X-ray Astrometric Confirmation of Association of the Candidate Tidal Disruption Event ASASSN-14li with its Host Nucleus

ATel #6834; W. P. Maksym (Alabama), J. M. Miller (Michigan), S. B. Cenko (NASA/GSFC), J. J. Drake (SAO), S. Gezari (Maryland), R. Mushotzky (Maryland), J. Irwin (Alabama), K. Gultekin (Michigan), J. Kaastra (SRON), F. Paerels (Columbia), E. Ramirez-Ruiz (UCSC), M. Reynolds (Michigan)
on 18 Dec 2014; 02:57 UT
Credential Certification: Peter Maksym (peter.maksym@gmail.com)

Tweet

We report on the use of the 0th order images from the Chandra HRC-LETG DDT observations of ASSASN-14li (Miller et al., 2014; ATel #6800) to locate this likely tidal disruption event (TDE; Jose et al., 2014; ATel #6777) with respect to its host nucleus. By referencing counterparts in other bands, it is in principle possible to obtain absolute astrometry accurate to 0.1"-0.2" with HRC.

Obtaining optimal absolute astrometry is challenging due to the lack of strong field sources and Chandra's non-uniform off-axis PSF. The uncorrected offset from the SDSS DR8 location of PGC 043234 was ~1.0" in epoch #1 (obsid 17566; RA, Dec = 12:48:15.238, +17:46:25.37) and ~0.8" in epoch #2 (obsid 17567; RA, Dec = 12:48:15.252, +17:46:25.74) from the centroid of the 0th order peak. This is larger than the ~90th percentile limit of ~0.6" for Chandra's absolute astrometry. No obvious non-target reference sources were visible in either epoch. We thus corrected the aspect solution for the relative offset between epochs, reprojected the photon events, and merged the two event files. Using the wavelet detection tool wavdetect, we found two X-ray sources corresponding to SDSS objects besides PGC 043234. SDSS J124825.03+174937.7 (obj1) is an unresolved SDSS "star" (possibly a quasar) with ~24.5 net counts and SDSS J124808.84+174440.6 (obj2) is a faint, somewhat red galaxy, possibly a Sy2 or elliptical, with ~38.1 net counts. Each has ~0.15" statistical uncertainty. Other sources have no counterparts in FIRST or SDSS. Three sources are necessary for Chandra catalog matching tools, so we determined a mean offset from the individual sources. SAOTrace and MARX simulations show that the centroid of obj1 is within ~0.1" of its centroid, so we used its wavdetect position. The Chandra PSF asymmetry is significant for obj2, so we measured the relative offset from SDSS using the PSF peak as a reference point from the actual source location, which should give results similar to formal PSF fitting.

SDSS has flagged a nearby blue (u-g=1.03) object about ~10" south of PGC 043234 as a star. X-ray emission from a bright nearby star would complicate XMM-Newton analysis, but there is no X-ray source in the 0th order image, with an 84% confidence upper limit of 2.5x10^-4 counts/s.

We find a best 0th order position of (RA,Dec)=(12:48:15.213, +17:46:26.20), which is within ~0.3" of the SDSS position of the galaxy, and ~0.2" of the FIRST radio source corresponding to the nucleus of PGC 043234. This radio source may indicate star formation or a prior weakly accreting (e.g. a RIAF: radiatively inefficient accretion flow) AGN (3.5e21 W/Hz at 1.4 GHz), and does not necessarily preclude a TDE. The positional offsets are comparable to the uncertainty due to error propagation. We thus confirm that the flare is consistent with the host nucleus, to <130 pc at z=0.0206 (vs. R~2.5 kpc for the host galaxy).

This may be the best limit on publicly available astrometry for this flare. Holoien (private communication) informs us that the offset from the host nucleus (Jose et al. 2014; ATel #6777) is a best position, and that the astrometric uncertainty is significantly higher. Already the unusual X-ray characteristics of this object disfavor a supernova in the host nucleus (Miller et al., 2014; ATel #6800, ATel #6825), but the positional certainty from Chandra alone reduces the expected supernova rate (see e.g. Strubbe & Quataert, 2011) to ~3x10^-5/galaxy/year, comparable to recent estimates of the TDE rate (e.g. Holoien et al., 2014; Maksym et al., 2013; Khabibullin & Sazonov 2014). Although a non-grating Chandra HRC observation would improve the astrometry with better source match statistics, higher-resolution optical and UV observations (e.g. AO or HST) would be significantly better. Radio observations are urged to distinguish the spatial distribution of the radio emission, and to monitor changes related to any pre-existing jet structure.