Wetland restoration enhances soil carbon sequestration in lake ecosystems: Integrating multi-source remote sensing and optimized ensemble machine learning to map soil organic carbon density

Lake wetlands play essential roles in flood regulation, water purification, and biodiversity, and restoring degraded wetlands can enhance their carbon-sink capacity. However, knowledge regarding the spatial patterns, temporal dynamics, and key drivers of soil organic carbon density (SOCD) in restored wetlands is limited. This study integrated multi-source remote sensing and machine-learning approaches to map SOCD in a restored lake wetland and to assess its spatiotemporal evolution from 2000 to 2023. Results showed that soil depth, soil physicochemical properties, and near-infrared spectral features were the major drivers of SOCD variability. Among nine models tested, CatBoost achieved the highest accuracy under random five-fold cross-validation (R 2 = 0.49 ± 0.06; MAE = 6.78 ± 0.58 MgC·ha −1 ; RMSE = 9.30 ± 1.17 MgC·ha −1 ; RPIQ = 1.56 ± 0.17), whereas the RF model performed most robustly under spatially grouped cross-validation(R 2 = 0.41 ± 0.16, MAE = 7.45 ± 0.33 MgC·ha −1 ; RMSE = 9.76 ± 0.64 MgC·ha −1 ; RPIQ = 1.48 ± 0.24), demonstrating the advantage of ensemble learning in capturing spatial heterogeneity. Spatially, SOCD decreased from the upland to the lakeshore zones. Temporally, SOCD declined from 2004 to 2017 due to water-level fluctuations and aquaculture disturbance, with high-SOCD areas shrinking by 38.34 km 2 . Following the “Return-Aquaculture-to-Wetland” restoration project launched in 2017, SOCD recovered markedly, reaching a two-decade maximum in 2022 (high-SOCD area: 125.54 km 2 ; regional mean: 47.60 MgC·ha −1 ). These findings provide a scientific basis for wetland carbon-stock assessments and ecological-restoration monitoring and demonstrate the potential of combining multi-source remote sensing and machine learning for large-scale SOCD mapping. • Spatiotemporal patterns of soil organic carbon density (SOCD) were reconstructed • Prediction accuracy of different machine learning models for SOCD was evaluated • The random forest model demonstrated the strongest generalization ability • Ecological restoration has significantly increased SOCD • ML-based modeling approaches are effective for wetland carbon stock assessments