Inconsistencies between CMR and OBPG subscription

[alaroy]

We are experimenting with using CMR to query for granules in a region. The results I'm getting from CMR and an identical OBPG subscription are inconsistent with each other.

Specifically here's a query we did to the OBPG subscription:

Code: Select all

$ curl -s -d "sdate=2025-05-18&subID=2130&results_as_file=1" https://oceandata.sci.gsfc.nasa.gov/api/file_search | grep 2025141
A2025141165500.L1A_LAC.bz2
A2025141182500.L1A_LAC.bz2
A2025141183000.L1A_LAC.bz2
A2025141200500.L1A_LAC.bz2
A2025141201000.L1A_LAC.bz2
A2025141214000.L1A_LAC.bz2

And here is its CMR equivalent:

Code: Select all

$ curl -s "https://cmr.earthdata.nasa.gov/search/granules.csv?collection_concept_id=C1570116979-OB_DAAC&page_size=2000&temporal%5B%5D=2025-05-18T00:00:00Z&bounding_box[]=-97.9,21.1,-63.3,45.9&day_night_flag=day&page_num=1" | grep 2025141
1_MODISA_L1_A2025141182500.L1A_LAC,A2025141182500.L1A_LAC,2025-05-21T18:25:01.000Z,2025-05-21T18:29:59.000Z,https://oceandata.sci.gsfc.nasa.gov/cmr/getfile/A2025141182500.L1A_LAC,,,DAY,206.6071214675903
1_MODISA_L1_A2025141183000.L1A_LAC,A2025141183000.L1A_LAC,2025-05-21T18:30:01.000Z,2025-05-21T18:34:59.000Z,https://oceandata.sci.gsfc.nasa.gov/cmr/getfile/A2025141183000.L1A_LAC,,,DAY,218.1552133560181
1_MODISA_L1_A2025141200500.L1A_LAC,A2025141200500.L1A_LAC,2025-05-21T20:05:01.000Z,2025-05-21T20:09:59.000Z,https://oceandata.sci.gsfc.nasa.gov/cmr/getfile/A2025141200500.L1A_LAC,,,DAY,208.101053237915
1_MODISA_L1_A2025141201000.L1A_LAC,A2025141201000.L1A_LAC,2025-05-21T20:10:01.000Z,2025-05-21T20:14:59.000Z,https://oceandata.sci.gsfc.nasa.gov/cmr/getfile/A2025141201000.L1A_LAC,,,DAY,221.886815071106

Note that OBPG found the 16:55 and 21:40 granules while CMR did not.

Is this some sort of (probably obscure) thing I'm doing wrong or is there an issue with the CMR database?

Thank you!
-Andrew L.

[OB.DAAC - SeanBailey]

The CMR result is more correct. The subscription service uses a cartesian box of the NSWE extremes for a granule in the search. This can lead to false positives as the bounding box associated with a granule includes area outside the granule.
The A2025141165500.L1A_LAC granule does not contain any data in the region of interest you specified....it's close, but not within the box.

Sean

[alaroy]

Thank you for your response, Sean. That makes sense.