This study investigates the stabilization of lateritic soil through partial replacement of cement with flue gas desulfurization (FGD) gypsum, aiming to enhance its engineering properties for pavement subgrade applications. Lateritic soils are known for their high plasticity and low strength, which limit their utility in infrastructure. To address these challenges, soil specimens were treated with varying cement contents (3%, 6%, 9%) and FGD gypsum additions (1%-6%). Laboratory tests were conducted to evaluate plasticity, compaction, permeability, unconfined compressive strength (UCS), California Bearing Ratio (CBR), and fatigue behaviour. The optimal mix 6% cement with 3% FGD gypsum demonstrated significant improvements: UCS increased by over 110% after 28 days, permeability reduced by 26%, and soaked CBR improved by 56% compared to untreated soil. Additionally, fatigue life showed remarkable enhancement under cyclic loading, indicating increased durability for high-traffic applications. To support predictive insights, machine learning models including Decision Tree, Random Forest, and Multi-Layer Perceptron (MLP) were trained on 168 data samples. The MLP and Random Forest models achieved high prediction accuracy (R2 approximate to 0.98), effectively capturing the non-linear interactions between mix proportions and UCS. SHAP (SHapley Additive exPlanations) analysis identified curing duration as the most influential factor affecting strength development. This integrated experimental-computational approach not only validates the feasibility of using FGD gypsum in sustainable soil stabilization but also demonstrates the effectiveness of machine learning in predicting key geotechnical parameters, reducing reliance on extensive laboratory testing and promoting data-driven pavement design.

Glaciers playa vital role in providing water resources for drinking, agriculture, and hydro-electricity in many mountainous regions. As global warming progresses, accurately reconstructing long-term glacier mass changes and comprehending their intricate dynamic relationships with environmental variables are imperative for sustaining livelihoods in these regions. This paper presents the use of eXplainable Machine Learning (XML) models with GRACE and GRACE-FO data to reconstruct long-term monthly glacier mass changes in the Upper Yukon Watershed (UYW), Canada. We utilized the H2O-AutoML regression tools to identify the best performing Machine Learning (ML) model for filling missing data and predicting glacier mass changes from hydroclimatic data. The most accurate predictive model in this study, the Gradient Boosting Machine, coupled with explanatory methods based on SHapley Additive eXplanation (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) analyses, led to automated XML models. The XML unveiled and ranked key predictors of glacier mass changes in the UYW, indicating a decrease since 2014. Analysis showed decreases in snow water equivalent, soil moisture storage, and albedo, along with increases in rainfall flux and air temperature were the main drivers of glacier mass loss. A probabilistic analysis hinging on these drivers suggested that the influence of the key hydrological features is more critical than the key meteorological features. Examination of climatic oscillations showed that high positive anomalies in sea surface temperature are correlated with rapid depletion in glacier mass and soil moisture, as identified by XML. Integrating H2OAutoML with SHAP and LIME not only achieved high prediction accuracy but also enhanced the explainability of the underlying hydroclimatic processes of glacier mass change reconstruction from GRACE and GRACE-FO data in the UYW. This automated XML framework is applicable globally, contingent upon sufficient high-quality data for model training and validation.

Floods pose a significant risk for Bangladesh due to the country's geographical and climatic conditions. Traditional methods of predicting flood risk often fail to do justice to the complex dynamics of flood vulnerability in this region. This report provides a comprehensive overview of the use of advanced machine learning (ML) algorithms for flood risk prediction in Bangladesh. It addresses four primary areas of research: (a) factors influencing floods considered in ML-based studies, (b) performance metrics of ML models, and (c) research gaps and future challenges in ML-based flood risk prediction. This review identified 42 unique factors that influence flooding, with precipitation, distance from the river, elevation, orientation, land use and land cover, and soil type emerging as the most important. ML models showed high predictive performance with an accuracy of 82% to 95%, depending on the algorithm and dataset used. However, there are still problems with data quality and regional variability that affect the reliability of the models. To improve flood forecasting, integrating real-time data, combining ML with physical models and promoting stakeholder engagement are crucial. Future research should focus on improving data quality, combining ML and physical models, and integrating future climate projections to refine flood hazard mapping. By considering these aspects, this study contributes to improving flood risk assessment and sustainable flood management strategies in Bangladesh, which could reduce socio-economic losses and environmental damage -in high-risk areas by 20-30.

