search

mnlevy1981 / MARBL

Marine Biogeochemistry Library
Other
0 stars 0 forks source link

test_set_interior: how to test? #5

Closed mnlevy1981 closed 5 years ago

mnlevy1981 commented 5 years ago
  1. Run CESM with nstep output
  2. Re-run notebook to generate initial conditions / forcing
  3. Compare output from compute_cols test with output from same column after single timestep in POP
mnlevy1981 commented 5 years ago

Test case directory: /glade/work/mlevy/codes/CESM/cesm2_2_beta02+pop_trunk/cases/g.e22b2.G1850ECO.T62_g17.gen_single_col.nstep_out with following modifications:

  1. POP_TAVG_R8 = True in env_build.xml
  2. lecosys_tavg_all = .true. in user_nl_pop
  3. lecovars_full_depth_tavg = .true. in user_nl_marbl
  4. SourceMods/src.pop/ocn.base.tavg.csh to switch stream 1 -> nstep

Notebook has been updated to generate initial conditions from g.e22b2.G1850ECO.T62_g17.gen_single_col.nstep_out.pop.h.0001-01-01-07200.nc, have not yet compared output.

mnlevy1981 commented 5 years ago

I think I need to be smarter about combining data from the first two time steps to generate the forcing files - if POP is outputting tracer state at the end of the time step, then I want to initialize to tracer state from the first history file but use the forcing from the second time step (and then compare tendency computation to the second time step as well?)

mnlevy1981 commented 5 years ago

Process is well-defined in #13 but that issue is still open until we include documentation in MARBL itself.