【作者】 刘佳；

【导师】 尹海伟；

【作者基本信息】 南京大学 ， 城乡规划学， 2018， 硕士

【摘要】 自改革开放以来,我国经济飞速发展,城镇人口比重从10.64%上升到57.35%,人口的快速增长促使城市用地不断扩张,致使城市生态绿色空间逐渐被侵蚀,生态环境不断恶化。另一方面,人类活动产生的大量污染物:汽车尾气、污水、垃圾等,均需要一定规模的自然环境来承载和吸纳,而城市绿色空间不断减小,侧面加重了城市生态环境的压力,导致热岛效应加剧、空气质量恶化、洪涝灾害频发等一系列城市问题。城市原有的生态系统平衡被打破,可持续发展能力受到限制。绿色基础设施(Green Infrastructure,GI)是指一系列的人工或自然的绿色空间所组成的网络体系,是城市宜居性和可持续发展的重要空间载体,其多功能特性可提升城市多项服务功能,是改善人居环境、提升城市服务能力及促进城市生态系统良性循环的有效手段。传统的GI构建方法在分析斑块廊道相对重要性、确定不同地理空间位置的廊道宽度及多路径廊道规划上均存有一定的局限。电路理论是近年来新起的一种GI构建方法,因将随机漫步理论融合到景观连通性的度量中,所以可弥补传统GI构建方法的不足。然而应用电路理论构建城市GI格局的案例较少,对电路理论边界效应的探讨更为少见。本文以南京市为例,基于电路理论构建并优化南京市GI格局。首先,进行电路理论的边界效应分析。边界效应是指在电路理论模拟过程中,由于人工边界的存在会限制电子随机游走的空间,从而影响景观面中可能路径的识别与预测。本文通过在研究区外围设置缓冲区的方法来解除人工边界对模拟结果产生的影响。通过计算不同缓冲区宽度下电流密度相关性及曼哈顿距离,从而确定最佳的缓冲区宽度。然后,采用电路理论对南京市主城区自然生态斑块之间的景观连通性进行了定量分析,获取了自然生态斑块之间所有潜在的物种迁移扩散路径及其相对重要性(根据电流密度的大小)。其次,借助GIS软件的Linkage Mapper工具识别了研究区的主要障碍点,并利用电路理论定量分析障碍点对于景观连通性的影响大小。最后,结合以上分析结果构建南京市GI网络并有针对性地提出了 GI网络格局优化的具体策略。研究结果表明:(1)人工边界的存在使得电阻被高估,且边缘区受到的影响远大于研究区整体或核心区。通过在研究区外围增加缓冲区的方法可以有效消除电路理论模拟与实际的偏差,即使一个很窄的缓冲区(研究区宽度的4%)都可有效消除边界效应的影响。当缓冲区宽度为3000m(研究区宽度的13.3%)时可基本消除边界效应对模拟结果产生的影响。当研究区缓冲区宽度大于7000m(研究区宽度30%)时,就可完全消除边界效应的影响。南京市最佳缓冲区宽度为3000m。(2)南京主城区景观破碎化程度较高,南北部景观连通性差异较大,现存多处景观障碍点,严重影响了研究区景观连通性。通过电路理论模拟发现,40%的生境斑块(约为28.18 km2)对连通性的贡献较低,南部重要廊道的数量最多,局部簇团成网,网络结构较为复杂,其次为中部,且河流廊道(秦淮河)是其主要廊道类型,北部廊道数量最少,斑块多呈孤岛分布。借助Linkage Mapper工具共识别155处障碍点,可以发现,障碍点多位于斑块之间或斑块边缘,且其中84.5%面积小于5hm2,说明障碍点清除的可操作性较高。再次通过电路理论模拟分析发现,障碍点清除后研究区最大电流值提升近2倍,电流值小于605A的区域减少了520.38hm2,表明障碍点对于研究区景观连通性影响较大,且清除障碍点可有效提升南京市主城区景观连通性。基于以上研究结果,提取重要斑块、重要廊道及障碍点清除后对景观连通性贡献较高的区域组成南京市GI网络,并结合南京市卫星图梳理GI网络,最终将南京市GI网络划分为生态重点区、生态维护区、生态恢复区及生态改善区,并提出分级别保护重要GI要素、恢复必要的景观廊道、降低障碍点通过阻力等具体的GI优化策略。在南京市构建GI网络对于生态文明建设的推进具有重要意义,可以在提高景观连通性水平的同时提升南京市生态环境质量、提高宜居度、恢复老城原有空间尺度与肌理,改善空气污染、水污染等城市问题,实现社会、经济、生态协调、永续发展。同时,本文基于电路理论制定了一套简单易行的GI格局构建与优化的分析框架与技术方法体系,丰富了城市GI景观格局的构建方法,对其他城市GI网络的规划与建设具有一定的借鉴意义。通过电路理论的模拟分析量化环境因素对景观连通性的影响程度有助于规划管理者更科学的制定环境管理目标,具体的GI优化措施有利于指导南京市主城区GI格局的规划建设,为规划决策者提供参考。本文正文共约33 579字,图表31幅。更多还原

