martin-boettcher
commented
12 years ago Hi Mark,
oops. The reason may be how we had handled partial failure when we had reingested ARC3 and NWP results. Instead of deleting the temparea directory of the erroneous sensor (that is the way we do it today) we had simply marked the reingestion task as done. This had dropped the second part of reingestion which is the registration of matchup NWP data: The following time periods may be affected:
1992/reingest3-1992-04-b.out:done 2006/reingest3-2006-03-d.out:done 2006/reingest3-2006-03-e.out:done 2006/reingest3-2006-10-a.out:done 2006/reingest3-2006-10-b.out:done 2006/reingest3-2006-10-c.out:done 2006/reingest3-2006-10-d.out:done 2006/reingest3-2006-10-e.out:done 2006/reingest3-2006-10-f.out:done 2007/reingest3-2007-10-a.out:done 2007/reingest3-2007-10-b.out:done 2007/reingest3-2007-10-c.out:done 2007/reingest3-2007-10-d.out:done 2007/reingest3-2007-10-e.out:done 2007/reingest3-2007-10-f.out:done
Hi Martin,
The MMD file
/exports/nas/exports/cse/geos/scratch/gc/sst-cci/archive/mmd/v1/2006/mmd-2006-10.nc
does not have any matchup NWP data (at least, it has no matchup.nwp.an.sea_surface_temperature and several other variables) - they are all fill value.
Mark