Astron. Astrophys. 345, 419-429 (1999)

4. Modeling narrow band indices of CSPs

If fine structures and/or signatures of interaction in early type galaxies imply additional star forming episodes of various intensities and ages, how they would reflect into the spectro-photometric properties of these systems? Do the line strength indices keep memory of this past activity? How can we model this complex dynamical picture from the point of view of spectro-photometry?

A viable approximation to the problem is to consider a galaxy of a certain mass, which underwent bulk star formation in the far past, acquired its own pattern of abundances and ages for the mix of stars, evolved passively ever since and at a certain age hosted an additional episode of star formation of suitable duration and intensity in which a certain amount of gas was turned into stars. The source of gas can be either internal, if any is left over and retained after the first initial activity, or acquired from outside during the interaction episode.

If the initial star forming activity took place on a time scale much shorter than the Hubble time, the host galaxy can be further approximated to a single entity, in which stars span a certain range of age and metallicity. In the context of the standard galactic wind scenario (Larson 1974), the bulk star forming activity is terminated within the first Gyr and the mean metallicity of the stars (galaxy) is from nearly two times solar to a third of solar as the galaxy mass decreases from 1 to 0.01 (see for instance the complete galaxy models by Bressan et al. 1994, 1996; Tantalo et al. 1996, 1998).

The bursting mode idea has long been around starting from the pioneer studies by Huchra (1977) and Larson & Tinsley (1978) to explain the scatter in broad band colors observed in normal and peculiar galaxies of the same morphological type, till the more specific attempt by Leonardi & Rose (1996) who investigated the effects of a burst of star formation on the indices H+K(CaII) and H.

Leonardi & Rose (1996) start from the spectrum of a normal galaxy, obtained as the average spectrum of about 70 elliptical galaxies observed in the core of the Coma cluster and add to this template the synthetic spectra of young SSPs with solar metallicity and different ages.

The age of the SSP stands for different burst ages, whereas the burst strength is simulated by changing the relative proportions in which old and young spectra concur to build the final spectrum of the bursting galaxy .

The indices H+K(CaII) and H/FeI are then measured on the composite spectrum of the post-star-burst models and are compared with the observational data. The location of a sample of E+A galaxies on the H+K(CaII) vs. H/FeI diagram provides a first estimate of the ages and intensity of the star forming activity characterizing E+A galaxies.

In this work we make use of the same technique. The main difference with respect to Leonardi & Rose (1996) is that also for the normal galaxy we adopt a theoretical spectrum of suitable age and metallicity. In all experiments we are going to present the age of the quiescent galaxy is 15 Gyr and the metallicity is solar. Tests however have been made at varying the age and metallicity of the host galaxy. Specifically, we have also adopted for the age the values of 12 and 18 Gyr and for the metallicity the values (20% of the solar value) and (2.5 times the solar value). As expected the results are little affected by the choice for the age of the host galaxy, whereas they are more sensitive to the metallicity.

To simulate the bursts of star formation we have added to the template spectrum of the host galaxy, spectra of SSPs of seven different ages, i.e. 10, 6, 4, 2, 0.9, 0.1 and 0.05 Gyr.

In each simulation, the spectra of the two components (host galaxy and SSP) are weighted by a factor expressing the percentage of stellar mass, relative to the total mass, created during the corresponding star formation episode.

Table 3 shows the correspondence between the fractionary mass in the burst and the contribution to the total luminosity by the two component. This correspondence has been calculated using the mass to light ratio characterizing each SSP at the chosen age.

Table 3. Simulations of bursting galaxies. T, , and are the age, mass and bolometric magnitude and mass to light ratios of the SSPs representing the burst. and refer to an ideal SSP whose initial mass function in number is normalized to one over the whole mass range. The second half of the table shows the ratio of the luminosity associated to the young SSP to the total luminosity of the system at varying the age and mass percentage of the SSP representing the burst.

The mass is given by the total mass in living stars plus the mass in White Dwarfs and Neutron stars for which we have assumed the typical values of 0.6 and 1.4, respectively. The dependence of the White Dwarf and Neutron Stars mass on the progenitor mass is neglected here for the sake of simplicity.

The luminosity factor is taken from the Padua library of SSPs (Bertelli et al. 1994, Bressan et al. 1994, Tantalo et al. 1996).

The percentages of mass supposedly stored in the young SSPs ranges from 0 (only the host galaxy is considered) to 100 (only the young population contributes to the spectrum).

For purposes of illustration, we show in Fig. 4 a few cases with different burst age and intensity (percentage of mass turned into stars). Each panel contains the spectrum of the hosting galaxy, the spectrum of the SSP representing the burst and the resulting total spectrum. Worthy of note is that the contribution from the young component grows at decreasing age of the burst and that even very small percentages of a young component may deeply alter the final spectrum. This is particularly evident looking at the spectral region short-ward of 4000 Å, where, for instance, the contribution by a SSP of 0.9 Gyr involving only the 5% of the total galaxy mass parallels that of the host galaxy, 15 Gyr old and containing the remaining 95% of the mass. Furthermore, a stellar population of 0.1 Gyr representing only the 5% of the total mass can produce 2/3 of the total luminosity emitted at Å. This comparison makes evident the usefulness of the spectral region short-ward of 4000 Å to unravel even small traces of recent star forming activity.

Fig. 4. Integrated spectral energy distribution (bold lines) of a mixed population made of an old (15 Gyr; thin solid line) and a young SSP (dotted lines) component born in a recent star forming event. Left panels refer to the case of a young component of 2 Gyr. The percentages of the mass involved in the star-burst episode are indicated at the top left of each sub-panel. Right panels show the same but for a young component with age of 0.1 Gyr. In all cases the metallicity is solar

© European Southern Observatory (ESO) 1999

Online publication: April 19, 1999
helpdesk.link@springer.de