This study investigated the infestation of tomato plants by the plant-parasitic nematode, M. incognita, and its accurate detection by plant electrophysiology (PE). Dedicated tests were done on whole plants to record electrophysiological signals from nematode infested and uninfested plants and to establish a trained model indicating nematode-induced stress. Monitoring nematode-induced stress by PE confirmed the results obtained by assessing root galls and quantifying xylem sap 3 to 4 weeks after infestation. The machine learning model captured the stress intensities and the time course of plant damage caused by nematodes. Stress caused by second-stage juveniles (J2) infestation appeared 3 to 5 days after infestation (DAI), whereas stress caused by egg infestation was detected 5 to 7 days later (10-13 DAI). For the first time, the real-time effectiveness of nematicides was recorded in further tests. Nematode infested plants treated preventatively with cyclobutrifluram (TYMIRIUM (R) technology) showed a delayed and short (about 3 days) period of low stress intensity, whereas infested but untreated plants showed a period of maximum stress for about 12 days. In addition, depending on the type of application (preventative or curative), different modes of biological activity of IRAC group N-2 and N-3 nematicides (fluopyram, abamectin) could be captured by PE signalling. PE offers a new way of monitoring plant health in real time, which is particularly valuable for accessing 'invisible' pests, such as plant-parasitic nematodes in the soil.

The rapid development of rural regions, the mountainous landscape, and frequent subtropical-typhoon-related rainfall have collectively contributed to a high incidence of cut slope-induced landslides in the coastal areas of eastern China. Despite the escalating risk, there has been a noticeable absence of comprehensive hazard assessments and targeted management measures for private housing and road construction in these rural environments. This paper introduces a novel approach for mitigating such risks by employing a susceptibility evaluation framework grounded in machine learning and uncertainty methods, combined with a double-index rainfall intensity-duration (I-D) threshold model. The proposed Intelligent Slope Prevention System operates through a sequential four-step process: (i) Site-specific landslide susceptibility is assessed through cut slope feature investigations and the use of three machine learning algorithms, namely, Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Network (ANN); (ii) the double-index model calculates rainfall thresholds, accounting for both prolonged continuous rainfall and short-term heavy rainfall events; (iii) the integration of rainfall thresholds with susceptibility assessments allows for the categorization of hazard levels; and (iv) tailored management strategies are deployed for data collection and early warning issuance. The study demonstrates that the SVM achieved the highest prediction accuracy across soil, rock-soil mixed, and rock slopes. The double-index model further enhanced the system's performance by predicting all 20 rainfall-induced landslides, with 15 of them falling under high or very high warning levels. An empirical evaluation during a heavy rainfall event on 29th June 2021 confirmed the system's effectiveness in identifying high-hazard areas and issuing timely warnings, thus significantly mitigating potential damage. Implemented in the coastal mountain basins of eastern China, the Intelligent Slope Prevention System leverages the gathered knowledge to manage and regulate slope hazards effectively, thereby enhancing the safety of both residential and infrastructural assets.

The presence of excavations or cavities beneath the foundations of a building can have a significant impact on their stability and cause extensive damage. Traditional methods for calculating the bearing capacity and subsidence of foundations over cavities can be complex and time-consuming, particularly when dealing with conditions that vary. In such situations, machine learning (ML) and deep learning (DL) techniques provide effective alternatives. This study concentrates on constructing a prediction model based on the performance of ML and DL algorithms that can be applied in real -world settings. The efficacy of eight algorithms, including Regression Analysis, k -Nearest Neighbor, Decision Tree, Random Forest, Multivariate Regression Spline, Artificial Neural Network, and Deep Neural Network, was evaluated. Using a Python -assisted automation technique integrated with the PLAXIS 2D platform, a dataset containing 272 cases with eight input parameters and one target variable was generated. In general, the DL model performed better than the ML models, and all models, except the regression models, attained outstanding results with an R 2 greater than 0.90. These models can also be used as surrogate models in reliability analysis to evaluate failure risks and probabilities.

