基于人居尺度的中国城市热岛强度时空变化及其驱动因子解析  

Estimation of the Urban Heat Island Intensity Change and Its Relationships with Driving Factors Across China Based on the Human Settlement Scale

作　　者：孙艳伟[1,2,3] 王润[4] 郭青海 高超[1] SUN Yan-wei;WANG Run;GUO Qing-hai;GAO Chao(Department of Geography and Spatial Information Techniques,Ningbo University,Ningbo 315211,China;Center for Land and Marine Spatial Utilization and Governance Research,Ningbo University,Ningbo 315211,China;Institute of East China Sea,Ningbo University,Ningbo 315211,China;School of Resources and Environmental Science,Hubei University,Wuhan 430062,China;School of Civil Engineering and Architecture,Zhejiang Sci-Tech University,Hangzhou 310018,China)

机构地区：[1]宁波大学地理与空间信息技术系,宁波315211 [2]宁波大学陆海国土空间利用与治理研究中心,宁波315211 [3]宁波大学东海研究院,宁波315211 [4]湖北大学资源环境学院,武汉430062 [5]浙江理工大学建筑工程学院,杭州310018

出　　处：《环境科学》2021年第1期501-512,共12页Environmental Science

基　　金：浙江省哲学社科规划项目(20NDJC077YB)。

摘　　要：全球变暖的大背景下城市地表热环境的日趋恶化是全球现代城市共同面临的生态环境问题之一,对城市地区居民的身心健康、空气质量改善以及植被生长发育等诸多方面造成不利影响.本文利用2001~2018年时间序列MODIS地表温度数据产品,基于一致性方法对我国1232个主要城镇人居斑块(面积>10 km2)的热岛强度及其时空变化模式进行了定量测度,并应用随机森林回归模型(RF)试图揭示人类活动、城市形态、地形、植被以及气象因素对我国城市热岛强度的非线性驱动机制.结果表明,在人居尺度上,我国超过90%的城镇存在显著的昼夜热岛现象;昼夜城市热岛强度分别达到(0.75±0.6)℃和(0.81±0.53)℃,且夏季日间热岛强度显著高于冬季,冬夏季节夜间热岛强度差异不大.2001~2018年间我国白天热岛强度呈现逐年轻微下降趋势,而冬季夜间热岛显著增加,在夏季和年均尺度上则无显著变化.在空间分布上,我国东部沿海省份的白天热岛强度显著高于西北和高原省份,而夜间热岛分布模式正好与之相反.RF回归结果表明,城镇的气候背景(年均降雨量)和地理位置(纬度)是我国昼夜城市热岛的最为重要的两个主控因子.此外,缩小城乡绿色植被覆被差异对白天热岛的减缓将起到一定作用,而优化调控中小型城镇建成区规模与人口密度对缓解我国夜间热岛的持续升高也将起到积极作用.Considering a background of global warming,the effect of urban heat island(UHI)has become a serious ecological and environmental problem that is commonly faced by modern cities worldwide,posing challenges to the physical and mental health of residents in urban areas,air quality,and the healthy growth of surface vegetation.Using time series MODIS land surface temperature data products from 2001 to 2018,we evaluated the urban heat island intensity(UHII)and its spatiotemporal patterns in 1232 major urban human settlement patches(area>10 km2)across China based on the consistency method.A random forest regression model(RF)was also used to reveal the driving mechanism of urban heat island intensity in China due to human activities,urban morphology,vegetation,topography,and meteorological factors.The results show that more than 90%of cities and towns in China have exhibited significant diurnal heat island phenomena on the scale of human settlements.The diurnal UHII reaches up to(0.75±0.6)℃and(0.81±0.53)℃,respectively.The daytime UHII in summer was significantly higher than that in winter,while there was little difference in nighttime UHII between summer and winter.During the past 18 years,the daytime UHII in China has shown a slight downward trend,while the nighttime UHII has increased significantly,with no significant change in summer and on mean annual scales.In terms of spatial distribution,the eastern coastal provinces have higher daytime UHII than those of the western or plateau provinces,while the nighttime UHII shows the opposite pattern.The RF regression results show that the average annual rainfall and latitudinal position of cities and towns are the two most important factors controlling the daytime and nighttime UHII in China.Narrowing the differences in vegetation growth between urban and rural environments will have a mitigating effect on the daytime UHII,and controlling the expansion of small and medium-sized towns will also play a positive role in the continuously rising nighttime UHII.

关 键 词：城市热岛(UHI) 人居尺度 时空变化 随机森林 驱动因子 

分 类 号：X16[环境科学与工程—环境科学]