3D spatial interpolation of soil heavy metals by combining kriging with depth function trend model

Advanced three-dimensional prediction model based on stable machine learning for soil pollution: A case study from a contaminated site in Southern China

With over five million contaminated sites worldwide, accurately characterizing the three-dimensional (3D) distribution of soil contamination is critical for effective risk assessment and site remediation. However, current 3D interpolation methodologies often fail to simultaneously account for spatial correlation and spatial heterogeneity, both of which are critical for capturing the complex spatial structure of subsurface contamination. This study developed a refined 3D interpolation model that integrates site characteristics, spatial position, spatial correlation, and spatial heterogeneity to simulate site contamination and quantify prediction uncertainty. The proposed machine learning (ML) model achieved high predictive performance, with coefficient of determination (R2) values above 0.73 for four heavy metals (HMs). To enhance model generalizability, a stability analysis framework was developed alongside a novel model selection strategy based on random dataset partitioning and random ordering of input covariates, and 1000 random simulations could provide a reliable basis for model screening. This study introduces a new, precise 3D spatial interpolation method. Owing to the easy accessibility of its covariates, it offers high versatility, making a significant contribution to site assessment and remediation efforts.

A spatial interpolation method based on 3D-CNN for soil petroleum hydrocarbon pollution

Petroleum hydrocarbon pollution causes significant damage to soil, so accurate prediction and early intervention are crucial for sustainable soil management. However, traditional soil analysis methods often rely on statistical methods, which means they always rely on specific assumptions and are sensitive to outliers. Existing machine learning based methods convert features containing spatial information into one-dimensional vectors, resulting in the loss of some spatial features of the data. This study explores the application of Three-Dimensional Convolutional Neural Networks (3DCNN) in spatial interpolation to evaluate soil pollution. By introducing Channel Attention Mechanism (CAM), the model assigns different weights to auxiliary variables, improving the prediction accuracy of soil hydrocarbon content. We collected soil pollution data and validated the spatial distribution map generated using this method based on the drilling dataset. The results indicate that compared with traditional Kriging3D methods (R2 = 0.318) and other machine learning methods such as support vector regression (R2 = 0.582), the proposed 3DCNN based method can achieve better accuracy (R2 = 0.954). This approach provides a sustainable tool for soil pollution management, supports decision-makers in developing effective remediation strategies, and promotes the sustainable development of spatial interpolation techniques in environmental science.

Migration modeling of metal(loid)s in soil-groundwater systems from an abandoned mine: Based on multimethod integration

Metal(loid)s contamination of mine has been a global environmental challenge. Traditional investigations of metal(loid) distribution patterns and migration behavior in soil-groundwater systems are constrained by the high costs of drilling and sampling limitations, leading to significant uncertainties in contamination assessment. This study presents an integrated approach combining three-dimensional (3D) visualization with Random Forest (RF) modeling and GIS mapping to investigate metal(loid) contamination characteristics and migration behavior in a mining area's soil-groundwater system. We developed an RF model with 1000 decision trees to expand limited drilling data for comprehensive spatial coverage. Model performance was validated using R2 and Root Mean Square Error (RMSE) metrics. The validated predictions were integrated into 3D visualization models and analyzed in conjunction with GIS mapping to characterize spatial patterns. Through analysis of temporal groundwater sampling data across wet, dry, and transitional hydrological periods, combined with RF modeling, we visualized metal(loid) distribution patterns and characterized their migration behavior in the soil-groundwater system. This integrated methodology provides a novel framework for investigating metal(loid) distribution and migration in mine soil-groundwater systems, effectively bridging traditional exploration techniques with advanced numerical simulation.

