On the Mechanisms Driving Latent Heat Flux Variations in the Northwest Tropical Atlantic: a Modeling Approach

Summary

This study examines the relative impacts of mesoscale SST variability, atmospheric stability and ocean currents on the latent heat flux over an ocean region off the coast of Brazil north of where the Amazon River flows into the Atlantic using a high-resolution coupled ocean-atmosphere regional model. The study also examines the components of the ocean mixed layer heat budget across the region. The main motivation seems to be to identify the role of ocean mesoscale processes, particularly SST variability on mesoscales, in determining variability in latent heat flux and upper ocean vertical structure. The Introduction does an adequate job of motivating the work, although it seems overly focused on the latent heat aspects of the study, while the motivation for the ocean mixed layer heat budget is barely mentioned. I found the terminology regarding the various coupling coefficients and latent heat flux derivations to be lacking in coherency, making it difficult to follow the discussions in Secs. 4.1, 4.2 and 4.3.  It was also difficult to follow the discussion in Sec. 4.4 of the various parameter dependencies on mesoscale SST anomalies (histogram plots) without a clear definition of the Amazon plume. Particularly since the authors at one point state that the plume is not shown, but then proceed to discuss its presence in the histogram plots. The ocean mixed layer heat budget discussion is also not well connected to the discussion on the role of mesoscale SST anomalies on the latent heating term across the domain. Throughout Sec 4 there are also discussions that are not necessarily supported by the figures. I recommend major revisions primarily working on improving the flow of the arguments the authors wish to make throughout the text and making sure all the sections flow from one another. Some minor grammatical errors and typos are also noted.

Major Comments

Line 138 - What is meant by “Amazon sub-region is influenced by the Amazon plume”? Do you mean there is advection of warm, fresh surface waters into the sub-region? Please be more specific as the influence the authors have in mind ends up being important throughout the text. Without stating this influence up front here, some of the later discussions are confusing.

Lines 203-205 - This sentence is unclear. It sounds like you will look into linkages between mesoscale SST anomalies and the Amazon plume, but the Amazon plume is not present in your EURECA box (per line 244). Also, as worded, it sounds like you are saying that the mesoscale SST anomalies lead to the Amazon plume, which does not seem to make sense. 

Lines 241-242 - Can the authors provide more discussion of the cold filament of surface water across your domain? This is a very prominent feature of your experiment region and should be discussed further as background for the remaining analyses. 

Lines 255-256 - This seems to contradict what is said above, as the low-salinity patch extends within the Amazon box, while the warm SSTs do not, but both are said to be related to the Amazon plume. 

Line 372 - Again, it is indicated that the Amazon plume is present in the EURECA domain, contrary to what is said on line 244. Discussions around the Amazon plume need clarifying throughout the text, as what remains of the plume in your study area and its impacts on the Amazon box are unclear and inconsistent.

Lines 356-357 - Wind variations in Fig. 2c are very difficult to see. Are the contributions to the relative wind variability mainly due to the differences in the surface currents in the three regions or is the variability in the winds higher in Amazon and Tradewind boxes? Showing time series of the winds and currents and/or the relative wind for the four regions might be more helpful than comparing their separate time means for the purposes of this discussion.

Lines 365-368 - If CFB always increases surface winds in the direction of the current then this particular process would result in relative winds that are smaller than the wind alone, correct? So why would this effect ever increase LHF? It seems like this discussion conflates the relative vs full surface wind impacts on LHF with the impacts of CFB alone. Or maybe I am not understanding CFB?

Line 385 - Is it not that the wind speed increases aloft and decreases at the surface over cold SSTs, but rather that the momentum transfer from aloft to the surface just does not occur. This is the “decoupling” of the surface layer from the free troposphere common in stable boundary layer situations. The wording here is misleading as it implies an opposite momentum transfer to what is happening over warm SSTs.

Line 387 - Why show saturation specific humidity rather than potential temperature to illustrate temperature differences?

Lines 407-408 - On line 355 you state the opposite, that the LHF variations due to SST mesoscale variations are mainly due to the dynamic contribution. I believe this confusion is due to the many ways these authors use the term “dynamic contribution”. On the one hand it seems to be used to describe part of the overall “thermodynamic contribution”, as on line 355, while on the other hand it is also used to describe the relative wind impacts which seem to be what is mean here? I think the terminology needs to be consistent throughout the text given how many effects are being examined. It is difficult as a reader to keep them all straight. And Fig. 1 only shows two of them. Perhaps the authors can provide a table of effects they are investigating along with a description of what they are and how they are isolated? In any case, for the discussion on Lines 407-408, can the authors return to their LHF naming convention and add the appropriate terms in parentheses after the words “thermodynamic contribution” or “SST changes”, “variations in specific humidity”, and “wind speed” or “dynamic contribution” so we know which terms to refer to in Figs. 4 & 5?

Line 425-429 - This entire discussion is difficult to follow. The authors refer to the distribution of the mesoscale SST anomalies in space but we only see a histogram with height in Fig. 6. Also, the Amazon plume is mentioned multiple times despite it not being within the EURECA domain. Please clarify what is meant by the “core of the Amazon plume” since Fig. 1c suggests the plume is mostly outside of this domain.

