Evaluation of cadmium (Cd) transfer from paddy soil to rice (Oryza sativa L.) using DGT in comparison with conventional chemical methods: derivation of models to predict Cd accumulation in rice grains

A holistic field experimental inquiry into cadmium's migration and translocation dynamics across the entire growth spectrum of five Japonica rice cultivars

The contamination of farmland soils with cadmium (Cd) poses a substantial threat to agricultural productivity, food security and safety, and ultimately human health. However, little research has been done on the Cd transport mechanisms in highly Cd polluted soil via field experiment. This study, from a field-scale perspective, examines the migration and transformation features of Cd throughout the growth cycle of five (C1, C2, C3, C4, H1) Japonica rice cultivars in Jiangsu Province, China. Analysis of pH, SOM, total Cd, DTPA-Cd, and microbial communities were conducted. C1 ~ C3 were classified as High Cd-accumulating rice (HC), while C4 and H1 were considered as low Cd-accumulating rice (LC) based on the Cd levels in their brown rice. Phloem was confirmed as the main pathway for Cd into rice grains in high-Cd soil. For the HC group, the Cd concentration in brown and polished rice was positively correlated with the Cd concentration in the leaves and spikes; while for the LC group, they were significantly positively correlated with the Cd concentration in both stem and spike (p < 0.05). The husks of the LC group were more effective in intercepting and sequestering Cd. It was revealed that 6 % ~ 9.09 % of the Cd content detected in the rice grains could be attributed to the internal translocation processes occurring within the plant itself, and approximately 90.91 % ~ 93.84 % of the Cd was traced back to the roots' absorption during grouting. Rice polishing decreased the Cd content from the level in the brown rice by 18 % ~ 47 %. Distinct microbial profiles separated rice rhizosphere from bulk soil, with the former favouring copiotrophs in nutrient-rich zones and the latter oligotrophs in lean conditions. This study delivers crucial data support from a field perspective for a deeper understanding and control of Cd migration and transformation processes in highly Cd-contaminated soil.

Prediction of the bioaccessibility and accumulation of cadmium in the soil-rice-human system based on optimized DGT and BCR coupled models

Cadmium, as a typical heavy metal, has the potential to induce soil pollution and threaten human health through the soil-plant-human pathway. The conventional evaluation method based on the total content in soil cannot accurately represent the content migrated from the food chain to plants and the human body. Previous studies focused on the process of plant enrichment of heavy metals in soil, and very few studies directly predicted human exposure or risk through the labile state of Cd in soil. Hence, a relatively accurate and convenient prediction model of Cd release and translocation in the soil-rice-human system was developed. This model utilizes available Cd and soil parameters to predict the bioavailability of Cd in soil, as well as the in vitro bioaccessibility of Cd in cooked rice. The bioavailability of Cd was determined by the Diffusive Gradients in Thin-films technology and BCR sequential extraction procedure, offering in-situ quantification, which presents a significant advantage over traditional monitoring methods and aligns closely with the actual uptake of heavy metals by plants. The experimental results show that the prediction model based on the concentration of heavy metal forms measured by BCR sequential extraction procedure and diffusive gradients in thin-films technique can accurately predict the Cd uptake in rice grains, gastric and gastrointestinal phase (R2=0.712, 0.600 and 0.629). This model accurately predicts Cd bioavailability and bioaccessibility across the soil-rice-human pathway, informing actual human Cd intake, offering scientific support for developing more effective risk assessment methods.

Use of diffusive gradients in thin-film technique to predict the mobility and transfer of nutrients and toxic elements from agricultural soil to crops-an overview of recent studies

The purpose of this review was to survey the recent applications of the diffusive gradients in thin films (DGT) technique in the assessment of mobility and bioavailability of nutrients and potentially toxic elements (PTEs) in agricultural soil. Many studies compared the capabilities of the DGT technique with those of classical soil chemical extractants used in single or sequential procedures to predict nutrients and PTE bioavailability to crops. In most of the published works, the DGT technique was reported to be superior to the conventional chemical extraction and fractionation methods in obtaining significant correlations with the metals and metalloids accumulated in crops. In the domain of nutrient bioavailability assessment, DGT-based studies focused mainly on phosphorous and selenium labile fraction measurement, but potassium, manganese, and nitrogen were also studied using the DGT tool. Different DGT configurations are reported, using binding and diffusive layers specific for certain analytes (Hg, P, and Se) or gels with wider applicability, such as Chelex-based binding gels for metal cations and ferrihydrite-based hydrogels for oxyanions. Overall, the literature demonstrates that the DGT technique is relevant for the evaluation of metal and nutrient bioavailability to crops, due to its capacity to mimic the plant root uptake process, which justifies future improvement efforts.

