9. Local Group galaxies

We now quickly review the status of the individual galaxies in the Local Group for which (non-photographic) digital data on known Cepheids have been obtained in the last few years. This discussion and that in the next section update and supersede similar overviews published earlier by Madore (1985) and Walker (1987).

For the LMC, we adopt hereafter 18.5 mag for its true modulus and 0.10 mag for E (B-V), and scale all other Cepheid-based distances assuming a value for the total-to-selective absorption of R_V = A_V / E (B-V) = 3.3 (appropriate for the later spectral types of Cepheids). In a later section we discuss the impact of recent parallax observations of Galactic Cepheids by the Hipparcos satellite, and conclude that these calibrations are consistent at the 10% level. All period-luminosity fits are done over the range 0.2 < log P < 1.8. Furthermore, as discussed in Freedman (1988b), all fits are carried out for a self-consistent set of stars in the LMC defined by the simultaneous availability of photometry at B, V and I wavelengths. Although the sample of Cepheids with B and V photometry alone is about a factor of two larger, the comparison of inconsistent data sets can lead to erroneous results. It should be noted that although we quote the sources of original data throughout the next two sections, the distances and reddenings given for each galaxy here are based on a new and homogeneous application of the multiwavelength fitting procedures discussed earlier in this review (and described in detail in Freedman et al. 1991). In this procedure, we have adopted the reddening law as determined by Cardelli et al. (1989). By definition, our multiwavelength fitting procedure does not yield information about the distance and/or reddening to the LMC, as they are adopted ab initio. For completeness, we also review data for Cepheids in the Magellanic Clouds.

For a relatively complete bibliography of modern photographic and photoelectric observations of Magellanic Cloud Cepheids, the reader is referred to Tables 2 and 4 in Madore (1985); they will not be repeated here. Moreover, Caldwell & Laney (1991) have reviewed progress in the southern hemisphere on calibrating the Cepheid PL relations. They use data available in Madore (1985), in addition to more recently published near-infrared JHK data available for many dozens of Cepheids observed by Welch et al. (1987) and also Laney & Stobie (1986a, b). Similar to Welch & Madore (1984), these latter papers concentrate, among other things, on using the near-infrared data to determine the back-to-front geometry of the two Magellanic Clouds based on the ability of near-infrared observations of Cepheids to give extremely precise distances to individual stars. In this regard, it should be noted that Visvanathan (1989) has calibrated the Cepheid PL relation at 1.05 µm and applied it to Cepheids in the SMC (Mathewson et al. 1986, 1988), also aiming to probe the structure of the SMC in the context of a tidal encounter/disruption model.

Sandage (1971) published photographic light curves and periods for 25 of the confirmed Cepheids discovered, but never published, by Baade. Both single-phase, near-infrared H-band observations of 10 of those Cepheids (McAlary, Madore & Davis 1984) and single-phase, multiwavelength BVRI observations (Freedman 1988b) of 11 of them have now been published. Freedman's work incorporated the near-infrared data. Using the new fitting procedure (rather than an earlier adopted linear extinction approximation), and excluding the lower-accuracy H band data, yields a total mean reddening of E (B-V) = 0.02 mag, and a true modulus of 24.42 ± 0.13 mag, corresponding to a distance of 765 kpc. To maintain homogeneity for comparison of relative distances, this value supersedes the distance modulus 0.12 mag lower, quoted by Freedman (1988b).

It is also worth noting at this point that Freedman's (1988b) conclusions concerning the universality of the slope of the PL relation (once brought in to some degree of doubt because of the uncertainties in the faint-magnitude calibration of the photographic data on the IC 1613 Cepheids) have subsequently been bolstered by the additional analysis of 16 newly confirmed Cepheids in IC 1613, as discussed in Carlson & Sandage (1990).

In comparison to any other member of the Local Group, IC 1613 offers the best opportunity for a ground-based telescope to provide a definitive calibration of the multiwavelength period-luminosity relation for Classical Cepheid variables. It is nearby and therefore offers the opportunity for accurate photometry; it has very low foreground (and internal) extinction and yet it is distant enough that back-to-front effects are negligible. Finally, IC 1613 (and its Cepheid population) is metal poor, both in an absolute sense and with respect to the Galactic Cepheids in clusters. Comparison of a low-metallicity calibration through IC 1613 with the high-metallicity calibrations through the LMC and Milky way, respectively, may shed additional light on the metallicity sensitivity of the Cepheid PL relation zero point.

One of the most extensive studies of the dwarf irregular Local Group galaxy NGC 6822 is the photographic work of Kayser (1967). This study built on the original work of Hubble (1925) who found several Cepheids and a number of bright irregular variables in this galaxy. Unfortunately, NGC 6822 is fairly close to the Galactic plane (b = -18 degrees), resulting in large, and still somewhat uncertain, foreground reddening estimates. Published estimates for the reddening to the Cepheids in NGC 6822 range from E (B-V) = 0.19 to 0.42 mag.

