Icepack: write ice temperatures to diagnostic files

[marte_maq]

Hi everyone. I am doing some testing with the Icepack model as stand-alone and would like to plot ice temperature profiles. However, I see the ice temperature is not written to the diagnostic files. My question is about how to add the ice temperatures to the diagnostic files. I see the ice temperatures are calculated in the icepack_therm_mushy script in the temperature_changes_salinity submodule. However, to write them in the diagnostic file should I add a tracer for the ice temperature? In a comparable way as is done for the salinity of the layers? (although I want the temperature for the ice layers and not for the layers on the biogrid). I hope someone can help me out with this! Marte

 

[dbailey]

Hi Marte,

This is one of the main failings of the Icepack model. It is not easy to add new variables to the log output. Have a look at the code in configuration/drivers/icedrv_diagnostics.F90. You will see code here for how the internal energy of the snow and sea ice is output. There is also a computation for the snow temperature here in the subroutine print_state. So, depending on which physics you are using, you have have to compute the ice temperature from the nt_qice tracer using the relevant formulas from either the ktherm=1 or ktherm=2 thermodynamics. Then add the code to the subroutine runtime_diags. You have to add some additional declarations and module imports. I've worked with Clara Burgard in Germany to do this for icepack.

 

[Alfred]

Hi Marte,

This is one of the main failings of the Icepack model. It is not easy to add new variables to the log output. Have a look at the code in configuration/drivers/icedrv_diagnostics.F90. You will see code here for how the internal energy of the snow and sea ice is output. There is also a computation for the snow temperature here in the subroutine print_state. So, depending on which physics you are using, you have have to compute the ice temperature from the nt_qice tracer using the relevant formulas from either the ktherm=1 or ktherm=2 thermodynamics. Then add the code to the subroutine runtime_diags. You have to add some additional declarations and module imports. I've worked with Clara Burgard in Germany to do this for icepack.

Hi, if there is some advances in this issue ?

 

[dbailey]

I haven't had a chance to work on this. I will try to up the priority.