Predictive statistical modelling of cadmium content in durum wheat grain based on soil parameters

Source-specific risk assessment for cadmium in wheat and maize: Towards an enrichment model for China

Cadmium (Cd) pollution of agricultural soil is of public concern due to its high potential toxicity and mobility. This study aimed to reveal the risk of Cd accumulation in soil and wheat/maize systems, with a specific focus on the source-specific ecological risk, human health risk and Cd enrichment model. For this we investigated more than 6100 paired soil and grain samples with 216 datasets including soil Cd contents, soil pH and grain Cd contents of 85 sites from China. The results showed that mining activities, sewage irrigation, industrial activities and agricultural practices were the critical factors causing Cd accumulation in wheat and maize cultivated sites. Thereinto, mining activities contributed to a higher Cd accumulation risk in the southwest China and Middle Yellow River regions; sewage irrigation influenced the Cd accumulation in the North China Plain. In addition, the investigated sites were classified into different categories by comparing their soil and grain Cd contents with the Chinese soil screening values and food safety values, respectively. Cd enrichment models were developed to predict the Cd levels in wheat and maize grains. The results showed that the models exhibited a good performance for predicting the grain Cd contents among safe and warning sites of wheat (R² = 0.61 and 0.72, respectively); while the well-fitted model for maize was prone to the overestimated sites (R² = 0.77). This study will provide national viewpoints for the risk assessments and prediction of Cd accumulation in soil and wheat/maize systems.

A New Sight of Influencing Effects of Major Factors on Cd Transfer from Soil to Wheat (Triticum aestivum L.): Based on Threshold Regression Model

Due to the high toxicity and potential health risk of cadmium (Cd), the influencing effects of major factors (like pH, OM, and clay, etc.) on Cd bioaccumulation and transfer from soil to crop grains are highly concerned. Multiple linear regression models were usually applied in previous literature, but these linear models could not reflect the threshold effects of major factors on Cd transfer under different soil environmental conditions. Soil pH and other factors on Cd transfer in a soil-plant system might pose different or even contrary effects under different soil Cd exposure levels. For this purpose, we try to apply a threshold regression model to analyze the effects of key soil parameters on Cd bioaccumulation and transfer from soil to wheat. The results showed that under different soil pH or Cd levels, several factors, including soil pH, organic matter, exchangeable Cd, clay, P, Zn, and Ca showed obvious threshold effects, and caused different or even contrary impacts on Cd bioaccumulation in wheat grains. Notably, the increase of soil pH inhibited Cd accumulation when pH > 7.98, but had a promotional effect when pH ≤ 7.98. Thus, threshold regression analysis could provide a new insight that can lead to a more integrated understanding of the relevant factors on Cd accumulation and transfer from soil to wheat. In addition, it might give us a new thought on setting regulatory limits on Cd contents in wheat grains, or the inhibitory factors of Cd transfer.

Cadmium accumulation in wheat and maize grains from China: Interaction of soil properties, novel enrichment models and soil thresholds

The cadmium (Cd) activity in soil has been widely studied. However, the interactive effects of soil properties (e.g. soil pH, CEC, and SOM) on Cd transfer from soil to grain are generally overlooked. In total 325 datasets including soil pH, CEC, SOM, and soil Cd content were used in this study. The descriptive statistics indicated that Cd content in wheat and maize soils ranged from 0.05 to 10.31 mg/kg and 0.02-13.68 mg/kg, with mean values of 0.87 and 1.14 mg/kg, respectively. Cd contents in wheat and maize grains were 0.01-1.36 mg/kg and 0.001-1.08 mg/kg with average values of 0.15 and 0.10 mg/kg, respectively. The results of SEM demonstrated that the interactive effects of soil properties contributed more to Cd transfer from soil to wheat grain than the soil Cd content. Subsequently, CITs-MLR indicated that the critical factors, including soil pH and total soil Cd content, could mask the contribution of other soil properties on Cd accumulation in grain; soil CEC may prevent Cd from leaching and therefore improve grain Cd level of wheat especially at acidic soil condition. The result of derived Cd thresholds revealed that current Cd thresholds are not completely suitable to wheat and maize grain at different soil conditions. This study provides a new model for further investigation on relationships between soil properties, soil Cd content and grain Cd level.

