Modelling bioaccumulation of heavy metals in soil-crop ecosystems and identifying its controlling factors using machine learning

Predicting the efficiency of arsenic immobilization in soils by biochar using machine learning

Arsenic (As) pollution in soils is a pervasive environmental issue. Biochar immobilization offers a promising solution for addressing soil As contamination. The efficiency of biochar in immobilizing As in soils primarily hinges on the characteristics of both the soil and the biochar. However, the influence of a specific property on As immobilization varies among different studies, and the development and application of arsenic passivation materials based on biochar often rely on empirical knowledge. To enhance immobilization efficiency and reduce labor and time costs, a machine learning (ML) model was employed to predict As immobilization efficiency before biochar application. In this study, we collected a dataset comprising 182 data points on As immobilization efficiency from 17 publications to construct three ML models. The results demonstrated that the random forest (RF) model outperformed gradient boost regression tree and support vector regression models in predictive performance. Relative importance analysis and partial dependence plots based on the RF model were conducted to identify the most crucial factors influencing As immobilization. These findings highlighted the significant roles of biochar application time and biochar pH in As immobilization efficiency in soils. Furthermore, the study revealed that Fe-modified biochar exhibited a substantial improvement in As immobilization. These insights can facilitate targeted biochar property design and optimization of biochar application conditions to enhance As immobilization efficiency.

Perfluoroalkyl compounds in groundwater alter the spatial pattern of health risk in an arsenic‑cadmium contaminated region

Integrated health risk assessment strategies for emerging organic pollutants and heavy metals that coexist in water/soil media are lacking. Contents of perfluoroalkyl compounds and potentially toxic elements in multiple media were determined by investigating a county where a landfill and a tungsten mine coexist. The spatial characteristics and sources of contaminants were predicted by Geostatistics-based and multivariate statistical analysis, and their comprehensive health risks were assessed. The average contents of perfluorooctane acid, perfluorooctanesulfonic acid, arsenic, and cadmium in groundwater were 3.21, 0.77, 1.69, and 0.14 μg L-1, respectively; the maximum content of cadmium in soils and rice highly reached 2.12 and 1.52 mg kg-1, respectively. In soils, the contribution of mine lag to cadmium was 99 %, and fertilizer and pesticide to arsenic was 59.4 %. While in groundwater, arsenic, cadmium and perfluoroalkyl compounds near the landfill mainly came from leachate leakage. Significant correlations were found between arsenic in groundwater and arsenic and cadmium in soils, as well as perfluoroalkyl compounds in groundwater and pH and sulfate. Based on these correlations, the geographically optimal similarity model predicted high-level arsenic in groundwater near the tungsten mine and cadmium/perfluoroalkyl compounds around the landfill. The combination of analytic network process, entropy weighting method and game theory-based trade-off method with risk assessment model can assess the comprehensive risks of multiple pollutants. Using this approach, a high health-risk zone located around the landfill, which was mainly attributed to the presence of arsenic, cadmium and perfluorooctanesulfonic acid, was found. Overall, perfluoroalkyl compounds in groundwater altered the spatial pattern of health risks in an arsenic‑cadmium contaminated area.

Development and application of machine learning models for prediction of soil available cadmium based on soil properties and climate features

Identifying the key influencing factors in soil available cadmium (Cd) is crucial for preventing the Cd accumulation in the food chain. However, current experimental methods and traditional prediction models for assessing available Cd are time-consuming and ineffective. In this study, machine learning (ML) models were developed to investigate the intricate interactions among soil properties, climate features, and available Cd, aiming to identify the key influencing factors. The optimal model was obtained through a combination of stratified sampling, Bayesian optimization, and 10-fold cross-validation. It was further explained through the utilization of permutation feature importance, 2D partial dependence plot, and 3D interaction plot. The findings revealed that pH, surface pressure, sensible heat net flux and organic matter content significantly influenced the Cd accumulation in the soil. By utilizing historical soil surveys and climate change data from China, this study predicted the spatial distribution trend of available Cd in the Chinese region, highlighting the primary areas with heightened Cd activity. These areas were primarily located in the eastern, southern, central, and northeastern China. This study introduces a novel methodology for comprehending the process of available Cd accumulation in soil. Furthermore, it provides recommendations and directions for the remediation and control of soil Cd pollution.

Quantifying the impact of factors on soil available arsenic using machine learning

Arsenic (As) can accumulate in edible plant parts and thus pose a serious threat to human health. Identifying the contributions of various factors to soil available As is crucial for evaluating environmental risks. However, research quantitatively assessing the importance of soil properties on available As is scarce. In this study, we utilized 442 datasets covering total As, available As, and properties of farmland soils. The five machine learning models were employed to predict soil available As content, and the model with the best predictive performance was selected to calculate the importance of soil properties on available As and interpret the model results. The Random Forest model exhibited the best predictive performance, with R2 for the test set of dryland and paddy fields being 0.83 and 0.82 respectively, while also outperforming other machine learning models in terms of accuracy. Concurrently, evaluating the contribution of soil properties to soil available As revealed that increases in soil total arsenic, pH, organic matter (OM), and cation exchange capacity (CEC) led to higher soil available As content. Among these factors, soil total As had the greatest impact, followed by CEC. The influence of pH on soil available As was greater in dryland compared to OM, while in paddy fields, it was smaller than OM (p < 0.01). Sensitivity analysis results indicated that reducing soil total As content had the greatest effect on available As. In both dryland and paddy field soils, reducing soil total As had the most pronounced effect on available As, leading to reductions of 10.09% and 8.48%, respectively. Therefore, prioritizing the regulation of soil total As and CEC is crucial in As contamination management practices to alter As availability in farmland soils.

Predicting bioavailability of potentially toxic elements (PTEs) in sediment using various machine learning (ML) models: A case study in Mahabad Dam and River-Iran

Considering the significant impact of potentially toxic elements (PTEs) on the ecosystem and human health, this paper, investigated the contamination level of four PTEs (Zn, Cu, Mo and Pb) and their mobility in sediments of Mahabad dam and river. Choosing the most effective machine learning algorithms is very important in accurately predicting bioavailability of PTEs. Therefore, four machine learning (ML) models including decision tree regression (DTR), random forest regression (RFR), multi-layer perceptron regression (MLPR) and support vector regression (SVR), were used and compared for estimating the selected PTEs bioavailability. For these models, 9 variables (total concentration, pH, EC, OM and five chemical forms F1 to F5 obtained by sequential extraction) in 100 sediment samples were considered. The results showed that contamination level decreases from Zn and Cu to Pb and Mo, but the order of the mobility coefficient of the elements in the sediment follows the trend of zinc > copper > molybdenum > lead, and variation coefficient indicated more variability of spatial distribution for Zn and Cu. Among the four tested models, DTR and RFR performed the best for predicting PTEs bioavailability variations (with roc_auc>0.9, R2 > 0.8 and MSE>0.5), followed by MLPR and SVR. Furthermore, the relevance of the factors controlling the metals availability, evaluated using the RFR-based feature importance method and Pearson correlation, revealed that the most important physicochemical property for Zn, Cu and Mo bioavailability was pH, whereas for Pb, EC was the determinant factor. In the case of chemical speciation, F5 had an inverse correlation with the target, while F1 and F2 had a direct correlation. These fractions contributed significantly to the prediction results. This study represents the potential successful application of ML to PTEs risk control in sediments and early warning for the surrounding water PTEs contamination.

Machine learning-based prediction of cadmium pollution in topsoil and identification of critical driving factors in a mining area

Mining activities have resulted in a substantial accumulation of cadmium (Cd) in agricultural soils, particularly in southern China. Long-term Cd exposure can cause plant growth inhibition and various diseases. Rapid identification of the extent of soil Cd pollution and its driving factors are essential for soil management and risk assessment. However, traditional geostatistical methods are difficult to simulate the complex nonlinear relationships between soil Cd and potential features. In this study, sequential extraction and hotspot analyses indicated that Cd accumulation increased significantly near mining sites and exhibited high mobility. The concentration of Cd was estimated using three machine learning models based on 3169 topsoil samples, seven quantitative variables (soil pH, Fe, Ca, Mn, TOC, Al/Si and ba value) and three quantitative variables (soil parent rock, terrain and soil type). The random forest model achieved marginally better performance than the other models, with an R2 of 0.78. Importance analysis revealed that soil pH and Ca and Mn contents were the most significant factors affecting Cd accumulation and migration. Conversely, due to the essence of controlling Cd migration being soil property, soil type, terrain, and soil parent materials had little impact on the spatial distribution of soil Cd under the influence of mining activities. Our results provide a better understanding of the geochemical behavior of soil Cd in mining areas, which could be helpful for environmental management departments in controlling the diffusion of Cd pollution and capturing key targets for soil remediation.

