Update from 26 August

[matthew-england-unsw]

Great to catch up today.

The plot thickens (in terms of uncovering the mechanism that gives rise to the DSW formation anomalies, along with what yields such a consistent circumpolar response).

We discussed a few possible additional experiments: one is re-running the same expt, only with wind anomalies applied only over the DSW regions. i.e. the opposite of the recent new expt. The other was to apply the full circumpolar wind anomalies only during key seasons - e.g. just during the sea-ice growth season (MJJ) or just during the DSW formation season (ASO).

Consensus seems to have built around one of two possible drivers, although these are by no means mutually exclusive. One is the 'Adele' mechanism, originating from up/downwelling anomalies along the coastal fringe. This drives stratification anomalies that advect into the DSW regions. The other is the 'Matt' mechanism, where the dominant driver is wind-driven ice advection creating sea-ice and FW flux anomalies (combined to a lesser extent with resultant air-sea E-P fluxes). Could of course be both.

In both mechanisms, there's a view that the sea-ice growth and cold (DSW formation) seasons hold the key. Hence the experiments mentioned above.

The summer peak in heat and FW fluxes that Andy recently showed appears to be second order to sea-ice related FW fluxes, perhaps setting up summertime stratification anomalies that enable the DSW and MLD anomalies in winter to play out, but this seems second-order to the above cold-season mechanisms.

Amplifying feedbacks seem possible / likely. e.g. an initial broader circumpolar response might drive the initial DSW anomalies, although the DSW anomalies can then re-enforce the original circumpolar anomalies by influencing CDW transport onto the shelf.

A late night bleary eyed recollection, so please add to as you see fit. :-)

[StephenGriffies]

Thanks for this summary @matthew-england-unsw . Sorry I missed the gathering...I listened to 90 minutes of Zoom from my son's high school admin explaining how great his high school experience will be. Fingers crossed they are correct!

About the possible new experiments: A/ re-running the same expt, only with wind anomalies applied only over the DSW regions. i.e. the opposite of the recent new expt. B/ Apply the full circumpolar wind anomalies only during the sea-ice growth season (MJJ) C/ Apply the full circumpolar wind anomalies only during the DSW formation season (ASO).

What are the chances all three could be run? I suspect we will get useful information from each.

[adele-morrison]

Thanks for the summary Matt. Looks like Wilma also put a summary on the Results Summary page (I made a few edits too), thanks Wilma! Just a note with the upwelling mechanism, I don't think it's driven by the upwelling/downwelling on the coastal fringe, but rather the large scale upwelling everywhere except the coastal fringe. In the UP perturbation, there is a strong downwelling along a very narrow coastal strip, but everywhere else there is upwelling. The hypothesis is that the lower overturning cell could be driven by the increase in wind-driven upwelling (i.e. how we commonly think about the upper overturning cell), rather than being driven by the local DSW formation. When the upwelling increases, the DSW formation of course increases in response due to the enhanced heat loss with more CDW on the shelf. But the DSW formation response is not the initial driver. This is exactly what was found by Stewart and Thompson 2012 in their idealised setup.

[adele-morrison]

For experiment C, I was thinking the opposite of the above, to only apply the wind anomalies outside the winter season, i.e. October to June only. Then we could see if the change in DSW formation still occurs driven by changes in upwelling or sea ice advection even with no forcing change applied directly in winter. Though first we could check if the upwelling (CDW transport across 1000m isobath) changes in the first 6 months of the UP simulation before the first winter season.

[matthew-england-unsw]

Oops I didn’t see Wilma’s post apologies. And thanks for clarifying the upwelling mechanism you’ve proposed. Kind of like Morrison et al 2020 then - only instead of DSW formation driving CDW on-shelf flow, the other way around?

[matthew-england-unsw]

Yes agreed — we’d likely gain new insight with the “all seasons bar months XYZ” approach. Might indeed be cleaner. In your 2020 paper you showed that the onshore CDW / offshore DSW operated on fast time-scales. So maybe zero-ing out the wind anomaly during DSW formation months will be enough to kill off the DSW response?

[PaulSpence]

Hi Folks,

Thanks for the summaries. I am starting to like the seasonal wind perturbation ideas. A DSW water focussed perturbation (the inverse pattern) might need to be increased in amplitude to reveal much.

Re from summmary page: "The easterly winds drive large-scale upwelling via Ekman pumping over nearly all the Antarctic margins (apart from a narrow coastal downward Ekman pumping)."

I will plot up some other regional plots, but I think in much of E. Antarctica, the downwelling often spans the narrower shelf regions ( https://user-images.githubusercontent.com/6644956/125725593-2a8604cc-6072-422f-9a00-89ff10256eb5.png ).

Thank you, Paul

On Fri, Aug 27, 2021 at 7:46 AM Matthew England @.***> wrote:

Oops I didn’t see Wilma’s post apologies. And thanks for clarifying the upwelling mechanism you’ve proposed. Kind of like Morrison et al 2020 then

• only instead of DSW formation driving CDW on-shelf flow, the other way around?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/adele157/easterlies-collaborative-project/issues/35#issuecomment-906764946, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABSWJXCXNVVPZF5HGX3Y6MTT62Y2XANCNFSM5C3NC7PA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

-- Paul Spence, PhD ARC Future Fellow School of Geosciences University of Sydney, Australia https://paulspence.github.io/

[matthew-england-unsw]

Actually (responding to myself 🤦‍♂️) it’s hard to imagine this occurring on such a fast time scale. The DSW push driving CDW pull can be fast, but the broader circumpolar CDW upwell is likely to take some time (and of course suitable WMT) to drive enhanced DSW flow offshore. Fascinating stuff.....

[StephenGriffies]

In response to @matthew-england-unsw saying it is tough to imagine a fast baroclinic response, here is one mechanism to remember.

Namely, arrested Ekman layer mechanism can lead to nearly instantaneous baroclinic adjustment along the sloping bottom. We invoked that mechanism in Spence et al 2017 and Webb et al 2019. The time scale for upwelling or downwelling along the bottom goes like tau ~ |f|/(S*N)^2, where S=topographic slope and N=ambient stratification. Wahlin et al (2010) find tau on the order of a few hours in her analysis of Amundsen measurements. That is incredibly fast for a baroclinic adjustment.

So with the winds sending barotropic waves around the continent, creating barotropic pressure gradients, there could be fast baroclinic responses up or down the slopes depending on the ambient stratification, barotropic pressure gradient direction, and direction of the wind.

[matthew-england-unsw]

Thanks for the reminder @StephenGriffies - I am perhaps too wedded to thinking of DSW as occurring in localized hot-spots (or I should say cold-saline spots) due to local buoyancy fluxes. But the experiment run recently with wind anomalies zero-ed out in those hotspots is a good reminder that it's more than just that - part of the process is what lies deeper in the water column, in terms of allowing top-to-bottom convection over the shelf.