Gis-based Southern Plateau Landscape Dynamics And Optimization

At present land use and land cover dynamics become a hot issue of domestic and international, take the land use/land cover change (land use/land cover change, LUCC) as the breakthrough point, use the historical period land use condition forecast the future situation, and according to the principles of local conditions to optimizes the land come into one of land science difficult questions. The landform of South Loess plateau is complex, the landscape type is bountiful, the city in south is crowded, the dense of population is high and the industry gathers compares with other areas of Loess plateau, These prombles causes the land ecosystem faced with the huge pressure for the past many years, the land use is not very intensive in the socio-economic development, also has the situation of sacrifice ecological environment benefit gain economic efficiency, the contradictory between human and land even more prominent.Carries on the landscape pattern dynamics and the optimized research, has the important value of theory and practice to land resource intensive reasonable use.This article take the south Loess plateau as the research area, through the open country on-the-spot investigation, use remote sensing, the GPS and the GIS technological, make use of mathematics model and space statistical analysis methods and so on, carries on the dynamic analysis and the optimization to south Loess plateau under supports of LUCC theory, landscape ecology theory, land optimization disposition theory, spatial analysis theory, mathematical statistic and so on.The main research content of this article includes:(1) pretreatment the images of remote sensing, classification and the interpretation, carry on the land use change dynamic analysis (2) has carried on scale response and scale choice research and landscape pattern dynamic analysis according to the best scale uses the GIS spatial analysis technology; (3) uses the vector grid integration data model construction landscape pattern and the influencing factor database, definite landscape pattern and influencing factor relativity; (4) carries on the landscape pattern forecast and the simulationl; (5) constructs the landscape pattern optimization model (LSOM), realizes the LOSM model through the programming in MATLAB, obtains the quantity optimization plan of landscape pattern and the space optimization distribution map of south Loess plateau.The main conclusion of this article includes:(1) The construction land area of south Loess plateau increases from 3555.99 Km2 to 4794.28 Km2 during 25 years, the cultivated area reduces from 100004.79 Km2 to 98561.57 Km2. The land use structure basically maintains stable, the main landscape type of south Loess plateau are fanning, lawn and forest.In different province territory unit and the land use regionalization unit, the land use type change important value is various, the land use type transformation show obvious region difference.(2) Discover landscape index granularity response in type level intense than landscape level; The granularity response of attribute mottling degree of fragmentation, the mottling shape, the mottling accumulation degree, the mottling connection is obvious,and obviously displays phenomenon of "the critical threshold value",150 meters and 600 meters are many kinds of landscape indices change granularity along with the granularity change, so selects 150 meters and 600 meters for the "the critical threshold value" granularity. When selects the critical threshold value granularity and below any granularity, will not have the obvious influence to the findings, this article will choose 150 meters as optimum size for the landscape pattern dynamic.(3) The landscape pattern space and time dynamic result of the class level showed that the PD of forest land, the construction land and the waters rises year by year during 25 years, the PD of virgin land reduced year by year, the PD of lawn and the cultivated land reduces first, elevates then. the PLAND of Forest land, lawn, construction land and the waters rises year by year, the PLAND of virgin land and the farming reduces year by year. The FRAC_MN of waters, virgin land, forest land, the lawn to be high equally, The FRAC MN of construction land and the cultivated land to be low. The FRAC MN of forest land, the construction land, virgin land rise year by year, The FRAC_MN of farming reduce first, elevates then. The COHESION of farming, the forest land and the lawn is high, The COHESION of construction land is lowest, The COHESION of the waters and virgin land also has not been high. The landscape pattern space and time dynamic result in the landscape level showed that the PD increases from 0.3049 to 0.31, the FRAC MN reduces from 1.0695 to 1.0693, the COHESION reduces from 99.8004 to 99.7036, the SHDI increases.(4) Uses the vector grid integration data model construct the database of south Loess plateau landscape pattern and the influence factors, discovered that uses this method can realize the multiple source data fusions, the demographic data, the GDP data and other social economy data may realize satisfactory fusion with the spatial data. (5) The autocorrelation of landscape pattern demonstrated that landscape pattern has the spatial relevance, various landscape spatial relevance changes along with the weight change.(6) Uses the classics regression model and the space regression model carries on the landscape pattern and the influencing factor relations quantitative analysis.The result indicated that the space regression model simulation precision surpasses the classics regression model, factors less associated with landscape were excluded; the visual impact of the factors of land from each other.(7) The result of landscape pattern forecast showed that in future 20 years, the landscape pattern of south Loess plateau will still have the sweeping change, the construction land area continues to expand, and still mainly distributed in plain areas for example, Guanzhong basin, Fenhe River valley land.(8) After LOSM model optimization regulation,The area of forest, lawn, waters and the construction has varying degree increase, the farming and virgin land has reduced.The innovation and improvement of this article are:(1) Carry out land use and landscape dynamic analysis in different spatial and temporal scales of south Loess plateau.(2) Carried on scale effect analysis and the best criterion choice research in south Loess plateau using the GIS spatial analysis technology (3) Used the vector grid integration model to construct landscape pattern and the influencing factor database, has solved the multiple source data fusion problem. (4) Constructed the LOSM model, has realized landscape pattern quantity optimization and the spatial pattern optimization under the GIS and MATLAB support.