Question about parameter settings in CESM-BGC (atm_alt_co2_const and atm_co2_const)

[Redlichia]

Hello, everyone！Thanks for opening this thread. I am running fully coupled climate simulation (use B1850) with CESM1.3. The ocean tracer module is set to "iage ecosys" for this simulation.
--------------------------------------------------------------------------
In the simulation for the PI period, the atmospheric CO2 concentration is 284.7 ppm. correspondingly, in the settings of &ecosys_nml, atm_alt_co2_const = 284.7; atm_alt_co2_opt = 'const'; atm_co2_const = 284.7; atm_co2_opt = 'const'.
Both parameters (atm_co2_const and atm_alt_co2_const) are set to the default atmospheric CO2 concentration for the PI period. I have completed the simulation with the default 284.7 ppm atmospheric CO2 concentration.
I would like to test the changes in the ocean biogeochemical field (CESM-BGC) for future scenarios with doubled CO2 x2 (284.7x2) and quadrupled CO2 x4 (284.7x4). If I am going to do both tests reasonably well. About how the following two parameters in the user_nl_pop2 should be set:

atm_alt_co2_const = 284.7
atm_alt_co2_opt = 'const'
atm_co2_const = 284.7
atm_co2_opt = 'const'
Is it possible to double CO2 x2 (284.7x2) and quadruple CO2 x4 (284.7x4) these two parameters? What is the difference between the settings of these two parameters (atm_co2_const and atm_alt_co2_const) and what do they each mean? How should I set these two parameters in my experiment?

--------------------------------------------------------------------------
I'm a little confused because I am a rookie about CESM-BGC.

I hope to get your help. I would really appreciate your help!

Thank you!


Best,Chen

 

[mlevy]

Hi Chen,

The BGC tracer package in CESM 1.3 has evolved into MARBL, and the best documentation available is Long et al. (2021). That said, I can give a general answer to your questions:

The BGC system has two carbon chemistry systems running simultaneously. The main one (controlled by atm_co2_opt) feeds back to the rest of the system, while the alternate system (controlled by atm_alt_co2_opt) is only used to evolve DIC_ALT_CO2 and ALK_ALT_CO2. By comparing these tracers to DIC and ALK, you can look at the anthropogenic oceanic carbon with a single run. For your specific experiments, you could run two cases - one with

atm_co2_const = 569.4 ! 284.7*2
atm_alt_co2_const = 284.7

and one with

atm_co2_const = 1138.8 ! 284.7*4
atm_alt_co2_const = 284.7

In both of these cases, the full model will progress using the higher atmospheric co2 concentration, and the difference DIC - DIC_ALT_CO2 is a reasonable approximation of the anthropogenic carbon. From Section 4.3 of Long et al.

Note that this definition differs subtly from subtracting a DIC field obtained from an 1850-control integration: changes in climate impact both DIC and DIC_ALT_CO2, so the resulting [anthropogenic carbon] does not include the impact of climate on natural CO2 as it would if the baseline DIC field were taken from an 1850-control integration. While climate impacts on natural CO2 are significant under future scenarios with strong radiative forcing, this feedback is modest over the historical period

If you've already done a B1850 control run with atm_co2_const = 284.7, you should just use that as your baseline for anthropogenic carbon and atm_alt_co2_const isn't as important in your experiments. (Also, you'll need the control run to provide a baseline for variables outside the carbonate system.)

I hope this helps, I may have gone off on a tangent without actually answering your question...
~Mike

 

[Redlichia]

Hi Chen,

The BGC tracer package in CESM 1.3 has evolved into MARBL, and the best documentation available is Long et al. (2021). That said, I can give a general answer to your questions:

The BGC system has two carbon chemistry systems running simultaneously. The main one (controlled by atm_co2_opt) feeds back to the rest of the system, while the alternate system (controlled by atm_alt_co2_opt) is only used to evolve DIC_ALT_CO2 and ALK_ALT_CO2. By comparing these tracers to DIC and ALK, you can look at the anthropogenic oceanic carbon with a single run. For your specific experiments, you could run two cases - one with

atm_co2_const = 569.4 ! 284.7*2
atm_alt_co2_const = 284.7

and one with

atm_co2_const = 1138.8 ! 284.7*4
atm_alt_co2_const = 284.7

In both of these cases, the full model will progress using the higher atmospheric co2 concentration, and the difference DIC - DIC_ALT_CO2 is a reasonable approximation of the anthropogenic carbon. From Section 4.3 of Long et al.



If you've already done a B1850 control run with atm_co2_const = 284.7, you should just use that as your baseline for anthropogenic carbon and atm_alt_co2_const isn't as important in your experiments. (Also, you'll need the control run to provide a baseline for variables outside the carbonate system.)

I hope this helps, I may have gone off on a tangent without actually answering your question...
~Mike

Hi！Michael. Thank you for your patient reply!

As you suggested, in addition to setting co2vmr = 569.4ppm (CO2 x2) and co2vmr = 1138.8ppm (CO2 x4) in the user_nl_cam file.

I will add the following line to the user_nl_pop2 file for each of the two experiments.

In the double CO2 x2 (284.7x2) experiment:
atm_co2_const=569.4

In quadruple CO2 x4 (284.7x4) experiment:
atm_co2_const=1138.8

Thank you for the help you provided, it was very useful to me.

Do you think this setup method can be applied to palaeoclimate scenarios? Because we know there were many high CO2 climate scenarios during the paleoclimate period?

Thank you! I'm excited to get a response from you！


Best,Chen

 

[mlevy]

Do you think this setup method can be applied to palaeoclimate scenarios? Because we know there were many high CO2 climate scenarios during the paleoclimate period?

Yup, the same namelist variables will control atmospheric CO2 in the paleo runs as well

 

[Redlichia]

Yup, the same namelist variables will control atmospheric CO2 in the paleo runs as well

Thanks for your help, Mike.

Best,
Chen

 

[zhangheng]

Hi Chen,

The BGC tracer package in CESM 1.3 has evolved into MARBL, and the best documentation available is Long et al. (2021). That said, I can give a general answer to your questions:

The BGC system has two carbon chemistry systems running simultaneously. The main one (controlled by atm_co2_opt) feeds back to the rest of the system, while the alternate system (controlled by atm_alt_co2_opt) is only used to evolve DIC_ALT_CO2 and ALK_ALT_CO2. By comparing these tracers to DIC and ALK, you can look at the anthropogenic oceanic carbon with a single run. For your specific experiments, you could run two cases - one with

atm_co2_const = 569.4 ! 284.7*2
atm_alt_co2_const = 284.7

and one with

atm_co2_const = 1138.8 ! 284.7*4
atm_alt_co2_const = 284.7

In both of these cases, the full model will progress using the higher atmospheric co2 concentration, and the difference DIC - DIC_ALT_CO2 is a reasonable approximation of the anthropogenic carbon. From Section 4.3 of Long et al.



If you've already done a B1850 control run with atm_co2_const = 284.7, you should just use that as your baseline for anthropogenic carbon and atm_alt_co2_const isn't as important in your experiments. (Also, you'll need the control run to provide a baseline for variables outside the carbonate system.)

I hope this helps, I may have gone off on a tangent without actually answering your question...
~Mike

Hi Levy, is it possible to config a well mixed atmospheric CO2 model with the G1850ECO compset? If so, I can analyze the individual impact of marine biogeochemical activities on the atmospheric CO2 content.
Because I noticed the `box_atm_co2` option for the `OCN_CO2_TYPE` in env_run.xml, it seems like a box atmospheric CO2 configuration method.