The Coastal Surveillance Through Observation of Ocean Color (COASTℓOOC) dataset

Massicotte, Philippe; Babin, Marcel; Fell, Frank; Fournier-Sicre, Vincent; Doxaran, David

doi:https://doi.org/10.5194/essd-15-3529-2023

Articles | Volume 15, issue 8

https://doi.org/10.5194/essd-15-3529-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/essd-15-3529-2023

© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 8

Data description paper

|

08 Aug 2023

Data description paper |

| 08 Aug 2023

The Coastal Surveillance Through Observation of Ocean Color (COASTℓOOC) dataset

Philippe Massicotte, Marcel Babin, Frank Fell, Vincent Fournier-Sicre, and David Doxaran

Download

Final revised paper (published on 08 Aug 2023)
Preprint (discussion started on 30 Mar 2023)

Interactive discussion

Status: closed

CC1:
'Comment on essd-2023-83', ZhongPing Lee, 14 Apr 2023
This manuscript describes the well known and widely used dataset collected by the COASTLOOC project. It is in general well described, but further polish is necessary. Specifically,
A few citations not available, such as line 30, “Sat, 2000”.

Font size of many figures (e.g., Fig. 2) should be larger.

Incomplete sentences (e.g., line 54, “… boat (some samples were collected in waters 55 as shallow as 1 m, Babin 2003”).

Line 119, R(0+), a typo. Also, please check the grammar of this sentence and the entire manuscript.

Line 121, “using a sum of two exponential functions to account for Raman scattering”. Need more information on why “two exponential functions”. On the other hand, from the figures in A1, it appears after ~1 m, the Eu(779) is no longer valid.

Further, for Eu data, did you carry out correction of instrument self shading?

The increase of Eu(683,706) with depth showing in Fig. 3 suggests this measurement is quite questionable.

Suggest to include a section to specifically to talk about uncertainties associated with the measured parameters.

Fig. 7, for Y axis, suggest to use dynamic range, so the R spectra can be presented more clearly.
Citation: https://doi.org/10.5194/essd-2023-83-CC1
- AC2: 'Reply on CC1', Philippe Massicotte, 28 Jun 2023
  
  Reviewer CC1
  Comment C1
  A few citations not available, such as line 30, “Sat, 2000”.
  Answer A1
  Thank you for pointing it out. We have fixed it and made sure all citations were correctly cited within the manuscript.
  Comment C2
  Font size of many figures (e.g., Fig. 2) should be larger.
  Answer A2
  Text font size has been increased from 10 pt to 12 pt in all figures.
  Comment C3
  Incomplete sentences (e.g., line 54, “… boat (some samples were collected in waters 55 as shallow as 1 m, Babin 2003”).
  Answer A3
  The sentence has been fixed.
  Comment C4
  Line 119, R(0+), a typo. Also, please check the grammar of this sentence and the entire manuscript.
  Answer A4
  The sentence has been fixed.
  Comment C5
  Line 121, “using a sum of two exponential functions to account for Raman scattering”. Need more information on why “two exponential functions”. On the other hand, from the figures in A1, it appears after ~1 m, the Eu(779) is no longer valid.
  Answer A5
  The whole paragraph has been rewritten as follows:
  Vertical irradiance profiles in the red and near-infrared channels at wavelengths beyond 600 nm were fitted using a sum of two exponential functions of which one represents the diffuse attenuation due to absorption and elastic scattering for the channel under consideration, while the other accounts for Raman scattering from the corresponding (shorter) excitation wavelengths significantly contributing to the upward light field (e.g., Sugihara et al. 1984). Not doing so, i.e., using just a single exponential function, leads to substantial underestimation of the extrapolated subsurface values due to significant Raman contribution from excitation wavelengths with significantly lower K values (see Appendix 1).
  Remark; “… from the figures in A1, it appears after ~1 m, the Eu(779) is no longer valid.”
  Figure A1 showing the near-surface depth dependence of EU at 779 nm demonstrates why the two-function approach is required. Below a depth of ca. 1.5 m, the signal is strongly dominated by Raman scattering and is therefore characterized by a diffuse attenuation coefficient very close to that of the corresponding excitation wavelengths.
  Comment C6
  Further, for Eu data, did you carry out correction of instrument self shading?
  Answer A6
  The following text has been added:
  Note that a correction to account for instrument self-shading of the upwelling irradiance was not applied. We assume the impact of this omission is relatively small except for highly absorbing waters (depending on constituent concentration and/or observation wavelength), where it may lead to an underestimation of near-surface upwelling irradiances.
  Comment C7
  The increase of Eu(683,706) with depth showing in Fig. 3 suggests this measurement is quite questionable.
  Answer A7
  As now described in the text, the observed increase of EU with depth at 683 nm and, to a lesser degree, at 664 and 706 nm is due to chlorophyll fluorescence associated with a deep chlorophyll maximum in the clear North-East Atlantic waters. This case has actually been chosen to demonstrate that the sensitivity of the radiometric measurements is sufficient to resolve such phenomena.
  Comment C8
  Suggest to include a section to specifically to talk about uncertainties associated with the measured parameters.
  Answer A8
  While uncertainties have not been assigned to individual values, statistical approaches may be applied by users to assign uncertainties on an aggregated level. In case a reprocessing of selected parameters is carried out, uncertainty related issues will receive a more thorough coverage (as far as the still information allows).
  Comment C9
  Fig. 7, for Y axis, suggest to use dynamic range, so the R spectra can be presented more clearly.
  Answer A9
  This was done according to the reviewer’s suggestions. Each panel in Fig. 7 has a free y-axis adjusted for its content. However, we have added a sentence to the caption to make sure that the reader is aware of the dynamic y-axis.
  The following text has been added to the caption of Fig. 07:
  Please also note the y-axes are adjusted to the data presented within each subplot.
  
