A GIS-based spatial extent in gradient analysis of urban green space pattern in Shenyang, China

Quantification of green space patterns is essential for the monitoring and assessment of ecological consequences of urbanization. Using RS, GIS and landscape structure analysis program FRAGSTATS, based on the urban green space type map by Quickbird image of Shenyang in 2008, combined gradient analysis with landscape metrics, this paper aimed to quantify the spatial pattern of urban green space in Shenyang city and investigate the effects of spatial extend in analyzing urban green space landscape pattern. The patch density, landscape shape index showed significant gradient characteristics, while the large patch index, Shannons diversity index, and the percentage of landscape, landscape aggregation index were in adverse. The landscape metrics had different variation patterns with the variation of spatial extend. The results of transect analysis with landscape-level metrics showed that the spatial pattern of urban green space could be reliably quantified using landscape metrics with a gradient analysis approach. It was suggested that a spatial extent of 5km might be more appropriate for urban green space pattern analysis by landscape metrics changing among different window sizes.