Appropiate turbulence model for natural convection
 

Hello, I'm going forward in the modelling of the bitumen tank. For the real case, the Rayleigh number is very high, so we have a turbulent free convection problem. I resolved the laminar case using a gravity constant 9.81e-3 instead 9.81 m/s2 so I have an initial guess. The problem arise when I have to choose the turbulent model. I did not find a recomentation in the documentation for this case.

So, anybody knows what should be the best turbulent model for turbulent free convection problems??

Thanks