Search

Sea-surface petroleum pollution is observed as "oil slicks" (i.e., "oil spills" or "oil seeps") and can be confused with "look-alike slicks" (i.e., environmental phenomena, such as low-wind speed, upwelling conditions, chlorophyll, etc.) in synthetic aperture radar (SAR) measurements, the most proficient satellite sensor to detect mineral oil on the sea surface. Even though machine learning (ML) has become widely used to classify remotely-sensed petroleum signatures, few papers have been published comparing various ML methods to distinguish spills from look-alikes. Our research fills this gap by comparing and evaluating six traditional techniques: simple (naive Bayes (NB), K-nearest neighbor (KNN), decision trees (DT)) and advanced (random forest (RF), support vector machine (SVM), artificial neural network (ANN)) applied to different combinations of satellite-retrieved attributes. 36 ML algorithms were used to discriminate "ocean-slick signatures" (spills versus look-alikes) with ten-times repeated random subsampling cross validation (70-30 train-test partition). Our results found that the best algorithm (ANN: 90%) was >20% more effective than the least accurate one (DT: ~68%). Our empirical ML observations contribute to both scientific ocean remote-sensing research and to oil and gas industry activities, in that: (i) most techniques were superior when morphological information and Meteorological and Oceanographic (MetOc) parameters were included together, and less accurate when these variables were used separately; (ii) the algorithms with the better performance used more variables (without feature selection), while lower accuracy algorithms were those that used fewer variables (with feature selection); (iii) we created algorithms more effective than those of benchmark-past studies that used linear discriminant analysis (LDA: ~85%) on the same dataset; and (iv) accurate algorithms can assist in finding new offshore fossil fuel discoveries (i.e., misclassification reduction). [ABSTRACT FROM AUTHOR]

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°, 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.

Mariners navigating within Canadian waters rely on Canadian Hydrographic Service (CHS) navigational charts to safely reach their destinations. To fulfil this need, CHS charts must accurately reflect the current state of Canadian coastal regions. While many coastal regions are stable, others are dynamic and require frequent updates. In order to ensure that important and potentially dangerous changes are reflected in CHS products, the organization, in partnership with the Canadian Space Agency, is exploring coastal change detection through satellite remote sensing (SRS). In this work, CHS examined a hybrid shoreline extraction approach which uses both Synthetic Aperture Radar (SAR) and optical data. The approach was applied for a section of the Mackenzie River, one of Canada’s most dynamic river systems. The approach used RADARSAT-2 imagery as its primary information source, due to its high positioning accuracy (5 m horizontal accuracy) and ability to allow for low and high water line charting. Landsat represented the primary optical data source due to its long historical record of Earth observation data. Additional sensors, such as Sentinel-2 and WorldView, were also used where a higher resolution was required. The shoreline extraction process is based on an image segmentation approach that uses both the radar and optical data. Critical information was collected using the automated approach to support chart updates, resulting in reductions to the financial, human and time factors present within the ship-based hydrographic survey techniques traditionally used for chart improvements. The results demonstrate the potential benefit of wide area SRS change detection within dynamic waterways for navigational chart improvements. The work also demonstrates that the approach developed for RADARSAT-2 could be implemented with data from the forthcoming RADARSAT Constellation Mission (RCM), which is critical to ensure project continuity.

This study aims to detect coastline changes using temporal synthetic aperture radar (SAR) images for the state of Kelantan, Malaysia. Two active images, namely, RADARSAT-1 captured in 2003 and RADARSAT-2 captured in 2014, were used to monitor such changes. We applied noise removal and edge detection filtering on RADARSAT images for preprocessing to remove salt and pepper distortion. Different segmentation analyses were also applied to the filtered images. Firstly, multiresolution segmentation, maximum spectral difference and chessboard segmentation were performed to separate land pixels from ocean ones. Next, the Taguchi method was used to optimise segmentation parameters. Subsequently, a support vector machine algorithm was applied on the optimised segments to classify shorelines with an accuracy of 98% for both temporal images. Results were validated using a thematic map from the Department of Survey and Mapping of Malaysia. The change detection showed an average difference in the shoreline of 12.5 m between 2003 and 2014. The methods developed in this study demonstrate the ability of active SAR sensors to map and detect shoreline changes, especially during low or high tides in tropical regions where passive sensor imagery is often masked by clouds.

