Systematic variation of the stellar initial mass function in early-type galaxies

Abstract

Much of our knowledge of galaxies comes from analysing the radiation emitted by their stars, which depends on the present number of each type of star in the galaxy. The present number depends on the stellar initial mass function (IMF), which describes the distribution of stellar masses when the population formed, and knowledge of it is critical to almost every aspect of galaxy evolution. More than 50 years after the first IMF determination, no consensus has emerged on whether it is universal among different types of galaxies. Previous studies indicated that the IMF and the dark matter fraction in galaxy centres cannot both be universal, but they could not convincingly discriminate between the two possibilities. Only recently were indications found that massive elliptical galaxies may not have the same IMF as the Milky Way. Here we report a study of the two-dimensional stellar kinematics for the large representative ATLAS(3D) sample of nearby early-type galaxies spanning two orders of magnitude in stellar mass, using detailed dynamical models. We find a strong systematic variation in IMF in early-type galaxies as a function of their stellar mass-to-light ratios, producing differences of a factor of up to three in galactic stellar mass. This implies that a galaxy's IMF depends intimately on the galaxy's formation history.

Dusty star-forming galaxies at high redshift

“…Auger et al, 2010;Treu et al, 2010;Spiniello et al, 2011;Cappellari et al, 2012Cappellari et al, , 2013Brewer et al, 2012;Dutton et al, 2013;Tortora et al, 2013), though we note that increased mass to light ratios can result from both bottom-light and bottom-heavy IMFs (the former owing to increased numbers of low mass stars, which the latter originating in increased numbers of stellar remnants). Considering both the indirect evidence of potential IMF variations at high-z, as well as observations of presentepoch massive galaxies (which are likely descendants of starbursts at high-z), it is fair to say that the form of the IMF in high-z systems is still a completely open issue.…”

Giant disk galaxies : Where environment trumps mass in galaxy evolution

“…Such behavior is also seen in massive early-type galaxies (Bernardi et al 2011). The systematic deviation might also hint at bottom heavy IMFs in more massive galaxies, again as seen in massive early type galaxies (van Dokkum & Conroy 2010;Cappellari et al 2012). The HI massive galaxies lie systematically above the BTF.…”

“…This error could arise because the stellar initial mass function (IMF) is more bottom heavy in the faster rotators. Although this hypothesis may seem farfetched, there is empirical evidence for this type of behavior among early type galaxies (van Dokkum & Conroy 2010;Cappellari et al 2012), where the IMF becomes progressively more bottom heavy as one considers galaxies with larger velocity dispersions. Evidence for IMF variations even extends to Local Group clusters (Zaritsky et al , 2014a, and so the possibility that such variations exist among disk galaxies as well should not be quickly dismissed.…”

The First Billion Years project: gamma-ray bursts at z > 5

“…The IMF for Population-II stars is usually directly measured for nearby star clusters and locally resolved stellar populations (Bastian, Covey & Meyer 2010), and occasionally indirectly via dynamical modelling (Cappellari et al 2012 LGRBs is redshift independent changing the IMF would only change the overall normalisation (or absolute value of LGRBs). This, however, has no effect on the results outlined in this paper and any change in the overall rate of LGRBs caused by changing the IMF would be much smaller than the other model parameters involved (we discuss this in more details in Sect.…”