Remote sensing monitoring and assessment of oil pollution from 2015 to 2020 in the Gulf of Guinea: The case of the two Congo and the Cabinda region (Angola)

Remote sensing monitoring and assessment of oil pollution from 2015 to 2020 in the Gulf of Guinea: The case of the two Congo and the Cabinda region (Angola) | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Remote sensing monitoring and assessment of oil pollution from 2015 to 2020 in the Gulf of Guinea: The case of the two Congo and the Cabinda region (Angola) Marius MASSALA MBINDZOUKOU, Sylvie Brizard ZONGO, Bruno NKOUMAKALI, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3860518/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Several countries in the Gulf of Guinea have economies based on the exploitation of raw materials, particularly oil. This oil exploitation is not without consequences for marine and coastal ecosystems. Maritime pollution by hydrocarbons is one of the factors affecting marine areas. Having a better understanding of this environment makes it possible to assess the impact of an oil spill on the environment or to study the effects of chronic pollution associated with shipping lanes, ports, oil platforms, pipelines or refineries. It is essential that oil spill detection and monitoring systems respond quickly to contain these hazards. This is why we believe that the use of remote sensing appears to be a potential avenue for the detection of oil slicks at sea. The present work is a retrospective study of oil slick pollution from 2015 to 2020. We propose a methodology based on radar imagery to highlight the presence of ocean oil slicks due to oil activities. Using the Sentinel 1 satellite, we mapped all the oil slicks, which enabled us to calculate the areas where the hydrocarbons and any oil spills spread. The results clearly show that the Congo region is subject to frequent oil spills. oil slick pollution remote sensing sea cartography Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction The Gulf of Guinea region in Africa is a vague area of contiguous maritime and continental space along the West African coast. In terms of oil geopolitics, some authors limit the Gulf of Guinea to the oil-producing countries along the West African coast, stretching from Côte d'Ivoire to Angola (Kounou 2009). This zone is characterized by the abundance of natural resources, particularly oil (Ndoutoume 2010). The economies of these countries heavily rely on oil exploitation. For example, in 2000, the oil production of the Gulf of Guinea countries accounted for approximately 3.8 billion tons, representing around 5% of global production. The good quality of its crude oil attracts the interest of numerous investors (Amoussou 2018). The intensification of oil activities in this region leads to an increase in anthropogenic activities, resulting in oil spills. Analysis of some studies in the field shows that between 2002 and 2012, the Republic of Congo had a high probability of hydrocarbon spills, following Nigeria and Cameroon (Amoussou 2018). Similarly, Ngoma (2015) shows that the Democratic Republic of Congo (DRC) experienced extensive hydrocarbon spills between January and April 2010. In Gabon, particularly at the Cap Lopez terminal (Port-Gentil), an incident in 2022 due to a storage tank leak resulted in an oil spill of approximately 50,000 cubic meters (Malouana Biggie 2022 ). These cases highlight the extent of the pollution problem related to oil exploitation in this region. It is important for scientists to guide decision-makers toward suitable solutions that would help strengthen the region's legislation. In addressing this issue, we believe that the use of remote sensing can monitor the expansion of oil spills in marine environments. The use of Earth observation tools, especially SAR imagery, appears to be a potential avenue for identifying and mapping areas with high oil pollution. To construct our reasoning, we relied on the approaches detailed by Najoui ( 2017 ) in his literature review. In his work, he explains that the detection of marine oil slicks includes four basic steps: 1) images preprocessing, 2) dark patches segmentation, 3) features extraction and 4) oil slicks classification (Solberg et al. 1999 ; Brekke and Solberg 2005 ; Topouzelis and Konstantinos 2008). Generally, in the literature, the image preprocessing (first step) is limited to speckle filtering on standard detected products. Unfortunately, so many heterogeneities remain in the radar images after such classical preprocessing hindering the "robustness" of the segmentation and the classification methods especially when working on large areas. The work of this paper focuses on image preprocessing and dark patches segmentation. Due to the incidence angle dependencies (SAR images tend to become darker with increasing range), upwind/downwind or crosswind, and swath width effects, brightness variations may occur in the SAR images and hence compromise the processing of SAR images. A variety of segmentation methods have been proposed and are listed below: adaptive thresholding (Solberg et al.1999), hysteresis thresholding (Kanaa et al. 2003 ), edge detection using Laplace of Gaussians or Difference of Gaussians (Chang et al. 2008 ), wavelets (Liu et al. 1997 ), mathematical morphology (Gasull et al. 2002 ), neural network (Garcia-Pineda et al. 2009; Angiuli et al. 2006 ; Del Frate et al. 2013), etc., etc. Even though a variety of segmentation methods have been applied, the most frequently used are based on a local analysis to overcome the brightness variations in the SAR images. In the study area, scientific work on the detection of oil slicks at sea by remote sensing is virtually non-existent. Previous studies have investigated the pollution problems (Kounou 2009; Malouana 2022; Mbaki 2003 ) and mapped local oil slicks in the Gulf of Guinea area (Najoui 2017 ; Najoui 2022a ; Najoui 2022b ; Najoui 2018a; Najoui 2018b; Okafor 2018 ). However, to our knowledge, this study is the first to carry out a statistical analysis followed by the proposal of a first intelligent GIS approach capable of monitoring the evolution of slick drift followed by their change of state. This work aims to evaluate particularly in the areas of the two Congos and Cabinda (Angola) from 2015 to 2020. Our methodology is based on the pre-processing of radar images to facilitate the semi-automatic detection of oil slicks and then to make the semi-automatic recognition of marine oil slicks more reliable. The main objective is to implement an intelligent GIS for data prediction and analysis. In this article, we will present the materials and methods section, followed by the presentation and discussion of the results in section 3 . Finally, we conclude with some perspectives. 2. Study area The study area of this work is located at the confluence of three countries: The Republic of Congo, the Democratic Republic of Congo (DRC), and the Republic of Angola through its Cabinda region. Regarding the Republic of Congo, it is situated in the center of Africa in the coastal zone. This country spans 162 km of coastline, featuring bays and points such as Pointe Noire, Indienne, and Kounda. The map below (Fig. 2 ) illustrates the selected study area for this work. It specifically covers the maritime frontage of both Congo and Cabinda. This region is of particular interest for several reasons, as it is situated at the convergence of three oil-producing countries in a confined space and hosts multiple oil platforms. 3. Methodology The analysis of radar images for hydrocarbon slick detection has been the subject of several studies. Generally, three approaches emerge: a manual approach conducted by trained human operators who analyze the images to detect hydrocarbon slicks. This approach is very rare. The semi-automatic approach where a computer detects all black objects in the radar image using various segmentation techniques, after which an experienced human operator classifies these objects as oil slicks or look-alikes. Finally, the automatic system that uses complex image processing and programming techniques to perform both segmentation and classification. In our case, we have opted for the semi-automatic approach considering the software tools at our disposal. 