3  Data Sources Overview

4 Data Sources Overview

4.1 Accidental spills

4.1.1 Integrated Satellite Tracking of Pollution (ISTOP)

The Integrated Satellite Tracking of Pollution (ISTOP) program (Environment and Climate Change Canada, 2024) utilizes satellite imagery to monitor and detect oil spills in coastal waters. Since 2006, daily satellite analyses have been conducted to identify and report both illegal and accidental marine oil pollution. These analyses detect sea-surface anomalies, which are vectorized into polygons representing affected areas. The data obtained span 2006 to 2023 and include 637 recorded incidents. The data was transferred manually to the Canadian Wildlife Service and are currently stored on a secure Google Cloud Storage bucket accessible through an authentication key and maintained by inSileco.

  • Source: Environment Canada Integrated Satellite Tracking of Pollution (ISTOP) Program
  • Accessibility: Restricted
  • Data Type: Vectorized Polygons
  • Coverage: North American coastal waters, 2006–2023
  • Processing Script: prc_istop.R
  • Output File: istop.gpkg

4.1.2 National Aerial Surveillance Program (NASP)

The National Aerial Surveillance Program (NASP) (Transport Canada, 2024) is a country-wide aerial surveillance initiative designed to detect pollution and protect Canadian waters. The program, covering pollution detections in the Atlantic region from April 2012 to March 2023, aims to safeguard marine environments, endangered marine life, and promote safe maritime transport along Canada’s coastlines. The dataset includes 3578 incidents. The data was transferred manually to the Canadian Wildlife Service and are currently stored on a secure Google Cloud Storage bucket accessible through an authentication key and maintained by inSileco.

  • Source: Transport Canada National Aerial Surveillance Program (NASP)
  • Accessibility: Restricted
  • Data Type: CSV
  • Coverage: Atlantic Region, Canada, 2012–2023
  • Processing Script: prc_nasp.R
  • Output File: nasp.gpkg

4.1.3 National Environmental Emergency Centre (NEEC)

The National Environmental Emergency Centre (NEEC) (National Environmental Emergency Centre, 2024), a program under Environment and Climate Change Canada, provides an inventory of environmental emergency incidents across Canada. The NEEC dataset provided to the Canadian Wildlife Service covers incidents impacting waterbodies in the Atlantic region, including Québec, from 2016 to 2023, comprising 9570 recorded incidents. The data was transferred manually to the Canadian Wildlife Service and are currently stored on a secure Google Cloud Storage bucket accessible through an authentication key and maintained by inSileco.

  • Source: National Environmental Emergency Centre (NEEC)
  • Accessibility: Restricted
  • Data Type: CSV
  • Coverage: Atlantic Region, Canada (including Québec), 2016–2023
  • Processing Script: prc_neec.R
  • Output Files: neec.csv and substances.csv

4.2 Offshore petroleum activities

4.2.1 Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB)

The Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) (Canada-Newfoundland and Labrador Offshore Petroleum Board, 2024) oversees and provides data on offshore petroleum activities in Newfoundland. This dataset includes multiple files covering exploration and production licenses, various well data, call for bids, land tenure sectors of the Scheduled Land Tenure System, and well info summaries (Table 4.1), providing a comprehensive view of the offshore petroleum landscape in Newfoundland. The dataset includes both spatial and tabular data. Key variables encompass license IDs, well IDs, geographic coordinates, vessel types, exploration phases, and parcel identifiers. Data files are provided in various formats, including shapefiles and Excel files, and have been processed to create spatial representations of offshore petroleum activity sites. For the purpose of this project, we only retained, if available, the classification, status, start and end dates when processing the data. The different categories of offshore petroleum activities are detailed in Table 4.2. Each entry in the integrated dataset was given a unique identifiers. Individual entries with dual classification (e.g. exploration and delineation) were duplicated, each retaining the same id to avoid overestimating the number of entries in the dataset. As such, the number of unique entries would correspond to the number of unique identifiers rather than the number of rows in the dataset. For data with well status, we simplified status categories for the processed data, which are detailed in Table 4.3. How date ranges were established for each entry is detailed in Table 4.1.