A novel prediction approach driven by graph representation learning for heavy metal concentrations

The potential risk of heavy metals (HMs) to public health is an issue of great concern. Early prediction is an effective means to reduce the accumulation of HMs. The current prediction methods rarely take internal correlations between environmental factors into consideration, which negatively affects the accuracy of the prediction model and the interpretability of intrinsic mechanisms. Graph representation learning (GraRL) can simultaneously learn the attribute relationships between environmental factors and graph structural information. Herein, we developed the GraRL-HM method to predict the HM concentrations in soil-rice systems. The method consists of two modules, which are PeTPG and GCN-HM. In PeTPG, a graphic structure was generated using graph representation and communitization technology to explore the correlations and transmission paths of different environmental factors. Subsequently, the GCN-HM model based on the graph convolutional neural network (GCN) was used to predict the HM concentrations. The GraRL-HM method was validated by 2295 sets of data covering 21 environmental factors. The results indicated that the PeTPG model simplified correlation paths between factor nodes from 396 to 184, reducing by 53.5 % graph scale by eliminating the invalid paths. The concise and efficient graph structure enhanced the learning efficiency and representation accuracy of downstream prediction models. The GCN-HM model was superior to the four benchmark models in predicting the HM concentration in the crop, improving R2 by 36.1 %. This study develops a novel approach to improve the prediction accuracy of pollutant accumulation and provides valuable insights into intelligent regulation and planting guidance for heavy metal pollution control.

Optimization of polyamine and mycorrhiza in sorghum plant for removal of hazardous cadmium

Eco-friendly and sustainable practices must be followed while using the right plants and microbes to remove harmful heavy metals from the soil. The goal of the current study was to ascertain how effectively sorghum plants removed cadmium (Cd) from the soil using polyamines and mycorrhiza. Plant-biochemicals such as free amino acids, ascorbic acids, anthocyanin, proline, and catalase, APX, peroxidase activities were considered as markers in this study which revealed the adverse plant growth performance under 70 and 150 ppm of Cd concentration (w/w) after 30,60, and 90 days of treatment. The plants showed a mitigating effect against high Cd-concentration with exogenous use of mycorrhiza and putrescine. The treatment T17 (mycorrhiza +5 mM putrescine) showed a substantial decrease in the content of total free amino acid, ascorbic acid, catalase, APX, peroxidase by 228.36%, 39.79%, 59.06%, 182.79% 106.97%, respectively after 90 days as compared to T12 (150 ppm Cd). Anthocyanin content was negatively correlated (-0.503, -0.556, and -0.613) at p < 0.01 with other studied markers, with an increase by 10.52% in T17 treated plant as compared to T12. The concentration of Cd in root increased by 49.6% (141 ppm) and decreased in the shoot by 71% (17.8 ppm) in T17 treated plant as compared to T12 after 90 days. The application of mycorrhiza and putrescine significantly increased BCF (>1) and decreased TF (<1) for Cd translocation. The administration of mycorrhiza and putrescine boosted the Cd removal efficiency of sorghum plants, according to FTIR, XRD, and DSC analysis. As a result, this study demonstrates novel approaches for induced phytoremediation activity of plants via mycorrhiza and putrescine augmentation, which can be a promising option for efficient bioremediation in contaminated sites.

Evaluation of the prediction effectiveness for geochemical mapping using machine learning methods: A case study from northern Guangdong Province in China

This study compares seven machine learning models to investigate whether they improve the accuracy of geochemical mapping compared to ordinary kriging (OK). Arsenic is widely present in soil due to human activities and soil parent material, posing significant toxicity. Predicting the spatial distribution of elements in soil has become a current research hotspot. Lianzhou City in northern Guangdong Province, China, was chosen as the study area, collecting a total of 2908 surface soil samples from 0 to 20 cm depth. Seven machine learning models were chosen: Random Forest (RF), Support Vector Machine (SVM), Ridge Regression (Ridge), Gradient Boosting Decision Tree (GBDT), Artificial Neural Network (ANN), K-Nearest Neighbors (KNN), and Gaussian Process Regression (GPR). Exploring the advantages and disadvantages of machine learning and traditional geological statistical models in predicting the spatial distribution of heavy metal elements, this study also analyzes factors affecting the accuracy of element prediction. The two best-performing models in the original model, RF (R2 = 0.445) and GBDT (R2 = 0.414), did not outperform OK (R2 = 0.459) in terms of prediction accuracy. Ridge and GPR, the worst-performing methods, have R2 values of only 0.201 and 0.248, respectively. To improve the models' prediction accuracy, a spatial regionalized (SR) covariate index was added. Improvements varied among different methods, with RF and GBDT increasing their R2 values from 0.4 to 0.78 after enhancement. In contrast, the GPR model showed the least significant improvement, with its R2 value only reaching 0.25 in the improved method. This study concluded that choosing the right machine learning model and considering factors that influence prediction accuracy, such as regional variations, the number of sampling points, and their distribution, are crucial for ensuring the accuracy of predictions. This provides valuable insights for future research in this area.

Controlling factors of heavy metal(loid) accumulation in rice: Main and interactive effects

Identifying the key determinants of heavy metal(loid) accumulation in rice and quantifying their contributions are critical for precise prediction of heavy metal(loid) concentrations in rice and the formulation of effective pollution control strategies. The accumulation of heavy metal(loid)s in rice can be influenced by both natural and anthropogenic factors, which may interact with each other. However, distinguishing the independent roles (main effects) from interactive effects and quantifying their impacts separately pose challenges. To address this knowledge gap, we employed TreeExplainer-based SHAP and random forest algorithms in this study to quantitatively estimate the primary influencing factors and their main and interactive effects on heavy metal(loid)s in rice. Our findings reveal that soil cadmium (SCd) and rice cultivation time (C_TIME) were the primary contributors to rice cadmium (RCd) and rice arsenic (RAs), respectively. Soil lead (SPb) and sampling distances from roads significantly contributed to rice lead (RPb). Additionally, we identified significant interactive effects of SCd and C_TIME, C_TIME and RCd, and RCd and rice variety on RCd, RAs, and RPb, respectively, emphasizing their significance. These insights are pivotal in improving the accuracy of heavy metal(loid) concentration predictions in rice and offering theoretical guidance for the formulation of pollution control measures.

Contaminated soil remediation with nano-FeS loaded lignin hydrogel: A novel strategy to produce safe rice grains while reducing cadmium in paddy field

Cadmium (Cd) contamination in agricultural soil has been an elevated concern due to the high health risks associated with the transfer through the soil-food chain, particularly in the case of rice. Recently, there has numerous researches on the use of nanoparticle-loaded materials for heavy metal-polluted soil remediation, resulting in favorable outcomes. However, there has been limited research focus on the field-scale application and recovery. This study was aimed to validate the Cd reduction effect of the nano-FeS loaded lignin hydrogel composites (FHC) in mildly polluted paddies, and to propose a field-scale application method. Hence, a multi-site field experiment was conducted in southern China. After the application for 94-103 days, the FHC exhibited a high integrity and elasticity, with a recovery rate of 91.90%. The single-round remediation led to decreases of 0.42-31.72% in soil Cd content and 1.52-49.11% in grain Cd content. Additionally, this remediation technique did not adversely impact rice production. Consequently, applying FHC in the field was demonstrated to be an innovative, efficient, and promising remediation technology. Simultaneously, a strategy was proposed for reducing Cd levels while cultivating rice in mildly polluted fields using the FHC.

Spatial distribution prediction of soil heavy metals based on sparse sampling and multi-source environmental data

Predicting the precise spatial distribution of heavy metals in soil is crucial, especially in the fields of environmental management and remediation. However, achieving accurate spatial predictions of soil heavy metals becomes quite challenging when the number of soil sampling points is relatively limited. To address this challenge, this study proposes a hybrid approach, namely, Light Gradient Boosting Machine plus Ordinary Kriging (LGBK), for predicting the spatial distribution of soil heavy metals. A total of 137 soil samples were collected from the Shengli Coal-mine Base in Inner Mongolia, China, and their heavy metal concentrations were measured. Leveraging environmental covariates and soil heavy metal data, we constructed the predictive model. Experimental results demonstrate that, in comparison to traditional models, LGBK exhibits superior predictive performance. For copper (Cu), zinc (Zn), chromium (Cr), and arsenic (As), the coefficients of determination (R²) from the cross-validation results are 0.65, 0.52, 0.57, and 0.63, respectively. Moreover, the LGBK model excels in capturing intricate spatial features in heavy metal distribution. It accurately forecasts trends in heavy metal distribution that closely align with actual measurements. ENVIRONMENTAL IMPLICATION: This study introduces a novel method, LGBK, for predicting the spatial distribution of soil heavy metals. This method yields higher-precision predictions even with a limited number of sampling points. Furthermore, the study analyzes the spatial distribution characteristics of Cu, Zn, Cr, and As in the grassland coal-mine base, along with the key environmental factors influencing their spatial distribution. This research holds significant importance for the environmental regulation and remediation of heavy metal pollution.