3.1 Downloading The collection of Sentinel-1 data was carried out on the ASF Data Search platform. This platform is a satellite data archiving platform developed by the University of Alaska's Institute of Geophysics and managed by National Aeronautics and Space Administration (NASA). Downloading the data is free for all users. We proceeded with downloading the data for the period from 2015 to 2020. This data acquisition step allowed us to collect approximately 644 Sentinel-1A sensor data on a day-to-day basis over the five years. These data were collected according to the following parameters: Table 1 Characteristics of downloaded data CHARACTERISTIC DATA TYPES Number of data collected 644 Mission SENTINEL 1_A Period 2015–2020 Acquisition mode IW Product type GRD Direction ASCENDANT Polarisation VV, VH Instrument name C-band synthetic aperture radar Swath 250 Kilometers Resolution 5X20 meters 3.2. Data pre-processing The downloading step was followed by a data pre-processing phase. This phase aimed to enhance the perception of certain details present in each of the images. The software used for these treatments was SNAP. Several reasons guided us in choosing this software: its free accessibility, its ability to perform quick pre-processing, and its reliability in detecting dark spots in radar imagery. The pre-processing step was preceded by a phase that involved reducing the sizes of the images. Indeed, with a Swat of approximately 250 km and a resolution of 5 meters by 20 meters, the size of the images was very high. This situation made the pre-processing phase very challenging. Faced with this difficulty, we created sub-images taking into account the areas of interest. This step facilitated the processing time of the algorithms. After this step, we moved on to the pre-processing phase. This process unfolded in the following steps: calibration, ellipsoid correction, multilook , land-sea mask application, conversion to decibels, and speackle filter . The complete description of these steps is as follows: Calibration : It consisted of obtaining values at the level of the images which are representative of the target surface sought. The multilook process, on the other hand, helped reduce the speckled effects on the images, which hinder the visualization of all the information present in each image. For this step, we utilized the multilook tool. This tool allows for the combination of multiple images incoherently, as if they corresponded to different views of the same scene. The multilook tool has the effect of improving the image interpretation process and enhancing the execution of the speckle filter. The ellipsoid correction was applied to ensure that the representation of the processed image in SNAP is as close as possible to the reality on the ground. The land-sea mask tool was used to mask terrestrial data and represent the entire marine domain of the target area. The speackle filter was used to mitigate the overall noise or shimmer present in the raw images. This noise can disturb the perception of objects or the detection of dark spots in the image. This process aims to highlight dark spots and contextual elements by reducing unwanted interference. The conversion to decibels was used to better visualize oil slicks and differentiate them from their look-alikes. We performed a reading of the processed image in decibels. This enhances the contrast of the image and characterizes the intensity across the entire image. Each area of the image corresponds to a specific intensity, which is high or positive in brighter areas and low or negative in darker areas. Through this conversion, one can read in decibel values the echoes recorded by the radar sensor. This recorded echo is representative of the backscattering intensity and the behavior of electromagnetic waves when interacting with the target area. Following the pre-processing in SNAP, we identified areas with a high presence of oil slicks solely of petroleum origin. However, some look-alikes may resemble hydrocarbon slicks with the same spectral signature. Depending on the look-alikes, the backscattering intensity in decibels read on the radar image after conversion varies. Generally, upon observation, only certain look-alikes, such as biological slicks, often exhibited spectral signatures quite similar to those of hydrocarbon slicks. However, we eliminated them since they typically develop in clusters, far from oil platforms. After this work, we proceeded to save the results in GeoTIFF format, which is useful for future processing with QGIS software. 3.3 Data processing After the pre-processing phase, we exported all the data to the Quantum GIS (QGIS) software for the digitization of detected pollution. In total, we identified and digitized 244 hydrocarbon slicks. This digitization led to the creation of vector data, which were subsequently classified by acquisition date. Next, the calculation of the areas of each vector's attributes was performed. The attribute data of each vector were then exported to a spreadsheet for the creation of various graphs. The last phase of these treatments was the production of an annual map of hydrocarbon pollution slicks. Figure 2 below schematically describes all the aforementioned steps. 3.4 GIS modelling To build the future intelligent GIS, we believe it is necessary to define certain key concepts. To do this, we consider that at a given moment an oil slick can be assimilated to a polygonal geographical object denoted O. We assume that the latter is compact and is characterised by the geographical position (X and Y) of its centroid. Based on this principle, we can say that O is a dynamic object as a function of time. Wagener (2006) describes that an oil slick at sea goes through 4 possible successive states, which we represent in the following figure: State 1 represents Film formation; state 2 represents Dissipation; state 3 represents Foam and finally state 4 Tar balls. It follows that the model discretises a continuous dynamic process. The transition from one state to another is governed by a transition or fragmentation function, which is a stochastic function. This function transforms the slick by taking it through the 4 states we have just described. These are divided into four groups: water, wind, the initial size of the slick and the configuration of the coast. These are: Concerning the water: there are 4 variables: the surface temperature, which is a random variable, the tides, the sea current if it exists, and the chemical characteristics of the sea water in the area crossed by the oil slick; Concerning the wind: this is characterised by variables such as the speed, direction and temperature of the air, which generates turbulence and therefore random behaviour; The initial size of the slick; The coastline: which is characterised by the nature of the coastline (sand, reef, mangrove, etc.) and the distance from the initial slick. These variables give the function its stochastic character and will influence the rate of dispersion and slick fragmentation as follows: Slick stretch: this is the first effect of the transition function. It will depend in particular on the initial size of the slick, the water temperature, the wind and the tides; Dissipation: in this state, the function will break up the slick. The size of the fragments will depend closely on the turbulence at the water surface, induced essentially by air turbulence; Foam: this state of the slick will be influenced to a large extent by the chemistry of the water and its temperature; Tar: the function ends its effect by determining the size of the pellets, which will depend on the nature of the coastline. The formal description of the function requires an in-depth statistical study accompanied by simulations on high-performance infrastructures, which we do not have. However, as an indication, the transition function can be broken down into a series of 3 functions: f1, f2, f3, which characterise the passage of the slick from one state to another (from state 1 to 2 for f1, then from 2 to 3 for f2, and finally from 3 to 4 for f3). The definite advantage of such a study is that it will enable us to predict the behaviour of hydrocarbon pollution in the Gulf of Guinea area, with a view to effectively implementing the appropriate responses likely to reduce its impact on ecosystems. Following the presentation of the methodology, we will now detail the results of our treatments. 4. Results The mapping of oil slicks was carried out step by step over the selected 5 years. The figure below illustrates one aspect of this work. After the mapping work from 2015 to 2020, we have summarized all of these results through the table below. Table 2 Repair per year of the quantity of slicks detected From this identification work, we subsequently calculated the pollution surfaces at the water tables. Year CB DRC Cabinda Total 2015 23 22 19 64 2016 17 17 2 36 2017 17 12 3 32 2018 13 24 8 45 2019 14 16 4 34 2020 12 24 0 36 Total 96 115 36 Table 3 Area in km² occupied by groundwater per year Year CB DRC Cabinda Total 2015 132,984 113,47 57,987 304,441 2016 35,032 156,006 15,923 206,961 2017 58,754 93,114 8,056 159,924 2018 50,438 106,464 42,171 199,073 2019 28,82 104,961 2,937 136,718 2020 52,882 131,71 0 184,592 Total 358,91 705,725 127,074 From the above results, we can deduce a regression in pollution due to hydrocarbons from 2015 to 2020. This drop is estimated at around 97.4 km², or a reduction of around 8% as mentioned above. Thus between 2016 and 2020 the spreading surfaces of hydrocarbon slicks from oil concessions vary between 150 and 200 km². We also clearly see that the most recurrent and highest pollution areas are observed in the DRC's EEZ. In 2015 we observed around in the EEZ of the Democratic Republic of Congo (ZEE COD) a surface area of spread of hydrocarbon slicks of approximately 110 km² compared to 130 and 50 for the EEZ of the Republic of Congo (ZEE COG) and Cabinda. However, between 2016 and 2020, the pollution areas observed in the DRC are almost greater than 100 km², up to 150 km² in 2016 and in the case of the exclusive economic zones of the Republic of Congo and Cabinda, the pollution areas are almost less than 50 km². Km². 5. Discussion The results obtained show large areas of spreading of oil slicks which may be due to frequent and uncontrolled spills. During data collection we used 644 pieces of data. Depending on the number of data downloaded per year, the probability of detecting slicks is greater. Thus, the data collected in 2015 and 2016 are not very representative compared to that collected between 2017 and 2020. The observation that emerges is that this trend is decreasing year after year. To reduce margins of error in detection, we used a stochastic approach which consists of obtaining a large number of samples or data in order to optimize the detection of oil slicks, this approach was carried out in the framework of certain works such as those of (Najoui 2017 ; Amoussou 2018). A more detailed analysis shows that the DRC is the most impacted country. The EEZ of the