KITP: extragalactic globulars

it is extragalactic day here at KITP
this could get exciting, if you like that sort of thing

unsolved problems in dynamical evolution, and why they are interesting...
for the five people who care.

This afternoon we tackle colour: red and blue, again...

All our data belongs to M87
'cause M87 has shitloads (technical term) of globulars.

Magic!


"Elowitz's Law: Sufficiently advanced image processing is not distinquishable from magic."

Yup, M87

SDSS image

What we know:

we ought to be able to calculate cluster mass loss rates, but we appear to be underestimating it by something like factor or two. Maybe rotation is key.

or we may just not understand the intiial globular cluster mass function...
though again we ought to, really.

there is no mass-radius relation for globulars
and there ought to be. why is this?
star formation processes?
stellar mass function?
environment?
when does this get burned in? after 100 Myrs or before?

Aside: it is very interesting to watch different peoples' facial expression when a theorist says "but it is just data" [paraphrasing]
really, the full range of emotional states comes through...
maybe we should start regular poker games with the observers,
and grad students...

back to it:

initial mass segregation. check
is it a function of cluster mass?
why don't stellar population people take into account (differential) dynamical mass loss...

on a different note, are we completely on the wrong track?
we keep thinking of the big clusters as homologous structures,
but what if they are not? what if the big clusters all form as agglomerations of littler clusters and we form bigger clusters only where enough little clusters form close to each other to rapidly merge... this used to be inconceivable, since cluster populations are so homogenous, but now we know they are not really, and the word clearly does not mean what I think it means.

I have a headache.

Orbits.
Steve said orbits!
If we go back to β then I will have to take a break.

ok, current outer clusters (R > 75 kpc) are definitely not on extremely radial orbits
because we'd have seen the projected velocity signature already in the data
- if you don't believe me I will say "kurtosis", which is numerically unstable, so we'd actually properly use h₄ but you all knew that, didn't you?

"the radial orbit thing is dead, it just does not work, it is dead"
resting! or just pining for the fjords...

metal rich (red) and metal poor (blue) M87 clusters do not have the same mass functions, even allowing for differential evolution (in prep - Chris and Steve).
Lots more faint blue ones - radial dependence???

There are also Intergalactic globulars

I get ahead of myself, but I like this old plot:

M87 globular colours - bright clusters