晴空下环渤海地区城市化增温效应的热红外遥感研究

It is an important issue to accurately quantify the influence of urbanization on climate change; however, the trends and magnitude of air temperature rise resulted by urbanization in regional and global scales is still controversial. IPCC (2001) pointed out the urgent need for long-term remote sensing-based land surface temperature (LST) data in global warming studies to improve the limits of surface air temperature observations. Currently, clear sky LST derived from thermal infrared images from different satellite platforms has been widely used in many fields, such as urban heat island, water resource, land-atmospheric interaction, however few study focus on urban induced warming due to the land’s high surface heterogeneities. Some studies already found that LST is reasonable for climate analysis after accurate evaluation, validation and correction to eliminate the effect of orbit drift, surface emissivity influence and cloud contamination and so on. In this study, we try to integrate the satellite images and the ground measurments to examine the effect of urbanization on climate. First, satellite-based thermal infrared images from different satellite platform was validated by field measurements or meteorological station measurements to check errors and inhomogeneous factors in sample scale (1km,3km,9km around stations) to build LST time series, then LST trends from 1980s-2010 among those samples area could be derived. According to classification of land cover type for sample area and regional area in different urban expansion period, we can quantify the difference of LST trends between urban, suburban and rural area.

如何准确量化城市化效应的气候影响是全球变化研究中的重要问题，目前在区域与全球尺度上其导致的增温贡献幅度尚存争议。2001年IPCC就曾提出使用卫星地表温度来辅助近地面气温观测。目前遥感方法已广泛应用于晴空下地表温度的分析与提取，但在如何用于区域尺度上气候增暖研究方面有待于进一步探索。研究发现对遥感地表温度信息进行合理的评估与校正，可以满足长时间序列的气候分析需求。本研究以京津唐典型城市为研究区，使用星地集成的方法，以多源热红外通道影像为主要数据源，辅助地表参数与气象指标观测，进行典型样区的卫星地表温度订正和地表温度时间序列构建，探讨典型样区尺度的城市化影响下的增温趋势差异，揭示城市、乡村区域间的增温差异的幅度与影响因素，从遥感对地观测角度进一步认识城市化的区域增暖趋势，为常规器测资料提供晴空状态下的分析参考，为区城市化气候效应研究提供新的思路和科学方法。

如何准确量化城市化气候效应的气候影响是全球变化研究中的重要问题，尤其是在区域尺度上关于其增温幅度等还存在许多争议。2001年IPCC就曾提出使用卫星地表温度来辅助近地面气温观测。遥感方法已经广泛用于了地表温度的提取与分析，而且在许多领域已经得到了很好应用，但是其在气候分析的应用研究一直存在欠缺。本研究选取了多种遥感数据集，结合地表订正和构建，构建了一套基于AVHRR的长时间序列的地表温度数据集。选取了京津唐核心区23个气象观测台站作为对比分析对象，将台站周边下垫面进行分类识别，得到了3组不同类型站点。然后将站点尺度得到的两套数据集对比分析，地表温度和气温有一致的增温趋势，且城市化所导致的地表温度增加程度约为0.415±0.793 ℃，分析发现城市化导致的气温增温速率约为0.14℃/10年，而地表温度的增温速率是气温的约10倍。