Support ATel At Patreon

[ Previous | Next | ADS ]

EVLA observations suggest that M15 X-2 is the currently flaring source in M15

ATel #3378; J. C.A. Miller-Jones (ICRAR - Curtin), G. R. Sivakoff (U. Alberta), C. O. Heinke (Alberta), D. Altamirano (Amsterdam), E. Kuulkers (ESA/ESAC, Spain), M. Morii (Tokyo Tech)
on 25 May 2011; 02:28 UT
Credential Certification: Gregory R Sivakoff (grs8g@virginia.edu)

Subjects: Radio, X-ray, Binary, Globular Cluster, Neutron Star, Transient, Variables

Referred to by ATel #: 3393, 5327, 15586, 15604

Following the detection of an X-ray flare in the globular cluster M15 (ATel #3356) and its subsequent Swift XRT localization to the cluster core (ATel #3363), we undertook EVLA observations to determine the source of the flaring event. The 1-hour observation on 2011 May 22 (12:11-13:11 UT) comprises of two 1024-MHz bands centred at 5.0 and 7.0 GHz. The array was in its BnA configuration, providing an angular resolution of 1.25 x 0.63 srq arcsec at 5 GHz and 0.92 x 0.50 sqr arcsec at 7 GHz.

We detected two radio sources within the cluster core, at positions consistent with the two known X-ray sources AC 211 (an accretion disc corona source; Auriere et al., 1984, A&A, 138, 415) and M15 X-2 (an ultracompact binary; Dieball et al., 2005, ApJ, 634, L105). For AC 211, we measured flux densities of 276 ± 11 microJy/beam at 5 GHz and 267 ± 10 microJy/beam at 7 GHz, yielding a spectral index of -0.10 ± 0.17. For M15 X-2, we measured flux densities of 156 ± 9 microJy/beam at 5 GHz and 199 ± 10 microJy/beam at 7 GHz, yielding a spectral index of 0.72 ± 0.35.

Previous radio observations of M15 (Kulkarni et al., 1990, ApJ, 363, L5; Johnston et al., 1991, ApJ 382, L89; Bash et al., 2008, AJ, 135, 182) have all detected AC 211, showing evidence for variability between epochs. Prior to our observations, M15 X-2 had never been detected in the radio band, despite the previous observations having had the sensitivity and resolution to detect it and resolve it from AC 211. The deepest available constraint on the radio emission from M15 X-2 (from Bash et al., 2008) is a 3-sigma upper limit of 25.5 microJy/beam at 8.4 GHz.

Given the known low-level radio variability of AC 211, the increase in flux of M15 X-2 by a factor >8, and the lack of any other detected radio source within the Swift error circle, we infer that M15 X-2 is likely the source of the ongoing X-ray flaring event. However, scheduled Chandra observations should confirm this. The inverted radio spectrum would suggest that the flare was not optically thin at 5GHz by the time of our observations.

To determine the accretion state at the time of the radio observations, we combined Swift XRT (May 22 11:55-12:06 UT and 13:39-13:46) and RXTE (May 23 02:46 - 03:34 UT) observations. The combination of a multi-colour disk black body model (diskbb with kT_in ~ 0.26 keV) and a Comptonization model (comptt with 0.37 keV seed photons Comptonized by a ~ 2.2 keV plasma with an optical depth of ~ 2.2), accounting for Galactic (N_H = 6.7e20 cm-2) and intrinsic absorption (N_H ~ 1.2E21 cm-2), fit the model well (chi^2 of 65.8 for 68 dof) in our preliminary fits. The fit corresponds to a 1.2E37 (7.1E36) erg/s observed luminosity in the 0.5-10 (2-10) keV band. Intrinsic luminosities are only marginally higher (<~ 12%). The source is dominated by the Comptonization component. Between 0.5-10 keV, the disk component is 30 ± 5%; however, this drops to < 1.5% in the 2.0-10 keV band. Thus, the source spectra is consistent with the hard state. However, RXTE/PCA observations do not show the variability in the 0.1-10 Hz range typically associated with the hard state; the measured fractional rms amplitude is 3 ± 3%.

Although AC 211 is the brighter radio source, these X-ray observations include only a minimal component from AC 211. X-ray emission from AC 211 is heavily absorbed by its accretion disk, and including the X-ray spectra of AC 211 reported in White & Angelini (2001,ApJ, 561, L101) neither improves the fit, nor accounts for more than 3% of the observed flux.

We encourage continued multi-wavelength observations of this flaring source.

We also thank the EVLA, RXTE, and Swift X-ray Observatory, for their rapid response.