中文摘要：

城乡结合部是城市化最为敏感、迅速的特殊地域实体,区域内土地快速城镇化导致耕地资源大量流失,尖锐的土地利用矛盾引发生态环境质量大幅下降,已成为社会各届关注的热点。该文以南京市郊江宁区为例,基于1999年、2010年的2期影像及其他数据,利用Logistic回归模型和地理加权逻辑回归（GWLR,geographically weighted logistic regression）模型,将邻近变量、邻域变量和政策变量3类因素在空间上予以整合,从全局、局域2个尺度探索城乡结合部土地资源城镇化的驱动模式。结果表明：Logistic模型与GWLR模型分别可解释研究区历史时期51%、64%的土地城镇化过程,后者在局部参数估计方面存在优势,各项评价参数都较优,更能够适用于研究区土地城镇化的模拟预测与机理探究;邻近变量中的距开发区、城市中心、次要干道的远近,邻域变量中城镇用地密度以及政策变量中土地交易样点密度是研究区城市扩展的重要解释变量,充分表征了城乡结合部的开发区建设热潮、已城市化地区的集聚效应与边缘式扩展模式,以及土地市场、内部交通网络对城市增长的引导与促进作用;借助GIS工具,GWLR模型能够可视化各驱动因素的空间非平稳性特征,可用来更直观、深刻地探究各变量对研究区土地城镇化的空间驱动模式,这也为耕地资源的保护提供新的视角。总体而言,以土地出让行为及开发区建设为代表的政府主导力量,是研究区土地城镇化的核心动力,这一定程度反应了中国城市郊区化的普遍特征,可为类似城乡结合部的土地管理工作提供有益借鉴。

英文摘要：

Urban expansion can be seen as a spatial non-stationary process because of the natural environment, social and economic characteristics are distinctive in different regions. The complexity of land use in urban fringe determines its unique urbanization process and mechanism, to explore the driving factors of land use change in urban fringe is not only helpful to the alleviation of land use conflicts and the optimization of urban space structure but also has a realistic significance in the protection of cultivated land resource and ecological environment. Based on remote sensing images in 1999 and 2010 and GIS data, we spatially integrated proximity variables, neighborhood variables and policy variables with global logistic regression model and local geographically weighted logistic regression （GWLR） model to explore driving factors of urban expansion in the urban fringe area with Jiangning District of Nanjing as an example. Results showed that： 1） logistic model and GWLR model respectively explained 51%, 64% of urban land change of the study area over historical periods, and the latter model had advantages in local estimation with better parameters which could be better applied to simulation, prediction and mechanism investigation of urban expansion in the study area. Moreover, land use data sampling integrated system sampling and random sampling, which could effectively reduce the influence of spatial autocorrelation on the research results. 2） Considering the two model results, X 3 , X 5 , X 1 which belong to proximity variables; X 6 which belongs to neighborhood variables and X 11 which belongs to policy variables were the key explanatory variables for urban expansion in the studied area. It indicated the agglomeration effects of industrial park and urban center in urban fringe area, as well as promotion role of the land market and internal transportation network, were important driving factors to the urban land change of urban fringe. 3） Using GIS tools, GWLR model could visualize the spatial non-s