基于AOD数据反演地面PM2.5浓度研究进展

Research Progress of Retrieval Ground-level PM_(2.5) Concentration Based on AOD Data

随着中国城市发展,城镇化进程的不断推进,能源消耗持续增加,空气中的污染物含量越来越高,空气污染事件频发,城市空气质量研究成为一个热点议题。PM_(2.5)作为表征空气质量的重要指标之一,越来越受到人们的关注,目前获取PM_(2.5)数据主要有地面监测和卫星遥感监测2种方式。传统的地面监测手段可以得到高精度的局部PM_(2.5)污染数据,但是由于其覆盖范围的局限性,并没有办法反映出整个区域的PM_(2.5)污染情况。遥感卫星监测恰到好处地弥补了这一缺陷,其中应用最为广泛的是使用卫星遥感数据产品大气气溶胶光学厚度AOD来反演地面的PM_(2.5)浓度。文章从AOD数据的多样性及其应用、反演地面PM_(2.5)浓度模型的选择以及反演模型的优化这3个方面对目前国内外利用遥感卫星AOD数据反演地面PM_(2.5)浓度的研究进行了归纳梳理。其中AOD数据分辨率的不同产生了不同精度的反演结果;而线性回归模型和非线性回归模型的反演精度也存在较为明显的差异;通过在模型中加入气象参数、气溶胶垂直分布特性以及地表信息等因素会显著地改善反演结果。上述研究为流行病学中PM_(2.5)人口暴露研究及健康影响提供方法论基础。

The pollutant is on the increase because of the rapid urbanization and the energy consumption in China. Urban air quality researches become the hot spot in the air quality field with more and more common pollutant days. As the PM2.5 concentration is one of the important indicators of the air quality, people pay more attention to the PM2.5. PM2.5 data could be obtained by two methods including ground observation and satellite remote sensing observation. The data acquired by traditional ground observation are accurate but its coverage areas aren＇t satisfying, while the satellite remote sensing can make up for this flaw. And using satellite data product AOD to retrieve PM2.5 is the most widely applied. This article based on many studies at home and abroad started with all kinds of AOD data, including the model choices for PM2.5 concentration and the model optimization. AOD data of different resolutions will result in distinct retrieval precisions. Besides, there are also obvious differences between linear regression and non-linear regression. The retrieval can be significantly improved by adding the meteorological factors, aerosol vertical profiles, land use information and other factors. Furthermore, the paper＇s focus and orientation in the future were prospected. And this paper provides a methodological basis for epidemiologic study.