Why are cgs units the norm in astrophysics? Other physics communities, e.g. the particle physics one, have their own set of units, custom-tailored to their own needs. Now, the astrophysics community is somewhat similar, in that a lot of measurements are often expressed in terms of corresponding numbers related to the Sun (solar mass, solar luminosity, solar radius, et cetera). However, when it comes to using down and dirty numbers, I've seen cgs units used way more often than SI units. Why is that? Is it just a matter of tradition or is there something more subtle at work?
 A: It is convention and laziness (and I'm as guilty as anyone). In fact cgs units should not be used (according to the International Astronomical Union), in the same sense as Pluto should not be called a planet. Both were decisions made by the IAU. In the case of units, the IAU unambiguously endorsed the use of SI units, except for a short list of defined units common in Astronomy (like solar mass, parsec etc., although the electron-volt is also included).
So there you have it - the IAU has a policy, but most astronomers ignore it.
http://www.iau.org/publications/proceedings_rules/units/
Of the main journals ApJ, AJ, A&A don't specify, but MNRAS does suggest using SI units and points to the IAU policy.
Interestingly, the American Astronomical Society machine readable table standards document also says it requires SI units, but then points to a table which includes gram as an allowable unit of mass, however the electrical units are SI (Teslas, Webers etc,)?
https://aas.org/authors/aas-journal-unit-standards-machine-readable-tables
A: It's partially convention, but I think it's much more that Gaussian units are nicer to work with for a lot of the common problems in astrophysics. It's not that we prefer cm, g, and erg over m, kg, and J; that's just to signal that we're using Gaussian units. To me, it's that Maxwell's equations have constants with intuitive physical meaning that we use all the time in astrophysics ($c$ and $4\pi$) instead of the permeability and permittivity of free space, which we never really think about in astrophysics.
For example, coming straight from Appendix C of Griffith's E&M textbook, the energy density in the magnetic field is
$$
U = \frac{1}{8\pi} \int (E^2 + B^2) d\tau,
$$
with no $\epsilon_0 E^2$ and $B^2 / \mu_0$ to make the units of electric and magnetic fields different. Since we deal with electric and magnetic fields all the time (both because we observe them and because they are dominant forces in many astrophysical contexts, such as in radiative stellar atmospheres and interstellar gas), having their units be the same both makes the physics more clear and makes them easier to work with.
All that said, in a field that still uses magnitudes, discounting convention too much would be a mistake!
