The Sleeping Monster: NuSTAR observations of SGR 1806-20, 11 years after the Giant Flare

George Younes, Matthew G. Baring, Chryssa Kouveliotou, Alice Harding, Sophia Donavan, Ersin Gogus, Victoria Kaspi, Jonathan Granot

Published 2017-10-31Version 1

We report the analysis of 5 NuSTAR observations of SGR 1806-20 spread over a year from April 2015 to April 2016, more than 11 years following its Giant Flare (GF) of 2004. The source spin frequency during the NuSTAR observations follows a linear trend with a frequency derivative $\dot{\nu}=(-1.25\pm0.03)\times10^{-12}$ Hz s$^{-1}$, implying a surface dipole equatorial magnetic field $B\approx7.7\times10^{14}$ G. Thus, SGR 1806-20 has finally returned to its historical minimum torque level measured between 1993 and 1998. The source showed strong timing noise for at least 12 years starting in 2000, with $\dot{\nu}$ increasing one order of magnitude between 2005 and 2011, following its 2004 major bursting episode and GF. SGR 1806-20 has not shown strong transient activity since 2009 and we do not find short bursts in the NuSTAR data. The pulse profile is complex with a pulsed fraction of $\sim8\%$ with no indication of energy dependence. The NuSTAR spectra are well fit with an absorbed blackbody, $kT=0.62\pm0.06$ keV, plus a power-law, $\Gamma=1.33\pm0.03$. We find no evidence for variability among the 5 observations, indicating that SGR 1806-20 has reached a persistent and potentially its quiescent X-ray flux level after its 2004 major bursting episode. Extrapolating the NuSTAR model to lower energies, we find that the 0.5-10 keV flux decay follows an exponential form with a characteristic timescale $\tau=543\pm75$ days. Interestingly, the NuSTAR flux in this energy range is a factor of $\sim2$ weaker than the long-term average measured between 1993 and 2003, a behavior also exhibited in SGR $1900+14$. We discuss our findings in the context of the magnetar model.