marcin-kupilas / MesoS2D-modelling

0 stars 0 forks source link

Timestep test #5

Open marcin-kupilas opened 1 week ago

marcin-kupilas commented 1 week ago

This test is to confirm if changing the timestep but keeping the same spatial resolution makes a difference to the model state. Ideally - it should not.


RESOURCES


CASE NAME f.e23.FWmaHIST.ne30pg3_me30pg3_mg17_L135.short_timestep

KEYWORDS 1deg, ne30, WACCM, 135L, tag cam6_4_028, derecho, cam7 physics

DESCRIPTION 1 degree SE WACCM 5 day test with dt=3.75 min to test if anything is changed to the model. E.g. in WACCM-RR paper, outside the CONUS region, both RR and Non-RR models have 1-degree resolution, but Non-RR has 30 min timestep and RR has 3.75 min timestep. There are global changes seen in the model, so I want to know what's up.

TAG DIR /glade/derecho/scratch/mmkupilas/tag_cam6_4_028

CASE DIR /glade/derecho/scratch/mmkupilas/cesm_sims/cam6_4_028/f.e23.FWmaHIST.ne30pg3_me30pg3_mg17_L135.short_timestep

RUN DIR /glade/derecho/scratch/mmkupilas/cesm_dirs/f.e23.FWmaHIST.ne30pg3_me30pg3_mg17_L135.short_timestep/run

ARCHIVE DIR /glade/derecho/scratch/mmkupilas/archive/f.e23.FWmaHIST.ne30pg3_me30pg3_mg17_L135.short_timestep

CREATE NEWCASE /glade/derecho/scratch/mmkupilas/tag_cam6_4_028/cime/scripts/create_newcase --compset FWmaHIST --res ne30pg3_ne30pg3_mg17 --case /glade/derecho/scratch/mmkupilas/cesm_sims/cam6_4_028/f.e23.FWmaHIST.ne30pg3_me30pg3_mg17_L135.short_timestep --project P93300043 --run-unsupported


ssh derecho

create newcase

cd CASE DIR


XML CHANGES

./xmlchange RUNDIR=/glade/derecho/scratch/mmkupilas/cesm_dirs/f.e23.FWmaHIST.ne30pg3_me30pg3_mg17_L135.short_timestep/run/ ./xmlchange RUN_REFDATE=1980-01-01 ./xmlchange RUN_STARTDATE=1980-01-01 ./xmlchange STOP_OPTION=nmonths ./xmlchange STOP_N=1 ./xmlchange REST_N=5 ./xmlchange REST_OPTION=ndays ./xmlchange CONTINUE_RUN=FALSE ./xmlchange RESUBMIT=4 ./xmlchange DOUT_S_SAVE_INTERIM_RESTART_FILES=TRUE

;env_mach_pes.xml ./xmlchange NTASKS=-20 ; 20*128=2160

Check env_workflow.xml for clock settings <"JOB_WALLCLOCK_TIME" value = "12:00:00"> <"JOB_WALLCLOCK_TIME" value = "03:00:00">

./case.setup


Edit env_build.xml: ./xmlchange CAM_CONFIG_OPTS="-phys cam7 -nlev 135 -age_of_air_trcs -chem waccm_ma_mam5 -cppdefs -Dwaccm_debug -pcols 9" For definitions see arguments to configure


TODO Edit timesteps

Need to see what timesteps have been chosen for the RR model - although it should be enough to just test the main timestep - the physics timestep, right? Everything else changes as a consequence anyway.


TODO Edit user_nl_cam


TODO ./preview_namelists TODO ./preview_run TODO ./case.build TODO ./case.submit