An integrated feature selection approach to high water stress yield prediction

The timely and precise prediction of winter wheat yield plays a critical role in understanding food supply dynamics and ensuring global food security. In recent years, the application of unmanned aerial remote sensing has significantly advanced agricultural yield prediction research. This has led to the emergence of numerous vegetation indices that are sensitive to yield variations. However, not all of these vegetation indices are universally suitable for predicting yields across different environments and crop types. Consequently, the process of feature selection for vegetation index sets becomes essential to enhance the performance of yield prediction models. This study aims to develop an integrated feature selection method known as PCRF-RFE, with a focus on vegetation index feature selection. Initially, building upon prior research, we acquired multispectral images during the flowering and grain filling stages and identified 35 yield-sensitive multispectral indices. We then applied the Pearson correlation coefficient (PC) and random forest importance (RF) methods to select relevant features for the vegetation index set. Feature filtering thresholds were set at 0.53 and 1.9 for the respective methods. The union set of features selected by both methods was used for recursive feature elimination (RFE), ultimately yielding the optimal subset of features for constructing Cubist and Recurrent Neural Network (RNN) yield prediction models. The results of this study demonstrate that the Cubist model, constructed using the optimal subset of features obtained through the integrated feature selection method (PCRF-RFE), consistently outperformed the RNN model. It exhibited the highest accuracy during both the flowering and grain filling stages, surpassing models constructed using all features or subsets derived from a single feature selection method. This confirms the efficacy of the PCRF-RFE method and offers valuable insights and references for future research in the realms of feature selection and yield prediction studies.

Predictive and estimation model of Cd, Ni, and Zn bioaccumulations in maize based on diffusive gradients in thin films

Consumption of maize contaminated with heavy metals such as cadmium, nickel, and zinc threaten human health. For situ measuring the bioavailability of heavy metals, the diffusive gradients in thin films (DGT) is superior to other traditional methods. It is also important to find a method for predicting heavy metal enrichment in maize based on the DGT method. In this study, field surveys were conducted in the main maize producing areas of Tianjin, China. Heavy metal concentrations in maize grains were predicted by coupling DGT with traditional extraction methods. The results show that coupling DGT with soil solution can significantly improve prediction accuracy (Cd-R² = 0.908, Ni-R² = 0.903, and Zn-R² = 0.904). This indicated that DGT and soil solution were feasible predictors of heavy metal concentration in maize. The DGT induced fluxes in soil/sediment (DIFS) model was used to simulate the uptake process of heavy metals by DGT, and better reveal the desorption processes of heavy metals in soils. DIFS-based desorption processes were employed to characterize the resupply ability of heavy metals in soils. The coupling of DGT and DIFS parameters provided the best prediction accuracy in this study (Cd-R² = 0.920, Ni-R² = 0.928, and Zn-R² = 0.908). Predictions are slightly weaker for Zn than for Cd and Ni (Cd-P < 0.01, Ni-P < 0.01, and Zn-P < 0.05). The reason is that the average resupply type of Cd and Ni in soil is partially sustained while Zn is resupplied via diffusion only. The desorption rate k_-1 can excellently improve the prediction accuracy of DGT, which avoids the disadvantage that soil solution does not consider desorption. The coupling of DGT and DIFS parameters provides an accurate and reliable method for predicting heavy metal enrichment in maize.

