the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Oct 2022
Seasonal cycle of sea surface temperature in the tropical Angolan upwelling system
Abstract. The Angolan shelf system represents a highly productive ecosystem. Throughout the year sea surface temperatures (SSTs) are cooler near the coast than further offshore. Lowest SSTs, the strongest cross-shore temperature gradient and maximum productivity occur in austral winter when seasonally prevailing upwelling favourable winds are weakest. Here, we investigate the seasonal mixed layer heat budget to analyse atmospheric and oceanic causes for heat content variability. By using different satellite and in-situ data, we derive monthly estimates of surface heat fluxes, mean horizontal advection and local heat content change. We calculate the heat budgets for the near coastal and offshore regions separately to explore processes that lead to the observed differences. The results show that the net surface heat flux warms the coastal ocean stronger than further offshore thus acting to damp spatial SST differences. Mean horizontal heat advection is dominated by meridional advection of warm water along the Angolan coast. However, its contribution to the heat budget is small. Ocean turbulence data suggests that the heat flux due to turbulent mixing across the base of the mixed layer is an important cooling term. This turbulent cooling that is strongest in shallow shelf regions is capable of explaining the observed negative cross-shore temperature gradient. The residuum of the mixed layer heat budget and uncertainties of budget terms are discussed.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(6192 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(6192 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-973', Anonymous Referee #1, 16 Nov 2022
Review of "Seasonal cycle of sea surface temperature in the tropical Angolan upwelling system" by Korner et al.
The seasonal cycle of the mixed layer heat budget is analyzed in the southeastern tropical Atlantic near the African coast. There is a significant residual in the budget when comparing the rate of change of mixed layer heat content and the sum of the surface heat flux and horizontal oceanic heat advection. The residual is larger near the coast in the annual mean, suggesting a larger contribution from vertical turbulent cooling through the base of the mixed layer. Direct measurements of turbulence from several cross-shore cruises reveal stronger mixing and turbulent cooling in the near-shore region, consistent with the heat budget results. It is also hypothesized that seaonal variations in temperature stratification my generate a seasonal cycle in turbulent cooling that drives seasonal differences in the cross-shore SST gradient.
The manuscript is well-written and organized, though there are numerous minor edits that are needed to the language/grammar (see detailed comments below). The results are interesting and will be useful for understanding the mechanisms of mixing and SST variability in the Angolan region and possibly more generally in coastal areas, and for validating models since many have large biases in the southeastern tropical Atlantic. The conclusions are supported well by the analysis and results. I have a few main comments for the authors to consider during their revision.
Main comments:Lines 301 and 381: the turbulent heat flux is averaged between 2 and 15 m below the ML: why use this depth range? Are results similar for smaller depth ranges? 15 m seems deep for mixing to affect SST. Is there any precedent for using this depth range?
As the authors mention, the use of climatological MLD introduces uncertainty. What are the expected errors introduced by the MLD climatology? How do they affect the uncertainties in the heat budget terms? Can you estimate them by comparing the clim. MLD that you used to the actual MLD calculated from the PIRATA mooring? With such a thin climatological ML, small errors could have a big impact on the magnitudes/errors of the heat budget terms and error bars on residual.
Lines 326-327: It looks like the combination of a relative maximum in dissipation (Fig. 7a) and minimum in dT/dz (Fig. 7c) causes the maximum in turbulent cooling 7 m below the ML (Fig. 7d). This might be worth pointing out. Without the max. in dissipation, I'm not sure the pronounced max. in cooling would be there.
Minor edits:
line 35: Delete 'Additionally' and change 'southeast' to 'southeastern'
line 53: Insert 'the' after 'reveals'
line 54: Change 'propagate' to 'propagates'
line 63: Change 'intrude' to 'intrudes'
line 65: Change to '...Current, with the Congo River an important...'
line 71: Either 'Southeast' or 'southeastern' (and in other places in the manuscript)
line 77: Change 'help' to 'helps'
line 84: Change to '...area different processes lead to the cooling of the ML compared to further offshore.'
line 102: Change 'location' to 'locations'
line 118: Change 'mention' to 'mentioned'
line 159: Change to '...within 1-deg of the coast.'
line 160: Change '...over the extend of the boxes.' to '...over the boxes.'
line 171: Change to 'It decays on length scales between 0 and 0.267 m.' and 'exponential' to 'exponentials'
line 191: Change to 'Integration of the shear wave number...'