Lines 446-448 - The colormap may be too hard to read for the OSS values, as it seems that they never get higher than 40% (cyan). A value of 50% would be green, and at least in the version of the figure provided in the manuscript there does not seem to be any green color. Also, is it correct to say that salinity is important to the stability when overall the OSS % is well below 50%? According to Line 235, that means salinity is not important to the ocean stratification. Also, the core of the plume appears to be from the surface to about 20m  depth, while the peak OSS values appear to be from about 15 to 40 m depth. So the peak OSS seems to occur at the base of the plume, not in the core of the plume.

Lines 440-464 - It might help this discussion (as at lines 424 and 426, for example) to provide an additional panel in Fig. 7 showing the SST across the Amazon box only with the SST mesoscale anomalies overlaid as contour lines. This will help to see spatially where the mesoscale SST anomalies are located within the domain. The dT/dz panel could be removed as it confirms no temperature inversions. This could simply be stated in the text without a figure. Also, or alternatively, the Amazon plume waters could be delineated in at least panel 7d.

Lines 452-455 - Still confused where the Amazon plume waters are with respect to the mesoscale SST anomalies. If it is represented by the most fresh SSS anomalies (mesoscale SST anomalies of ~-0.02 to 0.2), then the total heat flux over this plume is near zero, not transitioning to negative until mesoscale SST anomalies > 0.2. Heat tendency in the plume is also near zero, with some positive heating at the upper end of the mesoscale SST anomalies within the plume (>0.1). Again, this discussion might be clearer if we had a planar view of the mesoscale SST anomalies within the Amazon sub-region with the Amazon plume clearly delineated on the map either in Fig 6 or Fig. 7.

Fig. 8 - Please plot SSS over SST in one of these panels so we can see the salinity signature of the Amazon plume and its corresponding SST signature together. This will make earlier discussions of the plume influences easier to follow. You could also consider such a panel for Fig. 2 for the entire EURECA domain.

Fig.8a vs Fig. 2a - It is not clear that the SST contours in Fig. 8a match the filled contours in Fig. 2a. According to Fig. 2a, the warmest SSTs in the Amazon domain are near the 17.2 contour label for specific humidity and towards the northeast, where colors are more yellow. However, there is a clear tongue of warm SST extending from the southwest across to the northeast of this domain (the Amazon plume) in Fig. 8a. Can the authors use a different color bar in Fig 2a to better highlight the SST gradients across the region? 

Line 473 - Do the authors mean temperature advection from the east? The temperature contours appear to be oriented east-west, with temperature increasing to the west. Advection from the south would bring cold water northward I would think, just looking at the SST contours in panels (a) and (b). Or perhaps there are warmer waters below the surface to the south within the ML?

Lines 474-479 - The temperature tendency within the <35 PSU contour is not just negative, it is positive in the southwestern region of this contour, with heating due mainly to horizontal advection, not atmospheric forcing. The region defined by SST > 26.7 degC and SSS between 35 and 35.4 PSU also seems to not exactly match the very narrow region of positive temperature tendencies. It is not clear where the authors are referring to when they talk about the core of the plume. A panel with SST and SSS together with the plume marked on the figure would facilitate this discussion. Also, this discussion contradicts that on lines 453-454.

Fig. 8e,f - Panel (e) is not discussed and is an order of magnitude smaller than most of the other terms. Suggest removing this panel. Likewise, although panel (f) is briefly mentioned, this term is also an order of magnitude smaller than the others and could be left out along with discussion on lines 480-482.

Line 487-488 - If this statement were true, would not the temperature tendencies be zero? They are in fact small compared to the advection, residual and atmospheric forcing terms. Is that what the authors are trying to say, despite the discussion on Lines 474-479 describing the tendencies?

title - Suggest a change to the title as it seems to describe only one section of the manuscript. The latter part of the manuscript is spent understanding the ML budget. Maybe “On the Mechanisms Controlling SST and Ocean Mixed Layer Heat Content in the Northwest Tropical Atlantic: A Modeling Approach”.

Minor Edits

Figure 1 - Suggest splitting this figure into two different figures, one with panels (a) and (b) and one with panel (c). The current 3 panel layout is crowded and the text for panel (b) extends into panel (c).

Line 70-73 - Change “shortens” to “shorten” but also check sentence structure as it does not read well. 

Line 125 - Do the authors mean freshwater, heat, and momentum fluxes? Turbulent does not make sense in this context since momentum fluxes are also turbulent fluxes.

Line 240 & Fig. 2 - Can the authors add Trinidad and Tobago to these panels?

Line 249 - Typo, “wuch” should be “such”.

Line 263 - Change to “in the following sections.”

Lines 274-275 - Sentence is not grammatically correct. Please fix.

Line 328 - Typo, should be “among” or “amongst”.

Sec. 4.4 heading - should be “the Amazon”

Fig. 6 caption - Please add that the values shown are for the Amazon box only for clarity.

Fig. 7 - The labeling on these panels is overall confusing since the x-axis for all panels is only labeled in panels (g) and (h), but a color bar is shown beneath all the panels. It would be better to include the Mesoscale SST anomaly tick labels and axis label in all panels for readability.

Fig. 7 caption - Please state what the white arrows represent in panel (c). They are defined on line 446 but should also be defined in the figure caption. Also, what is their magnitude? Also, add that these panels are for the Amazon box only.

Line 472-473 - I think the authors mean to refer to Fig. 8c, the temperature tendency panel, and Fig. 8d, the horizontal temperature advection panel, in this sentence.