Chen Zhang

It remains unclear what changes have occurred in the distribution pattern of and trend in soil organic carbon (SOC) in China against the background of climate and land use change. Clarifying the dynamic changes in SOC and their driving factors in different regions of China is therefore crucial for assessing the global carbon cycle. In this study, we collected and supplemented a large amount of soil organic carbon density (SOCD) data in China from 2001 to 2020 and extracted data on environmental covariates (ECs) for the corresponding years. A random forest model was used to estimate the SOCD at a depth of 0–20 cm and 0–100 cm in China for the years 2001, 2005, 2010, 2015, and 2020, and we explored the trend of SOCD changes and their key driving factors. The results showed the following: (1) Compared with previous studies, the predictive ability of the 0–100 cm depth model was greatly improved; the coefficient of determination (R²) was 0.61 and Lin’s concordance correlation coefficient (LCCC) was =0.76. (2) From 2001 to 2020, China’s soil organic carbon stocks (SOCS) were 38.11, 39.11, 39.88, 40.16, and 41.12 Pg C for the 0–20 cm depth and 110.49, 112.67, 112.80, 113.06, and 114.96 Pg C for the 0–100 cm depth, respectively. (3) The effects of temperature and precipitation on SOCD in China showed obvious regional variability, and land use changes had mainly positive effects on SOCD in all regions of China, which was related to the large-scale implementation of ecological protection and restoration and the policy of returning farmland to forests and grasslands in China. This study provides strong scientific support for addressing climate change and rationalizing the use of land resources.

The dynamic replacement of soil organic carbon represents a pivotal mechanism through which water erosion modulates soil–atmosphere CO₂ fluxes. However, the extent of this dynamic replacement of soil inorganic carbon within this process remains unclear. In our study, we focused on Yuanmou County, China, a prototypical region afflicted by water erosion, as our study area. We leveraged the WaTEM/SEDEM model to quantify the dynamic replacement of soil carbon, accounted for the average annual net change in soil carbon pools, and used isotope tracer techniques to track and measure the process of the coupled carbon–water cycling. This comprehensive approach enabled us to scrutinize the dynamic replacement of soil carbon under water erosion and delineate its ramifications for the carbon cycle. Our findings unveiled that the surface soil carbon reservoir in the Yuanmou area receives an annual replacement of 47,600 ± 12,600 tons following water erosion events. A substantial portion, amounting to 39,700 ± 10,500 tons, stems from the dynamic replacement of soil inorganic carbon facilitated by atmospheric carbon. These results underscore the critical role of the dynamic replacement of soil inorganic carbon in altering the soil–atmosphere CO₂ fluxes under water erosion, thereby influencing the carbon cycle dynamics. Consequently, we advocate for the integration of water erosion processes into regional carbon sink assessments to attain a more comprehensive understanding of regional carbon dynamics.

Carbon absorption and storage in forests is one of the important ways to mitigate climate change. Therefore, it is essential to use a variety of remote-sensing resources to accurately estimate forest aboveground biomass (AGB) in dry-hot valley regions. In this study, satellite images from the Sentinel-1A, Sentinel-2A, and Gaofen-2 satellites were utilized to estimate the forest AGB in Yuanmou County, Yunnan Province, China. Different combinations of image data, based on selected variables of stepwise regression and their performance in constructing linear stepwise regression (LSR) and random forest (RF) models, were explored. The results showed that: (1) after adding the polarized values of the synthetic aperture radar backscatter coefficients, the combination fitting effect was significantly improved; (2) the fitting effect of the Sentinel-1A + Sentinel-2A + Gaofen-2 data combination was superior to the other combinations, indicating that the effective extraction of forest horizon and vertical information can improve the estimation effect of the forest AGB; and (3) the RF model exhibited superior fitting performance compared to the LSR model across all permutations of remotely sensed image datasets, with R² values of 0.71 and 0.65, and RMSE values of 30.67 and 33.79 Mg/ha, respectively. These findings lay the groundwork for enhancing the precision of AGB estimation in dry-hot valley areas by integrating Sentinel-2A, Sentinel-1A, and GF-2 imagery, providing valuable insights for future research and applications.

Comparing algorithms are crucial for enhancing the accuracy of remote sensing estimations of forest biomass in regions with high heterogeneity. Herein, Sentinel 2A, Sentinel 1A, Landsat 8 OLI, and Digital Elevation Model (DEM) were selected as data sources. A total of 12 algorithms, including 7 types of learners, were utilized for estimating the aboveground biomass (AGB) of Pinus yunnanensis forest. The results showed that: (1) The optimal algorithm (Extreme Gradient Boosting, XGBoost) was selected as the meta-model (referred to as XGBoost-stacking) of the stacking ensemble algorithm, which integrated 11 other algorithms. The R² value was improved by 0.12 up to 0.61, and RMSE was decreased by 4.53 Mg/ha down to 39.34 Mg/ha compared to the XGBoost. All algorithms consistently showed severe underestimation of AGB in the Pinus yunnanensis forest of Yunnan Province when AGB exceeded 100 Mg/ha. (2) XGBoost-Stacking, XGBoost, BRNN (Bayesian Regularized Neural Network), RF (Random Forest), and QRF (Quantile Random Forest) have good sensitivity to forest AGB. QRNN (Quantile Regression Neural Network), GP (Gaussian Process), and EN (Elastic Network) have more outlier data and their robustness was poor. SVM-RBF (Radial Basis Function Kernel Support Vector Machine), k-NN (K Nearest Neighbors), and SGB (Stochastic Gradient Boosting) algorithms have good robustness, but their sensitivity was poor, and QRF algorithms and BRNN algorithm can estimate low values with higher accuracy. In conclusion, the XGBoost-stacking, XGBoost, and BRNN algorithms have shown promising application prospects in remote sensing estimation of forest biomass. This study could provide a reference for selecting the suitable algorithm for forest AGB estimation.