Swift-XRT globular cluster monitor: April 2016 Terzan 5 observations

ATel #8996; M. Linares (IAC, Spain & NTNU, Norway), J. Chenevez (DTU Space, Denmark)
on 27 Apr 2016; 14:11 UT
Credential Certification: Manuel Linares (linares@iac.es)

Tweet

As part of our Swift-XRT monitoring program of Galactic globular clusters, publicly available here, Swift observed Terzan 5 on 2016 April 17 for 3.2 ksec. Two additional 1.5-ksec-long observations were taken on April 21 and 23 (ATel #8982).

In the full XRT dataset, the "quiescent" 0.5--10 keV X-ray luminosity of Terzan 5, integrated within the half-light radius (0.72'), is in the range Lx=[1--4]e34 erg/s (using a distance of 6.9 kpc, from Harris 2010). This integrated luminosity is in general due to a combination of faint X-ray sources near the core of Terzan 5, some of which are known to be variable (at least 19 sources are known within the 9.6" core radius; Heinke et al. 2006, ApJ, 651:1098). In the latest observations taken after the 2015 March-June outburst of EXO 1745-248 we find:

2015-07-02 obs: 00032148057 Lx= 2.2e+34 +/- 1.5e+34 erg/s
2015-07-06 obs: 00032148058 Lx= 2.4e+34 +/- 1.1e+34 erg/s
2015-07-09 obs: 00032148059 Lx= 1.5e+34 +/- 1.1e+34 erg/s
2015-07-12 obs: 00032148060 Lx= 1.9e+34 +/- 7.4e+33 erg/s
2015-08-10 obs: 00032148061 Lx= 2.4e+34 +/- 6.8e+33 erg/s
2016-04-17 obs: 00032148062 Lx= 5.0e+34 +/- 9.1e+33 erg/s
2016-04-21 obs: 00032148063 Lx= 5.5e+34 +/- 1.7e+34 erg/s
2016-04-23 obs: 00032148064 Lx= 2.7e+34 +/- 1.1e+34 erg/s

Inspecting closely the 2016-04-17 and 2016-04-21 observations after the report of a faint outburst (ATel #8982), and subtracting the 2015 quiescent emission (2.1e34 erg/s estimated from the weighted average of the 2015 July-August observations), we find that Lx increased by [2.9+/-0.9]e34 erg/s and [3.4+/-1.7]e34 erg/s, respectively. This is close to the lowest luminosity of an outburst state (1e34 erg/s in Linares 2014, ApJ, 795:72; but definitions may vary), and about ten times more luminous than the accretion disk state of known redback millisecond pulsars.

Moreover, we find that the XRT core emission during the 2016-04-17 observation is extended and shows evidence for two partly blended sources about 9" away (see this image). The UVOT-enhanced position is dominated by the brighter source in the pair, and is consistent with the neutron star transient EXO 1745-248 (T5X1, source C3 in Heinke et al. 2006; ATel #8982). Applying the same astrometric correction to the fainter source suggests that this may be associated with the unidentified source C38 in Heinke et al. (2006; which they locate to RA(J2000)=17:48:05.391, DEC(J2000)=-24:46:56.28).

We caution that XRT location and spectral results should be interpreted with care under these circumstances (a ~7" FWHM PSF and two blended sources ~9" away). Having said this, our results are consistent with T5X1 being in a prolonged low-level activity state since its 2015 outburst finished, and with a new and brief activity episode in 2016 April 17-21 associated with C38 or a nearby source. We encourage multi-wavelength follow-up of this event to identify its nature.

We thank Phil Evans for discussions on the UVOT astrometric correction, and the Swift team for making the globular cluster monitor possible. We acknowledge support from the COST Action MP1304 "NewCompStar". This work made use of data supplied by the UK Swift Science Data Centre at the University of Leicester.

The Swift-XRT globular cluster monitor