【作者】 孙振如；

【导师】 孔繁花；

【作者基本信息】 南京大学 ， 地图学与地理信息系统， 2012， 硕士

【摘要】 快速城市化使城市下垫面原有的自然环境被非渗透性地面代替,由此引起的城市热环境问题给经济、生态及城市居民的生产生活带来严重的负面影响。城市绿地作为城市结构中的自然生产力主体,在缓解城市热岛,调节城市气候和协助城市应对未来气候变化中扮演着极其重要的角色。本文借助RS和GIS软件平台,基于南京三个时期(1988,2000,2009)的TM/ETM+反演地表温度,解译土地利用类型,并通过2009年的iknos高分辨率遥感影像数据获取城市绿地信息；在此基础上结合景观指标、移动窗口和梯度分析等方法,分析绿量、绿地格局与冷岛效应之间的关系,进而从城市绿地降温效应的角度,提出了城市绿地规划建设的建议。本论文共分为6章。第一章绪论,阐明论文选题的背景与意义,研究思路与基本框架,并介绍了研究区的概况。第二章为国内外相关研究综述,从相关研究的数据应用,技术方法等不同角度进行了系统的总结和评述。第三章为数据与研究方法,对论文应用的数据进行了详细介绍,并重点对论文分析所采用的关键技术,例如地表温度反演的单窗算法,景观格局分析的移动窗口方法等进行了介绍。第四章和第五章为论文的主要研究内容,其中第四章重点对研究区绿地格局及冷岛格局的时空变化特征及联系进行了分析,并明确了冷岛定义及绿地降温效应形成的冷岛强度的量化方法；第五章对绿地降温效应及绿地降温效应形成的冷岛强度的影响要素进行了挖掘与分析,并在此基础上构建预测绿地降温效应形成的冷岛强度的模型。第六章为结论,系统总结了论文的主要研究结论和创新点,并提出论文的不足及将来研究与完善的方向。论文研究结果表明：(1)近二十年来南京市土地利用发生了巨大的变化,主要表现在城镇建设用地的增加和渗透性地面的减少,绿地的景观格局破碎程度、形状复杂程度和集聚程度均先减后增。这与城市化进程逐渐加快,并不断蚕食自然植被有很大关系。基于移动窗口和梯度变化的景观指数分析结果显示,空间上绿地景观形状指数(Landscape Shape Index, LSI)值从中心向外围逐渐降低,且绿地斑块形状的复杂程度较高的区域与城市建成区呈现较高的一致性；在时间轴上,随着时间的推近绿地斑块集聚程度(Aggregation Index, AI)在中心区域稍有增加,而在远离中心区域则明显降低,表明随着距离城市中心越远的区域将逐渐被城镇化,城市外围的绿地分布也越来越离散化,绿地形状复杂程度(LSI)也随城镇的扩张而增强。(2)冷岛格局及梯度分析表明,在时间轴上,高温区域随着城市范围的扩张而延伸,低温区域随着自然植被面积的减少而缩减,但冷岛的斑块数目在快速增加,这说明冷岛也随着绿地斑块的破碎化而逐渐破碎,形状也越来越复杂。(3)五种土地利用类型上的温度均值统计结果表明,除水体以外,绿地的平均温度最低,其次是农田。另外,基于流域分析模型的南京市温度流域的空间分布结果表明,随着自然植被和城镇区域面积的剧变,冷流的格局也发生了较大的变化。(4)以清凉山公园为例,分析了绿地斑块的降温效应引起的冷岛格局特征。结果表明,同一绿地斑块在各个不同方向上的降温范围不同,绿地斑块向外的降温梯度作用幅度受到周边地形的影响,总体而言清凉山绿地斑块边界的距离每增加10m,温度升高约0.37℃,降温的最大距离不超过距离边界130m的区域。以选择8条较为典型的冠层宽度绿化分析了城市中道路绿化带的降温效应。分析结果表明绿化带乔木冠层宽度为20m时,和周边地区相比降温幅度可达0.4℃,随着绿化带乔木冠层宽度每增加10m,降温幅度增加0.33℃。(5)景观尺度的统计分析表明绿量和冷岛强度呈现显著的正相关,类型水平上绿地的LSI、AI与冷岛强度分别呈现显著的负相关和显著的正相关关系。回归分析得到了以斑块面积、斑块形状(SHAPE).绿量与归一化冷岛强度(NDCⅡ)的关系模型,线性函数可以较好的模拟SHAPE与NDCⅡ的关系,而斑块面积和绿量与NDCⅡ的关系使用对数模型更为合理。论文最终使用以上四个因素回归得到的NDCⅡ和斑块内外温差的模型均达到了一定的精度。更多还原

