source_id registration of CAM-MPAS

[beharrop]

label = CAM-MPAS label_extended = CAM MPAS (Community Atmosphere Model – Model for Prediction Across Scales) source_id = CAM-MPAS institution_id = WACCEM-PNNL release_year = 2018 activity_participation = [HighResMIP]

aerosol: description = MACv2-SP prescribed aerosol optical properties and cloud droplet concentration modifiers (same grid as atmos ) nominal_resolution = 25 km atmos: description = CAM-MPAS (CAMv5.4 with Grell-Freitas deep convection; MPASv4, C-grid staggered centroidal Voronoi tesselation atmosphere 30 km mesh; 32 vertical levels) nominal_resolution = 25 km atmosChem: description = none nominal_resolution = 25 km land: description = CLM (v4.0, same grid as atmos), River Transport Model (v1.0) nominal_resolution = 25 km landIce: description = none nominal_resolution = none ocean: description = input4MIPs SSTs and SeaIce for HighResMIP data input nominal_resolution = 25 km ocnBgchem: description = none nominal_resolution = none seaIce: description = input4MIPs SSTs and SeaIce for HighResMIP data input nominal_resolution = 25 km

[durack1]

@beharrop thanks for the submission, normally the "description" fields are used to define the model version, grid and longitude x latitude information - for examples see CMIP6_source_id.html the "description" field is shown under each realm heading, so "ocean" is the ocean, description field for e.g.

An example from ACCESS-ESM1-5 ocean is ACCESS-OM2 (MOM5, tripolar primarily 1deg; 360 x 300 longitude/latitude; 50 levels; top grid cell 0-10 m), there are some other examples that use unstructured grids also contained in the link above

[beharrop]

@durack1 thanks for the clarification. I have modified the descriptors to better match the examples you provided. I am pinging my collaborator @kosaka90 to double check my cell and edge counts for the mesh we used. Does this work, or should I change ocean and ice to something like "prescribed"?

aerosol: description = MACv2-SP (same grid as atmos ) nominal_resolution = 25 km atmos: description = CAM-MPAS (CAMv5.4 with Grell-Freitas deep convection; MPASv4, C-grid staggered centroidal Voronoi tesselation atmosphere 30 km mesh with 655362 cells and 1966080 edges; 32 vertical levels) nominal_resolution = 25 km atmosChem: description = none nominal_resolution = 25 km land: description = CLM (v4.0, same grid as atmos), River Transport Model (v1.0) nominal_resolution = 25 km landIce: description = none nominal_resolution = none ocean: description = none nominal_resolution = 25 km ocnBgchem: description = none nominal_resolution = none seaIce: description = none nominal_resolution = 25 km

[durack1]

@beharrop so I assume that you are not running an interactive 3D ocean or sea-ice for this model? If so, there's no need to define a resolution as there is no model component to describe and these entries should be "none".

For the atmos, what is the atmos upper-most level, so following the ACCESS-ESM1-5 model again we have HadGAM2 (r1.1, N96; 192 x 145 longitude/latitude; 38 levels; top level 39255 m)

[beharrop]

@durack1 that's right, no interactive ocean or sea-ice for our model. I have modified these entries accordingly, as well as added the top level into the atmos entry.

aerosol: description = MACv2-SP (same grid as atmos ) nominal_resolution = 25 km atmos: description = CAM-MPAS (CAMv5.4 with Grell-Freitas deep convection; MPASv4, C-grid staggered centroidal Voronoi tesselation atmosphere 30 km mesh with 655362 cells and 1966080 edges; 32 vertical levels, model top 40363 m) nominal_resolution = 25 km atmosChem: description = none nominal_resolution = 25 km land: description = CLM (v4.0, same grid as atmos), River Transport Model (v1.0) nominal_resolution = 25 km landIce: description = none nominal_resolution = none ocean: description = none nominal_resolution = none ocnBgchem: description = none nominal_resolution = none seaIce: description = none nominal_resolution = none

