The spatial characteristics and relationships between landscape pattern and ecosystem service value along an urban-rural gradient in Xi’an city, China

Abstract

• Combined with concentric zone theory, buffer zones were built up in this study. • Landscape pattern indexes (LPIs) which are not influenced by area were selected. • The gradient changes of LPIs and ecosystem service values (ESVs) were analyzed. • Nine ESV types and ten LPIs were adopted to construct models. The spatial structure and function of Xi’an city is significantly affected by urbanization, a factor which can be considered as the main driver of landscape pattern and ecosystem service change. Due to these changes are a response to urban land use and land cover (LULC), remote sensing images are interpreted by the method of supervised classification and visual interpretation to obtain the LULC data for research on landscape pattern index (LPI) and ecosystem service value (ESV) of Xi’an city, China. Combined with urban planning theory, concentric buffer zones were used to explore the characteristics and relationships between LPIs and ESVs along an urban-rural gradient. Ten landscape indices and nine ecosystem service types were selected to analyze the correlation and construct multiple linear regression models, and principal component analysis (PCA) was used to eliminate the problem of multi-collinearity in the process of model construction. Results indicate that the highest landscape fragmentation was mainly distributed in the urban-rural fringe, 20–35 km from the urban center, and patch density (PD), edge density (ED), Shannon's diversity index (SHDI), and landscape shape index (LSI) recorded the highest values. Total ESV of Xi’an city was 20493.65 × 10 6 CNY in 2016, and the mean value of ESV increased from 0.06 × 10 6 CNY to 2.60 × 10 6 CNY along an urban-rural gradient. This finding indicates that a higher ESV was recorded for the natural landscape. Results also indicated that SHAPE_MN, FRAC_MN, PLADJ, and AI recorded a positive effect on total ESV whilst ED and PD have a negative effect on total ESV. Results from the regression models showed quantitative relationships between ESVs and LPIs which revealed how ecosystem service values were affected by the landscape pattern. This study will improve the quantitative assessment method on landscape ecology and provide a basis for further research on city’s landscape pattern and ecosystem balance, especially for detecting and evaluating urban-rural development.