3. CEPHEID VARIABLE STARS

Recent distance estimates based on Cepheid variable stars in the LMC have made use of the densely sampled light curves available from microlensing surveys like MACHO and OGLE. Theoretical pulsation models reproduce the detailed shapes of period-folded light curves, and best-fit models yield the fundamental parameters of each Cepheid, like luminosity and effective temperature, and hence the distance. The Cepheids used for this analysis are those which display a secondary luminosity maximum each pulsation cycle. These so-called "bump" Cepheids collectively represent the Hertzsprung Progression because the bumps occur at earlier phases in the pulsation cycle as period increases. Bono et al. (2002) modeled the OGLE I-band light curves for 2 bump Cepheids and found µ₀ = 18.48 mag. Keller & Wood (2002) modeled the MACHO V and R light curves for 20 bump Cepheids to derive µ₀ = 18.55±0.02_r. Both studies also found a level of core overshoot in good agreement with the calibration from eclipsing binaries (Ribas et al., these proceedings).

The use of pulsational models of bump Cepheids to derive the LMC distance is a promising method. However, only a small fraction of the available data have been used so far. I have identified 183 bump Cepheids with periods of 5 to 25 days in the MACHO database ¹, and out of these 117 are also found in the OGLE database ², and 89 in the catalog of four-color light curves published by Sebo et al. (2002). Furthermore, 153 of these bump Cepheids are found in the 2MASS database ³, which provides partial light curve coverage in the near-infrared. Models will eventually be constructed for all of the LMC bump Cepheids, and parameter estimation will employ more multiwavelength data.

Table 1 lists the average differences between the (intensity-weighted) mean magnitudes published by OGLE (Udalski et al. 1999) and Sebo et al. (2002) relative to the MACHO data for bump Cepheids. The implication is a V-band offset of -0.032 mag between Sebo et al. (2002) and OGLE; however, Sebo et al. (2002) report a +0.04 mag offset. Hence the different comparisons of the same two datasets yield different results. I note that their average is weighted differently and their comparison is only for Cepheids with periods greater than 10 days. See Bono et al. (2002b) for analysis of LMC and theoretical period-luminosity (PL) calibrations and their metallicity dependence.

The HST key project calculates true distance modulus using µ₀ = 2.48 µ_I -1.48 µ_V -0.2 [Fe/H], where µ_I and µ_V are apparent moduli relative to LMC PL calibrations (Freedman et al. 2001). As a check, consider the following application of this method to the Cepheids in the Galaxy. In this case, however, the Galactic Cepheids are represented by only one star, Cephei. Benedict et al. (2002) recalibrated the Hipparcos parallax of Cephei using HST/FGS3 and found M_V = - 3.47 ± 0.10 and M_I = - 4.14 ± 0.10 mag. Adopting the Sebo et al. (2002) PL calibrations and [Fe/H] = 0.4 dex yields µ₀ = 18.54±0.29 mag. See Alves (2003) for a review of the metallicity slope.

¹ http://wwwmacho.mcmaster.ca/Data/MachoData.html Back.

² http://bulge.princeton.edu/~ogle Back.

³ http://www.ipac.caltech.edu/2mass/ Back.