【Abstract】 Since reform and opening up,China has experienced a rapid urbanization process with remarkable achievements in social and economic development,in the meantime,the proportion of urban population has rose from 10.64%to 57.35%.The rapid growth of population and insufficient land area promoted urban land space expansion,which resulting in changing the nature of the underlying surface of the city and increasingly fragmenting the habitat patches.These phenomena seriously damaged ecological environment and weakened the urban ecosystem services and sustainable development ability of the city.The green infrastructure(GI)is a multifunctional green space network,which contains a variety of natural and artificial landscape and ecological elements.Green Infrastructure is the cornerstone of urban livability and sustainable development.The multifunctional characteristic of green infrastructure can improve many services of the city,and the connectivity of urban green infrastructure can significantly improve the biodiversity and sustainable development ability of cities,which is of great significance to maintaining the health and stability of urban ecosystems.However,the traditional methods of GI construction have some limitations in analyzing the importance of patches or corridores,determining corridor width of different geo-spatial position and planning of multiple pathways.The circuit theory is a new method of GI construction,which can make up for the deficiency of traditional GI construction because it integrates the random walk theory into the measurement of landscape connectivity.However,there are few cases to construct the urban GI pattern using the circuit theory,and rarely considered the effect of map boundary when using circuit theory.In the present study,we used circuit theory to map the landscape connectivity of GI in the main urban area of Nanjing.In the first step of our approach,we analyzed the effect of map boundary.Artificial boundaries on a map occur when the map extent does not cover the entire area of study;edges on the map do not exist on the ground.These artificial boundaries might bias the results of animal dispersal models by creatingartificial barriers to movement for model organisms where there are no barriers for real organisms.Here,we palced a buffer around the study area as a solution to this problem.The optimal buffer width is determined by calculating the current density correlation and the Manhattan distance under different buffer widths.Then,we used the circuit theory to quantitatively analyze the landscape connectivity between natural ecological patches in the main urban area of Nanjing,and identified all the potential pathways between natural ecological patches and analyzed the relative importance of patches and corridors according to current density.Next,we used a search window to detect barriers of the study area with the help of GIS software linkage Mapper tool,and then used the circuit theory to quantitatively analyze the influence of barriers on landscape connectivity.Finally,based on the above analysis,we constructed GI network in Nanjing and put forward a specific strategy of GI network pattern optimization landscape connectivity optimization strategy for Nanjing.The main conclusions of this study are as follows:1)Artificial map boundaries overestimate resistance values,and the map boundary caused more biases around the edge than the whole or core area of the research region.The method of placing buffer outside the research area can effectively remove the deviation of circuit theory simulation,and even a very narrow buffer(4%of the study area width)can effectively eliminate the effect of map boundary.A 3000m buffer(13.3%of the width of the study area)can basically remove the effect and when the buffer width was greater than 7000m(the width of the study area is 30%),the effect of map boundary can be completely eliminated.And the best buffer width of Nanjing is 3000m.2)Habitat fragmentation in the main urban area of Nanjing was significant.The total area of the patches was about 70.45 km2,and 40%of patches area had a low contribution to connectivity.The southern part of the study area exhibited the best landscape connectivity,followed by the middle and northern parts.In the south,numerous corridors clustered into net structure,and the network structure is complex.In the middle of the study area,the Qinhuai River mainly formed a circular corridor.Relatively few corridors were observed in the northern part of the study area,and the patches were isolated.There were 155 barriers in the study area,and 84.5%of those barriers were>5 ha,so the landscape connectivity in the main urban area of Nanjing can still improve greatly.Based on the above results,we combined the important patches,important corridors and the important area after removing barriers points into Nanjing GI network.Then,we adjusted the GI network according to satellite map and divided the GI network into ecological key area,ecological maintenance area,ecological restoration area and ecological improvement area.Finally,we put forward some specific GI optimization strategies,such as protecting important GI elements,restoring necessary landscape corridor and reducing barriers’resistance.Although many methods have been developed and used for predicting connectivity,circuit theory has rarely been used to map connectivity in urban areas.The present study demonstrates how circuit theory can be used to map connectivity in complex landscape and to identify important habitat patches and movement corridors for conservation planning,which is an advance in mapping the landscape connectivity of GI.The circuit theory incorporates all possible pathways and combines structural and functional corridors,which improves corridor redundancy and provides an efficient and cost-effective tool for mapping GI landscape connectivity.The results of this study provided an important reference for Nanjing GI networks.The whole thesis contains about33 579 words,31 pictures and charts.更多还原