ESCOMP
Remove ERA5 forcing option as it isn't functional
In our XML files we list ERA5 as an option, because it's an option in DATM, but that forcing option isn't valid for CTSM. There was some work at getting it to work for land and ocean, but that didn't pan out so far. Maybe it will be an option in the future, but the current version as non-functional should be removed.
@slevis-lmwg I'd like to assign this to you, and I'm thinking this should be part of your CRUJRA PR, because it helps compensate for the complexity that CRUJRA is adding. This might make more sense, after I finish the review there.
There aren't very many references so easy to do:
cime_config/buildnml: "clm4_5_ERA5": "clm4_5_CRUv7",
cime_config/buildnml: "clm5_0_ERA5": "clm5_0_GSWP3v1",
cime_config/buildnml: "clm6_0_ERA5": "clm6_0_GSWP3v1",
cime_config/config_component.xml: ERA5 (not tuned)
cime_config/config_component.xml: <!-- 10 forcing options for each CLM physics option: CRUv7, GSWP3, cam7.0, cam6.0, cam5.0, cam4.0, QIAN, 1PT, NLDAS2, ERA5 -->
cime_config/config_component.xml: <valid_values>clm5_0_cam6.0,clm5_0_cam7.0,clm5_0_cam5.0,clm5_0_cam4.0,clm5_0_GSWP3v1,clm5_0_CRUv7,clm5_0_QIAN,clm5_0_1PT,clm5_0_NLDAS2,clm5_0_ERA5,clm4_5_CRUv7,clm4_5_GSWP3v1,clm4_5_QIAN,clm4_5_cam6.0,clm4_5_cam7.0,clm4_5_cam5.0,clm4_5_cam4.0,clm4_5_1PT,clm4_5_NLDAS2,clm4_5_ERA5,clm6_0_CRUv7,clm6_0_GSWP3v1,clm6_0_cam6.0,clm6_0_cam7.0,clm6_0_cam5.0,clm6_0_cam4.0,clm6_0_QIAN,clm6_0_1PT,clm6_0_NLDAS2,clm6_0_ERA5</valid_values>
cime_config/config_component.xml: <!-- Options for atm forcing are: CRU, CRUv7, GSWP3, cam6.0 (also used for DATM%CPLHIST), cam5.0, cam4.0, QIAN, WISOQIA, 1PT, NLDAS2, and ERA5) -->
cime_config/config_component.xml: <value compset="DATM%ERA5_CLM45" >clm4_5_ERA5</value>
cime_config/config_component.xml: <value compset="DATM%ERA5_CLM50" >clm5_0_ERA5</value>
cime_config/config_component.xml: <value compset="DATM%ERA5_CLM60" >clm6_0_ERA5</value>
I think I'd prefer to not lump this issue in with the CRUJRA pr to prevent scope creep and to keep prs manageable and focused.
I also wonder if removing ERA5 is prudent at this point, given ESPAT interest in the capability? We can discuss.
I have used it. Currently I am have a regional subset on derecho. For global runs, there is a complication with generating a mesh file due to the treatment of the poles.
On Wed, Mar 12, 2025 at 7:41 PM will wieder @.***> wrote:
@swensosc https://github.com/swensosc have you run with ERA data before?
— Reply to this email directly, view it on GitHub https://github.com/ESCOMP/CTSM/issues/3010#issuecomment-2719530955, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57B45RIVIHCG6N6I5H32UDO3VAVCNFSM6AAAAABY4VSFFOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMJZGUZTAOJVGU . You are receiving this because you were mentioned.Message ID: @.***> [image: wwieder]wwieder left a comment (ESCOMP/CTSM#3010) https://github.com/ESCOMP/CTSM/issues/3010#issuecomment-2719530955
@swensosc https://github.com/swensosc have you run with ERA data before?
— Reply to this email directly, view it on GitHub https://github.com/ESCOMP/CTSM/issues/3010#issuecomment-2719530955, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57B45RIVIHCG6N6I5H32UDO3VAVCNFSM6AAAAABY4VSFFOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMJZGUZTAOJVGU . You are receiving this because you were mentioned.Message ID: @.***>
ERA5 is desired, so we want to keep it around, but will update based on Sean's work so it works correctly.
I created data for era5 that excluded the poles. I think to get it to run properly, i.e. using all the raw data, a 'polemod' mesh would need to be created, similar to what is used for fv09.
On Fri, Mar 14, 2025 at 7:16 PM Erik Kluzek @.***> wrote:
@swensosc https://github.com/swensosc could you point us to a case you have working with ERA5?
— Reply to this email directly, view it on GitHub https://github.com/ESCOMP/CTSM/issues/3010#issuecomment-2726088907, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57E24OKTBN5VVZO4IXD2UN5QPAVCNFSM6AAAAABY4VSFFOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMRWGA4DQOJQG4 . You are receiving this because you were mentioned.Message ID: @.***> [image: ekluzek]ekluzek left a comment (ESCOMP/CTSM#3010) https://github.com/ESCOMP/CTSM/issues/3010#issuecomment-2726088907
@swensosc https://github.com/swensosc could you point us to a case you have working with ERA5?
— Reply to this email directly, view it on GitHub https://github.com/ESCOMP/CTSM/issues/3010#issuecomment-2726088907, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57E24OKTBN5VVZO4IXD2UN5QPAVCNFSM6AAAAABY4VSFFOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMRWGA4DQOJQG4 . You are receiving this because you were mentioned.Message ID: @.***>
@ekluzek / @swensosc - Wondering if there's an update on this? Finally looking to start some ESPAT runs that use ERA5 forcing to spin up offline CLM. Do we have the capability to do an offline CLM in CESM2.2 with ERA5 forcing in DATM (and do the poles work okay now)? Or was the complication with ERA5 only introduced in CTSM6 development and this isn't an issue anyway?
@ekluzek / @swensosc - Wondering if there's an update on this? Finally looking to start some ESPAT runs that use ERA5 forcing to spin up offline CLM. Do we have the capability to do an offline CLM in CESM2.2 with ERA5 forcing in DATM (and do the poles work okay now)? Or was the complication with ERA5 only introduced in CTSM6 development and this isn't an issue anyway?
Well #3588 is about bringing in updated ERA5 forcing and to get it working in the latest model. So that's the issue you should follow.
We didn't previously get this working for CLM in CESM2.2. It was just something we wanted to try. There has been an option to use ERA5 forcing back then, but it only worked with the ocean model and not the land model. That's where this issue started when we realized that configuration wouldn't work.
Thanks Erik! I've been tracking the issue but was hopeful for a second that I'd just misunderstood the original issue! Thanks for clarifying.