Jiaqian Zhang

Real-time and accurate monitoring of heavy metal concentrations in rice is essential for ensuring food safety and supporting the safe utilization of contaminated agricultural land. Hyperspectral technology offers advantages such as rapid, nondestructive, cost-effective, and environmentally friendly monitoring. This study developed multiple models for estimating Pb content in rice leaves, compared their performance, and identified stable modeling pathways for continuous regional monitoring using hyperspectral remote sensing data. A meta-analysis was conducted to assess how different spectral preprocessing methods and modeling strategies affect model performance. The results indicated that the combination of feature band selection algorithms and modeling methods substantially affected model performance. Among the evaluated strategies, the model constructed using Competitive Adaptive Reweighted Sampling (CARS) and Partial Least Squares (PLS) showed the most consistent performance while maintaining a high level of estimation accuracy. In contrast, the combination of Whale Optimization Algorithm (WOA) and PLS achieved the highest validation estimation accuracy (R² = 0.7452). When applied to GF-5A hyperspectral data, the selected models achieved regional-scale predictions with R² values greater than 0.55. These findings provide a technical basis for regional monitoring of heavy metal contamination in crops using hyperspectral remote sensing.

With the rapid development of industrialization in China, significant economic benefits have been accompanied by varying degrees of threat to the soil environment, particularly from heavy metal pollution. The rapid quantitative inversion of heavy metal concentrations and assessment of pollution risks are urgent tasks. This study selected the affected area of a non-ferrous metal smelting slag yard in Gejiu City as the research subject. Farmland soil samples were collected from this area, and the basic physicochemical properties of the soil, along with the contents of Pb and Cd, were analyzed. Models for quantitatively inverting Pb and Cd contents based on soil physicochemical parameters were established using multiple linear regression (MLR), backpropagation neural network (BPNN), and genetic algorithm-optimized BPNN (GA-BPNN). Among the three inversion models, the GA-BPNN model demonstrated the highest accuracy, with inversion precision R² reaching 0.8980 and 0.9013 for Pb and Cd, respectively; the corresponding root mean square error (RMSE) values were 0.0001868 and 0.0001821. The inversion model established a nonlinear relationship between heavy metals and basic soil physicochemical properties, enabling further heavy metal pollution assessment based on the inversion results. This method provides a novel approach for inverting heavy metal content in soil and holds practical value for heavy metal remediation.

Industrial development has caused significant environmental damage, especially through potentially toxic element (PTE) pollution. Combining pollution indices, health risk assessment, spatial autocorrelation (Moran’s I), and receptor modeling (APCS/MLR), this study quantified sources and risks of heavy metals in smelting-adjacent farmland soils, facilitating targeted PTE pollution mitigation. Soil analysis revealed significantly elevated mean concentrations of As (326 mg/kg), Cd (23 mg/kg), Cr (104 mg/kg), Cu (106 mg/kg), Ni (73 mg/kg), Pb (274 mg/kg), and Zn (660 mg/kg), all exceeding Yunnan provincial background values. The average total non-carcinogenic risk index (HIadult = 2, HIchild = 11) and total carcinogenic risk index (TCRadult = 5.52 × 10–4, TCRChild = 6.44 × 10–4) for both adults and children exceeded the threshold (HI = 1, TCR = 1 × 10–04). The results of environmental pollution evaluation show that the overall pollution in the study area is a heavy pollution level. The ACPS-MLR model showed that Cd and Zn in soil mainly came from industrial activities (37%). Cu and Pb were derived from motor vehicle emissions and agricultural activities (20%). As may be derived from agricultural and industrial activities. Furthermore, based on the combination of source apportionalization and the spatial distribution of environmental pollution, the northeastern part of the study area and transportation hubs are the key pollution areas and need to be given priority for treatment. PTEs accumulate in the soil, will be enriched through the food chain, and seriously threaten human health and soil ecological environment. Therefore, this study can provide a basis for identifying, preventing, and controlling the risk of PTEs pollution in soil.

Due to the limitations inherent in SAR satellite imaging modes, utilizing time-series InSAR technology to process single-orbit satellite image data typically only yields one-dimensional deformation information along the LOS direction. This constraint impedes a comprehensive representation of the true surface deformation of landslides. Consequently, in this paper, after the SBAS-InSAR and PS-InSAR processing of the 30-view ascending and 30-view descending orbit images of the Sentinel-1A satellite, based on the imaging geometric relationship of the SAR satellite, we propose a novel computational method of fusing ascending and descending orbital LOS-direction time-series deformation to extract the landslide’s downslope direction deformation of landslides. By applying this method to Baige landslide monitoring and integrating it with an improved tangential angle warning criterion, we classified the landslide’s trailing edge into a high-speed, a uniform-speed, and a low-speed deformation region, with deformation magnitudes of 7~8 cm, 5~7 cm, and 3~4 cm, respectively. A comparative analysis with measured data for landslide deformation monitoring revealed that the average root mean square error between the fused landslide’s downslope direction deformation and the measured data was a mere 3.62 mm. This represents a reduction of 56.9% and 57.5% in the average root mean square error compared to the single ascending and descending orbit LOS-direction time-series deformations, respectively, indicating higher monitoring accuracy. Finally, based on the analysis of landslide deformation and its inducing factors derived from the calculated time-series deformation results, it was determined that the precipitation, lithology of the strata, and ongoing geological activity are significant contributors to the sliding of the Baige land-slide. This method offers more comprehensive and accurate surface deformation information for dynamic landslide monitoring, aiding relevant departments in landslide surveillance and management, and providing technical recommendations for the fusion of multi-orbital satellite LOS-direction deformations to accurately reconstruct the true surface deformation of landslides.