line 197: Change to '...spectra are...
line 241: Change to '...amplitudes and strengths...'
line 242: Change to 'The minimum in July is driven by the...'
line 242: Change to 'The largest differences...'
line 251: Change 'lead' to 'leads'
line 257: Change 'differences' to 'difference'
line 261: Change to '...coastal box temperature decreases...'
line 268: Delete 'in'
line 284: Change to '...amounts to 21.5...'
line 285: Change to '...heat advection is...'
line 290: Change 'indicates' to 'indicate'
line 318: Change 'shows' to 'show'
line 331: Change 'contribute' to 'contributes'
line 336: Insert comma between 'same' and 'leading'
line 343: Change 'shows' to 'show'
Citation: https://doi.org/10.5194/egusphere-2022-973-RC1 - AC1: 'Reply on RC1', Mareike Körner, 21 Dec 2022
-
RC2: 'Comment on egusphere-2022-973', Anonymous Referee #2, 21 Nov 2022
Summary
This study focusses on understanding the seasonal cycle of Sea Surface Temperature (SST) in the tropical Angolan Upwelling System and in particular look at the processes responsible for an increased cross-shore SST gradient in winter. To do so, the authors analyse the seasonal mixed layer heat budget using satellites and reanalysis datasets in two boxes: at the coast and offshore. By comparing the temperature rate of change to the surface heat fluxes, horizontal advection and residual, they found (1) a strong contribution of the surface heat fluxes that sets the seasonal cycle counterbalance by (2) a significant contribution of the residual; (3) The contribution of the horizontal advection is shown to be minor. Interestingly, the cooling contribution of the residual is larger near the coast than offshore suggesting a key role of the residual in driving the increased SST cross-shore gradient in winter. Turbulent heat fluxes estimated from shipboard measurements show a strong cooling effect of turbulent mixing, which is particularly strong at the coast. This process might explain the increased cross-shore SST gradient.
This is an important topic, and the paper is well-structured and clear. I do have reservations about the large uncertainties that might exist due to the use of various datasets, the gaps between data resolutions and the contribution of the other processes included in the residual but the major points are well discussed in the discussion and the main conclusions of the paper are still valid. There are a few general remarks and minor corrections of the text that I feel that the authors should address, but overall, it is a substantive piece of work of good quality and worthy of publication.
General Remarks
Introduction: L30-35. When reading the introduction, I will have found it nice to see a plot with the seasonal cross-shore SST gradient and primary productivity (the paper's main motivation). Maybe, as a last subplot of Fig. 1?
Introduction: L45 to 61. In this part of the introduction, you describe the SST seasonal cycle, emphasizing that the latter cannot be explained by the upwelling-favourable winds. What about the net heat flux that I guess is a major contributor to the seasonal cycle SST in the tAUS?
L160-163: Why did you decide to interpolate variables with coarser resolution onto a higher resolution grid? Usually, the inverse is done to prevent the generation of fake information. How much do you think this might affect the contribution of mean horizontal advection?
L254 – It is interesting that the SW is higher in the coastal box than in the offshore box. I will have thought the contrary due to the important cloud cover along the coast. Do you think that the bias in the SW could be more important along the coast?
Figure8: From the figure, it seems that the rate of change in winter is more negative offshore than inshore (except for April maybe). Should not be the cooling stronger at the coast, resulting in an increased SST cross-shore gradient in winter. Could you provide a plot showing the difference of heat content change in the two boxes?
Discussion: L494 onward: What about the other processes constrained in the residual (entrainment, vertical temperature velocity covariance)? They also represent an uncertainty for the contribution of the turbulent heat to the net budget that is not discussed here. In literature, are there studies that have shown that their contribution could be neglected?
Also, I wonder how much the contribution of the mean advection is smoothed (due to their low spatial and temporal resolution) compared to the turbulent heat loss derived using high resolution data. Are the terms comparable?
Overall document: In overall, I suggest to re-check the punctuation and the various acronyms in the text. For example, I found several times the term mixed layer instead of “ML”.
Line by Line Comments
L40, 53 Same sometime to define acronym you use capital letters, sometime no. To me it is better to do, one or the other
L8 – remove “the” before strongest
L17 - “This turbulent cooling, strongest in shallow shelf regions, can explain the observed negative cross-shore temperature gradient.”
L66 – Acronym not defined – it is defined later at L78.
