青藏高原荒漠区不同土壤类型碳密度分布图

通过在青藏高原荒漠区开展大范围、高精度的野外调查，获取土壤有机碳数据，基于经验贝叶斯克里金法（Empirical Bayesian Kriging，EBK），EBK的局部半变异函数建模能更好刻画养分含量的空间梯度变化。此外，野外土壤采样常受成本限制导致样本量有限，EBK在小数据集上的精度优势可提升制图准确性。EBK通过融合贝叶斯概率框架与局部建模思想，在土壤参数空间预测中展现出方法论优势。其自动化程度高、误差评估严谨的特点，使其成为中小尺度生态要素制图的优选工具。插值得出青藏高原荒漠区碳密度分布图，根据全国1：400万土壤类型分布图，运用区域统计工具（Zonal statistics）得出青藏高原荒漠区不同土壤类型碳密度分布图。

Large-scale and high-precision field surveys were conducted in the desert regions of the Qinghai-Tibet Plateau to acquire soil organic carbon data. Using Empirical Bayesian Kriging (EBK), the local semivariogram modeling implemented in EBK can better characterize the spatial gradient variations of soil nutrient contents. Moreover, field soil sampling is frequently constrained by costs, leading to limited sample sizes. The accuracy advantage of EBK on small datasets can enhance mapping accuracy. EBK integrates the Bayesian probability framework and local modeling philosophy, thereby exhibiting methodological superiority in spatial prediction of soil parameters. With its high automation level and rigorous error assessment, EBK has emerged as the preferred tool for mapping ecological elements at medium and small scales. Interpolation was performed to generate the carbon density distribution map of the desert regions of the Qinghai-Tibet Plateau. Additionally, based on the national 1:4,000,000 soil type distribution map, the Zonal Statistics tool was employed to derive carbon density distribution maps for different soil types within the desert regions of the Qinghai-Tibet Plateau.