Age Constraints for an M31 Globular Cluster from SEDs Fit

We have constrained the age of the globular cluster S312 in the Andromedagalaxy (M31) by comparing its multicolor photometry with theoretical stellarpopulation synthesis models. This is both a check on the age of this globularcluster, as well a check on our methodology. Main-sequence photometry has beenthe most direct method for determining the age of a star cluster. S312 wasobserved as part of the Beijing-Arizona-Taiwan-Connecticut (BATC) MulticolorSky Survey from 1995 February to 2003 December. The photometry of BATC imagesfor S312 was taken with 9 intermediate-band filters covering 5000-10000\AA.Combined with photometry in the near-ultraviolet (NUV) of {\sl GALEX},broad-band $UBVR$ and infrared $JHK_s$ of 2MASS, we obtained the accuratespectral energy distributions (SEDs) of S312 from 2267-20000\AA. A quantitativecomparison to simple stellar population models yields an age of$9.5_{-0.99}^{+1.15}$ Gyr, which is in very good agreement with the previousdetermination by main-sequence photometry. S312 has a mass of$9.8\pm{1.85}\times 10^5 \rm M_\odot$, and is a medium-mass globular cluster inM31. By analysis of errors of ages determined based on the SED fitting methodof this paper, secure age constraints are derived with errors of $< 3$ Gyr forages younger than 9 Gyr. In fact, the theoretical SEDs are not sensitive to thevariation of age for ages greater than $\sim 10$ Gyr. Therefore, for globularclusters as old as the majority of the Galactic GCs, our method do notdistinguish them accurately. We emphasize that our results show that even withmultiband photometry spanning NUV to $K_s$, our age constraints from SEDfitting are distressingly uncertain, which has implications for age derivationsin extragalactic globular cluster systems.Comment: Accepted for Publication in ApJ, 9 pages, 5 figure