I've collected data having 1 million rows of Gaia DR3. I am trying to make a dataset suitable for Machine Learning. I need a target variable which is the stellar object type. My issue is I can't seem to find any object in the simbad database that exactly matches with the 'ra' and 'dec' coordinates in Gaia DR3. Whenever I run a query, it gives me result.

This is what I use to query simbad in Python 3.11: Simbad.query_region(SkyCoord(ra=df['ra'].iloc[i], dec=df['dec'].iloc[i], unit=(u.deg, u.deg), frame='icrs'), radius=0.01*u.deg)

I am getting:

          MAIN_ID                  RA      ...    OTYPE     SCRIPT_NUMBER_ID
                                "h:m:s"    ...
---------------------------- ------------- ... ------------ ----------------
Gaia DR3 5952837905070694016 17 22 34.5927 ... LongPeriodV*                1
     2MASS J17224991-4449366 17 22 49.9096 ...      HighPM*                1
Gaia DR3 5953202251423606400 17 22 52.8099 ... LongPeriodV*                1
                   HD 156919 17 22 23.9468 ...         Star                1
Gaia DR3 5953201018811842944 17 23 10.2454 ... LongPeriodV*                1
Gaia DR3 5952837355356355072 17 22 34.7870 ... LongPeriodV*                1
                   HD 157063 17 23 12.7249 ...    alf2CVnV*                1
Gaia DR3 5952825187656206464 17 23 00.8129 ... LongPeriodV*                1
Gaia DR3 5953214071173664512 17 22 42.3904 ...       EclBin                1

Where Gaia DR3 5953201637243132416 is the object ID in Gaia DR3 dataset and 260.6787873092191 -44.78877169277981 is the coordinate of the object in degrees. Is there any other way to get the correct object? I am aiming to get the object type of each stellar object that matches the coordinates in Gaia DR3.

  • $\begingroup$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. $\endgroup$
    – Community Bot
    Commented Dec 13, 2022 at 12:22
  • $\begingroup$ This may help: Pre computed cross matches cosmos.esa.int/web/gaia-users/archive/combine-with-other-data $\endgroup$ Commented Dec 14, 2022 at 13:07
  • 1
    $\begingroup$ @GregMiller Thank you for this!! Your comment has made everything easier to work with. The website is a bit slow for queries though. i found a list of tables with cross-matched data from different catalogues. $\endgroup$
    – VADeR
    Commented Dec 14, 2022 at 19:01
  • $\begingroup$ I'm not sure what result you were expecting. You have searched on a radius of 36 arcseconds, which will in generalcontain lots of objects in the SIMBAD catalogue. A radius of 1 arcecond might be more appropriate. $\endgroup$
    – ProfRob
    Commented Jan 12, 2023 at 19:50

2 Answers 2


The right ascension and declination listed in the various Gaia data releases are the position at the reference epoch for that catalog and this is not J2000 and also varies for each Data Release (DR1 was epoch 2015.0, DR2 was 2015.5, DR3 is 2016.0; documentation of the DR3 gaia_source data model and fields). If you query SIMBAD, you will be getting J2000 positions and these will not match exactly as there are 4 years difference in the epoch and accumulated proper motion.

I suggest you use Vizier, available through the Python astroquery module which will allow the computation of J2000 positions, taking into the account the proper motion. This is shown in the screenshot below which shows the original RA_ICRS, DE_ICRS columns from the Gaia DR3 catalog and the calculated J2000 positions _RAJ2000, _DEJ2000 columns in brown.

Vizier query table result showing the difference between J2000 and J2016 positions in DR3

  • 2
    $\begingroup$ Thank you for this!! I didn't know the data varies like that. $\endgroup$
    – VADeR
    Commented Dec 14, 2022 at 18:59

I wrote a code queries Simbad gives different catalog IDs of an object. I hope this may help you. https://github.com/BedriKeskin/SimbadQueryID


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .