Rebrightening of GRS 1915+105 Observed by NICER

ATel #14792; J. Neilsen (Villanova), D. Altamirano (Southampton), J. Homan (Eureka Scientific), K. Gendreau (NASA GSFC), J. Steiner (SAO, CfA), Z. Arzoumanian (NASA GSFC)
on 16 Jul 2021; 18:43 UT
Distributed as an Instant Email Notice Transients
Credential Certification: Joey Neilsen (jneilsen@villanova.edu)

Tweet

On July 9, 2021, MAXI reported a sharp flux increase from the stellar-mass black hole GRS 1915+105. After 26 years of persistent high luminosity and structured variability, GRS 1915+105 faded exponentially in 2018, eventually settling (after a sudden final decrease in mid 2019) at a few percent of its prior apparent brightness (Negoro et al., ATel #11828; Homan et al., ATel #12742). Spectra from the Neil Gehrels Swift Observatory and NuSTAR suggested strong partial covering absorption with emission and absorption lines (Miller et al., ATel #12771, #12788). We now understand this "obscured state" to be driven by the appearance of a Compton thick obscurer whose origin is unclear (Miller et al. 2020; Neilsen et al. 2020; Balakrishnan et al. 2021). Rebrightening events are of interest because they (a) suggest the source is still active and (b) probe the obscuring medium; similar events were observed in April and August of 2020 (Aoki et al., ATel #13652; Lepingwell et al., ATel #13676; Reynolds et al., ATel #13974).

In response to the MAXI GCN alert, we increased our monitoring frequency of GRS 1915+105 with NICER; recent observations cover the range from July 2 to July 12 and show a transition from ~25 cts/s to 250 cts/s over this period. The average spectrum is dominated by the data taken since July 10, 15:37:15 UT, and is well described as absorbed (NH=6.7e22 cm^-2) thermal Comptonization (nthcomp), with a photon index of 1.43 and a very low electron temperature of 1.5 keV; we fit a bend at >8 keV with a smeared edge. This model is suitable for measuring the observed flux, which is 1.8E-9 erg/s/cm^2 and rises to 2.3E-9 erg/s/cm^2 in the last two days of data. The spectrum also contains over a dozen strong absorption lines from Si, S, Ar, Ca, Fe, and Ni, as well as an emission line consistent with Fe K-alpha. Preliminary analysis with zxipcf indicates an ionized absorber with NH>5e23, but there is no evidence of the cold Compton thick obscuration.

We also found that the slow (seconds) and fast (sub-second) variability has changed significantly as the intensity of GRS 1915 has increased. In two visits at low flux (July 2 and July 4) NICER detects an average 0.5-12 keV intensity of around 20 cts/sec and some flare-like variability on timescales of seconds. As the average intensity increases, the slow variability increases in strength, and we see changes of intensity by a factor of 5-7 on timescales of 100-200 seconds. In the later visits, GRS 1915 is observed at ~240 cts/s, showing sporadic dips where the intensity decreases by 50-70% for a few tens of seconds (similar to the omega class of Klein-Wolt et al. 2002).

We created 0.5-12 keV power spectra of all data available. Initially, when GRS 1915 was observed at an average rate of ~20 cts/s, there were no significant quasi-periodic oscillations (QPOs) in the power spectrum, though we did find broadband noise between 0.01 Hz and 1 Hz at a level of 21+-2% fractional rms. As the intensity increases, the broadband noise evolves to higher frequencies and a QPO appears. At an average count rate of 240 cts/s, the power spectrum shows a QPO at 0.17+-0.01 Hz with a fractional rms amplitude of 8.0+-0.4% and a quality factor Q=3.6+-0.5. The 0.01-10 Hz integrated fractional rms amplitude is on the order of 15% (15.1+-0.2%). For comparison to the low-intensity observation, the 0.01-1 Hz integrated fractional rms amplitude was 12.5+0.2% when GRS1915 was detected at 240 cts/s.

Given the unusual evolution of this source, further observations are encouraged.

Plots of Spectrum and Power Spectrum