Abstract: The construction of green space ecological network is a key point for urban and rural green space to play its compound ecological efficiency, which is of great significance for promoting the sustainable development of urban and rural ecological environment. In this study, we extracted the spatial distribution information of green space in the central urban area of Chengdu to quantify the spatial connectivity characteristics of green space ecological network, and examined the optimal structure of urban green space ecological network by using morphological spatial pattern analysis (MSPA) and graph theory. The results showed that MSPA method could effectively identify the systemically important core areas and connecting-bridge areas. The green space ecological network constructed by graph theory presented the following characteristics: (1) The south and north parts of Chengdu were dominated by river corridors, with complex network structure. (2) The west part of Chengdu was dominated by isolated island green patches, with relatively less corridors. (3) The connectivity of green space network had obvious spatial differences, with a decreasing trend from west to east and from periphery to center. Finally, based on the graph metrics, important ecological sources, stepping-stone patches and important connected ecological corridors were selected in the central Chengdu to optimize the structure of green space ecological network. The optimization of green space ecological network based on MSPA and graph theory has crucial theoretical reference and method support for urban green space system planning.

Key words: morphological spatial pattern, spatial connectivity, structural optimization, urban green space, ecolo gical network.