【作者】 王海涛；

【导师】 董雅；

【作者基本信息】 天津大学 ， 风景园林学， 2017， 博士

【摘要】 快速城市化进程中剧烈的城市景观格局变化导致土地资源的不合理布局、热环境恶化、空气环境污染严重、生物多样性遭到破坏等生态问题。从而严重威胁到高密度城市人群的生活环境和身体健康,影响区域生态安全和可持续发展。因此对城市景观格局的演变及其环境效应进行研究迫在眉睫。本论文选择天津地区作为研究对象,以大数据理论的视角,利用遥感、地理信息系统和统计分析等方法研究景观要素变化情况和相应的热环境和空气环境相关性。首先,对1992年、1999年、2006年和2013年四期遥感影像进行解译,采用最大似然法将遥感影像分为城镇建设用地、绿地、水体和其他用地。采用景观指数法和移动窗口法计算景观水平指数和类型水平指数,在面积/密度、形状指标、蔓延度指标和多样性指标范围内分析景观要素的时空分布。对研究区分布进行圆梯度分析和八方向梯度分析,得到各景观要素的景观比例、斑块密度和最大斑块指数等指标的空间变化规律。以2006年和2013年遥感影像为基础,通过CA-Markov模型模拟预测2020年景观格局分布图,分析景观要素的转移矩阵。其次,采用辐射传输方程法反演遥感图像的地表温度,划分热岛景观等级,分析热岛年际演化特征和季节演化特征。建立六条温度剖面线,探讨温度剖面线与景观类型的相关关系,得到年际温度剖面线、季节温度剖面线和景观类型的相关关系规律。统计温度剖面线与遥感影像对应点的数据,在SPSS中检验LST和NDBI、NDVI、NDWI的Pearson双侧显著性,得到两者的回归方程式。通过定量分析景观类型归一化指数、异质性指数等景观格局指数与地表热环境的相关关系,得出两者的定量耦合规律。再次,根据统计资料和遥感影像,采用空间插值法和遥感反演法反演NO₂、SO₂、PM₁₀、PM_2.5等主要污染物的时空分布。对影响城镇建设用地的社会经济驱动因素进行定性和定量分析,通过主成分分析法得到景观格局演变的驱动机制规律。最后,综合分析景观格局时空演变与热岛效应和环境污染的相关关系,提出城市规划建议。更多还原

【Abstract】 The change of the urban landscape pattern in the process of rapid urbanization has led to the irrational distribution of land resources,the deterioration of the thermal environment,the serious pollution of the air environment and the destruction of biodiversity.Thus,it seriously threatened the living environment and physical health of high-density urban population,affecting regional ecological security and sustainable development.Therefore,it is urgent to study the evolution of urban landscape pattern and its environmental effect.In this paper,we select Tianjin area as the research object,and use the methods of remote data,geographic information system and statistical analysis to study the change of landscape elements and the corresponding thermal environment and air environment.Firstly,the remote sensing images of 1992,1999,2006 and 2013 were interpreted,and the maximum likelihood method was used to divide the remote sensing images into urban construction land,green space,water and other land.The landscape index and the horizontal index were calculated by using the landscape index method and the moving window method.The spatial and temporal distribution of landscape elements was analyzed in the area of area/density,shape index,spread index and diversity index.The spatial distribution of the landscape scale,patch density and maximum patch index of the landscape elements were obtained by circular gradient analysis and eight direction gradient analyses.Based on the remote sensing images of 2006 and 2013,the CA-Markov model is used to predict the distribution pattern of landscape pattern in 2020,and the transfer matrix of landscape elements is analyzed.Secondly,the surface temperature of the remote sensing image is retrieved by the radiation transport equation method,and the thermal island landscape grade is divided,and the annual evolution characteristics and seasonal evolution characteristics of the heat island are analyzed.Six temperature profiles were established,and the correlation between temperature profile and landscape type was discussed.The correlation between annual temperature profile lines,seasonal temperature profile line and landscape pattern was obtained.The data of the corresponding point of the statistical temperature hatching and the remote sensing image were tested in SPSS,and the regression equation of LST and NDBI,NDVI and NDWI were obtained.By quantitative analysis of landscape type normalization index,heterogeneity index and other landscape pattern index and the relationship between the surface thermal environments,the quantitative coupling between the two laws.Select the typical park green space,analyze the relevant factors of its cooling effect,and get the cooling range,the temperature range and the parks.Thirdly,the temporal and spatial distributions of major pollutants such as NO₂,SO₂,PM₁₀ and PM_2.5 were retrieved by spatial interpolation and remote sensing inversion based on statistical data and remote sensing images.The qualitative and quantitative analysis of the socio-economic factors influencing the urban construction land is carried out,and the driving mechanism of the landscape pattern evolution is obtained by principal component analysis.Finally,according to the above study of the spatial and temporal evolution of urban landscape pattern and its environmental effect proposed to optimize the landscape pattern..更多还原