Environmental Earth Sciences最新文献

Accurate 3D geological modeling of Quaternary deposits is crucial due to their inherently heterogeneous structures, such as clay-silt lenses, channels, and slope deposits. These small- to medium-scale features significantly influence geological engineering and resource evaluation but are challenging to reconstruct with limited borehole data. In this study, a synthetic explicit 3D geological reference model containing typical Quaternary structures was constructed using GOCAD^®. Borehole data were systematically extracted at varying densities using both regular and irregular sampling layouts and used to build implicit models through Leapfrog Geo™ software. Quantitative comparisons between the explicit and implicit models were conducted using Jaccard distance and normalized City-Block Distance metrics. The results demonstrate that as borehole density increases, the accuracy of the implicit models improves significantly, particularly in capturing small-scale structures such as lenses and narrow channels. Meanwhile, the influence of sampling layout remains comparatively minor. The study identifies a critical borehole density threshold required to reliably reconstruct complex geological features, providing quantitative measures for optimizing borehole exploration strategies in Quaternary geological settings.

Saturated soil hydraulic conductivity (Ksat) is a key property for soil and water management in agricultural and environmental contexts. Due to the high cost and complexity of direct Ksat measurement, pedotransfer functions (PTFs) based on readily available soil variables are widely used. However, studies focusing on subtropical soils remain limited. This study aimed to develop PTFs for a monitored river basin with subtropical Brazilian soils and evaluate machine learning (ML) techniques for Ksat estimation. A dataset with 105 samples from the Ellert Creek watershed (Rio Grande do Sul, Brazil) was used. Twelve model sets were trained using different combinations of predictors, and six ML algorithms were tested: multiple linear regression, decision tree, random forest, support vector regression, artificial neural networks (ANN), and ANN combined with principal component analysis. Random forest and ANN showed the highest predictive performance, followed by linear regression and support vector regression. Decision trees performed least effectively. The best PTFs, using variables such as sand, clay, bulk density, and macroporosity, achieved R² = 0.75. These results represent a significant advancement for estimating Ksat in subtropical soils, supporting the use of ML-based PTFs where direct measurements are scarce. The developed models can improve hydrological simulations and contribute to sustainable water and soil resource management in subtropical regions.

Graphical abstract

Landfills are commonly employed for the disposal of solid waste; however, they pose a significant risk of groundwater contamination due to leachate. To address this concern, traditional impermeable systems composed of bentonite-sand composites are employed as liners at the base of landfills. This study investigates an alternative impermeable system featuring a composite material made from quarry stone chips and bentonite aimed at enhancing leachate containment. The primary objective is to assess the effects of varying bentonite dosage and various dry densities on the permeability performance of the stone chips-bentonite mixture. GDS permeation tests are conducted to assess the permeability performance of the soil mixture comprising stone chips and bentonite. Various dosages of bentonite (ranging from 3% to 13%) and different states of dry density (between 1.76 g/cm³ and 1.95 g/cm³) are considered. The experimental results demonstrate a significant reduction in the permeability coefficient with increasing bentonite content; specifically, an impressive decrease by three orders of magnitude is observed when the bentonite dosage reaches 7%, resulting in a permeability coefficient of 10^− 7 cm/s. Conversely, decreasing the dry density while maintaining a constant bentonite admixture leads to an increase in the permeability coefficient. Additionally, this study introduces a novel evaluation method for determining the permeability coefficient of mixed soil, thereby providing valuable insights into landfill design and leachate management.

Uniaxial compressive strength (UCS) is one of the most fundamental parameters used in rock mechanics and mining design; however, laboratory UCS testing is often time-consuming, costly, and impractical for continuous field applications. This study aims to develop a rapid and low-cost UCS estimation framework using two easily obtainable indices: Schmidt hammer rebound hardness (SHT) and point load strength (PLT). A total of 114 coal samples collected from the A1 panel of the Ömerler Mine were used to train and evaluate four machine-learning models; multiple linear regression (MLR), regression trees (RT), support vector regression (SVR with linear, polynomial, and RBF kernels), and artificial neural networks (ANN). Model performances were assessed through 5-fold cross-validation and statistically compared using the Friedman and Nemenyi tests. The ANN model achieved the highest predictive accuracy, with an R² value exceeding 0.85 and the lowest error metrics among all evaluated algorithms. SVR models also produced competitive results. Statistical rank comparisons confirmed the significant superiority of the ANN model over the RT method. The findings demonstrate that reliable UCS prediction can be achieved using only SHT and PLT, offering a practical and cost-effective alternative for preliminary geotechnical characterization in mining operations. The proposed framework provides field engineers with a fast decision-support tool for strength estimation when laboratory testing is limited or unavailable.

