Vegetation Mapping And Its Analysis In Wufendigou Site

Vegetation mapping and its spatial pattern were studied in Wufengdigou site by use ofIKONOS and ETM+ images. In order to show the advantage of high ground resolution of IKONOSand to demonstrate the characteristics of topographical fragmentation and obvious humandisturbances in the site, a lot of studies such as digital image processing of IKONOS, DEM, visualinterpretation of vegetation, extraction of specific vegetation types, vegetation mapping, andvegetation map generalization were carried out with the use of software as remote sensing imageprocessing, GIS, and Matlab. On the basis of these studies, this paper illustrated the spatial patternfeatures of vegetation in the site. The integrated characteristics of remotely sensed data withdifferent ground resolutions for the earth surface features were examined primarily as well. Theresults showed as follows. 1, According to the results of TWINSPAN classification and DCAordination, six plant community types of grassland in the site were identified, i.e. Stipa bungeana +forbs community, Stipa bungeana + Lsepedeza davurica + Cleistogenes squarrosa community,Thymus serpyllum + Lsepedeza davurica + Cleistogenes squarrosa community, Stipa bungeana +Astragalus melilotoides + Cleistogenes squarrosa community, Stipa clements community, andStipa grandis community. 2, A large scale vegetation map (1:1,5000) was mapped. With the 3-Dperspective view, the vegetation types and their distribution characteristics in the area werepresented candidly and subtly. 3, The spatial vegetation pattern derived from landscape indicesshowed that the fragmentation was severe, the diversity was high, and the shape of patches wassimple meant to towards stabilization. Planted woodlands, occupying the largest area with the mostnumber of patches, dominated the site. The smallest value of both fragmentation and dominancewere in croplands. The natural vegetation showed biggest value in fragmentation. From the slopepoint of view, frequent human activities were appeared under the slope of 25 degree, wherecharacterized by low patch fractal, fragmentation and dominance and by high diversity comparingto other slopes. Human activities impacted on spatial patterns of vegetation strongly. 4, Themethod of computer-aid recognizing Populus sp. woodland, Populus sp. open woodland, scatteredPopulus sp. trees, and planted Caragana intermedia shrub land was presented. These types enrichedvegetation information, and compensated the shortcomings of visual interpretation. 5, On the basisof 1:5,000 vegetation map, the small scale vegetation map of 1:50,000 was made using GIS tools interms of the principles of map generalization. Comparing with another vegetation map of the samescale generated by visual interpretation of ETM+ image, these two maps showed great similarity.Consequently, an automatic map generalization method for cartography was put forward. 6, Finally,the characteristics of ETM+ to synthesize features in IKONOS image was analyzed according totwo visual interpreted vegetation maps, a large scale (1:5,000) from IKONOS and a small scale(1:50,000) from ETM+.