Democratic Republic of Congo has a strong presence of oil slicks. We also observed that the average surface area covered by oil slicks is 117.6 km² in the DRC and 59.81 km² for Congo Brazzaville. However, some works, notably those of (Najoui 2022), have shown that the DRC had an average of low pollution areas (around 50 km²) compared to that of the Republic of Congo Brazzaville (250 km²) 2002 and 2012. We noted that the DRC has heavily polluted after 2015. From a general point of view, the results show a decreasing trend of around 8% corresponding to a reduction of around 97.4 km². This percentage is, however, put into perspective by the DRC which is an exception to the rule. Several potential avenues can explain this situation for this country. Firstly, the weakness of legislation in the face of the power of oil lobbies. Secondly, the insufficiency of controls by state authorities and finally thirdly, the obsolescence of production equipment leading to potential accidents. As we see in our results, the Congo Basin is strongly threatened by groundwater pollution. This threatening situation weighs on the preservation of coastal and marine ecosystems. In this area of preservation, scientific work has shown that when such phenomena occur, the presence time of hydrocarbons as a function of the volume spilled into the marine environment depends on its density (Wegener 2006). This work shows that for a density lower than 0.8 or between 0.8 and 0.85 the presence time could be expressed between 1 hour and 2 hours. These data differ slightly when it comes to light crude oil (density between 0.85 and 0.95) and heavy oil (density above 0.95), the dwell time could be estimated between a week and a year (Wegener 2006). These uncontrolled spills can often appear on coasts in the form of tar or foam, which cause numerous disruptions to coastal ecosystems and a reduction in the resilience of mangrove sites (Wegener 2006). Another consequence linked to this phenomenon is the threat to planktonic species, fish generally in the juvenile stage, marine birds whose soiling of their plumages is the most significant effect of this pollution, marine mammals which during their outings surface are left with damage to nasal tissues due to hydrocarbons (ITPFO 2013 ). In order to reduce the risk of error in identifying slicks in the results we have just presented, we relied on the principle that oil spreads very quickly on the surface of the sea. About twelve hours after a spill, the slick can cover an area extending over several kilometres (PNIU, 2012 ). To enable us to map oil slicks effectively, we relied on the sensor characteristics of the Sentinel 1 satellite, which is capable of observing an area once every 12 days. By combining the orbital characteristics of Sentinel 1_A and 1_B, this gives us an exact repeat cycle of 6 days at the equator. With this approach, we believe we have considerably reduced the risk of duplicates in the same area. However, this probability is not zero, since throughout the drift of the slicks, there are changes in the content and shape of the oil slicks. With regard to the methodology we have adopted, it is important to emphasise that failure to take into account factors such as wind speed, polarisation, incident angle, the effect of dielectric properties and the nature of oil slicks has a major influence on the process of detecting hydrocarbon products at sea. This is why we believe that any GIS modelling must take into account the states that a slick can experience. 6. Conclusion and perspectives As underlined by Najoui ( 2017 ), radar sensors are commonly used in oil spill monitoring systems because of their well because of their proven detection capability. The emergence of new satellites that are more efficient with larger volumes of data makes automatic oil spill oil slick detection a necessity. However, oil slick detection is a multivariate phenomenon that depends on several factors. Multivariate phenomenon that depends on several parameters. The present work, which is a retrospective study, made it possible to highlight the phenomenon of pollution between 2015 and 2020. This study highlights that the maritime coast of the two Congos and Cabinda is exposed to numerous oil spills with large sprawling areas of up to 300 km 2 . However, this work remains to be improved by integrating not only the automatic approach into the processing of slick detection, but also models for deriving the objects identified using artificial intelligence. Such an approach would reduce the risk of redundancy in the process of identifying oil slicks during the scanning phases. References Amoussou, N. L. 2019. Data capitalisation and oil slick detection in the Gulf of Guinea from Envisat radar images 2002-2012 . Sorbone2019. Angiuli, E., F. D. Frate, and L. Salvatori. 2006. “ Neural networks for oil spill detection using ERS and ENVISAT imagery .” In Proceedings of SeaSAR’06, 23-26 January 2006, Frascati, Italy, 1–6. Brekke, C. and A.h.S. Solberg. 2005. "Oil Spill Detection by Satellite Remote Sensing." Remote Sensing of Environment 95 (2005): 1-13. Chang, L., Z. S. Tang., S. H. Chang, and Yang-Lang, C. 2008. “ A region-based GLRT detection of oil spills in SAR Images .” Pattern Recognition Letters 29 (14): 1915–23. doi:10.1016/j.patrec.2008.05.022. Del Frate, Fabio, D. Latini, C. Pratola, and F. Palazzo. 2013. “ PCNN for automatic segmentation and information extraction from X-band SAR imagery .” International Journal of Image and Data Fusion 4 (1): 75–88. doi:10.1080/19479832.2012.713398. EGE. 2022. Oil, the key to power and influence in the Gulf of Guinea . 2022. ETOGA. 2023. Governance of Marine and Coastal Biodiversity in the Gulf of Guinea . University of Nice-Sophia Antipolis, 1-192. Garcia-Pineda, O. 2009. “Spatial and temporal analysis of oil slicks in the gulf of Mexico based on remote sensing” . Texas A&M University. Gasull, A., X. Fábregas, J. Jiménez, F. Marqués, V. Moreno, and M. A. Herrero. 2002. “Oil spills detection in SAR images using mathematical morphology.” In 11th European Signal Processing Conference (EUSIPCO 2002), 25–28. Toulouse, France. doi:10.1.1.81.3086. Houndegnonto Odillon J. 2022. Analysis of thermohaline variations in the intraseasonal scales of the freshwater beds of the Goffe de Guinee. HAL Open Science , 13-185. ITPFO. 2013. Technical Information Guide ,The Effect of Oil Pollution on the Environment. Cantebury: Impact PR & Design Limited. 2013 Kanaa, T. F. N., E. T. G . Mercier., V. P. Onana., J. M. Ngono., P. L. Frison., J. P. Rudant, and R. Garello. 2003. “Detection of oil slick signatures in SAR images by fusion of hysteresis thresholding responses.” In IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477), 4:2750–52. IEEE. doi:10.1109/IGARSS.2003.1294573. Kounou Michel 2009. Oil and Poverty South of the Sahara: Analysis of the Foundations of the Political Economy of Oil in the Gulf of Guinea, Yaoundé, Clé, 2006, p. 21. Liu, A. K., C. Y. Peng, and S. Y. Chang. 1997. “Wavelet analysis of satellite images for coastal watch.” IEEE Journal on Ocean Engineering 22 (1): 9–17 Malouana Biggie. 2022. Serious oil spill at the Cap Lopez terminal. Port-Gentil: Gabon Review. Mbaki Esther Pabou. 2003. The Congo unarmed in the face of oil pollution. VertigO - la revue électronique en sciences de l'environnement [On line], Regards / Terrain, online 01 May 2003, accessed 01 December 2023.. URL : http://journals.openedition.org/vertigo/4856 ; Monde Afrique. 2022. Congo Brazaville under power lines, villages in the dark. 12 October 2022. Najoui, Z., N.Amoussou., S. Riazanoff., G. Aurel, G., and F. Frappart. 2022. Oil slicks in the Gulf of Guinea – 10 years of Envisat Advanced Synthetic Aperture Radar observations. Earth Syst. Sci. Data 14 : 4569–4588, https://doi.org/10.5194/essd-14-4569-2022, 2022. Najoui, Z. 2017. Prétraitement optimal des images radar et modélisation des dérives de nappes d'hydrocarbures pour l'aide à la photo-interprétation en exploration pétrolière et surveillance environnementale,https://pdfs.semanticscholar.org/92b2/e8e06b49d7f31c0847c694f4b4f3bea41222.pdf?_ga=2.235046721.1549874629.1594648502-969427726.1594648502 (last access: 14 October 2022), 2017. Najoui, Z. 2022a. 100 geolocated oil spills in the Gulf of Guinea (1.0), Zenodo, https://doi.org/10.5281/ZENODO.6907743, 2022a. Najoui, Z. 2022b. Spatial distribution of oil slicks in the Gulf of Guinea between 2002 and 2012, Zenodo, https://doi.org/10.5281/ZENODO.6470470, 2022b. Najoui, Z., Riazanoff, S., Deffontaines, B., and Xavier, J.-P. 2018a. A Statistical Approach to Preprocess and Enhance C-Band SAR Images in Order to Detect Automatically Marine Oil Slicks, IEEE T. Geosci. Remote, 56, 2554–2564 https://doi.org/10.1109/TGRS.2017.2760516, 2018a Najoui, Z., Riazanoff, S., Deffontaines, B., and Xavier, J.-P. 2018b. Estimated location of the seafloor sources of marine natural oil seeps from sea surface outbreaks: A new “source path procedure” applied to the northern Gulf of Mexico, Mar. Petrol. Geol., 91, 190–201, https://doi.org/10.1016/j.marpetgeo.2017.12.035 Ngodi, E.: Gestion des ressources pétrolières et développement en Afrique, Présentation à la 11ème Assemblée générale du CODESRIA (6–10 December 2005), Maputo, Mozambique, 2005. Ndoutoume xxxxxx. 2010. Terrorism and piracy: what security for the seas of the Gulf of Guinea? FES, 159-182. Ngoma Khuabi Camille. 2015. Corporate social responsibility and repair of ecological damage due to oil pollution in the coastal city of Moanda in the Democratic Republic of Congo. KAS African Law Study Library, 9-41. Okafor-Yarwood, I.: The effects of oil pollution on the marine environment in the Gulf of Guinea–the Bonga Oil Field example, Transnatl. Leg. Theory, 9, 254–271, https://doi.org/10.1080/20414005.2018.1562287, 2018. Tchetchoua Tchokonte Severin. 