The Circinus galaxy is one of the nearest obscured AGN, making it an ideal target for detailed study. Combining archival Chandra and XMM-Newton data with new NuSTAR observations, we model the 2-79 keV spectrum to constrain the primary AGN continuum a
nd to derive physical parameters for the obscuring material. Chandras high angular resolution allows a separation of nuclear and off-nuclear galactic emission. In the off-nuclear diffuse emission we find signatures of strong cold reflection, including high equivalent-width neutral Fe lines. This Compton-scattered off-nuclear emission amounts to 18% of the nuclear flux in the Fe line region, but becomes comparable to the nuclear emission above 30 keV. The new analysis no longer supports a prominent transmitted AGN component in the observed band. We find that the nuclear spectrum is consistent with Compton-scattering by an optically-thick torus, where the intrinsic spectrum is a powerlaw of photon index $Gamma = 2.2-2.4$, the torus has an equatorial column density of $N_{rm H} = (6-10)times10^{24}$cm$^{-2}$ and the intrinsic AGN $2-10$ keV luminosity is $(2.3-5.1)times 10^{42}$ erg/s. These values place Circinus along the same relations as unobscured AGN in accretion rate-vs-$Gamma$ and $L_X$-vs-$L_{IR}$ phase space. NuSTARs high sensitivity and low background allow us to study the short time-scale variability of Circinus at X-ray energies above 10 keV for the first time. The lack of detected variability favors a Compton-thick absorber, in line with the the spectral fitting results.
A simple method for calculating a low-resolution power spectrum from data with gaps is described. The method is a modification of the $Delta$-variance method previously described by Stutzki and Ossenkopf. A Mexican Hat filter is used to single out fl
uctuations at a given spatial scale and the variance of the convolved image is calculated. The gaps in the image, defined by the mask, are corrected for by representing the Mexican Hat filter as a difference between two Gaussian filters with slightly different widths, convolving the image and mask with these filters and dividing the results before calculating the final filtered image. This method cleanly compensates for data gaps even if these have complicated shapes and cover a significant fraction of the data. The method was developed to deal with problematic 2D images, where irregular detector edges and masking of contaminating sources compromise the power spectrum estimates, but it can also be straightforwardly applied to 1D timing analysis or 3D data cubes from numerical simulations.
Emission from Active Galactic Nuclei is known to vary strongly over time over a wide energy band, but the origin of the variability and especially of the inter-band correlations is still not well established. Here we present the results of our X-ray
and optical monitoring campaign of the quasar MR2251-178, covering a period of 2.5 years. The X-ray 2-10 keV flux is remarkably well correlated with the optical B, V and R bands, their fluctuations are almost simultaneous with a delay consistent with 0 days and not larger than 4 days in either direction. The amplitude of variations shows an intriguing behaviour: rapid, large amplitude fluctuations over tens of days in the X-rays have only small counterparts in the optical bands, while the long-term trends over hundreds of days are stronger in the B band than in X-rays. We show that simple reprocessing models, where all the optical variability arises from the variable X-ray heating, cannot simultaneously explain the discrepant variability amplitudes on different time-scales and the short delays between X-ray and optical bands. We interpret the variability and correlations, in the optically-thick accretion disc plus corona scenario, as the result of intrinsic accretion rate variations modulating both X-ray and optical emission, together with reprocessing of X-rays by the accretion disc.
We compute Fourier-resolved X-ray spectra of the Seyfert 1 Markarian 766 to study the shape of the variable components contributing to the 0.3-10 keV energy spectrum and their time-scale dependence. The fractional variability spectra peak at 1-3 keV,
as in other Seyfert 1 galaxies, consistent with either a constant contribution from a soft excess component below 1 keV and Compton reflection component above 2 keV, or variable warm absorption enhancing the variability in the 1-3 keV range. The rms spectra, which shows the shape of the variable components only, is well described by a single power law with an absorption feature around 0.7 keV, which gives it an apparent soft excess. This spectral shape can be produced by a power law varying in normalisation, affected by an approximately constant (within each orbit) warm absorber, with parameters similar to those found by Turner et al. for the warm-absorber layer covering all spectral components in their scattering scenario. The total soft excess in the average spectrum can therefore be produced by a combination of constant warm absorption on the power law plus an additional less variable component. On shorter time-scales, the rms spectrum hardens and this evolution is well described by a change in power law slope, while the absorption parameters remain the same. The frequency dependence of the rms spectra can be interpreted as variability arising from propagating fluctuations through an extended emitting region, whose emitted spectrum is a power law that hardens towards the centre. This scenario reduces the short time-scale variability of lower energy bands making the variable spectrum harder on shorter time-scales and at the same time explains the hard lags found in these data by Markowitz et al.
We present a power spectral analysis of a 100 ksec XMM-Newton observation of the narrow line Seyfert 1 galaxy Ark~564. When combined with earlier RXTE and ASCA observations, these data produce a power spectrum covering seven decades of frequency whic
h is well described by a power law with two very clear breaks. This shape is unlike the power spectra of almost all other AGN observed so far, which have only one detected break, and resemble Galactic binary systems in a soft state. The power spectrum can also be well described by the sum of two Lorentzian-shaped components, the one at higher frequencies having a hard spectrum, similar to those seen in Galactic binary systems. Previously we have demonstrated that the lag of the hard band variations relative to the soft band in Ark 564 is dependent on variability time-scale, as seen in Galactic binary sources. Here we show that the time-scale dependence of the lags can be described well using the same two-Lorentzian model which describes the power spectrum, assuming that each Lorentzian component has a distinct time lag. Thus all X-ray timing evidence points strongly to two discrete, localised, regions as the origin of most of the variability. Similar behaviour is seen in Galactic X-ray binary systems in most states other than the soft state, i.e. in the low-hard and intermediate/very high states. Given the very high accretion rate of Ark 564 the closest analogy is with the very high (intermediate) state rather than the low-hard state. We therefore strengthen the comparison between AGN and Galactic binary sources beyond previous studies by extending it to the previously poorly studied very high accretion rate regime.
