Accurately mapping soil organic carbon (SOC) is conducive to evaluating carbon storage and soil quality. However, the high spatial heterogeneity of SOC caused by river-related factors and agricultural management brings challenges to digital soil mapping in floodplain farmland. Moreover, current studies focus on the non-linear relationship between SOC and covariates, but ignore the effective range of environmental variables on SOC, which prevents the revelation of the SOC differentiation mechanism. Using the 375 samples collected from the Jiangchang Town near Han River, we aim to determine the main controlling factors of SOC, reveal the effective range of environmental variables, and obtain the spatial map of SOC by using the gradient boosting decision tree (GBDT) model and partial dependence plots. Linear regression was used as a reference. Results showed that GBDT outperformed linear regression. GBDT results show that the distance from the river was the most important SOC factor, confirming the importance of the Han River to the SOC pattern. The partial dependence plots indicate that all environmental variables have their effective ranges, and when their values are extremely high or low, they do not respond to changes in SOC. Specifically, the influential ranges of rivers, irrigation canals, and rural settlements on SOC were within 4000, 200, and 50 m, respectively. The peak SOC was obtained with high clay (≥31%), total nitrogen (≥1.18 g/kg), and total potassium contents (≥11.1 g/kg), but it remained steady when these covariates further increased. These results highlight the importance of revealing the effective range of environmental variables, which provides data support for understanding the spatial pattern of SOC in floodplain farmland, achieving carbon sequestration in farmland and precision agriculture. The GBDT with the partial dependence plot was effective in SOC fitting and mapping.

Landslides frequently occur in the Xinjiang Uygur Autonomous Region of China due to its complex geological environment, posing serious risks to human safety and economic stability. Existing studies widely use machine learning models for landslide susceptibility prediction. However, they often fail to capture the threshold and interaction effects among environmental factors, limiting their ability to accurately identify high-risk zones. To address this gap, this study employed a gradient boosting decision tree (GBDT) model to identify critical thresholds and interaction effects among disaster-causing factors, while mapping the spatial distribution of landslide susceptibility based on 20 covariates. The performance of this model was compared with that of a support vector machine and deep neural network models. Results showed that the GBDT model achieved superior performance, with the highest AUC and recall values among the tested models. After applying clustering algorithms for non-landslide sample selection, the GBDT model maintained a high recall value of 0.963, demonstrating its robustness against imbalanced datasets. The GBDT model identified that 8.86% of Xinjiang’s total area exhibits extremely high or high landslide susceptibility, mainly concentrated in the Tianshan and Altai mountain ranges. Lithology, precipitation, profile curvature, the Modified Normalized Difference Water Index (MNDWI), and vertical deformation were identified as the primary contributing factors. Threshold effects were observed in the relationships between these factors and landslide susceptibility. The probability of landslide occurrence increased sharply when precipitation exceeded 2500 mm, vertical deformation was greater than 0 mm a⁻¹, or the MNDWI values were extreme (<−0.4, >0.2). Additionally, this study confirmed bivariate interaction effects. Most interactions between factors exhibited positive effects, suggesting that combining two factors enhances classification performance compared with using each factor independently. This finding highlights the intricate and interdependent nature of these factors in landslide susceptibility. These findings emphasize the necessity of incorporating threshold and interaction effects in landslide susceptibility assessments, offering practical insights for disaster prevention and mitigation.