Thermonuclear X-ray bursts on the surface of neutron stars (NSs) can enrich the photosphere with metals, which may imprint photoionization edges on the burst spectra. We report here the discovery of absorption edges in the spectra of the type I X-ray burst from the NS low-mass X-ray binary GRS 1747-312 in Terzan 6 during observations by the Rossi X-ray Timing Explorer. We find that the edge energy evolves from 9.45 +/- 0.51 to similar to 6 keV and then back to 9.44 +/- 0.40 keV during the photospheric radius expansion phase and remains at 8.06 +/- 0.66 keV in the cooling tail. The photoionization absorption edges of hydrogen-like Ni, Fe, or an Fe/Ni mixture and the bound-bound transitions of metals may be responsible for the observed spectral features. The ratio of the measured absorption edge energy in the cooling tail to the laboratory value of the hydrogen-like Ni(Fe) edge energy allows us to estimate the gravitational redshift factor 1 + z = 1.34 +/- 0.11(1 + z = 1.15 +/- 0.09). The evolution of the spectral parameters during the cooling tail are well described by metal-rich atmosphere models. The combined constraints on the NS mass and radius from the direct cooling method and the tidal deformability strongly suggest very high atmospheric abundance of the iron group elements and limit the distance to the source to 11 +/- 1 kpc.

We use ensemble machine learning algorithms to study the evolution of magnetic fields in magnetohydrodynamic (MHD) turbulence that is helically forced. We perform direct numerical simulations of helically forced turbulence using mean field formalism, with electromotive force (EMF) modeled both as a linear and non-linear function of the mean magnetic field and current density. The form of the EMF is determined using regularized linear regression and random forests. We also compare various analytical models to the data using Bayesian inference with Markov chain Monte Carlo (MCMC) sampling. Our results demonstrate that linear regression is largely successful at predicting the EMF and the use of more sophisticated algorithms (random forests, MCMC) do not lead to significant improvement in the fits. We conclude that the data we are looking at is effectively low dimensional and essentially linear. Finally, to encourage further exploration by the community, we provide all of our simulation data and analysis scripts as open source IPYTHON notebooks.

KTH, Centres, Nordic Institute for Theoretical Physics NORDITA.

Poutanen, Juri

KTH, Centres, Nordic Institute for Theoretical Physics NORDITA. Tuorla Observatory, Department of Physics and Astronomy, 20014 University of Turku, Finland e-mail: thjsal@utu.fi; Space Research Institute of the Russian Academy of Sciences, Profsoyuznaya str. 84/32, 117997 Moscow, Russia .

We present a Bayesian method to constrain the masses and radii of neutron stars (NSs) using the information encoded in the X-ray pulse profiles of accreting millisecond pulsars. We model the shape of the pulses using "oblate Schwarzschild" approximation, which takes into account the deformed shape of the star together with the special and general relativistic corrections to the photon trajectories and angles. The spectrum of the radiation is obtained from an empirical model of Comptonization in a hot slab in which a fraction of seed black-body photons is scattered into a power-law component. By using an affine-invariant Markov chain Monte Carlo ensemble sampling method, we obtain posterior probability distributions for the different model parameters, especially for the mass and the radius. To test the robustness of our method, we first analysed self-generated synthetic data with known model parameters. Similar analysis was then applied for the observations of SAX J1808.4-3658 by the Rossi X-ray Timing Explorer (RXTE). The results show that our method can reproduce the model parameters of the synthetic data, and that accurate constraints for the radius can be obtained using the RXTE pulse profile observations if the mass is a priori known. For a mass in the range 1.5-1.8 M, the radius of the NS in SAX J1808.4-3658 is constrained between 9 and 13 km. If the mass is accurately known, the radius can be determined with an accuracy of 5% (68% credibility). For example, for the mass of 1.7 M the equatorial radius is Req = 11.9+0.5 -0.4 km. Finally, we show that further improvements can be obtained when the X-ray polarization data from the Imaging X-ray Polarimeter Explorer will become available.