2019. Oil Issues and Games in Africa: Study of the Chinese Oil Offensive in the Gulf of Guinea, Master's Thesis in Political Science, University of Yaoundé II, 2007-2009. Ona Ona Jenifer Octavie. 2019. Sustainable management of fisheries resources in Central East Atlantic Africa: Cameroon-Congo-Gabon. Nantes: 19 December 2019. PNIU. 2012. Behaviour and Fate of Hydrocarbons. Tunisia Topouzelis, Konstantinos N. 2008. Oil spill detection by SAR Images: Dark formation detection, feature extraction and classification algorithms. Sensors 8 (10): 6642–59. doi:10.3390/s8106642. Solberg, A. H. S., G. Storvik, R. Solberg, and E. Volden., 1999. Automatic detection of oil spills in ERS SAR images. IEEE Transactions on Geoscience and Remote Sensing 37 (4): 1916–24. doi:10.1109/36.774704 Wald Lucien, Jean-Marie Monget, & Michel Albuisson. 2010. Oil pollution in the Mediterranean as seen by the Landsat satellite. Hal Open Science, 61-68 Wegener Angela. 2006. Marine pollution, case study. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 20 Jan, 2024 Reviewers invited by journal 19 Jan, 2024 Editor invited by journal 12 Jan, 2024 Editor assigned by journal 12 Jan, 2024 First submitted to journal 12 Jan, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3860518","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":268279764,"identity":"808dd0af-e8cb-4297-81e2-31905bfb9658","order_by":0,"name":"Marius MASSALA MBINDZOUKOU","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIiWNgGAWjYJCCA1Ca8QBDBZBiZm4gWguQcQakhZGwFoRexjawbfi16LaffXiY5882efn+ww8O/JxXG83fDtTyo2IbTi1mZ9INDvO23TbccCPN4GDvtuO5Mw4zNjD2nLmNW8uBNIbDvA23GTdIMBgc4N12LLcBqIWZsQ2PlvPPGIAOu20/v//4h4N/5xzLnU9Qyw2gLTxstxMbDuQAXdhQk7uBsJZnDAfntt1O3nAjp+CwzLEDuRuBWg7i9cv5NOYPb/7ctgU6bOPDNzV1ufPOHz744EcFbi3o4DCYPEC0eiCoI0XxKBgFo2AUjBAAALtfaJ6fp2aMAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-5867-2956","institution":"Université Omar Bongo: Universite Omar Bongo","correspondingAuthor":true,"prefix":"","firstName":"Marius","middleName":"MASSALA","lastName":"MBINDZOUKOU","suffix":""},{"id":268279765,"identity":"33b9388d-b331-42ff-9233-e131af6da21b","order_by":1,"name":"Sylvie Brizard ZONGO","email":"","orcid":"","institution":"Université des Sciences et Techniques de Masuku: Universite des Sciences et Techniques de Masuku","correspondingAuthor":false,"prefix":"","firstName":"Sylvie","middleName":"Brizard","lastName":"ZONGO","suffix":""},{"id":268279766,"identity":"766346d1-1d04-4fdb-affd-369194823e65","order_by":2,"name":"Bruno NKOUMAKALI","email":"","orcid":"","institution":"Université Omar Bongo: Universite Omar Bongo","correspondingAuthor":false,"prefix":"","firstName":"Bruno","middleName":"","lastName":"NKOUMAKALI","suffix":""},{"id":268279767,"identity":"296eef82-97e0-44a5-87e8-8881c0df74df","order_by":3,"name":"Aboubacar MAMBIMBA NDJOUNGUI","email":"","orcid":"","institution":"Agence Gabonaise d'Etudes et d'Observations Spatiale","correspondingAuthor":false,"prefix":"","firstName":"Aboubacar","middleName":"MAMBIMBA","lastName":"NDJOUNGUI","suffix":""},{"id":268279768,"identity":"7abc4940-e5ee-475c-b94a-71e0b5bc5df1","order_by":4,"name":"Johan Lain IVALA","email":"","orcid":"","institution":"Agence Gabonaise d'Observations Spatiales","correspondingAuthor":false,"prefix":"","firstName":"Johan","middleName":"Lain","lastName":"IVALA","suffix":""},{"id":268279769,"identity":"da53a3f0-e0f9-4a51-b355-6f0510a771c8","order_by":5,"name":"Loïc E BLEKPON HOUNGUEVOU","email":"","orcid":"","institution":"Ecole Nationale des Eaux et Forets","correspondingAuthor":false,"prefix":"","firstName":"Loïc","middleName":"E BLEKPON","lastName":"HOUNGUEVOU","suffix":""}],"badges":[],"createdAt":"2024-01-13 15:22:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3860518/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3860518/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":50116237,"identity":"2c489149-0a08-4ee3-aad6-8884e8d69a27","added_by":"auto","created_at":"2024-01-24 18:49:02","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":44595,"visible":true,"origin":"","legend":"\u003cp\u003eDelimitation of the area of interest in the Congo region.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/cd75f68ba16b48d19c9dd32c.jpg"},{"id":50116241,"identity":"1ca91145-ec4a-4c75-98b0-2a474ad9bb25","added_by":"auto","created_at":"2024-01-24 18:49:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":60689,"visible":true,"origin":"","legend":"\u003cp\u003eOverall diagram of the methodology for identifying oil slicks\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/2d70ac0b80ba6f6de9488982.png"},{"id":50116238,"identity":"2faeeff5-8021-4b78-9327-4b8dcf8640e1","added_by":"auto","created_at":"2024-01-24 18:49:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":20427,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentation of the different states of an oil slick Wagener (2006)\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/4a8cc2335896847e19bd4a37.png"},{"id":50116243,"identity":"f5accad2-e35a-4838-b834-d0864cc1d51f","added_by":"auto","created_at":"2024-01-24 18:49:02","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":50665,"visible":true,"origin":"","legend":"\u003cp\u003eExample of mapping the distribution of oil slicks in 2015\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/65b77a6dbd52e10d35da69f2.jpg"},{"id":50117384,"identity":"1c5f1ed4-0e76-4804-a657-bc39713d937e","added_by":"auto","created_at":"2024-01-24 18:57:02","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":60745,"visible":true,"origin":"","legend":"\u003cp\u003eExample of mapping the distribution of oil slicks in 2016\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/0ecf335e577578a7454e5e2a.jpg"},{"id":50116239,"identity":"cefa4c4b-4c12-4a28-9abb-c7662c02d568","added_by":"auto","created_at":"2024-01-24 18:49:02","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":55262,"visible":true,"origin":"","legend":"\u003cp\u003eExample of mapping the distribution of oil slicks in 2016\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/94920b0da908bb9de3c96282.jpg"},{"id":50117383,"identity":"91d86d3c-557a-47ce-be9a-5788b980fd5e","added_by":"auto","created_at":"2024-01-24 18:57:02","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":65461,"visible":true,"origin":"","legend":"\u003cp\u003eTrend of hydrocarbon pollution surfaces.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/aeb558f201e905ad3ecfe7c3.png"},{"id":50116244,"identity":"65490e40-2b89-4352-b561-dca1c05ae263","added_by":"auto","created_at":"2024-01-24 18:49:02","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":124525,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative surface areas of oil slicks by EEZ of each country\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/8854763abc2792f3fb463a31.png"},{"id":50118003,"identity":"6012c2e3-7637-49d0-a8c0-d3b894628495","added_by":"auto","created_at":"2024-01-24 19:05:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":905280,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3860518/v1/c09ae05d-646a-40bd-abe6-e0f47c7e4540.pdf"}],"financialInterests":"","formattedTitle":"Remote sensing monitoring and assessment of oil pollution from 2015 to 2020 in the Gulf of Guinea: The case of the two Congo and the Cabinda region (Angola)","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe Gulf of Guinea region in Africa is a vague area of contiguous maritime and continental space along the West African coast. In terms of oil geopolitics, some authors limit the Gulf of Guinea to the oil-producing countries along the West African coast, stretching from C\u0026ocirc;te d'Ivoire to Angola (Kounou 2009). This zone is characterized by the abundance of natural resources, particularly oil (Ndoutoume 2010). The economies of these countries heavily rely on oil exploitation. For example, in 2000, the oil production of the Gulf of Guinea countries accounted for approximately 3.8\u0026nbsp;billion tons, representing around 5% of global production. The good quality of its crude oil attracts the interest of numerous investors (Amoussou 2018).\u003c/p\u003e \u003cp\u003eThe intensification of oil activities in this region leads to an increase in anthropogenic activities, resulting in oil spills. Analysis of some studies in the field shows that between 2002 and 2012, the Republic of Congo had a high probability of hydrocarbon spills, following Nigeria and Cameroon (Amoussou 2018). Similarly, Ngoma (2015) shows that the Democratic Republic of Congo (DRC) experienced extensive hydrocarbon spills between January and April 2010. In Gabon, particularly at the Cap Lopez terminal (Port-Gentil), an incident in 2022 due to a storage tank leak resulted in an oil spill of approximately 50,000 cubic meters (Malouana Biggie \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These cases highlight the extent of the pollution problem related to oil exploitation in this region. It is important for scientists to guide decision-makers toward suitable solutions that would help strengthen the region's legislation. In addressing this issue, we believe that the use of remote sensing can monitor the expansion of oil spills in marine environments. The use of Earth observation tools, especially SAR imagery, appears to be a potential avenue for identifying and mapping areas with high oil pollution.