Heavy metals prediction in coastal marine sediments using hybridized machine learning models with metaheuristic optimization algorithm

This study proposes different standalone models viz: Elman neural network (ENN), Boosted Tree algorithm (BTA), and f relevance vector machine (RVM) for modeling arsenic (As (mg/kg)) and zinc (Zn (mg/kg)) in marine sediments owing to anthropogenic activities. A heuristic algorithm based on the potential of RVM and a flower pollination algorithm (RVM-FPA) was developed to improve the prediction performance. Several evaluation indicators and graphical methods coupled with visualized cumulative probability function (CDF) were used to evaluate the accuracy of the models. Akaike (AIC) and Schwarz (SCI) information criteria based on Dickey-Fuller (ADF) and Philip Perron (PP) tests were introduced to check the reliability and stationarity of the data. The prediction performance in the verification phase indicated that RVM-M2 (PBAIS = -o.0465, MAE = 0.0335) and ENN-M2 (PBAIS = 0.0043, MAE = 0.0322) emerged as the best model for As (mg/kg) and Zn (mg/kg), respectively. In contrast with the standalone approaches, the simulated hybrid RVM-FPA proved merit and the most reliable, with a 5 % and 18 % predictive increase for As (mg/kg) and Zn (mg/kg), respectively. The study's findings validated the potential for estimating complex HMs through intelligent data-driven models and heuristic optimization. The study also generated valuable insights that can inform the decision-makers and stockholders for environmental management strategies.

Transcriptomics, proteomics, and metabolomics interventions prompt crop improvement against metal(loid) toxicity

The escalating challenges posed by metal(loid) toxicity in agricultural ecosystems, exacerbated by rapid climate change and anthropogenic pressures, demand urgent attention. Soil contamination is a critical issue because it significantly impacts crop productivity. The widespread threat of metal(loid) toxicity can jeopardize global food security due to contaminated food supplies and pose environmental risks, contributing to soil and water pollution and thus impacting the whole ecosystem. In this context, plants have evolved complex mechanisms to combat metal(loid) stress. Amid the array of innovative approaches, omics, notably transcriptomics, proteomics, and metabolomics, have emerged as transformative tools, shedding light on the genes, proteins, and key metabolites involved in metal(loid) stress responses and tolerance mechanisms. These identified candidates hold promise for developing high-yielding crops with desirable agronomic traits. Computational biology tools like bioinformatics, biological databases, and analytical pipelines support these omics approaches by harnessing diverse information and facilitating the mapping of genotype-to-phenotype relationships under stress conditions. This review explores: (1) the multifaceted strategies that plants use to adapt to metal(loid) toxicity in their environment; (2) the latest findings in metal(loid)-mediated transcriptomics, proteomics, and metabolomics studies across various plant species; (3) the integration of omics data with artificial intelligence and high-throughput phenotyping; (4) the latest bioinformatics databases, tools and pipelines for single and/or multi-omics data integration; (5) the latest insights into stress adaptations and tolerance mechanisms for future outlooks; and (6) the capacity of omics advances for creating sustainable and resilient crop plants that can thrive in metal(loid)-contaminated environments.

Effects of selenium on biogeochemical cycles of cadmium in rice from flooded paddy soil systems in the alluvial Indus Valley of Pakistan

This study delves into the pollution status, assesses the effects of Se on Cd biogeochemical pathways, and explores their interactions in nutrient-rich paddy soil-rice ecosystems through 500 soil-rice samples in Pakistan. The results showed that 99.6 % and 12.8 % of soil samples exceeded the World Health Organization (WHO) allowable Se and Cd levels (7 and 0.35 mg/kg). In comparison, 23 % and 6 % of the grain samples exceeded WHO's allowable Se and Cd levels (0.3 and 0.2 mg/kg), respectively. Geographically Weighted Regression (GWR) model results further revealed spatial nonstationarity, confirming diverse associations between dependent variables (Se and Cd in rice grain) and independent variables from paddy soil and plant tissues (root and shoot), such as Soil Organic Matter (SOM), pH, Se, and Cd concentrations. High Se:Cd molar ratios (>1) and a negative correlation (r = -0.16, p < 0.01) between the Cd translocation factor (Cd in rice grain/Cd in root) and Se in roots suggest that increased root Se levels inhibit the transfer of Cd from roots to grains. The inverse correlation between Se and Cd in paddy grains was further characterized as Se deficiency, no risk, high Cd risk, Se risk, Cd risk, and Se-Cd co-exposure risk. There was no apparent risk for human co-consumption in 42.6 % of grain samples with moderate Se and low Cd. The remaining categories indicate differing degrees of risk. In the study area, 31 % and 20 % of grain samples with low Se and Cd indicate Se deficiency and risk, respectively. High Se and low Cd levels in rice samples suggest a potential hazard for severe Se exposure due to frequent rice consumption. This study not only systematically evaluates the pollution status of paddy-soil systems in Pakistan but also provides a reference to thoroughly contemplate the development of a scientific approach for evaluating human risks and the potential dangers associated with paddy soils and rice, specifically in regions characterized by low Se and low Cd concentrations, as well as those with moderate Se and high Cd concentrations. SYNOPSIS: This study is significant for understanding the effects of Se on Cd geochemical cycles and their interactions in paddy soil systems in Pakistan.

Using machine learning to predict selenium and cadmium contents in rice grains from black shale-distributed farmland area

Cadmium (Cd) and selenium (Se) are widely enriched in soil at black shale outcropping areas, with Cd levels exceeding the standard (2.0 mg/kg in 5.5 < pH ≤ 6.5) commonly. The prevention of Cd hazards and the safe development of Se-rich land resources are key issues that need to be urgently addressed. To ensure safe utilization of Se-rich land in the CdSe coexisting areas, 158 rice samples, their corresponding rhizosphere soils, and 8069 topsoil samples were collected and tested in the paddy fields of Ankang City, Shaanxi Province, where black shales are widely exposed. The results showed that 43 % of the topsoil samples were Se-rich soil (Se > 0.4 mg/kg) wherein 79 % and 3 % of Cd concentrations exceeded the screening value and control value, respectively, according to the GB15618-2018 standard. Meanwhile, 63 % of the rice samples were Se rich (Se > 0.04 mg/kg) and the Cd content exceeded the prescribed limit (0.2 mg/kg) in Se-rich rice by 26 %. There was no significant positive correlation between the Se and Cd contents in the rice grains and the Se and Cd contents in the corresponding rhizosphere soil. The factors influencing Se and Cd uptake in rice were SiO2, CaO, P, S, pH, and TFe2O3. Accordingly, an artificial neural network (ANN) and multiple linear regression model (MLR) were used to predict Cd and Se bioaccumulation in rice grains. The stability and accuracy of the ANN model were better than those of the MLR model. Based on survey data and the prediction results of the ANN model, a safe planting zoning of Se-rich rice was proposed, which provided a reference for the scientific planning of land resources.

Unintended mitigation effect of air pollutant regulation on the aquatic cadmium: Evidence from the 11-FYPEP in China

This paper evaluates the unintended mitigation effect of air pollutant regulation on aquatic cadmium (Cd) emissions in the China's Eleventh Five-Year Plan for Environmental Protection (11-FYPEP), by employing a continuous Difference-in-Difference-in-Difference (DDD) estimator. We find that: (1) Although the 11-FYPEP did not target to reduce Cd emission, the implementation of 11-FYPEP reduced the emissions by 2.8 %. (2) The Cd emission is closely related to the industrial level, because the reduction of Cd is 6.1 % higher in areas with lower industrial output, and the mediating effect of the number of industrial enterprises accounts for 6.8 % of the Cd reduction. Based on our findings, implications like improving production efficiency and modifying industrial structure are proposed, as the 11-FYPEP achieves Cd reduction in an unsustainable way.

Spatial and variety distributions, risk assessment, and prediction model for heavy metals in rice grains in China

In this study, 6229 brown rice grains from three major rice-producing regions were collected to investigate the spatial and variety distributions of heavy metals in rice grains in China. The potential sources of heavy metals in rice grains were identified using the Pearson correlation matrix and principal component analysis, and the health risks of dietary exposure to heavy metals via rice consumption were assessed using the hazard index (HI) and total carcinogenic risk (TCR) method, respectively. Moreover, 48 paired soil and rice samples from 11 cities were collected to construct a predicting model for Cd accumulation in rice grains using the multiple linear stepwise regression analysis. The results indicated that Cd and Ni were the main heavy metal pollutants in rice grains in China, with approximately 10% of samples exceeding their corresponding maximum allowable limits. The Yangtze River basin had heavier pollution of heavy metals than the Southeast Coastal Region and Northeast Plain, and the indica rice varieties had higher heavy metal accumulation abilities compared with the japonica rice. The Cu, Pb, and Cd mainly originated from anthropogenic sources, while As, Hg, Cr, and Ni originated from both natural and anthropogenic sources. The mean HI and TCR values of dietary exposure to heavy metals via rice consumption ranged from 2.92 to 4.31 and 9.74 × 10-3 to 1.44 × 10-2, respectively, much higher than the acceptable range, and As and Ni were the main contributor to the HI and TCR for Chinese adults and children, respectively. The available Si (ASi), total Cd (TCd), available Mo (AMo), and available S (AS) were the main soil factors determining grain Cd accumulation. A multiple linear stepwise regression model was constructed based on ASi, TCd, AMo, and AS in soils with good accuracy and precision, which could be applied to predict Cd accumulation in rice grains and guide safe rice production in contaminated paddy fields.

