Contamination vertical distribution and key factors identification of metal(loid)s in site soil from an abandoned Pb/Zn smelter using machine learning

Selection of pollution-safe head cabbage: Interaction of multiple heavy metals in soil on bioaccumulation and transfer

Screening for pollution-safe cultivars (PSCs) is a cost-effective strategy for reducing health risks of crops in heavy metal (HM)-contaminated soils. In this study, 13 head cabbages were grown in multi-HMs contaminated soil, and their accumulation characteristics, interaction of HM types, and health risks assessment using Monte Carlo simulation were examined. Results showed that the edible part of head cabbage is susceptible to HM contamination, with 84.62% of varieties polluted. The average bio-concentration ability of HMs in head cabbage was Cd> > Hg > Cr > As>Pb. Among five HMs, Cd and As contributed more to potential health risks (accounting for 20.8%-48.5%). Significant positive correlations were observed between HM accumulation and co-occurring HMs in soil. Genotypic variations in HM accumulation suggested the potential for reducing health risks through crop screening. G7 is a recommended variety for head cabbage cultivation in areas with multiple HM contamination, while G3 could serve as a suitable alternative for heavily Hg-contaminated soils.

Source apportionment of soil PTE in a northern industrial county using PMF model: Partitioning strategies and uncertainty analysis

Positive matrix factorization (PMF) has commonly been applied for source apportionment of potentially toxic elements (PTE) in agricultural soil, however, spatial heterogeneity of PTE significantly undermines the accuracy and reliability of PMF results. In this study, a representative industrial-agricultural hub in North China (Xuanhua district, Zhangjiakou City) was selected as the research subject, multiple partition processing (PP) strategies and uncertainty analyses were integrated to advance the PMF modeling and associated algorithm mechanisms were comparatively discussed. Specifically, we adopted three methods to split the research area into several subzones according to industrial density (PP-1), population density (PP-2), and the ecological risk index (PP-3) respectively, to rectify the spatial bias phenomenon of PTE concentrations and to achieve a more interpretable result. Our results indicated that the obvious enrichment of Cd, Pb, and Zn was found in the agricultural soil, with Hg and Cd accounted for 83.49% of the overall potential ecological risk. Combining proper PP with PMF can significantly improve the modelling accuracy. Uncertainty analysis showed that interval ratios of tracer species (Cd, Pb, Hg, and Zn) calculated by PP-3 were consistently lower than that of PP-1 and PP-2, indicating that PP-3 coupled PMF can afford the optimal modeling results. It suggested that natural sources, fertilizers and pesticides, atmosphere deposition, mining, and smelting were recognized as the major contributor for the soil PTE contamination. The contribution of anthropogenic activities, specifically fertilizers and pesticides, and atmosphere deposition, increased by 1.64% and 5.91% compared to PMF results. These findings demonstrate that integration of proper partitioning processing into PMF can effectively improve the accuracy of the model even at the case of soil PTE contamination with high heterogeneity, offering support to subsequently implement directional control strategies.

Pollution levels and probability risk assessment of potential toxic elements in soil of Pb-Zn smelting areas

Soil pollution around Pb-Zn smelters has attracted widespread attention around the world. In this study, we compiled a database of eight potentially toxic elements (PTEs) Pb, Zn, Cd, As, Cr, Ni, Cu, and Mn in the soil of Pb-Zn smelting areas by screening the published research papers from 2000 to 2023. The pollution assessment and risk screening of eight PTEs were carried out by geo-accumulation index (Igeo), potential ecological risk index (PERI) and health risk assessment model, and Monte Carlo simulation employed to further evaluate the probabilistic health risks. The results suggested that the mean values of the eight PTEs all exceeded the corresponding values in the upper crust, and more than 60% of the study sites had serious Pb and Cd pollution (Igeo > 4), with Brazil, Belgium, China, France and Slovenia having higher levels of pollution than other regions. Besides, PTEs in smelting area caused serious ecological risk (PERI = 10912.12), in which Cd was the main contributor to PREI (86.02%). The average hazard index (HI) of the eight PTEs for adults and children was 7.19 and 9.73, respectively, and the average value of total carcinogenic risk (TCR) was 4.20 × 10-3 and 8.05 × 10-4, respectively. Pb and As are the main contributors to non-carcinogenic risk, while Cu and As are the main contributors to carcinogenic risk. The probability of non-carcinogenic risk in adults and children was 84.05% and 97.57%, while carcinogenic risk was 92.56% and 79.73%, respectively. In summary, there are high ecological and health risks of PTEs in the soil of Pb-Zn smelting areas, and Pb, Cd, As and Cu are the key elements that cause contamination and risk, which need to be paid attention to and controlled. This study is expected to provide guidance for soil remediation in Pb-Zn smelting areas.

