PI Name: Eline Tolstoy

Title: A New Class of Dwarf Galaxy?

Abstract: The primary aim of this proposal is to determine accurate distances to two Extreme Gas-Rich Dwarfs selected by Blind 21cm Line surveys. These two galaxies are representative of a galaxy class with an apparent large HI gas mass fraction, which may be vastly under-represented in conventional catalogs. The contribution of these galaxies to the mass content of the Universe, and hence their cosmological significance, rests upon the accurate calculation of their number density. The determination of this is severly complicated by the fact that they can only be detected at nearby distances, where peculiar velocities with respect to the Hubble flow make it difficult to obtain precise distance determinations from a redshift; this uncertainty propagates into the computation of volume density. Thus it is crucial to find an independent distance estimator for at least a sub-sample of these galaxies. In this case their proximity helps us, because using a large aperture telescope like Gemini means that we can resolve the stellar populations in these systems and determine their distances using the tip of the red giant branch in a Colour-Magnitude diagram. We can then also use these Colour-Magnitude diagrams to study the recent star-formation properties and the reddening towards these galaxies, and see how this relates to what we know of other galaxy types. It is crucial to our understanding of the distribution of mass in the Universe that we verify whether or not we are looking at a new class of galaxy, and if they could contribute significantly to the mass budget of the Universe.

Proposal Contents

Principal Investigator

PI Institution

Co-Investigators

Science Justification

Here we concentrate on the most extreme cases of the ultra-LSB galaxies that are occasionally detected by blind HI 21cm line surveys that have been conducted at several radio telescopes (Green Bank, Arecibo, Nancay and Parkes) over the past decade (e.g., Zwaan et al. 1997 ApJ, 490, 173; Spitzak & Schneider 1998 ApJS, 119, 159; Staveley-Smith et al. 1998 AJ, 116, 2717). Some of these have HI masses that, although intrinsically small, appear to be greater than the luminous mass in stars by factor of 5 to 10. The stellar population of these extreme objects is unknown, and finding out more details will potentially have an important impact upon our understanding of galaxy evolution. Presumably star formation in such a gas rich environment has similarities to what must have occured in the distant past, at high redshift.

A good illustration of the most extreme objects and the difficulty of determining their space density is the Arecibo survey analyzed by Schneider et al. (1998 ApJL, 507, L9). This analysis has led to the construction of an HI luminosity function that has a diverging tail (with a slope of the Schecter function of -3.5, see Figure 1). This faint end is based solely on the detection of 3 dwarfs: SS47, SS73 and SS75. One of these, SS75, is within ~15'' of a 10th magnitude star, and no optical counterpart can be detected. The other two contain apparently HI masses of ~10^7Msun and have absolute blue magnitudes M(B)~ -11.5. These values are based on distances computed using measured redshift velocities that have been corrected for bulk deviations from a pure Hubble flow using POTENT methods, followed by adoption of Hubble constant of 70km/s/Mpc. These distances are 9.3 Mpc for SS47 and 8.3 Mpc for SS73. The distances inferred from the ``HI Tully-Fisher'' arguments lead to a shallower faint end slope indicated by the open circle points in Figure 1, providing strong incentive to find a way to derive distances from an independent indicator. This will be necessary in order to settle the question of the existence of this diverging tail population.

Figure 2 presents one of the tantalizing characteristics that must hold for these 3 tiny galaxies if their distances are correct - specifically, they rotate faster than the other objects that have been selected in a similar way. Does this mean that they are much more Dark Matter dominated than other galaxies? This would be an important discovery affecting our understanding of galaxy structure and formation. Or have the distances been underestimated, so that these objects actually ought to fall into the higher mass bins? The diagonal line in Figure 2 forms a boundary to the bulk of the points, and the line can be interpreted as the profile width that these galaxies would have if they were corrected to edge-on. The line is a sort of HI Tully-Fisher Relation and provides a conservative measure of possible distance corrections that might be required.

One plausible scenario is that these galaxies are of a new class charaterised by a particularly deep Dark Matter potential, and a strict interpretation of their number density in Figure 1 suggests that they could be very numerous, hither too largely unnoticed, population easily capable of dominating the mass content of the Universe. Thus it is extremely important to verify if these galaxies are really in the correct position in Figure 1. This can be done only by finding an independent measure of their distances upon which the above supositions rest.

The alternative interpretation is of course that the distances implied by their redshifts (corrected as best we can for bulk motions relative to a pure Hubble flow using POTENT) are wrong. To test how big an effect is implied, we have shifted SS47, SS73 and SS75 to the right by the minimum amount required to reach the diagonal line, computed the distance that is implied by the observed flux and the inferred HI mass, and then recomputed the HI mass function as discussed in Figure 1. These distance adjustments would move the galaxies from 8.3 to 11 Mpc for SS47, 9.3 to 11 Mpc for SS73, and 7.5 to 14 Mpc for SS75.

We propose to resolve the upper mass end of the stellar population in two of these galaxies, and by detecting the tip of the red giant branch (RGB) detemine an independent distance measurement. If stars cannot be resolved, or we do not reach the clearly distinguishable RGB tip, then we will have determined an accurate lower limit to the distances, which will be sufficient information to remove the uncertainty in the number densities that are computed based on this sample. With these data we will be able to use a number of different methods to obtain a distance to each of these galaxies if the distances are below ~16Mpc (which any reasonable estimate predicts them to be). The most reliable method will use the tip of the RGB (e.g., Lee, Freedman & Madore 1993 ApJ, 417, 553). We can also consider surface brightness fluctuations, and look at the magnitudes of the brightest stars we detect. All of these methods have uncertainties, especially at the probable distance of these galaxies, but attempting these observations will at least place stringent lower limits on the distance to these two systems, and provide significant information on the nature of the stellar populations in these galaxies.

From these deep images we will also get detailed morphological information about these two peculiar galaxies. From the morphology and the colour of the brightest RGB tip stars and the more massive younger stars we will be able to assess the the closest counterpart of the galaxy type in our Local Group (if it exists). This will include rough, upper limit estimates of the metallicity from the RGB colour. These observations will give us the information we need to decide the important question of whether these galaxies represent a new and significant population of galaxies coming in at the limits of our detection in HI surveys.

Keywords

• Cosmological distance scale
• Dark matter
• Dwarf galaxies
• Galaxy morphology
• Low surface brightness galaxies
• Stellar populations in external galaxies

Resources

• Instrument
  • GMOS
    • Filter
      • g' (500nm)
      • r' (660nm)

Scheduling Information

• 2001/11/1-2002/1/31

Requirements

Observations

Observing Conditions

Proposal Support

• Tolstoy et al. 1998 AJ, 116, 1244
• Tolstoy & Irwin 2000 MNRAS, 318, 1241
• Zwaan, Briggs, Sprayberry & Sorar 1997 ApJ, 490, 173
• Zwaan & Briggs 2000 ApJL, 530, 61

Technical Justification

This means that effectively we need to get images down to r=27 (and g=28) with a S/N > 5, and in good seeing. Assuming the best seeing conditions, and a dark sky, the GMOS ETC would suggest 3x3600sec exposure in r and 4x3600sec in g to reach these magnitude limits. Assuming an hour of overheads due to acquistion and readout, this means we need to request 8 hours of telescope per galaxy.