivity measurement revealed that the thermal conductivity of 15% coal gasification coarse slag or 30% and 50% coal gangue improved loess filler was considerably lower than that of plain loess, thereby enhancing the thermal insulation performance of coal-based solid waste improved loess subgrade.ConclusionsBased on the results of this research, the use of coal gangue and coal gasification coarse slag improved loess as subgrade filler has significant advantages. Compared with the plain loess filler, mixing 50% coal gangue or 15% coal gasification slag in loess can not only effectively improve the thermal insulation performance of the subgrade, reduce the change of water content, and then significantly reduce the deformation of the top surface of the subgrade, but also a large amount of coal gangue and coal gasification slag stockpile that can be obtained locally to make up for the current sand and gravel and other road construction materials are seriously lacking. However, in order to take full advantage of these benefits, the quality of the construction must be strictly controlled. Prior to construction, the proposed coal gangue, coal gasification coarse slag and loess must undergo a comprehensive test of basic physical properties such as grain size, density, moisture content and other indicators to ensure that the material meets the design requirements. During construction, the water content and compaction of the mixed fill is strictly controlled to ensure the overall quality and stability of the subgrade.

The general surface temperature is influenced by changes in air temperature and is affected by local factors. The Tibetan Plateau, characterized by its high altitude and intense solar radiation, shows substantial effects related to slope orientation. Slope orientation plays a crucial role in controlling the absorption and reflection of solar radiation, leading to variations in surface temperatures across various slope orientations and complicating the relationship between surface and air temperatures. This study focused on air temperature and slope orientation. Considering atmospheric parameters such as solar altitude angle (h), solar declination angle (delta), hour angle (psi), and local geographical parameters like latitude and longitude. We proposed a novel method for estimating surface temperature on slopes based on air temperature, validated using measured data from the established octagonal platform at the Huashixia permafrost observation station in the Tibetan Plateau. Results indicated that at a significance level of P < 0.05, the highest correlation coefficient and linear fitting coefficient (R2) calculated using daily average air temperature were 0.89 and 0.79, respectively. When applying monthly average air temperature, these values were 0.99 and 0.98. This demonstrated that using monthly average air temperature as an inversion parameter yielded high-accuracy near-surface temperatures. The study will provide important guidance for optimizing design parameters and setting boundary conditions for stability prediction models for railways, highways, and other infrastructure under different orientations.

Alpine wet meadow (AWM), an important wetland type on the Qinghai-Tibet Plateau (QTP), is sensitive to climate change, which alters the soil hydrothermal regime and impacts ecological and hydrological functions in permafrost regions. The mechanisms underlying extreme AWM degradation in the QTP and hydrothermal factors controlling permafrost degradation remain unclear. In this study, soil hydrothermal processes, soil heat migration, and the permafrost state were measured in AWM and extremely degraded AWM (EDAWM). The results showed that the EDAWM exhibited delayed onset of both soil thawing and freezing, shortened thawing period, and extended freezing period at the lower boundary of the active layer. The lower ground temperatures resulted in a 0.2 m shallower active layer thickness in the EDAWM compared with the AWM. Moreover, the EDAWM altered soil thermal dynamics by redistributing energy, modifying soil moisture, preserving soil organic matter, and adjusting soil thermal properties. As for energy budget, a substantial amount of heat in the EDAWM was consumed by turbulent heat fluxes, particularly latent heat flux, which reduced the amount of heat transferred to the ground. Additionally, the higher soil organic matter content in EDAWM decreased the annual mean soil thermal conductivity from 1.42 W m- 1 K-1 in AWM to 1.26 W m- 1 K-1 in EDAWM, slowing down heat transfer within the active layer and consequently mitigating permafrost degradation. However, with continued climate warming, the soil organic matter content in EDAWM will inevitably decline due to microbial decomposition in the absence of new organic inputs. As the soil organic matter content diminishes, soil heat transfer processes will likely accelerate, and the permafrost warming rate may surpass that in undistributed AWM. These findings enhance our understanding of how alpine ecosystem succession influences regional hydrological cycles and greenhouse gas emissions.