Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach
Distinguishing between natural and anthropic oil slicks is a challenging task, especially in the Gulf of Mexico, where these events can be simultaneously observed and recognized as seeps or spills. In this study, a powerful data analysis provided by machine learning (ML) methods was employed to deve...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-11-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/13/22/4568 |
_version_ | 1797508561702486016 |
---|---|
author | Ítalo de Oliveira Matias Patrícia Carneiro Genovez Sarah Barrón Torres Francisco Fábio de Araújo Ponte Anderson José Silva de Oliveira Fernando Pellon de Miranda Gil Márcio Avellino |
author_facet | Ítalo de Oliveira Matias Patrícia Carneiro Genovez Sarah Barrón Torres Francisco Fábio de Araújo Ponte Anderson José Silva de Oliveira Fernando Pellon de Miranda Gil Márcio Avellino |
author_sort | Ítalo de Oliveira Matias |
collection | DOAJ |
description | Distinguishing between natural and anthropic oil slicks is a challenging task, especially in the Gulf of Mexico, where these events can be simultaneously observed and recognized as seeps or spills. In this study, a powerful data analysis provided by machine learning (ML) methods was employed to develop, test, and implement a classification model (CM) to distinguish an oil slick source (OSS) as natural or anthropic. A robust database containing 4916 validated oil samples, detected using synthetic aperture radar (SAR), was employed for this task. Six ML algorithms were evaluated, including artificial neural networks (ANN), random forest (RF), decision trees (DT), naive Bayes (NB), linear discriminant analysis (LDA), and logistic regression (LR). Using RF, the global CM achieved a maximum accuracy value of 73.15. An innovative approach evaluated how external factors, such as seasonality, satellite configurations, and the synergy between them, limit or improve OSS predictions. To accomplish this, specific classification models (SCMs) were derived from the global ones (CMs), tuning the best algorithms and parameters according to different scenarios. Median accuracies revealed winter and spring to be the best seasons and ScanSAR Narrow B (SCNB) as the best beam mode. The maximum median accuracy to distinguish seeps from spills was achieved in winter using SCNB (83.05). Among the tested algorithms, RF was the most robust, with a better performance in 81% of the investigated scenarios. The accuracy increment provided by the well-fitted models may minimize the confusion between seeps and spills. This represents a concrete contribution to reducing economic and geologic risks derived from exploration activities in offshore areas. Additionally, from an operational standpoint, specific models support specialists to select the best SAR products and seasons for new acquisitions, as well as to optimize performances according to the available data. |
first_indexed | 2024-03-10T05:05:41Z |
format | Article |
id | doaj.art-10e960cf8b374cba9e7883325556fc1d |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T05:05:41Z |
publishDate | 2021-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-10e960cf8b374cba9e7883325556fc1d2023-11-23T01:19:30ZengMDPI AGRemote Sensing2072-42922021-11-011322456810.3390/rs13224568Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning ApproachÍtalo de Oliveira Matias0Patrícia Carneiro Genovez1Sarah Barrón Torres2Francisco Fábio de Araújo Ponte3Anderson José Silva de Oliveira4Fernando Pellon de Miranda5Gil Márcio Avellino6Software Engineering Laboratory (LES), Department of Informatics, Pontifical Catholic University (PUC-Rio), 225, Marquês de São Vicente Street, Gávea, Rio de Janeiro 22451-900, BrazilSoftware Engineering Laboratory (LES), Department of Informatics, Pontifical Catholic University (PUC-Rio), 225, Marquês de São Vicente Street, Gávea, Rio de Janeiro 22451-900, BrazilSoftware Engineering Laboratory (LES), Department of Informatics, Pontifical Catholic University (PUC-Rio), 225, Marquês de São Vicente Street, Gávea, Rio de Janeiro 22451-900, BrazilSoftware Engineering Laboratory (LES), Department of Informatics, Pontifical Catholic University (PUC-Rio), 225, Marquês de São Vicente Street, Gávea, Rio de Janeiro 22451-900, BrazilSoftware Engineering Laboratory (LES), Department of Informatics, Pontifical Catholic University (PUC-Rio), 225, Marquês de São Vicente Street, Gávea, Rio de Janeiro 22451-900, BrazilPetrobras Research