Read on Twitter
My latest work that was accepted for publication: https://arxiv.org/abs/2102.00629
A">https://arxiv.org/abs/2102.... lot of my work is related to tying fundamental magnetised fluid dynamics to astrophysical processes. This study is no exception where we extend a well know density variance - Mach number relation.. 1/n
A">https://arxiv.org/abs/2102.... lot of my work is related to tying fundamental magnetised fluid dynamics to astrophysical processes. This study is no exception where we extend a well know density variance - Mach number relation.. 1/n
to a regime where the flow is highly-anisotropic due to very strong, coherent magnetic fields. In this regime we find a distinct bimodal nature to the density fluctuations and think these form from two types of shocks that form parallel and perpendicular to the B-field: 2/n
Why do we care about shocks? Well we show how the volume-weighted density variance of the field can be simply related to over-densities, under-densities and logarithmic over-densities. For supersonic turbulence the over-densities make the largest contribution:
By solving for the shock jump conditions we are able to estimate the density contrast for each shock type, and are able to construct a density variance that has two terms, both which make different contributions to the total variance and give rise to the anisotropy 4/n
We test our model on a bunch of 3D, supersonic ideal MHD turbulence simulations, all in the anisotropic regime where the mean-field has more magnetic energy than the turbulent kinetic energy 5/n
If you compute the parallel and perpendicular divergence you can actually see the two types of shocks that we use in our model (top), which is very cool, and the corresponding density contrasts (bottom): 6/n
We compute the time-averaged density variance as a function of turbulent Mach number and show our model (blue curves) does a pretty amazing job with just a single free parameter, the volume-filling fraction of the hydrodynamical shocks that form along the magnetic field 6/n
Also, the model has some pretty nice properties… for example, as the B field becomes infinitely strong our model predicts that you get something similar to the hydro. variance relation, but reduced by a filling factor as the fluctuations are constrained to just along B
As the B field gets super weak we get back to the regular hydro. variance relation:
AND, perhaps one of the coolest aspects, as the Mach numbers gets infinitely large the magnetic field sets up upper bound on the total variance… this is the first model to predicts an upper bound for the density variance (hydro. is unbounded in Mach) :
So what, right? Well… if you are into predicting the star formation potential of a molecular cloud then one way of going about that is to compute a density PDF, define a critical density where the turbulent and gravitational energy are equal and integrate that PDF
To get the star formation rate per free fall time… the density variance describes the width of this PDF, so more-or-less the wider the PDF the more stuff in that high-density tail that can potentially form stars
Our new models suggests that if there is a strong coherent field present then there is an upper limit for the density variance, and hence for the star formation potential of that region in the MC.
But this is only very early days in developing star formation in this regime. We have plenty more work to do, including understand the nature of the scale in which gravity and turbulent energies balance each other out, which we believe will be highly anisotropic itself.
Oops, forgot to keep numbering things… ummm… n/n
