Hello Nicholas,
my guess would be that the assumption that rho>0 is showing in lots of places in GRhydro thus not likely to be easily removed.
Just to be sure: "rho" in GRHyro is the rest-mass density, basically the number density multiplied by the rest mass of the particles. For photons this seems to be an hard quantitiy to define.
The energy density ("tt-component of the stress energy tensor") on the other hand is not actually evolved by GRhydro and is instead either provided part of the stress energy tensor in the TmunuBase::eTtt variable.
GRHydro (and the Valencia formulation of hydrotdynamics that it uses) does involve (almost) this component of Tmunu in its "energy density" variable "E" but usually that one is not directly evolved and instead "tau" the difference between total energy "E" and relativistic rest-mass energy "D" is evolved.
Please see:
https://link.springer.com/article/10.12942/lrr-2008-7#Sec2.1.3
and
https://www.einsteintoolkit.org/thornguide/EinsteinEvolve/GRHydro/documentat...
for the definitions.
Since you are at Waterloo you should also be able to reach out to people at Perimeter institute (Luis Lehner, Erik Schnetter come to mind) who are experts on fluid dynamics and NR.
Yours, Roland
Hello there
I have been learning how to use and working with the Einstein Toolkit for a few months now and I have gotten to a point where I wanted to implement a photon gas with (locally) negative rho as part of my research, following a paper written by one of my supervisor's PhD students. While looking into it I found that GRHydro automatically discards negative values of rho and replaces them with rho_abs_min, and I can't set rho_ans_min to a negative value because they are ignored according to the documentation. I was wondering if there is a way to set rho_abs_min to be negative without it being ignored, or a way to circumvent the check for negative values of rho somehow to allow for negative values. I have some experience with rewriting thorns for my own purposes but GRHydro isn't as simple as the thorns I've been changing.
Thanks, Nicholas Olsen, University of Waterloo