Hi Quinten,
Wondering if there's a way to use different AOT values for each scene in a batch job.
Thanks.
Specifying AOT values for each scene when batch processing
Re: Specifying AOT values for each scene when batch processing
To clarify, in the available version of ACOLITE no explicit AOT value is used. The aerosol reflectance is estimated in the SWIR bands and then extrapolated exponentially to the VNIR bands according to the aerosol "epsilon" (SWIR1/SWIR2 ratio).
Is there any reason you want to force an AOT for your scenes? Maybe I can add a used specified model + AOT as an option. (E.g. for the Continental, Marine and Urban aerosol models in 6SV.)
Due to the way the ACOLITE GUI and distribution is currently organised, there is no way to provide any per scene specific settings to ACOLITE. If you were to handle the batch processing outside of ACOLITE (e.g. by shell or bat scripting) you could generate a settings file for each scene and run ACOLITE separately for each scene + settings file combination. This would however require an IDL license or a lot of clicking.
Quinten
Is there any reason you want to force an AOT for your scenes? Maybe I can add a used specified model + AOT as an option. (E.g. for the Continental, Marine and Urban aerosol models in 6SV.)
Due to the way the ACOLITE GUI and distribution is currently organised, there is no way to provide any per scene specific settings to ACOLITE. If you were to handle the batch processing outside of ACOLITE (e.g. by shell or bat scripting) you could generate a settings file for each scene and run ACOLITE separately for each scene + settings file combination. This would however require an IDL license or a lot of clicking.
Quinten
Re: Specifying AOT values for each scene when batch processing
Thanks Quinten!
I'm comparing atmospherically corrected images from different algorithms (ACOLITE inclusive) and validating them with in situ data from AERONET-OC station. So, for consistency among methods, I thought it might be a good idea to use constant AOT values (from AERONET-OC station/SeaDAS output) per scene across all atmospheric correction algorithms.
I'm comparing atmospherically corrected images from different algorithms (ACOLITE inclusive) and validating them with in situ data from AERONET-OC station. So, for consistency among methods, I thought it might be a good idea to use constant AOT values (from AERONET-OC station/SeaDAS output) per scene across all atmospheric correction algorithms.
Re: Specifying AOT values for each scene when batch processing
Ok, thanks for the information!
I'm not sure if imposing the AOT will give a proper representation of each method. Separating the aerosol and water signal is the aim of the atmospheric correction, and different methods will estimate the aerosol contribution differently. If you already provide the aerosol optical thickness, this will probably reduce the variability between methods and will maybe even avoid the strengths of the method altogether. The variability will then depend on the spectral shape/type of aerosol used, and on different processing steps that are or are not performed (e.g. glint or BRDF correction).
Quinten
I'm not sure if imposing the AOT will give a proper representation of each method. Separating the aerosol and water signal is the aim of the atmospheric correction, and different methods will estimate the aerosol contribution differently. If you already provide the aerosol optical thickness, this will probably reduce the variability between methods and will maybe even avoid the strengths of the method altogether. The variability will then depend on the spectral shape/type of aerosol used, and on different processing steps that are or are not performed (e.g. glint or BRDF correction).
Quinten
Re: Specifying AOT values for each scene when batch processing
Thanks for the insights Quinten!