L21- southeast to be consistent with the rest of the draft
L26 – “extent” instead of “extend”
Figure1 – Please add the arrow length on the figure
L65 – AC instead of Angola Current
L66, 67: Maybe here you want to use the acronyms you defined (SSS/SST)
L76- shows
L77-helps to reduce errors.
L89 – sections
L118 – mentioned
L120- radiations
L123 - fluxes
L128 – SST analyses are
L131 – decide if you prefer near-surface or near surface for consistency
L135 – remove coma
L140 – remove “the”
L160 – extent instead of extend
L161 – resolutions
L169 - ‘by’ applying
L172 - parts
L173 - exponentials
L173 – MLD
L189 - provided
L190 – are used to
L215 – is calculated instead of are calculated
L217 – level
L220 – The arrow
L228 – impacting
L229 - analysis
L236 – The black line shows or Black lines show
L240 – Surface heat fluxes show
L251- increased wind speed away from the … leads
L255 – dampen
L257 – in the coastal
L261 – decreases
L268 – is generally weak
L271 – AC
L274 – centered
L275 – annually averaged
L288-290 - , microstructure profiles available in this study (section 2.1) indicate instead of “the microstructure profiles available to this study (section 2.1) indicates”
L297 – to the coast
L299 - “close to” or “at the surface”
L318 – “Results in Fig. 7 clearly show” instead of “The results in Fig. 7 clearly shows”
L321 - exhibits
L327 – strengths
L331 – contributes
L339 – In shallow water, fluxes …
L343-345: Here the sentence could be reformulated
L347 – exhibit
L349 – a possible explanation
L365 - and is only
L383-384 – Microstructure profiles suggest
L388 – revealed large differences in monthly averaged ..
L404/405 – As the warming [...], the resulting
L406 – dampen
L407 – Do you mean Austral winter?
L408- The mean […] to the warming
L413 – to the coast
L417 – twice as large as in the offshore box
L436 - exhibits
L447 – acts to dampen
L454 – suggest
L464 - show
L466 – the breaking
L467 – values
L477 – suggests
L479 – in this context are
L494 – Results […] show
L496 – represents
L511 – datasets
Figure A. 2: Correct the title of sensible heat flux; Be consistent in the paper for shortwave / longwave
L532, 534, 535, 538 – exhibits, reveals, shows, differs
L547 – in better agreement with
L559 – “than” instead of “then”
L564 – PIRATA-SEE instead of TropFlux
Figure A.3 - In the legend of the figure add a description of the Net flux.
L570 – The smallest
L578 – A minimum southward velocity
L605 – the institute
Citation: https://doi.org/10.5194/egusphere-2022-973-RC2 - AC2: 'Reply on RC2', Mareike Körner, 21 Dec 2022
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-973', Anonymous Referee #1, 16 Nov 2022
Review of "Seasonal cycle of sea surface temperature in the tropical Angolan upwelling system" by Korner et al.
The seasonal cycle of the mixed layer heat budget is analyzed in the southeastern tropical Atlantic near the African coast. There is a significant residual in the budget when comparing the rate of change of mixed layer heat content and the sum of the surface heat flux and horizontal oceanic heat advection. The residual is larger near the coast in the annual mean, suggesting a larger contribution from vertical turbulent cooling through the base of the mixed layer. Direct measurements of turbulence from several cross-shore cruises reveal stronger mixing and turbulent cooling in the near-shore region, consistent with the heat budget results. It is also hypothesized that seaonal variations in temperature stratification my generate a seasonal cycle in turbulent cooling that drives seasonal differences in the cross-shore SST gradient.
The manuscript is well-written and organized, though there are numerous minor edits that are needed to the language/grammar (see detailed comments below). The results are interesting and will be useful for understanding the mechanisms of mixing and SST variability in the Angolan region and possibly more generally in coastal areas, and for validating models since many have large biases in the southeastern tropical Atlantic. The conclusions are supported well by the analysis and results. I have a few main comments for the authors to consider during their revision.
Main comments:Lines 301 and 381: the turbulent heat flux is averaged between 2 and 15 m below the ML: why use this depth range? Are results similar for smaller depth ranges? 15 m seems deep for mixing to affect SST. Is there any precedent for using this depth range?