Heavy metals concentration in soils across the conterminous USA: Spatial prediction, model uncertainty, and influencing factors

Assessment and proper management of sites contaminated with heavy metals require precise information on the spatial distribution of these metals. This study aimed to predict and map the distribution of Cd, Cu, Ni, Pb, and Zn across the conterminous USA using point observations, environmental variables, and Histogram-based Gradient Boosting (HGB) modeling. Over 9180 surficial soil observations from the Soil Geochemistry Spatial Database (SGSD) (n = 1150), the Geochemical and Mineralogical Survey of Soils (GMSS) (n = 4857), and the Holmgren Dataset (HD) (n = 3400), and 28 covariates (100 m × 100 m grid) representing climate, topography, vegetation, soils, and anthropic activity were compiled. Model performance was evaluated on 20 % of the data not used in calibration using the coefficient of determination (R2), concordance correlation coefficient (ρc), and root mean square error (RMSE) indices. Uncertainty of predictions was calculated as the difference between the estimated 95 and 5 % quantiles provided by HGB. The model explained up to 50 % of the variance in the data with RMSE ranging between 0.16 (mg kg-1) for Cu and 23.4 (mg kg-1) for Zn, respectively. Likewise, ρc ranged between 0.55 (Cu) and 0.68 (Zn), respectively, and Zn had the highest R2 (0.50) among all predictions. We observed high Pb concentrations near urban areas. Peak concentrations of all studied metals were found in the Lower Mississippi River Valley. Cu, Ni, and Zn concentrations were higher on the West Coast; Cd concentrations were higher in the central USA. Clay, pH, potential evapotranspiration, temperature, and precipitation were among the model's top five important covariates for spatial predictions of heavy metals. The combined use of point observations and environmental covariates coupled with machine learning provided a reliable prediction of heavy metals distribution in the soils of the conterminous USA. The updated maps could support environmental assessments, monitoring, and decision-making with this methodology applicable to other soil databases, worldwide.

Utilizing machine learning algorithm for finely three-dimensional delineation of soil-groundwater contamination in a typical industrial park, North China: Importance of multisource auxiliary data

Intensive industrial activities cause soil contamination with wide variations and even perturb groundwater safety. Precision delineation of soil contamination is the foundation and precondition for soil quality assurance in the practical environmental management process. However, spatial non-stationarity phenomenon of soil contamination and heterogeneous sampling are two key issues that affect the accuracy of contamination delineation model. Taking a typical industrial park in North China as the research object, we constructed a random forest (RF) model for finely characterizing the distribution of soil contaminants using sparse-biased drilling data. Results showed that the R2 values of arsenic and 1,2-dichloroethane predicted by RF (0.8896 and 0.8973) were greatly higher than those of inverse distance weighted model (0.2848 and 0.2908), indicating that RF was more adaptable to actual non-stationarity sites. The back propagation neural network algorithm was utilized to establish a three-dimensional visualization of the contamination parcel of subsoil-groundwater system. Multiple sources of environmental data, including hydrogeological conditions, geochemical characteristics and anthropogenic industrial activities were integrated into the model to optimize the prediction accuracy. The feature importance analysis revealed that soil particle size was dominant for the migration of arsenic, while the migration of 1,2-dichloroethane highly depended on vertical permeability coefficients of the soil. Contaminants migrated downwards with soil water under gravity-driven conditions and penetrated through the subsoil to reach the saturated aquifer, forming a contamination plume with groundwater flow. Our findings afford a new idea for spatial analysis of soil-groundwater contamination at industrial sites, which will provide valuable technical support for maintaining sustainable industry.