RADARSAT and ERS-2 data collected at multiple incidence angles are used to characterize the seasonal variations in the backscatter of snow-covered landscapes in the northern Hudson Bay Lowlands during the winters of 1997/98 and 1998/99. The study evaluates the usefulness of C-band SAR systems for retrieving the snow water equivalent under dry snow conditions in the forest–tundra ecotone. The backscatter values are compared against ground measurements at six sampling sites, which are taken to be representative of the land-cover types found in the region. The contribution of dry snow to the radar return is evident when frost penetrates the first 20 cm of soil. Only then does the backscatter respond positively to changes in snow water equivalent, at least in the open and forested areas near the coast, where 1-dB increases in backscatter for each approximate 5–10 mm of accumulated water equivalent are observed at 20–31° incidence angles. Further inland, the backscatter shows either no change or a negative change with snow accumulation, which suggests that the radar signal there is dominated by ground surface scattering (e.g., fen) when not attenuated by vegetation (e.g., forested and transition). With high-frequency ground-penetrating radar, we demonstrate the presence of a 10–20-cm layer of black ice underneath the snow cover, which causes the reduced radar returns (−15 dB and less) observed in the inland fen. A correlation between the backscattering and the snow water equivalent cannot be determined due to insufficient observations at similar incidence angles. To establish a relationship between the snow water equivalent and the backscatter, only images acquired with similar incidence angles should be used, and they must be corrected for both vegetation and ground effects.

We present a flexible, integrated statistical-based modeling approach to improve the robustness of soil moisture data predictions. We apply this approach in exploring the consequence of different choices of leading predictors and covariates. Competing models, predictors, covariates and changing spatial correlation are often ignored in empirical analyses and validation studies. An optimal choice of model and predictors may, however, provide a more consistent and reliable explanation of the high environmental variability and stochasticity of soil moisture observational data. We integrate active polarimetric satellite remote-sensing data (RADARSAT-2, C-band) with ground-based in-situ data across an agricultural monitoring site in Canada. We apply a grouped step-wise algorithm to iteratively select best-performing predictors of soil moisture. Integrated modeling approaches may better account for observed uncertainty and be tuned to different applications that vary in scale and scope, while also providing greater insights into spatial scaling (upscaling and downscaling) of soil moisture variability from the field- to regional scale. We discuss several methodological extensions and data requirements to enable further statistical modeling and validation for improved agricultural decision-support. [ABSTRACT FROM AUTHOR]

RADARSAT and ERS-2 data collected at multiple incidence angles are used to characterize the seasonal variations in the backscatter of snow-covered landscapes in the northern Hudson Bay Lowlands during the winters of 1997/98 and 1998/99. The study evaluates the usefulness of C-band SAR systems for retrieving the snow water equivalent under dry snow conditions in the forest-tundra ecotone. The backscatter values are compared against ground measurements at six sampling sites, which are taken to be representative of the land-cover types found in the region. The contribution of dry snow to the radar return is evident when frost penetrates the first 20 cm of soil. Only then does the backscatter respond positively to changes in snow water equivalent, at least in the open and forested areas near the coast, where 1-dB increases in backscatter for each approximate 5-10 mm of accumulated water equivalent are observed at 20-31° incidence angles. Further inland, the backscatter shows either no change or a negative change with snow accumulation, which suggests that the radar signal there is dominated by ground surface scattering (e.g., fen) when not attenuated by vegetation (e.g., forested and transition). With high-frequency ground-penetrating radar, we demonstrate the presence of a 10-20-cm layer of black ice underneath the snow cover, which causes the reduced radar returns (-15 dB and less) observed in the inland fen. A correlation between the backscattering and the snow water equivalent cannot be determined due to insufficient observations at similar incidence angles. To establish a relationship between the snow water equivalent and the backscatter, only images acquired with similar incidence angles should be used, and they must be corrected for both vegetation and ground effects. [ABSTRACT FROM AUTHOR]

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°, 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. [ABSTRACT FROM AUTHOR]