Prediction model for Cd accumulation of rice (Oryza sativa L.) based on extractable Cd in soils and prediction for high Cd-risk regions of southern Zhejiang Province, China

Soil environmental quality in China for agricultural land always considers the effect of total cadmium (Cd) in soil, ignoring the bioavailability of soil Cd. The 139 paired rice (Oryza sativa L.) and soil samples were collected from the Cd-contaminated paddy fields of southern Zhejiang Province, China. The objectives of this study were to establish accurate prediction models for Cd accumulation in brown rice based on bioavailable Cd and physiochemical properties of soils and to evaluate the safety of rice production in Cd-contaminated paddy. The bioavailable Cd in soils was extracted and evaluated by using CaCl₂, HNO₃, diethylenetriamine pentaacetic acid (DTPA), diffusive gradients in thin-films technique (DGT), and sequential extraction method proposed by the European Community Bureau of Reference; 100 pairs of data were used as training sets, and the remaining 39 sets were used as validation sets. Stepwise multiple linear regression analysis and random forest analysis showed that total Cd in soil could roughly indicate the content of Cd in rice, while extractable Cd could better explain the accumulation of Cd in rice grain and DTPA and DGT extractive technologies are the most evaluative. The validation sets also showed that the prediction model has a good fit. Based on the prediction model for Cd in brown rice based on soil pH and DGT extractive Cd, the Monte Carlo simulation showed that 74.32% and 89.35% of the estimated brown rice hazard quotient (HQ) of the daily Cd intake of adults and children in safe utilization paddy sites could exceed the safe level of 1, respectively. Additionally, the threshold values for extractable Cd by DGT or DTPA for rice safe production were 3.4 μg/kg or 0.13 mg/kg when the pH in soils was below 5.5. The results further proved the threshold concentration of extractable Cd for predicting high-risk soils of Cd contamination in brown rice.

The determination of regulating thresholds of soil pH under different cadmium stresses using a predictive model for rice safe production

Regulating soil pH becomes a crucial practice to alleviate cadmium (Cd) contamination. However, little is known about the threshold of soil pH for the safe production of rice at various soil Cd levels. In this paper, the relationships between soil pH values and the contents of available Cd extracted by calcium chloride (CaCl₂-Cd) in neutral and acidic soils were studied by mandatory acidification with H⁺ addition or neutralization with lime at various soil Cd levels. The results showed that the soil CaCl₂-Cd contents dramatically decreased with increasing soil pH, and a logarithmic function could well describe the relations of soil CaCl₂-Cd contents and soil pH at constant total Cd (CaCl₂-Cd model). The Cd contents in rice grain (grain-Cd) in relation to soil CaCl₂-Cd was further established through modified rice pot experiments. A model for the prediction of Cd content in rice grains (grain-Cd model) was set up, though which the grain-Cd content could be predicted based on soil pH and total Cd content. 122 data pairs of rice grain-Cd contents obtained at various soil total Cd contents and pH were employed from the literature to verify the reliability of the established model, approximately 95.08% of those data favorably located within the 1:1 line ± 0.5 unit area of the grain-Cd model. Notably, this model can be applied to determine the thresholds of soil pH at a specific Cd pollution level. For instance, to achieve a rice grain-Cd contents matching the Chinese national food safety limit of 0.2 mg kg^-1, the soil pH thresholds were estimated to be 5.05, 5.70, and 6.02 at soil Cd contents of 0.3, 0.6, and 0.8 mg kg^-1, respectively. In addition, the established model can also be used to estimate the health risk from rice in broad regions with various soil pH values and Cd contents.

Seawater intrusion induced cadmium activation via altering its distribution and transformation in paddy soil

Seawater intrusion can cause environmental risks to paddy soils around estuaries, but the impacts on the availability of heavy metals are still unclear. River water and sea water were collected along the river of an estuary. A stirred-flow experiment was conducted to examine the Cd desorption behavior in Cd-contaminated paddy soil. While the pH increased with increasing salinity levels, more Cd was released with increasing salinity, suggesting that Cd competition by cations and complexation by anions, but not pH, dominated the release of Cd from soils. Moreover, paddy soil was incubated at different salinities under alternating redox conditions. The availability of Cd, as indicated by the diffusive gradients in thin film (DGT), became relatively high with increasing salinity levels during the initial anaerobic and later aerobic stages. The available Cd fractions substantially decreased under anaerobic condition, and then rapidly increased under aerobic condition. When oxygen was introduced into the system, Cd associated with organic matter and Fe-Mn oxides were released, and oxidative dissolution of Cd sulfides was observed, especially in the high salinity treatment. Seawater intrusion affects biogeochemical cycles and can promote rapid export of NH₄⁺, Fe²⁺, and SO₄^2- in paddy soils, especially in soils with high salinity. Our findings demonstrated that the high salinity content in paddy soil significantly enhanced the availability of Cd, especially during the drainage stage.

