Search for: All records

Abstract Active galactic nuclei (AGN) light curves observed with different wave bands show that the variability in longer wavelength bands lags the variability in shorter wavelength bands. Measuring these lags, or reverberation mapping, is used to measure the radial temperature profile and extent of AGN disks, typically with a reprocessing model that assumes X-rays are the main driver of the variability in other wavelength bands. To demonstrate how this reprocessing works with realistic accretion disk structures, we use 3D local shearing box multifrequency radiation magnetohydrodynamic simulations to model the UV-emitting region of an AGN disk, which is unstable to the magnetorotational instability and convection. At the same time, we inject hard X-rays (>1 keV) into the simulation box to study the effects of X-ray irradiation on the local properties of the turbulence and the resulting variability of the emitted UV light curve. We find that disk turbulence is sufficient to drive intrinsic variability in emitted UV light curves and that a damped random walk model is a good fit to this UV light curve for timescales >5 days. Meanwhile, X-ray irradiation has negligible impact on the power spectrum of the emitted UV light curve. Furthermore, the injected X-ray and emitted UV light curves are only correlated if there is X-ray variability on timescales >1 day, in which case we find a correlation coefficientr= 0.34. These results suggest that if the opacity for hard X-rays is scattering dominated as in the standard disk model, hard X-rays are not the main driver of reverberation signals. 

Abstract UsingAthena++, we perform 3D radiation-hydrodynamic calculations of the radiative breakout of the shock wave in the outer envelope of a red supergiant (RSG) that has suffered core collapse and will become a Type IIP supernova. The intrinsically 3D structure of the fully convective RSG envelope yields key differences in the brightness and duration of the shock breakout (SBO) from that predicted in a 1D stellar model. First, the lower-density “halo” of material outside of the traditional photosphere in 3D models leads to a shock breakout at lower densities than 1D models. This would prolong the duration of the shock breakout flash at any given location on the surface to ≈1–2 hr. However, we find that the even larger impact is the intrinsically 3D effect associated with large-scale fluctuations in density that cause the shock to break out at different radii at different times. This substantially prolongs the SBO duration to ≈3–6 hr and implies a diversity of radiative temperatures, as different patches across the stellar surface are at different stages of their radiative breakout and cooling at any given time. These predicted durations are in better agreement with existing observations of SBO. The longer durations lower the predicted luminosities by a factor of 3–10 (L_bol∼ 10⁴⁴erg s⁻¹), and we derive the new scalings of brightness and duration with explosion energies and stellar properties. These intrinsically 3D properties eliminate the possibility of using observed rise times to measure the stellar radius via light-travel time effects. 

Abstract We explore the three-dimensional properties of convective, luminous (L≈ 10^4.5–10⁵L_⊙), hydrogen-rich envelopes of red supergiants (RSGs) based on radiation hydrodynamic simulations in spherical geometry usingAthena++. These computations comprise ≈30% of the stellar volume, include gas and radiation pressure, and self-consistently track the gravitational potential for the outer ≈3M_⊙of the simulatedM≈ 15M_⊙stars. This work reveals a radius,R_corr, around which the nature of the convection changes. Forr>R_corr, though still optically thick, diffusion of photons dominates the energy transport. Such a regime is well studied in less luminous stars, but in RSGs, the near- (or above-)Eddington luminosity (due to opacity enhancements at ionization transitions) leads to the unusual outcome of denser regions moving outward rather than inward. This region of the star also has a large amount of turbulent pressure, yielding a density structure much more extended than 1D stellar evolution predicts. This “halo” of material will impact predictions for both shock breakout and early lightcurves of Type IIP supernovae. Inside ofR_corr, we find a nearly flat entropy profile as expected in the efficient regime of mixing-length theory (MLT). Radiation pressure provides ≈1/3 of the support against gravity in this region. Our comparisons to MLT suggest a mixing length ofα= 3–4, consistent with the sizes of convective plumes seen in the simulations. The temporal variability of these 3D models is mostly on the timescale of the convective plume lifetimes (≈300 days), with amplitudes consistent with those observed photometrically.