Comparison of three spatial interpolation methods in predicting time-dependent toxicities of single substances and mixtures

This study aims to optimize the time-dependent toxicity assessments for both single substances, particularly those causing hormesis, and mixtures that exhibit toxicological interactions. To achieve this, three time-dependent toxicity prediction methods were developed using geologic interpolation techniques: Inverse distance weighted (IDW), Kriging, and linear interpolation based on Delaunay triangulation (LDT). The toxicity of 7 single substances and 80 mixtures on Vibrio qinghaiensis sp.-Q67, along with 6 single substances and 19 mixtures on Microcystis aeruginosa, were assessed to evaluate the predictive accuracy of these methods. The coefficient of determination (R2), mean absolute error (MAE), and root-mean-square error (RMSE) were employed as performance metrics during cross-validation. The results showed that IDW underperformed LDT and Kriging in terms of both RMSE and MAE, indicating that LDT and Kriging had superior accuracy compared to IDW. Although LDT and Kriging demonstrated comparable predictive capabilities, LDT was identified as the more practical option for time-dependent toxicity prediction due to its simplicity and no requirement for parameter tuning. Consequently, LDT was presented as a new, efficient, and user-friendly tool for assessing the time-dependent toxicity of both individual chemicals and chemical mixtures. LDT will help to better assess the ecological risks of chemicals.

Relationships between high-concentration toxic metals in sediment and evolution of microbial community structure and carbon-nitrogen metabolism functions under long-term stress perspective

Microorganisms are highly sensitive to toxic metal pollution and play an important role in the material cycling and energy flow of the water ecosystem. Herein, 13 sediment samples from Junchong Reservoir (Guangxi Province, China) were collected in December 2021. The spatial distribution of pollution levels for toxic metals and the effects of toxic metals on the composition, functional characteristics, and metabolism of microorganisms were investigated. The results demonstrated that the area is a proximate area to industrial zones with severity of toxic metal pollution. Their mean concentrations of As, Cu, Zn, and Pb were up to 128.79 mg/kg, 57.62 mg/kg, 594.77 mg/kg, and 97.12 mg/kg respectively. There was a strong correlation between As, Cu, Zn, and Pb, with the highest correlation coefficient reaching 0.94. As the level of toxic metal pollution increases, the diversity and abundance of microorganisms gradually decrease. Compared to those with lower pollution levels, the Shannon index in regions with higher pollution levels decreases by up to 0.373, and the Chao index decreases by up to 143.507. However, the relative abundance of Bacteroidota, Patescibacteria, and Chloroflexi increased by 23%, 20%, and 5%, respectively, indicating their higher adaptability to toxic metals. Furthermore, microbial carbon and nitrogen metabolism were also affected by the presence of toxic metals. FAPROTAX analysis demonstrated an abundant reduction of ecologically functional groups associated with carbon and nitrogen transformations under high toxic metal pollution levels. KEGG pathway analysis indicated that carbon fixation and nitrogen metabolism pathways were inhibited with increasing toxic metal concentrations. These findings would contribute to a better understanding of the effects of toxic metal pollution on sediment microbial communities and function, shedding light on the ecological consequences of toxic metal contamination.

Quantitative heterogeneous source apportionment of toxic metals through a hybrid method in spatial random fields

Toxic metals in soils pose hazards to food security and human health. Accurate source apportionment provides foundation for pollution prevention. In this study, a novel hybrid method that combines positive matrix factorization, Bayesian maximum entropy and integrative predictability criterion is proposed to provide a new perspective for exploring the heterogeneity of pollution sources in spatial random fields. The results suggest that Cd, As and Cu are the predominant pollutants, with exceedance rates of 27%, 12% and 11%, respectively. The new method demonstrates superiority in predicting toxic metals when combined major and all sources as auxiliary information., with the improvements of 44% and 46%, respectively, Although the major sources identified with the hybrid method are the primary contributors to the accumulation of toxic metals (e.g. coal combustion for Hg, traffic emission for Pb and Zn, industrial activities for As, agricultural activities for Cd and Cu and natural sources for Cr and Ni), the impact of nonmajor sources on toxic metal sin specific regions should not be ignored (e.g. industrial activities on Ni, Pb and Zn in the north and natural sources on Cd, Cu, As, Pb and Zn in the south). For better pollution control, specific local sources should be considered.