Machine learning prediction and interpretation of the impact of microplastics on soil properties

The annual microplastic (MP) release into soils is 4-23 times higher than that into oceans, significantly impacting soil quality. However, the mechanisms underlying how MPs impact soil properties remain largely unknown. Soil-MP interactions are complex because of soil heterogeneity and varying MP properties. This lack of understanding was exacerbated by the diverse experimental conditions and soil types used in this study. Predicting changes in soil properties in the presence of MPs is challenging, laborious, and time-consuming. To address these issues, machine learning was applied to fit datasets from peer-reviewed publications to predict and interpret how MPs influence soil properties, including pH, dissolved organic carbon (DOC), total P, NO3--N, NH4+-N, and acid phosphatase enzyme activity (acid P). Among the developed models, the gradient boost regression (GBR) model showed the highest R2 (0.86-0.99) compared to the decision tree and random forest models. The GBR model interpretation showed that MP properties contributed more than 50% to altering the acid P and NO3--N concentrations in soils, whereas they had a negligible impact on total P and 10-20% impact on soil pH, DOC, and NH4+-N. Specifically, the size of MPs was the dominant factor influencing acid P (89.3%), pH (71.6%), and DOC (44.5%) in soils. NO3--N was mainly affected by the MP type (52.0%). The NH4+-N was mainly affected by the MP dose (46.8%). The quantitative insights into the impact of MPs on soil properties of this study could aid in understanding the roles of MPs in soil systems.

Improving the mapping accuracy of soil heavy metals through an adaptive multi-fidelity interpolation method

Soil heavy metal pollution poses a serious threat to environmental safety and human health. Accurately mapping the soil heavy metal distribution is a prerequisite for soil remediation and restoration at contaminated sites. To improve the accuracy of soil heavy metal mapping, this study proposed an error correction-based multi-fidelity technique to adaptively correct the biases of traditional interpolation methods. The inverse distance weighting (IDW) interpolation method was chosen and combined with the proposed technique to form the adaptive multi-fidelity interpolation framework (AMF-IDW). In AMF-IDW, sampled data were first divided into multiple data groups. Then one data group was used to build the low-fidelity interpolation model through IDW, while the other data groups were treated as high-fidelity data and used for adaptively correcting the low-fidelity model. The capability of AMF-IDW to map the soil heavy metal distribution was evaluated in both hypothetical and real-world scenarios. Results showed that AMF-IDW provided more accurate mapping results compared with IDW and the superiority of AMF-IDW became more evident as the number of adaptive corrections increased. Eventually, after using up all data groups, AMF-IDW improved the R² values for mapping results of different heavy metals by 12.35-24.32%, and decreased the RMSE values by 30.35%-42.86%, indicating a much higher level of mapping accuracy relative to IDW. The proposed adaptive multi-fidelity technique can be equally combined with other interpolation methods and provide promising potential in improving the soil pollution mapping accuracy.

Risk source identification and diffusion trends of metal(loid)s in stream sediments from an abandoned arsenic-containing mine

Stream sediments from mine area are a converging source of water and soil pollution. The risk and development trends of metal(loid)s pollution in sediments from an abandoned arsenic-containing mine were studied using modelling techniques. The results showed that the combined techniques of geographic information system (GIS), random forest (RF), and numerical simulation (NS) could identify risk sources and diffusion trends of metal(loid)s in mine sediments. The median values of As, Cd, Hg, and Sb in sediments were 5.01, 3.02, 5.67, and 3.20 times of the background values of stream sediments in China, respectively. As (14.09%) and Hg (18.64%) pollution in mine stream sediments were severe while As is the main potential risk source with a strong spatial correlation. High-risk blocks were concentrated in the landfill area, with the surrounding pollution shows a decreasing trend of "step-type" pollution. The risk correlation between Hg and As (55.37%) in the landfill area is high. As a case of arsenic, the diffusion capacity of As within 500m is strong and stabilizes at 1 km when driven by the flows of 0.05, 0.5, and 5 m³/s, respectively. With the worst-case scenario flow (86 m³/s), it would take only 147 days for the waters within 3 km to become highly polluted. The high pollution levels in a stream under forecast of different distance intervals (500, 1500, 2000 m) within 6.5 km is arrived at approximate 344, 357, and 384 days, respectively. The study suggested the combined technique of GIS, RF, and NS can serve the risk source identification of contaminated sites and risk forecast of toxic element diffusion in emergency situations.

