the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Oil slicks in the Gulf of Guinea – 10 years of Envisat ASAR observations
- 1VisioTerra, 14 rue Albert Einstein Champs-sur-Marne, France
- 2Université Gustave Eiffel – Institut Gaspard Monge -IGM, 5 boulevard Descartes, Champs sur Marne
- 3Université de Sorbonne Université - Laboratoire d’Océanographie et du Climat : Expérimentations et Approches Numériques (LOCEAN), 4 Place Jussieu, 75005 Paris
- 4INRAE, ISPA, UMR 1391 INRAE/Bordeaux Sciences Agro, Villenave d'Ornon, France
- 1VisioTerra, 14 rue Albert Einstein Champs-sur-Marne, France
- 2Université Gustave Eiffel – Institut Gaspard Monge -IGM, 5 boulevard Descartes, Champs sur Marne
- 3Université de Sorbonne Université - Laboratoire d’Océanographie et du Climat : Expérimentations et Approches Numériques (LOCEAN), 4 Place Jussieu, 75005 Paris
- 4INRAE, ISPA, UMR 1391 INRAE/Bordeaux Sciences Agro, Villenave d'Ornon, France
Abstract. Gulf of Guinea is a very active area regarding maritime traffic as well as oil and gas exploitation (platforms). As a result of some actors of both sectors that fail to comply with environmental standards, the region is subject to a large number of oil pollutions. This study aims to detect oil slicks spilled in the Gulf of Guinea and analyse their spatial distribution using Synthetic Aperture Radar (SAR) images. If previous works have already locally mapped oil slicks in this area, this study is the first one to achieve a global statistical analysis based on a very high number of radar images covering 17 Exclusive Economic Zones of the Gulf of Guinea. To carry out the present study, a database of 3,644 SAR images, collected between 2002 and 2012 by the Advanced SAR (ASAR) sensor onboard the European Spatial Agency (ESA) Envisat mission has been used. This database allowed the identification of 18,063 oil slicks. These "Oil slicks" herein detected regroup: "oil spills" – of anthropogenic origin- and "oil seeps" – of natural origin (natural oil reservoir leaks).
Zhour Najoui et al.
Status: closed
-
RC1: 'Comment on essd-2022-126', Anonymous Referee #1, 05 Jun 2022
the paper requires editing for English - it is currently poorly written - especially the first part
specifics - see attached
-
AC1: 'Reply on RC1', Zhour Najoui, 28 Jul 2022
RC1-1 : the paper requires editing for English - it is currently poorly written - especially the first part
Thank you. English proofreading of the article is in progress.
RC1-2 : specifics - see attached
Thank you for the corrections. They have been integrated
-
AC1: 'Reply on RC1', Zhour Najoui, 28 Jul 2022
-
RC2: 'Comment on essd-2022-126', Anonymous Referee #2, 06 Jun 2022
The authors presented a study to detect oil slicks in Gulf of Guinea and analyze their distribution using synthetic aperture radar (SAR) data provided by Envisat mission. It identified 18,063 oil slicks in 3,644 SAR images, covering 17 exclusive economic zones of the Gulf of Guinea. The authors have put a good effort into mapping oil slicks and creation of a large database of SAR imagery. However, the referee has some specific concerns about this data, which need to be addressed before considering the manuscript for publication.
- In line 140, the authors stated that the oil slicks categorization has been done based on morphological and textural criteria. The referee is wondering whether it is enough to consider only morphology and texture to divide the detected/located oil slicks into natural and anthropogenic? There should be some examples that can highlight the limitation of the used categorization criteria and detection approach.
- In line 156-157, the authors stated that they have used some auxiliary data for validation of their analysis. Details about the auxiliary information used alongside manual detection approach should be clearly stated.
- In line 176, ‘for each class X of oil slick among (s) seepage’ should be replaced with ‘for each class X of oil slick among (e) seepage’.
- In line 211, ‘spill form ships’ should be replaced with ‘spill from ships’.
- In fig 7-9, please add a unit near to the color bar and its numerical values.
