Inconsistency between documenation and code for topo pond scheme

[davidclemenssewall]

I believe there is a mismatch between the Icepack documentation and what is implemented in the code for the topo melt pond scheme. The 'Melt Ponds' figure in section 2.7.2.1 of the topo formulation illustrates a calculation of hydraulic head (\DeltaH) in which the top surface of the thinner ice categories can sit below sea level and the caption states: "A positive hydraulic head (pond surface above sea level) will flush melt water through the sea ice into the ocean; a negative hydraulic head can drive percolation of sea water onto the ice surface." I think this statement is inconsistent or misleading compared to what is implemented in the code in the following ways:

1. I don't see anywhere that it is implemented that "a negative hydraulic head can drive percolation of sea water onto the ice surface." It looks like this functionality should be implemented in the pond_area subroutine in icepack_meltpond_topo.F90. But in the permeability section of that subroutine, line 492 (if (ktherm /= 2 .and. pressure_head > c0) then) permits drainage/percolation only when there is a positive pressure head.

2. When mushy thermodynamics is active, the way that hydraulic head is calculated is different from what is described in the topo pond scheme. The topo documentation states: "The hydraulic head of melt water on sea ice (i.e., its height above sea level) drives flushing of melt water through the porous sea ice and into the underlying ocean. The mushy thermodynamics scheme (ktherm = 2) handles flushing." For someone reading the documentation but not checking the code, I think this could mislead them into thinking that the hydraulic head is calculated by the topo scheme, and then the permeability and flushing computation is handled by the mushy thermodynamics. In reality, the hydraulic head is also calculated by the mushy scheme.

I think that we could fix issue 1 by simply eliminating the statement "a negative hydraulic head can drive percolation of sea water onto the ice surface" from the documentation. I don't see a simple resolution to issue 2. In the mushy thermodynamics, flushing is handled by the temperature_changes_salinity subroutine which is called in the thermo_vertical subroutine. thermo_vertical only has access to information about one category at a time, whereas the hydraulic head calculation in the topo scheme requires simultaneous knowledge about all categories within the grid cell. So modifying the code to achieve what the documentation states would be nontrivial. I think at a minimum we should update the documentation to clearly explain how the hydraulic head is being calculated. I would also support adding a warning or possibly having the code abort if the topo scheme and mushy thermodynamics are both turned on.

Are there any groups that are using (or are planning to use) the topo pond scheme with the mushy thermodynamics?