To enhance the sorption efficacy of attapulgite for heavy metals, mercapto-functionalized attapulgite (ATP-SH) was synthesized with mercaptan functional groups. When the mass-to-volume ratio of calcined attapulgite (ATP-C) to 3-Mercapropyltrimethoxysilane (MPTMS) was 1 g:0.5 mL (ATP-SH-0.5) and the pH was set to 8, a strong adsorption capacity for Cd (II) and Pb (II) was demonstrated. This indicates excellent adsorption performance for these heavy metals. ATP-SH-0.5 exhibited a maximum adsorption capacity of 43.81 mg/g and 274.83 mg/g for Cd (II) and Pb (II), respectively, in a single ion system. In a binary ion system, the maximum adsorption capacity was 31.86 mg/L and 254.45 mg/L for Cd (II) and Pb (II), respectively. Various characterizations and experiments showed that the adsorption of Cd (II) and Pb (II) onto ATP-SH-0.5 involves ion exchange reactions involving hydroxyl and thiol functional group complexation reactions. This adsorption process follows a single-molecule layer adsorption mechanism. XPS results indicate that hydroxyl and grafted thiol functional groups on the surface of mercapto-functionalized attapulgite participated in surface complexation reactions with Cd (II) and Pb (II), resulting in the formation of Cd-S and Pb-S species. Overall, this study provides a promising mercapto-functionalized modification material for the remediation of polluted water and soil.

Vis-NIR and XRF spectroscopy are widely used in monitoring heavy metals in soil due to their advantages of being fast, non-destructive, cost-effective, and non-polluting. However, when used individually, XRF and vis-NIR may not meet the accuracy requirements for Cd determination. In this study, we focused on the impact area of a non-ferrous metal smelting slag site in Gejiu City, Yunnan Province, fused the pre-selected vis-NIR and XRF spectra using the Pearson correlation coefficient (PCC), and identified the characteristic spectra using the competitive adaptive reweighted sampling (CARS) method. Based on this, a quantitative model for soil Cd concentration was established using partial least squares regression (PLSR). The results showed that among the four fusion spectral quantitative models constructed, the model combining vis-NIR spectral second-order derivative transformation and XRF spectral first-order derivative transformation (D2(vis-NIR) + D1(XRF)) had the highest coefficient of determination (R² = 0.9505) and the smallest root mean square error (RMSE = 0.1174). Compared to the estimation models built using vis-NIR and XRF spectra alone, the average computational time of the fusion models was reduced by 68.19% and 63.92%, respectively. This study provides important technical means for real-time and large-scale on-site rapid estimation of Cd content using multi-source spectral fusion.

Traditional methods for obtaining soil heavy metal content are expensive, inefficient, and limited in monitoring range. In order to meet the needs of soil environmental quality evaluation and health status assessment, visible near-infrared spectroscopy and XRF spectroscopy for monitoring heavy metal content in soil have attracted much attention, because of their rapid, nondestructive, economical, and environmentally friendly features. The use of either of these spectra alone cannot meet the accuracy requirements of traditional measurements, while the synergistic use of the two spectra can further improve the accuracy of monitoring heavy metal lead content in soil. Therefore, this study applied various spectral transformations and preprocessing to vis-NIR and XRF spectra; used the whale optimization algorithm (WOA) and competitive adaptive re-weighted sampling (CARS) algorithms to identify feature spectra; designed a combination variable model (CVM) based on multi-layer spectral data fusion, which improved the spectral preprocessing and spectral feature screening process to increase the efficiency of spectral fusion; and established a quantitative model for soil Pb concentration using partial least squares regression (PLSR). The estimation performance of three spectral fusion strategies, CVM, outer-product analysis (OPA), and Granger-Ramanathan averaging (GRA), was discussed. The results showed that the accuracy and efficiency of the CARS algorithm in the fused spectra estimation model were superior to those of the WOA algorithm, with an average coefficient of determination (R²) value of 0.9226 and an average root mean square error (RMSE) of 0.1984. The accuracy of the estimation models established, based on the different spectral types, to predict the Pb content of the soil was ranked as follows: the CVM model > the XRF spectral model > the vis-NIR spectral model. Within the CVM fusion strategy, the estimation model based on CARS and PLSR (CARS_D1+D2) performed the best, with R² and RMSE values of 0.9546 and 0.2035, respectively. Among the three spectral fusion strategies, CVM had the highest accuracy, OPA had the smallest errors, and GRA showed a more balanced performance. This study provides technical means for on-site rapid estimation of Pb content based on multi-source spectral fusion and lays the foundation for subsequent research on dynamic, real-time, and large-scale quantitative monitoring of soil heavy metal pollution using high-spectral remote sensing images.