2002ApJ...580L.117M

We discuss a 1.5 day long observation of the bright variable Seyfert 1 galaxy NGC 4051 that was obtained with the X-ray and ultraviolet telescopes on XMM-Newton and extended in the X-ray band using the Rossi X-Ray Timing Explorer. The ultraviolet (2900 Å) and X-ray (2-10 keV) light curves varied with peak-to-peak amplitudes of 5% and a factor of ∼4, respectively. A simple interband cross-correlation function shows three broad features: a positive correlation with coefficient r_M=+0.63 at a lag (T_opt-T_X–ray) τ~+0.2 days, an anticorrelation with r_M=-0.65 at τ~+0.7 days, and a broad but slightly weaker feature with r_M≥0.53 over τ=1.2-1.4 days. Because of the asymmetric sampling of the optical and X-ray data, only positive lags could be probed. Monte Carlo simulations based on synthetic uncorrelated light curves derived from the power spectrum of the real data give an ∼85% confidence in features at this level. If the X-ray and UV light curves are indeed correlated, the lag is highly significant (i.e., not consistent with zero), which is suggestive of a scenario in which X-rays are being reprocessed into UV light at a distance from the X-ray source. A simple model in which the reprocessor is an inclined optically thick ring at a radius of 0.14 lt-days from the X-ray source reproduces the UV data well. A comparison of the modulated fractions in the ultraviolet and X-ray bands suggests that 5% of the 2900 Å flux is produced in this way. The reprocessing site is close enough to the X-ray emitter for X-ray heating from a source with the luminosity of NGC 4051 to produce a brightening in the UV band, and the amount of UV flux emitted suggests that the covering fraction of the reprocessor is about 6%. We briefly discuss the possible physical nature of the reprocessing ring structure.