OB.DAAC L2_OC OLCI-Sentinel3A/B L1 source files

[slsmit34]

Hello,
I have a question about the L1 source files for OB.DAAC's OLCI S3A/S3B Ocean Color products.

I am working with a custom script for l2gen. The OCSSW parts of this script currently works with MODIS-A/T by retrieving names of MODIS OB.DAAC L2 OC scenes (query with earthData), and then uses the timestamp on the granule identifier to download the corresponding L1 data from OB.DAAC. This works well because OB.DAAC's MODIS L1 files can be found at a link using the granule identifier + the '.L1A_LAC.bz2' extension.

I am adapting this to work with OLCI S3A/S3B, but I cannot find a collection in OB.DAAC that contains the L1 source files that were processed to create the L2 scenes. Do you have any advice on how to find them? Through googling, I was directed to LAADS (https://ladsweb.modaps.eosdis.nasa.gov/archive), but this L1 OLCI collection does not seem to have scenes that match the granule timestamp identifiers of the OB.DAAC L2 OC OLCI scenes.

Thank you, and apologies if this question sounds confusing.
Best,
Sami

[OB ODPS - towens]

You can find the source L1B files here:
https://oceandata.sci.gsfc.nasa.gov/directdataaccess/Level-1B/S3A-OLCI/
https://oceandata.sci.gsfc.nasa.gov/directdataaccess/Level-1B/S3B-OLCI/

Tommy

[slsmit34]

Hi Tommy,
Thanks for the response. Is there a way to know which L1B source file was used to create the L2 scenes? For example, when looking through the directory for the corresponding L1B scene to "S3A_OLCI_EFRNT.20210405T144930.L2.OC.nc", there were no L1B scenes with a matching date and time in the name. The closest I could find was "S3A_OL_1_EFR____20210405T144912_20210405T145212_20210406T193322_0179_070_196_2340_LN1_O_NT_002.zip" (see pictures below)

The L2 scene of interest

l2_id.PNG (6.93 KiB) Not viewed yet

There is no L1B scene with a matching date and time in the name. These are the closest times to the L2 scene.

closest_l1b.PNG (27 KiB) Not viewed yet

Best,
Sami

[OB ODPS - towens]

The L2 file names are derived from the start time of the data in the L1B granule, which is later than the start time in L1B file name

Can you add logic in your script to search forward by incrementing the L1B time by 1 sec in the expected L2 file name and check until it finds a match or reaches a limit (+30s ?) before it quits?

S3A_OL_1_EFR____20210405T144612_20210405T144912_20210406T193258_0179_070_196_2160_LN1_O_NT_002.zip
produced:
S3A_OLCI_EFRNT.20210405T144630.L2.OC.nc

S3A_OL_1_EFR____20210405T144912_20210405T145212_20210406T193322_0179_070_196_2340_LN1_O_NT_002.zip
produced:
S3A_OLCI_EFRNT.20210405T144930.L2.OC.nc

S3A_OL_1_EFR____20240405T142815_20240405T143115_20240406T145305_0179_111_039_1980_PS1_O_NT_003.zip
produced:
S3A_OLCI_EFRNT.20240405T142828.L2.OC.nc

Tommy

[slsmit34]

Hi Tommy,
Yes, that is a good solution. Thank you for the support!
Best,
Sami

[erickduck]

Sure, I can help you add that logic to your script. Below is a Python script that takes the L1B filename, extracts the start time, and then searches for the corresponding L2 filename by incrementing the start time by 1 second until it either finds a match or reaches a limit of +30 seconds.

S3A_OL_1_EFR____20210405T144612_20210405T144912_20210406T193258_0179_070_196_2160_LN1_O_NT_002.zip

[slsmit34]

Hi Eric,
Thanks so much for your help. I do not see the python script that you mentioned. Did you include it as an attachment to your post?
Best,
Sami

[OB ODPS - towens]

The forum software may have filtered the code attachment to prevent distribution of malicious code.
Do not run the code if you cannot verify its actions.

Tommy