关键词: 城市生态福利绩效, 时空演变, 动态模拟, 灰色BP神经网络模型

Abstract: The promotion of urban ecological well-being performance is essential for sustainable urban development. We used the super-efficiency slacks-based measure (SBM) model based on undesired output and the global Malmquist-Luenberger (GML) index model to quantify the temporal and spatial variations in urban ecological well-being performance from static and dynamic perspectives. The panel data covering 282 cities in China from 2005 to 2020 was utilized for the analysis. Furthermore, the panel Tobit model was employed to explore the factors influencing urban ecological well-being performance. Lastly, a grey BP neural network model was used to predict the dynamics of urban ecological well-being performance from 2025 to 2035. The results showed that: (1) From 2005 to 2020, the ecological well-being performance of Chinese cities was characterized by an N-shaped pattern, with a certain degree of heterogeneity in the distribution of performance values. In terms of spatial distribution, low-level cities were primarily distributed in the northeast, central and southwest regions of China. Relatively low-level cities were mostly concentrated. Relatively high-level cities were mainly found in the northwest and southeast regions of China. High-level cities were mainly situated in the southwest region of Inner Mongolia, the northern regions of the three northeast provinces, and the southeast coast of China. There were significant differences in the number of cities of different types across these regions. (2) The annual average value of the GML index of urban ecological well-being performance was greater than 1, and the overall level of urban ecological well-being performance showed an upward tendency from 2005 to 2020. Both technical efficiency change and technical progress change were credited as having a positive impact on urban ecological well-being performance, with technical progress change being the primary influencing factor. (3) The mechanism of urban ecological well-being performance was clarified from the natural and socio-economic dimensions. Urban ecological well-being performance was positively impacted by technological progress, environmental regulation, openness, and consumption level. In contrast, industrial structure and urban-rural integration had an inhibitory effect on urban ecological well-being performance. (4) During 2025 to 2035, China’s urban ecological well-being performance levels would exhibit an upward trend with specific clustered evolution characteristics. During this period, the growth rate of low-level cities was gradually slowing down, the number of relatively low-level cities was decreasing year by year, while the number of relatively high-level and high-level cities was increasing annually. Accordingly, policy recommendations are put forward in three aspects, including the path of enhancement, the mechanism of influence, and the direction of development, to provide theoretical references for the enhancement of the performance level of ecological well-being in different types of cities and the realization of the goal of regional high-quality development.

Key words: urban ecological well-being performance, spatiotemporal evolution, dynamic simulation, grey BP neural network model