Pollution Risk Prediction for Cadmium in Soil from an Abandoned Mine Based on Random Forest Model

It is highly uncertain as to the potential risk of toxic metal(loid)s in abandoned mine soil. In this study, random forest was used to predict the risk of cadmium pollution in the soils of an abandoned lead/zinc mine. The results showed that the random forest model is stable and precise for the pollution risk prediction of toxic metal(loid)s. The mean of Cd, Cu, Tl, Zn, and Pb was 6.02, 1.30, 1.18, 2.03, and 2.08 times higher than the soil background values of China, respectively, and their coefficients of variation were above 30%. As a case study, cadmium in the mine soil had "slope" hazard characteristics while the ore sorting area was the major source area of cadmium. The theoretical values of the random forest model are similar to the practical values for the ore sorting area, metallogenic belt, riparian zone, smelting area, hazardous waste landfill, and mining area. The potential risk of soil Cd in the ore sorting area, metallogenic belt, and riparian zone are extremely high. The tendency of pollution risk migrates significantly both from the ore sorting area to the smelting area and the mining area, and to the hazardous waste landfill. The correlation of soil pollution risk is significant between the mining area, the smelting area, and the riparian zone. The results suggested that the random forest model can effectively evaluate and predict the potential risk of the spatial heterogeneity of toxic metal(loid)s in abandoned mine soils.

Potential Mechanism of Long-Term Immobilization of Pb/Cd by Layered Double Hydroxide Doped Chicken-Manure Biochar

Layered double hydroxide (LDH)-doped chicken-manure biochar (CMB) with long-term stability was synthesized to immobilize Pb/Cd. MgAl-Cl-LDH-doped CMB (MHs) showed prominent long-term oxidation resistance and the least biodegradation sensitivity. Efficient Pb/Cd adsorption was observed on MHs, and the maximum adsorption capacities of Pb(II)/Cd(II) reached 1.95 mmol/g and 0.65 mmol/g, respectively. Precipitation and isomorphous substitution were identified as the key adsorption mechanisms, which formed highly stable Pb/Cd species (PbAl-CO₃-LDH, Pb₃(OH)₂CO₃, CdAl-Cl-LDH and CdCO₃). Pb(II) and Cd(II) precipitated with CO₃^2- in MHs; meanwhile, Mg(II) and Ca(II) in LDH layers were substituted by Pb(II) and Cd(II) respectively. Therefore, MHs had the potential for long-term stability of Pb/Cd. Moreover, complexation and electrostatic adsorption also contributed to the Pb/Cd immobilization to a certain extent. When 5% MHs (w/w) was applied to Pb/Cd contaminated smelting site soils, the soil pH increased from 5.9 to 7.3. After applying MHs for 25 d, the content of bioavailable Pb(II) and Cd(II) decreased by 98.8% and 85.2%, respectively, and the content of soluble Pb and Cd dropped by 99.5% and 96.7%. This study paves the way for designing a novel LDH doped CMB as efficient Pb/Cd immobilizers for smelting site soils.

Application of Machine Learning in a Mineral Leaching Process-Taking Pyrolusite Leaching as an Example

In this study, several machine learning models were used to analyze the process variables of electric-field-enhanced pyrolusite leaching and predict the leaching rate of manganese, and the applicability of those models in the leaching process of hydrometallurgy was compared. It showed that there was no correlation between the six leaching conditions; in addition to the leaching time, the concentrations of sulfuric acid and ferrous sulfate had great influences on the leaching of pyrolusite. The results of the prediction models showed that the support vector regression model has the best prediction performance, with regression index (R ²) = 0.92 and mean square error = 25.04, followed by the gradient boosting regression model (R ² > 0.85). In this research, machine learning models were applied to the optimization of the manganese leaching process, and the research process and methods were also applicable to other hydrometallurgical processes for majorization and result prediction.