DynamicssimulationandanalysisofYiRiverWatershedlandscapepattern

The technologies of 3S were employed to analyze spatial/temporal changes of landscape types and landscape indices at Yiluo River watershed. Using multi-band satellite remote sensing image of the area in 1985, 1995, 2002 and 2008 as the main data resource, combing with land use maps, topographic maps and GPS field survey, we built a landscape classification system and the interpretation of signs, and divided the land use type into 7 categories according to the order of research. We then established a database building upon the study area vector data, raster data and the statistical Yearbook data.. In order to simulate the landscape pattern change,Karlman Filtering method was applied for simulation using software Matlab. My thesis includes the following sections:Part I: The current situation of Simulation of landscape pattern change, and the research method.Part II: Overview of the study area.Part III: Data collation, analysis.Part IV: Landscape pattern simulation and correction.Part V: Conclusions and prospects.The results showed that using the filtering method of Karlman is feasible to simulate the landscape pattern, especially to simulate the whole basin of the study. One of the greatest features of the landscape is spatial autocorrelation (spatial autocorrelation). The observation that landscape pattern had a high degree of autocorrelation in a certain gradient possibly indicates that some kind of ecological process plays an important role. . According to the simulation results, by 2020, the forest and agriculture landscape are relative stable at the landscape of this area and the land-use of construction has increased. Other types of land use change in different direction, but all of them share in a small proportion of all land types, which results in that their change can not have a significant effect on the all valley landscape. Taken together, the stability of ecosystem in the whole region was slightly affected.