- The authors only shared the spatial distribution map of the detected oil slicks that is only useful for visualization purposes. The geocoded map should be shared along with the geographic location and time stamps of verified oil spills. Furthermore, the processed SAR data should be openly available as a benchmark dataset to develop some good manual/automated oil spill detection approaches.
- The authors should share any additional material associated with the verified oil spills that would help to model the flow of the spill. The time stamps of verified oil spills can help in tracing the oceanic parameters and accessing imagery over the same area from other spaceborne sensors.
- AC2: 'Reply on RC2', Zhour Najoui, 28 Jul 2022
-
RC3: 'Comment on essd-2022-126', Chuanmin Hu, 19 Jun 2022
The paper describes how the 11-year ENVISAT ASAR data were used to establish a database of oil slicks (both anthropogenic and natural origins) in the Gulf of Guinea, from which statistics for each country’s EEZ and for the entire Gulf were obtained.
Overall, the writing is relatively clear, and the information is useful for monitoring and mitigating oil pollution. But before final publication I’d like to ask to authors to consider the following comments.
The paper needs a thorough English proof reading. There are countless grammar errors, for example: “these oil slicks detected regroup” is awkward. Line 102: there is extra “ASAR”. Line 106: “free-oil” should be “oil-free.” There are others.
Fig. 1. Are these EEZs labeled by number or by country in the database? For example, if we want to find EEZ for Togo, which is it on the map? Lat lon labels are too small to read, so are the notations. Also, what is the data source of these EEZs?
Line 98, there is this most recent one to add: Dong et al. (2022), chronic oiling in global oceans. It would be nice to compare the findings from these two studies for the Gulf in a follow-on study.
Fig. 4. What date? What location? What length scale? How is it stretched? Where is the platform?
Line 138 – 155. This description is good, but please some examples to show how these different forms are differentiated. How is the boundary of each slick extracted manually? By drawing the boundary with a steady hand? If so, show examples of the original images and the manually drawn boundaries.
Line 159, what is “photo-interpretation”? This is no photo used in the study, right? I suppose you wanted to say “image-interpretation”, correct? There are several other places with the same word.
Line 159 – 164. This normalization is very good.
Eqs. 4-5: what is GG?
Line 202 – 2016. I couldn’t find how slicks due to ships, platforms, oil seeps are differentiated manually.
Fig. 5. It’s better to show the image itself, and then have a call-out box to show the oil platforms with yellow dots.
Figs. 7 – 9: What does density mean? For example, if a location has a number of 200, what does it mean?
Fig. 10. This is an important figure. Can we interpret the values as, for a given year, if there is just 1 huge image covering the entire EEZ, the value represents the oil slick area from this imaginary image? I think this is what “normalization” means.
Fig. 10. What caused the sudden drop in ship-induced oil spill after 2008?
Figs. 11 & 12. All fonts need to be enlarged. Also, it’s better to use a different y-axis range for different regions so the annual variations can be seen. Same for Figs. 13 & 14.
- AC3: 'Reply on RC3', Zhour Najoui, 28 Jul 2022
Status: closed
-
RC1: 'Comment on essd-2022-126', Anonymous Referee #1, 05 Jun 2022
the paper requires editing for English - it is currently poorly written - especially the first part
specifics - see attached
-
AC1: 'Reply on RC1', Zhour Najoui, 28 Jul 2022
RC1-1 : the paper requires editing for English - it is currently poorly written - especially the first part
Thank you. English proofreading of the article is in progress.
RC1-2 : specifics - see attached
Thank you for the corrections. They have been integrated
-
AC1: 'Reply on RC1', Zhour Najoui, 28 Jul 2022
-
RC2: 'Comment on essd-2022-126', Anonymous Referee #2, 06 Jun 2022
The authors presented a study to detect oil slicks in Gulf of Guinea and analyze their distribution using synthetic aperture radar (SAR) data provided by Envisat mission. It identified 18,063 oil slicks in 3,644 SAR images, covering 17 exclusive economic zones of the Gulf of Guinea. The authors have put a good effort into mapping oil slicks and creation of a large database of SAR imagery. However, the referee has some specific concerns about this data, which need to be addressed before considering the manuscript for publication.
