Digital mapping of soil organic carbon density in China using an ensemble model.

The soil organic carbon stock (SOCS) is considered as one of the largest carbon reservoirs in terrestrial ecosystems, and small changes in soil can cause significant changes in atmospheric CO2 concentration. Understanding organic carbon accumulation in soils is crucial if China is to meet its dual carbon target. In this study, the soil organic carbon density (SOCD) in China was digitally mapped using an ensemble machine learning (ML) model. First, based on SOCD data obtained at depths of 0-20 cm from 4356 sampling points (15 environmental covariates), we compared the performance of four ML models, namely random forest (RF), extreme gradient boosting (XGBoost), support vector machine (SVM), and artificial neural network (ANN) models, in terms of coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE) values. Then, we ensembled four models using Voting Regressor and the principle of stacking. The results showed that ensemble model (EM) accuracy was high (RMSE = 1.29, R2 = 0.85, MAE = 0.81), so that it could be a good choice for future research. Finally, the EM was used to predict the spatial distribution of SOCD in China, which ranged from 0.63 to 13.79 kg C/m2 (average = 4.09 (±1.90) kg C/m2). The SOC storage amount in surface soil (0-20 cm) was 39.40 Pg C. This study developed a novel, ensemble ML model for SOC prediction, and improved our understanding of the spatial distribution of SOC in China.

Machine learning-based prediction of cadmium bioaccumulation capacity and associated analysis of driving factors in tobacco grown in Zunyi City, China.

Tobacco grown in areas with high-geochemical backgrounds exhibits considerably different cadmium (Cd) bioaccumulation abilities due to regional disparities and environmental changes. However, the impact of key factors on the Cd bioaccumulation ability of tobacco grown in the karst regions with high selenium (Se) geochemical backgrounds is unclear. Herein, 365 paired rhizospheric soil-grown tobacco samples and 321 topsoil samples were collected from typical karst tobacco-growing soil in southwestern China and analyzed for Cd and Se. XGBoost was used to predict and evaluate the Cd bioaccumulation ability of tobacco and potential influencing factors. Results showed that regional geochemical characteristics, such as soil Cd and Se contents, soil type, and lithology, have the highest influence on the Cd bioaccumulation ability of tobacco, accounting for 46.5% of the overall variation. Moreover, soil Se contents in high-geochemical background areas considerably affect Cd bioaccumulation in tobacco, with a threshold for the mutual suppression effects of Cd and Se at a soil Se content of 0.8 mg/kg. According to the results of bivariate local indicators of spatial association analysis, tobacco cultivated in the central, northeast, and southeast regions of Zunyi City carries a lower risk of soil Cd contamination. This study provides new insights for managing tobacco cultivation in karst regions.

Variation in rhizosphere microbiota correlates with edaphic factor in an abandoned antimony tailing dump.

The distribution pattern of root-associated bacteria in native plant growth in tailing dumps with extreme conditions remains poorly understood and largely unexplored. Herein we chose a native plant, Bidens bipinnata, growing on both an Sb tailing dump (WKA) and adjacent normal soils (WKC) to in-depth understand the distribution pattern of root-associated bacteria and their responses on environmental factors. We found that the rhizosphere microbial diversity indices in the tailing dump were significantly different from that in the adjacent soil, and that such variation was significantly related with soil nutrients (TC, TOC, TN) and metal(loid) concentrations (Sb and As). Some dominant genera were significant enriched in WKA, suggesting their adaption to harsh environments. Notably, these genera are proposed to be involved in nutrient and metal(liod) cycling, such as nitrogen fixing (Devosia, Cellvibrio, Lysobacter, and Cohnella), P solubilizing (Flavobacterium), and Sb and As oxidation (Paenibacillus, Bacillus, Pseudomonas, and Thiobacillus). Our results suggest that certain root-associated bacteria in tailing dump were governed by soil edaphic factors and play important ecological roles in nutrient amendments and metal cycling for the successful colonization of Bidens bipinnata in this tailing dump.