Understanding soil organic carbon dynamics is crucial for climate change mitigation and developing strategies for sustainable land management. This research aimed to quantify SOC, analyze the effect of physicochemical properties on SOC, and evaluate different statistical models to identify the effective predictors for SOC. A total of 280 soil samples were analyzed for physicochemical properties using standard protocols. The study applied several statistical models, including Linear Mixed, Random Forest, Bayesian Linear, Generalized Additive, Multivariate Regression models, and Principal Component Analysis (PCA). Undisturbed soils exhibited significantly higher SOC stocks, averaging 74.71±8.65 Mg ha⁻1, compared to 53.58±7.13 Mg ha⁻1 in disturbed soils. The LMM and GAM indicated a significant baseline SOC but showed no notable effect of altitude on SOC (p = 0.703 and 0.62-0.93, respectively). RFR identified bulk density as the strongest predictor, with the highest node purity (2269.57). PCA accounted for 78.35% of the variance, showing the critical role of soil texture in stabilizing SOC. Altitude showed a minimal effect; soil bulk density is the key factor in SOC variability, while clay content is crucial for SOC stabilization. Advanced models like RFR provide better SOC predictions, aiding sustainable land management while incorporating additional variables that could further enhance SOC forecasting.