Can Parallel Offline Radiative Transfer software tool be used for CESM1.2.2?

[11438023@zju_edu_cn]

Dear all,

I am running CESM1.2.2 (with CAM5) and have two 20-year transient simulations:

1) Default historical simulation ranging from 1870-1890;
2) Same as 1) except that we added a fixed amount of additional aerosol emissions from 1870 to 1890.

We would like to know how much radiative forcing is produced during this period by these additional aerosols. One way is to use the Parallel Offline Radiative Transfer software tool. However, based on this online document (PORT - PORT - Parallel Offline Radiative Transfer Tool for CAM - wiki.ucar.edu), it seems that PORT is not supported for cesm1.2.2.

I am wondering if there is any way to calculate the radiative forcing? Is PORT updated to support CESM1.2.2 with CAM5 now?

Thanks very much!

Best wishes,
Lei

 

[andrew]

I would suggest using CESM2 with the CAM5 atmosphere, and run PORT there: PORT is supported in CESM2 and should run with the -phys cam5 option to the cam configure command. I am not sure PORT is supported in CESM1.2.2.

This should get you close.

Dear all,

I am running CESM1.2.2 (with CAM5) and have two 20-year transient simulations:

1) Default historical simulation ranging from 1870-1890;
2) Same as 1) except that we added a fixed amount of additional aerosol emissions from 1870 to 1890.

We would like to know how much radiative forcing is produced during this period by these additional aerosols. One way is to use the Parallel Offline Radiative Transfer software tool. However, based on this online document (PORT - PORT - Parallel Offline Radiative Transfer Tool for CAM - wiki.ucar.edu), it seems that PORT is not supported for cesm1.2.2.

I am wondering if there is any way to calculate the radiative forcing? Is PORT updated to support CESM1.2.2 with CAM5 now?

Thanks very much!

Best wishes,
Lei

 

[11438023@zju_edu_cn]

Thanks, Andrew. It's good to know CESM2 supports PORT calculations.

Since all our simulations including future RCP simulations (which took several months to finish) used CESM1.2.2, it would be good if we can stick with it for now.

But I will try CESM2 for historical simulations and see if it can run RCP scenarios.

Thanks again.

 

[gchiodo]

Dear all, I have the exact same issue as Lei: I'd like to run offline PORT experiments with CAM5, but have not bee able to do this in the CESM1.x distributions (PORT only works with CAM4 physics there).

I can confirm that in CESM2, PORT works with CAM5 physics via the "PC5" compset. However, the problem with PORT_CAM5 in CESM2 is that there are virtually no CAM5 compsets available this code distribution, to produce the corresponding "base run" whose radiation data is needed to drive PORT.

The only way to use CAM5 in CESM2 is by setting up a present-day CAM6 experiment (F2000climo compset) and overwriting its env_run.xml settings with "-phys cam5". However, this will *not* produce the CAM5 climate that one would obtain from this same model, using the CESM1.2.2 distribution instead, since lots of tuning parameters for the cloud parameterizations are still optimized for CAM6. I have been trying to "revert" CAM6 to CAM5 within CESM2, but without success.

I am afraid this limits the usability of PORT in CESM2 for radiation schemes that are different from CAM6... and posted about it on the CAM section of the BB forum: https://bb.cgd.ucar.edu/cesm/threads/running-cam5-in-cesm2-1-3.5416/

The two possible solutions are: (1) create some CAM5 compset within CESM2, which truly reproduces the CAM5 climate in CESM1.x, or (2) make the PORT flag work for RRTMG in any CESM1.2.x distribution. Do you happen to know if the latter has been done for some unreleased alpha/beta distribution of CESM1.2.x, e.g. to allow the production of the CAM5 kernels (see paper)?

Best regards,

Gabriel Chiodo

 

[andrew]

Hi Gabriel,

I sympathize with your plight.

I do not know if anyone has gotten PORT to work in CESM1.2.X. We can ask around about that (your option 2).

Regarding #1, our policy has been that we are not trying to reproduce the CESM1-CAM5 climate when you run CESM2-CAM5. As you said, you could try by re-setting all the namelist settings. But I am not sure anyone has done this.

Note that you should be able to run almost any CESM2 experiment with CAM5, even if we don't have a 'comp set'. The comp sets map to a 'long name' which tells you which components are being used. You can copy this name from any compset definition and change the atmosphere to CAM5. I forget exactly the syntax, but this should be in the user's guide.

Hope that helps.

Cheers,

Andrew