基于潜力—约束和SLEUTH模型松散耦合的南京城市扩展模拟

Urban sprawl simulation based on the loose coupling between potential-limitation and SLEUTH model in Nanjing city

将潜力—约束与SLEUTH模型进行松散耦合,在建设用地适宜性评价的基础上,将不同生态安全格局情景融入SLEUTH模型的排除图层,对研究区2013-2040年的城市用地空间扩展进行多情景模拟。研究表明:(1)两个模型的松散耦合能够有效提高像元尺度上模型校正的精度。(2)用地扩展在以"边缘增长"和"填充增长"为主的同时,在城市外围形成了较为明显的"跳跃式"发展,表明模型的松散耦合能够更好地捕捉城市发展政策所导致的新城市增长中心。(3)三种生态安全格局情景方案的城市用地模拟结果均呈增长趋势,但高生态安全格局情景的新增城市用地面积和增长率均最小,表明将生态敏感性作为城市发展的约束图层,能够有效保护研究区的自然生态空间,大幅降低生态安全风险。

Urban land growth is driven by many factors entailing spatial complexity and regional specificity. Because of difficulties in bridging the divide between transformation rules derived from bottom-up processes and top-down urban development policies using the SLEUTH model, accurate assessment of the impacts of top-down urban development policies on urban expansion is a challenging task. In contexts of top-down growth-oriented urban land management, urban land growth mainly manifests as the growth of new regions, especially as clumping growth. It is, therefore, necessary to incorporate other urban land growth models to improve the predictive ability of the SLEUTH model relating to urban development policies. We used a loose coupling of the potential-limitation and SLEUTH models to investigate urban change in Nanjing and its surroundings. Applying the potential-limitation model, we first integrated potential factors associated with urban development and ecological constraints into city development strategies, quantitatively evaluating the suitability of construction land in relation to different ecological security patterns in the study area. We subsequently, blended different scenarios for these ecological security patterns into the excluded layers, and used the SLEUTH model to simulate the urban space sprawl of this area from 2013 to 2040. We obtained the following results. First, the loose coupling of the two models improved the accuracy of calibration at the pixel scale. Moreover, reconstruction results obtained with the SLEUTH model were consistent with the actual situation. Second, ＂marginal growth＂ and ＂fill- in growth＂ patterns dominated the urban land sprawl. Simultaneously, a number of new city development centers were emerging outside the city, presenting rapid and conspicuous development. These findings demonstrate that loosely coupled models are better than the SLEUTH model, used alone, to capture the movement of the city center induced by urban development policies and the locations for the emergence of new urban growth centers. Consequently, the SLEUTH model performs better in relation to urban development policies, and is able to bridge the divide, to some extent, between conversion rules derived through bottom-up processes and top-down urban development policies. Third, while the simulation results for the three ecological security pattern scenarios indicated an increasing trend for urban sprawl, differences in their growth rates were significant. For the scenario with a high ecological security pattern, the newly added urban land area and the growth rate were both minimal, indicating that ecological sensitivity analysis applied as a layer constraining urban development could effectively protect the natural ecological space in the research area, and dramatically reduce the ecological security risk. These results can be used to guide and support future urban space sprawl management, urban planning, and land use planning decisions in the study area.