X-ray luminosity and absorption column fluctuations in the H2O maser galaxy NGC 4258 from weeks to years

We report monitoring of the 0.3 - 10 keV spectrum of NGC 4258 with the XMM-Newton Observatory at five epochs over 1.5 yr. We also report reprocessing of an overlapping four-epoch series of archival Chandra observations (0.5 - 10 keV). By including earlier ASCA and BeppoSAX observations, we present a new, 9 year time series of models fitted to the X-ray spectrum of NGC 4258. We model the Chandra and XMM-Newton data self-consistently with partially absorbed, hard power-law, soft thermal plasma, and soft power-law components. Over the 9 years, the photoelectric absorbing column (similar to 10(23) cm(-2)) did not vary detectably, except for a similar to 40% drop between two ASCA epochs separated by 3 years ( in 1993 and 1996) and a similar to 60% rise between two XMM-Newton epochs separated by just 5 months ( in 2001 and 2002). In contrast, factor of 2 - 3 changes are seen in absorbed flux on the timescale of years. These are uncorrelated with changes in absorbing column and indicative of central engine variability. The most rapid change in luminosity ( 5 - 10 keV) that we detect ( with XMM-Newton and Chandra) is similar to 30% over 19 days. The warped disk, a known source of H2O maser emission in NGC 4258, is believed to cross the line of sight to the central engine. We propose that the variations in absorbing column arise from inhomogeneities sweeping across the line of sight in the rotating disk at the radius at which the disk crosses the line of sight. We estimate that the inhomogeneities are similar to 10(15) cm in size at the crossing radius of 0.29 pc, slightly smaller than the expected scale height of the disk. This result thus provides strong evidence that the warped accretion disk is the absorber in this ( and possibly other) active galactic nuclei (AGNs). This is the first direct confirmation that obscuration in type 2 AGNs may, in some cases, arise in thin, warped accretion disks rather than in geometrically thick tori. Some previous studies report detection of weak Fe K alpha emission in NGC 4258. We do not detect this line emission in any of our XMM-Newton spectra with a 90% upper limit to the equivalent width of similar to 49 eV for one observation. Weak, time-variable Fe line absorption has also been reported for a previous Chandra study. We do not observe evidence of absorption lines in the XMM-Newton or reprocessed Chandra data. The absence of Fe line emission is consistent with the disk being optically thin to hard photons, as well as subtending a small solid angle as seen from the central engine because of the known shallowness of the warp.