Models based on galactic winds

Monolithic models assume that ellipticals suffer a strong star formation and quickly produce galactic winds when the energy from SNe injected into the ISM equates to the potential energy of the gas. Star formation is assumed to halt after the development of the galactic wind and the galaxy is assumed to evolve passively afterwards. The original model of Larson (1974) suggested that galactic winds should occur later in more-massive objects due to the assumption of a constant efficiency of star formation in ellipticals of different masses and to the increasing depth of the potential well in more-massive ellipticals. Unfortunately, this prediction is at variance with the observation that the ([Mg/Fe])* ratio increases with galactic mass, which instead suggests a shorter period of star formation for larger galaxies. This was first suggested by Trager et al. (1993), and Matteucci (1994, hereafter M94), who also computed models for ellipticals by assuming a shorter period of star formation in big ellipticals. In order to obtain that, an increasing efficiency of star formation with galactic mass was assumed, with the consequence of obtaining a galactic wind occuring earlier in the massive than in the small galaxies. She called this process "inverse wind" and showed that such a model was able to reproduce the increase of ([Mg/Fe])* with galactic mass.

More recently, PM04 presented a revised monolithic model that allows for the formation of ellipticals by a fast merger of gas lumps at high redshift. The model is multi-zone and predicts that each elliptical forms "outside-in" (star formation stops in the outer before the inner regions owing to a galactic wind). In other words, the galactic wind develops outside-in. Following the original suggestion by M94, they assumed an increasing efficiency of star formation with the galactic mass. They also suggested a shorter timescale t for the gas assembly with increasing galactic mass. In Figure 44.1 we show the predicted histories of star formation in the "inverse-wind scenario" of M94 and PM04. As one can see, the most-massive ellipticals undergo a shorter and more intense episode of star formation than the less-massive ones.

In Figure 44.2 we show the predictions of Matteucci (1994) concerning ([Mg/Fe])* versus the galactic mass (stellar) in the inverse-wind scenario, in Larson's classical scenario, and in the case of a variable initial mass function (IMF). The last case assumes that more-massive ellipticals should have a flatter IMF. However, this particular scenario requires too flat an IMF for massive ellipticals, redshift

redshift

Lookback Time (Gyr)

Figure 44.1. The predicted star-formation histories (star-formation rate per unit stellar mass) for galaxies of masses 1012 (dashed line), 1011 (continuous line), and 1010M0 (dotted line). Such a behavior is obtained by assuming that the efficiency of star formation is increasing with galactic mass whereas the timescale for the assembly of the gaseous lumps giving rise to the galaxies is a decreasing function of mass (downsizing both in star formation and in mass assembly). In these models (PM04) the galactic wind occurs earlier in the more-massive galaxies than it does in less-massive ones.

Lookback Time (Gyr)

Figure 44.1. The predicted star-formation histories (star-formation rate per unit stellar mass) for galaxies of masses 1012 (dashed line), 1011 (continuous line), and 1010M0 (dotted line). Such a behavior is obtained by assuming that the efficiency of star formation is increasing with galactic mass whereas the timescale for the assembly of the gaseous lumps giving rise to the galaxies is a decreasing function of mass (downsizing both in star formation and in mass assembly). In these models (PM04) the galactic wind occurs earlier in the more-massive galaxies than it does in less-massive ones.

whichis at variance with observational properties (e.g. M/Lratio, colour-magnitude diagram).

PM04 recomputed the relation < [Mg/Fe]) * versus mass (velocity dispersion) and compared it with the data published by Thomas et al. (2002), who showed how hierarchical semi-analytical models cannot reproduce the observed <[Mg/Fe])* versus velocity-dispersion trend, since in this scenario massive ellipticals have longer periods of star formation than smaller ones. In Figure 44.3, we have plotted the predictions of PM04 (continuous line) compared with data and hierarchical-clustering predictions.

More recently, Thomas et al. (2005) presented a suggestion about the starformation histories in ellipticals in the cases of high- and low-density environment (clusters and field) and suggested that the formation of ellipticals in the field might have started 2 Gyr after that of ellipticals in clusters. This suggestion is based on recent data relating to [a/Fe] and ages in ellipticals and is shown in Figure 44.4. As one can see, their suggestion for the star formation in ellipticals in clusters is similar to that of PM04.

Figure 44.2. The predicted ([Mg/Fe]}+ versus log Mf (final mass) for ellipticals, under several of the assumptions considered by M94. The curves labeled IWM0.95 and IWM1.35 correspond to models with star-formation histories similar to those shown in Figure 44.1, the only difference being that the ellipticals are considered as closed-box systems until the occurrence of a galactic wind. The case IWM0.95 assumes for all galaxies an IMF with x = 0.95, whereas the case IWM1.35 assumes a Salpeter (1955) IMF. The curve labeled CWM indicates classic wind models in which the galactic wind occurs earlier in less-massive than it does in more-massive galaxies. Finally, the curve labeled VARIMF assumes that the IMF varies as a function of the galactic mass (see the text).

Figure 44.2. The predicted ([Mg/Fe]}+ versus log Mf (final mass) for ellipticals, under several of the assumptions considered by M94. The curves labeled IWM0.95 and IWM1.35 correspond to models with star-formation histories similar to those shown in Figure 44.1, the only difference being that the ellipticals are considered as closed-box systems until the occurrence of a galactic wind. The case IWM0.95 assumes for all galaxies an IMF with x = 0.95, whereas the case IWM1.35 assumes a Salpeter (1955) IMF. The curve labeled CWM indicates classic wind models in which the galactic wind occurs earlier in less-massive than it does in more-massive galaxies. Finally, the curve labeled VARIMF assumes that the IMF varies as a function of the galactic mass (see the text).

0 100 200 300 400

Figure 44.3. The [Mg/Fe] versus a relationship for ellipticals (continuous line) as predicted by PM04 compared with observations (dots) and with the predictions of hierarchical semi-analytical models (shaded area). Figure adapted from Thomas et al. (2002).

0 100 200 300 400

Figure 44.3. The [Mg/Fe] versus a relationship for ellipticals (continuous line) as predicted by PM04 compared with observations (dots) and with the predictions of hierarchical semi-analytical models (shaded area). Figure adapted from Thomas et al. (2002).

Redshift

Redshift

Lookback Time (Gyr)

Figure 44.4. Thomas etal.'s view of the star-formation history (star-formation rate per unit mass of gas) in ellipticals of different masses and in different environments. Figure from Thomas et al. (2005).

Lookback Time (Gyr)

Figure 44.4. Thomas etal.'s view of the star-formation history (star-formation rate per unit mass of gas) in ellipticals of different masses and in different environments. Figure from Thomas et al. (2005).

Was this article helpful?

0 0

Post a comment