the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Nov 2024
Relating North Atlantic Deep Water transport to ocean bottom pressure variations as a target for satellite gravimetry missions
Abstract. The Atlantic Meridional Overturning Circulation (AMOC) is a salient feature of the climate system, observed for its strength and variability with a wide range of offshore installations and expensive sea-going expeditions. Satellite-based measurements of mass changes in the Earth system, such as from the Gravity Recovery and Climate Experiment (GRACE) mission, may help monitor these transport variations at large scale, by measuring associated changes in ocean bottom pressure (OBP) at the boundaries of the Atlantic remotely from space. However, as these signals are mainly confined to the continental slope and small in magnitude, their detection using gravimentry will likely require specialised approaches. Here we use the output of a fine-resolution (1/20°) regional ocean model to assess the connection between OBP signals at the western boundary of the North and South Atlantic. We find that North Atlantic Deep Water (NADW) transports in the ~1–3-km depth range can be reconstructed using spatially averaged OBP signals with correlations of 0.75 (0.72) for the North (South) Atlantic and root-mean-square errors of ~1 Sverdrup on monthly to interannual time scales. We further create a synthetic dataset containing only OBP signals due to NADW transport anomalies at the western boundary, which can be included in dedicated satellite gravimetry simulations to assess the AMOC detection capabilities of future mission scenarios and to develop specialised recovery strategies that are needed to track those weak signatures in the time-variable gravity field.
- Preprint
(21567 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 14 Feb 2025)
-
RC1: 'Comment on egusphere-2024-3660', Rory Bingham, 07 Jan 2025
reply
As a key regulatory component of the climate system, with the potential to disrupt global climate and weather patterns, the Atlantic Meridional Overturning Circulation (AMOC) merits our close attention. Since 2004, AMOC strength has been monitored continuously by the RAPID array at 26°N. However, such in-situ monitoring is labour and resource intensive and only measures the AMOC at a single latitude. Thus, the ability to measure changes in AMOC strength remotely from space could have substantial scientific, societal and economic benefit, especially if part of a system simultaneously measuring many other aspects of the climate system. Inspired, in part, by previous work demonstrating a strong link between ocean bottom pressure (OBP) changes along the North Atlantic western boundary and changes in AMOC strength, the planned NASA-ESA joint twin-pair satellite gravimetry mission offers the potential for such a system. However, detecting the relatively weak and small-scale mass and OBP changes associated with the AMOC, will still prove challenging and likely require the development of novel post-processing techniques. A prerequisite step towards assessing the potential of proposed mission concepts, refining specifications and developing the required post-processing techniques is end-to-end simulations that generate synthetic observational products from a complete Earth System Model of mass redistribution, including an ocean component, and associated gravity changes.
The aims of the present work are twofold: First, the authors examine the extent that zonally-integrated meridional transport variations at various depths, but with a focus on 1000-3000m, can be recovered using pressure on the western boundary. Finding that it can, the work confirms earlier work by Bingham and Hughes (2008). However, the authors go further than this early work by testing two average-based approaches that may be more feasible with relatively low-resolution GRACE-like gravity data. The affirmative results for these cases, and a similar analysis for the South Atlantic, along with the use of a model with a nested, very high resolution North Atlantic, are the most novel aspects of the work. Second, the authors propose a simple method for extracting a meridional transport-related OPB signal from the model - a known signal - that can be used, although the authors are vague on the details, in an end-to-end simulation.
The work is extremely well written, with very few grammatical or typographic errors. It is logically structured and the figures are of a high quality. My overall impression is that the work is sufficiently novel and certainly timely enough to warrant publication, subject to some, mostly, minor points listed below being addressed. In addition, I think the authors should consider the following two broader concerns, which should also be relatively easily addressed:
- The authors refer throughout to North Atlantic Deep Water (NADW) transport as the quantity being measured/estimated. Although this makes for a neat narrative, speaking as an oceanographer, I think it relies on a simplistic, cartoonish view of the ocean’s circulation, one that perhaps applies to some extent to the time-mean flow, but certainly not the time-variable component. The ocean’s actual circulation and the anomalous zonally-integrated meridional flows over a particular depth range at a particular time, while perhaps incorporating some fluctuating NADW signal, are likely composed of a much messier mixture of water masses. And, indeed, the net southward flow probably results from an imbalance of northward and southward flows of different water masses, some not readily classified. In my view, it would be more accurate and scientifically correct to remain agnostic as to the water masses involved, and refer simply to the zonally-integrated meridional transport, in the title and elsewhere. The fact that this range likely includes some fraction of NADW, and the fact that meridional transport fluctuations may, to some undetermined extent, reflect more or less NADW flowing south, could be discussed in the introduction. There is a link but a highly qualified and uncertain one without a lot of further work.
- My second, “major” comment concerns the second aim of the work: The authors do not make clear why an extracted AMOC-related OPB signal is required for the end-to-end simulations. Why not just use an ocean model that naturally includes this signal along with other OPB variability? Although I think I know the reason, that motivation needs to be made clear for this section to make sense to the wider readership.
Minor comments
L20: Incorrectly formed citation.