\u003c/p\u003e \u003cp\u003eTo construct our reasoning, we relied on the approaches detailed by Najoui (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) in his literature review. In his work, he explains that the detection of marine oil slicks includes four basic steps: 1) images preprocessing, 2) dark patches segmentation, 3) features extraction and 4) oil slicks classification (Solberg et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Brekke and Solberg \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Topouzelis and Konstantinos 2008). Generally, in the literature, the image preprocessing (first step) is limited to speckle filtering on standard detected products. Unfortunately, so many heterogeneities remain in the radar images after such classical preprocessing hindering the \"robustness\" of the segmentation and the classification methods especially when working on large areas. The work of this paper focuses on image preprocessing and dark patches segmentation. Due to the incidence angle dependencies (SAR images tend to become darker with increasing range), upwind/downwind or crosswind, and swath width effects, brightness variations may occur in the SAR images and hence compromise the processing of SAR images. A variety of segmentation methods have been proposed and are listed below: adaptive thresholding (Solberg et al.1999), hysteresis thresholding (Kanaa et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), edge detection using Laplace of Gaussians or Difference of Gaussians (Chang et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), wavelets (Liu et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1997\u003c/span\u003e), mathematical morphology (Gasull et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), neural network (Garcia-Pineda et al. 2009; Angiuli et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Del Frate et al. 2013), etc., etc. Even though a variety of segmentation methods have been applied, the most frequently used are based on a local analysis to overcome the brightness variations in the SAR images. In the study area, scientific work on the detection of oil slicks at sea by remote sensing is virtually non-existent.\u003c/p\u003e \u003cp\u003ePrevious studies have investigated the pollution problems (Kounou 2009; Malouana 2022; Mbaki \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2003\u003c/span\u003e) and mapped local oil slicks in the Gulf of Guinea area (Najoui \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Najoui \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022a\u003c/span\u003e; Najoui \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022b\u003c/span\u003e; Najoui 2018a; Najoui 2018b; Okafor 2018 ). However, to our knowledge, this study is the first to carry out a statistical analysis followed by the proposal of a first intelligent GIS approach capable of monitoring the evolution of slick drift followed by their change of state. This work aims to evaluate particularly in the areas of the two Congos and Cabinda (Angola) from 2015 to 2020. Our methodology is based on the pre-processing of radar images to facilitate the semi-automatic detection of oil slicks and then to make the semi-automatic recognition of marine oil slicks more reliable. The main objective is to implement an intelligent GIS for data prediction and analysis. In this article, we will present the materials and methods section, followed by the presentation and discussion of the results in section \u003cspan refid=\"Sec3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Finally, we conclude with some perspectives.\u003c/p\u003e"},{"header":"2. Study area","content":"\u003cp\u003eThe study area of this work is located at the confluence of three countries: The Republic of Congo, the Democratic Republic of Congo (DRC), and the Republic of Angola through its Cabinda region. Regarding the Republic of Congo, it is situated in the center of Africa in the coastal zone. This country spans 162 km of coastline, featuring bays and points such as Pointe Noire, Indienne, and Kounda. The map below (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) illustrates the selected study area for this work. It specifically covers the maritime frontage of both Congo and Cabinda. This region is of particular interest for several reasons, as it is situated at the convergence of three oil-producing countries in a confined space and hosts multiple oil platforms.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"3. Methodology","content":"\u003cp\u003eThe analysis of radar images for hydrocarbon slick detection has been the subject of several studies. Generally, three approaches emerge: a manual approach conducted by trained human operators who analyze the images to detect hydrocarbon slicks. This approach is very rare. The semi-automatic approach where a computer detects all black objects in the radar image using various segmentation techniques, after which an experienced human operator classifies these objects as oil slicks or look-alikes. Finally, the automatic system that uses complex image processing and programming techniques to perform both segmentation and classification. In our case, we have opted for the semi-automatic approach considering the software tools at our disposal.\u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Downloading\u003c/h2\u003e \u003cp\u003eThe collection of Sentinel-1 data was carried out on the \u003cem\u003eASF Data Search\u003c/em\u003e platform. This platform is a satellite data archiving platform developed by the University of Alaska's Institute of Geophysics and managed by National Aeronautics and Space Administration (NASA). Downloading the data is free for all users. We proceeded with downloading the data for the period from 2015 to 2020. This data acquisition step allowed us to collect approximately 644 Sentinel-1A sensor data on a day-to-day basis over the five years. These data were collected according to the following parameters:\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCharacteristics of downloaded data\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCHARACTERISTIC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDATA TYPES\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of data collected\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e644\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMission\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSENTINEL 1_A\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeriod\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2015\u0026ndash;2020\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAcquisition mode\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIW\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProduct type\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGRD\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDirection\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eASCENDANT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePolarisation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVV, VH\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInstrument name\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-band synthetic aperture radar\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSwath\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e250 Kilometers\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResolution\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5X20 meters\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Data pre-processing\u003c/h2\u003e \u003cp\u003eThe downloading step was followed by a data pre-processing phase. This phase aimed to enhance the perception of certain details present in each of the images. The software used for these treatments was SNAP. Several reasons guided us in choosing this software: its free accessibility, its ability to perform quick pre-processing, and its reliability in detecting dark spots in radar imagery. The pre-processing step was preceded by a phase that involved reducing the sizes of the images. Indeed, with a Swat of approximately 250 km and a resolution of 5 meters by 20 meters, the size of the images was very high. This situation made the pre-processing phase very challenging. Faced with this difficulty, we created sub-images taking into account the areas of interest. This step facilitated the processing time of the algorithms.\u003c/p\u003e \u003cp\u003eAfter this step, we moved on to the pre-processing phase. This process unfolded in the following steps: calibration, ellipsoid correction, \u003cem\u003emultilook\u003c/em\u003e, \u003cem\u003eland-sea\u003c/em\u003e mask application, conversion to decibels, and \u003cem\u003espeackle filter\u003c/em\u003e. The complete description of these steps is as follows:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eCalibration\u003c/em\u003e: It consisted of obtaining values at the level of the images which are representative of the target surface sought.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe \u003cem\u003emultilook\u003c/em\u003e process, on the other hand, helped reduce the speckled effects on the images, which hinder the visualization of all the information present in each image. For this step, we utilized the multilook tool. This tool allows for the combination of multiple images incoherently, as if they corresponded to different views of the same scene. The multilook tool has the effect of improving the image interpretation process and enhancing the execution of the speckle filter.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe \u003cem\u003eellipsoid correction\u003c/em\u003e was applied to ensure that the representation of the processed image in SNAP is as close as possible to the reality on the ground. The land-sea mask tool was used to mask terrestrial data and represent the entire marine domain of the target area.