Main menu

Navigation

Aquaplanets with slab ocean (by Brian Medeiros 2013)

6 posts / 0 new
Last post
valeriarudloff@...
Aquaplanets with slab ocean (by Brian Medeiros 2013)

Hello,

 

i'm working on an Aquaplanet with slab ocean, and i'm following the instructions of Madeiros (http://www.cesm.ucar.edu/models/simpler-models/aquaplanet.html).

Everything has been well documented until i started to work with the inputdata related to qfluxes, SST and depth ocean.

 

That file (tipically for D and E compset, named as pop_frac.1x1d.090130.nc) has those variables, but normally that information is written for a different grid... i mean, for a grid different to f19_f19, which one is used in the Madeiros's Instructions.

 

May i create the file, using the same grid size of other input files that have been changed (particularly, domain.ocn.1.9x2.5_gx1v6_090403.nc)?


Is there someone else who create this file?

 

Regards

brianpm

This is an old question, but I wanted to let people know that the released CESM2 has a standard aquaplanet with slab-ocean. It can be run using the compsets called:

QSC4, QSC5, QSC6

where the number indicates the CAM physics version. 

 

The files that are needed for the slab-ocean file are a domain file and a forcing file. Examples can be seen in the CESM inputdata:

forcing: /glade/p/cesmdata/cseg/inputdata/ocn/docn7/SOM/default.som.forcing.aquaplanet.Qflux0_h30_sstQOBS.1degFV_c20170421.nc

domain: /glade/p/cesmdata/cseg/inputdata/share/domains/domain.ocn.fv0.9x1.25_gx1v7.151020.nc

But specifying the domain file is no longer necessary in CESM2 if running the compsets listed above.

.brian
_________
Brian Medeiros
Project Scientist
NCAR/CGD/AMP

mpb20@...

Hi all,

 

I'm working on getting a slab-ocean aquaplanet version of cesm1_2_2 (CAM4) running on the UK's ARCHER supercomputer, following Brian Medeiros' instructions (http://www.cesm.ucar.edu/models/simpler-models/files/CAM_aqua_som_descri...).

The model compiles ok but I get a runtime error related to initialising the solar radiation calculation:

(GETFIL): attempting to find local file abs_ems_factors_fastvx.c030508.nc
(GETFIL): using /work/ecse0116/shared/CESM1.0/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc
(shr_sys_abort) ERROR: rad_solar_var_init: solar input file must have reference irradiance spectrum
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping

 

The input file (abs_ems_factors_fastvx.c030508.nc) does exist at the specified location, but doesn't seem to contain "reference irradiance spectrum" data. Should rad_solar_var.F90 even be called for this simple perpetual equinox case? Any suggestions would be gratefully received, the ATM ans CESM log files are attached.

Cheers,

Mike

 

Dr Mike Byrne
Lecturer (University of St Andrews)
Marie Curie Research Fellow (University of Oxford)

mpb20@...

Update: The model is looking for "reference irradiance spectrum" data in the solar input file, which in my case is ape_solar_ave_tsi_1365.nc. However this input file is based on the Aqua-Planet Experiment design, and does not include a reference irradiance spectrum. Has anyone running aquaplanets encountered this issue?

Dr Mike Byrne
Lecturer (University of St Andrews)
Marie Curie Research Fellow (University of Oxford)

brianpm

Hi Mike,

I haven't hit that error, but I have run CAM4 aquaplanets with ape_solar_ave_tsi_1365.nc many times. I suspect that some logic has been triggered that isn't usually. 

I just set up a fixed-SST CAM4 aquaplanet following the instructions at http://www.cesm.ucar.edu/models/simpler-models/fixed_sst_aqua.html#cam4

In my atm_in namelist, I see:

absems_data  = '/glade/p/cesmdata/cseg/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'

and also:

radiation_scheme = 'camrt'
solar_const = 1365.0
solar_htng_spctrl_scl = .false.

Is your run also setting these values? 

.brian
_________
Brian Medeiros
Project Scientist
NCAR/CGD/AMP

mpb20@...

Thanks Brian for the help! Comparing to your fixed-SST CAM4 case, in my atm_in file I have the same absems_data input file specified:

absems_data = '/work/ecse0116/shared/CESM1.0/inputdata/atm/cam/rad/abs_ems_factors_fastvx.c030508.nc'

But the solar_const is not specified, and in solar_inparm I have:

&solar_inparm

  solar_data_file                = '/work/ecse0116/shared/CESM1.0/inputdata/atm/cam/solar/ape_solar_ave_tsi_1365.nc'

  solar_htng_spctrl_scl          = .true.

/

So for some reason in the CAM4 slab aquaplanet setup that I have tried, the solar radiation is not being setup correctly. I based my build on your notes for a CAM5 slab-ocean aquaplanet (http://www.cesm.ucar.edu/models/simpler-models/files/CAM_aqua_som_descri...), so maybe it's a CAM4/5 issue? One idea would be to invoke the aquaplanet use case when I build, though your notes suggests this creates other issues when setting up the SOM

Dr Mike Byrne
Lecturer (University of St Andrews)
Marie Curie Research Fellow (University of Oxford)

Log in or register to post comments

Who's new

  • lwheele@...
  • afroberts@...
  • richard.allard@...
  • tracy@...
  • 498749303@...