Not logged inOcean Color Forum
Up Topic Products and Algorithms / Satellite Data Access / Earthdata CMR WIP?
- By b.detracey Date 2017-11-07 11:01
I recently noticed the SeaDAS python script which led me to the Earthdata CMR search API. The following collection search for OB.DAAC L1/L2 datasets gives:
$ wget -q "\
&processing_level_id=1&processing_level_id=2&pretty=true" -O - | grep name | sort

I have two questions:
Are there plans to add the MODIS and VIIRS L1 metadata to the CMR?
At present there seems to be no way, on the collection level, to discriminate between full and reduced resolution MERIS, between full and reduced resolution OLCI, or between GAC and MLAC SeaWiFS. Are there plans to allow discrimination of these as separate collections? Or should I 'manually' discriminate by filtering the results of a granule search?

Sorry. I guess that was three questions... :eek:
- By SeanBailey Date 2017-11-13 09:08
1) Maybe.  Currently, the MODIS L1 data are in CMR, but cataloged by MODAPS/LAADS.   I had thought they would do the same for VIIRS...
2) No
3) Yes

- By b.detracey Date 2017-11-21 08:38
Got it. Thanks Sean.
Searching on Earthdata the only L1 VIIRS product is VIIRS/NPP Moderate Resolution Terrain Corrected Geolocation 6-Min L1 Swath 750m Light V001. It would be helpful if your MODIS L1 assets were added as my understanding is that they are not the same as those of MODAPS/LAADS.

A few (hopefully) last questions, mostly directed at understanding whether a CMR query can tell me if a collection and/or granule has been fixed/reprocessed/updated.
-It looks like only your refined products are in the CMR. Makes sense to me but could you please verify?
-Within CMR, will the granule search parameter 'updated_since' or 'revision_date' tell me if a granule itself has been updated? Or can this not discriminate from a metadata update? I am trying to understand if CMR can be used to determine, at my year's end reprocessing, whether individual granules have been altered/fixed by your group. Or does it make more sense to stick with checksums from your own file search api?
-What is the reason that the CMR granule metadata "Granule/Collection/VersionID" matches only the major processing version and not the netcdf pversion attribute?
- By SeanBailey Date 2017-11-21 16:05
The MODIS L1A files we distribute are identical to the LAADS data.  L1B is not, but we don't distribute L1B (LAADS does, though).

Yes, currently we only put the refined data in CMR.
We generally do not alter files between reprocessings (except for the quicklook/refined delineation, but as we're only putting refined in CMR, a non-issue).
The collection versionID refers to a consistent set.  The sub versions in the netCDF pversion attribute typically are just calibration updates that affect a subset of the data.
The calibration is consistent within the larger collection, but we may have had to reprocess a portion of the data to ensure that consistency.

Up Topic Products and Algorithms / Satellite Data Access / Earthdata CMR WIP?

Powered by mwForum 2.29.7 © 1999-2015 Markus Wichitill