Users wishing to look up the E/DR3 record for an astrophysical source identified in DR2 must NOT simply extract the record from E/DR3 having the same source identifier.
As described in the detailed description of attribute designation in gaia_source it is not guaranteed that the same astronomical source will always have the same source identifier in different Data Releases. Hence the only safe way to compare source records between different Data Releases in general is to check the records of proximal source(s) in the same small part of the sky. This table provides the means to do this via a precomputed crossmatch of such sources, taking into account the proper motions available at E/DR3.
Within the neighbourhood of a given E/DR3 source there may be none, one or (rarely) several possible counterparts in DR2 indicated by rows in this table. This occasional source confusion is an inevitable consequence of the merging, splitting and deletion of identifiers introduced in previous releases during the DR3 processing and results in no guaranteed one–to–one correspondence in source identifiers between the releases.
For more details of the procedure used to create this crossmatch, see Chapter 11 in the online documentation.
Source identifier assigned during DR2 processing for an astrophysical object in close proximity to (possibly) the same object in DR3.
Source identifier assigned during DR3 processing for an astrophysical object in close proximity to one or more (including possibly the same) objects in DR2.
Angular (great–circle) distance on sky between the source pairing between DR2 and DR3.
G band magnitude difference between the sources based on the individual magnitudes materialised in gaia_source at the time of the respective data releases. The sense of the difference is DR3 minus DR2.
This flag indicates whether the DR3 source has proper motion estimates. If yes, they have been used to propagate DR3 coordinates to the DR2 epoch (J2015.5). A simple linear correction has been applied, due to the epoch proximity between both catalogues (0.5 yr).