Notes on CICE5 in ESM1.6 #426
Conversation
|
I think i will leave it, the iceberg code is within the cice code, and the coupler page could be more about oasis3-mct |
ccarouge
left a comment
ccarouge
left a comment
There was a problem hiding this comment.
Looks good. I spotted a few typos, nothing major. I would keep the meltwater portion and decide later if it needs to move depending on what documentation gets written.
I would approve but fixing the typos is going to cancel my approval anyway, I believe.
|
For next time, my GitHub handle is @ccarouge , "ClareCR" is the handle of Clare, our previous data manager (avoiding giving full name here). |
oops! - I do this all the time |
Co-authored-by: Claire Carouge <[email protected]>
blimlim
left a comment
blimlim
left a comment
There was a problem hiding this comment.
Thanks @anton-seaice, just a couple small questions
| - Zero-layer thermodynamics (Semtner 1976) | ||
| - One layer of snow and one layer of ice | ||
| - UM calculates ice surface temperature, and conductive heat flux into the sea-ice | ||
| - Ice transport - Lipscomb (2001) and ridging – Rothrock (1975) | ||
| - Internal Ice Stress follow EVP (Hunke and Dukowicz, 2002) |
There was a problem hiding this comment.
Should the references here added to the bibliography?
|
|
||
| ## Meltwater Runoff | ||
|
|
||
| Like ESM1.5, the OASIS3-MCT coupler is used and the sea ice model acts as the interface between the atmosphere and ocean models. The only significant change to this interface since ESM1.5 is changes to meltwater from Antarctica and Greenland. As there is no ice sheet model, the volume of meltwater dischange from Antarctica and Greenland is equal to the instantaneous precipitation over each continent. In ESM1.6, this is partially discharged at the coastline of each continent (to represent ice shelf basal melt) and partially spread in open ocean (to represent melt from icebergs). In ESM1.5 all meltwater is at the coastlines. In addition, the latent heat to melt this water is now taken from the ocean. Meltwater runoff is configured in the `input_ice.nml` [namelist](https://github.com/ACCESS-NRI/access-esm1.6-configs/blob/dev-preindustrial%2Bconcentrations/ice/input_ice.nml#L14-L25) with a presrcibed pattern from the [lice dischange](https://github.com/ACCESS-NRI/access-esm1.6-configs/blob/13cc7d229b0d4bda193879b8b30cde3441d61bec/config.yaml#L98) input file. |
There was a problem hiding this comment.
Just commenting on "the volume of meltwater dischange from Antarctica and Greenland is equal to the instantaneous precipitation over each continent"
Wondering if we need to be 100% accurate or whether it gets too complicated. My understanding is that when the snow depth on a grid point over antarctica/greenland exceeds a limit of 1100m, the excess is fed to the UM's river model and makes its way to the coast. Once it arrives it then gets treated as the meltwater and sent to CICE: https://github.com/CABLE-LSM/CABLE/blob/b2d3f8df5c055a5942f1f5e86288f736582afd74/src/science/soilsnow/cbl_surfbv.F90#L71-L98
There was a problem hiding this comment.
Maybe something like "As there is no ice sheet model, excess snow over Antarctica and Greenland is redistributed as meltwater"?
There was a problem hiding this comment.
ah - we've always said there's no time lag though, so is the routing through the river model only a timestep or two ?
There was a problem hiding this comment.
I also thought the snow wasn't removed, just an equivalent volumne was added as meltwater
There was a problem hiding this comment.
It would be good to document this properly somewhere, it does get discussed often
There was a problem hiding this comment.
I'll double check on these points and get back to you!
There was a problem hiding this comment.
It would be good to get @DaveBi to comment on what you write.
There was a problem hiding this comment.
My understanding is that when the snow depth on a grid point over antarctica/greenland exceeds a limit of 1100m, the excess is fed to the UM's river model and makes its way to the coast.
Ok I think something like this is happening, I tried adding an extra 30m to one of the snow depth variables (s00i240) over the center of Antarctica in the UM restart:
I'm not sure how the snow variables interact, but the added snow makes it's way onto the snow mass variable, which subsides slowly over ~ 7 days.
Plotting the river storage (s26001 - this needed some code changes to get the daily means coming out correctly), the extra snow appears to be fed into the river system, and it takes several days for the water to reach the coast:
Would be good to get confirmation from Dave though
There was a problem hiding this comment.
There may be a timescale in the river scheme @blimlim for runoff you have been looking at water budgets more, we had one way back in CSIRO Mk3 days, especially for large river systems (Congo, Amazon, O (months) so I am sure UM has something too within Jules/Cable code as well to check.
4 participants
Some notes related to #80 . This is a very short summary of the CICE5 configuration for esm1.6, the main reference will be the not-yet-written paper
@ClareCR @blimlim - should we have a seperate page for coupling, and put the meltwater stuff there?