L20: “contributed” -> “contributes”.
L28: It would be good to specifically introduce the RAPID array here as it is the most comprehensive of the observational systems.
L75: Rather than stating “quite well” it would be good to quantify the agreement found by BH08.
L76: It would be worth emphasising that we can only recover anomalies by this method, and mention the prime notation.
L81: Do you mean the Deep Western Boundary Current in this depth range?
L86: The strong relationship with US east coast sea level, as first demonstrated by Bingham and Hughes (2009) and subsequently supported by many studies, also lend credence to the dominance of the WB pressure.
L95: Bingham and Hughes (2008) also showed a strong correspondence between the upper and lower layer transports.
L114-115: Place commas after “computed” and “component”.
L115-116: Description of OBP calculation could be clearer. It is not entirely clear what is meant by 2-D surface density.
L117: “filtered” -> “filtering”.
L121-2: Sentence starting “Overall” is unclear/awkwardly phrased.
L147-9: The sentence starting “Although” is too long and difficult to parse. The z vs. sigma issue is rather a distraction. I would delete or make it clear by rephrasing that Biastoch is an example of the latter (if that’s the case).
Figure 4 caption, last sentence: “with previous analyses” sounds better.
L185: “1.13 Sv only” -> “only 1.13 Sv”. Is the “only” really justified here given it is quite close to the RMS itself? % of variance accounted for (skill) might be a useful additional (better) metric to provide.
L190: With regard to the statement beginning “Instead” it would be useful here to support this statement with the results shown in Figure 9 of Pail et al (2015) and Figure 13 of Daras et al (2024). If not here, then I think the results shown in these two Figures deserve a more detailed discussion/examination in the introduction, being the only two studies (to my knowledge) that have examined how the slope bottom pressure is impacted by spherical harmonic truncation and noise. It is not as though examinations of these two are not included because there is a lot of additional/better relevant literature to be considered.
L196: Remove the 0.02 value, or just give the correlation itslef.
L210: It is worth noting that when expressed in Sv / cm the shelf scale factor -0.42 is similar to the -0.59 scale factor for the relationship between coastal sea level and the AMOC at 42N found by Bingham and Hughes (2009), with the higher magnitude value found in that study plausibly explained by the larger amplitude of coastal sea level compared with bottom pressure averaged over the shelf. Note, that paper (Figure 1a) also shows a very clear OPB signal along the slope between 1300 and 3000 m.
L210: In BH08, we removed the depth average slope pressure before computing the meridional transport in the upper and lower layers. To an extent, the gradient method achieves the same thing - removing the common signal, although not as effectively as computing the slope average pressure directly (which, or course, is not possible with GRACE like data). The reduced amplitude of the OPB recovered transport (5c) and the lower RMSE of the gradient approach compared to the average OPB transport (5b) seems to support that. I wonder if the gradient approach could be further improved by first removing the average signal across the shelf and slope.
L224: “may allow to monitor variations” -> “may allow variations to be monitored”
L225: Delete “already”.
L233: Would be good to show this time series.
L238: “as noted above” - please be more specific.
L294: Perhaps it is beyond the scope of work, but it would be interesting to consider why the scale factors are so different for the SA.
L300: “suggest” - “confirm previous analyses” is perhaps more accurate.
L318-9: While I agree that the dataset will primarily contain transport related to OPB, and as such will be fit for the proposed purpose, it is perhaps an overstatement to say that the resulting dataset only has OBP variations due to the upper NADW (zonally-integrated meridional) transport, since it may also contain OPB variations that are correlated with the transport but are not dynamically/geostrophically related to the transport variations (eg balanced, basin-wide OPB patterns and other wind-driven signals).
L345: Perhaps “provide continuous large-scale monitoring” should be qualified with the appendage “should satellite gravimetry become operational” or similar.
L349: “previous model-based studies” - cite them.
L352: “as it” -> “, which”; “when” -> “, by”.
L360: “only the variations due to transport variations” - as before, strictly speaking this is not necessarily true - there could be other correlated factors compensating or driving OBP.
L362: Not clear what is meant by “we can supply the data”.
L365: “precludes the removal of pressure signals associated with basin-wide modes” - as mentioned above, the gradient approach does this to some extent which may be the reason for the improved agreement.
L366-7: Incorrectly formatted citation.
L379: “is deemed an important first step” - in light of Pail et al (2015) and Daras et al (2024) it is perhaps a little inaccurate to say the work is a “first step”. It is an important contribution; though I always feel it is better to let the reader judge the importance of one's work, rather than it be the authors doing the deeming! In fact, since the final sentence is also too long, I would recommend a more elegant ending comprised of two shorter sentences.
Citation: https://doi.org/10.5194/egusphere-2024-3660-RC1
Data sets
Relating North Atlantic Deep Water transport to ocean bottom pressure variations as a target for satellite gravimetry missions Linus Shihora et al. https://doi.org/10.5281/zenodo.13985568
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 107 | 37 | 4 | 148 | 1 | 2 |
- HTML: 107
- PDF: 37
- XML: 4
- Total: 148
- BibTeX: 1
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1