Oil-slick category discrimination (seeps vs. spills): A linear discriminant analysis using RADARSAT-2 backscatter coefficients (σ°, β°, and γ°) in Campeche Bay (Gulf of Mexico)

Gustavo de Araújo Carvalho, Peter J. Minnett, Eduardo T. Paes, Fernando P. de Miranda, Luiz Landau

Research output: Contribution to journalArticle

Abstract

A novel empirical approach to categorize oil slicks' sea surface expressions in synthetic aperture radar (SAR) measurements into oil seeps or oil spills is investigated, contributing both to academic remote sensing research and to practical applications for the petroleum industry. We use linear discriminant analysis (LDA) to try accuracy improvements from our previously published methods of discriminating seeps from spills that achieved ~70% of overall accuracy. Analyzing 244 RADARSAT-2 scenes containing 4562 slicks observed in Campeche Bay (Gulf of Mexico), our exploratory data analysis evaluates the impact of 61 combinations of SAR backscatter coefficients (σ°, β°, and γ°), SAR calibrated products (received radar beam given in amplitude or decibel, with or without a despeckle filter), and data transformations (none, cube root, log10). The LDA ability to discriminate the oil-slick category is rather independent of backscatter coefficients and calibrated products, but influenced by data transformations. The combination of attributes plays a role in the discrimination; combining oil-slicks' size and SAR information is more effective. We have simplified our analyses using fewer attributes to reach accuracies comparable to those of our earlier studies, and we suggest using other multivariate data analyses-cubist or random forest-to attempt to further improve oil-slick category discrimination.

Original languageEnglish (US)
Article number1652
JournalRemote Sensing
Volume11
Issue number14
DOIs
StatePublished - Jan 1 2019
Externally publishedYes

Fingerprint

RADARSAT
discriminant analysis
backscatter
synthetic aperture radar
oil seep
oil
oil spill
sea surface
radar
filter
remote sensing
oil slick
gulf
product
attribute

Keywords

  • Campeche Bay (Gulf of Mexico)
  • Linear discriminant analysis (LDA)
  • Ocean remote sensing
  • Oil seeps
  • Oil slicks
  • Oil spills
  • Physical oceanography
  • RADARSAT
  • Satellite image classification and segmentation
  • Synthetic aperture radar (SAR)

ASJC Scopus subject areas

  • Earth and Planetary Sciences(all)

Cite this

Oil-slick category discrimination (seeps vs. spills) : A linear discriminant analysis using RADARSAT-2 backscatter coefficients (σ°, β°, and γ°) in Campeche Bay (Gulf of Mexico). / Carvalho, Gustavo de Araújo; Minnett, Peter J.; Paes, Eduardo T.; de Miranda, Fernando P.; Landau, Luiz.

In: Remote Sensing, Vol. 11, No. 14, 1652, 01.01.2019.

Research output: Contribution to journalArticle

@article{15107f760e5a49ae814a2c06cd334ae3,
title = "Oil-slick category discrimination (seeps vs. spills): A linear discriminant analysis using RADARSAT-2 backscatter coefficients (σ°, β°, and γ°) in Campeche Bay (Gulf of Mexico)",
abstract = "A novel empirical approach to categorize oil slicks' sea surface expressions in synthetic aperture radar (SAR) measurements into oil seeps or oil spills is investigated, contributing both to academic remote sensing research and to practical applications for the petroleum industry. We use linear discriminant analysis (LDA) to try accuracy improvements from our previously published methods of discriminating seeps from spills that achieved ~70{\%} of overall accuracy. Analyzing 244 RADARSAT-2 scenes containing 4562 slicks observed in Campeche Bay (Gulf of Mexico), our exploratory data analysis evaluates the impact of 61 combinations of SAR backscatter coefficients (σ°, β°, and γ°), SAR calibrated products (received radar beam given in amplitude or decibel, with or without a despeckle filter), and data transformations (none, cube root, log10). The LDA ability to discriminate the oil-slick category is rather independent of backscatter coefficients and calibrated products, but influenced by data transformations. The combination of attributes plays a role in the discrimination; combining oil-slicks' size and SAR information is more effective. We have simplified our analyses using fewer attributes to reach accuracies comparable to those of our earlier studies, and we suggest using other multivariate data analyses-cubist or random forest-to attempt to further improve oil-slick category discrimination.",
keywords = "Campeche Bay (Gulf of Mexico), Linear discriminant analysis (LDA), Ocean remote sensing, Oil seeps, Oil slicks, Oil spills, Physical oceanography, RADARSAT, Satellite image classification and segmentation, Synthetic aperture radar (SAR)",
author = "Carvalho, {Gustavo de Ara{\'u}jo} and Minnett, {Peter J.} and Paes, {Eduardo T.} and {de Miranda}, {Fernando P.} and Luiz Landau",
year = "2019",
month = "1",
day = "1",
doi = "10.3390/rs11141652",
language = "English (US)",
volume = "11",
journal = "Remote Sensing",
issn = "2072-4292",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "14",