Effect of bamboo biochar on reducing grain cadmium content in two contrasting wheat genotypes

Wheat is the second most important food crop worldwide, which is prone to accumulate cadmium (Cd). Accumulation of Cd in wheat grains depends not only on wheat genotype, but also largely on the availability of soil Cd and its internal distribution. In this study, several experiments were used to achieve low-grain Cd content: a field trial for wheat genotype screening, a soil incubation experiment to test passivation effect of bamboo biochar on soil Cd, and a soil pot experiment to examine bamboo biochar effect on wheat grain accumulation. The results showed that of the 243 wheat cultivars tested, the variation range of grain Cd content was 0.365-1.243 mg/kg, in a field with soil Cd of 3 mg/kg. The application of bamboo biochar reduced soil Cd availability, among which 5.0% bamboo biochar treatment had the greatest effect. The content of available Cd in soil treated with 5.0% bamboo biochar decreased by 0.32 mg/kg compared with the control in a 120-day incubation experiment. Effect of bamboo biochar (0, 0.1%, 1.0%, and 5.0%) on reducing grain Cd content in two wheat genotypes (Mianyou-1 and 1279-9) was investigated. The application of bamboo biochar decreased Cd uptake by plants, while distribution of Cd in different wheat plant parts was more controlled by the plant genetic characteristics. Compared with the control, Cd content in roots, straw, and grains was decreased by 34.06% (P < 0.05), 21.57%, and 23.33%, respectively, in low-grain Cd wheat cultivar 1279-9 by 5% bamboo biochar application. Overall, the combination of low-grain Cd accumulation wheat and bamboo biochar may be a feasible strategy to lessen grain Cd accumulation in Cd-contaminated soils.

Efficient models for predicting durum wheat grain Cd conformity using soil variables and cultivars

Contamination of durum wheat grain by cadmium (Cd) threatens food safety and is of increasing concern because regulations concerning Cd are becoming stricter due to its toxicity. This work aimed at using soil variables and cultivar types to build models to predict whether durum wheat grain Cd will conform with current and possibly lower regulatory thresholds. We combined multiple Gaussian and logistic regressions and the random forest algorithm to take advantage of their strength. Models tested using cross-validation produced excellent performances including for the lowest regulatory threshold of 0.1 mg Cd/kg, half of the current one: 79-85% of the non-conformity cases were detected and the reliability of predictions was 69-82%. The models enabled identification of a x1.4 variability in grain Cd content between cultivars that do not have the low Cd accumulation allele of the Cdu1 gene. The models confirmed that for the grain Cd content, the between-cultivar variability had much less influence than the phytoavailability of Cd in soil, the critical contexts of which were characterized by the models. For farmers, these models are valuable tools to predict whether durum wheat production will conform with existing and future Cd regulation in foodstuffs.

Prediction models for rice cadmium accumulation in Chinese paddy fields and the implications in deducing soil thresholds based on food safety standards

Deducing soil cadmium (Cd) safety thresholds should be different for rice cultivars with different capacities to accumulate Cd to guarantee safe rice production in China. This study developed prediction models based on soil properties and deduced soil safety thresholds for Cd translocation from thirty-three paddy soils by two contrasting rice cultivars, Yelicanghua (high Cd accumulator, HCd) and Longhuamaohu (low Cd accumulator, LCd). A total of 330 paired field validation samples were used to examine the accuracy of prediction models and soil safety thresholds. The average soil Cd concentration was 0.26 (range 0.057-0.72) mg kg^-1. The average brown rice Cd concentrations were 0.14 (0.043-0.55) mg kg^-1 in HCd and 0.024 (0.007-0.15) mg kg^-1 in LCd in 2017, with corresponding values of 0.16 (0.016-0.66) and 0.027 (0.009-0.10) mg kg^-1 in 2018. Soil total Cd and pH were the two most important variables exhibiting direct effects on Cd concentrations in HCd, explaining 66% of the variance across the 33 soils. Soil total Cd, pH and organic carbon (OC) were the three most important variables in LCd, explaining 75% of the variance. Soil safety thresholds ranged from 0.27 to 1.00 mg kg^-1 for HCd and from 4.52 to 46.9 mg kg^-1 for LCd with pH ranging from 4.5 to 8.0. The validation results suggest ∼60% for HCd or the current soil quality standard (SQS) and 88% for LCd of the validation samples were suitable to meet the food quality standard (FQS), with 6.4% and 12%, respectively, of the validation soils unsuitable for rice cultivation. The current Chinese SQS is too strict for LCd which may be grown safely in moderately polluted soils and the derivation of soil thresholds should therefore consider the abilities of different cultivars to accumulate Cd.

Cd(II) retention and remobilization on δ-MnO2 and Mn(III)-rich δ-MnO2 affected by Mn(II)

