Processing VIIRS L1A to L2 SST Using Extracted VIIRS L1A file

[monger]

Hi --- When I process VIIRS L1A file to L2 CHL and SST using the full L1A data granule and full GEO file I obtain nice L2 results for BOTH CHL and SST.  However, when I extract the L1A file, and generate a new GEO file from the corresponding extracted L1A file, I get a nice L2 CHL output, but the L2 SST have problems that seem like they could be related to GEO locations problems.  It seems weird that L2 CHL is fine with extraction, but L2 SST is not fine with the same extraction. 

PS-- I am using SeaDAS 7.4 and updated all processing (today) modules before before running all of this!

Below are the relevant calls I used to generated the new GEO file and then the calls I used to generate the L2 CHL and SST..  Any help would be appreciated!!

Bruce

Generating a new GEO file (fname_geo) from an extracted L1A file (l1a_extract_fname)
subprocess.call(['geolocate_viirs',
                      'verbose=True',
                      'ifile='               +  l1a_extract_fname,
                      'geofile_mod='  +  fname_geo,
                      'terrain_path='  +  os.environ['OCDATAROOT']+'/viirsn/ncdem'])

Generating L2 CHL and SST from extracted L1A VIIRS and Corresponding GEO file..
call(['l2gen',
       'ifile='          + l1a_extract_fname,
       'ofile1='       + color_l2_file_fname,
        'l2prod1='   + prod_list_color,            
        'ofile2='      + sst_l2_file_fname,
        'l2prod2='   + prod_list_sst,        
        'geofile='     + fname_geo,
        'oformat='  + 'netcdf4',
        'par='         + fname_ancil_list,
        'proc_sst='  + '1'])

---------------
If it helps... the link below contains the L1A viirs file I used and the L2 output for the case of 1) using the full granule and 2) the case of the extracting Level1 with extraction lat-lon bounds (S,W, N, E) of:  37,-72, 46,-63.
http://www.geo.cornell.edu/ocean/viirs_test_monger.zip
--------------

[OB.DAAC-EDL - SeanBailey]

Bruce,

Does indeed appear to be an issue with L1A extract as input...:eek:  Until we can fix that, if you process the L1A extract to L1B (using calibrate_viirs) and use that as input to l2gen, all seems to be as expected (i.e. it works :grin:)

Sean