Influencing factors of selenium transformation in a soil-rice system and prediction of selenium content in rice seeds: a case study in Ninghua County, Fujian Province

Selenium (Se) is an essential element for human and animal health and has antioxidant, anticancer, and antiviral effects. However, more than 100 million people in China do not have enough Se in their diets, resulting in a state of low Se in the human body. Since the absorption of Se by crop seeds depends not only on the Se content in soil, there are many omissions and misjudgments in the division of Se-rich producing areas. Soil pH, total iron oxide content (TFe2O3), soil organic matter (SOM), and P and S contents were the main factors affecting Se migration and transformation in the soil-rice system. In this study, we compared the performance of the back propagation neural network (BP network) and multiple linear regression (MLR) using 177 pairs of soil-rice samples. Our results showed that the BP network had higher accuracy than MLR. The accuracy and precision of the prediction data met the requirements, and the prediction data were reliable. Based on the Se data of surface paddy fields, 26,900 ha of Se-rich rice planting area was planned using this model, accounting for 77% of the paddy field area. In the planned Se-rich area for rice, the proportion of soil Se content greater than 0.4 mg·kg-1 was only 5.29%. Our research is of great significance for the development of Se-rich lands.

A novel framework for improving soil organic matter prediction accuracy in cropland by integrating soil, vegetation and human activity information

Remote sensing is an important tool for monitoring soil information. However, accurate spatial modeling of soil organic matter (SOM) in areas with high vegetation coverage, typically represented by agroecosystems, remains a challenge for field-scale estimation using remote sensing. To date, studies have focused on using single-period or multi-temporal vegetation information to characterize SOM. Thus, the relationship between SOM content and time-series vegetation biomass has not yet been fully explored. In addition, most studies have ignored the effects of critical soil properties and human activities (e.g., soil salinization, soil particle size fractions, history of land-use changes) on SOM. By integrating information on vegetation, soil, and human activities, we propose a novel framework for assessing SOM in cotton fields of artificial oases in northwest China, where returned straw is one of the primary sources of SOM coming from vegetation. We developed an Annual Maximum Biomass Accumulation Index (AMBAI) using time-series Landsat images from 1990 to 2019. Subsequently, we quantified the information of the planting years (PY) of cropland using spectral index threshold and incorporated proximal sensing data (soil hyperspectral and apparent conductivity data) and soil particle size fractions to establish a predictive model of SOM using partial least squares regression (PLSR), random forest (RF), and convolutional neural network (CNN). The results revealed that AMBAI had the highest correlation coefficient (r) with SOM (0.76, P < 0.01). AMBAI, soil hyperspectral data, and PY were the most relevant predictors for estimating SOM. The CNN model integrating vegetation, soil, and human activity information performed best, with coefficient of determination (R2), relative analysis error (RPD), and root mean square error (RMSE) of 0.83, 2.38 and 1.38 g kg-1, respectively. This study confirmed that AMBAI and PY had great potential for characterizing SOM in arid and semi-arid regions, providing a reference for other relevant studies.

Simulation and prediction for the spatial heterogeneity of soil selenium bioavailability at different stratigraphic scales

Stratigraphic lithology strongly influences the spatial heterogeneity of soil available selenium (ASe), however, it is often neglected in regional simulation. Therefore, taking the Jiangjin District, where the soil is richer in selenium (Se), as the research area, the changes of soil ASe at different spatial scales have been simulated by combining Geodetector and three popular models (Multiple linear regression (MLR), Random forest (RF) and BP neural network (BPN)). The results showed that modelling with 'Formation' as the spatial scale could reduce the influence of stratum lithology difference on the spatial heterogeneity of soil ASe and improve the model's prediction accuracy. Compared with the MLR (R2 = 0.52, root mean squares error (RMSE) = 13.217 μg kg-1) and BPN (R2 = 0.55, RMSE = 13.79 μg kg-1), the RF (R2 = 0.67, RMSE = 10.85 μg kg-1) exhibited higher R2 and smaller RMSE, and the simulation effect of soil ASe is the best in the Middle Jurassic Shaximiao Formation (J2s). The outcomes of variable importance analysis revealed that soil total selenium (TSe) and soil organic matter (SOM) were the imperative factors for predicting ASe. The scenario simulation prediction showed that in the next 40 years, due to the combined influence of SOM and pH, the content of ASe in soil developed in the J2s would decrease from 40.8 μg kg-1 to 37.8 μg kg-1, a 7.8 percent drop. The main areas of soil ASe loss were in the western farming areas. The ASe content in dry land and paddy fields decreased by 12.0% and 4.9%, respectively. Therefore, long-term agricultural production activities would lead to soil ASe loss. The present results could provide a new scheme for the simulation and prediction of regional soil ASe, which is helpful for scientific planning, utilization of selenium-rich soil resources, and development of regional agricultural economy.

Prediction of cadmium contents in rice grains from Quaternary sediment-distributed farmland using field investigations and machine learning

The Quaternary sediment-distributed regions of South China are suitable for rice cultivation, which is crucial for ensuring food security. Spatial correlations between soil cadmium (Cd) and rice Cd contents are generally poor, making the evaluation of rice quality and associated health risks challenging. In this study, we developed machine learning algorithms for predicting rice Cd contents using 654 data pairs of soil-rice samples collected in Guangxi province, China. After a comprehensive comparison, our results showed that the random forest (RF) had the better performance than artificial neural network (ANN) based on all the data (RMSEtesting 0.066 vs. 0.099 and R2testing 0.860 vs. 0.688). The feature importance analysis showed that soil CaO, Cd, elevation, and rainfall were the four most important features affecting the rice Cd contents in the study area. Using the established RF-predicated model, the rice Cd contents were predicted at the provincial level with an additional dataset of 1176 paddy soil samples. The prediction result revealed about 23 % of farmland cultivated rice with Cd content over 0.2 mg kg-1 in the study area. Therefore, it is recommended to implement strict measures by local agricultural departments to reduce rice Cd contents and ensure food safety in these areas. Our study provides valuable insights into the prediction of rice Cd contents, thus contributing to ensuring food safety and preventing Cd exposure-associated health risks.

Ensemble learning-based applied research on heavy metals prediction in a soil-rice system

Accurate prediction of heavy metal accumulation in soil ecosystems is crucial for maintaining healthy soil environments and ensuring high-quality agricultural products, as well as a challenging scientific task. In this study, we constructed a dataset containing 490 sets of multidimensional environmental covariate data and proposed prediction models for heavy metal concentrations (HMC) in a soil-rice system, EL-HMC (including RF-HMC and GBM-HMC), based on Random Forest (RF) and Gradient Boosting Machine (GBM) ensemble learning (EL) techniques. To reasonably evaluate the effectiveness of each model, Multiple linear and Bayesian regressions were selected as benchmark models (BM), and mean absolute error (MAE), root mean square error (RMSE), and determination coefficient R2 were selected as evaluation indicators. In addition, sensitivity and spatial autocorrelation (SAC) analyses were used to examine the robustness of the model. The results showed that the R2 values of RF-HMC and GBM-HMC for modeling available cadmium (Cd) concentrations in soil were 0.654 and 0.690, respectively, with an average increase of 48.0 % compared to the BMs. The R2 values of RF-HMC and GBM-HMC for predicting Cd, lead (Pb), chromium (Cr), and mercury (Hg) concentrations in rice ranged from 0.618 to 0.824 and 0.645 to 0.850, respectively, with an average increase of 58.2 % compared with the BMs. The corresponding MAEs and RMSEs of RF-HMC and GBM-HMC had low error levels. Sensitivity analysis of the input features and the SAC of the prediction bias showed that the EL-HMC models have excellent robustness. Therefore, the EL technology-based prediction models for HMCs proposed herein are practical and feasible, demonstrating better accuracy and stability than the traditional model. This study verifies the application potential of EL technology in pollution ecology and provides a new perspective and solution for sustainable management and precise prevention of heavy metal pollution in farmland soil at the regional scale.

Health risk assessment of trace metal(loid)s in agricultural soil using an integrated model combining soil-related and plants-accumulation exposures: A case study on Hainan Island, South China

Traditional health risk assessment of trace metal(loid)s (TMs) in agricultural soil exclusively considers direct soil-related exposure and may underestimate the health risks they pose. In this study, the health risks of TMs were evaluated using an integrated model that combined soil-related and plant-accumulating exposures. A detailed investigation of common TMs (Cr, Pb, Cd, As, and Hg) coupled with probability risk analysis based on a Monte Carlo simulation was conducted on Hainan Island. Our results showed that, except for As, the non-carcinogenic risk (HI) and carcinogenic risk (CR) of the TMs were all within the acceptable ranges (HI < 1.0, and CR < 1E-06) for direct soil-related exposure to bio-accessible fractions and indirect exposure via plant accumulation (CR substantially lower than the warning threshold 1E-04). We identified crop food ingestion as the essential pathway for TM exposure and As as the critical toxic element in terms of risk control. Moreover, we determined that RfDo and SFo are the most suitable parameters for assessing As health risk severity. Our study demonstrated that the proposed integrated model combining soil-related and plant-accumulating exposures can avoid major health risk assessment deviations. The results obtained and the integrated model proposed in this study can facilitate future multi-pathway exposure research and could be the basis for determining agricultural soil quality criteria in tropical areas.

