【作者】 顾凤霞；

【导师】 刘文宝；

【作者基本信息】 山东科技大学 ， 地图制图学与地理信息工程， 2011， 博士

【摘要】 当今世界正在经历着一种前所未有的城市化浪潮,中国正处于城市化的快速发展阶段,城市用地的动态变化异常剧烈。伴随着城市化的持续推进和经济的快速腾飞,用地格局的动态演化也不断加速。城市用地格局是否合理可以影响城市的可持续发展、结构功能的实现和生态环境建设。天津滨海新区和深圳市是中国经济增长的重要两极,在大力发展经济时,两城市用地的规模和结构都在快速演变。同时,两地又都致力于建设现代化的生态新区。因此,本研究将滨海新区和深圳行政区作为研究对象,以地理学、规划学、生态学等多学科理论为指导,根据区域对比法、数学度量法、景观指数法等多种研究方法,利用Fragstats、ArcGIS、Photoshop.CAD、SPSS等多种计算机软件的绘图、分析和统计计算功能,对两地的城市土地利用格局的时空演变特征和规律开展研究。在研究方法上,主要从景观类型和景观指数两个角度展开。首先,以景观类型为基础,研究了两区域各种用地类型的空间分布特征,通过计算景观转移矩阵分析了各种用地之间的转入转出关系,并提出了景观收益度、景观损失度和景观波动度三个新指标,能更准确地表达景观演化的动态过程特征。其次,以景观指数为基础,提出了区域剖分理论和方法,并提出利用分布式度量和集成式度量数学指标表征景观格局随时间的演化规律。主要研究结论如下：(1)在用地景观格局的总体特征上,天津滨海新区的主要用地呈带状形态,而深圳行政区的主要用地呈环状和枝状相结合的网络形态。(2)海岸线对两城市的用地格局产生了重要影响,而且对天津滨海新区用地格局的影响更为明显。(3)在整个研究期限内,两城市都呈现出林草地、湿地等高功能生态用地减少,而城市建设用地等低功能生态用地大量增加的特点。(4)波动度比动态度能更准确地描述用地格局的动态变化。(5)土地利用在景观干预梯度下,沿研究样带形成农业景观—城郊景观—城镇景观—城郊景观—农业景观的梯度分布。(6)区域整体研究发现景观指数可以分为两大类：敏感性指数和不敏感性指数。更多还原

【Abstract】 Nowadays, under an unprecedented tide of urbanization in the world, China has experienced drastic changes of the urban landuse. With the continuous development of urbanization and economy, the evolution of landuse patterns in China is speeding up. Weather the pattern of urban landuse is rational or not will largely influence urban sustainable development, structure functioning and eco-environmental protection. Tianjin Binhai New Area and Shenzhen are two important poles of economic growth in China. The scale and patterns of landuse are changing dramatically with the economic development. Meanwhile, the two areas are devoting themselves to set up a new modern livable area, separately. Therefore, this dissertation took Tianjin Binhai New Area and Shenzhen as case study sites and investigated the regularities of their landuse change patterns in the basis of geography, urban planning and ecology using the methods of regional comparison, mathematical measure and landscape index. Some new developed software, such as Fragstats, ArcGIS, Photoshop, CAD and Excel, are used in this research to process, analyze and mapping data.The spatial-temporal changes of Landuse patterns were investigated from two perspectives:landscape classes and landscape metrics. First, the spatial-temporal distribution of landuse classes was analyzed by calculating the transition matrix that shows the relationships between move in or out of landuse classes. Meanwhile, three new landuse change measures, the degree of landuse gain, the degree of landuse loss and the degree of landuse variation were proposed in this study, which can more accurately convey the characteristics of evolution of changing landuse classes. Second, a new theory of regional subdivision was proposed to examine the spatial-temporal patterns of changing landscape metrics. The theory includes distributed measures and integrated measures of mathematical indicators to characterize landscape pattern evolution with time.Major conclusion from this research follows:(1) The general characteristics of landuse change pattern in Tianjin Binhai New Area shows a zonal pattern, while in Shenzhen it shows ring and branched forms of a combination of networks.(2) Coastal lines have profound influences on the landuse change patterns of the two sites; however in Tianjin Binhai New Area, the impact is more apparent.(3) Throughout the study period, the areas of forest/grasslands and wetlands with higher ecological functions were largely reduced, while the areas of urban land use with lower ecological functions were rapidly increased.(4) The degree of variation was a better measure than the degree of dynamic in accurate characterizing the dynamic changes of landuse patterns.(5) Between two urban centers, the spatial patterns of landuse changes show a gradient pattern: agricultural landscape-suburban landscape-urban landscape-suburban landscape-agricultural landscape.(6) Landscape metrics can be classified into two categories:sensitivity index and non-sensitive to the changes of landuse classes.更多还原