Mariners navigating within Canadian waters rely on Canadian Hydrographic Service (CHS) navigational charts to safely reach their destinations. To fulfil this need, CHS charts must accurately reflect the current state of Canadian coastal regions. While many coastal regions are stable, others are dynamic and require frequent updates. In order to ensure that important and potentially dangerous changes are reflected in CHS products, the organization, in partnership with the Canadian Space Agency, is exploring coastal change detection through satellite remote sensing (SRS). In this work, CHS examined a hybrid shoreline extraction approach which uses both Synthetic Aperture Radar (SAR) and optical data. The approach was applied for a section of the Mackenzie River, one of Canada's most dynamic river systems. The approach used RADARSAT-2 imagery as its primary information source, due to its high positioning accuracy (5 m horizontal accuracy) and ability to allow for low and high water line charting. Landsat represented the primary optical data source due to its long historical record of Earth observation data. Additional sensors, such as Sentinel-2 and WorldView, were also used where a higher resolution was required. The shoreline extraction process is based on an image segmentation approach that uses both the radar and optical data. Critical information was collected using the automated approach to support chart updates, resulting in reductions to the financial, human and time factors present within the ship-based hydrographic survey techniques traditionally used for chart improvements. The results demonstrate the potential benefit of wide area SRS change detection within dynamic waterways for navigational chart improvements. The work also demonstrates that the approach developed for RADARSAT-2 could be implemented with data from the forthcoming RADARSAT Constellation Mission (RCM), which is critical to ensure project continuity. [ABSTRACT FROM AUTHOR]

This study aims to detect coastline changes using temporal synthetic aperture radar (SAR) images for the state of Kelantan, Malaysia. Two active images, namely, RADARSAT-1 captured in 2003 and RADARSAT-2 captured in 2014, were used to monitor such changes. We applied noise removal and edge detection filtering on RADARSAT images for preprocessing to remove salt and pepper distortion. Different segmentation analyses were also applied to the filtered images. Firstly, multiresolution segmentation, maximum spectral difference and chessboard segmentation were performed to separate land pixels from ocean ones. Next, the Taguchi method was used to optimise segmentation parameters. Subsequently, a support vector machine algorithm was applied on the optimised segments to classify shorelines with an accuracy of 98% for both temporal images. Results were validated using a thematic map from the Department of Survey and Mapping of Malaysia. The change detection showed an average difference in the shoreline of 12.5 m between 2003 and 2014. The methods developed in this study demonstrate the ability of active SAR sensors to map and detect shoreline changes, especially during low or high tides in tropical regions where passive sensor imagery is often masked by clouds. [ABSTRACT FROM AUTHOR]

Mapping the historical occurrence of flood water in time and space provides information that can be used to help mitigate damage from future flood events. In Canada, flood mapping has been performed mainly from RADARSAT imagery in near real-time to enhance situational awareness during an emergency, and more recently from Landsat to examine historical surface water dynamics from the mid-1980s to present. Here, we seek to integrate the two data sources for both operational and historical flood mapping. A main challenge of a multi-sensor approach is ensuring consistency between surface water mapped from sensors that fundamentally interact with the target differently, particularly in areas of flooded vegetation. In addition, automation of workflows that previously relied on manual interpretation is increasingly needed due to large data volumes contained within satellite image archives. Despite differences between data received from both sensors, common approaches to surface water and flooded vegetation mapping including multi-channel classification and region growing can be applied with sensor-specific adaptations for each. Historical open water maps from 202 Landsat scenes spanning the years 1985–2016 generated previously were enhanced to improve flooded vegetation mapping along the Saint John River in New Brunswick, Canada. Open water and flooded vegetation maps were created over the same region from 181 RADARSAT 1 and 2 scenes acquired between 2003–2016. Comparisons of maps from different sensors and hydrometric data were performed to examine consistency and robustness of products derived from different sensors. Simulations reveal that the methodology used to map open water from dual-pol RADARSAT 2 is insensitive to up to about 20% training error. Landsat depicts open water inundation well, while flooded vegetation can be reliably mapped in leaf-off conditions. RADARSAT mapped approximately 8% less open water area than Landsat and 0.5% more flooded vegetation, while the combined area of open water and flooded vegetation agreed to within 0.2% between sensors. Derived historical products depicting inundation frequency and trends were also generated from each sensor’s time-series of surface water maps and compared. [ABSTRACT FROM AUTHOR]