As the authors mention, the use of climatological MLD introduces uncertainty. What are the expected errors introduced by the MLD climatology? How do they affect the uncertainties in the heat budget terms? Can you estimate them by comparing the clim. MLD that you used to the actual MLD calculated from the PIRATA mooring? With such a thin climatological ML, small errors could have a big impact on the magnitudes/errors of the heat budget terms and error bars on residual.
Lines 326-327: It looks like the combination of a relative maximum in dissipation (Fig. 7a) and minimum in dT/dz (Fig. 7c) causes the maximum in turbulent cooling 7 m below the ML (Fig. 7d). This might be worth pointing out. Without the max. in dissipation, I'm not sure the pronounced max. in cooling would be there.
Minor edits:
line 35: Delete 'Additionally' and change 'southeast' to 'southeastern'
line 53: Insert 'the' after 'reveals'
line 54: Change 'propagate' to 'propagates'
line 63: Change 'intrude' to 'intrudes'
line 65: Change to '...Current, with the Congo River an important...'
line 71: Either 'Southeast' or 'southeastern' (and in other places in the manuscript)
line 77: Change 'help' to 'helps'
line 84: Change to '...area different processes lead to the cooling of the ML compared to further offshore.'
line 102: Change 'location' to 'locations'
line 118: Change 'mention' to 'mentioned'
line 159: Change to '...within 1-deg of the coast.'
line 160: Change '...over the extend of the boxes.' to '...over the boxes.'
line 171: Change to 'It decays on length scales between 0 and 0.267 m.' and 'exponential' to 'exponentials'
line 191: Change to 'Integration of the shear wave number...'
line 197: Change to '...spectra are...
line 241: Change to '...amplitudes and strengths...'
line 242: Change to 'The minimum in July is driven by the...'
line 242: Change to 'The largest differences...'
line 251: Change 'lead' to 'leads'
line 257: Change 'differences' to 'difference'
line 261: Change to '...coastal box temperature decreases...'
line 268: Delete 'in'
line 284: Change to '...amounts to 21.5...'
line 285: Change to '...heat advection is...'
line 290: Change 'indicates' to 'indicate'
line 318: Change 'shows' to 'show'
line 331: Change 'contribute' to 'contributes'
line 336: Insert comma between 'same' and 'leading'
line 343: Change 'shows' to 'show'
Citation: https://doi.org/10.5194/egusphere-2022-973-RC1 - AC1: 'Reply on RC1', Mareike Körner, 21 Dec 2022
-
RC2: 'Comment on egusphere-2022-973', Anonymous Referee #2, 21 Nov 2022
Summary
This study focusses on understanding the seasonal cycle of Sea Surface Temperature (SST) in the tropical Angolan Upwelling System and in particular look at the processes responsible for an increased cross-shore SST gradient in winter. To do so, the authors analyse the seasonal mixed layer heat budget using satellites and reanalysis datasets in two boxes: at the coast and offshore. By comparing the temperature rate of change to the surface heat fluxes, horizontal advection and residual, they found (1) a strong contribution of the surface heat fluxes that sets the seasonal cycle counterbalance by (2) a significant contribution of the residual; (3) The contribution of the horizontal advection is shown to be minor. Interestingly, the cooling contribution of the residual is larger near the coast than offshore suggesting a key role of the residual in driving the increased SST cross-shore gradient in winter. Turbulent heat fluxes estimated from shipboard measurements show a strong cooling effect of turbulent mixing, which is particularly strong at the coast. This process might explain the increased cross-shore SST gradient.
This is an important topic, and the paper is well-structured and clear. I do have reservations about the large uncertainties that might exist due to the use of various datasets, the gaps between data resolutions and the contribution of the other processes included in the residual but the major points are well discussed in the discussion and the main conclusions of the paper are still valid. There are a few general remarks and minor corrections of the text that I feel that the authors should address, but overall, it is a substantive piece of work of good quality and worthy of publication.
General Remarks
Introduction: L30-35. When reading the introduction, I will have found it nice to see a plot with the seasonal cross-shore SST gradient and primary productivity (the paper's main motivation). Maybe, as a last subplot of Fig. 1?
Introduction: L45 to 61. In this part of the introduction, you describe the SST seasonal cycle, emphasizing that the latter cannot be explained by the upwelling-favourable winds. What about the net heat flux that I guess is a major contributor to the seasonal cycle SST in the tAUS?
L160-163: Why did you decide to interpolate variables with coarser resolution onto a higher resolution grid? Usually, the inverse is done to prevent the generation of fake information. How much do you think this might affect the contribution of mean horizontal advection?
