2010MNRAS.403..196M

Long-exposure Suzaku X-ray observations of the nearby active galaxy NGC 4051 from 2005 and 2008 are analysed, in an attempt to reach a self-consistent understanding of both the spectral variability on long time-scales and the broad-band variability at high time resolution. The techniques of principal components analysis and a maximum likelihood method of power spectrum analysis are used. In common with other type I active galactic nuclei (AGN), the spectral variability is dominated by a varying-normalization power-law component together with a quasi-steady, hard-spectrum offset component that contains Fe K atomic features. NGC 4051 displays a strong excess over a power law at energies of above 20keV, some fraction of which also appears to vary with the power-law continuum. The fast time-scale power spectrum has a shape generally consistent with previous determinations, with the previously known dependence on broad-band photon energy, but in the new data significant differences are found between the low and high flux states of the source, demonstrating that the power spectrum is non-stationary. Frequency-dependent time lags between the hard and soft bands of up to 970±225s are measured. The existence of the observed time lags excludes the possibility that the hard spectral component originates as reflection from the inner accretion disc. We instead show that the time lags and their frequency and energy dependence may be explained simply by the effects of reverberation in the hard band, caused by reflection from a thick shell of material with maximum lags of about 10000s. If the reflecting material surrounds the AGN, it extends to a distance of about 1.5x10¹⁴cm, 600 gravitational radii, from the illuminating source and the global covering factor is C_g≳ 0.44, confirming previous suggestions that type I AGN have high covering factors of absorbing and reflecting material. Given the spectral and timing similarities with other type I AGN, we infer that this source structure is common in the type I population.