AGASC 1p8

• Introduction
• Data Products
• Analysis and Results
• Validation and Testing
• Other Issues
• AGASC Update
• Approval

Introduction

AGASC 1p8 is the fifth update to the AXAF Guide and Acquisition Star Catalog (AGASC) since launch. This update improves ACA magnitudes and positions using the Gaia DR3 catalog and more than 94,000 stars observed with the ACA.

The AGASC 1p8 update was motivated by inaccuracies in the catalog:

• Magnitude uncertainties are large for some stars. The median magnitude uncertainty is 0.4.
• Positions are based on pre-1995 measurements. Of all stars used in daily operations, 16% have no proper motion information, and only 37% have proper motion information in both RA and dec. This means tha position uncertainties grow with time. The largest observed outlying angular offset is 29 arcsec.

It is difficult to account for these inaccuracies in daily operations, and as a consequence:

• Stars that are fainter than in the catalog might not be acquired
• A star that is brighter than expected might be rejected while tracking if mag < 5.2.
• A star located more than 5 arcsec from the expected location will be rejected while tracking.
• A spoiler that is brighter than expected can affect the attitude determination.

The AGASC catalog has undergone several updates, where a few external catalogs have been progressively incorporated. As of AGASC 1.7, more than 98% of stars used for attitude control and aspect determination come from either the Tycho-2 Catalog (based on ESA's Hipparcos mission) or the GSC2.3 Catalog.

AGASC 1.8 includes information from the Gaia DR3 catalog. Gaia is a European Space Agency mission providing high-precision astrometry and multi-band photometry of more than 1000 million stars in the Milky Way. The spectral response of Gaia matches the ACA response very well, thus providing a good estimate of the ACA magnitudes in different spectral bands. Gaia also has high astrometric accuracy (200 times more accurate than Hipparcos) and 99% of surveyed stars have measured proper motion. All this makes it ideal to update the AGASC.

8903379 stars were matched with stars in Gaia DR3, and their magnitudes and positions were updated in AGASC 1p8. This includes more than 98% of stars used in daily operations. The median magnitude uncertainty of the updated stars is reduced from 0.4 to 0.02, and their median position error is reduced from 100 milli-arcsec to less than 1 milli-arcsec.

For comments or questions about this release, contact Aspect Help

Data Products

The main data products of the AGASC 1p8 update are the following:

• /data/agasc1p8 . The updated AGASC catalog with positions and mag_ACA estimates based on Gaia photometry stored as a collection of FITS files. These files are used in the ground pipeline.
• Files used in daily operations, distributed as part of the Ska package:
  • agasc1p8.h5 . The updated AGASC catalog with positions and mag_ACA estimates based on Gaia photometry.
  • proseco_agasc_1p8.h5 . A subset of AGASC consisting of bright stars, plus all faint stars in their vicinity.
  • miniagasc_1p8.h5 . A subset of AGASC consisting of bright stars.

Analysis and Results

What follows is a brief summary of the analysis. More information can be found in the analysis page.

Cross-match

The cross-match between AGASC and Gaia was performed on 8903379 stars with defined GSC1.1 and Tycho2 IDs. Only stars with CLASS 0, 2 or 6 are cross-matched. These correspond to star, blend or member of incorrectly resolved blend and known multiple system. We used two independent algorithms:

• The tycho2-tdsc and GSC2.3 cross-matching tables included in Gaia DR3. This algorithm considers star positions, not magnitudes.
• A bespoke cross-matching algorithm that takes into account position, proper motion and magnitude. This model included the following features:
  • A matching probability density function of magnitude and position.
  • An algorithm to deal with complex cases in difficult star fields matching more than one AGASC star at once.
  • A numerical estimate of the p-value of any given match.
  • A prescription for determining the p-value threshold for accepting matches.

Outlier analysis was performed to verify that the cross-matches are reasonable, with a discrepancy rate less than 0.01%

ACA Magnitude Model

The ACA magnitude is estimated from the measured Gaia magnitudes in the three Gaia passbands (mag_G, mag_Rp, mag_Bp). To derive this model, we used a sample of 94,000 stars observed with the ACA. The fit procedure only used magnitudes smaller than 8.5, as instrumental effects become important for fainter stars.

The chosen ACA magnitude model, as a function of Gaia magnitudes, is of the form: \[ \begin{eqnarray} \text{mag}_{ACA} &=& \text{mag}_{G} + Pol(color, N=2) \\ color &=& \text{mag}_{Bp} - \text{mag}_{Rp} \end{eqnarray} \]

The mean residuals and their standard deviation are summarized in the following figure as a function of magnitude and color. The horizontal lines represent the 1-sigma uncertainty given by the base uncertainty described below.