  Citation: https://doi.org/10.5194/essd-2023-83-AC2
RC1: 'Comment on essd-2023-83', Emmanuel Boss, 28 Apr 2023

This paper is very important as it provides an introduction to a unique dataset to the community in its full glory. I strongly support publishing it. However I think the text can be significantly improved. In particular mentioning figures in sentences should be avoided and rather they should be cited in brackets while the text includes only what we learn from the data in those figures. Also, some figure captions (Fig. 3) include description of data characteristics that are never mentioned in the text. There are some 'Frenchisms' that could be made clearer (I made suggestions in the attached annotated PDF. Finally, the conclusion section does not do justice to the importance of the data set and the impact it had and likely will have in the future on the field of ocean optics in general and ocean color remote sensing in particular.
All the best, Emmanuel Boss, UMaine

Citation: https://doi.org/10.5194/essd-2023-83-RC1
RC2:
'Comment on essd-2023-83', David McKee, 02 May 2023

The Coastlooc project was a trailblazer for the marine optics community, providing an unusually rich and complete data set which remains relatively rare to this date. THe list of publictions that have previously used the data set is testament to its importance and this submission of a fully quality controlled version is a welcome step to ensure that it is available to the community for the future. I am certain that it will continue to be very highly valued and exploited as a result.
The authors have done a good job of describing the data set and I only have relatively minor comments in the attached marked up version. I would highlight the request to include uncorrected AC9 data so that users may select their own scattering correction.
Great job and delighted to see this data set being made available to the community.

Citation: https://doi.org/10.5194/essd-2023-83-RC2
- AC3:
  'Reply on RC2', Philippe Massicotte, 28 Jun 2023
  Reviewer RC1
  Comment C10
  This paper is very important as it provides an introduction to a unique dataset to the community in its full glory. I strongly support publishing it. However I think the text can be significantly improved. In particular mentioning figures in sentences should be avoided and rather they should be cited in brackets while the text includes only what we learn from the data in those figures. Also, some figure captions (Fig. 3) include description of data characteristics that are never mentioned in the text. There are some 'Frenchisms' that could be made clearer (I made suggestions in the attached annotated PDF. Finally, the conclusion section does not do justice to the importance of the data set and the impact it had and likely will have in the future on the field of ocean optics in general and ocean color remote sensing in particular.
  Answer A10
  Thank you very much for this review. Accordingly, we have significantly improved the text by addressing all the comments provided. Finally, we have completely rewritten the conclusion to better reflect the impacts of this data set on future ocean color remote sensing studies (see also A21). Please see below the answers to all the specific points raised. The caption to Figure 3 has been simplified, explanatory descriptions have been moved to the main text. The final version of the article will be checked by a native English speaker to minimize the presence of “Frenchisms” potentially confusing readers.
  Comment C11
  Line 38: Text improvement.
  Answer A11
  homogenous has been replaced by consistent
  
