calculation of saturated frozen thermal conductivity of soil

[yis2@mcmaster_ca]

In module SoilTemperatureMod of Community Land Model 3, there a line to calculate saturated frozen thermal conductivity of soil,


> dksat = tkmg(c,j)*0.249**(fl*watsat(c,j))*2.29**watsat(c,j)
>
> 2.29 is for thermal conductivity of ice
> 0.023 for air
> 0.6 for liqid water

what is 0.249 for.


Thanks

------------------------------------
I cannot find the appropriate category for my question, so I just put in bug reporting. Why not open a category for land surface modeling ?

 

[mvertens]

The constant 0.249 is the thermal conductivity of water divided by the thermal conductivity of ice. You can rearrange the equation documented in the CLM tech note (eq. 6.60) to get the equation above. However, for one thing, this number shouldn't be hard coded (and neither should 2.29 which is the thermal conductivity of ice). We should be using the variables used in the code for physical constants. For another thing, the value 0.57 is actually used for the thermal conductivity of water instead of 0.60 as it is elsewhere. 0.249 = 0.57/2.29. This was discovered when writing the tech note and is a legacy from the precurser model, the Common Land Model. Since the code was frozen at the time, no changes were made. A future version of the model will probably include changing from the hard coded values to the fixed physical constants
used in the model.

 

[yis2@mcmaster_ca]

Thanks