The retrieval of aerosol over land surface from GF-1 16m camera with Deep Blue algorithm

As Chinese new satellite of high resolution for earth observation, the application of air pollution monitoring for GF-1 data is a key problem which need to be solved. In the paper, based on Hsu et al (2004) Deep Blue algorithm, by taking count of the characteristic of GF-1 16m camera, the contribution of land surface was removed by MODIS surface reflectance product, and aerosol optical depth (AOD) was retrieved from apparent reflectance in blue band. So, Deep Blue algorithm was applied to GF-1 16m camera successfully. Then, we collected GF-1 16m camera data over Beijing area between August and November, 2014, and the experiment of AOD was processed. It is obvious that the retrieved image showed the distribution of AOD well. At last, the AOD was validated by ground-based AOD data of AERONET/PHOTONS Beijing site. It is showed that there are good agreement between GF-1 AOD and AERONET/PHOTONS AOD, (R>0.7). But GF-1 AOD is obviously larger than ground-based AOD which may be brought by the difference of filter response function between MODIS and GF-1 camera. (paper)