NuSTAR looks at Swift J0243.6+6124 during the rise of the 2023 outburst

On 8 Apr 23, MAXI detected an outburst from Swift J0243.6+6124 and the X-ray flux is exhibiting a steady increase and is being monitored by various instruments in X-rays like Swift, NICER and by the VLA at radio wavelengths (ATels #15983, #15984, #15987, #16076, #16101, #16121). NuSTAR observed the source for ~46 ks starting at MJD 60122.54 (2023-06-27, 12:56:48 UTC, ObsID 90901321002) in an intermediate brightness phase when the Swift/BAT rate in the 15-50 keV was around 1 Crab. Although there were 5 NuSTAR observations of the source during the 2017 outburst, they covered the low and high luminosity states, and this is the first time that NuSTAR covered the rising phase of the outburst with a luminosity of 1.6 x 10^38 erg/s assuming a source distance of 6.8 kpc (Bailer-Jones et al. 2018, AJ, 156, 58).

We extracted the NuSTAR light curves and spectra using a source extraction region of ~120 arc-sec around the source centroid, and a similar sized background extraction region far away from the source. The lightcurves were background subtracted and times were corrected to solar system barycenter. From the lightcurves, we find that the hardness ratio between the energy bands of 12-70 keV and 3-12 keV, remained constant throughout the NuSTAR observation. The NuSTAR light curves show pulsations of 9.7991 (+/-0.0001) seconds. This is consistent with the fact that the pulsar is spinning up as luminosity increases, as evident from the long-term Fermi/GBM spin period evolution of the source. The pulse profile exhibits a single peak throughout the NuSTAR energy range with a notch during the falling flank, presumably consistent with the location of a secondary peak before MJD 60117 (Atel #16121). We fit the pulse profiles with two Gaussians and calculate their pulse fraction as (Imax-Imin)/(Imax+Imin), where Imax, Imin are the maximum and minimum intensities, respectively. The pulse fraction increases with energy, and the peak dominating the pulse profile gets narrower with increasing energy. The pulse fraction below 7 keV is ~ 25%, from 7-30 keV is ~35%, and from 20-70 keV is ~52%. No orbital corrections were performed on the lightcurves.

We also looked for QPO-like features using the Stingray software (Huppenkothen et al. 2019 ApJ 881 39). We split the NuSTAR event files into segments of 1024 s with a binning of ~120 microseconds and produced a Fourier Amplitude Difference corrected power density spectrum (PDSs; Bachetti & Huppenkothen 2018, ApJ, 853, 21) for each 1024s segment. The average PDS does not show any significant QPO, however a weak, broad feature is visible between 20-30 mHz. Note that earlier NICER observations during the 2017 outburst displayed a QPO at 50 mHz (Wilson-Hodge et al 2018 ApJ 863 9).

We fit the 3-79 keV spectra with a phenomenological model consisting of an absorbed cutoff power law with a black body component and a Gaussian component for the Fe K\alpha line, i.e., TBabs(cutoffpl + gaussian + bbodyrad) . The hydrogen column density is 1.3 (+/- 0.4)x10^22, the photon index is 1.06 (+/-0.02), the cutoff energy is 22.0 (+/- 0.4) keV, the iron line centroid is 6.42 (+/- 0.03) keV with a width of 0.19 (+/- 0.05) keV and an equivalent width of 0.039 (+/- 0.006) keV. The blackbody temperature is 0.48 (+/- 0.02) keV, and the normalisation is 4265 (+/- 2000). The 3-70 keV flux is 2.8 (+/-0.1) x 10^-8 erg/s/cm^2. Uncertainties are reported at 90% confidence. The spectrum exhibits complex residuals around the Fe K\alpha line and also a hump around 10-15 keV reminiscent of a Compton hump or a relativistic disk reflection component.

Further multiwavelength observations are encouraged.

We thank the NuSTAR PI Fiona Harrison for approving this DDT request and the NuSTAR Operations team for the prompt scheduling of this observation