Modern Cosmological Observations and Problems

6.4.3. Cosmological Releance of Low Surface Brightness Galaxies

The discovery of a substantial population of LSB galaxies at z = 0 is relevant to a number of cosmological issues. These issues are enumerated below and are more fully discussed in Impey and Bothun (1997) or Bothun et al. (1997).

1. QSO absorption lines: Increasing the number density of galaxies is a very good first step towards understanding the origin of QSO absorption lines. In addition, the discovery of Malin 1, which has 10⁵ kpc² of gaseous cross section at a level of N_h geq 3 x 10¹⁹ cm^-2 is a helpful addition to the zoology of objects that might produce damped Lyman systems.

2. Large Scale Structure (LSS): To date, most studies of LSS are based on redshift surveys of optically selected samples, which, by definition, do not contain many LSB galaxies. If LSB galaxies, for some reason, are better tracers of the mass distribution, then the biasing factor between mass and light has a surface brightness dependence. This would be an ugly complication to the various sophisticated attempts to determine the parameter b Omega ^.6 (see Strauss and Willick 1995). In an attempt to shed light on this, Bothun et al. (1985,1986) performed a redshift survey of approx 400 LSB disk galaxies, primarily selected from the UGC. Few of these galaxies were in the CFA redshift survey at that time. That data, coupled with a thorough analysis of the clustering properties of LSB galaxies by Bothun et al. 1993 and Mo et al. 1994 provides the following insights:

On scales geq 5 h^-1 Mpc LSB galaxies trace the identical structure as HSB galaxies.

LSB galaxies generally avoid virialized regions.

On scales leq 2 h^-1 Mpc LSB galaxies are significantly less clustered. In fact, there is a significant deficit of companions within a projected distance of 0.5 h^-1 Mpc indicating that LSB disks are generally isolated.

When found in the group or cluster environment, LSBs are usually near the edge of the galaxy distribution. The results of O'Neil et al. (1997a) show that, in the cluster environment, there is a limiting value of µ₀ which depends on local galaxy density. If one considers that the highest surface brightness galaxies of all, ellipticals, tend to occur in virialized regions then we have the surface brightness extension of the morphology-density relation of Dressler (1980). That is, the local galaxy density determines the surface mass/luminosity density of of the morphology-density relation of Dressler (1980). That is, the local galaxy density determines the surface mass/luminosity density of galaxies.

LSB galaxies are not preferentially found in large scale voids. For instance, there are none in either the CFA bubble (Bothun et al. 1992) or the Bootes Void (Aldering et al. 1997).

This results allow the following scenario to be advanced. If the initial spectrum of density perturbations which eventually form galaxies is Gaussian in nature, then there should be many more low density perturbations than high density ones. Many of these low density perturbations are subject to disruption or assimilation into other perturbations and hence will not produce individual galaxies. However, if a substantial percentage of them do survive to produce individual galaxies then we would expect them to dominate the galaxy population. We tentatively identify our discovery of LSB galaxies with these 1-2 sigma peaks in the initial Gaussian perturbation spectrum. If that is the case, we expect LSB galaxies to 1) have formed in isolation and 2) be fair tracers of the mass distribution on large scales. This isolation on small scales must clearly affect formed in isolation and 2) be fair tracers of the mass distribution on large scales. This isolation on small scales must clearly affect their evolution as, compared to HSB galaxies, LSB galaxies have experienced fewer tidal encounters with nearby galaxies over the last Hubble time. Tidal encounters are effective at clumping gas and driving global star formation. Without this external hammer, LSB galaxies would continue to evolve slowly and passively.

3. After correcting for surface brightness selection effects, the faint end slope of the galaxy luminosity function is significantly steeper than is derived when only HSB galaxies are selected. This is best seen in the Virgo and Fornax clusters where the kinds of LSB dwarfs that were missed in earlier surveys but detected using the Malin's method were very diffuse but fairly large. Thus, the integrated luminosity of these galaxies, although they are diffuse, was brighter then many of the Sandage diffuse dwarfs owing to their larger scale length. Traditionally, luminosity functions are corrected for incompleteness on the basis of apparent magnitude only. As discussed above, this is conceptually incorrect for galaxies since they are selected on the basis of surface brightness and not total luminosity. The Virgo / Fornax surveys found galaxies that were up to 3 magnitudes brighter than the nominal magnitude limit of the Sandage survey but which were not included in that survey. Figure 6-16 shows that when properly accounting for these galaxies, there is a significant increase in the faint end slope of the GLF. LSB galaxies dominate numerically in clusters and probably in the field (although this has not yet been established); if M / L increases with decreasing surface brightness, they might form a significant component of baryonic mass in the universe. It is interesting to speculate in the case of cluster LSBs that they might represent galaxy evolution following a phase of intense baryonic blow out at higher redshifts. These relic galaxies then might be the source of the gas seen in clusters as well as the enrichment of that gas. In the case of the Virgo cluster, there are approximately twice as many dwarfs as brighter galaxies.

Figure 6-16: The corrected luminosity function that results from applying surface brightness selection effects as opposed to erroneously applying only corrections for mangitude incompleteness. Data represents the Virgo cluster data of Impey etal (1988).

4. LSB galaxies have many of the same properties - color, luminosity, mean surface brightness, clustering amplitude - as the enormous number of FBGs seen in deep CCD surveys. Indeed, it may well be these LSB galaxies, in their initial phase of active star formation, that are what is being discovered in deep CCD surveys. However, the current space density of known LSBs remains too low to fully explain the FBGs. This might imply that were are only seeing the tip of the iceberg and that a much larger population of faded LSBs lurks below the current sensitivity limits. This possibility remains viable in the face of the rather difficult to understand color distribution of LSBs. Practically all of the LSB galaxies discovered to date are blue to very blue despite the lack of star formation. Furthermore, there is no correlation between surface brightness level and overall color as would be expected in any scenario which suggests that LSB galaxies are the faded remnants of HSB galaxies after star formation has subsided. These blue colors remain difficult to understand and, in many cases, they demand that the galaxy has recently formed. This is consistent with the observed low densities of these objects as the dynamical timescale is an order of magnitude larger than HSB galaxies. Thus we expect most LSB galaxies to have late collapse times and hence delayed formation of their first stars. Perhaps our surveys for LSB galaxies have recovered this population but we are still missing a component that should be present. The still outstanding question is: where is the red LSB galaxy population?

Very recently, a new CCD survey for LSB galaxies (see O'Neil et al. 1996) has finally found some examples of red LSB objects, though they do not dominate the kinds of LSBs found in this new survey. Figure 6-17 shows the data which indicates this new population. On average, the red LSBs are no different in scale length or µ₀ than the blue LSBs. The discovery of this component is heartening in the sense that we know, once the star formation stops due to gas depletion in disk galaxies, they will begin to fade. At at some future point the typical disk galaxy in the Universe will be red and of low µ₀. Since the evolutionary timescale of a disk galaxy is set both by its surface gas density and its environment, it is logical to assume that some galaxies have already evolved to this endpoint. Surveys have now recovered this population.

Figure 6-17: Color-surface brightness relation for the CCD selected sample of LSB galaxies from O'Neil et al. (1997) where genuinely red LSB galaxies have been detected for the first time. Filled objects are the CCD selected LSB galaxies and the crosses are LSBs found in previous, photographically-based surveys.