2008A&A...483..437M

The SeyfertI galaxy MCG-6-30-15 shows one of the best examples of a broad ``red wing'' of emission in its X-ray spectrum at energies 2<E<6.4keV, commonly interpreted as being caused by relativistically-blurred reflection close to the event horizon of the black hole. We aim to test an alternative model in which absorption creates the observed spectral shape, explains the puzzling lack of variability of the red wing and reduces the high reflection albedo, substantially greater than unity, that is otherwise inferred at energies E>20keV. We compiled all the available long-exposure, high-quality data for MCG-6-30-15: 522ks of Chandra hetgs, 282ks of XMM-Newton pn/rgs and 253ks of Suzaku xis/pin data. This is the first analysis of this full dataset. We investigated the spectral variability on timescales >20ks using principal components analysis and fitted spectral models to ``flux state'' and mean spectra over the energy range 0.5-45keV (depending on detector). The absorber model was based on the zones previously identified in the high-resolution grating data. Joint fits were carried out to any data that were simultaneous. Multiple absorbing zones covering a wide range of ionisation are required by the grating data, including a highly ionised outflowing zone. A variable partial-covering zone plus absorbed low-ionisation reflection, distant from the source, provides a complete description of the variable X-ray spectrum. A single model fits all the data. We conclude that these zones are responsible for the red wing, its apparent lack of variability, the absorption structure around the FeKα line, the soft-band ``excess'' and the high flux seen in the hard X-ray band. A relativistically-blurred Fe line is not required in this model. We suggest the partial covering zone is a clumpy wind from the accretion disk.