【Abstract】 Rapid urbanization resulted in the the change between natural environmentand the impervious ground, and which has caused the servious thermal environment problems. The increase of the thermal environment can bring negative effect on the economy, ecology, as well as the living of city residents. The urban green spaces play an important role in the mitigation of urban heat, and accordingly can help the city adapt to the urban climate change. With the help of RS and GIS, three periods (1988,2000,2009) of TM/ETM+were used to retrieve land surface temperature, and interpret land use patterns, in addition, urban green spaces information were obtained through high resolution data of IKNOS in2009; then landscape metrics, moving window and gradient analysis are employed to find the relationship between green biomass, green space pattern and cold island effect, accordingly a model were developed to predict the influence of differenct landscape characteristic of green spaces. At last, based on the research of this thesis, suggestions on the urban green space planning and development for their better cooling effect were given.This thesis was constructed by six chapters. The first chapter introduced the, s paper’s background, significance, ideas, the basic framework of research as well as the description of the study area. The second chapter reviewed the related research. The third chapter gave a general expalnaton on the data and research method used in this thesis, such as single window algorithm of retrieving land surface temperature and moving window of landscape patter analysis. The fourth chapter and the fifth chapter are the main parts of this thesis. The fourth chapter analyzed the landscape characteristics the green spaces of study area and the cold island pattern from spatio-temporal scale. The fifth chapter discovered and analyzed the factors which may affect the cold island intensity caused by the cooling effect of green space, and on the basis of which, a prediction model to calculate the cold island intensity was developed. The sixth chapter is the conclusions. It systematically summarized the main conclusions and innovations of this research, and put forward the disadvantages and the direction future research would study and perfect. The results showed as follows:(1) During the past two decades, the land-use in Nanjing city has undergone tremendous changes, mainly showing as the in the increase of urban areas and decrease of urban green spaces and farm land. However, the landscape spatial pattern analysis shows that the fragmentation, landscape shape index and aggregation of the urban green space had a trend of decreasing at the first10years and then increased in the recent10years. The moving window and the gradient analysis showed that landscape shape index (Landscape Shape Index, LSI) on green class level decreased gradually from the center outwards, which coincident with urban built-up exploration; while in the temporal scale, the aggregation of green (AI) increased slightly in the central region, while decreased away from the central region, indicating that with the outlying region being urbanized, green space distribution became more and more discrete, and the green shape complexity (LSI) was becoming more complexed.(2) The cold island landscape and gradient analysis showed that the high temperature region expanded with extension of the built-up, the low temperature region was reduced with the reduction in the area of natural vegetation, but the cold island patch number was increasing rapidly, indicating that as the fragmentation of green cool island raising gradually, the shape becomes more and more complex.(3) Two different ways were tried to describe the intensity of cold island, that were NDCII (Normalized Difference Cool Island Intensity) and the temperature difference between inside and outside of green patch. Statistical results of mean temperature of five land-use classes showed that except for water areas, green space are the lowest, and next is the farmland. In addition, the watershed analysis model based on digital elevation model was introduced to analyze the distribution of the temperature watershed in Nanjing, and the results showed that with the upheaval of the natural vegetation and the urban area, the pattern of low temperature lines changed significantly.(4) Take Qingliangshan Park as example, the characteristics of cold island landscape caused by cooling effect of the green space had been analyzed. The results showed that its cooling ranges were different in different directions for surrounding terrain. The cooling gradient magnitude of Qingliangshan Park was affected by the surrounding terrain. Over all, distance from the park border increased for each additional10m temperature raises about0.368℃, and the cooling distance was no more than130m. To study the cooling effect of the road green belt, we analyzed the cooling effect of eight samples with typical canopy road green belt. The results showed that the green belt canopy with the width of20m, temperature was0.4℃lower compared to the surrounding area. With the Green Belt tree canopy width increasing for each additional10m, the cooling rate increased by0.33℃.(5) On the scale of landscape, the statistical analysis showed that green biomass presented a significant positive correlation with the cooling effect, in addition, LSI and AI showed a significant negative and a positive correlation respectively. Finally, the model between green patch NDCII and the temperature difference inside and outside green patch was developed, and the accuracy of this model was proved.更多还原