Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

-chem trop_ghg

I could not find the -check trop_ghg configure directive in the config_compsets.xml of cesm1_2_2. Is that still used? I ran it with aquaplanet and I got this:(shr_sys_abort) ERROR: list_init2 ERROR: cannot find physics buffer field sulf

My goal is to have aquaplanet + GHG but CO2 prescribed from a concentration file.What's the -chem option I should start from?Thank you!
 

eaton

CSEG and Liaisons
I reproduced this failure in a "-phys cam4  -chem trop_ghg  -ocn aquaplanet" configuration.  The problem is that build-namelist is not putting the necessary prescribed aerosol settings into the namelist.This can be fixed by modifying line 1021 in
build-namelist (version in cesm1_2_2) from
    if ($chem eq 'none' or $chem eq 'waccm_ghg') {
to
    if ($chem eq 'none' or $chem =~ /_ghg/) {

Alternatively, you could just add the missing prescribed aerosol
specification to your namelist customization.  I found the missing info by
generating the namelist for a default cam4 configuration.

prescribed_aero_cycle_yr=2000
prescribed_aero_datapath='/fs/cgd/csm/inputdata/atm/cam/chem/trop_mozart_aero/aero'
prescribed_aero_file='aero_1.9x2.5_L26_2000clim_c091112.nc'
prescribed_aero_specifier='sulf:SO4','bcar1:CB1','bcar2:CB2','ocar1:OC1','ocar2:OC2','sslt1:SSLT01',
 'sslt2:SSLT02','sslt3:SSLT03','sslt4:SSLT04','dust1:DST01','dust2:DST02','dust3:DST03','dust4:DST04'
prescribed_aero_type ='CYCLICAL'


 
Thank you, Brian.Now it runs but I get this in the atm log: CHECK_VAR INFO: variable CH4 is not on file
 Warning:  Not reading CH4              from IC file.
           CH4              initialized by "chem_init_cnst"
 CHECK_VAR INFO: variable N2O is not on file
 Warning:  Not reading N2O              from IC file.
           N2O              initialized by "chem_init_cnst"
 CHECK_VAR INFO: variable CFC11 is not on file
 Warning:  Not reading CFC11            from IC file.
           CFC11            initialized by "chem_init_cnst"
 CHECK_VAR INFO: variable CFC12 is not on file
 Warning:  Not reading CFC12            from IC file.
           CFC12            initialized by "chem_init_cnst"

I guess the model is not finding an init file.Does it use hardcoded 3D concentrations?I looked at the h1 file produced with rad_data_output=.true.    float rad_CFC11(time, lev, lat, lon) ;
    float rad_CFC12(time, lev, lat, lon) ;
    float rad_CH4(time, lev, lat, lon) ;
    float rad_N2O(time, lev, lat, lon) ;And they seem to have the same 3D pattern but different absolute amounts (attachment)whereas these are spatially constant:    float rad_CO2(time, lev, lat, lon) ;
    float rad_O2(time, lev, lat, lon) ;

I want to prescribe CO2. Do I need to use these? How?
 

eaton

CSEG and Liaisons
The messages from CHECK_VAR are just telling you that the GHGs are not present in the initial file and that they are being initialized by the chem_init_cnst method.  The trop_ghg chemistry is a simple GHG chemistry, so  the constituents CH4, N2O, CFC11, and CFC12 are prognostic.  You see variations in those fields because they're prognostic.  CO2 and O2 are specified constant.  The CO2 values are set by the module ghg_data which uses module chem_surfvals.  I believe that any of the namelist variables available in the chem_surfvals module can be used to set the CO2 values.  Note that that functionality is restricted to specifying a time varying but spatially constant field.  It would require code modification to implement a spatially 3D CO2 field.
 
Thank you, Brian.I get that CO2 values are set by ghg_data using chem_surfvals but, would there be a way to pass CO2 value from file using other chemistry configurations?
 

eaton

CSEG and Liaisons
I'm not sure.  There is a prescribed_ghg module in src/chemistry/utils which appears to handle a CO2 field.  But I don't know how this interacts with the chemistry schemes.  Francis Vitt is the expert on this topic.  From looking at the build-namelist script it appears possible to have it read a single species.  This capability is used to provide CH4 in the super_fast_llnl_* schemes.  I'd have a look at that module.
 
Top