The Study Of The Constructing And Simulation Of Multi-Objective Land Use Change Model Based On The Target Motad-Clue-s Model

The soil erosion of Loess Plateau is very serious in China, irrational land use and human activity further deteriorated the local fragile ecological environment, posing a grave threat to ecological security and sustainable development of the regional economy of the Loess Plateau. Combine bottom-up farmers land use decision and top-down allocation of land, can help us to a deeper understanding of the macroscopic land use change’s microcosmic drive mechanism.The land use map Gao Xi-gou village is obtained based on aerial photograph of Gao Xi-gou village of2000and GPS survey data of2009-2011. Combining3S techniques, this article analysed the land cover change in the past50years. After that this article calculated different types land demand using the target-MOTAD model. On this basis, using the CLUE-S model to simulate the future Gao Xi-gou village land use change. We can get following conclusions from this paper:(1) The land use types of Gao Xi-gou village take dramatic changes in the past50years. In this period, cultivated land area was greatly reduced and zoology forest、grassland increased significantly. Little change in residential land and transportation land area, and a slight increase in ecological forest area.(2) The conversion of and use types have significant difference between different periods and the flow characteristics significantly different, the land use structure change of1978-2000is the most intense period throughout the study period. Cultivated land is the main land use change type, and the main flow is from cultivated land to the grass and zoology forest.(3) In accordance with different sources of income to farmers, the households are classified as pure peasant households, I concurrent households and II concurrent households. Different types of households have significant difference in agricultural labour input, land investment and capital invested.(4) For the purposes of maximization of benefits and minimize the risk, using the target MOTAD model to simulate different target income scenarios to optimize agricultural production. The results show that the economic scenario of land use prediction is higher goodness of fit.(5) Separately analysis various land use types and the driving factors with Logistic model, the ROC test results are greater than0.7, indicating that the simulation goodness of fit is well.(6)Using the CLUE-S model to simulate the2011land-use change, and the correct rate is0.87and Kappa index is0.85.2015land-use change simulation shows that the future land use structure will not significantly change.Arable land will reduce a little and grassland forest area will a slight increase, and ecological forest will increase significantly.