Unconfined compressive strength (UCS) is an important parameter of rock and soil mechanical behavior in foundation engineering design and construction. In this study, salinized frozen soil is selected as the research object, and soil GDS tests, ultrasonic tests, and scanning electron microscopy (SEM) tests are conducted. Based on the classification method of the model parameters, 2 macroscopic parameters, 38 mesoscopic parameters, and 19 microscopic parameters are selected. A machine learning model is used to predict the strength of soil considering the three-level characteristic parameters. Four accuracy evaluation indicators are used to evaluate six machine learning models. The results show that the radial basis function (RBF) has the best UCS predictive performance for both the training and testing stages. In terms of acceptable accuracy and stability loss, through the analysis of the gray correlation and rough set of the three-level parameters, the total amount and proportion of parameters are optimized so that there are 2, 16, and 16 macro, meso, and micro parameters in a sequence, respectively. In the simulation of the aforementioned six machine learning models with the optimized parameters, the RBF still performs optimally. In addition, after parameter optimization, the sensitivity proportion of the third-level parameters is more reasonable. The RBF model with optimized parameters proved to be a more effective method for predicting soil UCS. This study improves the prediction ability of the UCS by classifying and optimizing the model parameters and provides a useful reference for future research on salty soil strength parameters in seasonally frozen regions.

Gullying is one of the problems that cause soil degradation in semi-arid areas and should be predicted to mitigate its damaging effects. Three machine learning models have been employed in this work to map the susceptibility to gully erosion in the N'fis river basin in the Moroccan High Atlas. Utilizing high-resolution images from Google Earth alongside fieldwork data, we digitized 434 gully erosion events to construct the comprehensive inventory map. These data were divided into two groups: training (70%) and test (30%). Based on the literature research and the multicollinearity test, 11 conditioning factors were selected. The receiver operating characteristic (ROC) approach and other statistical measures were used to quantify the model's accuracy. The study findings highlight the significance of drainage density, slope, NDVI, and distance from roads as crucial factors influencing gully erosion in the study area. Among the evaluated machine learning algorithms, the random forest (RF) model exhibited the highest performance, with an area under the curve (AUC) value of 0.932. It was followed by adaptive boosting (AB) with an AUC of 0.902 and gradient-boosted decision trees (GBDT) with an AUC of 0.893. The maps produced reveal that the southern and central regions of the study area have the classes of very high and high gully erosion susceptibility. The outputs of the current study can be used by decision-makers to improve prevention planning and mitigation techniques against gully erosion damage.

Climate warming has aggravated the occurrence of thaw settlement in permafrost region, but the associated risk has not been precisely assessed or understood. This study applied four machine learning models to explore and compare the spatial distribution of thaw settlement risk in the Wudaoliang-Tuotuohe region, Qinghai-Tibet Plateau, namely, naive Bayesian, k-nearest neighbor, logistic model tree and random forest models. A total of 853 thaw settlement locations and 12 conditioning factors were used to train and validate the above four models. The results indicated that random forest model performed best with the highest accuracy. The risk map produced by random forest model implied that about 76.55% of thaw settlements were located in very high-risk regions, which only occupied 6.85% of study area. The volume ice content, active layer thickness and thawing degree days were the main factors leading thaw settlement. By further comparing the performances between random forest model and other three models, the overestimated and underestimated risk regions (Beiluhe and Tuotuohe basins), and imbalanced conditioning factors (altitude and slope angle) were determined. In contrast with similar studies, this research performed better in model construction and accuracy. The results can help designers to implement precautionary measures in thaw settlement risk management.