Trade-offs and synergistic relationships of ecosystem services under land use change in Xinjiang from 1990 to 2020: A Bayesian network analysis

Abstract

Ecosystem service value (ESV) refers to the value of benefits provided by the ecosystem to people, and can reflect the quality of regional ecological environment. There have been few studies on ESV in arid regions experiencing dramatic land use changes. Also, many past ESV studies have obtained distorted results by using a simple linear function to examine the trade-offs between driving factors. This study quantified ESV in Xinjiang from 1990 to 2020 based on value equivalent method. Differences in ESV among ecosystem services in Xinjiang under different scenarios were simulated using a Bayesian network model. The results demonstrated land use changes in Xinjiang from 1990 to 2020, with construction land expanding the most significantly (dynamic index: 224.63 %), whereas grassland area decreased (dynamic index: -1.31 %) due to transformation to unused and cultivated land. ESV in Xinjiang presented an N-shaped variation trend from 1990 to 2020 and decreased by 309.6 × 10⁸ CNY, with a variation rate of -20.35 %. The rank of the four categories of ecological services from 1990 to 2020 in terms of ESV was: regulating services > support services > cultural services > supply services. There was a gradual reduction in ESV in Xinjiang from 1990 to 2020. The rank of the different regions in terms of the reduction in ESV was: Northern Xinjiang (295.24 × 10⁸ CNY) > Southern Xinjiang (280.94 × 10⁸ CNY) > Eastern Xinjiang (109.76 × 10⁸ CNY). Land use change was a direct driver of changes in ESV, whereas natural and social factors, such as precipitation, temperature, population, and policy factors, were indirect drivers. This study can act as a reference for sustainable management of ecosystem services in arid regions.

Keywords: Bayesian network; Ecosystem services; Land use changes; Spatiotemporal pattern; Trade-offs.