NMSUAstronomy

Skip to: [content ] [navigation]

Nicole Vogt » The Tully-Fisher Relation at High Redshifts


We have observed a sample of more than 100 spiral galaxies in the distant Universe (sampling uniformly throughout 0.2 < z < 1.1), using Keck+LRIS multi-slit mask spectroscopy to obtain individual, spatially resolved velocity profiles (rotation curves) of optical emission lines for each galaxy. These data provide a unique, mass-driven view of high redshift spirals, as only by determining circular velocities can we distinguish between brightening due to instantaneous bursts of star formation, and more gradual changes in disk size, luminosity and mass due to growth in the underlying stellar populations.

The Tully-Fisher relation, shown below, is a two dimensional projection of fundamental plane space, showing the linear relationship between circular velocity and luminosity (absolute magnitude in B restframe, in this case) for spiral galaxies. We present the data binned by redshift, with ~ 20 galaxies in each of the first five bins. The solid points represent galaxies with the highest quality data. The local Tully-Fisher relationship and its three sigma limits are drawn in yellow in each of the six bins. The thin blue line shows the offset to the distant galaxies, showing no significant offsets in the first five bins. We find that a significant fraction (85%) of the galaxies have luminosities, sizes, morphologies and masses analogous to those found in local spirals today.

A small fraction of the galaxies are found to lie below the yellow lines by redshift z = 0.5, offset from the local relation by three magnitudes but still below M_B of -20. As we move towards redshift z = 1 we are less successful at detecting fainter galaxies (below M_B of -20), and so we become less sensitive to the region in which these outliers might be found. The selection criteria of the observations thus become a critical factor in understanding the distribution of the total ensemble of galaxies at high redshift. The outlying galaxies can be characterized (see next figure) as compact, high surface-brightness galaxies undergoing bursts of star formation (high luminosity), or as morphologically disturbed systems with only a superficial similarity to edge-on disks (low velocity).

The final redshift bin is populated by only two galaxies (open points) which were detected from optical spectra of the [OII]3727 emission line. The remaining three galaxies (solid points) were observed with the VLT+ISAAC infrared spectrograph. They were selected by using photometric redshifts determined from the broadband optical and IR photometry of the HDF-S field to target galaxies for which the Ha line falls within the H band. As this was a pilot program these galaxies were selected for their spiral appearance and their blue colors, which suggested a high rate of star formation, to make the Ha detection as probably as possible. We would expect a sample targeted this way to lie off of the Tully-Fisher relationship, in the same region in which local blue, star-forming galaxies fall today.

In the local Universe, the circular velocity of a spiral galaxy is typically measured by examining the velocity distribution of HI gas. Because the bulk of the HI gas is found at large radii, beyond the optical radius, a fit to the leading and trailing edges of the bihorn line profile (flux versus frequency, or velocity) is a good estimate of the terminal behavior of the velocity profile of the disk. Alternatively, one can observe the optical Ha emission line by placing a narrow slit along the major axis of the galaxy. The spatially resolved velocity profile is then fit to determine the circular velocity by one of a number of techniques, evaluating the velocity at a set number of disk scale lengths along the disk, measuring the maximum velocity of the profile, or fitting the profile at large radii to determine its characteristic terminal behavior.

The current limitations on measuring circular velocities with HI gas make it feasible only out to redshift z ~ 0.1. At larger redshifts sensitivity becomes a problem, and it is difficult to isolate the signal of the galaxy from the background thermal noise. In addition, the rest-frame 21 cm. line of the HI gas shifts into an unprotected region of the radio spectrum, becoming one with a myriad number of garage door openers and TV transmitters. The Ha emission line also shifts slowly into a region of the optical spectrum which is dominated by strong [OH] emission from the atmosphere. However, the lower wavelength emission lines of [OIII]5007, Hb and [OII]3727 remain accessible in the optical windows up to redshifts z ~ 1. At higher redshifts it becomes quite challenging to isolate the [OII]3727 doublet from the [OH] emission lines, and fringing in the red sensitive CCDs becomes a real problem in the reduction process. The Ha emission line has then begun to shift into the infrared J, H and K windows (centered about redshifted Ha at redshift z = 0.8, 1.1 and 2.3), and the latest generation of IR spectrographs is beginning to be used to make circular velocity measurements there.