[Applications of 2D and 3D landscape pattern indices in landscape pattern analysis of mountainous area at county level].

Landscape pattern indices are the commonly used tools for the quantitative analysis of landscape pattern. However, the traditional 2D landscape pattern indices neglect the effects of terrain on landscape, existing definite limitations in quantitatively describing the landscape patterns in mountains areas. Taking the Qixia City, a typical mountainous and hilly region in Shandong Province of East China, as a case, this paper compared the differences between 2D and 3D landscape pattern indices in quantitatively describing the landscape patterns and their dynamic changes in mountainous areas. On the basis of terrain structure analysis, a set of landscape pattern indices were selected, including area and density (class area and mean patch size), edge and shape (edge density, landscape shape index, and fractal dimension of mean patch), diversity (Shannon's diversity index and evenness index) , and gathering and spread (contagion index). There existed obvious differences between the 3D class area, mean patch area, and edge density and the corresponding 2D indices, but no significant differences between the 3D landscape shape index, fractal dimension of mean patch, and Shannon' s diversity index and evenness index and the corresponding 2D indices. The 3D contagion index and 2D contagion index had no difference. Because the 3D landscape pattern indices were calculated by using patch surface area and surface perimeter whereas the 2D landscape pattern indices were calculated by adopting patch projective area and projective perimeter, the 3D landscape pattern indices could be relative accurate and efficient in describing the landscape area, density and borderline, in mountainous areas. However, there were no distinct differences in describing landscape shape, diversity, and gathering and spread between the 3D and 2D landscape pattern indices. Generally, by introducing 3D landscape pattern indices to topographic pattern, the description of landscape pattern and its dynamic change would be relatively accurate.