2005MNRAS.357..669S

The accurate determination of galaxy relative distances is extremely important for the empirical calibration of the uncertain metallicity dependence of some standard candles like Cepheid stars, or for studying the galaxy space distribution and peculiar velocities. Here we have investigated the reliability of the widely used I-band Tip of the Red Giant Branch (TRGB) relative distances for a sample of Local Group galaxies with complex star formation histories (SFHs) and age-metallicity relationships (AMRs) namely the Large Magellanic Cloud (LMC), Small Magellanic Cloud (SMC) and LGS3. The use of the K band is also discussed. By employing theoretical stellar population synthesis techniques, we find that using actual determinations of SFH and AMR of the LMC and SMC, their RGB is populated by stars much younger (by ∼9 Gyr) than the Galactic globular cluster counterparts, on which the I-band (and K-band) TRGB absolute magnitude is calibrated. This age difference induces a bias in both the photometric metallicity estimates based on the comparison of RGB colours with globular cluster ones, and the TRGB distances. The extent of the distance bias - which is not influenced by the actual value of the TRGB absolute magnitude zero-point - is strongly dependent on the specific TRGB technique applied, and on the assumed I-band bolometric correction (BC_I) scale adopted; the correction to apply to the SMC-LMC distance modulus ranges from 0 up to +0.10 mag. LGS3 is an example of a galaxy populated mainly by old stars, so that photometric metallicity and distance estimates using globular cluster calibrations are reliable. However, the relative distance moduli between the Magellanic Clouds and LGS3 are affected by the population effects discussed for the LMC and SMC. The correction to apply to the LGS3-LMC distance modulus ranges between -0.05 and +0.14 mag, whereas in the case of the LGS3-SMC distance modulus it goes from -0.07 to +0.04 mag. In the case of all three relative distances discussed before, the correction to apply to the K-band TRGB distances are larger than the I-band case.

Our results clearly show that the presence of a well-developed RGB in the colour - magnitude diagram of a stellar system with a complex SFH does not guarantee that it is populated by globular cluster-like red giants, and therefore the TRGB method for distance determination has to be applied with caution. A definitive assessment of the appropriate corrections for population effects on TRGB distances has, however, to wait for a substantial reduction in the uncertainties on the BC_I scale for cold stars.