This study quantifies the seismic fragility assessment of shallow-founded buildings in liquefiable and treated soils, enhanced by drainage and densification, considering both short-and long-term behaviors. A conceptual framework is proposed for developing seismic fragility curves based on engineering demand parameters (EDPs) of buildings subjected to various earthquake magnitudes. The framework for establishing seismic fragility curves involves three essential steps. First, nonlinear dynamic analyses of soil-building systems are performed to assess both the short-term response, which occurs immediately following an earthquake, and the longterm response, when excess pore water pressure completely dissipates, and generate a dataset of building settlements. The seismic responses are compared in terms of excess pore water pressure buildup, immediate and residual ground deformation, and building settlement to explore the dynamic mechanisms of soil-building systems and evaluate the performance of enhanced drainage and densification over short-and long-term periods. Second, 38 commonly used and newly proposed intensity measures (IMs) of ground motions (GMs) are comprehensively evaluated using five statistical measures, such as correlation, efficiency, practicality, proficiency, and sufficiency, to identify optimal IMs of GMs. Third, fragility curves are developed to quantify probability of exceeding various capacity limit states, based on structural damage observed in Taiwan, for both liquefaction-induced immediate and residual settlements of buildings under different levels of IMs. Overall, this study proposes a rapid and straightforward probabilistic assessment approach for buildings in liquefiable soils, along with remedial countermeasures to enhance seismic resilience.

As a potential source of damage, earthquake-induced liquefaction is a major concern for embankment safety and serviceability. Densification has been a popular method for improving the performance of liquefiable soils. Understanding embankment settlement mechanisms plays a fundamental role in determining densification remediation. In this work, nonlinear dynamic analysis of embankments on liquefiable soils is conducted by the finite-difference program FLAC3D (version 6.0) with the simple anisotropic sand constitutive model. Numerical models are validated via dynamic centrifuge test results reported in the literature. The effects of densification countermeasures on the mean and differential settlements are explored in this study. Furthermore, the effects of the densification spacing and width are investigated to optimize the geometry of the densified regions. The development of pore pressure and the movement of the surrounding loose soil are discussed. The results show that both the mean settlement and differential settlement should be simultaneously utilized to comprehensively assess the overall effectiveness of densification treatment. The mean settlement is influenced by the densification spacing and width, but the differential settlement is highly associated with the inner edge of the densified region. This study provides insight for improving the design of the location and lateral extent of densification regions to prevent excessive embankment settlement.

After sand liquefaction, buried underground structures may float, leading to structural damage. Therefore, implementing effective reinforcement measures to control sand liquefaction and soil deformation is crucial. Stone columns are widely used to reinforce liquefiable sites, enhancing their resistance to liquefaction. In this study, we investigated the mitigation effect of stone columns on the uplift of a shield tunnel induced by soil liquefaction using a high-fidelity numerical method. The liquefiable sand was modeled using a plastic model for large postliquefaction shear deformation of sand (CycLiq). A dynamic centrifuge model test on stone column-improved liquefiable ground was simulated using this model. The results demonstrate that the constitutive model and analysis method effectively reproduce the liquefaction behavior of stone column-reinforced ground under seismic loading, accurately reflecting the time histories of excess pore pressure ratio and acceleration. Subsequently, numerical simulations were employed to analyze the liquefaction resistance of saturated sand strata and the response of a shield tunnel before and after reinforcement with stone columns. Additionally, the effects of densification and drainage of the stone columns were separately studied. The results show that, after installing stone columns, the excess pore pressure ratio at each measurement point significantly decreased, eliminating liquefaction and mitigating the uplift of the tunnel. The drainage effect of the stone columns emerged as the primary mechanism for dissipating excess pore pressure and reducing tunnel uplift. Furthermore, the densification effect of stone columns effectively reduces soil settlement, particularly pronounced around the stone columns, i.e., at a distance of three times the diameter of the stone column.

