地形复杂度的多因子综合评价方法

Terrain Complexity Assessment Based on Multivariate Analysis

地形复杂度指标(Terrain Complexity Index,TCI)是评价地表起伏和褶皱程度的指标,广泛应用于DEM数字地形分析、数据综合和建模、地貌分类以及DEM精度研究等领域。然而目前地形复杂度指标多数采用单一地形指标或区域统计指标,缺乏局部的复合评价指标。为此,引入多因子分析方法和局部窗口分析方法,探讨一种基于格网DEM数据的复合地形复杂度指标(Compound Terrain Complexity Index,CTCI)的建模方法。首先利用多因子评价方法选取4种局部地形因子(局部高差、局部标准差、局部褶皱度、局部全曲率),之后利用局部窗口分析方法获取各指标的计算值,最后融合4种因子得到每个格网的CTCI。在实验分析中,选取了3个典型地貌样区和1个混合地貌样区,实验结果表明:CTCI能从整体上区分不同典型地貌区的地形复杂程度,同时CTCI在局部范围与混合地貌样区的等高线的密度和变化程度有较好的吻合,表明CTCI能从整体和局部反映地表的起伏和褶皱变化,是较好的地形复杂度评价指标。

Terrain complexity,which describes rugousness and complexity of the terrain surface, is not only an im- portant terrain parameter in digital terrain analysis, but also widely applied in the fields of reduction of topographic data and DEM modeling, terrain classification and DEM accuracy modeling and assessment. Unfortunately, terrain complexity is an ambiguous terrain feature that tills now no optimal index to quantify it. The traditional terrain com- plexity indices evaluate terrain complexity only from one perspective of geomorphometry or just give a statistical val- ue （global value）, which will cause more or less prejuclice when modeling the real landform. This paper wants to seek a local compound terrain complexity index to quantitatively evaluate topographic complexity based on grid DEM. According to multivariate analysis method, we firstly selected 4 different traditional indices that can easily be evaluated by local kernel window, then deduce the compound terrain complexity index （CTCI） using indices fu- sion. In order to validate model, we selecte 4 study areas with 3 typical landform and 1 hybrid landform, then com- pute the CTCI of each test areas and compare the CTCIs with contour lines and DEM hill shade map, tile result shows the CTCI can roughly differentiate the contour line where is cliff and dense and where is smooth and sparse, which indicates CTCI is a sound terrain parameter to evaluate terrain complexity.