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe \u003cem\u003espeackle filter\u003c/em\u003e was used to mitigate the overall noise or shimmer present in the raw images. This noise can disturb the perception of objects or the detection of dark spots in the image. This process aims to highlight dark spots and contextual elements by reducing unwanted interference.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe \u003cem\u003econversion to decibels\u003c/em\u003e was used to better visualize oil slicks and differentiate them from their look-alikes. We performed a reading of the processed image in decibels. This enhances the contrast of the image and characterizes the intensity across the entire image. Each area of the image corresponds to a specific intensity, which is high or positive in brighter areas and low or negative in darker areas. Through this conversion, one can read in decibel values the echoes recorded by the radar sensor. This recorded echo is representative of the backscattering intensity and the behavior of electromagnetic waves when interacting with the target area.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eFollowing the pre-processing in SNAP, we identified areas with a high presence of oil slicks solely of petroleum origin. However, some look-alikes may resemble hydrocarbon slicks with the same spectral signature. Depending on the look-alikes, the backscattering intensity in decibels read on the radar image after conversion varies. Generally, upon observation, only certain look-alikes, such as biological slicks, often exhibited spectral signatures quite similar to those of hydrocarbon slicks. However, we eliminated them since they typically develop in clusters, far from oil platforms. After this work, we proceeded to save the results in GeoTIFF format, which is useful for future processing with QGIS software.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Data processing\u003c/h2\u003e \u003cp\u003eAfter the pre-processing phase, we exported all the data to the Quantum GIS (QGIS) software for the digitization of detected pollution. In total, we identified and digitized 244 hydrocarbon slicks. This digitization led to the creation of vector data, which were subsequently classified by acquisition date. Next, the calculation of the areas of each vector's attributes was performed. The attribute data of each vector were then exported to a spreadsheet for the creation of various graphs. The last phase of these treatments was the production of an annual map of hydrocarbon pollution slicks. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e below schematically describes all the aforementioned steps.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.4 GIS modelling\u003c/h2\u003e \u003cp\u003eTo build the future intelligent GIS, we believe it is necessary to define certain key concepts. To do this, we consider that at a given moment an oil slick can be assimilated to a polygonal geographical object denoted O. We assume that the latter is compact and is characterised by the geographical position (X and Y) of its centroid. Based on this principle, we can say that O is a dynamic object as a function of time. Wagener (2006) describes that an oil slick at sea goes through 4 possible successive states, which we represent in the following figure:\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eState 1 represents Film formation; state 2 represents Dissipation; state 3 represents Foam and finally state 4 Tar balls. It follows that the model discretises a continuous dynamic process. The transition from one state to another is governed by a transition or fragmentation function, which is a stochastic function. This function transforms the slick by taking it through the 4 states we have just described. These are divided into four groups: water, wind, the initial size of the slick and the configuration of the coast. These are:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eConcerning the water: there are 4 variables: the surface temperature, which is a random variable, the tides, the sea current if it exists, and the chemical characteristics of the sea water in the area crossed by the oil slick;\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eConcerning the wind: this is characterised by variables such as the speed, direction and temperature of the air, which generates turbulence and therefore random behaviour;\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe initial size of the slick;\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe coastline: which is characterised by the nature of the coastline (sand, reef, mangrove, etc.) and the distance from the initial slick.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThese variables give the function its stochastic character and will influence the rate of dispersion and slick fragmentation as follows:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eSlick stretch: this is the first effect of the transition function. It will depend in particular on the initial size of the slick, the water temperature, the wind and the tides;\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eDissipation: in this state, the function will break up the slick. The size of the fragments will depend closely on the turbulence at the water surface, induced essentially by air turbulence;\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eFoam: this state of the slick will be influenced to a large extent by the chemistry of the water and its temperature;\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eTar: the function ends its effect by determining the size of the pellets, which will depend on the nature of the coastline.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe formal description of the function requires an in-depth statistical study accompanied by simulations on high-performance infrastructures, which we do not have. However, as an indication, the transition function can be broken down into a series of 3 functions: f1, f2, f3, which characterise the passage of the slick from one state to another (from state 1 to 2 for f1, then from 2 to 3 for f2, and finally from 3 to 4 for f3). The definite advantage of such a study is that it will enable us to predict the behaviour of hydrocarbon pollution in the Gulf of Guinea area, with a view to effectively implementing the appropriate responses likely to reduce its impact on ecosystems.\u003c/p\u003e \u003cp\u003eFollowing the presentation of the methodology, we will now detail the results of our treatments.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Results","content":"\u003cp\u003eThe mapping of oil slicks was carried out step by step over the selected 5 years. The figure below illustrates one aspect of this work.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAfter the mapping work from 2015 to 2020, we have summarized all of these results through the table below.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eRepair per year of the quantity of slicks detected From this identification work, we subsequently calculated the pollution surfaces at the water tables.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDRC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCabinda\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2015\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e64\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2016\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e36\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2017\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e32\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2018\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e45\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2019\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e34\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2020\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e36\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTotal\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e96\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e115\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e36\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eArea in km\u0026sup2; occupied by groundwater per year\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCB\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDRC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCabinda\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2015\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e132,984\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e113,47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e57,987\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e304,441\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2016\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e35,032\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e156,006\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e15,923\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e206,961\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2017\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e58,754\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e93,114\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8,056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e159,924\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2018\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e50,438\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e106,464\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e42,171\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e199,073\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2019\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e28,82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e104,961\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2,937\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e136,718\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2020\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e52,882\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e131,71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e184,592\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTotal\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e358,91\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e705,725\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e127,074\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFrom the above results, we can deduce a regression in pollution due to hydrocarbons from 2015 to 2020. This drop is estimated at around 97.4 km\u0026sup2;, or a reduction of around 8% as mentioned above. Thus between 2016 and 2020 the spreading surfaces of hydrocarbon slicks from oil concessions vary between 150 and 200 km\u0026sup2;.