Bacterial seed endophyte and abiotic factors influence cadmium accumulation in rice (Oryza sativa) along the Yangtze River area

Cadmium (Cd) contamination in rice (Oryza sativa) is particularly problematic due to its high risk to human health. Investigating the hidden roles of seed endophytes of rice in influencing Cd accumulation is essential to comprehensively understand the effects of biotic and abiotic factors to food security. Here, the content of Cd in soils and rice (Huanghuazhan) seeds from 19 sites along the Yangtze River exhibited considerable differences. From a biotic perspective, we observed the dominant endophytic bacteria, Stenotrophomonas (7.25 %), contribute to Cd control of rice (below 0.2 mg kg-1). Partial Least Squares (PLS) analysis further suggested that Enterobacteriaceae (15.48 %), altitude and pH were found to be the strong variables that might reduce the Cd uptake of rice. In contrast, Cytophagaceae (0.58 %), latitude and mean annual air pressure had the opposite effect. In pot experiments, after respectively inoculating the isolated endophytic bacteria Stenotrophomonas T4 and Enterobacter R1, N1 (f_Enterobacteriaceae), the Cd contents in shoot decreased by 47.6 %, 21.9 % and 33.0 % compared to controls. The distribution of Cd resistant genes (e.g., czcABC, nccAB, cznA) of Stenotrophomonas, Enterobacteriaceaea and Cytophagaceae further suggested their distinct manners in influencing the Cd uptake of rice. Overall, this study provides new insights into the food security threatened by globally widespread Cd pollution.

A novel AI-based approach for modelling the fate, transportation and prediction of chromium in rivers and agricultural crops: A case study in Iran

Chromium (Cr) pollution caused by the discharge of industrial wastewater into rivers poses a significant threat to the environment, aquatic and human life, as well as agricultural crops irrigated by these rivers. This paper employs artificial intelligence (AI) to introduce a new framework for modeling the fate, transport, and estimation of Cr from its point of discharge into the river until it is absorbed by agricultural products. The framework is demonstrated through its application to the case study River, which serves as the primary water resource for tomato production irrigation in Mashhad city, Iran. Measurements of Cr concentration are taken at three different river depths and in tomato leaves from agricultural lands irrigated by the river, allowing for the identification of bioaccumulation effects. By employing boundary conditions and smart algorithms, various aspects of control systems are evaluated. The concentration of Cr in crops exhibits an accumulative trend, reaching up to 1.29 µg/g by the time of harvest. Using data collected from the case study and exploring different scenarios, AI models are developed to estimate the Cr concentration in tomato leaves. The tested AI models include linear regression (LR), neural network (NN) classifier, and NN regressor, yielding goodness-of-fit values (R2) of 0.931, 0.874, and 0.946, respectively. These results indicate that the NN regressor is the most accurate model, followed by the LR, for estimating Cr levels in tomato leaves.

Using multi-technology to characterize transboundary Hg pollution in the largest presently active Hg deposit in China

Active Hg mines are primary sources of Hg contamination in the environment of mining districts and surrounding areas. Alleviation of Hg pollution requires knowledge of pollution sources, migration, and transform pathways across various environmental media. Accordingly, the Xunyang Hg-Sb mine, the largest active Hg deposit in China, presently was selected as the study area. GIS, TIMA, EPMA, μ-XRF, TEM-EDS, and Hg stable isotopes were adopted to investigate the spatial distribution, mineralogical characteristics, in situ microanalysis, and pollution sources of Hg in the environment medium at the macro- and micro-levels. The total Hg concentration in samples showed a regional distribution, with higher levels in areas close to the mining operations. The in situ distribution of Hg in soil was mainly associated with the mineralogical phases of quartz, and Hg was also correlated with Sb and S. Hg was also found to be rich mainly in quartz minerals in the sediment and showed different distributions of Sb. Hg hotspots had S abundances and contained no Sb and O. The contributions from the anthropogenic sources to soil Hg were estimated to be 55.35%, among which 45.97% from unroasted Hg ore and 9.38% from tailing. Natural input of soil Hg due to pedogenic processes accounted for 44.65%. Hg in corn grain was mainly derived from the atmosphere. This study will provide a scientific basis for assessing the current environmental quality in this area and minimizing further impacts that affect the nearby environmental medium.

Source apportionment and transfer characteristics of Pb in a soil-rice-human system, Jiulong River Basin, southeast China

The source apportionment and transfer of Pb in a paddy soil-rice-human system within the Jiulong River Basin in southeast China was investigated by analyzing (1) the chemical fractionation of Pb in paddy soils using a modified BCR four-step sequential extraction procedure, and (2) the bioaccessibility of Pb in both paddy soils and rice grains using a Simple Bioaccessibility Extraction Test method. In addition, a qualitative Pb isotopic model was used in combination with IsoSource software to quantify the contribution of potential Pb sources. The results show the enrichment of Pb in agro-ecosystems in the Jiulong River Basin. Contaminant Pb in paddy soils was mainly present in the reducible (42.9%) and the residual fractions (27.1%). The average bioaccessibility of Pb in rice grains was significantly higher than that in paddy soil, with values of 77.85% and 37.44%, respectively. Lead in paddy soils was primarily derived from agricultural (35.3%), natural (25.5%), industrial (24.5%) and coal combustion sources (14.7%), while Pb in rice grains was primarily derived from coal combustion (54.1%), agricultural (35.1%), industrial (6.0%) and natural sources (4.8%). The bioaccessible Pb was mainly derived from anthropogenic sources [agricultural (42.3% for soil and 25.3% for grain) and coal combustion sources (25.3% for soil and 59.3% for grain)]. Lead isotopic ratios are an effective tracer of Pb transfer from potential sources to rice plants and within the rice plants. Rice plants absorb Pb from the soil and the atmosphere through the roots and leaves, respectively. Most of the Pb was accumulated in roots. The integrated use of chemical fractionation, bioaccessibility and Pb isotopic data provides an effective method to study the source apportionment and transfer characteristics of Pb in paddy soil-rice-human systems.

Rapid identification of high and low cadmium (Cd) accumulating rice cultivars using machine learning models with molecular markers and soil Cd levels as input data

Excessive accumulation of cadmium (Cd) in rice grains threatens food safety and human health. Growing low Cd accumulating rice cultivars is an effective approach to produce low-Cd rice. However, field screening of low-Cd rice cultivars is laborious, time-consuming, and subjected to the influence of environment × genotype interactions. In the present study, we investigated whether machine learning-based methods incorporating genotype and soil Cd concentration can identify high and low-Cd accumulating rice cultivars. One hundred and sixty-seven locally adapted high-yielding rice cultivars were grown in three fields with different soil Cd levels and genotyped using four molecular markers related to grain Cd accumulation. We identified sixteen cultivars as stable low-Cd accumulators with grain Cd concentrations below the 0.2 mg kg^-1 food safety limit in all three paddy fields. In addition, we developed eight machine learning-based models to predict low- and high-Cd accumulating rice cultivars with genotypes and soil Cd levels as input data. The optimized model classifies low- or high-Cd cultivars (i.e., the grain Cd concentration below or above 0.2 mg kg^-1) with an overall accuracy of 76%. These results indicate that machine learning-based classification models constructed with molecular markers and soil Cd levels can quickly and accurately identify the high- and low-Cd accumulating rice cultivars.

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.

Phytoremediation as an Effective Remedy for Removing Trace Elements from Ecosystems

The pollution of soil by trace elements is a global problem. Conventional methods of soil remediation are often inapplicable, so it is necessary to search intensively for innovative and environment-friendly techniques for cleaning up ecosystems, such as phytoremediation. Basic research methods, their strengths and weaknesses, and the effects of microorganisms on metallophytes and plant endophytes resistant to trace elements (TEs) were summarised and described in this manuscript. Prospectively, bio-combined phytoremediation with microorganisms appears to be an ideal, economically viable and environmentally sound solution. The novelty of the work is the description of the potential of "green roofs" to contribute to the capture and accumulation of many metal-bearing and suspended dust and other toxic compounds resulting from anthropopressure. Attention was drawn to the great potential of using phytoremediation on less contaminated soils located along traffic routes and urban parks and green spaces. It also focused on the supportive treatments for phytoremediation using genetic engineering, sorbents, phytohormones, microbiota, microalgae or nanoparticles and highlighted the important role of energy crops in phytoremediation. Perceptions of phytoremediation on different continents are also presented, and new international perspectives are presented. Further development of phytoremediation requires much more funding and increased interdisciplinary research in this direction.

Interactions and quantification of multiple influencing factors on cadmium accumulation in soil-rice systems at a large region

The accumulation of Cd in soil-rice systems at a large region is often extremely complicated due to environmental heterogeneity and the interactions of multiple influencing factors. However, the interactive effects and quantification of the contributions of influencing factors on Cd accumulation in large regions remain unclear. In this study, conditional inference trees and random forest analysis were used to identify the interactions of various factors (soil properties, topography and demographic-economic), and quantify their contributions to Cd accumulation in soil-rice systems of Sichuan-Chongqing region, China. The results showed that Cd content in the soil was the most significant influencing factor on Cd accumulation in soil-rice systems, especially bioavailable Cd in soil contributed to 35.73 % and 54.78 % for soil total Cd (Cd_soil) and brown rice Cd (Cd_rice), respectively. Population density (PD) and elevation contributed 31.16 % and 27.40 % to Cd_soil content, respectively, and their interaction promoted the increase in Cd_soil content. Moreover, PD played a leading role in Cd_soil content when the elevation exceeded 324 m. The relative importances of slope and elevation for Cd_rice content were 16.81 % and 8.49 %, respectively, and their interaction facilitated the increment of Cd_rice content. As soil pH, gross domestic product (GDP) and slope decreased, the interaction of soil pH with GDP led to the increase of bioavailability factor (BAF), and that with slope enhanced the bioaccumulation factor (BCF). In addition, soil pH, PD and elevation were of considerable importance for the migration and transformation of Cd, with contributions of 22.11 %, 12.90 % and 12.52 % to BAF, and 5.05 %, 5.62 % and 5.50 % to BCF, respectively. This study is hopeful to provide a scientific insight into the prevention and control of Cd contamination in soil-rice systems at a large region.

