Mushy Layer
1.0
|
The inputs files in this directory are designed to
To run the code you'll need to have compiled the code /mushy-layer/execSubcycle/
, and also the code in /mushy-layer/setupNewRun/
Assuming your compiled mushy layer executable is called mushyLayer2d.Linux.64.mpiCC.gfortran.OPT.MPI.ex
(it may not be), you can start growing sea ice via
This is a very coarse simulation, which should grow a sensible ammount of ice in a few minutes. Once you're happy that you have enough ice, cancel the execution and move on to the next step.
Adding a melt pond is done via a seperate program. You'll need to edit the setupNewRunAddPond
file to ensure that the inFile
, run_inputs
, and outfile
paths are correct. Then run the program via
This should create a restart.2d.hdf5
file at the location specified in the setupNewRunAddPond
inputs file.
Make sure the parameter main.restart_file
in inputsEvolveMeltPond
points to the newly created file, then start simulations like before but with the new inputs file
The key difference between growing ice and evolving the melt pond simulations are:
We increase the enthalpy boundary at the top of the domain in the vertical direction, specify the pressure head, change velocity boundary condition at the top of the domain to use this pressure head, and make the plotting period more frequent (as we now have much stronger flow, so smaller timesteps).