atmosphere

MITgcm includes an atmospheric physics as a package and it is in aim_v23. The following is the description of this package you can find in MITgcm/pkg/aim_v23.

aim_v23_description.tex

Package aim_v23 contains code for an intermediate complexity atmospheric physics scheme derived from the physics in the SPEEDY model of Franco Molteni. The package contains the SPEEDY physics routines, re-cast to work with pressure coordinate dynamics, together with code that interfaces the SPEEDY internal arrays to the MITgcm dynamical kernel.

The following is the description of the verification named aim.5l_cs in MITgcm’s user manual.

aim.5l_cs

5-level intermediate atmospheric physics, global configuration on cube sphere grid (32x32 grid points per face, roughly 2.8° resolution). Also contains an additional setup with a slab-ocean and thermodynamic sea ice.

First, open the terminal and go to the example directory cd /MITgcm/verification/aim.5l_cs

We want to use more than 1 cpu, we use SIZE.h_mpi.

• go to code directory
• rename SIZE.h as SIZE.h_single
• rename SIZE.h_mpi as SIZE.h \(\rightarrow\) It specifies the number of cpus as 6.
• go to build directory
• do ../../../tools/genmake2 -mods ../code -optfile ../../../tools/build_options/linux_amd64_gfortran -mpi
• if it is finished without a severe error, do make depend
• if it is finished without a severe error, do make
• if you have mitgcmuv, then the compliation is successful

The default setting integrates the model for only 10 time steps. We want to integrate the model for a couple of years. Here is the step for the model integration

• go to run directory
• copy all input files to here: cp ../input/* .
• create a sympolic link of the executable file to here: ln -s ../build/mitgcmuv .
• overwrite the data file to here by
  • hover the pointer here and do right-click
  • choose Save link as...
  • save this file in run as data

Now, let’s run the model!

• execute the run: mpirun -np 6 ./mitgcmuv

We want to see the model output using python. What I usually do is to make a new directory called analysis to analyze the model results.

I have prepared a simple JupyterLab code to start with. Hover the pointer here and do right-click, then Save link as.... If you decide do work in the analysis directory, save it to there.

In order to launch this file, open up the terminal, go to the directory where you saved this file (or analysis). Then

jupyter lab

You will have the web browser showing the JupyterLab interface. You can find this file in the left panel. Then double click it to open it.

What if we increase the CO2 concentration in the atmosphere?

• go to code directory
• ../../../pkg/aim_v23/AIM_OPTIONS.h .
• open AIM_OPTIONS.h and change from #undef ALLOW_AIM_CO2 to #define ALLOW_AIM_CO2
• go to build directory
• do make CLEAN
• compile the code again as in compile
• prepare the simulation as in preparation: you may change the directory name from run to run_co2 or similar.
• edit data.aimphys \(\rightarrow\) add

  aim_select_pCO2 = 1,
  aim_fixed_pCO2 = 420.E-6,
  

  in &AIM_PARAMS
• run the model as in model integration