Concentrations and bioconcentration factors of leaf microelements in response to environmental gradients in drylands of China

Determining response patterns of plant leaf elements to environmental variables would be beneficial in understanding plant adaptive strategies and in predicting ecosystem biogeochemistry processes. Despite the vital role of microelements in life chemistry and ecosystem functioning, little is known about how plant microelement concentrations, especially their bioconcentration factors (BCFs, the ratio of plant to soil concentration of elements), respond to large-scale environmental gradients, such as aridity, soil properties and anthropogenic activities, in drylands. The aim of the present study was to fill this important gap. We determined leaf microelement BCFs by measuring the concentrations of Mn, Fe, Ni, Cu and Zn in soils from 33 sites and leaves of 111 plants from 67 species across the drylands of China. Leaf microelement concentrations were maintained within normal ranges to satisfy the basic requirements of plants, even in nutrient-poor soil. Aridity, soil organic carbon (SOC) and electrical conductivity (EC) had positive effects, while soil pH had a negative effect on leaf microelement concentrations. Except for Fe, aridity affected leaf microelement BCFs negatively and indirectly by increasing soil pH and SOC. Anthropogenic activities and soil clay contents had relatively weak impacts on both leaf microelement concentrations and BCFs. Moreover, leaf microelement concentrations and BCFs shifted with thresholds at 0.89 for aridity and 7.9 and 8.9 for soil pH. Woody plants were positive indicator species and herbaceous plants were mainly negative indicator species of leaf microelement concentrations and BCFs for aridity and soil pH. Our results suggest that increased aridity limits the absorption of microelements by plant leaves and enhances leaf microelement concentrations. The identification of indicator species for the response of plant microelements to aridity and key soil characteristics revealed that woody species in drylands were more tolerant to environmental changes than herbaceous species.

Convolutional graph neural networks-based research on estimating heavy metal concentrations in a soil-rice system

Estimating heavy metal concentrations in soil-rice systems is of great significance to identify the factors controlling heavy metal transfer in soil-crop ecosystems. Recent research utilizes the advantage of convolutional calculations to extract and learn complicated information from 17 environmental covariates in rice and achieve promising results. However, as the complexity and interconnectivity in soil-crop ecosystem, just relying on convolutional calculations and a deep network structure is far from enough. The data processed by traditional deep learning technologies even with convolutional calculations are limited to Euclidean space; these architectures do not have the ability to extract information from the relationships in graph structures, which may contain rich information. Thus, in this paper, we try to integrate graph information into convolutional calculations for heavy metal prediction and propose a model named ConvGNN-HM. ConvGNN-HM combines the advantages of graph learning and convolutional calculations to predict heavy metal concentrations in a soil-rice system with analysis of 17 environmental factors. For comparison, we conduct an experiment to compare ConvGNN-HM with techniques with convolutional neural networks, multilayer perceptron, back-propagation neural networks, support vector machine, random forest, Bayesian ridge regression, and multiple linear regression. The experimental results illustrate that ConvGNN-HM got the best prediction values; the R² values of ConvGNN-HM for cadmium (Cd), plumbum (Pb), chromium (Cr), arsenic (As), and hydrargyrum (Hg) in rice were 0.84, 0.75, 0.79, 0.49, and 0.83, respectively, and the MAE values were also acceptable. We further conduct sensitivity analysis to demonstrate the stability and robustness of ConvGNN-HM. This study demonstrates the usefulness of combining graph learning and convolutional calculations in the prediction of heavy metal concentrations and provides a new perspective to build multidimensional and multi-scale complex ecosystem models.

Insights into the effects of tea pruning litter biochar on major micronutrients (Cu, Mn, and Zn) pathway from soil to tea plant: An environmental armour

A field study was conducted from 0 to 360 days to investigate the effect of tea pruning litter biochar (TPLBC) on the accumulation of major micronutrients (copper: Cu, manganese: Mn, and zinc: Zn) in soil, their uptake by tea plant (clone: S.3 A/3) and level of contamination in soil due to TPLBC. To evaluate the level of contamination due to TPLBC, a soil pollution assessment was carried out using the geo-accumulation index (I_geo), enrichment factor (EF), contamination factor (CF), potential ecological risk factor (PERF), individual contamination factor (ICF), and risk assessment code (RAC). The total content of Cu, Mn, and Zn gradually increased with increasing doses of TPLBC at 0D, and then decreased with time. The fractionation of the three micronutrients in soil changed after the application of TPLBC. The contamination risk assessment of soil for Cu, Mn, and Zn based on the I_geo, EF, CF, PERF,ICF, and RAC suggested that the application of TPLBC does not have any adverse effect on soil. Except for Mn, the bioconcentration and translocation factors were less than one for Cu and Zn. Results from this study revealed that the application of 400 kg TPLBC ha^-1 is significantly better than the other treatments for Cu, Mn, and Zn at a 5% level of significance.

Modeling phytoremediation of heavy metal contaminated soils through machine learning

As an important subtopic within phytoremediation, hyperaccumulators have garnered significant attention due to their ability of super-enriching heavy metals. Identifying the factors that affecting phytoextraction efficiency has important application value in guiding the efficient remediation of heavy metal contaminated soil. However, it is challenging to identify the critical factors that affect the phytoextraction of heavy metals in soil-hyperaccumulator ecosystems because the current projections on phytoremediation extrapolations are rudimentary at best using simple linear models. Here, machine learning (ML) approaches were used to predict the important factors that affecting phytoextraction efficiency of hyperaccumulators. ML analysis was based on 173 data points with consideration of soil properties, experimental conditions, plant families, low-molecular-weight organic acids from plants, plant genes, and heavy metal properties. Heavy metal properties, especially the metal ion radius, were the most important factors that affect heavy metal accumulation in shoots, and the plant family was the most important factor that affect the bioconcentration factor, metal extraction ratio, and remediation time. Furthermore, the Crassulaceae family had the highest potential as hyperaccumulators for phytoremediation, which was related to the expression of genes encoding heavy metal transporting ATPase (HMA), Metallothioneins (MTL), and natural resistance associated macrophage protein (NRAMP), and also the secretion of malate and threonine. New insights into the effects of plant characteristics, experimental conditions, soil characteristics, and heavy metal properties on phytoextraction efficiency from ML model interpretation could guide the efficient phytoremediation by identifying the best hyperaccumulators and resolving its efficient remediation mechanisms.

Conceptual methodological framework for the resilience of biogeochemical services to heavy metals stress

The idea of linking stressors, services providing units (SPUs), and ecosystem services (ES) is ubiquitous in the literature, although is currently not applied in areas contaminated with heavy metals (HMs), This integrative literature review introduces the general form of a deterministic conceptual model of the cross-scale effect of HMs on biogeochemical services by SPUs with a feedback loop, a cross-scale heuristic concept of resilience, and develops a method for applying the conceptual model. The objectives are 1) to identify the clusters of existing research about HMs effects on ES, biodiversity, and resilience to HMs stress, 2) to map the scientific fields needed for the conceptual model's implementation, identify institutional constraints for inter-disciplinary cooperation, and propose solutions to surpass them, 3) to describe how the complexity of the cause-effect chain is reflected in the research hypotheses and objectives and extract methodological consequences, and 4) to describe how the conceptual model can be implemented. A nested analysis by CiteSpace of a set of 16,176 articles extracted from the Web of Science shows that at the highest level of data aggregation there is a clear separation between the topics of functional traits, stoichiometry, and regulating services from the typical issues of the literature about HMs, biodiversity, and ES. Most of the resilience to HMs stress agenda focuses on microbial communities. General topics such as the biodiversity-ecosystem function relationship in contaminated areas are no longer dominant in the current research, as well as large-scale problems like watershed management. The number of Web of Science domains that include the analyzed articles is large (26 up to 87 domains with at least ten articles, depending on the sub-set), but thirteen domains account for 70-80% of the literature. The complexity of approaches regarding the cause-effect chain, the stressors, the biological and ecological hierarchical level and the management objectives was characterized by a detailed analysis of 60 selected reviews and 121 primary articles. Most primary articles approach short causal chains, and the number of hypotheses or objectives by article tends to be low, pointing out the need for portfolios of complementary research projects in coherent inter-disciplinary programs and innovation ecosystems to couple the ES and resilience problems in areas contaminated with HMs. One provides triggers for developing innovation ecosystems, examples of complementary research hypotheses, and an example of technology transfer. Finally one proposes operationalizing the conceptual methodological model in contaminated socio-ecological systems by a calibration, a sensitivity analysis, and a validation phase.

