Annual report - 2007

Objectives planned for the reporting period

Observation will be analyzed by our group, and in parallel we plan to develop further the XSPEC model KY by enhancing the treatment of reflection continuum. The main objective here is to better constrain the origin of the broad iron line and to decide between the relativistic disc and the Compton broadening hypotheses. A number of ways have been proposed, also by our group, of how to deduce the properties of black holes and measure their parameters. While the first rough estimates of their mass were based on energy considerations and limits implied by the shortest time-scales of variability, more precise and elaborate methods became possible with modern observational techniques. Here we plan assessing the accuracy of black hole spin measurements improving the present estimates.

The first two years of the project duration will be mainly devoted to new observations with XMM-Newton and to data interpretation. We assume that a natural continuation of our present research, collaboration and students training will be the main content of the project. Cooperation with foreign teams will be substantially intensified with the financial support of the proposed PECS project.

Research activities within the scope of the PECS project

1. Physical context of the on-going research

An attractive possibility to explain the X-ray spectrum of active galactic nuclei (AGN) and its variability is given by the magnetic flare model (pioneered by Galeev, Rosner, Vaiana 1979). This model assumes that localized magnetic reconnection events occur above the accretion disk and produce compact sources of hard X-ray radiation. Similar events on a much lower luminosity scale are seen in the sun (e.g. for example Liu et al. 2006). In the so-called loop-top source of a flare event, hard X-ray emission revealing a power-law spectrum is produced. A fraction of this primary radiation shines down on the disk and is reprocessed. It is thus necessary to consider two different constituents of the composite flare spectrum:

The primary emission from the loop-top source has been extensively modeled in the past (see for instance Sunyaev & Titarchuk 1980). It is explained by inverse Compton scattering of soft X-ray photons coming from the disk in the hot plasma of the reconnection site. From the observed power-law shape of the primary spectrum it is not possible to conclude about the energy distribution of the electrons, both thermal and power-law distributions are discussed in the literature (see e.g. Zdziarski et al. 2001). No coherent model of magnetic reconnection does exist to the present and the injection mechanism for the high energy electrons is still unclear.

As for the reprocessed component we contributed to the most up-to-date modeling attempts. They include the computation of the vertical structure of an accretion disk in hydrostatic equilibrium, multi-angle radiative transfer within the disk, consideration of geometrical time delays, and relativistic ray-tracing between the emitting accretion disk and a distant observer.

2. Modeling the X-ray reprocessing for individual X-ray flare

We add significant sophistication to previous modeling of the X-ray reprocessing due to magnetic flares above AGN accretion disks. We investigate in more detail the effects of a compact non-thermal source of radiation situated at a given height above the disk. The source creates an irradiated hot-spot on the disk in which the reprocessed component is re-emitted. The incident angle and the flux arriving at the surface depend on the position within the hot spot producing different ionization states and thus different re-emitted spectra. Besides, the disk and therefore the hot-spot can be evaluated at different viewing angles. Our computations of the reprocessed spectrum re-emitted by the spot take these geometrical details into account.

In order to construct the observed spectrum for a given inclination of the system we apply the code KY (Dočiak, Karas, & Yaqoob 2004; Dovčiak 2004) with the model spectra for the spot re-emission. For a given viewing direction KY integrates the photon trajectories coming from the disk and arriving at a distant observer. It considers the time delay effects between separate trajectories, gravitational shift of photon energy, and modifications of the flux due to relativistic lensing. In Goosmann et al. (2007a) we report the theoretical spectral time-evolution for a flare spot around a Schwarzschild black hole at two different distances from the disk center. We provide simulated 'observations' for XMM-Newton and for the planned XEUS mission for such a setup. For the time being, XMM-Newton with its large effective collecting area is the most powerful X-ray observatory to investigate the reprocessed iron line complex in AGN. Nevertheless, the detailed irradiation pattern across AGN accretion disks cannot yet be determined even for the most X-ray-luminous objects. A limiting case is the XMM-Newton data of NGC 3516 analyzed by Iwasawa, Miniutti, & Fabian (2004). Much better observational constraints for the appearance of individual magnetic flares in AGN will be obtained from the planned XEUS mission.

For an observed strong X-ray flare in the Seyfert-1 galaxy MCG-6-30-15 we suggest a simple toy model to reproduce time-delays measured between the light-curves over different spectral bands (Goosmann et al. 2007b). This analytical model is based on the idea that a distant observer detects the primary and the reprocessed radiation of the flare as two consecutive pulses. We model the primary and the reprocessed spectrum in a simplified way assuming a power-law shape for both and different spectral slopes. The time-like behavior of both components is realized by Lorentzian functions that are shifted one from the other by an intrinsic delay. This intrinsic delay gives a rough measure of the distance between the compact primary source and the reprocessing hot-spot. The toy model reproduces correctly the time delays obtained by Ponti et al. (2004) from cross-correlation analysis of the observed MCG-6-30-15 light-curve during the flare. In the future, we intend to extend this modeling by replacing the rather simplified function for the spectrum of the reprocessed component by a detailed flare model as described in (1). The time-lags can then be computed more realistically applying the KY code. A further development might include the reprocessed component due to relativistic particles traveling from the reconnection site along the flux tubes and arriving at the disk. This component is mostly neglected in the flare modeling of AGN accretion disks.

3. Modeling X-ray variability of AGN

The detailed flare modeling can be applied to the analysis of variability data that is currently available. Therefore we implemented a Monte-Carlo code to sample random distributions of flares across an accretion disk. The radial distribution of flares and their luminosities can be varied by model parameters. The flares co-orbit with the accretion disk and have a limited lifetime. They continuously appear and disappear across the disk and their total number fluctuates around the average flare number defined. From such orbiting flare distributions we compute the time evolution of the spectra seen at large distances, thereby including corrections for the Doppler effects and for general relativistic effects in the vicinity of the black hole. Then, variability and time-averaged spectra can be computed and compared to the observations.

We obtain the `rms' variability, power density, and time-averaged spectra. The model setup can be described by the following global parameters: mass, spin, and accretion rate of the black hole, the fraction of energy dissipated in the disk corona, the illumination structure across the disk. Constraining these main parameters of accreting black holes remains a crucial subject for the X-ray community. In Goosmann et al. (2006) we are able to reproduce the relative lack of spectral variability across the broad iron K-alpha line that is observed by XMM-Newton in MCG-6-30-15 (see the analysis by Ponti et al. 2004) and in other Seyfert galaxies.