Photocenter-barycenter offsets for Gaia observations with high-precision asteroid shape models

ESA’s Gaia mission provides astrometry of asteroids at an unprecedented accuracy measured, at best, in milli- or even microarcseconds. However, the accuracy decreases for larger objects due to shape and size effects, such as the offset of the apparent photocenter and the true barycenter of the object. The effect has been studied with a highly accurate shape model for one target, (21) Lutetia [1], and in a statistical sense for the over 156,000 asteroids in Gaia Focused Product Release (FPR) [2] using a spherical shape approximation [3]. Both studies show that the residuals in the along-scan direction improve when the photocenter-barycenter offset is accounted for in the orbital fitting.

To take a step further, we first derive photocenter-barycenter offsets for 35 large main-belt asteroids using shape models reconstructed with the ADAM method from VLT/SPHERE/ZIMPOL images and available lightcurve data [4]. In addition, we apply a correction for the offset to Gaia Data Release 3 (DR3) astrometry and inspect the effect on orbit computation for asteroids (21) Lutetia and (511) Davida by fitting their orbits in a least-squares sense using the open-source orbit-computation software OpenOrb [5]. The offsets are computed following e.g., [6] for each transit epoch of a given target asteroid in Gaia DR3. The correction is performed by subtracting the offsets from the corresponding observations: each transit contains up to 9 observations, and it is assumed that the offset remains constant throughout a transit.

We find that the photocenter-barycenter offset is on average 4–7% of the diameter of the asteroid, which in many cases translates to an angular offset of some milliarcseconds. Considering Gaia’s astrometric precision, such an offset is significant. We also find a nearly linear dependence between the average magnitude of the offset and the size of the asteroid. In Fig. 1 the magnitude of the offset is computed as an average over all the DR3 transits of a given target. As expected, the magnitude of the offset increases as the diameter of the asteroid increases. The linear nature of the dependence suggests that there is a possibility to derive a useful empirical model for the offset based on its size. However, it is likely that such a model would have to take into account the shape of an asteroid in addition to its size as well as possible albedo variations. Additionally, we show that the orbital fits for (21) Lutetia and (511) Davida improve in terms of the number of rejected (outlier) observations and along-scan residuals.

Figure 1: Magnitude of the photocenter-barycenter offset averaged over DR3 transits as a function of the diameter of the asteroid for 35 large main-belt asteroids.

Finally, we will report on asteroid mass estimation based on asteroid-asteroid close encounters. Our goal of achieving mass estimates with improved and realistic uncertainties should be possible given the highly accurate Gaia astrometry and the addition of a photocenter-barycenter correction, which is relevant for the target asteroids for mass estimation, as they are often relatively large. The new mass estimates and the accurate shape models, that are used for the photocenter-barycenter offset correction, allow us to derive more realistic bulk density estimates.

References:

[1] Gaia collaboration, P. Tanga et al. 2023, Astronomy and astrophysics, 674, A12 

[2] Gaia collaboration, P. David et al. 2023, Astronomy and astrophysics, 680, A37 

[3] O. Fuentes-Muñoz et al. 2024, The Astronomical journal, 167, 290 

[4] P. Vernazza et al. 2021, Astronomy and astrophysics, 654, A56 

[5] M. Granvik et al. 2009, Meteoritics and Planetary Science, 44, 1853 

[6] K. Muinonen and K. Lumme 2015, Astronomy and astrophysics, 584, A23