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe also clearly see that the most recurrent and highest pollution areas are observed in the DRC's EEZ. In 2015 we observed around in the EEZ of the Democratic Republic of Congo (ZEE COD) a surface area of spread of hydrocarbon slicks of approximately 110 km\u0026sup2; compared to 130 and 50 for the EEZ of the Republic of Congo (ZEE COG) and Cabinda. However, between 2016 and 2020, the pollution areas observed in the DRC are almost greater than 100 km\u0026sup2;, up to 150 km\u0026sup2; in 2016 and in the case of the exclusive economic zones of the Republic of Congo and Cabinda, the pollution areas are almost less than 50 km\u0026sup2;. Km\u0026sup2;.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"5. Discussion","content":"\u003cp\u003eThe results obtained show large areas of spreading of oil slicks which may be due to frequent and uncontrolled spills. During data collection we used 644 pieces of data. Depending on the number of data downloaded per year, the probability of detecting slicks is greater. Thus, the data collected in 2015 and 2016 are not very representative compared to that collected between 2017 and 2020. The observation that emerges is that this trend is decreasing year after year. To reduce margins of error in detection, we used a stochastic approach which consists of obtaining a large number of samples or data in order to optimize the detection of oil slicks, this approach was carried out in the framework of certain works such as those of (Najoui \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Amoussou 2018). A more detailed analysis shows that the DRC is the most impacted country. The EEZ of the Democratic Republic of Congo has a strong presence of oil slicks. We also observed that the average surface area covered by oil slicks is 117.6 km\u0026sup2; in the DRC and 59.81 km\u0026sup2; for Congo Brazzaville. However, some works, notably those of (Najoui 2022), have shown that the DRC had an average of low pollution areas (around 50 km\u0026sup2;) compared to that of the Republic of Congo Brazzaville (250 km\u0026sup2;) 2002 and 2012. We noted that the DRC has heavily polluted after 2015. From a general point of view, the results show a decreasing trend of around 8% corresponding to a reduction of around 97.4 km\u0026sup2;. This percentage is, however, put into perspective by the DRC which is an exception to the rule. Several potential avenues can explain this situation for this country. Firstly, the weakness of legislation in the face of the power of oil lobbies. Secondly, the insufficiency of controls by state authorities and finally thirdly, the obsolescence of production equipment leading to potential accidents. As we see in our results, the Congo Basin is strongly threatened by groundwater pollution. This threatening situation weighs on the preservation of coastal and marine ecosystems. In this area of preservation, scientific work has shown that when such phenomena occur, the presence time of hydrocarbons as a function of the volume spilled into the marine environment depends on its density (Wegener 2006). This work shows that for a density lower than 0.8 or between 0.8 and 0.85 the presence time could be expressed between 1 hour and 2 hours. These data differ slightly when it comes to light crude oil (density between 0.85 and 0.95) and heavy oil (density above 0.95), the dwell time could be estimated between a week and a year (Wegener 2006). These uncontrolled spills can often appear on coasts in the form of tar or foam, which cause numerous disruptions to coastal ecosystems and a reduction in the resilience of mangrove sites (Wegener 2006). Another consequence linked to this phenomenon is the threat to planktonic species, fish generally in the juvenile stage, marine birds whose soiling of their plumages is the most significant effect of this pollution, marine mammals which during their outings surface are left with damage to nasal tissues due to hydrocarbons (ITPFO \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn order to reduce the risk of error in identifying slicks in the results we have just presented, we relied on the principle that oil spreads very quickly on the surface of the sea. About twelve hours after a spill, the slick can cover an area extending over several kilometres (PNIU, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). To enable us to map oil slicks effectively, we relied on the sensor characteristics of the Sentinel 1 satellite, which is capable of observing an area once every 12 days. By combining the orbital characteristics of Sentinel 1_A and 1_B, this gives us an exact repeat cycle of 6 days at the equator. With this approach, we believe we have considerably reduced the risk of duplicates in the same area. However, this probability is not zero, since throughout the drift of the slicks, there are changes in the content and shape of the oil slicks.\u003c/p\u003e \u003cp\u003eWith regard to the methodology we have adopted, it is important to emphasise that failure to take into account factors such as wind speed, polarisation, incident angle, the effect of dielectric properties and the nature of oil slicks has a major influence on the process of detecting hydrocarbon products at sea. This is why we believe that any GIS modelling must take into account the states that a slick can experience.\u003c/p\u003e"},{"header":"6. Conclusion and perspectives","content":" \u003cp\u003eAs underlined by Najoui (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), radar sensors are commonly used in oil spill monitoring systems because of their well because of their proven detection capability. The emergence of new satellites that are more efficient with larger volumes of data makes automatic oil spill oil slick detection a necessity. However, oil slick detection is a multivariate phenomenon that depends on several factors. Multivariate phenomenon that depends on several parameters. The present work, which is a retrospective study, made it possible to highlight the phenomenon of pollution between 2015 and 2020. This study highlights that the maritime coast of the two Congos and Cabinda is exposed to numerous oil spills with large sprawling areas of up to 300 km\u003csup\u003e2\u003c/sup\u003e. However, this work remains to be improved by integrating not only the automatic approach into the processing of slick detection, but also models for deriving the objects identified using artificial intelligence. Such an approach would reduce the risk of redundancy in the process of identifying oil slicks during the scanning phases.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAmoussou, N. L. 2019. \u003cem\u003eData capitalisation and oil slick detection in the Gulf of Guinea from Envisat radar images 2002-2012\u003c/em\u003e. Sorbone2019.\u003c/li\u003e\n\u003cli\u003eAngiuli, E., F. D. Frate, and L. Salvatori. 2006. \u0026ldquo;\u003cem\u003eNeural networks for oil spill detection using ERS and ENVISAT imagery\u003c/em\u003e.\u0026rdquo; In Proceedings of SeaSAR\u0026rsquo;06, 23-26 January 2006, Frascati, Italy, 1\u0026ndash;6.\u003c/li\u003e\n\u003cli\u003eBrekke, C. and A.h.S. Solberg. 2005. \"Oil Spill Detection by Satellite Remote Sensing.\" \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 95 (2005): 1-13.\u003c/li\u003e\n\u003cli\u003eChang, L., Z. S. Tang., S. H. Chang, and Yang-Lang, C. 2008. \u0026ldquo;\u003cem\u003eA region-based GLRT detection of oil spills in SAR Images\u003c/em\u003e.\u0026rdquo; Pattern Recognition Letters 29 (14): 1915\u0026ndash;23. doi:10.1016/j.patrec.2008.05.022.\u003c/li\u003e\n\u003cli\u003eDel Frate, Fabio, D. Latini, C. Pratola, and F. Palazzo. 2013. \u0026ldquo;\u003cem\u003ePCNN for automatic segmentation and information extraction from X-band SAR imagery\u003c/em\u003e.\u0026rdquo; International Journal of Image and Data Fusion 4 (1): 75\u0026ndash;88. doi:10.1080/19479832.2012.713398.\u003c/li\u003e\n\u003cli\u003eEGE. 2022. \u003cem\u003eOil, the key to power and influence in the Gulf of Guinea\u003c/em\u003e. 2022.\u003c/li\u003e\n\u003cli\u003eETOGA. 2023. \u003cem\u003eGovernance of Marine and Coastal Biodiversity in the Gulf of Guinea\u003c/em\u003e. University of Nice-Sophia Antipolis, 1-192.\u003c/li\u003e\n\u003cli\u003eGarcia-Pineda, O. 2009. \u003cem\u003e\u0026ldquo;Spatial and temporal analysis of oil slicks in the gulf of Mexico based on remote sensing\u0026rdquo;\u003c/em\u003e. Texas A\u0026amp;M University.