Earthdata Forum

Hi,

I am wondering if anyone has used l2gen with Wang's NIR/SWIR switching method with OLCI data? I've tried:
         'aer_opt= -9',
   'aer_swir_short = Oa20_radiance.nc',
          'aer_swir_long = Oa21_radiance.nc',
    'aer_wave_short = Oa16_radiance.nc',
     'aer_wave_long = Oa17_radiance.nc'
But get an error saying "-E- clo_parseOption: Invalid integer value=Oa20_radiance.nc for option key=aer_swir_short". Should I be using other bands or actual integers?

Thank you,
Anna

Use integers:
  aer_wave_short (int) (default=765) = shortest sensor wavelength for aerosol
   aer_wave_long (int) (default=865) = longest sensor wavelength for aerosol
  aer_swir_short (int) (default=-1) = shortest sensor wavelength for
   aer_swir_long (int) (default=-1) = longest sensor wavelength for SWIR-based

Thanks Sean.

Would this be correct? Do I need all four NIR/SWIR wavelengths for aer_corr=-9 or just two wavelengths?

aer_wave_short=779
aer_wave_long=865
aer_swir_short=940
aer_swir_long=1012 (realized it is not OLCI band 1020 after reading this post: https://oceancolor.gsfc.nasa.gov/forum/oceancolor/topic_show.pl?pid=26855;hl=msl12%20olci

The aer_wave_[short,long] are defaulted to 779 and 865 (or 866 depending on S3B vse S3A), so those don't need to be explicitly set by you.  Won't hurt if you do, though.
940 is NOT a band you want to use in the atmospheric correction as it is heavily contaminated by atmospheric water vapor.  The point of the SWIR bands in the aerosol selection is to avoid the problem of NIR water leaving reflectance in highly turbid waters where the assumption of a "black pixel" doesn't hold for the NIR bands (but may for SWIR due to the increased water absorption in that region of the spectrum).  Technically speaking,  OLCI doesn't have SWIR bands (700-1100nm is NIR), so probably not a path you want to explore.  You can try, or perhaps you could just use
aer_wave_short=865
aer_wave_long=1012
and stick with the standard aer_opt=-3

Sean