Aperture Effects and Limitations on the Accuracy of Ground-Based Spectrophotometry of Active Galactic Nuclei

We use narrow-band images of the [O III] λ 5007 emission line and adjacent continuum in the well-studied Seyfert galaxies NGC 4151 and NGC 5548 to produce models of the surface-brightness distributions of the narrow-line region and host-galaxy starlight distribution in these galaxies. We use these models to compute the expected magnitude of seeing-induced aperture effects that can lead to systematic errors in broad emission-line and continuum flux measurements made from spectra which are flux calibrated based on the narrow emission lines. We find that small spectrograph apertures (e. g., 2" X 10") are highly undesirable as photometric errors as large as 10%-20% can result. Moreover, photometric corrections based on image modeling are not likely to offer great improvement since the corrections are a sensitive function of the seeing. Use of larger apertures (e. g., 5" XT".5) can reduce photometric errors to the few percent range, and in principle seeing corrections based on models of the surface-brightness distributions of the narrow-line region and starlight from the host galaxy can reduce the errors to about the 1% level, at which point miscentering errors probably dominate. The limitation in application of seeing corrections is probably our inability to characterize with sufficient accuracy the point-spread function for the spectroscopic observations, although uncertainties in the surface-brightness models are also likely to play a role.