应用土壤-景观定量模型预测土壤属性空间分布及制图

Prediction and Mapping of Spatial Distribution of Soil Attributes by Using Soil-Landscape Models

以土壤-景观定量模型为基础的土壤制图方法在世界范围内得到了广泛研究。本文在皖南宣城的丘陵地带内选择研究区,从该区的数字高程模型(DEM)中获取景观信息:地形因子,定量地分析了土壤属性与地形因子之间的相关关系,并建立基于该关系的线性土壤-景观定量模型,最后应用该模型来预测土壤属性在空间上的分布并制图。结果表明:土体厚度和表层有机质含量与地形因子之间有着显著相关性;建立的线性回归模型分别能解释土体厚度、表层有机质含量空间变异的32.2%和35.3%;依据该模型预测的土体厚度和表层有机质含量具有较高的准确度,并能制图表达土壤属性在空间上的自然连续性。

Soil mapping based on quantitative soil-landscape model has been studied all over the world. This paper chose a hilly area in Xuancheng, southern part of Anhui province, as the study site, and produced soil mapping based on quantitative soil-landscape models relying on quantitative relationships between soil and landscapes. The results showed that there were strong correlations between solum thickness, organic material content in topsoil and some of topographic indexes. Quantitative soil-landscape models could explain 32.2% and 35.3% of the spatial variations of solum thickness and organic content in topsoil, respectively. The predicted values of solum thickness and organic content in topsoil on validation points by the two soil-landscape models matched the determined in a high degree. Additionally, the maps produced by the two soil-landscape models kept the natural spatial continuity of soil properties, which cannot be captured on traditional soil maps.