基于MODIS数据的PM2.5反演在大气污染监测中的应用  

Application of PM2.5 retrieval in air pollution monitoring with MODIS data

作　　者：张智[1] 白穆[1] 游浩妍[1] 

机构地区：[1]国家测绘地理信息局陕西基础地理信息中心,西安710054

出　　处：《测绘科学》2016年第9期42-46,共5页Science of Surveying and Mapping

基　　金：国家测绘地理信息局科技发展计划项目(2011-03);地理空间信息工程国家测绘局重点实验室经费资助项目(201212)

摘　　要：针对基于传统10km气溶胶光学厚度产品计算PM2.5无法获得更为精细的PM2.5空间分布特征的问题,提出采用6S模型计算获得不同气溶胶光学厚度条件下辐射传输模型方程参数查找表,该查找表可以采用原始光谱波段信息进行数据反演,并结合暗像元法计算得到的地表反射率信息,在辐射传输模型方程计算结果的基础上进行线性插值获得1km气溶胶光学厚度,再利用地面实测PM2.5浓度数据,通过回归分析的方法,获得1km的PM2.5监测数据。以西安市为例,1km分辨率的PM2.5更进一步表现了西安市域内PM2.5的空间分布特征,具有明显的地形分异特征,而且呈现出从城区向郊区逐渐衰减的趋势。研究结果为小区域的PM2.5来源和变化分析提供了一种有效的技术途径。As traditional PM2.5 monitoring result from Aerosol Optical Depth（AOD）data of MODIS with 10 km spatial resolution is unable to obtain finer spatial distribution characteristic of PM2.5, the second simulation of the satellite signal in the solar spectrum model（6S）was proposed to obtain lookup table for the parameters of radiation transfer model equation under different aerosol optical depths, which help to retrieve data from information of original spectral bands in the MODIS product. The Dense Dark Vegetation method（DDV）was used to calculate the surface albedo. The results from the radiation transfer model equation were calculated for aerosol optical depth with 1 km spatial resolution by linear interpolation. Then the ground measured PM2.5 data was combined by regression analysis method to obtain PM2.5 monitoring data with 1 km spatial resolution. Taking Xi＇an city as an example, the PM2.5 data of 1 km spatial resolution further showed spatial distribution features of PM2.5 by detail. Result showed that PM2.5 in Xi＇an could be demonstrated by the data of 1 km resolution with more spatial differential features with topographic variation and a gradual attenuation trend from urban to suburb, which would provide a reference for source and change analysis of PM2.5 in small regions.

关 键 词：PM2.5 监测 高分辨率 气溶胶光学厚度 MODIS DDV 

分 类 号：P208[天文地球—地图制图学与地理信息工程]