Matrix Land Use Analysis - a Case Study of Landscape Changes in Israeli Carmel Area

The spatial redistribution of the land uses can be measured by remote sensing conventional methods in the form of the matrices of the land uses redistributions within a given set of regions in a given time period. Two new methods of analysis of such land uses redistribution matrices are proposed. The first method represent the geometric and analytical algorithm of decomposition of the land use redistribution matrix into the convex combination of the land use matrices which represents the main tendencies of land use redistributions in a given set of regions in a given time period. Thus, each empirically given land use redistribution can be presented as a superposition of the land uses redistributions connected with the optimal solutions of some extreme land use redistributions corresponding to the parsimonious behavior of land users in a given set of regions in a given time period. The second method represents the construction of the artificial land use landscape corresponding to the minimum information land use redistribution with fixed initial and final land use distributions. The comparison of the empirical land uses landscape with the artificial oner can be acheived by measurement of the entropy limitator of homogenuity of land uses redistribution. This limitator represents the spatial specifics of an actual land use redistributions connected with different parsimonious behavior of the land users themselves. As an empirical validation of these new methods the set of 10 different regions in the vicinity of Haifa Carmel area is chosen for different time intervals and the main tendencies of land use redistributions are identified together with their minimum information artificial landscapes. The new methodology and modeling approach will assist future planning in the rural-urban fringe. Optimal solutions for nature conservation and urban development conflicts can be learned through the application of these models.