- In line 140, the authors stated that the oil slicks categorization has been done based on morphological and textural criteria. The referee is wondering whether it is enough to consider only morphology and texture to divide the detected/located oil slicks into natural and anthropogenic? There should be some examples that can highlight the limitation of the used categorization criteria and detection approach.
- In line 156-157, the authors stated that they have used some auxiliary data for validation of their analysis. Details about the auxiliary information used alongside manual detection approach should be clearly stated.
- In line 176, ‘for each class X of oil slick among (s) seepage’ should be replaced with ‘for each class X of oil slick among (e) seepage’.
- In line 211, ‘spill form ships’ should be replaced with ‘spill from ships’.
- In fig 7-9, please add a unit near to the color bar and its numerical values.
- The authors only shared the spatial distribution map of the detected oil slicks that is only useful for visualization purposes. The geocoded map should be shared along with the geographic location and time stamps of verified oil spills. Furthermore, the processed SAR data should be openly available as a benchmark dataset to develop some good manual/automated oil spill detection approaches.
- The authors should share any additional material associated with the verified oil spills that would help to model the flow of the spill. The time stamps of verified oil spills can help in tracing the oceanic parameters and accessing imagery over the same area from other spaceborne sensors.
- AC2: 'Reply on RC2', Zhour Najoui, 28 Jul 2022
-
RC3: 'Comment on essd-2022-126', Chuanmin Hu, 19 Jun 2022
The paper describes how the 11-year ENVISAT ASAR data were used to establish a database of oil slicks (both anthropogenic and natural origins) in the Gulf of Guinea, from which statistics for each country’s EEZ and for the entire Gulf were obtained.
Overall, the writing is relatively clear, and the information is useful for monitoring and mitigating oil pollution. But before final publication I’d like to ask to authors to consider the following comments.
The paper needs a thorough English proof reading. There are countless grammar errors, for example: “these oil slicks detected regroup” is awkward. Line 102: there is extra “ASAR”. Line 106: “free-oil” should be “oil-free.” There are others.
Fig. 1. Are these EEZs labeled by number or by country in the database? For example, if we want to find EEZ for Togo, which is it on the map? Lat lon labels are too small to read, so are the notations. Also, what is the data source of these EEZs?
Line 98, there is this most recent one to add: Dong et al. (2022), chronic oiling in global oceans. It would be nice to compare the findings from these two studies for the Gulf in a follow-on study.
Fig. 4. What date? What location? What length scale? How is it stretched? Where is the platform?
Line 138 – 155. This description is good, but please some examples to show how these different forms are differentiated. How is the boundary of each slick extracted manually? By drawing the boundary with a steady hand? If so, show examples of the original images and the manually drawn boundaries.
Line 159, what is “photo-interpretation”? This is no photo used in the study, right? I suppose you wanted to say “image-interpretation”, correct? There are several other places with the same word.
Line 159 – 164. This normalization is very good.
Eqs. 4-5: what is GG?
Line 202 – 2016. I couldn’t find how slicks due to ships, platforms, oil seeps are differentiated manually.
Fig. 5. It’s better to show the image itself, and then have a call-out box to show the oil platforms with yellow dots.
Figs. 7 – 9: What does density mean? For example, if a location has a number of 200, what does it mean?
Fig. 10. This is an important figure. Can we interpret the values as, for a given year, if there is just 1 huge image covering the entire EEZ, the value represents the oil slick area from this imaginary image? I think this is what “normalization” means.
Fig. 10. What caused the sudden drop in ship-induced oil spill after 2008?
Figs. 11 & 12. All fonts need to be enlarged. Also, it’s better to use a different y-axis range for different regions so the annual variations can be seen. Same for Figs. 13 & 14.
- AC3: 'Reply on RC3', Zhour Najoui, 28 Jul 2022
Zhour Najoui et al.
Data sets
Spatial distribution of oil slicks in the Gulf of Guinea between 2002 and 2012 Zhour NAJOUI https://doi.org/10.5281/zenodo.6470470
Zhour Najoui et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 337 | 112 | 20 | 469 | 4 | 6 |
- HTML: 337
- PDF: 112
- XML: 20
- Total: 469
- BibTeX: 4
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1