  complete has been replaced by comprehensive
  
  Comment C12
  Line 214: Citation typo.
  Answer A12
  Line 214: Reviewer is right, the year of the citation has been corrected and now reads Aiken et al., 2000.
  Comment C13
  Line 224: I don't think you can claim, given the number of samples measured, that you got the correct average IOP for each area.
  Answer A13
  To provide a better idea of the spectral variability of the measured IOPs across the sampled areas, individual absorption spectra have been plotted in light gray in Fig. 5A-D behind the averaged spectra.
  Comment C14
  Line 225: Text improvement.
  Answer A11
  Averaged has been replaced by measured
  
  by has been replaced with in each
  
  Comment C15
  Line 228: This is the first time this concept is mentioned.
  Answer A15
  “water color” has been removed from the sentence.
  Comment C16
  Line 234: Text improvement.
  Answer A16
  The word strengthened has been replaced with emphasized.
  Comment C17
  Line 241: This is interesting as it Kd is driven mostly by absorption. Does it means that the same particles that absorb at 443 are also responsible to scattering (e.g. chlorophyll containing ones?)
  Answer A17
  This is indeed interesting. The regressions between bp and a - aw, and between ap and bp shown at different wavelengths in the two figures below, confirm the strong correlation between these two IOPs. This is due to both the fact that Chl, Spm and CDOM all covary in this dataset (see Babin et al. 2003, JGR), and the fact that the particles that scatter light also absorb because of phytoplankton pigments (in the blue and red), and/or because of non-algal particles (at all wavelengths in the visible).
  Comment C18
  Line 265: Your style of writing could be significantly improved. This sentence should read: “Total chlorophyll a co-varied with POC (Fig. 4) and their relationship was highly variable (Fig. 9a).”
  Answer A18
  The sentence has been replaced with the suggested sentence.
  Comment C19
  Line 280: Are POC molecule more colorful than phytoplankton?
  Answer A19
  The sentence has been removed because we have no appropriate normalization at our disposal to figure this out. POC includes both NAP and phytoplankton, but we have no way to determine their respective contribution to POC.
  Comment C20
  Line 282: Text improvement.
  Answer A20
  “associated with a particular type of measure” as been replaced with “associated with each measurement”
  Comment C21
  Line 293: You should add something about the number of papers that have cited this dataset and why it is important for the community for this dataset to be released publically and consistently.
  Answer A21
  In the conclusion, we now refer to Table A1, which shows that nearly 40 studies have already been used in peer-reviewed journals, despite the COASTlOOC data not being easily accessible on a public and open repository. We also highlight the uniqueness of this dataset, which opens the door for future development and evaluation of new bio-optical models adapted for optically-complex waters.
  Comment C22
  Figure 3A: This issue is not mentioned once in the text. Why raise it only in the caption if at all?
  Answer A22
  Text has been added to the manuscript (see section 2.4.3 Irradiance depth merging). We also modified the caption and the figure accordingly. There were also requests to change some aspects of the figures which have been done.
  
  Citation: https://doi.org/10.5194/essd-2023-83-AC3
- AC4: 'Reply on RC2', Philippe Massicotte, 28 Jun 2023
  
  Reviewer RC2
  
  Comment C23
  
  The Coastlooc project was a trailblazer for the marine optics community, providing an unusually rich and complete data set which remains relatively rare to this date. THe list of publictions that have previously used the data set is testament to its importance and this submission of a fully quality controlled version is a welcome step to ensure that it is available to the community for the future. I am certain that it will continue to be very highly valued and exploited as a result.
  
  The authors have done a good job of describing the data set and I only have relatively minor comments in the attached marked up version. I would highlight the request to include uncorrected AC9 data so that users may select their own scattering correction.
  
  Great job and delighted to see this data set being made available to the community.
  
  Answer A23
  
  Thank you very much for the review. See below point by point all the corrections we have made based on your constructive comments.
  
  Comment C24
  
  Line: 97: The SPMR was intended to be used as a freefall profiler. Was it used in that mode from the helicopter? Looks like it was attached to the IOP package? Might be worth making this clear here too.
  
  Answer A24
  
  Good point. The first and second paragraphs of this section have been rewritten as follows to provide more details:
  
  The SeaWiFS Profiling Multichannel Radiometer (SPMR, Satlantic Inc, Canada) was used to measure downward (Ed, W m-2 µm-1) and upward (Eu, W m-2 µm-1) irradiance in the water column. Irradiance was measured at 13 wavelengths matching the MERIS channels of direct relevance for ocean observations (411, 443, 456, 490, 509, 531, 559, 619, 664, 683, 706, 779, 866 nm, except for COASTlOOC 1 operating at 590 nm instead of 619 nm) ranging from the blue part of the spectrum to the near-infrared at an acquisition rate of 6 Hz. The actual wavelengths differ slightly between upward and downward observations.
  
