Skip to content

Latest commit

 

History

History
55 lines (33 loc) · 5.32 KB

Results_summary.md

File metadata and controls

55 lines (33 loc) · 5.32 KB

The story so far

Motivation for this study

Many studies assume (or have found at particular regions / times) that there is a relationship between the ASC strength and the cross slope heat transport. But this assumption hasn't been tested comprehensively (across all time and space scales). That's our plan for this study.

Results

  1. We find only weak correlations between the depth average ASC and cross slope heat transport (both integrated across different regimes):
Screenshot 2023-11-23 at 10 46 39 am

(Note there are a couple of exceptions at depth where the correlations are not so weak - 300-800m depth for the seasonality in the surface regime, and perhaps 700-900m depth for the reverse regime. Maybe check significance for these?) Is there some way we can communicate this message more simply than the above figure? e.g. Maybe lead people in slowly with a scatter plots for a few sample regions showing weak correlation before diving into this much detail?

  1. There are reasonable correlations between the ASC and cross slope heat transport at the same depth.
Screenshot 2023-11-23 at 11 00 11 am

The correlations are particularly high for the seasonality (r $\ge$ 0.8, lower row), and are around r=0.5-0.6 at depth for the interannual correlations (middle row). The regressions are in the right direction at depth for the surface and deep regimes - a weaker (less westward) ASC results in a southward heat transport anomaly. The reverse regime relationship seems to be in the wrong direction? In the reverse regime the ASC is eastward (positive), so a negative regression means that a stronger eastward ASC is linked to stronger southward transport. How does that work?

  1. Information about the relationship on different timescales. The relationship is strongest for determining seasonality of heat transport. The surface correlations disappear for the interannual variability.

  2. Correlations between eddy heat transport and the ASC are low (except for the deepest point in the Deep regime - maybe due to overflows?).

  3. Correlations between layerwise ASC and CSHT hold in regions of Totten, WAP and Ross, producing plots that resemble the regime averaged correlations. Correlations are also generally higher when averaging over specific regions, instead of over whole regimes like in 2. image image

Planned work we would like to add to this:

  • Test how correlations change if we keep the regime locations constant in time.

    • Next steps: If we do this, we could also look at correlations between daily variations in ASC and CSHT. (Will)

  • Regional analysis. People care about single ice shelves. Do we find stronger or weaker correlations at depth for smaller regions? Do a few regional case studies in different regimes to test. e.g. Ross Sea, Totten basin, Wilkins etc.

    • Next steps: Check that the same results hold/if correlations are better with daily data (Ellie)

  • Following from 5. and Wilma's analysis, averaging over smaller regions could retain some of the variability that is removed using regime averaged velocity/CSHT.

    • Next steps: Check if averaging velocities and CSHT over specific regions (e.g. 3, 10 deg longitude bins, or spliting reverse regime into 3 sections) before computing correlations would improve our results. ( Fabio)

Other analysis ideas that we may or may not need to follow up on:

  • How does the heat transport variability change with ASC strength? i.e. does the standard deviation (over a month or so timescale?) correlate with ASC strength? e.g. similar dynamics to Ellie's paper.

  • Can we get some observational evidence? e.g. a single mooring where we find that the ASC at depth is correlated with the onshore heat transport (but not the depth average ASC).

  • Is there any way to get more at the dynamics of why the ASC is related to the heat transport? Dynamical possibilities:

    1. The ASC strength change is a passive response to changes in the depth of isopycnals. Upward movement of isopycnals allows more access of warm CDW on deeper isopycnals. This would explain the negative correlations in both the reverse and surface/deep regimes, because upward moving isopycnals correspond to both weaker westward transport (surface / deep) and stronger eastward transport (reverse).
    2. A weaker ASC allows more eddy heat transport (eddy diffusion) along isopycnals to reach the shelf.
    3. A weaker ASC co-occurs with a period of stronger eddy activity that brings more heat onto the shelf (i.e. Ellie's mechanism).

    These seem hard to distinguish?

  • Repeat analysis with daily data to look at the ASC / heat transport relationship on sub-seasonal timescales.

  • Repeat for different isobaths (650m, 2000m etc)