Abstract: Supported by remote sensing (RS) and the geographic information system (GIS) and the data of high-resolution QuickBird satellite images in 2002 and 2005, the dynamic changes of green space landscape pattern in Shenzhen special economic zone (Shenzhen SEZ) were analyzed  using landscape index and transfer matrix methods. The Markov Chain model was used to predict the trends of green space in the subsequent six years. The results showed that the primary characteristics of green space landscape were increasing artificial green space and decreasing natural green space, increasing fragmentation degree of green space patches, increasing index of diversity and evenness with patch shape regularized, decreasing dominance, and trends of diversification and evenness of the landscape. It is predicted that the total area of green space would increase and the ecological benefit of natural green space area would decrease by 2011.