Synthetic stellar photometry - I. General considerations and new transformations for broad-band systems

After a pedagogical introduction to the main concepts of synthetic photometry, colours and bolometric corrections in the Johnson-Cousins, 2MASS, and HST-ACS/WFC3 photometric systems are generated from MARCS synthetic fluxes for various [Fe/H] and [α/Fe] combinations, and virtually any value of E(B - V) ≤ 0.7. The successes and failures of model fluxes in reproducing the observed magnitudes are highlighted. Overall, extant synthetic fluxes predict quite realistic broad-band colours and bolometric corrections, especially at optical and longer wavelengths: further improvements of the predictions for the blue and ultraviolet spectral regions await the use of hydrodynamic models where the microturbulent velocity is not treated as a free parameter. We show how the morphology of the colour-magnitude diagram (CMD) changes for different values of [Fe/H] and [α/Fe]; in particular, how suitable colour combinations can easily discriminate between red giant branch and lower main-sequence populations with different [α/Fe], due to the concomitant loops and swings in the CMD. We also provide computer programs to produce tables of synthetic bolometric corrections as well as routines to interpolate in them. These colour-T_eff-metallicity relations may be used to convert isochrones for different chemical compositions to various bandpasses assuming observed reddening values, thus bypassing the standard assumption of a constant colour excess for stars of different spectral type. We also show how such an assumption can lead to significant systematic errors. The MARCS transformations presented in this study promise to provide important constraints on our understanding of the multiple stellar populations found in globular clusters (e.g. the colours of lower main-sequence stars are predicted to depend strongly on [α/Fe]) and of those located towards/in the Galactic bulge.