基于InSAR优化的DEM空间分辨率对坡度、坡长因子的影响  

Effect of DEM spatial resolution on slope and slope length factors based on InSAR optimization

作　　者：杨荣康 张静[1] 高驭洋 梁明月 王晨沣 王小平[1] 刘梦云[1] Yang Rongkang;Zhang Jing;Gao Yuyang;Liang Mingyue;Wang Chenfeng;Wang Xiaoping;Liu Mengyun(Northwest A&F University,Yangling,Shanxi 712100,China)

机构地区：[1]西北农林科技大学,陕西杨陵712100

出　　处：《水土保持研究》2025年第3期188-195,205,共9页Research of Soil and Water Conservation

基　　金：国家重点研发计划(2022YFE0115300);一般项目—青年项目(2024JC-YBQN-0263)。

摘　　要：[目的]优化数字高程模型(DEM)数据精度,探究不同尺度效应下的DEM数据对地形因子(LS)计算及修正土壤流失方程(RUSLE模型)的影响,为黄土高原的生态治理提供理论依据。[方法]选取黄土高原地区的皇甫川流域作为研究区,利用合成孔径雷达(Synthetic Aperture Radar,SAR)获取的皇甫川流域2018—2022年逐年15 m空间分辨率DEM数据,以15 m正方形栅格作为DEM基准,设置梯度下降的栅格尺寸探究LS因子的不同尺度效应。[结果](1)随着DEM的空间分辨率从15 m下降至90 m,坡度因子(S因子)的效应量增加,坡长因子(L因子)的效应量则减小;(2)LS因子的归一化结果表明:RUSLE模型的整体平均绝对误差在DEM空间分辨率为50 m时达到计算阈值,此分辨率下LS两因子效应量接近,所计算的地形因子参与的土壤侵蚀模数评估具有优势;(3)皇甫川流域土壤侵蚀模数在水系周围出现高值,且南部土壤侵蚀模数高于北部。[结论]DEM数据的精度在一定范围内会出现阈值,在阈值处计算LS因子可显著提高其精度,将高精度的LS因子代入RUSLE模型中,能够进一步提升土壤侵蚀模数的计算精度,这对黄土高原的生态治理具有重要的支撑意义。[Objective]The study aims to optimize the accuracy of Digital Elevation Model(DEM)data and explore the effects of DEM data at different scales on topographic factor(LS)calculations and the Revised Universal Soil Loss Equation(RUSLE model),thereby providing a theoretical basis for ecological management on the Loess Plateau.[Methods]The Huangfuchuan Basin,located in the Loess Plateau,was designated as the study area.Synthetic Aperture Radar(SAR)data was used to obtain DEM data with a 15-meter spatial resolution for the basin from 2018 to 2022.A 15-meter square grid was established as the DEM reference,and decreasing resolution grids were evaluated to examine the scale effects on the LS factors.[Results](1)As the DEM spatial resolution decreased from 15 m to 90 m,the effect of the slope factor(S factor)increased,while the effect of the slope length factor(L factor)decreased.(2)The normalized results of the LS factor showed that the overall average absolute error of the RUSLE model reached the accuracy threshold when the DEM spatial resolution was 50 m.At this resolution,the effects of both LS factors were nearly equal,and the calculated topographic factors provided more accurate estimates of soil erosion modulus.(3)The soil erosion modulus in the Huangfuchuan Basin showed higher values around the water system,with the southern region exhibiting greater soil erosion modulus compared to the northern region.[Conclusion]Within a specific range,the accuracy of DEM data reaches a threshold.Calculating LS factors at this threshold significantly enhances their precision.Incorporating high-precision LS factors into the RUSLE model further improves the accuracy of soil erosion modulus calculations,which is crucial for the ecological management of the Loess Plateau.

关 键 词：数字高程模型 尺度效应 LS因子 RUSLE 土壤侵蚀 

分 类 号：P237[天文地球—摄影测量与遥感] S157.1[天文地球—测绘科学与技术]