Paper
High Resolution Aerosol Optical Depth Retrieval and Its Spatial Distribution Characteristic over Fuzhou
- Authors:
- Weihua Pan; Hui Chen; Chungui Zhang; Jiajin Chen
- Abstract
- Aerosol retrieval was an important way to estimate the aerosol and microphysical properties of the atmosphere and the aerosol optical depth (AOD) retrieval algorithms had been developed for some decades. The Dense Dark Vegetation algorithm (DDV) advocated by Kaufman was mainly one of methods to retrieve the AOD by using MODIS data, which must be possessed with two basic rules that correlation of the AOD and planetary albedo of red band and blue band were sensitive. On the opposite, the AOD of mid-infrared band was no significant change with the planetary albedo of mid-infrared band. Because the unreasonable selection of aerosol model in DDV method may cause significant error in retrieving of urban scale AOD of high spatial resolution, the error may even be higher than that caused by the surface reflectance uncertainty. To solve the problem, the sensitivity test experiments were carried out by selecting different atmospheric and aerosol models in the study and the surface reflectance was made for the planetary albedo of three channels of MODIS (red channel, blue channel and mid-infrared channel) by 6S (second simulation of satellite signal in the solar spectrum) atmospheric radiation models. As a result, the mid-latitude atmospheric model and urban aerosol model were accord with the two basic rules of DDV algorithms over the Fuzhou city. Furthermore, the AOD over Fuzhou city was retrieved based on MODIS data of November 11th, 2006, and the aerosol properties were described by using the MODIS 500 m resolution images. According the Look Up Tables (LUT) constructed among the planetary albedo, surface reflectance and the AOD, the values of AOD could be computed in different regions of Fuzhou. Moreover, the AOD images of red, blue and mid-infrared bands were derived by DDV and the 550 nm AOD image was made to analyze the spatial distribution characteristic. The resulting 500 m AOD images were highly correlated (r = 0.9) with the PM10 concentration of the air quality monitoring stations of Fuzhou. The results showed that the AOD over Fuzhou had notable spatial distribution features and the AOD was significantly affected by the terrain. The high value areas were mainly located in the plain areas, while the low value areas were mainly in the mountains. Furthermore, there were two high values areas in outline of AOD, one lied in metropolitan area, where was the commercial trade centre in Fuzhou, and the other laid in Mawei development area where many factories and shipbuilding mills were in existence. However, the minimum of AOD mainly laid in high mountains such as Gushan and Qishan mountains, where owned dense vegetation and forest. In conclusion, the AOD layouts retrieved from MODIS data were agreed with the distribution of urban surface types in Fuzhou, and the export and transfer of aerosol grains was slow because of landform. The study demonstrated the applicability of AOD retrieval at fine resolution scales (500 m) in urban areas, and the fine AOD images could be used to analyze the spatial distributions of pollutant sources and urban air quality in Fuzhou city.
- Keywords
- Aerosol Retrieval; Distribution; DDV Method ; Aerosol Optical Depth ; Fuzhou; Satellite Image
- StartPage
- 4
- EndPage
- 9
- Doi