A Long-Term Study of Broad Emission-Line Profile Variability
in NGC 5548
I. Wanders and B.M. Peterson: 1996, ApJ 466, 174
In this paper we examine five years worth of spectral monitoring data on the Seyfert-1 galaxy NGC 5548. The variable H-beta emission-line profile is studied in detail. Cross-correlation analysis of a number of time series constructed from the data indicate that, while the emission-line flux variations follow the continuum variations closely, emission-line profile variations are not due to the variations in the irradiating ionizing continuum flux. Rather, profile variations appear to be independent of the continuum source, and therefore do not trace reverberation effects through the broad-line region (BLR), but are due to a different physical process, probably dynamical changes. This result considerably complicates the use of the transfer function (TF) as a characteristic of the BLR, because the TF is no longer a stationary entity on the longer time scales. Also, cross-correlation results like "the lag" (emission-line response time scale) become difficult to interpret: on account of changes in the geometry and kinematics as indicated by profile variability, cross-correlation analysis of the light curves yields different solutions for different monitoring campaigns on the same object. The emission-line profile changes occur on a time scale that is much longer than the flux-variability time scale. Therefore, averaging techniques can be used to quantify the long-term profile changes. On the shorter time scales the TF is approximately stationary and an attempt to derive the velocity-dependent TF is made using the SOLA method. Contrary to previous results using the same data, though calibrated differently, it is no longer found that the response of the TF is close to zero at zero lag. It is argued that this may have been an artifact in previous data sets, due to the less precise calibration process employed previously. The reconstructed TFs confirm previous claims that the dominant motions in the BLR are not radial. Details in the TFs are not resolved numerically and daily sampling will be necessary for future monitoring campaings to resolve the TF structure on short time scales. The long-term profile changes indicate the BLR is clumpy and only relatively few clumps of clouds or filaments exist. There are three identifiable components in the line profile that vary independently of one another, a relatively narrow central component and somewhat broader red- and blue-wing shoulders at approximately $\pm 2500$ km/s from the broad-line center. These three components change their relative intensity over the years, but not their positions in the line profile. The observed emission-line profile and TF characteristics are consistent with the simple model in which the continuum source is irradiating the BLR anisotropically and the BLR material is moving along randomly inclined Keplerian orbits. Profile changes then occur as different clumps of matter pass through the continuum beams.