  Comment C25
  
  Line 97: Would be good to list the wavelength here.
  
  Answer A25
  
  See C24.
  
  Comment C26
  
  Line 106: This is a sensible decision, but you might want to note that this means there is potential for a degree of variability due to varying cloud cover to remain unresolved in those measurements.
  
  Answer A26
  
  The following text has been added to provide more information:
  
  Under stable atmospheric conditions, no significant error is expected to result from this approach, while it will contribute to increased variability in the retrieved parameters when sampling under varying cloud cover. The practical impact of this, however, is limited due to the short duration of helicopter-based sampling typically lasting between one and three minutes per profile.
  
  Comment C27
  
  Line 118: This is unclear. Can you provide more detail?
  
  Answer A27
  
  We looked into the old code again to clarify. Actually, a constant, channel-independent air-sea transmission factor has been used to account for losses of the downwelling irradiance (direct + diffuse) due to Fresnel reflection at the rough sea surface.
  
  The following text has been added to the sentence.
  
  “... by applying a constant factor of 0.943 to account for losses of the downwelling irradiance (direct + diffuse) due to Fresnel reflection at the rough sea surface.”
  
  Comment C28
  
  Line 170: What optical pathlength was used?
  
  Answer A28
  
  The optical path length was 10 cm. The precision has been added to the text.
  
  Comment C29
  
  Line 199: Other AC9 scattering correction approaches are available and are often preferred. Do you provide the uncorrected data? Would allow other scattering corrections to be applied.
  
  Answer A29
  
  The reviewer is right, there are other methods for scattering correction. As mentioned in the text, to correct absorption measurements for incomplete recovery of scattered light, a(715) was subtracted from a(<715). At this time, the raw data is not available. We will attempt to recover the complete AC9 data set and make it available on the SEANOE web page. We have added a precision about this in section 5.
  
  Comment C30
  
  Line 230: Typo
  
  Answer A30
  
  “estuaries” has been replaced with “estuarine”.
  
  Comment C31
  
  Line 234: Text suggestion.
  
  Answer A31
  
  See comment C16 where we replaced “strengthened” with “emphasized” given the comment of another reviewer.
  
  Comment C32
  
  Line 235: Text suggestion.
  
  Answer A32
  
  “distanced” has been replaced with “distant”.
  
  Comment C33
  
  Line 241: Text suggestion
  
  Answer A33
  
  The text now reads “with median bp(440) values varying” instead of “which median values varied”.
  
  Comment C34
  
  Line 248: Typo.
  
  Answer A34
  
  “or” has been replaced with “and”.
  
  Citation: https://doi.org/10.5194/essd-2023-83-AC4
- AC6: 'Reply on RC2', Philippe Massicotte, 28 Jun 2023
  
