Setting up a simulation with constant climate but transient natural forcing

[stevenDH]

Hi All

I've been trying to set up a coupled simulation using CESM2.1.2. with a constant 2015 climate however keeping the natural forcing (i.e. solar and aerosols) transient over time.
I have been encountering some technical issues however when trying to set the solar and aerosol forcing in the namelist files, I added the following lines in the namelist

for solar forcing I add these lines to the namelist:

solar_irrad_data_file = '/scratch/leuven/101/vsc10147/cesm/inputdata/atm/cam/solar/SolarForcingCMIP6_18491230-22991231_c171031.nc'
solar_data_type='SERIAL'
solar_data_ymd=20150101

Which gives the following error when I try to run the code:

solar_data_readnl: ERROR cannot set solar_data_ymd or solar_data_tod with solar _data_type=SERIAL

for aerosol forcing I add these lines to the namelist:

prescribed_strataero_datapath = '/scratch/leuven/101/vsc10147/cesm/inputdata/atm/cam/ozone_strataero'
prescribed_strataero_file = 'ozone_strataero_WACCM_L70_zm5day_18500101-21010201_CMIP6histEnsAvg_SSP370_c190403.nc'
prescribed_strataero_type = 'SERIAL'

Which gives the following error when I try to run the code:

ERROR: trcdata_init: Cannot specify data_cycle_yr if data type is not CYCLICAL

In short my questions pour down to the following:
- Is it possible to do a coupled simulation keeping everything but the natural forcing (aerosols and solar) constant?
- If so, what would be the best approach for this? Starting from a BHIST compset and changing all other forcing to cyclical or rather from a B1850 (or B2000) compset and changing only the natural forcing to serial (this is the approach I am currently trying and that is giving me the errors above)?
- Are the namelist commands I am using above correct or am I missing some dependencies with other namelist variables?

If anyone would have any insights on this or some experience with similar configurations, I would very much appreciate all input!

Cheers
Steven

 

[tilmes]

Hi Steven,
you cannot use these setting together:
solar_data_type='SERIAL'
solar_data_ymd=20150101
solar_data_cycle_yr = ...

as the error suggest. setting solar_data_ymd and *data_cycle_yr requires setting solar_data_type='CYCLICAL' not 'SERIAL'

You can setup a transient model simulation with some cyclical forcings.
For example, keeping aerosols and solar constant requires that you set the following to cyclical:
Vertical emissions:
ext_frc_type = 'CYCLICAL'
ext_frc_cycle_yr = Year you want

However, you have to decide what you do for example for surface emissions, some of them are natural emissions some of them are not.
You cannot set some to cyclical and some to serial, so that makes it difficult.
You can modify emission files so they are still transient (changes with years but have the same emissions every year)
You also have to decide what to do for the lbc, which includes greenhouse gases.

Hope this is helpful?
Simone

 

[stevenDH]

Hey Simone,

Thanks for replying, indeed these are the issues I am facing, I've been trying more things out and the issue I end up which is the following:

I have been using and customising a coupled constant climate compset (B1850 adapted for present day climate), and now I would like to put some of the natural forcing (like solar) transient.
Indeed as you say it should be quite straight forward, I change the solar data type to SERIAL and the solar irrad data file to a transient file without adapting any of the other namelist parameters, however I still end up with the same same error as above (despite not changing any of the namelist variables mentioned in the error).

So thats the main issue that I can't uncover and I hope to get some feedback of the developers through this channel, it should be rather straight forward to change one forcing to be transient in a constant climate compset but it is giving some unexpected errors still.

Cheers
Steven