Results on the Cepheid distance to NGC 6822 have been appearing slowly. Both Hodge (1977), and then van den Bergh & Humphreys (1979) have reported photoelectric BV observations of the 65-day Cepheid, V7 in NGC 6822. Multiwavelength BVRI CCD data have been obtained by the authors, but are as yet unpublished; while Schmidt, Spear & Simon (1986), Schmidt & Spear (1987), Schmidt & Simon (1987), and Schmidt & Spear (1989) have published some CCD observations of Cepheids in NGC 6822, which were obtained for other reasons. Of late, the only directed study of the Cepheid distance to NGC 6822 is the paper by McAlary et al. (1983) on near-infrared H-band aperture photometry of 9 Cepheids. Unfortunately, an independent determination of the foreground/internal reddening was not attempted because only one wavelength was involved in the new study. Visvanathan (1989) observed three Cepheids in NGC 6822 once each at 1.05 µm and derived a true modulus of 23.26 mag (scaled to an LMC true modulus of 18.5 mag). On the other hand, random-phase I-band CCD data (Lee, Freedman & Madore 1993) yield a true modulus of 23.59 mag.

Pending complete publication of the new CCD data, we can use the photographic observations of Kayser (1967) in combination with the H-band observations of McAlary et al. (1983) to provide a multiwavelength fit and solve for the true distance modulus and reddening to NGC 6822. In this application we have rederived all apparent moduli with respect to our internally self-consistent set of LMC Cepheid data. Excluding Kayser's variable No. 30 (which falls many sigma above the mean B PL relation), we find (m-M) _B = 24.66 ± 0.06 mag, (m-M) _V = 24.50 ± 0.08 mag and (m-M) _H = 23.77 ± 0.17 mag, resulting in (m-M) ₀ = 23.66 mag with E (B-V) = 0.26 mag. The reddening is well within the range of previous estimates quoted above and is remarkably close to Kayser's original estimate of 0.27 mag. Despite this formal solution, it was clear that the Cepheids in NGC 6822 could profit from a modern investigation at several wavelengths. Accordingly, Gallart et al. (1996) using single-phase (multiwavelength) CCD data of the Cepheids in NGC 6822 have derived a new estimate of distance and reddening. They have also compared their result with the the distance calculated using the tip of the red giant branch as a distance indicator. A distance modulus of (m-M) ₀ = 23.49 ± 0.08 mag and a reddening of E (B-V) = 0.24 ± 0.03 mag are derived.

Hubble (1926) discovered 35 Cepheids in the inner regions of M33 and determined their periods. These photographic data were later recalibrated by Sandage (1983). Sandage & Carlson (1983a) added identifications, periods and photographic mean magnitudes for 13 new Cepheids in an outer region of this galaxy, while Kinman, Mould & Wood (1987) then surveyed the main body of M33 using photographic plates taken at the prime focus of the KPNO 4m in search of long-period variables, and in the process discovered 54 new Cepheids. All four of these publications include finder charts for their variables.

Building on these discovery papers, single phase H-band observations (Madore et al. 1985) were made of 15 Cepheids in M33. These were then augmented by and incorporated into a multiwavelength study of 19 Cepheids (Freedman, Wilson & Madore 1991) using CCD observations to obtain light curves and therefore time-averaged magnitudes and colors. It was the H-band study of the M33 Cepheids that first noted the basic limitation due to crowding and confusion on the application of aperture techniques to the infrared Cepheid distance scale. JHK array imaging of individual Cepheids in M33 and M31 is being done by the authors using the Palomar 200-inch, so some of those early limitations should soon be lifted. Until those data are available, the CCD observations alone indicate that for M33, A_B = 0.41 mag and the true modulus is 24.63 ± 0.09 mag, corresponding to a linear distance of 840 kpc.

Baade & Swope (1963, 1965) and Gaposchkin (1962) used 200-inch plate material to catalog Cepheid variables in M31. With the exception of the outermost Field IV (for which there is both B and V data) only blue photographic magnitudes are available, making reddening estimates for the Cepheids both circumstantial and rather unreliable. Nearly a quarter of a century later Welch et al. (1986) obtained single-phase H-band observations of 22 Cepheids in M31. More recently, Freedman & Madore (1990) have published multiwavelength PL relations for 38 Cepheids in three of the Baade & Swope M31 fields resulting in independent reddenings and true distance moduli for each of the regions. A consistent true distance modulus of 24.44 ± 0.10 mag (corresponding to 770 kpc) is determined here with mean reddenings ranging from E (B-V) = 0.00 mag in Field IV to 0.25 mag in Field III. Those data are shown plotted in Figure 6.