Abstract:
This study was conducted on the basis of the open archive for Synthetic Aperture Radar (SAR) images collected over the lease block Green Canyon 600 (GC600) and its surrounding region from January 2003 to November 2011. The objective of this work was to estimate the surface residence-time of the oil slicks, and to determine the importance of wind and surface currents on the trajectory and fate of the released oil. SAR imagery was collected by eight SAR satellites: Cosmo-Skymed-1, -2, and -3; RADARSAT-1 and -2; ENVISAT ASAR, ALOS PALSAR, and ERS-2. Image sources included European Space Agency (ESA), National Aeronautics and Space Administration (NASA) with support from the Alaska Satellite Facility, and the Center for Spatial Technologies and Remote Sensing (CSTARS). We matched the location of the GC600 against the coverage of the SAR images and determined whether the oil slicks were visible in archived SAR images. Some of the image scenes were un-usable due to poor image quality, or did not contain oil. After preliminary screening, a Texture Classifying Neural Network Algorithm (TCNNA) was used to delineate georectified polygons for oil slicks from 41 SAR images. Trajectories of these oil slicks were investigated by employing a two-dimensional Lagrangian particle-tracking surface oil drift model. The evolution of each oil slick on the surface of the ocean was simulated by computing the two-dimensional trajectories of 100 particles every 15 minutes. The initial positions of the particles were randomly distributed near the Oil Slick Origin (OSO) point within the seeding radius. The seeding radius was defined as 1/3 width of the oil slick at the OSO estimated from SAR image. The trajectories of the particles were computed within a time-varying velocity field that is some combination of ocean surface currents derived from the HYbrid Coordinate Ocean Model (HYCOM) and 10-m winds derived from Cross-Calibrated Multi-Platform Ocean Surface Wind Vectors (CCMP). A set of numerical simulations was performed by increasing hindcast interval in reverse time order from the image collection time in order to obtain the closest resemblance between the simulated oil pathways and the length and shape of the oil slicks observed in SAR images. A surface residence-time estimate for an observed oil slick was the hindcast interval that best reproduced the length and shape of that oil slick in the simulated representation. The wind scaling coefficient of 0.035 and the wind deflection angle of 20° to the right of the wind direction yielded the most accurate trajectory of the oil particles compared to SAR data. To study the effect of wind and surface currents on the trajectory and fate of the natural oil slicks, the following simulations with the surface oil drift model were performed: 1) wind and surface currents combined; 2) surface currents alone (wind scaling coefficient was set to 0), and 3) wind alone (ocean currents were set to 0). In addition, a wind-powered autonomous surface vehicle (SailDrone) deployment provided in-situ observations at the lease block Green Canyon 574 (GC574) and its surrounding region to verify the importance of wind and surface currents in deriving the trajectory of oil slicks on the sea surface.
Data Parameters and Units:
This dataset provides a zip file containing: 1- SAR imagery info & sources: an excel file (.xlsx) including date and time of the 41 SAR images used in this study, SAR footprint coordinates (upper left, lower left, lower right, and upper right in decimal degrees), SAR data filenames, instrument name, name of the SAR satellite, TCNNA extracted oil slicks’ filenames, and the source of SAR images. Date (YYYY-MM-DD), Time (HH:MM:SS), Latitude (decimal degrees), Longitude (decimal degrees), SAR data filenames, Instrument, Satellite, TCNNA extracted oil slicks filenames, SAR data source 2- ArcGIS shapefiles for the TCNNA extracted oil slicks: a folder of 41 ArcGIS shapefiles (.shp) which are named according to the TCNNA extracted oil slicks’ filenames, and outlining TCNNA extracted oil slicks from the 41 SAR images in the study area. 3- ArcGIS shapefiles for the OSO points (Oil Slick Origin points): a folder of 41 ArcGIS shapefiles (.shp) which are named according to the SAR data filenames, and outlining OSO points in the 41 SAR images. 4- HYCOM & CCMP Data Sources: an ASCII file (.txt) including the metadata of HYCOM & CCMP data sources. 5- Surface Oil Drift Model setup info & output: a folder of 41 Excel files (.xlsx) which are named according to the SAR data filenames. Each of the excel files include 6 sheets: i) Setup Information: specifies the input parameters of the surface oil drift model for the corresponding oil slick when wind and surface currents were driving forces of the surface oil drift model, ii) Output (Wind + Surface Current): contains the geographic coordinates (longitude and latitude in decimal degrees) and ages (sec) of the simulated trajectory when wind and surface currents were driving forces of the surface oil drift model, iii) Setup Info (Surface Current): specifies the input parameters of the surface oil drift model for the corresponding oil slick when surface currents alone were driving forces of the surface oil drift model, iv) Output (Surface Current): contains the geographic coordinates (longitude and latitude in decimal degrees) and ages (sec) of the simulated trajectory when surface currents alone were driving forces of the surface oil drift model, v) Setup Info (Wind): specifies the input parameters of the surface oil drift model for the corresponding oil slick when wind alone was driving force of the surface oil drift model, vi) Output (Wind): contains the geographic coordinates (longitude and latitude in decimal degrees) and ages (sec) of the simulated trajectory when wind alone was driving force of the surface oil drift model. Number of particles seeded per time step (Np), Half-life (Th, Hrs), Maximum life (Tmax, Hrs), Seeding radius-the radius within which the particles are seeded around the OSO (Rs, m), Model time step (dt, Hrs), Wind scaling coefficient (Cw), Wind deflation angle (q, degrees), Start time (HH:MM), Date (MM/DD/YYYY) 6- Spatial measurements & estimated volume of the oil slicks: an Excel file (.xlsx) including date and time of the 41 SAR images, SAR data filenames, geographic coordinates of the OSO points (OSO_longitude and OSO_latitude in decimal degrees), and estimated oil slick lengths (km), oil slick areas (km2), oil slick volumes (m3), and oil slick ages (h), Date (YYYY-MM-DD), Time (HH:MM:SS), SAR data filenames 7- SailDrone in situ observations: an Excel file (.xlsx) which has two sheets: i) 2015-10-10 to 2015-10-12 per 15 min: includes date (YYYY-MM-DD) and time (HH:MM:SS), geographic coordinates of the SailDrone location (latitude and longitude in decimal degrees), the level of fluorescence signal, wind direction (degrees), and wind speed (knots) every 15 minutes from 10 to 12 October 2015, and ii) 2015-10-12 per second: includes time, geographic coordinates of the SailDrone location (latitude and longitude in decimal degrees), and the level of fluorescence signal every seconds on 12 October 2015.
View Metadata