Pollution Risk Assessment and Sources Analysis of Heavy Metal in Soil from Bamboo Shoots

In order to investigate the pollution situation and sources analysis of heavy metals in bamboo shoot soil in Guangdong Province, a total of 175 soil samples were collected at 46 sites. Atomic fluorescence spectrophotometer and inductively coupled plasma mass spectrometry were used to determine the content of five heavy metals: lead (Pb), cadmium (Cd), arsenic (As), mercury (Hg), and chromium (Cr). In addition, the soil environmental quality was evaluated through different index methods, including single-factor pollution, Nemeiro comprehensive pollution, geoaccumulation, and potential ecological risk. Furthermore, the correlation coefficients were also discussed. The results showed that the soils collected were acidic or slight alkaline. The maximum content of Pb and As from some areas exceeded the standard limit value. The coefficient of variation value from six areas exceeded 100%. The index method mentioned above confirmed that the soil within study areas was divided into three pollution levels: no, slightly, and mild. Additionally, there was a very significant correlation between pH and Pb, Hg; the correlation between heavy metal As and Pb, Cr also reached a very significant level. The principal component analysis results show that PC1 accounts for 39.60% of the total variance, which includes Pb, Cd, and As. PC2 mainly includes Hg and Cr.

Compositional features of Pb in agricultural soils and geochemical associations conditioning Pb contents in plants

Soil geochemical data is compositional. Hence the studies targeting the potential of accumulation of toxic elements (TE) in plants have to consider the compositional nature of soil chemical environment. In this study, the combined application of compositional data analysis and geospatial mapping was used to investigate Pb geochemical associations in agricultural soils, revealing the link between these associations and Pb contents in plants, as well as identifying source-specific transfer of Pb from soil to plants. The obtained results showed that soil chemical composition was conditioned by the geological peculiarities of the study area and the potential sources of chemical elements' release. Particularly, k-means clustering and CoDa-biplot allows to identify three distinct subsamples and the application of HCA showed that both Pb soil and plants contents were in the same cluster in all subsamples. However, the geochemical association of elements in subsamples I and III suggested that Pb contents in plants were conditioned by the geochemical behaviors of carbonates whereas in subsample II Pb plant contents were presented in a geochemical association (K, Rb, Pb, and Zn) typical for both fertilizers and the potassium feldspar. The transfer factor (TF) for the comparatively higher values is observed for the subsample linked to K, Rb, Pb, and Zn geochemical association. At the same time, the negative influence of carbonates on the Pb availability in the plants was evidenced. The results of this study can serve as a good example for other investigations targeting the role of soil chemical elements compositional features in elements transfer to plant.

Heavy metal habitat: A novel framework for mapping heavy metal contamination over large-scale catchment with a species distribution model

Heavy metal(loid)s (HMs) have been consistently entering the food chain, imposing great harm on environment and public health. However, previous studies on the spatial dynamics and transport mechanism of HMs have been profoundly limited by the field sampling issues, such as the uneven observations of individual carriers and their spatial mismatch, especially over large-scale catchments with complex environment. In this study, a novel methodological framework for mapping HMs at catchment scale was proposed and applied, combining a species distribution model (SDM) with physical environment and human variables. Based on the field observations, we ecologicalized HMs in different carriers as different species. This enabled the proposed framework to model the 'enrichment area' of individual HMs in the geographic space (termed as the HM 'habitat') and identify their 'hotspots' (peak value points) within the catchment. Results showed the output maps of HM habitats from secondary carriers (soil, sediment, and wet deposition) well agreed with the influence of industry contaminants, hydraulic sorting, and precipitation washout process respectively, indicating the potential of SDM in modeling the spatial distributions of the HM. The derived maps of HMs from secondary carriers, along with the human and environmental variables were then input as explanatory variables in SDM to predict the spatial patterns of the final HM accumulation in river water, which was observed to have largely improved the prediction quality. These results confirmed the value of our framework to leverage SDMs from ecology perspective to study HM contamination transport at catchment scale, offering new insights not only to map the spatial HM habitats but also help locate the HM transport chains among different carriers.

Driving effects and transfer prediction of heavy metal(loid)s in contaminated courtyard gardens using redundancy analysis and multilayer perceptron

The distribution and migration of heavy metal(loid)s in the soil-vegetable systems of courtyard gardens near mining areas have rarely been investigated, leading to potential food safety risks for residents. Moreover, the existing research is mainly focused on the total content of heavy metal(loid)s (tMetals) rather than the bioavailable contents (aMetals). In this study, 26 and 28 pairs of soil and vegetable samples were collected from the courtyard gardens near the Realgar mine in Baiyun Town and the lead-zinc (Pb-Zn) mine in Shuikoushan Town, respectively. The tMetal and aMetal of cadmium (Cd), mercury (Hg), arsenic (As), Pb, chromium (Cr), nickel (Ni), copper (Cu), Zn, manganese (Mn), iron (Fe), and calcium (Ca) in the samples were analyzed in this study. The results showed that courtyard gardens were polluted by various heavy metal(loid)s at varying degrees. The bioavailabilities of different metals varied significantly, among which Cd has the highest bioavailability (> 30%). In the transfer process of heavy metal(loid)s, the transfer rate (Tf) was ranked as soil-roots (1.50) > stems-leaves (1.07) > roots-stems (0.46) > stems-fruits (0.33). Redundancy analysis was used to evaluate the driving effects, and the results revealed that aCa, aZn, and aFe in soil could inhibit the absorption of aCd by plant roots. Soil organic matter was the inhibiting factor regarding the transfer of aAs and aCu, whereas it was also the promoting factor for transferring aPb, aNi, and aCr. Furthermore, the multilayer perceptron (MLP) could effectively predict the Tf of heavy metal(loid)s based on the aMetal. The R² values of the MLP were ranked as follows: 0.91 for As, 0.88 for Zn, 0.85 for Hg, 0.83 for Cu, 0.79 for Cr, 0.66 for Cd, 0.65 for Pb, and 0.52 for Ni. This study emphasizes the aMetal-based ecological characteristics and prediction ability. The study results are significant for guiding residents to strategize appropriate crop planting and ensure the safe production and consumption of vegetables.

Machine learning can identify the sources of heavy metals in agricultural soil: A case study in northern Guangdong Province, China

Source tracing of heavy metals in agricultural soils is of critical importance for effective pollution control and targeting policies. It is a great challenge to identify and apportion the complex sources of soil heavy metal pollution. In this study, a traditional analysis method, positive matrix fraction (PMF), and three machine learning methodologies, including self-organizing map (SOM), conditional inference tree (CIT) and random forest (RF), were used to identify and apportion the sources of heavy metals in agricultural soils from Lianzhou, Guangdong Province, China. Based on PMF, the contribution of the total loadings of heavy metals in soil were 19.3% for atmospheric deposition, 65.5% for anthropogenic and geogenic sources, and 15.2% for soil parent materials. Based on SOM model, As, Cd, Hg, Pb and Zn were attributed to mining and geogenic sources; Cr, Cu and Ni were derived from geogenic sources. Based on CIT results, the influence of altitude on soil Cr, Cu, Hg, Ni and Zn, as well as soil pH on Cd indicated their primary origin from natural processes. Whereas As and Pb were related to agricultural practices and traffic emissions, respectively. RF model further quantified the importance of variables and identified potential control factors (altitude, soil pH, soil organic carbon) in heavy metal accumulation in soil. This study provides an integrated approach for heavy metals source apportionment with a clear potential for future application in other similar regions, as well as to provide the theoretical basis for undertaking management and assessment of soil heavy metal pollution.

Insights into site-specific influences of emission sources on accumulation of heavy metal(loid)s in soils by wheat grains

Excessive accumulation of heavy metal(loid)s in agricultural environment usually originates from anthropogenic activities. Both large diversities of emission sources and complexity of plant accumulation challenge the understanding of the site-specific effects of emission sources on heavy metal(loid)s in wheat grains. Herein, both soil samples and wheat grain samples (n = 80) were collected from the farmland of Jiyuan City, China. Soil and grain burdens of heavy metal(loid)s were determined by inductively coupled plasma mass spectrometry (ICP-MS) and/or X-ray fluorescence spectrometry (XRF). The quotients (Q) were developed to indicate relative impacts of industrial plants and traffic to soil sites. Principal component analysis-absolute principal component scores-multivariate linear regression (PCA-APCS-MLR) analysis was conducted to reveal the source contributions to heavy metal(loid)s in grains, considering Q values, soil, and wheat grain data. Results showed that contributions of main sources and factors drastically varied with soil sites, and usually overlapped to different extents. For grain Cd and grain Pb, natural soil silicate (0.066/0.104 mg/kg) and iron-bearing minerals (- 0.044/ - 0.174 mg/kg) contributed to high extents, while metal smelting activities (0.018/0.019 mg/kg) and agronomic activities (- 0.017/ - 0.019 mg/kg) unexpectedly posed low or moderate contributions. The pH-mediated availability of soil Cd (0.035 mg/kg) and the sand-dust weather (0.028 mg/kg) also made considerable contributions to grain Cd. For grain As, both natural soil iron-bearing (- 0.048 mg/kg) and silicate minerals (- 0.013 mg/kg) made negative contributions. The results benefit to the decision-making of pollution remediation of farmland soils in the regional scales.