Birnessite owing to its negative surface charge and defective structure exhibits high sorption affinities for Cd(II). However, Mn(II) can not only compete for the sorption sites with Cd(II), but also react with structural Mn(IV) in birnessite to form Mn(III), and thus, affect Cd(II) immobilization by birnessite. Herein, we investigate effects of Mn(II) on Cd(II) retention and remobilization on two birnessite δ-MnO₂ and Mn(III)-rich δ-MnO₂ (denoted as HE-MnO₂). At pH 5.5, Cd(II) sorption to birnessite was inhibited by Mn(II) addition. Mn(II) addition to δ-MnO₂ led to Cd(II) migration from vacant sites to edge sites, forming double-corner sharing (DCS) complexes. Mn(II) introduction to δ-MnO₂ led to less stable Cd(II) species formed on birnessite, indicating that Cd(II) was more firmly bound to vacant sites than edge sites of birnessite. Cd(II) formed double-edge sharing (DES) and DCS complexes on HE-MnO₂. Mn(II) addition to HE-MnO₂ increased the CdMn distance in DES complexes. The stability of adsorbed Cd(II) on HE-MnO₂ was slightly elevated due to Mn(II) addition. At pH 7.5, Mn(II) had no effect on Cd(II) sorption and desorption amounts on birnessite. However, low concentration of Mn(II) added to δ-MnO₂ induced partial migration of Cd(II) from vacant sites to edge sites while high concentration of Mn(II) added to birnessite led to the formation of amorphous Cd(II)-Mn(III) coprecipitate. These findings imply that aqueous Mn(II) is an important factor in influencing Cd(II) immobilization by birnessite in the environment.

Long-term organic matter application reduces cadmium but not zinc concentrations in wheat

Wheat is a staple food crop and a major source of both the essential micronutrient zinc (Zn) and the toxic heavy metal cadmium (Cd) for humans. Since Zn and Cd are chemically similar, increasing Zn concentrations in wheat grains (biofortification), while preventing Cd accumulation, is an agronomic challenge. We used two Swiss agricultural long-term field trials, the "Dynamic-Organic-Conventional System Comparison Trial" (DOK) and the "Zurich Organic Fertilization Experiment" (ZOFE), to investigate the impact of long-term organic, mineral and combined fertilizer inputs on total and phytoavailable concentrations of soil Zn and Cd and their accumulation in winter wheat (Triticum aestivum L.). "Diffusive gradients in thin films" (DGT) and diethylene-triamine-pentaacetic acid (DTPA) extraction were used as proxies for plant available soil metals. Compared to unfertilized controls, long-term organic fertilization with composted manure or green waste compost led to higher soil organic carbon, cation exchange capacity and pH, while DGT-available Zn and Cd concentrations were reduced. The DGT method was a strong predictor of shoot and grain Cd, but not Zn concentrations. Shoot and grain Zn concentrations correlated with DTPA-extractable and total soil Zn concentrations in the ZOFE, but not the DOK trial. Long-term compost fertilization led to lower accumulation of Cd in wheat grains, but did not affect grain Zn. Therefore, Zn/Cd ratios in the grains increased. High Zn and Cd inputs with organic fertilizers and high Cd inputs with phosphate fertilizers led to positive Zn and Cd mass balances when taking into account atmospheric deposition and fertilizer inputs. On the other hand, mineral fertilization led to the depletion of soil Zn due to higher yields and thus higher Zn exports than under organic management. The study supports the use of organic fertilizers for reducing Cd concentrations of wheat grains in the long-term, given that the quality of the fertilizers is guaranteed.

Estimating cadmium availability to the hyperaccumulator Sedum plumbizincicola in a wide range of soil types using a piecewise function

Estimating the bioavailability and predicting the uptake of metals to hyperaccumulators is very important in developing the field application of phytoextraction. A pot experiment was conducted using 108 agricultural soils covering a wide range of soil properties by the cadmium (Cd) hyperaccumulator Sedum plumbizincicola. The contributions of a range of soil properties to Cd uptake were quantified. Soil total, soluble, CaCl₂-extractable and diffusive gradients in thin films (DGT)-extractable Cd concentrations (Cd_total, Cd_soln, Cd_CaCl2 and Cd_DGT) were used to estimate Cd bioavailability and predict shoot Cd concentration (Cd_shoot) using a piecewise function. Cd_total and pH were the two major contributors to Cd uptake. Cd_shoot showed a logarithmic increase with Cd_total from 0.30 to 10.0 mg kg^-1 but no further increase when Cd levels exceeded 10 mg kg^-1. Soil pH had a discernible negative effect on Cd bioavailability from pH 5.5 to 7.5 but a weak influence at pH < 5.5 or pH > 7.5. This indicates that the optimum pH for phytoextraction with S. plumbizincicola was ~5.5 and lower pH produced little increase in shoot Cd uptake. DGT gave the best estimation of Cd bioavailability across all the data. When Cd_total > 10 mg kg^-1, none of the four measures was accurate enough to predict Cd_shoot but when pH > 7.5 all the four measures were well correlated with Cd_shoot. Piecewise equations in different ranges of Cd_total or pH significantly improved the prediction of Cd_shoot compared with the global equations derived from all the data. Compared with the piecewise equations, when pH > 7.5 Cd_shoot was greatly overestimated with the global equation of Cd_total. Our study provides useful information on the soils in which phytoextraction with S. plumbizincicola is feasible in the field.

CAPSULE

Cd availability to S. plumbizincicola was estimated by a piecewise function in soils with wide ranges of total Cd concentration and pH.