Cooling Antarctica by increasing snow albedo

[wwieder]

@gustavo-marques and @dlawrenncar asked if we could cool Antarctica by increasing land albedo to improve Antarctic deep water formation and reduce Southern Ocean warm biases in recent CESM dev runs.

@slevis-lmwg cloned 315 and started a new run on LMWG-dev that applies spatially distributed land parameters enabled by PR from @swensosc. From discussions with @olyson, we opted to reduce the parameter xdrdt from 5 to 1 for latitudes south of -60. In land only runs we'd expect a 3-5% increase albedo and cool temperatures, but the magnitude of this change in coupled simulations hasn't been evaluated.

Preliminary results suggest we are increasing albedo and may be cooling temperatures, but the run is short (just 3 yeras) and not for an overlapping time period (the new run with adjusted albedo was a hybrid from year 100 of the simulation in 315). Full LDF for years 1-3 in the test case vs years 80-99 of the base case is here, with a snapshot of TSA for DJF below

Results seem promising and I've asked @slevis-lmwg to continue this run.

I also wonder if @cecilehannay could continue # 315 for another 10 years so we have a cleaner comparison?

sI'll also need additional diagnostics from other components to help evaluate if this is something we want to enable for CESM development runs.

[adamrher]

What's the problem exactly? That there's too much surface melt and therefore liquid runoff in Antarctica?

[dlawrenncar]

The goal, as far as I understand it, is to cool the Antartic region generally, and particualrly the surface near coastal Antarctic ocean, which may help reduce the ACC oscillations further. This is what the GFDL did, apparently, to reduce their oscillations.

…

On Thu, Mar 12, 2026 at 10:12 AM Adam Herrington ***@***.***> wrote: What's the problem exactly? That there's too much surface melt and therefore liquid runoff in Antarctica? — Reply to this email directly, view it on GitHub <#302 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFABYVDS7UGXVMQJCY5EGBT4QIFGPAVCNFSM6AAAAACWOXMYCSVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTMMBZGA3TAOA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[adamrher]

OK. I'm still not understanding the mechanism where cooling leads to reduced ACC oscillations, but I would imagine that increasing sea-ice albedo would also be effective at lowering coastal temperatures, considering sea-ice hugs the coasts.

[wwieder]

A cleaner comparison of overlapping years is now here. I'm waiting on the base case to extend further, as we only have 4 overlapping years right now, but DJF cooling looks similar (below),

Although annual TSA shows pretty subtle differences:

[wwieder]

Final land diags are here, with 12 overlapping years from the test case with higher Antarctic albedo with the default case from #_315. Climatology plots (maps) are for the last 7 years.

Thanks to @slevis-lmwg & @cecilehannay for doing / continuing these runs.

Albedo changes look as expected (here for DJF), but overall effect on temperatures seems subtle. Maybe looking at other components would be helpful to see if this has any desired impacts on ocean circulation, @gustavo-marques, or sea ice ( @dabail10 @duvivier)?

Note, I won't be at the CESM project meeting this week, but if this capability to support spatially distributed parameter is something we want in the CESM3 simulations let the LWMG know ASAP so we can move it up the SE queue.

[gustavo-marques]

Thanks, @wwieder. Below is the slide I added for tomorrow. Overall, increasing snow albedo cools the offshore SSTs around Antarctica. The run is too short to determine whether this affects the MOC AABW cell. It will be good to see how the sea ice compares in these runs.

[wwieder]

If you want this run extended please ask Sam L and Cecile

[slevis-lmwg]

@gustavo-marques @dlawrenncar
When @cecilehannay starts the pre-spinup with alpha08k or alpha08l OR when she starts the actual spinup with beta08, I could see myself starting a parallel case with the slightly cooler Antarctica. As Gustavo showed, it may take multiple decades for the cooler Antarctic to have a meaningful effect on the ocean. I'm not suggesting tweaking Antarctica's albedo further, though this is also still an option. In any case, I will not start a simulation unless I hear back.

[dlawrenncar]

@slevis-lmwg The albedo changes look fairly modest, on order of 1%. Am I seeing that right and is that what we would expect. From what I have seen, I don't think a new run with this albedo setting is worthwhile. The impact seems quite small. It might be worth trying an experiment with a stronger change to albedo, but overall, I don't think it is very likely that we will pursue this path to solve the problem.

[olyson]

Here, we changed xdrdt (a multiplicative factor on the snow aging rate) from 5 globally to 1 (just over Antarctica). Based on previous land-only simulations (which implemented changes in the opposite direction, changing xdrdt from 1 to 5 globally), I was expecting a bit larger changes than we see in this simulation. Here are the land-only results for ANN albedo:

https://webext.cgd.ucar.edu/I1850/ctsm51_ctsm51d150_ne30pg3ne30pg3mg17_GSWP3V1_ABCDE_blk_A5BCD_1850AD/lnd/ctsm51_ctsm51d150_ne30pg3ne30pg3mg17_GSWP3V1_ABCDE_blk_A5BCD_1850AD.271_280-ctsm51_ctsm51d150_ne30pg3ne30pg3mg17_GSWP3V1_ABCDE_blk_ABCD_1850AD.271_280/set2/set2_ANN_ASA.png

The albedo changes do look larger in DJF than ANN in the coupled simulation (Will's plot above, maybe 2-5% along some of the coastlines), as they also do in the land-only simulation:

https://webext.cgd.ucar.edu/I1850/ctsm51_ctsm51d150_ne30pg3ne30pg3mg17_GSWP3V1_ABCDE_blk_A5BCD_1850AD/lnd/ctsm51_ctsm51d150_ne30pg3ne30pg3mg17_GSWP3V1_ABCDE_blk_A5BCD_1850AD.271_280-ctsm51_ctsm51d150_ne30pg3ne30pg3mg17_GSWP3V1_ABCDE_blk_ABCD_1850AD.271_280/set2/set2_DJF_ASA.png

Maybe the snowpack characteristics or the precipitation frequency/amount are different in the coupled simulations, not sure.
I'm not sure it is worth pursuing this either, however, here are some thoughts on this that I had sent to @wwieder :

Regarding increasing albedo, our best bets are probably xdrdt and upplim_destruct_metamorph. Increasing xdrdt from 1 to 5 decreased albedo by maybe 2-4% over Antarctica so reversing that would hopefully result in an increase by that amount. Decreasing upplim_destruct_metamorph from 250 to 100 decreased albedo by 1-2%, so maybe we could go to 400 to increase albedo (we are at 250 now). fresh_snw_rds_max is also a possibility, it is now 400 and an increase from about 204 to 400 decreased albedo by 1-2% also, so we could try going back down to 200 or so. Not clear that the combined effects of these would be a simple sum, there must be some kind of upper limit on fresh snow albedo. Maybe first try a couple of these in land-only mode to make sure we get the magnitude and direction we are expecting in Antarctica.

[slevis-lmwg]

(I hadn't seen Keith's response when I posted mine.)

Sounds good.

Regarding the 1% change, yes, that's correct; however, it may be harder to determine what to expect:
We got the change by reducing xdrdt from 5 to 1 south of -60. This parameter is an arbitrary factor applied to the snow aging rate and in particular we multiply xdrdt by the rate of change of the snow's effective radius. The latter (effective radius) then affects the snow albedo...