}

TY - JOUR

T1 - Oil-slick category discrimination (seeps vs. spills)

T2 - A linear discriminant analysis using RADARSAT-2 backscatter coefficients (σ°, β°, and γ°) in Campeche Bay (Gulf of Mexico)

AU - Carvalho, Gustavo de Araújo

AU - Minnett, Peter J.

AU - Paes, Eduardo T.

AU - de Miranda, Fernando P.

AU - Landau, Luiz

PY - 2019/1/1

Y1 - 2019/1/1

N2 - A novel empirical approach to categorize oil slicks' sea surface expressions in synthetic aperture radar (SAR) measurements into oil seeps or oil spills is investigated, contributing both to academic remote sensing research and to practical applications for the petroleum industry. We use linear discriminant analysis (LDA) to try accuracy improvements from our previously published methods of discriminating seeps from spills that achieved ~70% of overall accuracy. Analyzing 244 RADARSAT-2 scenes containing 4562 slicks observed in Campeche Bay (Gulf of Mexico), our exploratory data analysis evaluates the impact of 61 combinations of SAR backscatter coefficients (σ°, β°, and γ°), SAR calibrated products (received radar beam given in amplitude or decibel, with or without a despeckle filter), and data transformations (none, cube root, log10). The LDA ability to discriminate the oil-slick category is rather independent of backscatter coefficients and calibrated products, but influenced by data transformations. The combination of attributes plays a role in the discrimination; combining oil-slicks' size and SAR information is more effective. We have simplified our analyses using fewer attributes to reach accuracies comparable to those of our earlier studies, and we suggest using other multivariate data analyses-cubist or random forest-to attempt to further improve oil-slick category discrimination.

AB - A novel empirical approach to categorize oil slicks' sea surface expressions in synthetic aperture radar (SAR) measurements into oil seeps or oil spills is investigated, contributing both to academic remote sensing research and to practical applications for the petroleum industry. We use linear discriminant analysis (LDA) to try accuracy improvements from our previously published methods of discriminating seeps from spills that achieved ~70% of overall accuracy. Analyzing 244 RADARSAT-2 scenes containing 4562 slicks observed in Campeche Bay (Gulf of Mexico), our exploratory data analysis evaluates the impact of 61 combinations of SAR backscatter coefficients (σ°, β°, and γ°), SAR calibrated products (received radar beam given in amplitude or decibel, with or without a despeckle filter), and data transformations (none, cube root, log10). The LDA ability to discriminate the oil-slick category is rather independent of backscatter coefficients and calibrated products, but influenced by data transformations. The combination of attributes plays a role in the discrimination; combining oil-slicks' size and SAR information is more effective. We have simplified our analyses using fewer attributes to reach accuracies comparable to those of our earlier studies, and we suggest using other multivariate data analyses-cubist or random forest-to attempt to further improve oil-slick category discrimination.

KW - Campeche Bay (Gulf of Mexico)

KW - Linear discriminant analysis (LDA)

KW - Ocean remote sensing

KW - Oil seeps

KW - Oil slicks

KW - Oil spills

KW - Physical oceanography

KW - RADARSAT

KW - Satellite image classification and segmentation

KW - Synthetic aperture radar (SAR)

UR - http://www.scopus.com/inward/record.url?scp=85071524698&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071524698&partnerID=8YFLogxK

U2 - 10.3390/rs11141652

DO - 10.3390/rs11141652

M3 - Article

AN - SCOPUS:85071524698

VL - 11

JO - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 14

M1 - 1652

ER -