| 摘要: |
| 在全球气候变暖和城市化进程加速的
背景下，城市热岛效应日益加剧。本研究耦合
地表温度、土地利用覆被、城市建成区边界和
数字高程模型等数据，基于GEE（Google Earth
Engine）平台对逐8天MODIS LST（Land Surface
Temperature）产品进行时间线性插值并生产全国
无缝LST数据，进一步利用动态简化城市边界算
法，研发2005—2020年逐年1 km空间分辨率中
国夏季地表城市热岛空间数据集。在此基础上，
使用城市热岛空间扩张指数揭示2005—2020年
夏季昼夜中国城市热岛空间扩张特征。研究结果
表明，2005—2020年夏季昼夜中国地表城市热岛
面积分别增长1.95和2.49倍。2020年夏季昼夜中
国地表城市热岛强度分别为1.36 ℃和1.33 ℃，较
2005年增长0.08 ℃和0.38 ℃。2005—2020年夏
季昼夜中国地表城市热岛空间扩张均以边缘型为主，2015—2020年城市热岛空间扩张程度在各时期最高。填充型城市热岛空间扩张城市热岛强度最
高。本研究所研发的时间线性插值算法和动态简化城市边界算法为长时间序列城市热岛效应空间识
别和城市热岛强度表征提供技术范式，中国地表城市热岛空间数据集为主动适应和减缓城市热环境
风险与促进城市可持续发展提供数据支持。 |
| 关键词: 城市热岛效应 地表温度 时间线性插值 动态简化城市边界算法 空间扩张 GEE |
| DOI：10.13791/j.cnki.hsfwest.20240621 |
| 分类号: |
| 基金项目:国家自然科学基金项目（52270187）；天津市自然科
学基金项目（21JCYBJC00390） |
|
| Annual 1-kilometer spatial dataset of summer urban heat island and spatial expansionanalysis in China from 2005 to 2020 |
|
JIA Ruoyu,LIU Luo,XU Xinliang,HAN Dongrui,QIAO Zhi
|
| Abstract: |
| The Earth's climate system is undergoing global climate change characterized by warming,
which is influenced by both natural climate and human activities. In the context of global warming and
accelerated urbanization, extreme climate risks such as heat waves are intensifying, leading to serious
consequences such as deaths. As an important manifestation of the disturbance of the Earth’s climate
system, Urban Heat Island (UHI) is a typical phenomenon of the combined effects of global climate
change and human activities. Therefore, establishing a long time series and high spatiotemporal resolution
dataset of UHI effects is of great scientific significance and practical value for establishing a systematic
high-temperature response framework, such as high-temperature response plans, mitigation and
adaptation guidelines, decision support systems, policy incentive guidelines, etc.
This study coupled data of land surface temperature, land use and cover, urban built-up
boundary, and digital elevation model, and based on the Google Earth Engine (GEE) platform,
performed a temporal linear interpolation on the 8-day MODIS LST (Land Surface Temperature)
product, filled the missing data of MODIS from the temporal dimension, and used the values
obtained by the temporal linear interpolation to fill the missing values, which are the average
temperatures of the adjacent times under the missing values, producing a seamless LST data for the
whole country. Furthermore, using a dynamic simplified urban boundary algorithm, within the urban
built-up boundary, according to the different types of land use and cover, the urban built-up areas and
water bodies such as rivers were excluded, and the cases with large differences in digital elevation
were also excluded, obtaining the rural areas, and then calculating the average rural temperature, and
then using the average rural temperature as the background, calculating the UHI intensity according
to the temperature within the urban built-up boundary, obtaining a spatial dataset of summer land
surface UHI in China with an annual 1 km spatial resolution from 2005 to 2020. And according to
the morphological relationship of UHI changes, the UHI morphological changes were expressed by
the UHI spatial expansion index, and according to the size of the index, they were divided into edge
type, filled type, and enclave type, and finally the spatial expansion characteristics of the summer
day and night UHI in China from 2005 to 2020 were revealed by the UHI spatial expansion index.
The research results show that the summer daytime and nighttime land surface UHI area in China
increased by 1.95 and 2.49 times, respectively, from 2005 to 2020. The summer daytime and nighttime
land surface UHI intensity in China in 2020 were 1.36°C and 1.33°C, respectively, an increase of 0.08°C
and 0.38°C compared to 2005. The UHI intensity is relatively stable in the eastern region, but high and
fluctuates greatly in the western region. In summer 2005, the surface UHI intensity was higher duringdaytime than at night. In summer 2020, the nighttime surface UHI intensity increased significantly, especially in the central and eastern regions, which was higher than
the daytime surface UHI intensity. The spatial expansion of the summer daytime and nighttime land surface UHI in China from 2005 to 2020 was dominated by the
edge type, and the degree of UHI spatial expansion was the highest in 2015–2020. The filled type UHI spatial expansion had the highest UHI intensity. This study used
the land surface temperature temporal linear interpolation algorithm, which improved the temporal accuracy of the original MODIS LST temperature data, ensuring
that the land surface temperature data had no missing data, and used the GUB data of multiple years to identify the UHI effects of the corresponding years, and
dynamically updating the GUB data was the most important guarantee for improving the spatial identification accuracy of the UHI effects. And based on the research,
it proposed to use the spatiotemporal interpolation algorithm and annual GUB data to further improve the accuracy of the spatial dataset of UHI in China.
The temporal linear interpolation algorithm and the dynamic simplified urban boundary algorithm used in this study provide a technical paradigm for the
quantitative identification of UHI effects in long time series, and the spatial dataset of land surface UHI in China provides data support for actively adapting and
mitigating urban thermal environmental risks and promoting urban sustainable development |
| Key words: urban heat island land surface temperature time linear interpolation dynamic simplified urban-extent algorithm spatial expansion google earth engine |