L254 – It is interesting that the SW is higher in the coastal box than in the offshore box. I will have thought the contrary due to the important cloud cover along the coast. Do you think that the bias in the SW could be more important along the coast?
Figure8: From the figure, it seems that the rate of change in winter is more negative offshore than inshore (except for April maybe). Should not be the cooling stronger at the coast, resulting in an increased SST cross-shore gradient in winter. Could you provide a plot showing the difference of heat content change in the two boxes?
Discussion: L494 onward: What about the other processes constrained in the residual (entrainment, vertical temperature velocity covariance)? They also represent an uncertainty for the contribution of the turbulent heat to the net budget that is not discussed here. In literature, are there studies that have shown that their contribution could be neglected?
Also, I wonder how much the contribution of the mean advection is smoothed (due to their low spatial and temporal resolution) compared to the turbulent heat loss derived using high resolution data. Are the terms comparable?
Overall document: In overall, I suggest to re-check the punctuation and the various acronyms in the text. For example, I found several times the term mixed layer instead of “ML”.
Line by Line Comments
L40, 53 Same sometime to define acronym you use capital letters, sometime no. To me it is better to do, one or the other
L8 – remove “the” before strongest
L17 - “This turbulent cooling, strongest in shallow shelf regions, can explain the observed negative cross-shore temperature gradient.”
L66 – Acronym not defined – it is defined later at L78.
L21- southeast to be consistent with the rest of the draft
L26 – “extent” instead of “extend”
Figure1 – Please add the arrow length on the figure
L65 – AC instead of Angola Current
L66, 67: Maybe here you want to use the acronyms you defined (SSS/SST)
L76- shows
L77-helps to reduce errors.
L89 – sections
L118 – mentioned
L120- radiations
L123 - fluxes
L128 – SST analyses are
L131 – decide if you prefer near-surface or near surface for consistency
L135 – remove coma
L140 – remove “the”
L160 – extent instead of extend
L161 – resolutions
L169 - ‘by’ applying
L172 - parts
L173 - exponentials
L173 – MLD
L189 - provided
L190 – are used to
L215 – is calculated instead of are calculated
L217 – level
L220 – The arrow
L228 – impacting
L229 - analysis
L236 – The black line shows or Black lines show
L240 – Surface heat fluxes show
L251- increased wind speed away from the … leads
L255 – dampen
L257 – in the coastal
L261 – decreases
L268 – is generally weak
L271 – AC
L274 – centered
L275 – annually averaged
L288-290 - , microstructure profiles available in this study (section 2.1) indicate instead of “the microstructure profiles available to this study (section 2.1) indicates”
L297 – to the coast
L299 - “close to” or “at the surface”
L318 – “Results in Fig. 7 clearly show” instead of “The results in Fig. 7 clearly shows”
L321 - exhibits
L327 – strengths
L331 – contributes
L339 – In shallow water, fluxes …
L343-345: Here the sentence could be reformulated
L347 – exhibit
L349 – a possible explanation
L365 - and is only
L383-384 – Microstructure profiles suggest
L388 – revealed large differences in monthly averaged ..
L404/405 – As the warming [...], the resulting
L406 – dampen
L407 – Do you mean Austral winter?
L408- The mean […] to the warming
L413 – to the coast
L417 – twice as large as in the offshore box
L436 - exhibits
L447 – acts to dampen
L454 – suggest
L464 - show
L466 – the breaking
L467 – values
L477 – suggests
L479 – in this context are
L494 – Results […] show
L496 – represents
L511 – datasets
Figure A. 2: Correct the title of sensible heat flux; Be consistent in the paper for shortwave / longwave
L532, 534, 535, 538 – exhibits, reveals, shows, differs
L547 – in better agreement with
L559 – “than” instead of “then”
L564 – PIRATA-SEE instead of TropFlux
Figure A.3 - In the legend of the figure add a description of the Net flux.
L570 – The smallest
L578 – A minimum southward velocity
L605 – the institute
Citation: https://doi.org/10.5194/egusphere-2022-973-RC2 - AC2: 'Reply on RC2', Mareike Körner, 21 Dec 2022
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 248 | 80 | 13 | 341 | 2 | 2 |
- HTML: 248
- PDF: 80
- XML: 13
- Total: 341
- BibTeX: 2
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Peter Brandt
Marcus Dengler
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(6192 KB) - Metadata XML