Abstract: Rational green space landscape pattern and proper plant community composition ar e essential to the ecological effect of urban green space system, and also, a ve ry important aspect in the planning and design of the system, especial ly for its construction in eco-factory. Under the supports of remote sensing an d GIS, the green space landscape pattern and the plant community composition in twelve functional units in Baoshan Iron & Steel Co., Ltd. were analyzed, and the results demonstrated that there were three dominant green space landscape types in the Company, i.e., tree-shrub, tree-shrub-grass, and shrub-grass. In par ts of the 12 functional units, higher degrees of landscape fragmentation and of the isolation between green patches were observed, which would apparently affect the ecological functions and ecosystem services of the green space. Aiming at t hese problems, some suggestions including enlarging green patch area, increasing landscape connectivity with green corridors, complicating plant community struc ture,increasing more indigenous species, and enhancing green biomass in three dimensions were put forward to further improve and optimize the green space land scape pattern in the Company.

Key words: Fargesiadenudata, Soil nutrient pool, Available nutrient dynamics, Density