\u003c/li\u003e\n\u003cli\u003eGasull, A., X. F\u0026aacute;bregas, J. Jim\u0026eacute;nez, F. Marqu\u0026eacute;s, V. Moreno, and M. A. Herrero. 2002. \u0026ldquo;Oil spills detection in SAR images using mathematical morphology.\u0026rdquo; In 11th European Signal Processing Conference (EUSIPCO 2002), 25\u0026ndash;28. Toulouse, France. doi:10.1.1.81.3086.\u003c/li\u003e\n\u003cli\u003eHoundegnonto Odillon J. 2022. Analysis of thermohaline variations in the intraseasonal scales of the freshwater beds of the Goffe de Guinee. \u003cem\u003eHAL Open Science\u003c/em\u003e, 13-185.\u003c/li\u003e\n\u003cli\u003eITPFO. 2013. Technical Information Guide ,The Effect of Oil Pollution on the Environment. Cantebury: Impact PR \u0026amp; Design Limited. 2013\u003c/li\u003e\n\u003cli\u003eKanaa, T. F. N., E. T. G . Mercier., V. P. Onana., J. M. Ngono., P. L. Frison., J. P. Rudant, and R. Garello. 2003. \u0026ldquo;Detection of oil slick signatures in SAR images by fusion of hysteresis thresholding responses.\u0026rdquo; In IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477), 4:2750\u0026ndash;52. IEEE. doi:10.1109/IGARSS.2003.1294573.\u003c/li\u003e\n\u003cli\u003eKounou Michel 2009. Oil and Poverty South of the Sahara: Analysis of the Foundations of the Political Economy of Oil in the Gulf of Guinea, Yaound\u0026eacute;, Cl\u0026eacute;, 2006, p. 21.\u003c/li\u003e\n\u003cli\u003eLiu, A. K., C. Y. Peng, and S. Y. Chang. 1997. \u0026ldquo;Wavelet analysis of satellite images for coastal watch.\u0026rdquo; \u003cem\u003eIEEE Journal on Ocean Engineering\u003c/em\u003e 22 (1): 9\u0026ndash;17\u003c/li\u003e\n\u003cli\u003eMalouana Biggie. 2022. Serious oil spill at the Cap Lopez terminal. Port-Gentil: Gabon Review.\u003c/li\u003e\n\u003cli\u003eMbaki Esther Pabou. 2003. The Congo unarmed in the face of oil pollution. VertigO - la revue \u0026eacute;lectronique en sciences de l'environnement [On line], Regards / Terrain, online 01 May 2003, accessed 01 December 2023..\u0026nbsp;URL : http://journals.openedition.org/vertigo/4856 ;\u003c/li\u003e\n\u003cli\u003eMonde Afrique. 2022. Congo Brazaville under power lines, villages in the dark. 12 October 2022.\u003c/li\u003e\n\u003cli\u003eNajoui, Z., N.Amoussou., S. Riazanoff., G. Aurel, G., and F. Frappart. 2022. Oil slicks in the Gulf of Guinea \u0026ndash; 10 years of Envisat Advanced Synthetic Aperture Radar observations. \u003cem\u003eEarth Syst. Sci. \u003c/em\u003e\u003cem\u003eData\u003c/em\u003e 14\u0026nbsp;: 4569\u0026ndash;4588, https://doi.org/10.5194/essd-14-4569-2022, 2022.\u003c/li\u003e\n\u003cli\u003eNajoui, Z. 2017. Pr\u0026eacute;traitement optimal des images radar et mod\u0026eacute;lisation des d\u0026eacute;rives de nappes d'hydrocarbures pour l'aide \u0026agrave; la photo-interpr\u0026eacute;tation en exploration p\u0026eacute;troli\u0026egrave;re et surveillance environnementale,https://pdfs.semanticscholar.org/92b2/e8e06b49d7f31c0847c694f4b4f3bea41222.pdf?_ga=2.235046721.1549874629.1594648502-969427726.1594648502 (last access: 14 October 2022), 2017.\u0026ensp;\u003c/li\u003e\n\u003cli\u003eNajoui, Z. 2022a. 100 geolocated oil spills in the Gulf of Guinea (1.0), Zenodo, https://doi.org/10.5281/ZENODO.6907743, 2022a.\u0026ensp;\u003c/li\u003e\n\u003cli\u003eNajoui, Z. 2022b. Spatial distribution of oil slicks in the Gulf of Guinea between 2002 and 2012, Zenodo, https://doi.org/10.5281/ZENODO.6470470, 2022b.\u0026ensp;\u003c/li\u003e\n\u003cli\u003eNajoui, Z., Riazanoff, S., Deffontaines, B., and Xavier, J.-P. 2018a. A Statistical Approach to Preprocess and Enhance C-Band SAR Images in Order to Detect Automatically Marine Oil Slicks, IEEE T. Geosci. Remote, 56, 2554\u0026ndash;2564 https://doi.org/10.1109/TGRS.2017.2760516, 2018a\u0026ensp;\u003c/li\u003e\n\u003cli\u003eNajoui, Z., Riazanoff, S., Deffontaines, B., and Xavier, J.-P. 2018b. Estimated location of the seafloor sources of marine natural oil seeps from sea surface outbreaks: A new \u0026ldquo;source path procedure\u0026rdquo; applied to the northern Gulf of Mexico, Mar. Petrol. Geol., 91, 190\u0026ndash;201, https://doi.org/10.1016/j.marpetgeo.2017.12.035\u003c/li\u003e\n\u003cli\u003eNgodi, E.: Gestion des ressources p\u0026eacute;troli\u0026egrave;res et d\u0026eacute;veloppement en Afrique, Pr\u0026eacute;sentation \u0026agrave; la 11\u0026egrave;me Assembl\u0026eacute;e g\u0026eacute;n\u0026eacute;rale du CODESRIA (6\u0026ndash;10 December 2005), Maputo, Mozambique, 2005.\u0026ensp;\u003c/li\u003e\n\u003cli\u003eNdoutoume xxxxxx. 2010. Terrorism and piracy: what security for the seas of the Gulf of Guinea? FES, 159-182.\u003c/li\u003e\n\u003cli\u003eNgoma Khuabi Camille. 2015. Corporate social responsibility and repair of ecological damage due to oil pollution in the coastal city of Moanda in the Democratic Republic of Congo. KAS African Law Study Library, 9-41.\u003c/li\u003e\n\u003cli\u003eOkafor-Yarwood, I.: The effects of oil pollution on the marine environment in the Gulf of Guinea\u0026ndash;the Bonga Oil Field example, Transnatl. Leg. Theory, 9, 254\u0026ndash;271, https://doi.org/10.1080/20414005.2018.1562287, 2018.\u0026ensp;\u003c/li\u003e\n\u003cli\u003eTchetchoua Tchokonte Severin. 2019. Oil Issues and Games in Africa: Study of the Chinese Oil Offensive in the Gulf of Guinea, Master's Thesis in Political Science, University of Yaound\u0026eacute; II, 2007-2009.\u003c/li\u003e\n\u003cli\u003eOna Ona Jenifer Octavie. 2019. Sustainable management of fisheries resources in Central East Atlantic Africa: Cameroon-Congo-Gabon. Nantes: 19 December 2019.\u003c/li\u003e\n\u003cli\u003ePNIU. 2012. Behaviour and Fate of Hydrocarbons. Tunisia\u003c/li\u003e\n\u003cli\u003eTopouzelis, Konstantinos N. 2008. Oil spill detection by SAR Images: Dark formation detection, feature extraction and classification algorithms. Sensors 8 (10): 6642\u0026ndash;59. doi:10.3390/s8106642.\u003c/li\u003e\n\u003cli\u003eSolberg, A. H. S., G. Storvik, R. Solberg, and E. Volden., 1999. Automatic detection of oil spills in ERS SAR images. IEEE Transactions on Geoscience and Remote Sensing 37 (4): 1916\u0026ndash;24. doi:10.1109/36.774704\u003c/li\u003e\n\u003cli\u003eWald Lucien, Jean-Marie Monget, \u0026amp; Michel Albuisson. 2010. Oil pollution in the Mediterranean as seen by the Landsat satellite. Hal Open Science, 61-68\u003c/li\u003e\n\u003cli\u003eWegener Angela. 2006. Marine pollution, case study.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"estuaries-and-coasts","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"esco","sideBox":"Learn more about [Estuaries and Coasts](https://www.springer.com/journal/12237)","snPcode":"12237","submissionUrl":"https://www.editorialmanager.com/esco/","title":"Estuaries and Coasts","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"oil slick, pollution, remote sensing, sea, cartography","lastPublishedDoi":"10.21203/rs.3.rs-3860518/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3860518/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSeveral countries in the Gulf of Guinea have economies based on the exploitation of raw materials, particularly oil. This oil exploitation is not without consequences for marine and coastal ecosystems. Maritime pollution by hydrocarbons is one of the factors affecting marine areas. Having a better understanding of this environment makes it possible to assess the impact of an oil spill on the environment or to study the effects of chronic pollution associated with shipping lanes, ports, oil platforms, pipelines or refineries. It is essential that oil spill detection and monitoring systems respond quickly to contain these hazards. This is why we believe that the use of remote sensing appears to be a potential avenue for the detection of oil slicks at sea. The present work is a retrospective study of oil slick pollution from 2015 to 2020. We propose a methodology based on radar imagery to highlight the presence of ocean oil slicks due to oil activities. Using the Sentinel 1 satellite, we mapped all the oil slicks, which enabled us to calculate the areas where the hydrocarbons and any oil spills spread. The results clearly show that the Congo region is subject to frequent oil spills.\u003c/p\u003e","manuscriptTitle":"Remote sensing monitoring and assessment of oil pollution from 2015 to 2020 in the Gulf of Guinea: The case of the two Congo and the Cabinda region (Angola)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-24 18:48:57","doi":"10.21203/rs.3.rs-3860518/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2024-01-20T05:08:52+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-01-19T21:37:43+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Estuaries and Coasts","date":"2024-01-12T22:00:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-01-12T14:12:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Estuaries and Coasts","date":"2024-01-12T07:40:52+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"estuaries-and-coasts","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"esco","sideBox":"Learn more about [Estuaries and Coasts](https://www.springer.com/journal/12237)","snPcode":"12237","submissionUrl":"https://www.editorialmanager.com/esco/","title":"Estuaries and Coasts","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"a1ff85e1-da24-4cb3-bb39-6f91c34a8e41","owner":[],"postedDate":"January 24th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-01-29T14:00:36+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-24 18:48:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3860518","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3860518","identity":"rs-3860518","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}