the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 May 2024
Monitoring the coastal-offshore water interactions in the Levantine Sea using ocean color and deep supervised learning
Abstract. Understanding and tracking the surface circulation of the Levantine Sea presents significant challenges, particularly close to the coast. This difficulty arises due to two main factors: the limited availability of in-situ observations and the increasing inaccuracies in altimetry data close to the coastline. Here, we propose a new approach to monitor the interaction between offshore and coastal waters. In this approach, we develop a pattern detection model using deep learning by training the U-Net model on ocean color data to track the interactions between the coastal and offshore water in the Levantine Sea.
The results showed the presence of notable variations in the behavior of coastal currents as they progress northward beyond 33.8° E. As these coastal currents become increasingly unstable, they exhibit continuous pinching-off events that are missed by conventional observational tools. These pinching-off events, observed especially along the Lebanese coast, manifest in various patterns evolving simultaneously. Typically, these patterns have a relatively short lifespan of a few weeks, appearing and disappearing rapidly. However, these structures can evolve into larger eddies that endure over four months in some years, especially in the Northern part of the Lebanese coasts. Although these structures could be observed during all the seasons, spring consistently records the lowest activity of these structures. Overall, we showed that the pinching-off events were always observed in the eastern part of the Levantine Sea. On the contrary, in the southern part along the Egyptian coasts, the coastal flow is more stable in the southern region, where these events are less frequently observed, with more than 63 % of the total observations not exhibiting any pinching-off events. Moreover, when these events occur in the south, their spatial extent is notably limited.
This research not only sheds light on previously missed (or underestimated) coastal current dynamics in the Levantine Sea but also highlights the crucial need to increase in-situ observations to advance our understanding of this region’s complex oceanographic processes.
-
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
(8366 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(8366 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-1168', Anonymous Referee #1, 29 May 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1168/egusphere-2024-1168-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Georges Baaklini, 09 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1168/egusphere-2024-1168-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Georges Baaklini, 09 Sep 2024
-
RC2: 'Comment on egusphere-2024-1168', Jack Barth, 17 Aug 2024
Motivated by the desire to identify and quantify the interaction of “coastal” water with the adjacent deep Levantine Sea, the authors develop a pattern detection model using deep learning applied to ocean color data to track the interactions between the coastal and offshore water. They choose to use ocean color because using satellite altimetry is prone to errors near the coast. The authors discover that pinching-off events that send coastal water offshore occur more frequently to the north along the coast where the coastal flows become more unstable. They also consider the temporal evolution, persistence, spatial scale and year-to-year variability of these features that transport coastal water seaward. Overall, this is a nice application of a novel technique, and the authors do a nice job of analyzing the results to gain insight into the connection of the coastal and adjacent deep sea. I recommend that this paper be published in the “Ocean Science” after minor revision.
Suggestions to improve the manuscript:
- I agree with the authors that satellite altimetry is likely to miss small-scale features, especially near the coast but I do not think the right-hand panel in Figure 2 makes the case well. First, the text says “velocity field” while the figure shows vorticity. Second, have the chlorophyll and altimetry fields been closely matched in time? The altimetry field has the correct sized positive and negative vorticity features that, if simply displaced by a mismatch in time, could match up with drifter field (see my comment that a symbol is needed at the start of the drifter track so the reader can tell the direction of flow). Please clear up the use of the altimetry data.
- More needs to be said about choosing the region of the Nile and eastern part of the Levantine Sea to train the learning procedure. This region should be indicated on one of the maps. What are the consequences if a larger or different region are chosen for training? What biases or errors might be introduced by the choice of training region?
- Lines 130-131 present the model performance. Please state whether these are “good” scores, perhaps by giving typical values for a good performance from other studies. The 3% error is indeed impressive.
- Section 4, figure 7: how is the dark blue mask along the coastline chosen? It evidently masks out the coastal water that is on the continental shelf (inshore of some isobath? 200 m?), so that only the coastal water making it into the deep ocean is highlighted in color. Does the sentence “The along-slope coastal circulation has been removed from the analysis to isolate and highlight the deviations or pinching-off events” have something to do with the mask?
- In the paragraph about potential instability of the along-coast flow (lines 157-164), a simple explanation is that when the coastal current is over sloping bottom topography the flow is more stable as water parcels tend to follow isobaths. So downstream of where the continental shelf narrows (the start of the eastern block), more of the flow is over the slope and stratification isolates it from the stabilizing influence of bottom topography. Offshore flow can also be caused by flow-topography interaction. I cannot tell from Figure 1 if there are coastline or bottom topographic features that might promote offshore flow. The authors should comment about this possibility.
- Section 4.2 about spatial scales and temporal persistence is interesting. Regarding the spatial scale, how many eddies based on an estimate of the local internal Rossby radius of deformation can “fit” along the coast in the eastern block? This would help explain how many different structures might be observed.
- I am not sure of the purpose of section 4.3 and suggest it could be omitted.