and Development Center (CENPES), Av. Horácio Macedo 950, Cidade Universitária, Federal University of Rio de Janeiro, Rio de Janeiro 21941-915, BrazilPetrobras Research and Development Center (CENPES), Av. Horácio Macedo 950, Cidade Universitária, Federal University of Rio de Janeiro, Rio de Janeiro 21941-915, BrazilDistinguishing between natural and anthropic oil slicks is a challenging task, especially in the Gulf of Mexico, where these events can be simultaneously observed and recognized as seeps or spills. In this study, a powerful data analysis provided by machine learning (ML) methods was employed to develop, test, and implement a classification model (CM) to distinguish an oil slick source (OSS) as natural or anthropic. A robust database containing 4916 validated oil samples, detected using synthetic aperture radar (SAR), was employed for this task. Six ML algorithms were evaluated, including artificial neural networks (ANN), random forest (RF), decision trees (DT), naive Bayes (NB), linear discriminant analysis (LDA), and logistic regression (LR). Using RF, the global CM achieved a maximum accuracy value of 73.15. An innovative approach evaluated how external factors, such as seasonality, satellite configurations, and the synergy between them, limit or improve OSS predictions. To accomplish this, specific classification models (SCMs) were derived from the global ones (CMs), tuning the best algorithms and parameters according to different scenarios. Median accuracies revealed winter and spring to be the best seasons and ScanSAR Narrow B (SCNB) as the best beam mode. The maximum median accuracy to distinguish seeps from spills was achieved in winter using SCNB (83.05). Among the tested algorithms, RF was the most robust, with a better performance in 81% of the investigated scenarios. The accuracy increment provided by the well-fitted models may minimize the confusion between seeps and spills. This represents a concrete contribution to reducing economic and geologic risks derived from exploration activities in offshore areas. Additionally, from an operational standpoint, specific models support specialists to select the best SAR products and seasons for new acquisitions, as well as to optimize performances according to the available data.https://www.mdpi.com/2072-4292/13/22/4568synthetic aperture radar (SAR)machine learning (ML)exploratory data analysis (EDA)classification model (CM)oil slicks source (OSS)oil seeps |
spellingShingle | Ítalo de Oliveira Matias Patrícia Carneiro Genovez Sarah Barrón Torres Francisco Fábio de Araújo Ponte Anderson José Silva de Oliveira Fernando Pellon de Miranda Gil Márcio Avellino Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach Remote Sensing synthetic aperture radar (SAR) machine learning (ML) exploratory data analysis (EDA) classification model (CM) oil slicks source (OSS) oil seeps |
title | Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach |
title_full | Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach |
title_fullStr | Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach |
title_full_unstemmed | Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach |
title_short | Improved Classification Models to Distinguish Natural from Anthropic Oil Slicks in the Gulf of Mexico: Seasonality and Radarsat-2 Beam Mode Effects under a Machine Learning Approach |
title_sort | improved classification models to distinguish natural from anthropic oil slicks in the gulf of mexico seasonality and radarsat 2 beam mode effects under a machine learning approach |
topic | synthetic aperture radar (SAR) machine learning (ML) exploratory data analysis (EDA) classification model (CM) oil slicks source (OSS) oil seeps |
url | https://www.mdpi.com/2072-4292/13/22/4568 |
work_keys_str_mv | AT italodeoliveiramatias improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach AT patriciacarneirogenovez improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach AT sarahbarrontorres improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach AT franciscofabiodearaujoponte improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach AT andersonjosesilvadeoliveira improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach AT fernandopellondemiranda improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach AT gilmarcioavellino improvedclassificationmodelstodistinguishnaturalfromanthropicoilslicksinthegulfofmexicoseasonalityandradarsat2beammodeeffectsunderamachinelearningapproach |