Soil liquefaction is a major contributor to earthquake damage. Evaluating the potential for liquefaction by conventional experimental or empirical methods is both time-intensive and laborious. Utilizing a machine learning model capable of precisely forecasting liquefaction potential might diminish the time, effort, and expenses involved. This research introduces an innovative predictive model created in three phases. Initially, correlation analysis determines essential elements affecting liquefaction. Secondly, predictions are produced using Convolutional Neural Networks (CNN) and Deep Belief Networks (DBN), verified by K-fold cross-validation to guarantee resilience. Third, Ant Colony Optimization (ACO) improves outcomes by increasing convergence efficiency and circumventing local minima. The suggested EC + ACO model substantially surpassed leading approaches, such as SVM-GWO, RF-GWO, and Ensemble Classifier-GA, attaining a very low False Negative Rate (FNR) of 2.00 % when trained on 90 % of the data. A thorough performance evaluation shown that the model achieved a cost value of 1.133 % by the 40th iteration, exceeding the performance of other models such SVMGWO (1.412 %), RF-GWO (1.305 %), and Biogeography Optimized-Based ANFIS (1.7439 %). The model exhibited significant improvements in convergence behavior, with a steady decline in cost values, especially between the 20th and 50th iterations. Additional validation using empirical data from the Tohoku-oki, Great East Japan earthquake substantiated the EC + ACO model's enhanced accuracy and dependability in mirroring observed results. These findings underscore the model's resilience and efficacy, providing a dependable method for forecasting soil liquefaction and mitigating its seismic effects.

This study investigates the application of machine learning (ML) algorithms for seismic damage classification of bridges supported by helical pile foundations in cohesive soils. While ML techniques have shown strong potential in seismic risk modeling, most prior research has focused on regression tasks or damage classification of overall bridge systems. The unique seismic behavior of foundation elements, particularly helical piles, remains unexplored. In this study, numerical data derived from finite element simulations are used to classify damage states for three key metrics: piers' drift, piles' ductility factor, and piles' settlement ratio. Several ML algorithms, including CatBoost, LightGBM, Random Forest, and traditional classifiers, are evaluated under original, oversampled, and undersampled datasets. Results show that CatBoost and LightGBM outperform other methods in accuracy and robustness, particularly under imbalanced data conditions. Oversampling improves classification for specific targets but introduces overfitting risks in others, while undersampling generally degrades model performance. This work addresses a significant gap in bridge risk assessment by combining advanced ML methods with a specialized foundation type, contributing to improved post-earthquake damage evaluation and infrastructure resilience.

The properties of soils are highly complex, and therefore, the classification system should be based on multiple perspectives of soil properties to ensure effective classification in geotechnical engineering. The current study of research demonstrates a lack of correlation between classification systems based on soil plasticity and those based on in-situ mechanical properties of soils. A CPTu-based plasticity classification system is proposed using the soil behaviour type index (Ic), with its reliability and limitations discussed. The results indicate that (1) Ic has the capacity to predict the stratigraphic distribution from the in-situ mechanical properties of soils. It showed a significant linear correlation with wL, which soil classification zone was similar to that of clay factor (CF); (2) A CPTu-plasticity classification system is proposed to characterize both plasticity and in-situ mechanical properties of soils. This system allows for the initial classification of soils solely based on CPTu data. Furthermore, it has been established that Ic = 2.95 can delineate the boundary between high- and low-compressibility soils. (3) The error is only 25.2% relative to the Moreno-Maroto classification chart, and the system tends to classify soils of intermediate nature as clay or silt. The distance between the data points and both the C-line and the new C-line (Delta Ip, Delta IpIc) showed a significant positive correlation. Only one data point was misclassified, considering human error in measuring Ip. (4) The new classification chart has been found to be more applicable to offshore and marine soils.

Background and AimsSoil salinization is a major cause of land degradation and ecological damage. Traditional soil salinity monitoring techniques are limited in coverage and scalability, while remote sensing offers broader applicability and efficiency. This study addresses spatiotemporal variations in soil salt content (SSC) inversion across crop types in Tongliao City, Inner Mongolia, China, using an innovative integration of multi-temporal data and crop cover types, improving remote sensing monitoring accuracy.MethodsField sampling data and Sentinel-2 images from June to September in 2021 and 2022 were utilized. The deep learning U-net model classified key crops, including sunflower (33%), beet (12%), and maize (55%), and analyzed the effects of crop coverage on SSC across multiple time series. Six spectral variables were selected using the SVR-RFE model (R2 = 0.994, MAE = 0.016). SSC prediction models were developed using three machine learning methods (DBO-RF, PSO-SVM, BO-BP) and a deep learning method (Transformer).ResultsConsidering crop coverage variations improved the sensitivity of spectral variables to SSC response, enhancing predictive accuracy and model stability. Crop classification showed that the salinity index (SIs) correlated more strongly with SSC than the vegetation index (VIs), with SI6 having the highest correlation coefficient of 0.50. The Transformer model, using multi-time series data, outperformed other algorithms, achieving an average R2 of 0.71. The SSC inversion map from the Transformer model closely matched field survey trends.ConclusionThis research provides a novel approach to soil salinity prediction using satellite remote sensing, offering a scalable solution for monitoring salinization and valuable insights for environmental management and agricultural planning.

