The spatial granularity effect, changing landscape patterns, and suitable landscape metrics in the Three Gorges Reservoir Area, 1995-2015

The Three Gorges Project (TGP) is a mega-water conservancy and hydropower project that have attracted worldwide attention. Under the influence of human activities, the landscape pattern has changed greatly. Using landscape metrics to analyze landscape pattern changes is the critical to understanding landscape pattern and ecological process. In this study, Landsat TM/ETM+/OLI images are used to extract indices for area edge, shape, spatial distribution, and diversity. Additionally, 19 landscape indices are taken to scale the granularity of the landscape level and the class level. An analysis of appropriate size is used to quantify the response of landscape indices to change landscape patterns during the four stages and a Crosstab model is used for pattern presentation. Finally, landscape indices are analyzed utilizing of Pearson correlation method to identify the landscape indicators suitable for the Three Gorges Reservoir. We have found that both the class level and landscape level indices are sensitive to grain size changes. Our results suggest that the suitable grain size is 30-60 m. Furthermore, the indicators suitable applied to analyzing changes in the landscape patterns in the TGR are Class Area (CA), Number of Patch (NP), Largest Patch Index (LPI), Landscape Shape Index (LSI), Effective Mesh Size (MESH), and Patch Density (PD). The main reasons for the change of landscape pattern in the Three Gorges Reservoir are land consolidation project, returning farmland to forest project, nature reserve construction, hydropower station construction and urbanization development.