青藏高原土壤容重数据

容重（BD）是理解土壤压实、元素储量和生态过程的重要土壤属性。然而，获取大面积异质性区域的准确BD数据仍然是一个艰巨的挑战，这主要是由于测量工作费时费力，特别是在青藏高原（QTP）等复杂环境中。本研究通过提出新的土壤转换函数（PTF）与数字土壤制图技术相结合，在剖面尺度上预测BD值，绘制了青藏高原BD的空间分布图。基于递归特征消除法选择最优环境变量。结果表明，我们的方法显著提高了BD预测的准确性。新开发的PTF对BD的预测精度R²为0.70。我们制作了6个深度（0-5、5-15、15-30、30-60、60-100和100-200 cm）空间分辨率为90 m的BD数据产品。预测准确度从良好到中等（R²从0.58到0.34）。BD数据产品填补了QTP土壤数据库中BD数据的空白，为评估土壤保水能力和养分估算提供了有力支持。

Bulk density (BD) is a critical soil property for understanding soil compaction, element stocks, and ecological processes. However, acquiring accurate BD data over large, heterogeneous areas remains a formidable challenge, primarily because field measurements are time-consuming and labor-intensive, especially in complex environments such as the Qinghai-Tibet Plateau (QTP). In this study, we combined a newly proposed soil pedotransfer function (PTF) with digital soil mapping techniques to predict BD values at the profile scale and produce the spatial distribution map of BD across the QTP. Optimal environmental covariates were selected via recursive feature elimination (RFE). The results demonstrate that our approach significantly improves the accuracy of BD prediction. The newly developed PTF achieved a coefficient of determination (R²) of 0.70 for BD forecasting. We generated BD data products with a spatial resolution of 90 m for six soil depth layers: 0–5, 5–15, 15–30, 30–60, 60–100, and 100–200 cm. The prediction accuracy ranges from good to moderate, with R² values varying from 0.58 to 0.34. This BD dataset fills the gap in BD records within the QTP soil database, providing robust support for soil water-holding capacity evaluation and nutrient stock estimation.

搜集汇总

背景与挑战

背景概述

该数据集提供了青藏高原土壤容重的空间分布数据，覆盖6个不同深度（0-200cm），空间分辨率为90m，数据大小为5.65 GB，采用开放获取方式共享。数据集通过新的土壤转换函数和数字土壤制图技术提高了预测准确性，填补了青藏高原土壤数据库中容重数据的空白。