Landslides significantly threaten natural and built environments, necessitating accurate prediction models for effective hazard mitigation. There is an urgent need to further improve the performance of machine learning algorithms in predicting landslide susceptibility by monitoring the impact of optimization algorithms on the performance of these models. This study evaluates the performance of various machine learning classifiers, including k-Nearest Neighbors (kNN), Multi-Layer Perceptron (MLP), and Extreme Gradient Boosting (XGBoost), for landslide susceptibility mapping. Additionally, Particle Swarm Optimization (PSO) is employed to enhance model performance by optimizing hyperparameters. Mountainous areas in the eastern Mediterranean (the northern Kabir River basin in western Syria) were identified as a result of the high frequency of landslide events over the past two decades. Nineteen factors causing landslides were identified, with no factor excluded, as a result of a multicollinearity test. The results indicate that XGBoost achieves the highest performance among traditional models. When integrated with PSO, the PSO-XGBoost model further improves classification performance, demonstrating its robustness in handling complex spatial patterns. Feature importance analysis using SHAP confirms slope as the dominant factor, followed by TRI, rainfall, Aspect, TWI, and curvature, highlighting the role of topography and hydrology in landslide occurrence. Moderate lithology, NDVI, and LULC contributions and lower importance of Flow Accumulation and Soil Depth suggest complex environmental interactions. Model predictions show varying susceptibility distributions. PSO-MLP assigns the highest very high susceptibility (44.09%), while PSO-XGBoost provides a balanced classification (31.13%). The PSO-XGBoost model demonstrates superior predictive capability, offering reliable landslide susceptibility maps for disaster risk management and land-use planning.

Phosphorus (P) loss from agricultural soils is a major contributor to freshwater eutrophication, yet soil-specific thresholds for P leaching remain poorly defined, particularly in reservoir catchments. This study assessed P migration and environmental risk in the Xiangxi River watershed, a representative tributary of the Three Gorges Reservoir, China. Column leaching experiments were performed on four typical soils—purple soil (PS), yellow soil (YS), calcareous soil (CS), and yellow-brown soil (YBS)—under controlled rainfall and fertilization scenarios, complemented by sorption isotherm and flooding incubation tests. Distinct soil-specific responses were observed: PS and YS exhibited substantially higher P leaching losses, while CS and YBS demonstrated greater sorption capacity and retention potential. Sorption behavior followed the Langmuir model, with maximum sorption capacity (Q_m) ranked as CS > YBS > PS > YS. Although the degree of P saturation (DPS < 5%) remained low, the equilibrium P concentration (EPC₀ = 0.7–1.2 mg·L^− 1) exceeded established eutrophication thresholds for freshwater systems. Critical Olsen-P values, determined via split-line regression, ranged from 10.5 to 23.2 mg·kg^− 1, above which leaching risk increased markedly. Overall, the findings demonstrate that P leaching potential is strongly soil-dependent. Under current fertilization practices, PS and YS likely exceed their safe Olsen-P thresholds, whereas CS and YBS remain less vulnerable. These results emphasize the need for soil-specific P management strategies—such as threshold-based fertilization and hydrological mitigation—to reduce diffuse P pollution and protect water quality in the Three Gorges Reservoir region.

