Is the stellar mass function static?

[Jerry]

The stellar initial mass function of metal-rich populations

Pavel Kroupa

http://arxiv.org/PS_cache/astro-ph/pdf/0703/0703282.pdf

The stellar initial mass function (IMF), _(m) dm, where m is the stellar mass, is the parent distribution function of the masses of stars formed in one event...

Does the IMF vary? Is it significantly different in metal-rich environments than in metal-poor ones? Theoretical work predicts this to be the case.

There are two basic arguments suggesting that the IMF ought to be dependent on the physical conditions of star formation:

The Jeans-mass argument:

A region of a molecular cloud undergoing gravitational collapse will have over-dense sub-regions within it which are also Jeans-unstable collapsing independently to form smaller structures which themselves may again sub-fragment. Ultimately stars result.

…While the concept of the Jeans mass is very natural and allows one to nicely visualise the physical process of fragmentation, it has the problem that the densest regions of a pre-star-cluster cloud core ought to have the smaller fragment masses, but instead the most massive stars are seen to form in the densest regions.

This theory of star and galaxy evolution directly supports the notion that as the Fe/H ratio increases, the mass of the system decreases.

And so does this one:

…A rather convincing physical model of the IMF which avoids the problem with the Jeans mass argument has been suggested by Adams & Fatuzzo (1996) and Adams & Laughlin (1996). The argument here is that the Jeans mass has virtually nothing to do with the final mass of a star because structure in a molecular cloud exists on all scales. Therefore, no characteristic density can be identified, and “no single Jeans mass exists”. When a cloud region becomes unstable a hydrostatic core forms after the initial free-collapse...

The important point however is that this theory also expects a variation of the resulting characteristic mass with metallicity as above…

So this competing basic formation scenario also supports the prediction that as the Fe/H ratio increases, the mass of the system decreases. However:

…The calculations that do exist demonstrate time and again that the IMF is invariant: There exists no statistically meaningful evidence for a variation of the IMF from metal-poor to metal-rich populations. This means that currently existing star-formation theory fails to describe the stellar outcome. Indirect evidence, based on chemical evolution calculations, however indicate that extreme star-bursts that assembled bulges and elliptical galaxies may have had a top-heavy IMF…

It would have seemed that when two mutually more or less exclusive theories of the origin of stellar masses make the same basic prediction concerning the variation of the average stellar mass with physical conditions, that this expected variation would be very robust and born out in observational data.

Alas, the observational data on the IMF are resilient - they do not yield what we desire to see. The stellar IMF is invariant and can be best described by a two-part power-law form. This holds true for metallicities
ranging from those of globular clusters to super-solar values near the Galactic centre, and for densities less than about 106 stars/pc3. Even the maximal stellar mass of about 150M⊙ seems to be independent of metallicity for Z>∼ 0.008.

Given the quite total failure to account for the resilience of the stellar IMF towards changes, it is clear that this IMF-conservatism poses some rather severe challenges on liberal star formation theory.

If there is no dependancy, no measurable decrease in the total mass of systems as the metallicity increases, wherein lies the cosmic aging mechanism?

[StupendousMan]

It is no surprise to me that we don't understand the physics of star formation. It's a very complicated business.

I was not aware that astronomers had actually measured the initial mass function (IMF) well enough, in enough different environments, to place strong constraints on its form. It's very difficult to measure stellar masses (problem number one), and almost all stars we can see formed a very long time ago (not yesterday, as required to measure the INITIAL mass function).

Some people try to observe the current luminosity function of stars in some population, then go from luminosity to mass, then evolve all the stars backwards in time to find the INITIAL mass function. Oh, wait, they also (usually) have to make huge corrections for incompleteness at the low-mass end, which is where most of the action is.

I don't put a lot of weight in most of those efforts.

This is absolutely one of the areas in astronomy that needs a LOT of work.

[Nereid]

Why post this in the ATM section?

As StupendousMan notes, this is an active area of research ... and you haven't presented any ATM alternative, have you Jerry?