Microbial induced carbonate precipitation contributes to the fates of Cd and Se in Cd-contaminated seleniferous soils

Bioremediation based on microbial induced carbonate precipitation (MICP) was conducted in Cd-contaminated seleniferous soils with objective to investigate effects of MICP on the fates of Cd and Se in soils. Results showed that soil indigenous microorganisms could induce MICP process to stabilize Cd and mobilize Se without inputting exogenous urease-producing strain. After remediation, soluble Cd (SOL-Cd) and exchangeable Cd (EXC-Cd) concentrations were decreased respectively by 59.8% and 9.4%, the labile Cd measured by the diffusive gradients in thin-films technique (DGT) was decreased by 14.2%. The MICP stabilized Cd mainly by increasing soil pH and co-precipitating Cd during the formation of calcium carbonate. Compared with chemical extraction method, DGT technique performs better in reflecting Cd bioavailability in soils remediated with MICP since this technique could eliminate the interference of Ca²⁺. The increase in pH resulted in Se conversion from nonlabile fraction to soluble and exchangeable fractions, thus improving Se bioavailability. And Se in soil solution could adsorb to or co-precipitate with the insoluble calcium carbonate during MICP, which would partly weaken Se bioavailability. Taken together, MICP had positive effects on the migration of Se. In conclusion, MICP could stabilize Cd and improve Se availability simultaneously in Cd-contaminated seleniferous soils.

Predictive model for cadmium uptake by maize and rice grains on the basis of bioconcentration factor and the diffusive gradients in thin-films technique

It is possible for heavy metals in soils to be adsorbed by crop roots and then accumulated in crops, which eventually causes great health risk when the crops are ingested by humans. Thus, it is valuable to understand the enrichment model of heavy metals in crops. Diffusive gradients in thin-films (DGT) technique, as an in-situ passive sampling method, can be used to evaluate the bioavailable heavy metals contents in soils. In this study, data of the bioavailable cadmium (Cd) in soils determined by DGT and Cd contents uptake in rice and maize grains in Tianjin, Zhejiang and Guangxi provinces of China were collected from previous references in Web of Science. By comparing bioconcentration factors, it was found that the heavy metal concentrations accumulated in rice and maize followed a general order roots > stems or leaves > grains. An accurate and robust model for the prediction of Cd content in maize and rice grains was established based on bioconcentration factor (BCF) and the bioavailable Cd content determined by DGT method, with R² 0.986 and root mean square error (RMSE) 0.128. This result suggests that the DGT method can be good tool for predicting heavy metals uptake in crops.

The Cd phytoextraction potential of hyperaccumulator Sedum alfredii-oilseed rape intercropping system under different soil types and comprehensive benefits evaluation under field conditions

Soil trace elements (TEs) contamination has become a worldwide problem in arable lands and poses great risk to human health via food chain. Intercropping of hyperaccumulator and cash crops is now proposed as a promising alternative phytoremediation technique to address the issue. However, the effect of intercropping in different soil types and field-scale benefits evaluation are rarely reported. A greenhouse pot experiment and a field trial were therefore designed to explore the effects of intercropping Sedum alfredii (hyperaccumulative population) and oilseed rape on Cd phytoextraction potential, Cd transport and crop production, as well as establishing a feasible assessment framework on the basis of benefits evaluation in contaminated soils. Compared with oilseed rape monoculture, intercropping with S. alfredii significantly and consistently increased biomass, seed yield and Cd accumulation in oilseed rape in five typical soil types. Accumulations of Cd varied with soil types, ranging from 22.8 to 4000 μg pot^-1. Stepwise multiple linear regression analysis (SMLRA) showed Cd concentrations in plants were related to available phosphorus (AP), pH, soil organic matter (OM), available potassium (AK), silt and sand; R² values varied from 0.834 to 0.994 (P < 0.05). A field trial also verified that intercropping could significantly enhance Cd phytoextraction. The highest index for comprehensive benefits evaluation was 0.61 observed in the S. alfredii and oilseed rape intercropping system. This system presented higher Cd phytoextraction potential and comprehensive benefits index whilst allowing ongoing agricultural activities in slightly and moderately Cd-contaminated soils. These results provide a possible technical approach for phytoremediation practice and give new insights into theoretical reference for development of Cd phytoextraction and benefits evaluation.