The ACA magnitude uncertainty is given by a variance model with several additive contributions:

Model Uncertainty

• base uncertainty. A constant determined for bright stars, below magnitude 8.5, is applied to all stars.
• variability. An estimate of variability is taken directly from Gaia's Median Absolute Deviation (MAD) for G FoV transits (vari_summary.mad_mag_g_fov). This contribution to the magnitude uncertainty has no tunable parameters and was verified to be a good estimate of the uncertainty of variable stars observed with the ACA.
• missing magnitudes. 99.97% of stars have magnitudes measured in all three Gaia passbands. In the remaining cases, the ACA magnitude is estimated using the available magnitudes, and the mean variance estimate is increased to account for this.

The final uncertainty is then given by \[ \sigma = \sqrt{\text{var}_{base} + \text{var}_{variable} + \text{var}_{missing}} \]

Validation and Testing

The magnitude model and catalog cross-match were developed using a 80/20 train/test split. The finalized dataset was tested and validated independently by the Aspect team.

Other Issues

Repeated entry. The AGASC 1p7 catalog has one repeated ID (AGASC ID 154534513). One of these two is removed in 1p8.

Potential duplicates. Looking at cross-matches with Gaia, it became apparent that there might be duplicate stars in the AGASC catalog. This can happen when the union of two catalogs is taken without accounting for stars present in both catalogs with slightly different positions and magnitudes. These potential duplicates were not removed.

AGASC Update

AGASC catalog will be updated in the following fields. The fields are updated only for the stars that have updated magnitude, unless otherwise stated:

• XREF_ID1, XREF_ID5. The ID of the Gaia counterpart (a 64-bit integer). The highest significant 32 bits of the Gaia ID are stored in XREF_ID1, and the lowest significant 32 bits are stored in XREF_ID5. The default Gaia ID value is -9999, which translates to the following default values XREF_ID1=-1 and XREF_ID5=-9999. The full Gaia ID can be reconstructed as follows:

  agasc["XREF_ID1"] * (1 << 32) + agasc["XREF_ID5"].astype(dtype=np.uint32)

• RSV4. Long integer which is the star number within the region in the AGASC Version 1.0 (= GSC1.1). This is not a unique identifier. Default value of -9999. In AGASC version 1p7 and earlier, this was stored in XREF_ID1.

• MAG, MAG_ERR, MAG_BAND, MAG_CATID. Gaia magnitude and its error.

  MAG_BAND = 23, 24 or 25, depending on whether it is the G, Rp or Bp magnitude.
  MAG_CATID = 7

• MAG_ACA, MAG_ACA_ERR. Estimated ACA magnitude and its error.

  MAG_ACA_ERR = mag_aca_err_gaia * 100

• COLOR1, COLOR2, COLOR1_ERR, COLOR2_ERR, C1_CATID, C2_CATID. Estimated B-R color and its error.

  COLOR1 = bp_mag - rp_mag
  COLOR1_ERR = sqrt(bp_mag_error**2 + rp_mag_error**2)
  COLOR2 = bp_mag - rp_mag
  COLOR2_ERR = sqrt(bp_mag_error**2 + rp_mag_error**2)
  C1_CATID = 7
  C2_CATID = 7

• RA, DEC, EPOCH, POS_CATID, POS_ERR. Position and its error.

  RA = gaia_ra
  DEC = gaia_dec
  EPOCH = 2016.0
  POS_CATID = 7
  POS_ERR = sqrt((gaia_ra_error/cos(gaia_dec))**2 + gaia_dec_error**2) * 1000

• PM_RA, PM_DEC, PM_CATID. Proper motion and its error.

  PM_RA = gaia_pm_ra
  PM_DEC = gaia_pm_dec
  PM_CATID = 7

• ASPQ1. Updated for all entries in catalog

• ASPQ2. Boolean cast as int:

  pm = np.sqrt(
      (agasc_gaia["pm_ra"] / np.cos(np.deg2rad(agasc_gaia["DEC"]))) ** 2
      + agasc_gaia["pm_dec"] ** 2
  )
  ASPQ2 = np.where(pm[sel] > 500, 1, 0)

• VAR. only for stars that are flagged as variables in Gaia.

  VAR[(mad_mag_g_fov < 0.2)] = 5
  VAR[(mad_mag_g_fov >= 0.2) & (mad_mag_g_fov < 2)] = 3
  VAR[(mad_mag_g_fov >= 2)] = 4
  VAR_CATID = CAT_ID

Approval

SSAWG discussed and approved the proposed AGASC 1p8 update on 2024-06-18 .

The updated was approved by the flight director on...