Spatial decomposition of on-nucleus spectra of quasar host galaxies

Jahnke, K. ; Wisotzki, L. ; Courbin, F. ; Letawe, G.

In: Monthly Notices of the Royal Astronomical Society, 2007, vol. 378, no. 1, p. 23-40

Add to personal list
    Summary
    In order to study the host galaxies of type 1 (broad-line) quasars, we present a semi-analytic modelling method to decompose the on-nucleus spectra of quasars into nuclear and host galaxy channels. The method uses the spatial information contained in long-slit or slitlet spectra. A routine determines the best-fitting combination of the spatial distribution of the point like nucleus and extended host galaxy. This is fully complementary to a numerical spatial deconvolution technique that we applied to the same data in a previous analysis, which allows a cross-calibration of the two methods. Inputs are a simultaneously observed point spread function, and external constraints on galaxy morphology from imaging. We demonstrate the capabilities of the method to two samples of a total of 18 quasars observed with EFOSC at the ESO 3.6-m telescope and FORS1 at the ESO VLT. ∼50 per cent of the host galaxies with successful decomposition show distortions in their rotation curves or peculiar gas velocities above normal maximum velocities for discs. This is consistent with the fraction from optical imaging. All host galaxies have quite young stellar populations, typically 1-2 Gyr. For the disc dominated hosts these are consistent with their inactive counterparts, the luminosity-weighted stellar ages are much younger for the bulge dominated hosts, compared to inactive early-type galaxies. While this presents further evidence for a connection of galaxy interaction and active galactic nucleus (AGN) activity for half of the sample, this is not clear for the other half. These are often undistorted disc dominated host galaxies, and interaction on a smaller level might be detected in deeper high-resolution images or deeper spectroscopic data. The velocity information does not show obvious signs for large scale outflows triggered by AGN feedback - the data are consistent with velocity fields created by galaxy interaction