MULTIBAND PHOTOMETRY OF NGC 1399 GLOBULAR CLUSTER SYSTEM

- A CRUCIAL TEST FOR THE ORIGIN OF COLOR BIMODALITYHak-Sub Kim1, Sangmo Tony Sohn1,2, Chul Chung1 , Sang-Yoon Lee1, and Suk-Jin Yoon1

1Dept of Astronomy & Center for Space Astrophysics, Yonsei University, Seoul 120-749, Korea E-mail: agapiel96@galaxy.yonsei.ac.kr2California Institute of Technology, MC 405-47 1200E California Blvd. Pasadena, CA 91125 U.S.A.

2. Observation

Instrument - CTIO 4-m Blanco Telescope - Prime Focus 8K Mosaic CCD Imager

Target - NGC 1399 GC system

Observing run - 2006 Nov 24, 28

Filter system & Exposure time - U : 1800s X 15 - B : 900s X 6 - V : 240s X 3 + 360s X 3 - I : 300s X 6

ABSTRACTThe bimodal color distribution is one of the most conspicuous features of extragalactic globular cluster (GC) systems. It has been commonly accepted that the color bimodality comes from two GC sub-populations with different mean metallicities within individual galaxies. However, a new explanation is proposed recently, in which the non-linear color-metallicity relations can produce the color bimodality even from the unimodal metallicity distribution. In this study, we use multi-band (U, B, V and I) photometry of NGC 1399 GCs obtained with the CTIO 4-m telescope to test the hypothesis that color bimodality arises from a metallicity-to-color "projection effect". After carefully selecting GC candidates, we show that different sets of colors (e.g., B-I versus U-B) for the same GC sample exhibit considerably different blue-to-red GC number ratios. We find that these observed results are well reproduced by our Monte Carlo simulations when a unimodal (rather than a bimodal) metallicity distribution and the new theoretical color-metallicity relations are used. We discuss the advantage of using U-band both in selecting GC candidates and in placing constraints on the physical origin of GC color bimodality.

3. Data Reductions

Bias subtraction and flat-fielding were done through the IRAF mscred package. After the preprocessing, we performed standard psf photometry using DAOPHOT II/ALLFRAME (Stetson 1987, 1994). Reddening corrections were applied using the reddening maps of Schlegel et al. (1998).

4. GC Candidate Selection

REFERENCESBrodie, J. P., & Strader, J. 2006, ARA&A, 44, 193Dirsch, B. et al. 2004, AJ, 127, 2114Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525

1. Introduction

Fig.4. GC selection areas in U-B vs. B-I diagram (upper) and B-V vs. V-I (lower) depicted by green boxes are empirically determined based on (1) spectroscopically-confirmed GCs (red dots; Dirsch et al. 2004), and (2) size information derived from HST ACS data (yellow dots). Blue loci are the Yonsei Evolutionary Population Synthesis (YEPS) model. Note that, compared to the lower panel, background galaxy contamination is reduced by a factor of three in the upper panel (U-B color).

Fig.2. Four-color composite image of the central part of NGC1399

(1) Chi, Sharpness & Error cut (1) Variation of blue to red GC number ratio

5. Results

Fig.7. (a) Blue lines represent theoretical color-metallicity relations from YEPS for 13-Gyr model GCs and red line is assumed unimodal metallicity distribution (centered at [Fe/H] = -1.0) that is used for Monte Carlo simulations. (b) Color-magnitude diagrams of NGC 1399 GC candidates. (c) Observed color histograms for U-B and B-I. Considerable variation of blue-to-red GC number ratio is evident. (d) Theoretical color histograms for U-B and B-I match well with the observations shown in (c).

Fig.1. Two different interpretations for color bimodality. The conventional explanation is to interpret the bimodal color distribution as the existence of two or more sub-populations with different mean meatallicities within individual galaxies (Brodie & Strader 2006 for a review). The new explanation offered by Yoon et al. (2006) is that bimodal color distribution can be produced even from unimodal metallicity distribution due to the non-linear color-metallicity relation. This figure is from Ken Freeman’s Science commentary (2006) on Yoon et al. (2006).

Fig.3. Chi & Sharpness cuts varying with magnitude are given by red curves. Objects of which photometric error is > 0.15 mag were also excluded.

(3) Magnitude cut

(4) Masking around neighboring galaxies

(2) Selection from color-color diagram

Fig.5. Objects of which U-band magnitude is < 21.5 were excluded in order to avoid stellar contamination.

Fig.6. Objects within radius of 1.5’ from the NGC 1399 center where galaxy light dominates are excluded. GC candidates close (< 4’) to NGC 1404 and NGC 1387 are also rejected to avoid the contamination from GCs around these neighboring galaxies.

Stetson, P. B. 1987, PASP, 99, 191Stetson, P. B. 1994, PASP, 106, 250Yoon, S.-J., Yi, S. K., & Lee, Y. -W. 2006, Science, 311, 1129