Grain Size Effect of Landscape Metrics in Wuxi City
WU Wei, FAN Shi-wei, XU Li-ping, ZHANG Min, OU Ming-hao
2016, 31 (3):
413-424.
doi: 10.11849/zrzyxb.20150303
Wuxi city in the Yangtze River Delta Area is taken as a representative case of urbanizing regions of China with great pressure of biodiversity conservation and environmental protection. The multi-weight factors model based on natural and artificial factors was applied in the process of identifying local ecological contribution pattern patches. Five main factors, i.e., slope, height, land use type, distances to settlements and traffic network, and their relative weights were obtained from previous results. The basic spatial cell unit is 30 m. By the scale the landscape pattern was converted to ArcGrid formats with grain sizes of 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, 30, 40, 50, 60, 70, 80, 90 and 100 cells. Above two steps were computed in ArcGIS 10.0 environment based on the dataset interpreted from TM images. Landscape metrics were used to detect the grain size effects of ecological pattern. Five landscape connectivity indicators, including number of links (NL), number of components (NC), integral index of connectivity (IIC), probability of connectivity (PC) and importance value of PC (dPC), were computed in ConeforSensinode 2.2 environment. Landscape metrics at class and landscape levels, including total class area (CA), number of patches (NP), patch density (PD), largest patch index (LPI), landscape shape index (LSI), perimeter area fractal dimension (PAFRAC), aggregation index (AI), splitting index (SPLIT), mean shape index (MSI), area-weighted patch fractal dimension (AWMPFD), cohesion index (COHESION), division index (DIVISION), Shannon's diversity index (SHDI), Shannon's evenness index (SHEI), were computed in Fragstats 4.0 environment. The results showed that with the increase of grain size, these metrics changed dramatically, and there existed scale domains of landscape metrics. The scale domains of landscape metrics at class and landscape levels were 2-30 cells and 2-10 cells respectively, and that of landscape connectivity indicators was 2-7 cells. The scale domain of 2-7 cells, i.e. 60-210 m, was recommended. The scale domain of landscape connectivity indicators was more precise compared with those of other landscape metrics. Landscape connectivity indicators were suitable for the research of grain size effect. However, it should be noticed that the response degrees of different landscape connectivity indicators were different.
References |
Related Articles |
Metrics
|