[durack1]

@beharrop you may want to also check the aerosol and atmosChem fields for tweaks, look like they also need a refresh. The MACv2-SP is one of the input4MIPs datasets, not an atmospheric model component, so that field should also be "none"

[beharrop]

@durack1 Good catch, thanks. I have updated the aerosol and atmosChem fields.

aerosol: description = none nominal_resolution = none atmos: description = CAM-MPAS (CAMv5.4 with Grell-Freitas deep convection; MPASv4, C-grid staggered centroidal Voronoi tesselation atmosphere 30 km mesh with 655362 cells and 1966080 edges; 32 vertical levels, model top 40363 m) nominal_resolution = 25 km atmosChem: description = none nominal_resolution = none land: description = CLM (v4.0, same grid as atmos), River Transport Model (v1.0) nominal_resolution = 25 km landIce: description = none nominal_resolution = none ocean: description = none nominal_resolution = none ocnBgchem: description = none nominal_resolution = none seaIce: description = none nominal_resolution = none

[durack1]

@beharrop all done, please take a peek at CMIP6_source_id.html and let me know of any further tweaks required

[beharrop]

@durack1 This looks great. I do have a question regarding the aerosol portion. I thought you had suggested removing the MACv2-SP specification, but maybe I misunderstood you. I noticed that quite a few models in the source_id link you posted have "prescribed MACv2-SP" for their aerosol component. If this is the appropriate way to specify how aerosols were handled, could we have the same text appear for our aerosol treatment?

[durack1]

@beharrop the registration was from the content in https://github.com/WCRP-CMIP/CMIP6_CVs/issues/913#issuecomment-620250456. As the model doesn't have an interactive aerosol component, I was suggesting leave this as "none", but if you want to follow other model entries we could add none, prescribed MACv2-SP if you like? Let me know

[beharrop]

@durack1 That sounds good to me.

[beharrop]

@durack1 one other thing. @kosaka90 pointed out that many of the models in this list distinguish HR and LR in their source_id names. We are also planning to publish a lower resolution version of the same model. Do you think it would be best to adopt a similar naming convention to other modeling groups? In other words, this entry would be CAM-MPAS-HR and then, presumably, we would open another issue for a CAM-MPAS-LR that is nearly identical except for the grid information.

[durack1]

@beharrop @kosaka90 good call, it's best that we plan ahead as much as possible so that it creates less work for you down the road. I'll update the source_id = CAM-MPAS-HR while updating aerosol too

[durack1]

@beharrop @kosaka90 take a peek at CMIP6_source_id.html which should now be all fixed

[beharrop]

Thanks @durack1 Does it matter that the label and source_id no longer match?

[durack1]

That’s really up to you. The source_id is used as the infrastructure identifier, and labels are intended for use in documentation with no character limits or limitations. Consequently, you could have “CAM MPAS” for a label, with the label_extended a more verbose identifier. If this is your default model, then “CAM-MPAS” is fine, the low res could be “CAM-MPAS (low resolution)” for e.g.

[beharrop]

@durack1 Thanks for the clarification. I think it would be more transparent to use CAM-MPAS-HR for both source_id and label.

[kosaka90]

Thanks, @durack1 . I agree with @beharrop to make source_id and label consistent. Could you do that?

[durack1]

Sure, please take another look over all the entries and confirm the only change is the label update. Once this is confirmed I’ll drop it in the to-do queue

[beharrop]

Looks good to me; do you see anything else @kosaka90. Thanks for all your help with this @durack1 . I hope we haven't been too difficult to work with :)

[kosaka90]

Same here, the only thing to update is the label. Thank you so much for your help, @durack1 !

[durack1]

@beharrop @kosaka90 all done. Take one last peek at CMIP6_source_id.html to double check, but I assume we're all good now

[beharrop]

@durack1 Looks great. Thanks again!

[kosaka90]

Yes, thank you so much @durack1 !