Exploring Disk Galaxy Dynamics Using IFU Data

In order to test the basic equations believed to dictate the dynamics of disk galaxies, we present and analyze deep two-dimensional spectral data obtained using the PPAK integral field unit for the early-type spiral systems NGC 2273, NGC 2985, NGC 3898 and NGC 5533. We describe the care needed to obtain and process such data to a point where reliable kinematic measurements can be obtained from these observations, and a new more optimal method for deriving the rotational motion and velocity dispersions in such disk systems. The data from NGC 2273 and NGC 2985 show systematic variations in velocity dispersion with azimuth, as one would expect if the shapes of their velocity ellipsoids are significantly anisotropic, while the hotter disks in NGC 3898 and NGC 5533 appear to have fairly isotropic velocity dispersions. Correcting the rotational motion for asymmetric drift using the derived velocity dispersions reproduces the rotation curves inferred from emission lines reasonably well, implying that this correction is quite robust, and that the use of the asymmetric drift equation is valid. NGC 2985 is sufficiently close to face on for the data, combined with the asymmetric drift equation, to determine all three components of the velocity ellipsoid. The principal axes of this velocity ellipsoid are found to be in the ratio sigma_z:sigma_phi:sigma_R ~ 0.7:0.7:1, which shows unequivocally that this disk distribution function respects a third integral of motion. The ratio is also consistent with the predictions of epicyclic theory, giving some confidence in the application of this approximation to even fairly early-type disk galaxies.

The rotation curves of flattened Sersic bulges

I present a method to deproject the observed intensity profile of an axisymmetric bulge with arbitrary flattening to derive the 3D luminosity density profile and to calculate the contribution of the bulge to the rotation curve. I show the rotation curves for a family of fiducial bulges with Sersic surface brightness profiles and with various concentrations and intrinsic axis ratios. Both parameters have a profound impact on the shape of the rotation curve. In particular, I show how the peak rotation velocity, as well as the radius where it is reached, depend on both parameters. I also discuss the implications of the flattening of a bulge for the decomposition of a rotation curve and use the case of NGC 5533 to show the errors that result from neglecting it. For NGC 5533, neglecting the flattening of the bulge leads to an overestimate of its mass-to-light ratio by approximately 30% and an underestimate of the contributions from the stellar disc and dark matter halo in the regions outside the bulge-dominated area.

The high mass end of the Tully-Fisher relation

We study the location of massive disk galaxies on the Tully-Fisher relation. Using a combination of K-band photometry and high-quality rotation curves, we show that in traditional formulations of the TF relation (using the width of the global HI profile or the maximum rotation velocity), galaxies with rotation velocities larger than 200 km/s lie systematically to the right of the relation defined by less massive systems, causing a characteristic `kink in the relations. Massive, early-type disk galaxies in particular have a large offset, up to 1.5 magnitudes, from the main relation defined by less massive and later-type spirals. The presence of a change in slope at the high-mass end of the Tully-Fisher relation has important consequences for the use of the Tully-Fisher relation as a tool for estimating distances to galaxies or for probing galaxy evolution. In particular, the luminosity evolution of massive galaxies since z = 1 may have been significantly larger than estimated in several recent studies. We also show that many of the galaxies with the largest offsets have declining rotation curves and that the change in slope largely disappears when we use the asymptotic rotation velocity as kinematic parameter. The remaining deviations from linearity can be removed when we simultaneously use the total baryonic mass (stars + gas) instead of the optical or near-infrared luminosity. Our results strengthen the view that the Tully-Fisher relation fundamentally links the mass of dark matter haloes with the total baryonic mass embedded in them.