Spectrophotometric evolution of elliptical galaxies. 1: Ultraviolet excess and color-magnitude-redshift relations

Abstract

We present new chemical-spectrophotometric models of population synthesis to predict the spectral evolution of galaxies, in which stars span wide ranges of ages and chemical compositions. The library of stellar models assembled here supersedes other existing libraries for the number of tracks and the coverage of stellar phases and chemical parameters. All evolutionary phases, from the main sequence until the white dwarf or C-ignition stage, as appropriate for the star mass, are included. The range of metallicity and helium content goes from Z = 0.0004 and Y = 0.230 to Z = 0.1 and Y = 0.475. Six combinations of Z and Y are considered, however, scaled according to the law {DELTA}Y/{DELTA}Z = 2.5. The inclusion of stellar models of both low and high metallicity allows us to follow in a great detail the evolution of spectral properties as function of the chemical enrichment and age. The library of stellar spectra stands on the most recent release of the Kurucz library, however, implemented at the low effective temperatures by means of observed spectra for stars of the latest spectral types. The population synthesis technique stems from the calculation of accurate isochrones in the color-magnitude diagram. Detailed results and useful calibrations are presented for single stellar populations aimed at providing, first, the building blocks of galaxy models and, second, the basic tools for studies of star clusters. The chemical-spectrophotometric model presented here is particularly designed for elliptical galaxies. It includes the presence of dark matter and galactic winds triggered by supernova explosion and energy injection by stellar winds from massive stars and allows for different laws of star formation. This model naturally follows the gradual enrichment in metallicity as a function of time both for the gas and the stars, thus providing the distribution in metallicity among the various stellar populations of a galaxy. The chemical model is the entry point for the model of spectral synthesis both for the detailed spectra and the broadband colors of the Johnson system. The results are used to study the color-magnitude relation for nearby galaxies and the origin of the UV excess in elliptical galaxies together with its dependence on metallicity and Mg_2_ index. We confirm and extend previous predictions that the main sources of the UV excess are the old, hot horizontal-branch (HB) and asymptotic giant branch (AGB) manque stars of high metallicity present in varying percentages in the stellar content of a galaxy. Since in our model the mean and maximum metallicity are ultimately driven by the mass of the galaxy, this provides a natural explanation for the observed correlation between UV excess and metallicity. We seek to disentangle the effects of metallicity and age on the global properties of a galaxy and address the question whether the prototype galaxy M 32 suffered from a recent episode of star formation. With the aid of the new calibrations for the magnitudes and colors of single stellar populations in particular stages of evolution and the new chemical-spectrophotometric model, we find that spectrum, turnoff color, integrated colors, brightest red giant branch (RGB) and AGB stars are all compatible with the notion that M 32 is an old system that underwent a prominent initial episode of star formation followed by later activity at about 5 X 10^9^ yr ago. Finally, we examine the photometric evolution of the models as a function of the redshift, and, in particular, we address the question whether there are prominent features of the evolving spectra of galaxies that might be used as age probes. We suggest that the color (1550 - V) measuring the intensity of the UV emission of a galaxy is a good age indicator. Specifically, this color suffers from a sudden variation from red to blue at the onset of the hot HB and AGB manque stages of the old, metal-rich stars, most likely present in massive galaxies. In the framework of the stellar models in use, this transition occurs at about 7.6 X 10^9^ yr. Equivalently, it is expected to be observed at redshifts perhaps accessible to present-day instrumentation. The detection of this feature at a certain redshift would impose firm constraints on the underlying cosmological model of the universe. Finally, we present the preliminary comparison of the magnitudes and colors as a function of the redshift, with the observational ones limited to a sample of radio galaxies. The implications of the present models for the study of galaxy evolution are shortly summarized in the concluding remarks.