DEM空间插值方法对土壤侵蚀模拟的影响研究——以USPED分析干热河谷典型冲沟为例

Effects of different DEM spatial interpolation methods on soil erosion simulation:A case study of a typical gully of dry-hot valley based on USPED

为探索不同空间插值方法得到的DEM如何影响土壤侵蚀模拟效果,本文选择金沙江干热河谷区典型冲沟为研究对象,利用野外测量高程数据,采用反距离加权(IDW)、析取克里格(DK)、局部多项式(LPI)和张力样条函数(ST)4种方法构建高精度DEM。基于USPED模型模拟冲沟的土壤侵蚀,对比不同空间插值方法的精度、土壤侵蚀的空间分布,采用相对差系数对比不同插值在土壤侵蚀研究中的相似性。结果表明:DEM空间插值的精度排序为ST<IDW<LPI<DK。基于USPED模拟的土壤侵蚀结果显示,DK模拟了主要的侵蚀、沉积分布,IDW突出了局部细节,LPI和ST则介于其间。相对差系数结果显示,IDW得到的DEM侵蚀模拟结果与其他插值方法有较高的相似性。在实测数据布局合理、密度较高的前提下,IDW更适合应用于土壤侵蚀模拟研究。

Digital elevation models （DEMs） have been widely used in soil erosion simulation, but how different interpolation methods affect the simulation results is not well researched. In this article, a typical gully in the dryhot valley of the Jinsha River was selected as the case study area. First, field measurement of elevations of the gully was preformed. Inverse Distance Weighted （IDW）, Disjunctive Kriging （DK）, Spline with Tension （ST）, and Local Polynomial Interpolation （LPI） methods were then used to produce high-precision DEMs of the gully. The United Stream Power-based Erosion Deposition （USPED） model was used to simulate the spatial distribution of erosion and deposition in the gully. The coefficient of relative difference was used to analyze the similarity of USPED results. The result shows that the precision of DEMs from high to low is： DK, LPI, IDW, and ST. The results of USPED simulation show that DK revealed the main distribution pattern of erosion and deposition, but IDW highlighted the details, and LPI and ST were in-between. The coefficient of relative difference shows that IDW was similar with other interpolation methods in soil erosion simulation, but when the layout of sampling points is reasonable and the density of the sampling points is sufficiently high, IDW is more suitable for soil erosion simulation using USPED.