基于高分辨率遥感的城市景观格局演变及其热岛效应研究

Urban areas are strikingly heterogeneous, representing a mix of natural and built components at different densities and arrangements in the landscape. Over the past decade, research in urban systems has increasingly focused on understanding the link between spatial heterogeneity and ecological processes in urban ecosystems. To develop such an ecological understanding of urban systems, it is necessary to first quantify the fine-scale heterogeneity of their built and natural components. This study aims to develop a new approach to characterizing and quantifying the fine-scale heterogeneity in urban landscapes that capitalizes on the strengths of two approaches - visual interpretation and object-based image analysis, with high-resolution remotely sensed imagery. This new approach integrates the ability of humans to delineate ecologically meaningful patches with an object based image analysis that accurately and efficiently describes the fine-scale land cover features within those patches. The overall objective of this study is to develop an ecologically meaningful and efficient approach to quantifying the fine-scale heterogeneity in urban landscapes. The method will involve two steps. First, patches will be generated through visual interpretation. These patches will serve as pre-defined boundaries for finer-scale segmentation and classification of within-patch land cover features, using an object-based classification. Patches will then be classified based on the within-patch proportion of land cover features. By developing a two-level hierarchical classification system, we will define and describe urban landscape as a two-level hierarchical patch mosaic. We will apply this approach to Beijing metropolitan areas for three years, 2002, 2008, and 2014, to capture the changes in landscape patterns before and after the 2008 Olympic Game. In addition, by applying this approach to characterize and quantify the spatial pattern of land cover, we will investigate the effects of land cover pattern, especially spatial configuration, on urban surface heat islands in the Beijing metropolitan areas from 2002-2014.

明确城市景观格局演变及其热岛效应，可为减缓城市热岛效应，提高城市人居环境质量提供科学保障。本项目选取城市扩张迅速、城市热岛问题突出的北京作为研究对象，针对城市生态系统精细尺度上空间异质性和社会经济自然复合特性，采用基于对象的图像分析技术和目视解译景观制图，构建基于高分辨率遥感数据的城市景观格局表征和精细量化方法，将城市生态系统描述成两个等级的斑块镶嵌体 - 土地覆盖斑块镶嵌体和社会经济自然复合斑块镶嵌体，研究北京2002－2014年城市景观格局演变特征。应用空间统计方法，分析北京城市热岛时空分布特征，研究精细尺度上城市景观的组成和空间配置（如斑块大小、形状、空间邻接关系）对城市热岛的影响；探讨对于给定的景观类型组成（如绿地覆盖率），如何通过优化其空间配置（如绿地斑块大小、形状），进一步减缓城市热岛效应，为城市生态规划与景观设计，提供科学依据。

本项目围绕“城市景观格局精细尺度空间异质性和社会经济自然复合特征定量表征”和“精细尺度上城市景观类型的空间配置对城市热岛强度和空间分布的影响”两个关键科学问题，选取城市扩张迅速、城市热岛问题突出的北京作为研究对象，重点开展了三个方面的研究：（1）基于高分辨率遥感的城市生态系统景观格局定量化方法；（2）城市生态系统景观格局演变；（3）城市景观格局对城市热岛效应的影响，并取得了以下主要进展。.1）构建了基于高空间分辨率遥感影像的城市景观格局定量化方法，用于定量表征城市景观格局空间异质性和社会自然复合特性。提出了耦合自然属性和社会经济属性的城市景观格局表征与量化的分类体系，将城市景观表征为一个两个等级的斑块镶嵌体，分别刻画其自然属性与社会经济属性；在此基础上，发展了城市景观精细表征和量化的技术方法，提出了基于景观类型和景观要素的等级分类概念框架，揭示了支持向量机（SVM）具有较好的分类精度和效率，并明确了SVM的最优参数设置。.2）定量分析了研究区主要建成区精细景观的特征与演变，揭示了北京城市建成区绿地景观高度破碎、以及高度动态的特征。研究发现：2005-2009年，北京市五环内新建绿地面积近60 km2（面积占比10.36%），但原有绿地减少近30 km2（4.91%）；新增绿地斑块以小斑块为主，平均面积仅为676.31m2。与中等分辨率的TM数据的对比研究发现，TM数据会严重低估城市绿地的覆盖率，并且无法准确揭示其高度的动态度。.3）定量研究了精细尺度景观格局对城市热环境的影响，揭示了城市景观要素（如绿地）的空间配置对城市热环境具有显著的影响，但影响程度在不同城市存在差异；进一步探讨了局地尺度建设用地比例与城市规模对城市热岛影响的相对重要性，发现当城市规模较小时，局地尺度人工表面比例对气温的影响更大，而当城市规模较大时其对气温的影响更加重要。