Clarification of meltpond tracers in Icepack

[spencerwong]

Hi everyone,

I'm working on coupling CICE (currently cice6_4_1_20230620) to an atmosphere model. We're using the topo meltpond scheme, and will be exporting the effective pond area `apeffn` and the pond depth `trcrn(:,:,nt_hpnd,:,:)` to the atmosphere. I'm hoping to clarify whether my interpretation of these variables in the Icepack code is correct, to make sure that I'm sending the right thing.

Based on the Icepack documentation, my understanding is that for the topo scheme, the pond fraction tracer `trcrn(:,:,nt_apnd,:,:)` is the fraction of sea ice in each category covered by ponds (and `apeffn` would also be given as a fraction of the ice), and `trcrn(:,:,nt_hpnd,:,:)` is just the raw depth of the ponds.

I might be misinterpreting the discussion here, which mentions that these tracers can be multiplied or divided by the ice fraction `aicen` at different stages, and so I just was hoping to check that my understanding of these tracers based on the documentation is correct.

Many thanks,
Spencer

 

[dbailey]

Hi Spencer,

I think this is best to look at the code. If you look at:

cicecore/cicedyn/analysis/ice_history_pond.F90

this should make it clearer. In terms of the trcrn tracers, the nt_apnd is the total pond area even including snow covered water. So, the total pond water is:

volpn = apnd*hpnd

whereas apeff is only the pond area that shows through and is the actual pond fraction used in the radiation.

 

[spencerwong]

Hi Dbailey, thank you for your reply.

I've taken a look at the ice_history_pond.F90 routines. My understanding based on this is that the trcrn(:,:,nt_apnd,:,:) would be a fraction of the sea ice, and trcrn(:,:,nt_hpnd,:,:) would be the raw depth of the ponds, while the history variable `hpondn` would be the depth averaged over the ice areas.

Let me know if I've made any mistakes!

Thank you,
Spencer

 

[dbailey]

nt_apnd is only the pond fraction. You can see in the ice_history_pond.F90 file how the output is determined. This is because the pond variables have dependences on sea ice fraction and potentially level ice area (when tr_pond_lvl = .true.)

 

[spencerwong]

Hi Dbaily, thank you for your reply! Apologies, I realise I worded my last message quite poorly, and had meant to say "trcrn(:,:,nt_apnd,:,:) would the fraction of the sea ice covered by ponds"

Thank you,
Spencer