• Support PF! Buy your school textbooks, materials and every day products Here!

Computational Astrophysics

  • Thread starter psholtz
  • Start date
  • #1
136
0

Main Question or Discussion Point

Which schools have the best programs in computational astrophysics?

Judging by their websites, Princeton and Univ. of Chicago seem to have strong programs, but I was wondering if people knew of others as well??
 

Answers and Replies

  • #2
133
2
All the good astrophysics programs are probably strong on the computational side as well.

Go to USNews physics graduate school rankings website and you'll see the astrophysics rankings.
 
  • #3
6,814
12
Which schools have the best programs in computational astrophysics?
Look for schools with a supercomputer center on-campus, and strong CS departments. Off the top of my head this includes UIUC, UCSD, University of Texas at Austin, University of Tennessee Knoxville, and University of North Carolina Chapel Hill.

It's useful to do this as a graduate student, because if it doesn't have an onsite center, then they have to get supercomputer time through competitive grant applications. If they have something onsite, then it means that the university likely put up money for the center, which means that you can much more easily get an account to "play".
 
  • #4
6,814
12
All the good astrophysics programs are probably strong on the computational side as well.
I wouldn't think this to be the case. At least when I think of computational astrophysics, I'm thinking in terms of problems like how do you apply GPU programming to fluid problems. That involves as much expertise in computer science as it does in astrophysics.

The other thing is that high performance computing tends to be interdisciplinary since you find that the problems that you find in astrophysics tend to be exactly the same as in other fields (like computational finance).

Go to USNews physics graduate school rankings website and you'll see the astrophysics rankings.
Which IMHO are totally useless and ought to be ignored. As far as I can tell the editors of USNews have no idea how to evaluate an astrophysics department.

The best thing to do is to go into the journals and websites, figure out what research excites you and go from there. Most junior and senior undergraduate physics majors are much better at evaluating what is interesting and useful research than the people that create the bogus rankings in USNews.
 
  • #5
491
2
Just to note, if you read the methodology of the US News Rankings, you'll find that it's basically a popularity contest. They ask faculty to rate what they think are the best departments, so it's rather fishy in that respect. Though I'd say it helps to get a general feel as to who's pretty good, though looking too closely (like comparing 4th and 5th place) is rather pointless.

GaTech recently got together an astronomy/astrophysics department. GT's computer science is top 10 (for what it's worth) and is a top engineering school, with pretty good physics and mathematics departments as well. Just to add to twofish's list of schools. You might expect engineering schools with good graduate programs in certain fields (like aerodynamics, engines, acoustics/vibrations, materials to some extent, and some others I can't think of right now) to have sufficient access to high performance computers, just as something to help you narrow down your search.
 
  • #6
Simfish
Gold Member
818
2
What also matters is the degree of collaboration between the departments on campus.

UWashington, for example, has an applied math program that's ranked #2 in the nation (frankly it's not one of those schools whose rating people will systematically overestimate). Its CS program is also top-tier. But none of those really have any effect on the astronomy department here (it has the potential of creating an effect, of course, but it isn't happening right now - in fact - the profs I'm working with collaborate with CS people from CMU instead). That being said, the AMath department at UW is a lot more closely tied in with the atmospheric sciences program here, which is probably one reason why the atmospheric sciences department is also ranked at around #2 in the nation. The statistics department here is also top-tier according to NRC, but most of the interdisciplinary statisticians here collaborate with biology and social science instead (same thing with the CS department actually)

On the other hand, http://escience.washington.edu/ will probably help the department a lot in the future. Plus, Washington is a pretty high-tech state to begin with. But for now, there are really only 1-2 professors here who really are into computational astrophysics (more postdocs though).

twofish's observation is still quite interesting to me though. There really aren't a lot of universities that have tier-1 departments in applied math, statistics, *and* computer science (which are the three departments that have the greatest potential for helping all the other departments here as well). And the influence of those three departments is definitely helping A LOT with the bioscience departments here (especially in terms of all the new things that have came out in the last 5 years).
 
Last edited:
  • #7
How are Maryland and Arizona for computational astro? Based on my limited research, Maryland isn't that great since its computational astrophysics professors have hardly any publications or faculty awards
 
Last edited:
  • #8
Simfish
Gold Member
818
2
Arizona is what one computational astrophysicist suggested to me for the planetary science + computational astro combo.

But I don't want to go there since Arizona is going to have some pretty massive water shortage problems in the near future (plus climate change is warming it up a lot faster than most states). And its current political climate hardly looks encouraging for scientific funding.
 
  • #9
99
0
Maryland doesn't have a fantastic computational astrophysics contingent. I took a course in the astronomy department called "computational astrophysics." One option for the final project was to redo a simply N-body simulation using a GPU. I can name two astronomy professors who are interested in the computational side of things. There are probably more that are slipping my mind at the moment.

Maryland does, however, have a solid applied math and scientific computation department (AMSC dept). They also have a center for scientific computation/math modeling (cscamm). You can search for these acronyms if you want more information about these programs.
 
  • #10
107
1
Arizona State University: http://sese.asu.edu/astrophysics [Broken]
 
Last edited by a moderator:
  • #11
Look for schools with a supercomputer center on-campus, and strong CS departments.
UC Santa Cruz is regarded as a top-tier astro program, including in computational astro, but their CS dept isn't exactly top-tier
 

Related Threads on Computational Astrophysics

  • Last Post
Replies
1
Views
1K
Replies
2
Views
3K
Replies
1
Views
3K
Replies
15
Views
224
Replies
13
Views
4K
  • Last Post
Replies
3
Views
1K
Top