I am an undergraduate in Physics, and I have an opportunity to work with a some of my seniors to learn Computational Fluid Dynamics (CFD) and Magnetohydrodynamics (MHD).
I would like to choose my first individual project in order to learn about how these techniques are used in Astrophysics.
The answer to your question depends on what you mean here. I assume you mean that you want to do an exercise, when you say "individual project." You want a topic within these fields that are well understood - that is, you are not interested in currently open questions/problems.
Further, since you're an undergraduate, you likely have not studied classical electrodynamics at an advanced level (perhaps you have?), which adds a difficulty in even understanding CFD/MHD exercises.
A big part of CFD/MHD is coding. It's very simulation heavy, unless you're doing pure theory, but I do not suspect that you intend to advance theory (as discussed above).
So, you could try to
- find an exercise that you think is a) tractable/solvable and b) interesting to you.
- Try to solve the problem first yourself, and ask your fellows for help finding an appropriate problem... it's easy to find one that's too difficult for your purposes.
- Use a coding language, e.g., Wolfram Mathematica, Python, etc..., to solve some equations and to explore the solutions. Ask yourself what you expect the solution to be before, and then ask yourself if the solution(s) you find make physical sense, etc... You can use extreme/limiting cases to explore these questions more simply.
In terms of finding an "exercise," this article/book by Spruit might be too advanced, perhaps your senior fellows can help you find an appropriate exercise, but there are many (see chapter 3). Here's the most recent version of it. Since you're an undergrad, I recommend just doing choosing a problem in electrodynamics, e.g., from Griffith's textbook, since a lot of MHD/CFD problems are applications of Maxwell's laws, etc... See here for some standard methods in CFD.