城市河道及周临用地对地表热环境的影响研究  

Influence of urban rivers and their surrounding land on the surface thermal environment

作　　者：冯晓刚[1] 赵毅 李萌[1] 周在辉[1] 李凤霞[1] 王园 杨永佺 FENG Xiaogang;ZHAO Yi;LI Meng;ZHOU Zaihui;LI Fengxia;WANG Yuan;YANG Yongquan(College of Architecture,Xi’an University of Architecture and Technology,Xi’an 710055,China;CCC Urban and Rural Construction Planning and Design Institute,Wuhan 430050,China;Shandong Province Metallurgical Engineering Co.,Ltd.,Jinan 250101,China)

机构地区：[1]西安建筑科技大学建筑学院,西安710055 [2]中交城乡建设规划设计研究院有限公司,武汉430050 [3]山东省冶金设计院股份有限公司,济南250101

出　　处：《自然资源遥感》2023年第3期264-273,共10页Remote Sensing for Natural Resources

基　　金：“十三五”国家重点研发计划子课题“建筑节能设计基础参数研究-海洋气候条件建筑节能基础气象参数”(编号:2018YFC0704505);国家自然科学基金青年基金项目“黄土高原地区绿地景观格局时空变异特征及其对碳汇响应的机制研究”(编号:51608419);陕西省自然科学基金面上基金项目“基于大数据融合的一带一路地域性城镇热生态环境安全格局规划研究”(编号:2018JM5073);西安建筑科技大学自然科学专项基金项目“面向区域地表热环境改善的城市多元景观要素结构配置模式与优化策略研究”(编号:ZR19001)共同资助。

摘　　要：水体作为城市生态系统的重要组成部分,对改善城市热岛效应和人居热环境具有重要的生态学意义。文章以多时相Landsat和SPOT为实验数据,提出了一种针对混合像元的支持向量机最优端元子集构建原理的地表比辐射率确定方法,采用单窗算法,以灞河及周临用地为研究对象,剖析了河道水体、周临用地与地表温度间的耦合关系。结果表明:①支持向量机的混合像元最优端元子集构建方法提取地表比反射率结果与MODIS LSE产品误差小于0.005(R=0.832),表明该方法精度较高,可用于地表比辐射率的提取;②1992—2019年27 a间灞河两岸用地类型与局地地表温度格局变化显著,建设用地急剧增加,变热趋势明显;不同时期河道周临用地类型对地表温度影响的幅度不同,建设用地、草地、水体及林地为影响灞河两岸热环境的主要用地类型,水体、林地、草地和耕地均表现为不同程度的降温效应,降温幅度依次为:水体>林地>草地>耕地;③同一时期灞河两岸用地类型对局地温度的影响存在空间差异。灞河以东水体、林地、草地和耕地降温效果显著,而灞河以西则为水体、林地及草地。该研究有助于正确认知城市河道对局地热环境的影响特征,能够为城市河道及其周临局地热环境改善提供科学参考。As an integral component of the urban ecosystem,water bodies hold considerable ecological significance for mitigating the urban heat island effect and the thermal environment of human habitat.With multi-temporal Landsat and SPOT data as experimental data,this study proposed a method for determining surface emissivity for mixed pixels based on the principle behind the construction of the support vector machine(SVM)optimal endmember subset.Then,this study employed the surface emissivity determination method to analyze the coupling relationship of the water bodies and surrounding land of the Bahe River with the surface temperature using a mono-window algorithm.The results are as follows:①The SVM optimal endmember subset construction method for mixed pixels yielded an error of surface emissivity less than 0.005(R=0.832)relative to the MODIS LSE product.This result indicates that the method has high accuracy and thus can be used to extract surface emissivity.②Over the past 27 years,the land types and local surface temperature patterns on both sides of the Bahe River have changed significantly,with a sharp increase in construction land and a significant warming trend.The effects of land use types surrounding the Bahe River on surface temperature varied in different periods,with construction land,grassland,water bodies,and forest land being the principal land use types affecting the thermal environment on both sides of the Bahe River.The cooling effects of water bodies,forest land,grassland,and cultivated land are in the order of water bodies>forest land>grassland>cultivated land.③The effects of land use types on both sides of the Bahe River on local temperatures exhibited spatial differences during the same period.To the east of the Bahe River,the water bodies,forest land,grassland,and cultivated land show significant cooling effects.In contrast,to the west of the river,only water bodies,forest land,and grassland showed significant cooling effects.This study contributes to the proper understanding of the influenc

关 键 词：地表比辐射率 热环境 城市热岛效应 水体 灞河 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] TP701[自动化与计算机技术—控制科学与工程]