The Gaia Catalogue does not only produce a wealth of data, it also represents
a complex processing before a Catalogue can be issued.
The main data processing is being handled by three DPAC Coordination Units, CU3 for
the astrometric data, CU5 for the photometric data and CU6 for the spectroscopic data.
Then three Coordination Units analyse the processed data, CU4 for optical
or binary stars, solar system objects and extended objects, CU7 for variable stars,
and CU8 for classification.
Finally, CU9 takes care of the intermediate and final publication of the Gaia data.
Before proceeding to the validation of the data, the information obtained from different
Coordination Units (or even different processes inside a single Coordination Unit, such as
AGIS and IDU for CU3) have to be integrated and combined into a coherent unit.
For Gaia DR2 this also implies combining the information already processed during cycle 1
that was the base for the Gaia DR1 catalogue with updates from cycle 2.
These integrated records contain all the available information on the observed sources in a
format which is convenient for processing, but not for being stored and queried in a database.
Therefore these records must be converted to a more convenient format for validation
and presentation before being ingested into the Archive database. At the same time, sources
that do not meet some minimum astrometric or photometric quality are filtered out.
Once the data are in the Archive database providing convenient access to the data,
a crossmatch with external catalogues is computed, and some dedicated validation processes
are run to provide statistics, assess their consistency and flag sources that do not pass the
validation criteria. These sources ar then filtered out before the catalogue is released to the public.
This chapter describes these successive steps which are being followed
for the consolidation of the Catalogue.