Probing the Shape of the Glabular Cluster Luminosity Function

J.J. Kavelaars

Department of Physics and Astronomy
McMaster University, Hamilton ON
Electronic-mail: kavelaars@physics.mcmaster.ca

and

Brett Gladman

Canadian Institute for Theoretical Astrophysics
University of Toronto, Toronto ON
Electronic-mail: gladman@cita.utoronto.ca

Abstract:

Understanding the shape of the globular cluster luminosity function (GCLF) and the dependence of that shape on the host galaxy type and the location of the cluster system within the host galaxy, will provide valuable insight into the process of galaxy formation and evolution. Additionally the universality of the GCLF will allow the determination of distances out to 100s of Mpc with the availability of telescopes such as the NGST. Local calibration of the luminosity function continues to be the main weakness of this distance indicator. We present here the results of our analysis of the GCLF of the calibrator galaxy NGC 4697. These data were obtained with the UH8k camera in April 1997 and indicate that, like E galaxies in Fornax, the peak and width of the GCLF are $M_V^{TO} \sim -7.4$ and $\sigma \sim 1.2$.

Introduction

  The early stages of galaxy formation are likely dominated by the conglomeration of proto-galactic fragments (Searle & Zinn, 1978). Globular clusters formed in these fragments and contain information on the initial formation and evolution of proto-galaxies. Studies of globular cluster systems (GCSs) of external galaxies have revealed that the total population of globular clusters which form around a galaxy are a good indicator of the total gas which was present during the cluster formation epoch (Kavelaars 1998, McLaughlin 1998). Additionally, consistent and useful models of globular cluster formation indicate that the shape of the globular cluster luminosity function (GCLF) is in fact a residual of dynamic processes which destroy low mass clusters, thus resulting in a peak in the luminosity function (for example Murali & Weinberg, 1997). If the picture of cluster evolution shaping the luminosity function (as first proposed in Fall & Ress 1977 ) is correct then there should be some variation in the shape of the GCLF when examined at different distances form the center of the host galaxy. Although some evidence of this evolution is present it does not appear to effect the location of the peak in the GCLF (Harris 1991, van den bergh 1992, Kavelaars & Hanes 1997). Thus, although the total population of clusters formed is very much a tracer of the environment of galaxy formation, the GCLF is not as sensitive to these conditions and certainly the peak of the distribution (perhaps determined in some dynamical way) is not indicative of environment in any strong sense.

As globular clusters near the bright end of the luminosity function are indeed quite bright ( V_AB = -9.8 ) they make obvious targets as distance indicators (Hanes 1977). As such there have been numerous articles on the usefulness of these objects in the cosmological distance business (see for example van den Bergh 1991 and Jacoby et al. 1992). As all of these authors have indicated the use of the GCLF to determine distance hinges on accurate and relevant calibration of the zero-point of the GCLF turnover and shape.

Understanding the shape of the GCLF and the dependence of that shape on the host galaxy type and the location of the cluster system within the host galaxy, will provide valuable insights into the process of galaxy formation and evolution. In order to use this power tool we must probe the GCLF of external galaxies to well ( $\sim 1.5$ mag) passed the peak. To be most useful these measurements should not involve the use of data beyond the 100% completeness limit of the data. Additionally, the role of dynamical evolution can be probed by examining the dependence of GCLF shape on distance from the parent galaxies center and so wide field imaging, like that offered by the CFH12k is essential.

Observations

 To explore the use of the GCLF as a tracer of the evolution of shape of the GCLF under the influence of dynamical erosion and to verify its usefulness as a distance indicator we have acquired a long series of short exposures of the E4 galaxy NGC 4697. (480 x 22 exposures). These exposures were obtained in a "piggy back" survey mode in conjunction with a separate science run (see Gladman and Kavelaars elsewhere at this meeting). As a result of these exposures we have precise photometric observation of this Virgo E galaxy which probe the GCLF at a variety of radii ( 1kpc < R_GC < 30 kpc) and to well beyond the peak of the GCLF.

Results

  Our preliminary results indicate that NGC 4697 possesses a GCS with a luminosity function whose Gaussian width, $\sigma = 1.2\pm 0.05$, and peak luminosity, $m^{TO}_{R} = 22.9 \pm 0.1$. Assuming a distance modulus to Virgo of $(m - M) = 31.0\pm 0.2$ (Harris et al. 1998 H98) and <V-R> = 0.6 implies $M^{TO}_V = -7.5 \pm 0.3$,much like that of other elliptical systems (Harris 1991 H91).

 

Figure 1: All detected ``stellar'' objects brighter than the 100% completeness limits (R < 23.8). The inner annulus is the area used to produce the GCLF in Fig 2, the background was computed using objects beyond 20 kpc from NGC 4697 but still on chip-1 of the UH8k array. 

 

Figure 2:   The globular cluster luminosity function for NGC 4697. Panel a) give the luminosity functions for the inner annulus and the background, b) is the inner lf after subtracting the background lf, ie. the GCLF. Panel c) shows the contours in $\chi_\nu ^2$, indicating that the turnover and width of the dispersion are well constrained by our data.