Minor/editorial comments:
- Figure 1 caption: define AW; ShE not She
- Figure 2: put a mark on the start location of the drifter
- Figure 8 caption: “… (orange) blocks from 2003 to 2023.”
- Figure 13: how was the top-to-bottom order of the seasons chosen? Not by total, since spring is the least of all. Why not order them by season: fall, winter, spring, summer?
- Figure 14: “… chlorophyll images are overlaid by the average …”
Citation: https://doi.org/10.5194/egusphere-2024-1168-RC2 -
AC2: 'Reply on RC2', Georges Baaklini, 09 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1168/egusphere-2024-1168-AC2-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-1168', Anonymous Referee #1, 29 May 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1168/egusphere-2024-1168-RC1-supplement.pdf
-
AC1: 'Reply on RC1', Georges Baaklini, 09 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1168/egusphere-2024-1168-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Georges Baaklini, 09 Sep 2024
-
RC2: 'Comment on egusphere-2024-1168', Jack Barth, 17 Aug 2024
Motivated by the desire to identify and quantify the interaction of “coastal” water with the adjacent deep Levantine Sea, the authors develop a pattern detection model using deep learning applied to ocean color data to track the interactions between the coastal and offshore water. They choose to use ocean color because using satellite altimetry is prone to errors near the coast. The authors discover that pinching-off events that send coastal water offshore occur more frequently to the north along the coast where the coastal flows become more unstable. They also consider the temporal evolution, persistence, spatial scale and year-to-year variability of these features that transport coastal water seaward. Overall, this is a nice application of a novel technique, and the authors do a nice job of analyzing the results to gain insight into the connection of the coastal and adjacent deep sea. I recommend that this paper be published in the “Ocean Science” after minor revision.
Suggestions to improve the manuscript:
- I agree with the authors that satellite altimetry is likely to miss small-scale features, especially near the coast but I do not think the right-hand panel in Figure 2 makes the case well. First, the text says “velocity field” while the figure shows vorticity. Second, have the chlorophyll and altimetry fields been closely matched in time? The altimetry field has the correct sized positive and negative vorticity features that, if simply displaced by a mismatch in time, could match up with drifter field (see my comment that a symbol is needed at the start of the drifter track so the reader can tell the direction of flow). Please clear up the use of the altimetry data.
- More needs to be said about choosing the region of the Nile and eastern part of the Levantine Sea to train the learning procedure. This region should be indicated on one of the maps. What are the consequences if a larger or different region are chosen for training? What biases or errors might be introduced by the choice of training region?
- Lines 130-131 present the model performance. Please state whether these are “good” scores, perhaps by giving typical values for a good performance from other studies. The 3% error is indeed impressive.
- Section 4, figure 7: how is the dark blue mask along the coastline chosen? It evidently masks out the coastal water that is on the continental shelf (inshore of some isobath? 200 m?), so that only the coastal water making it into the deep ocean is highlighted in color. Does the sentence “The along-slope coastal circulation has been removed from the analysis to isolate and highlight the deviations or pinching-off events” have something to do with the mask?
- In the paragraph about potential instability of the along-coast flow (lines 157-164), a simple explanation is that when the coastal current is over sloping bottom topography the flow is more stable as water parcels tend to follow isobaths. So downstream of where the continental shelf narrows (the start of the eastern block), more of the flow is over the slope and stratification isolates it from the stabilizing influence of bottom topography. Offshore flow can also be caused by flow-topography interaction. I cannot tell from Figure 1 if there are coastline or bottom topographic features that might promote offshore flow. The authors should comment about this possibility.
- Section 4.2 about spatial scales and temporal persistence is interesting. Regarding the spatial scale, how many eddies based on an estimate of the local internal Rossby radius of deformation can “fit” along the coast in the eastern block? This would help explain how many different structures might be observed.
- I am not sure of the purpose of section 4.3 and suggest it could be omitted.
Minor/editorial comments:
- Figure 1 caption: define AW; ShE not She
- Figure 2: put a mark on the start location of the drifter
- Figure 8 caption: “… (orange) blocks from 2003 to 2023.”
- Figure 13: how was the top-to-bottom order of the seasons chosen? Not by total, since spring is the least of all. Why not order them by season: fall, winter, spring, summer?
- Figure 14: “… chlorophyll images are overlaid by the average …”
Citation: https://doi.org/10.5194/egusphere-2024-1168-RC2 -
AC2: 'Reply on RC2', Georges Baaklini, 09 Sep 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1168/egusphere-2024-1168-AC2-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 344 | 82 | 42 | 468 | 26 | 43 |
- HTML: 344
- PDF: 82
- XML: 42
- Total: 468
- BibTeX: 26
- EndNote: 43
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Georges Baaklini
Julien Brajard
Leila Issa
Gina Fifani
Laurent Mortier
Roy El Hourany
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(8366 KB) - Metadata XML