In carbon cycle and soil fertility management, SOCS play an important role. A better understanding of spatial variation and temporal decline of SOCS is indispensable for sustainable soil resource management. The purpose of this study was to assess the SOCS spatial distribution and quantify its temporal decline in the Tombel (South-West, Cameroon) area over 32 years (between 1985 and 2017) in relation to land use changes. Quantile Regression Forest (QRF) was applied to map the spatial distribution of SOCS in topsoil (0–30 cm), and its temporal dynamics over 32 years was assessed. A set of 91 and 100 soil samples were included for models calibration for 1985 and 2017, respectively, and models performances were assessed using a 10-fold cross-validation, as well as prediction uncertainties. Over three decades, agricultural and built up area (area such as a town or city with a lot of buildings) experienced a fast expansion, with respective land area increasing rates of 83 % and 81 %. The total SOCS in the study area changed from 22.33 Tg (Tera grams) to 14.68 Tg. Accuracies in the assessment showed good performances for 1985 (R² = 0.62, ME = - 0.06 and RMSE = 11.61 Mg ha⁻¹) and 2017 (R² = 0.58, ME = - 0.03 and RMSE = 13.15 Mg ha⁻¹). Climatic and terrain attributes were the most important predictors of the SOCS distribution. There was on average, about 1.3 Mg ha^-1 yr^-1 of SOCS decrease between 1985 and 2017, likely due to land use change and controlled by exacerbated agricultural land extension and rapid urbanization. The study provided a sound baseline for land management and decision making for soil quality, carbon accounting and carbon sequestration potential.

在碳循环和土壤肥力管理中，SOCS 发挥着重要作用。更好地了解 SOCS 的空间变化和时间下降对于可持续的土壤资源管理是必不可少的。本研究的目的是评估 SOCS 空间分布并量化其在 32 年间（1985 年至 2017 年）在 Tombel（喀麦隆西南）地区与土地利用变化相关的时间下降。应用分位数回归森林（QRF）来绘制表土（0-30 cm）中 SOCS 的空间分布图，并评估其 32 年的时间动态。分别为 1985 年和 2017 年的模型校准包括了一组 91 个和 100 个土壤样本，并使用 10 倍交叉验证以及预测不确定性评估了模型性能。三十多年来，农业和建成区（如城镇或建筑较多的地区）快速扩张，土地面积分别增长了83%和81%。研究区域的总 SOCS 从 22.33 Tg（Tg）变为 14.68 Tg。评估的准确性表明 1985 年表现良好（R² = 0.62，ME = - 0.06 和 RMSE = 11.61 Mg ha^-1 ) 和 2017 (R² = 0.58, ME = - 0.03 和 RMSE = 13.15 Mg ha ^-1 )。气候和地形属性是 SOCS 分布的最重要预测因子。1985 年至 2017 年间 SOCS 平均减少约 1.3 Mg ha ^-1 yr ^-1，这可能是由于土地利用变化以及农业用地扩展加剧和快速城市化的控制。该研究为土地管理和土壤质量、碳核算和碳固存潜力的决策提供了良好的基线。