skip to main content

gaia data release 3 documentation

4 Astrometric data

4.4 Processing steps

Author(s): Jose Hernandez

The main objective of the AGIS processing is the accurate estimation of the source positions, their parallaxes and proper motions using as main input the observations obtained by Gaia. Each field of view transit is packed in an object named AstroElementary which contains information (times and across scan coordinate) of each individual transit of the star across the CCDs. Other inputs needed in AGIS are the working catalogue, the match records linking observations with sources, and auxiliary information like the Gaia and Solar System ephemeris, clock calibrations to convert the on board time to barycentric time, etc.

The main processing steps which are performed in AGIS are:

  • AGIS pre-processing: This phase takes all the input data available to AGIS (observations, working catalogue, matches, etc.) and processes them transforming the input data to a form better suited for the processing, sorting it so that all the observations of each source are put together. During the pre-processing some filtering on the input data is done so that transits where the quality is dubious are filtered out. The last step of the pre-processing is the selection of the primary source set which will be used to determine the attitude and calibration.

  • Primary source processing: This phase consists of the determination through an iterative scheme of the astrometry of the sources which were selected as primaries, the spacecraft attitude and the instrument calibration parameters all with the best possible accuracy for the given models and observations available. When the solution has been computed the primary sources and the attitude are aligned to the ICRF.

  • Secondary source processing: This phase consists of the determination of the astrometry for the rest of the sources not considered as primaries using their observations and the attitude and calibration computed during the Primary source processing.

  • AGIS post-processing: This phase takes care of merging the results of the primary and secondary processing. It also puts the AGIS outputs sent to the MDB in a format more suited for the consumers of the data, converting the times back to OBMT and stripping auxiliary data not useful outside AGIS.