Page 1 of 1
Posted: Tue Oct 27, 2020 11:44 pm America/New_York
I am running l2gen to get L1 to L2 for OLCI images. I have set maskcloud=0 in my parameter file but where CLDICE mask is present I get all Rhos as NaN but just Rhos_940 has values! At certain locations Rhos_490 has negative values! Just when I think I have things somewhat figured out there come images that confuse me!
In this particular image there is a vertical band (blue color is where CLDICE flag is on) seen on the right side (east) of the image - that is where all the values are NaN but Rhos_940.
Posted: Wed Oct 28, 2020 11:22 am America/New_York
Pixels can have NaN values for products for a multitude of reasons and NaN pixels can have a multitude of flags. You can view all the flags for a pixel by hovering over it and observing all its flag values within the Pixel info tool window (... or by placing a pin on it and using this tool). I believe the vertical flagged area to the right of your image is likely associated with high sensor zenith angle (HISATZEN flag).
Negative rhos values will likely be flagged LOWLW. Cloud shadow can be one cause for negative rhos values, but I will leave clarification regarding the reasons for negative rhos values to others who may wish to add in on this.
Posted: Thu Oct 29, 2020 12:41 pm America/New_York
The values of rhos should never be negative, unless the TOA radiance is less than the calculated Rayleigh radiance - which isn't likely. NaN can happen in the SeaDAS display of the data, if the data are outside the expected range defined in a valid data expression for the product. Attach (do not paste inline) the output of
ncdump -h for one of the OLCI L2 files for which you're seeing this issue.