Prediction models for monitoring selenium and its associated heavy-metal accumulation in four kinds of agro-foods in seleniferous area

Se-rich agro-foods are effective Se supplements for Se-deficient people, but the associated metals have potential risks to human health. Factors affecting the accumulation of Se and its associated metals in Se-rich agro-foods were obscure, and the prediction models for the accumulation of Se and its associated metals have not been established. In this study, 661 samples of Se-rich rice, garlic, black fungus, and eggs, four typical Se-rich agro-foods in China, and soil, matrix, feed, irrigation, and feeding water were collected and analyzed. The major associated metal for Se-rich rice and garlic was Cd, and that for Se-rich black fungus and egg was Cr. Se and its associated metal contents in Se-rich agro-foods were positively correlated with Se and metal contents in soil, matrix, feed, and matrix organic contents. The Se and Cd contents in Se-rich rice grain and garlic were positively and negatively correlated with soil pH, respectively. Eight models for predicting the content of Se and its main associated metals in Se-rich rice, garlic, black fungus, and eggs were established by multiple linear regression. The accuracy of the constructed models was further validated with blind samples. In summary, this study revealed the main associated metals, factors, and prediction models for Se and metal accumulation in four kinds of Se-rich agro-foods, thus helpful in producing high-quality and healthy Se-rich.

Recognition method for the health risks of potentially toxic elements in a headwater catchment

The spatial association of potentially toxic elements (PTEs) in soil-crop-groundwater systems is poorly recognised. In this study, the contents of arsenic (As), cadmium (Cd), copper (Cu) and lead (Pb) in paddy soils, rice and groundwater in the Xiancha River catchment were determined. The intrinsic effects of PTEs in soils on their spatial distribution in groundwater and rice were explored. Also, the potential sources and health risks of PTEs in multi-media were investigated. Results showed that the mean contents of As and Cd in soils were 23.86 and 0.26 mg kg^-1, respectively. In groundwater, the maximum (average) content of As reached 6.55 (1.84) μg L^-1. Moreover, As contents in soils and groundwater showed a sound spatial correlation (q = 0.81), and this is supported by the result of the soil column experiment, indicating homology and the strong vertical migration capacity of As. The non-homologous patterns of Pb, Cu and Cd contaminations in soil-groundwater system suggested that geogenic processes influenced the distribution of these PTEs. Cd presented a poor spatial correlation in soil-rice system, as multiple factors controlled its transfer process. Multivariate statistical analysis results demonstrated that As, Cu and Pb in soils mainly came from agricultural sources, whereas high Cd levels were from mining activities. Additionally, direct consumption of As-contaminated groundwater and Cd-contaminated rice posed significant health risks to local residents. This study, which proposes a risk recognition method used to investigate target PTEs in multi-media, may serve as a valuable reference for further research involving catchments.

Mitigating the accumulation of arsenic and cadmium in rice grain: A quantitative review of the role of water management

Arsenic exposure through rice consumption is a growing concern. Compared to Continuous Flooding (CF), irrigation practices that dry the soil at least once during the growing season [referred to here as Alternate Wetting and Drying (AWD)] can decrease As accumulation in grain; however, this can simultaneously increase grain Cd to potentially unsafe levels. We modelled grain As and Cd from field studies comparing AWD and CF to identify optimal AWD practices to minimize the accumulation of As and Cd in grain. The severity of soil drying during AWD drying event(s), quantified as soil water potential (SWP), was the main factor leading to a reduction in grain total As and inorganic As, compared to CF. However, lower SWP levels were necessary to decrease grain inorganic As, compared to total As. Therefore, if the goal is to decrease grain inorganic As, the soil needs to be dried further than it would for decreasing total As alone. The main factor driving grain Cd accumulation was when AWD was practiced during the season. Higher grain Cd levels were observed when AWD occurred during the early reproductive stage. Further, higher Cd levels were observed when AWD spanned multiple rice growth stages, compared to one stage. If Cd levels are concerning, the minimum trade-off between total As and Cd accumulation in rice grain occurred when AWD was implemented at a SWP of -47 kPa during one stage other than the early reproductive. While these results are not meant to be comprehensive of all the interactions affecting the As and Cd dynamics in rice systems, they can be used as a first guide for implementing AWD practices with the goal of minimizing the accumulation of As and Cd in rice grain.

Convolutional neural networks-based health risk modelling of some heavy metals in a soil-rice system

The long-term consumption of heavy metal-rich rice can cause serious harm to human health. However, the existing health risk assessment (HRA) can only be performed after the rice has been harvested, and this approach belongs to a passive and lagging pattern. This study is the first to explore the feasibility of health risk (HR) prediction by proposing the indirect model CNNHR-IND and the direct model CNNHR-DIR based on the convolutional neural network (CNN) technology. The dataset included 390 pairs of soil-rice samples collected from You County, China, with 17 environmental covariates. The R² values for CNNHR-IND for non-carcinogenic and carcinogenic risks were 0.578 and 0.554, respectively, and those for CNNHR-DIR were 0.647 and 0.574, respectively. The results demonstrated that both models performed well, especially CNNHR-DIR had a higher estimation accuracy. The spatial autocorrelation analysis indicated that CNNHR-DIR exerted no systematic bias in the prediction results for health risks, confirming the rationality of the CNNHR-DIR model. The sensitivity analysis further confirmed the generalizability and robustness of CNNHR-DIR. This study proved the feasibility of HR prediction and the potential of CNN technology in HRA, and is significant regarding early risk warnings of rice planting and the sustainable development of public health.

Chain modeling for the biogeochemical nexus of cadmium in soil-rice-human health system

This paper presents a novel chain model named soil-food-human (SFH) for clarifying the biogeochemical cascades among the triple challenges of cadmium contamination, food safety, and related public health effect. The model was developed based on the integration of spatial distribution pattern of soil environment and the biogeochemical process of cadmium in soil-rice-human health, and it was validated through a case study. In soil environment terms, SFH predicted the spatial distribution of soil properties with an average prediction accuracy of 82.28%. In food production terms, the SFH can identify the safe production zones for planting rice and unsafe area for adjusting croppingsystems with a relative error of 39.41%. In food consumption terms, SFH mapped the high-resolution map of cadmium exposure dose, which gives a new solution to assess the food safety risks for self-sufficient populations. For the health effect of rice cadmium exposure, SFH simulated the spatiotemporal pattern of urinary cadmium based on toxicokinetic which revealed the health effect of rice cadmium exposure. The chain model provides a new insight in understanding the biogeochemical cascades between food production, food safety, and public health, making it possible to develop a comprehensive strategy to tackle cadmium pollution in soil-rice-human health system.

Environmental health and risk assessment metrics with special mention to biotransfer, bioaccumulation and biomagnification of environmental pollutants

The environment pollutants, which are landed up in environment because of human activities like urbanization, mining and industrializations, affects human health, plants and animals. The living organisms present in environment are constantly affected by the toxic pollutants through direct contact or bioaccumulation of chemicals from the environment. The toxic and hazardous pollutants are easily transferred to different environmental matrices like land, air and water bodies such as surface and ground waters. This comprehensive review deeply discusses the routes and causes of different environmental pollutants along with their toxicity, impact, occurrences and fate in the environment. Environment health and risk assessment tools that are used to evaluate the harmfulness, exposure of living organisms to pollutants and the amount of pollutant accumulated are explained with help of bio-kinetic models. Biotransfer, toxicity factor, biomagnification and bioaccumulation of different pollutants in the air, water and marine ecosystems are critically addressed. Thus, the presented survey would be collection of correlations those addresses the factors involved in assessing the environmental health and risk impacts of distinct environmental pollutants.

Tracking the origin of trace metals in a watershed by identifying fingerprints of soils, landscape and river sediments

The identification of the spatial distribution of soil trace-elements and the contribution of different sources to the sediment yield is necessary for a better watershed and river water quality management. Until now, less attention has been paid to comprehensive assessments of sediment sources and soil trace-elements with respect to the suspended sediment production. The present study aimed at modelling the spatial distribution of soil trace-elements, quantifying the sediment sources apportionment and relating the landforms to polluted soils. Different techniques and approaches such as the Nemerow pollution index, machine learning algorithms (Random Forest (RF), generalised boosting methods (GBM), generalised linear models (GLM) and sediment fingerprinting were applied to the Kan watershed. A total of 79 soil samples having different Nemerow index values were considered for spatial modelling. Using statistical methods (Range test, Kruskal-Wallis and discrimination function analysis), an optimal set of tracers was selected. An unmixing model was applied to calculate the relative contribution of landforms for eight rainfall events. The results of the soil trace-element mapping showed that RF had the best performance with an accuracy of 83%. The evaluation of polluted soil areas showed that the landforms 'steep hills' and 'valley' contributed the most with 51% and 27% in the riparian zone, respectively. In addition, these landforms give a high contribution to sediment production in late-winter-spring events (29%) with a GOF (goodness of fit) of 0.65. The landform 'plain' had the highest contribution (28%) in sediment yield with a GOF of 0.72 in early-winter events. This means that the valley and steep hill landforms accelerate the transport of trace-elements across the watershed. Interestingly, the contribution of landforms varies during the year. Overall, the new proposed approach enables to better trace the origin of suspended sediments and trace-elements discharge into the river environment.