Complex interactions between land cover change, climate variability, and seasonal recharge processes shape groundwater dynamics in urban regions. This study investigates the hydrological impacts of urbanization and rainfall variability on groundwater levels (GWL) across four rapidly urbanizing districts in Eastern India, i.e., Kolkata, Khorda, Patna, and Ranchi, over a decadal period (2011–2021). Using a linear mixed-effects modeling (LMM) framework, the effects of built-up expansion, population growth, and precipitation on seasonal groundwater fluctuations were assessed, with groundwater levels disaggregated into four temporal categories: pre-monsoon minimum and maximum, and post-monsoon minimum and maximum. The model incorporates both fixed effects (urban and climatic indicators) and random effects (district-level heterogeneity), achieving a strong fit (R² = 0.961). Built-up area emerged as a statistically significant predictor of GWL, with its effect varying seasonally, particularly suppressing recharge during the pre-monsoon maximum phase. Population and rainfall showed limited direct influence, but exhibited significant interactions during dry periods, indicating that extraction intensity and rainfall-runoff dynamics jointly affect groundwater depletion. Spatial analysis revealed that model performance varied across districts, with highly urbanized areas, such as Kolkata, exhibiting more complex and less predictable aquifer responses. These findings demonstrate that urbanization modifies the seasonal timing and effectiveness of natural recharge, often decoupling rainfall from aquifer replenishment in densely built environments. The study highlights the importance of incorporating seasonal hydrological sensitivity into urban water management, providing valuable insights applicable to other data-scarce, rapidly urbanizing regions worldwide.

The quantitative identification of nitrate sources is of great significance for water resources management. Stable isotopes combined with Bayesian isotope mixing model (SIAR) model were widely used to identify nitrogen sources. However, limited attention has been paid to how variations in the isotopic composition of nitrate sources affect the estimated source contributions in isotope mixing models. Here, the δ¹⁵N-NO₃⁻ and δ¹⁸O-NO₃⁻ isotopes, the SIAR model, and the uncertainty and sensitivity analysis were used to quantify the contributions and uncertainties of nitrate sources in Huashan watershed. 60 surface water (SW) samples and 82 groundwater (GW)samples were collected from November 2021 to October 2022, and atmospheric deposition (AD), chemical nitrogen fertilizer (NF), soil nitrogen (SN), and manure and sewage (M&S) were determined as the potential nitrate sources. Source identification by SIAR indicated that in November 2021 the M&S was the main contributor of nitrate to SW, while NF was the main contributor to nitrate in groundwater. In April 2022, NF contributed the most to nitrate in surface water, while nitrate in groundwater mainly originated from SN and MS. The variation between November 2021 and April 2022 sources is due to spring fertilization and rainfall. The uncertainty analysis showed that the greatest uncertainties were in SN and NF. Sensitivity analysis showed that the changes in the nitrate isotopic composition of M&S had the greatest effect on the results for δ¹⁵N, whereas only the mean values of oxygen isotope values of AD had a greater effect on the results for δ¹⁸O. Fertilizer application and changes in soil fertility due to agricultural rotations and cropping practices are intrinsic to the high level of uncertainty in SN. Sensitivity analysis highlighted the necessity of accurately measuring the isotopic end-members of potential nitrate sources to reduce the uncertainty in SIAR-based nitrate source apportionment results. Management strategies for the Huashan watershed should focus on domestic sewage treatment, optimization of agricultural practices, and integrated management of strongly connected surface water–groundwater systems to proactively prevent nitrogen pollution risks.

Knowledge of water quality is crucial for sustainable management of the natural resource, yet such information is often limited by data scarcity and high monitoring costs. This study proposes a novel approach, leveraging underutilized hydrogeochemical data from (potentially) contaminated sites to characterize natural groundwater composition at the mesoscale. Although originally collected for remediation purposes, these data, if properly processed, can yield valuable insights into natural groundwater conditions, offering a cost-effective sustainable alternative to extensive monitoring programs. The proposed workflow is applied in the Ferrara province (Po Plain, Italy), a region affected by both anthropogenic and natural groundwater quality issues. Aggregated data processed through reproducible steps underwent multivariate analysis, confirming the method’s ability to depict natural geochemical heterogeneity. Results were validated against the official regional monitoring network, demonstrating improved spatial and temporal resolution. This approach provides a scalable solution for enhancing groundwater quality assessment and supporting sustainable groundwater management, without requiring significant new data collection.