基于反地形坡长地表汇流模拟逼近算法

Inverse topography-based surface confluence simulation approximation algorithm of slope length

坡长是土壤侵蚀领域中重要的地形因子之一。为寻找一种逼近真实坡长的算法,本文提出了一种基于反地形的地表汇流模拟近似算法来进行模拟演算,并与LS_TOOL和SAGA工具中的坡长计算方法进行比较,以验证该新算法的有效性。本研究基于5 m分辨率的DEM数据,利用LS_TOOL、SAGA工具中的坡长算法以及本文提出的坡长逼近算法分别提取陕西黄土高原甘泉和延川2个典型样区的坡长结果,利用相对差异系数和线性回归来对3种算法的结果进行相关性分析。结果表明,3种算法的计算结果整体差异较小,每2种算法间的相对差异系数以及线性回归R2值均在0.9以上,表明近似算法与其他2种算法的结果吻合度较高。本研究为估计土壤侵蚀状况中坡长计算提供了借鉴,为获取更高精度的坡长结果,可以计算上述3种方法的坡长平均值作为最逼近真实的坡长结果,以此来提高土壤侵蚀研究的效率。

Slope length is one of the important topographic factors in the field related to soil erosion.In order to find a new algorithm to approximate the real slope length results,this paper proposes an inverse topography-based surface confluence simulation approximation algorithm to perform simulation calculations and compare it with the slope length calculation methods in LS_TOOL and SAGA tools to verify the effectiveness of the new algorithm.In this study,based on the 5 m resolution DEM data,the slope length results of two typical sample areas,Ganquan and Yanchuan,in Shaanxi Loess Plateau,were extracted using LS_TOOL algorithm,the slope length algorithm in SAGA tool and the slope length approximation algorithm proposed in this paper,in total,and the relative coefficient of variation and linear regression were used to analyze the correlation of the results of the three algorithms.The results showed that the overall differences between the results of the three algorithms were small,and the R^(2) values of the relative coefficients of variation and linear regression were above 0.9 for each of the two algorithms,indicating that the results of the approximation algorithm and the other two algorithms were in good agreement.This provides a reference for the calculation of slope length in estimating soil erosion conditions,and the average of the slope length results of these three methods can be calculated as the closest to the real slope length results in order to obtain more accurate slope length results,which can improve the efficiency of soil erosion research.