Gap between MODIS Aqua 2019-01-01

[jvaldezch]

Hi,

I'm working with some MODIS A images in order to extract Chlor_a but there is a gap between not sure if I'm doing something wrong when using l2mapgen or any previous steps. On the other hand with MODIS Terra there is no gap.

Commands used:

l2mapgen ifile=A2019001191000_L2_OC.nc ofile=A2019001191000_chlor_a.tif prod=chlor_a outmode=tiff flaguse=ATMFAIL,LAND,HILT,CLDICE,LOWLW,CHLFAIL,CHLWARN,ATMWARN,FILTER,NAVWARN,NAVFAIL,ABSAER,MAXAERITER stype=1 width=2900
l2mapgen ifile=A2019001190500_L2_OC.nc ofile=A2019001190500_chlor_a.tif prod=chlor_a outmode=tiff flaguse=ATMFAIL,LAND,HILT,CLDICE,LOWLW,CHLFAIL,CHLWARN,ATMWARN,FILTER,NAVWARN,NAVFAIL,ABSAER,MAXAERITER stype=1 width=2900

Example images:

MODIS Aqua l2 product - Chlor_a [date 2019-01-01]


MODIS Terra l2 product - Chlor_a [date 2019-01-01]


Any help would be appreciated. Thanks.

[OB.DAAC - SeanBailey]

Your images are not accessible.  Can you describe what you mean by "there is a gap"?

[jvaldezch]

Thanks Sean, I just updated images above.

[bruce]

Have you looked to see what flags are set in the area of the "missing" data?

[OB.DAAC - SeanBailey]

The gap you show does not appear when I process the L2 file we distribute.  Follow Bruce's suggestion to check the flags in the L2 file (or grab our processed version of the L2).

Sean

[jvaldezch]

Hi again Sean, I just downloaded the files https://oceandata.sci.gsfc.nasa.gov/cgi/getfile/A2019001091000.L2_LAC_OC.nc and https://oceandata.sci.gsfc.nasa.gov/cgi/getfile/A2019001090500.L2_LAC_OC.nc and using the following commands:

l2mapgen ifile=A2019001191000.L2_LAC_OC.nc ofile=A2019001191000_OC_chlor_a.tif prod=chlor_a outmode=tiff flaguse=ATMFAIL,LAND,HILT,CLDICE,LOWLW,CHLFAIL,CHLWARN,ATMWARN,FILTER,NAVWARN,NAVFAIL,ABSAER,MAXAERITER stype=1 width=2900
l2mapgen ifile=A2019001190500.L2_LAC_OC.nc ofile=A2019001190500_OC_chlor_a.tif prod=chlor_a outmode=tiff flaguse=ATMFAIL,LAND,HILT,CLDICE,LOWLW,CHLFAIL,CHLWARN,ATMWARN,FILTER,NAVWARN,NAVFAIL,ABSAER,MAXAERITER stype=1 width=2900

Still getting same results.



Also I noticed that the same issues is presented on NASA WorldView webapp as you can see below:

[OB WebDev - norman]

I notice that the files you say you downloaded are different from the files
in your l2mapgen calls.  The files you downloaded are over the Indian Ocean.

https://oceancolor.gsfc.nasa.gov/cgi/browse.pl?file=A2019001090500.L2_LAC_OC
https://oceancolor.gsfc.nasa.gov/cgi/browse.pl?file=A2019001091000.L2_LAC_OC

Norman

[OB.DAAC - SeanBailey]

OK,  I had thought you were seeing a gap within a 5 min granule...after realizing you were concatenating two scenes together, I was able to replicate the problem.  Yes, there's a gap.  No, there shouldn't be one.   There seems to be an issue with our L1 processing that drops a scan occasionally.   I've got "top men" (well, OK "top women") working on this one :grin:  With luck, we'll resolve the issue soon (looking like a configuration problem, so a tractable one).

In the meantime, you have a couple of options.  You could grab the L0 data, concatenate them into a 10 minute granule and process that to L1A->GEO->L1B->L2->map.
or grab the associated L1A granules from LAADS (their processing does not have the dropped scan).

Sorry for the inconvenience, but thanks for pointing out the problem :grin:d

Sean

[jvaldezch]

Thank you Sean, sorry I didn't mention that I was mosaicking both images and there was the gap.

I will try using recommended images :)

Jaime E. Valdez

[OB.DAAC - SeanBailey]

Just to close this one out - we've resolved the issue with our processing.  We'll be regenerating the affected L1 files soon.
The ones you had issue with have already been regenerated.

Regards,
Sean