Both Aqua and Terra provide AOD value combining Deep Blue and Dark Target algorithm with corresponding Quality Flag value. They also have Aerosol cloud fraction data for corresponding grids. I want to use these AOD value for estimation of ambient PM2.5. I discarded the cells with QF less than 3 (as suggested by research). My query is whether further filtration of AOD values are required based on Cloud fraction data. That is, whether I have to discard those AOD value for which corresponding Cloud Fraction Value is approx.1.