Evaluate river and calving heat fluxes

jkrasting commented 6 years ago

During the development of the 0.25° ocean used in CM4, explicit land model heat fluxes from rivers and calving were not used. Instead, the heat fluxes on the ocean side assume a temperature equal to the SST for liquid and frozen runoff mass fluxes.

The 0.5° version of the ocean that will be used in ESM4 will use the calving and river heat fluxes. They are controlled by the parameters:

USE_RIVER_HEAT_CONTENT = True USE_CALVING_HEAT_CONTENT = True

A few notes about these parameters:

Regardless of their value, they are always set to .false. when running with KPP. This is a non-issue in the current model configurations but it is something to keep in mind if we revisit KPP.
In the commit message for 02b171e, @adcroft mentions that the fluxes should be used in OM4_025.

Talked with @StephenGriffies and @slm7826 and we came up with a list of things to verify:

Confirm that heat content is being passed off to the ocean is relative to 0 °C, as MOM6 expects.
The diagnosed heat content of the land model ocean discharge (rv_d_c_het ; units W m-2 per unit cell area) should match what the ocean receives
The heat budget internal to the ocean model should be self-consistent. @StephenGriffies wrote a Python notebook to diagnose the surface fluxes and compute budgets. This notebook can be adapted for the 0.5° model.
Run a 2-day test of OM4_05 coupled example using the diag_table tailored for the surface budget analysis

StephenGriffies commented 6 years ago

I suggest that the diag_table "tailored for surface budget analysis" become a standard diag_table for the 0.5 degree model. We never know when it may be important to perform a heat budget analysis including heat from liquid and solid runoff. As the surface fluxes are two-dimensional, they introduce a minor hit to the archive usage.

Hallberg-NOAA commented 6 years ago

While I agree that it is a good idea to have a default diag_table for certain configurations that allows the surface heat budget to be analyzed, this is an issue for NOAA-GFDL/MOM6-examples, not MOM6.

There are no diag_table files in the MOM6 repository. Moreover, the laudable to-do list that @jkrasting gives when opening this comment chain will necessarily be specific to the FMS-based coupled model configurations; verifying that each of these measures of cross-component consistency will check out for GFDL coupled model configurations is no guarantee that they will also work for other configurations using MOM6 with other couplers.

StephenGriffies commented 3 years ago

@jkrasting can we close this issue or is there a remaining task?

jkrasting commented 3 years ago

I think we can go ahead and close this from the ocean perspective. Not sure where this issue should be logged though. The ocean-ice and land systems are using slightly different freezing temperatures in their enthalpy calculations (SIS2=273.15; FMS constants=273.16).

StephenGriffies commented 3 years ago

Although @Hallberg-NOAA makes a good point about logging this issue under MOM6-examples, I note that there are possible code implications to handle the temperature of frozen precip. Those code issues are currently being considered by @gustavo-marques at NCAR and linked to issue #894

I thus suggest we keep the current issue open.

mom-ocean / MOM6

Evaluate river and calving heat fluxes #629

I think we can go ahead and close this from the ocean perspective. Not sure where this issue should be logged though. The ocean-ice and land systems are using slightly different freezing temperatures in their enthalpy calculations (SIS2=273.15; FMS constants=273.16).