Swift/XMM-Newton detection of a large glitch in the bursting transient Anomalous X-ray Pulsar CXO J164710.2-455216

ATel #932; G. L. Israel and S. Dall'Osso (INAF - AO Roma), S. Campana (INAF - AO Brera), M. Muno (Caltech), and L. Stella (INAF - AO Roma)
on 3 Nov 2006; 21:58 UT
Credential Certification: GianLuca Israel (gianluca@mporzio.astro.it)

Tweet

Swift observed the initial phases of the AXP CXO J164710.2-455216 outburst at 13 epochs between the 2006 September 21 and October 28 (see also ATel #893 and #896). Our timing analysis of the pulsations of this source included also two XMM-Newton observations carried out 5 days before and 2 days after the burst detected by BAT on 2006 September 21st (ATel #894, #902; GCN Circ. #5581) .

Our analysis unambiguously reveals the occurrence of a large glitch (spin-up) within the 7 day interval that separates the two XMM observations and contained the bursting event. Recovery from the glitch took place in about one week. The best fit exponential to the pulse arrival time residuals during the recovery phase gave a e-folding timescale of about 1.5 days and an amplitude Delta nu =1 e-5 Hz. Thus the glitch had an exceptionally large Delta nu/nu of about 1e-4, more than an order of magnitude higher than all previous glitches seen from AXPs and comparable to the largest ever recorded glitch from a radio pulsar. Subtracting this exponential component we were able to phase connect all the Swift and XMM observations, singling out a coherent solution with P=10.610648(4)s and Pdot=5.6(1.9) e-12\,s/s (1 sigma uncertainties). Both values above are consistent with those reported in ATel \#929. However, the uncertainties are significantly larger, because of the number of free parameters required to model both the spin-down and the glitch.

We note that the glitch component contaminates the phases of all the Swift observations carried out within 7-9 days from the BAT event. Future monitoring observations of this source will permit to separate completely the effects of the glitch from the secular trend and will provide more accurate P and Pdot measurements.