Hi,

I use Cormix model to calculate heated water discharge at the bank of a river.

The resulting plume size (both vertical and horizontal sizes) is smaller than that computed by a CFD model.

You know what is going on here?

Thanks

Ger

Cormix is not a detailed model. The entrainment will be calculated by empirical methods that might not be appropriate your your system.

Local turbulence and shear are going to affect this and should be better resolved by CFD (if your physics and grid are correct.

Then, if you have experience with Cormix model, do you know how accurate is the result of the model? such as the width and depth of the plume (near field) of a temperature jet shooting a the bank of a river?

Thanks for your valuable input.

Ger

Dont really know, it was quite a few years ago.

I seem to remember thsat the mid and far-field calculautions weren't that good, but maybe the plume-stuff was ok.

Hi Ger,

well I am not an expert in this but do know a bit. The two approaches are fundamentally different and it is therefore hard to make general comparisons. The Cormix model was developed by Dr. G. Jirka who is at the IfH in Karlsruhe. There is a lot of info on it if you go to the IfH web-page.

As far as a CFD simulation the main problem is appropriatly modeling the turbulent transport. Plumes have highly anisotropic turbulence fields and are not well predicted by a k-eps (or anything similar) turbulence model. There have been a lot of publications in this area. A good place to start is with Rodi's Habilitation work on turbulence models applied to Hydraulics.

If you are using a k-eps model it will likely over-predict the turbulent transport (Reynolds Stresses and Fluxes) thereby giving a broader plume...which is wrong!

Best of luck,

Bak_Flow