Uncertainties in carbon storage estimates for disturbance-prone dryland ecosystems hinder accurate assessments of their contribution to the global carbon budget. This study examines the effects of land-use change on carbon storage in an African savanna landscape, focusing on two major land-use change pathways: agricultural intensification and wildlife conservation, both of which alter disturbance regimes. By adapting tree inventory and soil sampling methods for dryland conditions, we quantified aboveground and belowground carbon in woody vegetation (AGC and BGC) and soil organic carbon (SOC) across these pathways in two vegetation types (scrub savanna and woodland savanna). We used Generalized Additive Mixed Models to assess the effects of multiple environmental drivers on AGC and whole-ecosystem carbon storage (C-total). Our findings revealed a pronounced variation in the vulnerability of carbon reservoirs to disturbance, depending on land-use change pathway and vegetation type. In scrub savanna vegetation, shrub AGC emerged as the most vulnerable carbon reservoir, declining on average by 56% along the conservation pathway and 90% along the intensification pathway compared to low-disturbance sites. In woodland savanna, tree AGC was most affected, decreasing on average by 95% along the intensification pathway. Unexpectedly, SOC stocks were often higher at greater disturbance levels, particularly under agricultural intensification, likely due to the preferential conversion of naturally carbon-richer soils for agriculture and the redistribution of AGC to SOC through megaherbivore browsing. Strong unimodal relationships between disturbance agents, such as megaherbivore browsing and woodcutting, and both AGC and C-total suggest that intermediate disturbance levels can enhance ecosystem-level carbon storage in disturbance-prone dryland ecosystems. These findings underline the importance of locally tailored management strategies-such as in carbon certification schemes-that reconcile disturbance regimes in drylands with carbon sequestration goals. Moreover, potential trade-offs between land-use objectives and carbon storage goals must be considered.

Soil and water conservation structures are vital for environmental resilience but present maintenance challenges due to their wide distribution and remote locations. To tackle these issues, a method using unmanned aerial vehicles (UAVs) combined with 360 degree photography was developed. UAVs captured images that were processed into panoramic and 3D models, enabling precise inspections of structural damage. These models were integrated into the disaster environment review and update (DER&U) rating system, enhanced by a fuzzy inference classification mechanism for improved damage estimation. Additionally, a management platform was created to boost overall efficiency and provide decision-making support for relevant authorities. The UAV-assisted inspection method demonstrated promising results, though certain limitations were also noted.

The substantial development of desiccation cracks profoundly impacts the mechanical and hydraulic properties of clayey soils, potentially leading to various engineering challenges such as slope failures. Therefore, identifying the soil's cracking potential is crucial for guiding engineering design and construction processes. The aim of this study was to propose a method for cracking potential classification for clayey soils. To this end, standard cyclic wet-dry tests, capable of maximizing the soil's cracking potential, were proposed based on an analysis of the cracking behavior of lateritic soils under different wet-dry conditions. Subsequently, the cracking characteristics of several typical clayey soils (i.e., lateritic soil, kaolinite, bentonite, and attapulgite) were examined by standard cyclic wet-dry tests. Finally, a novel method for cracking potential classification of clayey soils was proposed incorporating the entropy weighting method. The results show that the most significant degree of cracking in lateritic soil is observed under vacuum saturation and 60 degrees C oven-drying, which is identified as the standard wet-dry condition. When the crack development stabilizes after multiple standard wet-dry cycles, the cracking potential of the soil is characterized by parameters such as the total crack length, maximum crack width, surface crack rate and the fractal dimension of the cracks. On this basis, a classification method is proposed to categorize the cracking potential of clayey soils into five levels: extremely weak, weak, medium, strong, and extremely strong. The cracking potential of different clayey soils was evaluated using this method, revealing that bentonite exhibited the highest cracking potential, classified as extremely strong.