  Reviewer RC2
  Comment C23
  The Coastlooc project was a trailblazer for the marine optics community, providing an unusually rich and complete data set which remains relatively rare to this date. THe list of publictions that have previously used the data set is testament to its importance and this submission of a fully quality controlled version is a welcome step to ensure that it is available to the community for the future. I am certain that it will continue to be very highly valued and exploited as a result.
  The authors have done a good job of describing the data set and I only have relatively minor comments in the attached marked up version. I would highlight the request to include uncorrected AC9 data so that users may select their own scattering correction.
  Great job and delighted to see this data set being made available to the community.
  Answer A23
  Thank you very much for the review. See below point by point all the corrections we have made based on your constructive comments.
  Comment C24
  Line: 97: The SPMR was intended to be used as a freefall profiler. Was it used in that mode from the helicopter? Looks like it was attached to the IOP package? Might be worth making this clear here too.
  Answer A24
  Good point. The first and second paragraphs of this section have been rewritten as follows to provide more details:
  The SeaWiFS Profiling Multichannel Radiometer (SPMR, Satlantic Inc, Canada) was used to measure downward (Ed, W m-2 µm-1) and upward (Eu, W m-2 µm-1) irradiance in the water column. Irradiance was measured at 13 wavelengths matching the MERIS channels of direct relevance for ocean observations (411, 443, 456, 490, 509, 531, 559, 619, 664, 683, 706, 779, 866 nm, except for COASTlOOC 1 operating at 590 nm instead of 619 nm) ranging from the blue part of the spectrum to the near-infrared at an acquisition rate of 6 Hz. The actual wavelengths differ slightly between upward and downward observations.
  Comment C25
  Line 97: Would be good to list the wavelength here.
  Answer A25
  See C24.
  Comment C26
  Line 106: This is a sensible decision, but you might want to note that this means there is potential for a degree of variability due to varying cloud cover to remain unresolved in those measurements.
  Answer A26
  The following text has been added to provide more information:
  Under stable atmospheric conditions, no significant error is expected to result from this approach, while it will contribute to increased variability in the retrieved parameters when sampling under varying cloud cover. The practical impact of this, however, is limited due to the short duration of helicopter-based sampling typically lasting between one and three minutes per profile.
  Comment C27
  Line 118: This is unclear. Can you provide more detail?
  Answer A27
  We looked into the old code again to clarify. Actually, a constant, channel-independent air-sea transmission factor has been used to account for losses of the downwelling irradiance (direct + diffuse) due to Fresnel reflection at the rough sea surface.
  The following text has been added to the sentence.
  “... by applying a constant factor of 0.943 to account for losses of the downwelling irradiance (direct + diffuse) due to Fresnel reflection at the rough sea surface.”
  Comment C28
  Line 170: What optical pathlength was used?
  Answer A28
  The optical path length was 10 cm. The precision has been added to the text.
  Comment C29
  Line 199: Other AC9 scattering correction approaches are available and are often preferred. Do you provide the uncorrected data? Would allow other scattering corrections to be applied.
  Answer A29
  The reviewer is right, there are other methods for scattering correction. As mentioned in the text, to correct absorption measurements for incomplete recovery of scattered light, a(715) was subtracted from a(<715). At this time, the raw data is not available. We will attempt to recover the complete AC9 data set and make it available on the SEANOE web page. We have added a precision about this in section 5.
  Comment C30
  Line 230: Typo
  Answer A30
  “estuaries” has been replaced with “estuarine”.
  Comment C31
  Line 234: Text suggestion.
  Answer A31
  See comment C16 where we replaced “strengthened” with “emphasized” given the comment of another reviewer.
  Comment C32
  Line 235: Text suggestion.
  Answer A32
  “distanced” has been replaced with “distant”.
  Comment C33
  Line 241: Text suggestion
  Answer A33
  The text now reads “with median bp(440) values varying” instead of “which median values varied”.
  Comment C34
  Line 248: Typo.
  Answer A34
  “or” has been replaced with “and”.
  
  Citation: https://doi.org/10.5194/essd-2023-83-AC6
RC3:
'Comment on essd-2023-83', Jenny Lovell, 08 May 2023

General Comments

The COASTlOOC project dataset is a comprehensive and useful collection and clearly has been widely used. To standardize the QC, version control and documentation is an important task and serves to improve the standing and usability of the dataset. This paper has sufficient detail to describe the data and collection methods, and directs the reader to original sources for more detailed information. The paper includes useful plots for a user to decide whether the data is suitable for them as well as the full description of variables in Table 1.
Specific Comments

I would suggest that a text version of Table 1 would be useful to include with the dataset as a self-descriptor.
I followed the link in the paper and was able to access the data archive successfully. The default file-names of the downloaded files are not useful and would be better if they were descriptive and with some indication of version (in case of any future updates or errata).
The quality of the figures is generally good, but font sizes are too small.
Technical Comments

Additional detailed comments are annotated in the attached file.

Citation: https://doi.org/10.5194/essd-2023-83-RC3
- AC5: 'Reply on RC3', Philippe Massicotte, 28 Jun 2023
  
