search

adele-morrison / easterlies-collaborative-project

7 stars 3 forks source link

Cross-isobath heat transport #40

Closed wghuneke closed 2 years ago

wghuneke commented 3 years ago

Plot cross-isobath (1000-m isobath) heat transport for year 1 to check for increased onshore heat transport in UP, which is possibly related to widespread upwelling on the continental shelf.

wghuneke commented 3 years ago

The total heat transport is calculated as the sum of cross-isobath heat transport and zonal heat transport. We have to take zonal heat transport into account to get the total (net) heat flux as the ASC might carry heat across the isobath multiple times and we don't want to account for it multiple times.

Sign convention: -ve heat flux is onshore.

Fig_total_heat_transport_1000m_isobath_hovmoeller

wghuneke commented 3 years ago

The heat flux is pretty noisy (see hovmoeller diagram above), but the circumpolar sum for each month (Figure below, panel a) tells us that there is more onshore heat transport in UP, mostly in spring/summer. Panel b shows the circumpolar cumulative sum for the annual averaged total heat transport, showing again the increased heat transport in UP.

Updated: added DOWN case

Fig_total_heat_transport_1000m_isobath

PaulSpence commented 3 years ago

Hi Wilma,

Can you please push that cross isobath code to your github branch so that I can play with it re sea ice transports?

Thank you, Paul

On Fri, Sep 10, 2021 at 10:13 AM Wilma Huneke @.***> wrote:

The heat flux is pretty noisy (see hovmoeller diagram above), but the circumpolar sum for each month (Figure below, panel a) tells us that there is more onshore heat transport in UP, mostly in spring/summer. Panel b shows the circumpolar cumulative sum for the annual averaged total heat transport, showing again the increased heat transport in UP.

[image: Fig_total_heat_transport_1000m_isobath] https://user-images.githubusercontent.com/15355753/132778520-28c7dc49-b6b9-4c17-9f8f-1f9bfc35d765.png

— 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/40#issuecomment-916526897, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABSWJXHEQBEFVYQSQGI5TIDUBFETHANCNFSM5DYHFTAA . 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, Assoc. Prof. ARC Future Fellow Institute for Marine and Antarctic Studies https://www.imas.utas.edu.au/ University of Tasmania, Hobart, Australia https://paulspence.github.io/

adele-morrison commented 3 years ago

Wilma, these numbers seem large, it should be ~40TW for the total circumpolar heat transport. Is there a factor of 10 missing somewhere? Good to see it increases though. Is the DOWN perturbation symmetric in the time series? The monthly plot might be better as anomalies from the control given the large seasonality there?

wghuneke commented 3 years ago

@PaulSpence : I pushed the code to my branch.

@adele157 : that's probably because I didn't "scale" the heat transport for the binned product. I didn't do that because the scaling factor for July is a little off as they have different signs for the binned and unbinned product. I have some figures in the notebook showing what I mean by that. So while the absolute numbers might be wrong, I think we can still compare the two experiments. I forgot to mention that when I uploaded the figures to this issue. If we want to add a figure to the paper, we obviously have to fix this.

adele-morrison commented 3 years ago

Hi Wilma, Can you please push that cross isobath code to your github branch so that I can play with it re sea ice transports? Thank you, Paul

@PaulSpence, let me know if you want to chat to go through the cross-isobath transport code. It expects the transports to be on the vhrho/uhrho grid, so depending what grid the sea ice velocities are on, you may need to interpolate them first (and also the sea ice thickness).

wghuneke commented 3 years ago

Similar figure as above, but showing on the left the anomaly for the monthly time series of the circumpolar sum. Looks quite symmetric, and I'm actually quite impressed by how quickly you see the changes in the heat transport after perturbing the winds.

Fig_total_heat_transport_1000m_isobath_anomaly

adele-morrison commented 3 years ago

Possibly it's so fast because it's dominated by the Ekman layer transport change? i.e. There is more southward transport in the surface layer in UP. Even if these surface waters are cold (e.g. -1 degC), that would still contribute a net southward heat transport relative to the freezing point.

Might be interesting to split this into transport in the top 100 or 200m and the rest of the water column to separate where the change is coming from (without needing to do the full density binning using daily output).

Also, while it's largely symmetric, there is that large negative deviation in the DOWN simulation (and smaller positive deviations in the UP) during April - July. To check if these changes are robust differences from the control, perhaps we could plot on the range of variation in heat transport each month in the control over a 10 or 20 year period?