VECTOR ANALYSIS THEORY ON LANDSCAPE PATTERN （VATLP） IN SANJIANG PLAIN MARSH, CHINA

Landscape indices are popular for the quantification of landscape pattern. But all landscape indices being used so far are scalar quantity, which measure patterns without considering sufficiently the pattern size and the directionality together. Based on planar characteristics defined in mechanics such as centroid, moment of inertia, product of inertia, principal axes and so on, vector analysis theory on landscape pattern (VATLP) is explored here. Firstly we establish a coordinate system of centroidal principal axes (CSCPA) of a patch or patches. Some related new indices including those describing the direction of pattern distribution (patch orientation (PO), vectorial patch orientation (VPO)), and those indicating the shape of patch‘s equivalent ellipse (major axis (MJA), minor axis (M1A) and eccentric rate (ER)) are deduced. These landscape metrics are then applied to the pattern analysis of Sanjiang plain marsh, the study area. Two temporal vector-based data sets of the study area come from interpretation of remote sensing images M SS (1980) and T M (2000). The application of the theory captures some shape properties of riparian wetland in Sanjiang plain marsh. The dissymmetrieal featires of Sanjiang plain marsh around principal axes due to agricultural development could also be explained.

VECTOR ANALYSIS THEORY ON LANDSCAPE PATTERN (VATLP) IN SANJIANG PLAIN MARSH,CHINA

ndscape indices are popular for the quantification of landscape pattern. But all landscape indices being used so far are scalar quantity, which measure patterns without considering sufficiently the pattern size and the directionality together. Based on planar characteristics defined in mechanics such as centroid, moment of inertia, product of inertia, principal axes and so on, vector analysis theory on landscape pattern ( VATLP) is explored here. Firstly we establish a coordinate system of centroidal principal axes (CSCiPA) of a patch or patches. Some related new indices including those describing the direction of pattern distribution (patch orientation (PO), vectorial patch orientation ( VPO)), and those indicating the shape of patch's equivalent ellipse (major axis (MJA), minor axis (MIA) and eccentric rate (ER)) are deduced. These landscape metrics are then applied to the pattern analysis of Sanjiang plain marsh, the study area. Two temporal vector-based data sets of the study area come from interpretation of remote sensing images MSS (1980) and TM (2000). The application of the theory captures some shape properties of riparian wetland in Sanjiang plain marsh. The dissymmetrieal featires of Sanjiang plain marsh around principal axes due to agricultural development could also be explained.