How to change CLM50%SP to CLM50%BGC

[Paul]

Hello everyone, I'm using FC2010climo. The land model in this compset is CLM50%SP, and I want to change this to CLM50%BGC without switching to another compset. I've tried appending "- bgc bgc" in the CLM_CONFIG_OPTS, but that doesn't seem to work. Does anyone know how to complete this conversion? Thanks.

 

[oleson]

You could change the long name of that compset in your create_newcase command.
I.e., the long name for FC2010climo in release-cesm2.1.3 is:

2010_CAM60%CCTS_CLM50%SP_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV

To get the BGC model, you can do this:

2010_CAM60%CCTS_CLM50%BGC_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV

or this if you want to include the crop model:

2010_CAM60%CCTS_CLM50%BGC-CROP_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV

I *think* your approach of including "-bgc bgc" would also work, but it needs to be appended to CLM_BLDNML_OPTS, not CLM_CONFIG_OPTS.

 

[Paul]

Thanks for your reply. The create_newcase with long name is worked. I will also try the other method you mentioned.

 

[CGL]

You could change the long name of that compset in your create_newcase command.
I.e., the long name for FC2010climo in release-cesm2.1.3 is:

2010_CAM60%CCTS_CLM50%SP_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV

To get the BGC model, you can do this:

2010_CAM60%CCTS_CLM50%BGC_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV

or this if you want to include the crop model:

2010_CAM60%CCTS_CLM50%BGC-CROP_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV

I *think* your approach of including "-bgc bgc" would also work, but it needs to be appended to CLM_BLDNML_OPTS, not CLM_CONFIG_OPTS.

Hi!@oleson.

I want to know should i change the finitdata or fsurfdata when i change the F2000climo(2000_CAM60_CLM50%SP_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV) to 2000_CAM60_CLM50%BGC_CICE%PRES_DOCN%DOM_MOSART_CISM2%NOEVOLVE_SWAV?

In my points, we use the BGC model means the CESM model simulate the vegetation growth and phenology and the SP model means the fixed phenological periods and growth segments. I already modified the surfdata in Fclimo2000(related to this post) and I'm thinking if I need to change SP to BGC. If I use the SP model, the vegetation growth and phenology for the area i modified will be fix? What's the better choice for the SP or BGC for the modified area?

 

[slevis]

You shouldn't have to change the fsurdat file. It's more subjective whether you should change the finidata file or not. You will have to decide if you like how the model gets initialized with the existing finidat file.

 

[CGL]

You shouldn't have to change the fsurdat file. It's more subjective whether you should change the finidata file or not. You will have to decide if you like how the model gets initialized with the existing finidat file.

I don't get it. My aim is to replace the 2000 land cover data of the Amazon region with the 1850 land cover data. So, I change the fsurdat file.
The finitdat file is the initial field and boundary conditions of the model.The fsurdat file is the landuse type map of the model.Do you mean I should change the finitdat file but fsurdat file? Why not fsurdat file?
Other, SP model or BGC model can use the same finitdat or fsurdat file,as you said?

 

[CGL]

I change the Fclimo2000 SPmodel to BGC model. And i got this error when i set the user_nl_clm and set up the model
err=ERROR : CLM build-namelist::CLMBuildNamelist::setup_logic_initial_conditions() : use_init_interp is set BUT finidat is NOT, need to set both.

I remove the use_init_interp = .true and the case can be setup. I guess if i choose the BGC model, the the vegetation growth gird and phenology will be fixed or depend on the finidat to simulate. So, we don't need to interpolate to other resolution？

 

[slevis]

Sorry if I'm confused and I am confusing you further. I will continue to respond based on my best understanding.

In your last post, it sounds as though you have run successfully with
- bgc
- use_init_interp removed
- finidat removed

Starting this way, without a spun-up finidat file, means that you do not need to interpolate, but that you should spin up the model. This is fine and you will find many posts in these Forums to read about that.

 

[CGL]

Sorry if I'm confused and I am confusing you further. I will continue to respond based on my best understanding.

In your last post, it sounds as though you have run successfully with
- bgc
- use_init_interp removed
- finidat removed

Starting this way, without a spun-up finidat file, means that you do not need to interpolate, but that you should spin up the model. This is fine and you will find many posts in these Forums to read about that.

Thanks,slevis. I appreciate your help :)