  Reviewer RC3
  Comment C35
  The COASTlOOC project dataset is a comprehensive and useful collection and clearly has been widely used. To standardize the QC, version control and documentation is an important task and serves to improve the standing and usability of the dataset. This paper has sufficient detail to describe the data and collection methods, and directs the reader to original sources for more detailed information. The paper includes useful plots for a user to decide whether the data is suitable for them as well as the full description of variables in Table 1.
  Answer A35
  Thank you very much for the useful comments. Please find, below, detailed responses for each comment.
  Comment C36
  I would suggest that a text version of Table 1 would be useful to include with the dataset as a self-descriptor.
  Answer A36
  We do not understand the question. The table is already present in the manuscript. We are willing to reconsider if the editor judges it so.
  Comment C37
  I followed the link in the paper and was able to access the data archive successfully. The default file-names of the downloaded files are not useful and would be better if they were descriptive and with some indication of version (in case of any future updates or errata).
  Answer A37
  The COASTlOOC data has been uploaded to SEANOE (Sea Open Scientific Data Publication) funded and operated by Ifremer (France) under DOI 10.17882/93570. If the data is to be updated, the DOI will also be updated, so that users have the full history of the data set.
  The names of the individual COASTlOOC data files were assigned by SEANOE. We agree that they are not very helpful to the users, and will therefore try to convince the SEANOE custodian to rename the files. If this is not possible, we will make sure that file names mentioned in the article are consistent with those in the repository.
  Comment C38
  The quality of the figures is generally good, but font sizes are too small.
  Answer A38
  This has been resolved in C2, the font size has been increased from 10 pt to 12 pt in all figures.
  Comment C39
  Line 10: Text suggestion.
  Answer A39
  still today has been deleted.
  Comment C40
  Line 5: Text suggestion.
  Answer A40
  has been was replaced with was.
  Comment C41
  Line 54: Text suggestion.
  Answer A41
  efficient sampling stations replace to efficiently sample stations.
  Comment C42
  Line 54: Typo.
  Answer A42
  The parenthesis has been closed.
  Comment C43
  Line 61: Text suggestion.
  Answer A43
  immediately processed has been replaced with processed immediately.
  Comment C44
  Line 64: Typo.
  Answer A44
  “s” from litters has been deleted.
  Comment C45
  Line 65: Text suggestion.
  Answer A45
  no longer now reads for no longer.
  Comment C46
  Line 75: Text correction.
  Answer A46
  onto now read on.
  Comment C47
  Line 106: perhaps comment on the typical time in-water so that the reader has an idea of the lag between in-air and in-water Ed.
  Answer A47
  Good point, see C26 where this comment was answered.
  Comment C48
  Line 154: Text suggestion.
  Answer A48
  It was suggested to replace downcast with down-cast. It appears that downcast is a valid English word. We are happy to reconsider if the editor judges it so.
  Comment C49
  Line 156: Text suggestion.
  Answer A49
  larger has been replaced with greater.
  Comment C50
  Line 157: Text suggestion.
  Answer A50
  do therefore has been replaced with therefore do.
  Comment C51
  Line 163: Text suggestion.
  Answer A51
  resolving has been replaced with resolution of.
  Comment C52
  Line 163: Text suggestion.
  Answer A52
  for example has been deleted.
  Comment C53
  Figure 12 and other font size comments.
  Answer A53
  Font size in all figures has been increased from 10 pt to 12 pt.
  Comment C54
  Table 1: it would make more sense to me if the Variable was the first column in the table. The source file name could be after the units, or the final column.
  Answer A54
  As suggested, columns have been swapped.
  Comment C55
  Table 1: for what? I assume this informatin is in meta-data associated with the dataset ?
  Answer A55
  We have provided a better description of this variable in Table 1. It now reads as:
  “Non-algal absorption adjusted so that baseline background is equal to that of ap”
  Comment C56
  Table 1: Typo.
  Answer A56
  organig has been replaced with organic.
  Comment C57
  Table 1: Date format? Just date or date-time? If including time, is it UTC?
  Answer A57
  It is just the date.
  Comment C58
  Table 1: Typo.
  Answer A58
  celsius now spells correctly.
  Comment C59
  Table 1: Is this correct? - density units would normally be kg/m^3 or kg/L
  Answer A59
  Sigma-t represents the density of a sea-water at atmospheric pressure, i.e., at the sea surface. It is defined as density minus 1000, where the density is measured in kg/m^3 and is typically expressed as a unitless number.
  
  Citation: https://doi.org/10.5194/essd-2023-83-AC5
AC1: 'Comment on essd-2023-83', Philippe Massicotte, 28 Jun 2023

Dear editors and reviewers. Please find in the attached PDF our points-by-points answers.
Regards,
Philippe

Citation: https://doi.org/10.5194/essd-2023-83-AC1

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

AR by Philippe Massicotte on behalf of the Authors (29 Jun 2023) Author's response Author's tracked changes

EF by Sarah Buchmann (30 Jun 2023) Manuscript

ED: Publish as is (04 Jul 2023) by Giuseppe M.R. Manzella

AR by Philippe Massicotte on behalf of the Authors (04 Jul 2023) Manuscript

Short summary

The COASTlOOC oceanographic expeditions in 1997 and 1998 studied the relationship between seawater properties and biology and chemistry across the European coasts. The team collected data from 379 stations using ships and helicopters to support the development of ocean color remote-sensing algorithms. This unique and consistent dataset is still used today by researchers.