Linking subglacial hydrology to Heat Equation

Extension of Elmer in computational glaciology
Post Reply
joeatodd
Posts: 36
Joined: 02 Feb 2012, 18:49
Antispam: Yes

Linking subglacial hydrology to Heat Equation

Post by joeatodd »

Hi all,

I'm seeking some advice on linking the residual of the limited heat equation to subglacial hydrology at the bed.

My model setup has a Dirichlet temperature condition at the bed (Pressure Melting Point) and, if I understand correctly, the heat sinks and sources imposed in satisfying this Dirichlet condition should be directly related to the melting/refreezing at the bed by the value of the latent heat of fusion.

From reading the documentation, I understand that

Code: Select all

 Calculate Loads = Logical True 

can be set, which makes available the "Temperature Loads" variable.

My plan would be then to have the source term set such that:

Code: Select all

 
$LHF = latentheatoffusion

Water Flux = Variable Temperature Loads
   Real MATC "tx/$LHF"
Can anyone advise me if this will achieve the desired result or if I've missed something?

Thanks!

Joe
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Linking subglacial hydrology to Heat Equation

Post by raback »

Hi Joe

The nodal loads are in Units of power, e.g. Watts. So it would be natual to use the them directly as loads in the coupled equation as well. So this "Water Flux" should then be directly a nodal quantity and you would add it directly to the r.h.s. vector. This is actually automatically available as a library function. Assuming that you would apply it to the N-S equation you could then set the source term for continuity equation as

$Cmelt=330.0e3
Flow Solution 4 Load = Variable Temperature Loads
Real MATC "tx/Cmelt"

Or something like that...

In order to couple equations on a discrete level the mesh must of course be the same. If the method is applicable then it is often the preferred method.

-Peter
Post Reply