  • Source: Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB)
  • Accessibility: Open, Creative Commons Attribution License
  • Data Type: Shapefiles, XLSX
  • Coverage: Offshore regions of Newfoundland, Canada
  • Processing Script: prc_offshore_petroleum_nfl.R
  • Output File: offshore_petroleum_nfl.gpkg
Table 4.1: Description of licenses, call for bids and sectors for the petroleum industry in Newfoundland and Labrador.
Type Description Period
Active Exploration Licenses Offshore areas in Newfoundland and Labrador that are currently licensed for petroleum exploration. An Exploration License (EL) grants rights to conduct exploratory drilling and other activities aimed at discovering petroleum resources within a designated offshore area. These licenses are typically issued for a fixed period, encouraging exploration activities that assess the resource potential in Newfoundland and Labrador’s offshore regions. The beginning is the effective date (effective) and the end is the second exploration period specified (period2exp).
Active Significant Discovery Licenses Offshore areas within Newfoundland and Labrador where significant petroleum discoveries have been identified and are actively licensed. An Active Significant Discovery License (SDL) allows the holder to retain rights over a discovery area with proven, potentially commercial quantities of petroleum, preserving the right to future exploration and development. The beginning is the date of first exploration activity (spud_date) and the termination of the last renewal phase (term_date).
Active Production Licenses Offshore areas in Newfoundland and Labrador where petroleum resources are actively being produced. A Production License (PL) authorizes the holder to extract petroleum within a specified offshore area, following proven resource discoveries. This license allows companies to conduct production activities while adhering to regulatory requirements for resource management, environmental protection, and safety. The beginning is the date of first exploration activity (spud_date) and the termination of the last renewal phase (term_date).
Production Installations Locations of offshore production installations used in petroleum extraction activities in Newfoundland and Labrador’s offshore areas. A Production Installation refers to infrastructure such as platforms, rigs, or other facilities designed to extract, process, and transport petroleum resources from beneath the seafloor. These installations play a critical role in supporting production operations within licensed areas. The beginning is the date of first exploration activity (spud_date) and the termination of the last renewal phase (term_date).
Delineation Wells Locations of delineation wells drilled in offshore areas of Newfoundland and Labrador. A Delineation Well is drilled to determine the extent and characteristics of a known petroleum discovery, helping to assess the size, quality, and commercial viability of the resource. These wells are essential in refining the understanding of hydrocarbon deposits discovered by exploration wells. The beginning is the date of first exploration activity (spud_date) and the termination of the last renewal phase (term_date).
Development Wells Locations of development wells in offshore areas of Newfoundland and Labrador. Development Wells are drilled within proven petroleum fields to facilitate resource extraction. These wells are strategically placed to optimize production from known hydrocarbon deposits, making them critical for efficient and sustainable petroleum extraction. The beginning is the date when the well first began producing oil (first_oil); there is no end date specified.
Dual Classified Wells Offshore wells in Newfoundland and Labrador that have been classified under more than one category during their lifecycle. Dual Classified Wells may serve multiple purposes, such as transitioning from an exploration role to development or delineation, depending on the results of drilling and assessments. This dual classification reflects the adaptive approach taken in offshore petroleum exploration and production. The beginning is the date when the date on which information about the well or license was publicly posted (date_posted) and the end is the specified call for bids closing year (cfb_clse_dte) and assumed to be on January 1st to allow for date formatting.
Exploration Wells Locations of offshore wells in Newfoundland and Labrador that are designated for exploration. Exploration Wells are drilled to investigate potential petroleum deposits in unexplored or underexplored areas. These wells are the first step in assessing hydrocarbon presence and play a crucial role in determining the viability of further development. The beginning is the date when the date on which information about the well or license was publicly posted (date_posted) and the end is the specified call for bids closing year (cfb_clse_dte) and assumed to be on January 1st to allow for date formatting.
Call for Bids - Eastern Newfoundland NL23-CFB01 The Call for Bids NL23-CFB01 pertains to the Eastern Newfoundland Region and was issued by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) in 2023. This call offered 28 parcels, totaling approximately 7,222,551 hectares, for exploration licensing. Interested parties were invited to submit sealed bids by November 1, 2023. However, the C-NLOPB announced that no bids were received for this call. The beginning is the date when the date on which information about the well or license was publicly posted (date_posted) and the end is the specified call for bids closing year (cfb_clse_dte) and assumed to be on January 1st to allow for date formatting.
Call for Bids - South Eastern Newfoundland NL23-CFB02 The Call for Bids NL23-CFB02 was issued by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) in 2023, offering 19 parcels totaling approximately 4,982,275 hectares in the South Eastern Newfoundland Region for exploration licensing. Interested parties were invited to submit sealed bids by November 1, 2023. However, the C-NLOPB announced that no bids were received for this call. The beginning is the date when the date on which information about the well or license was publicly posted (date_posted) and the end is the specified call for bids closing year (cfb_clse_dte) and assumed to be on January 1st to allow for date formatting.
Call for Bids - Eastern Newfoundland NL24-CFB01 The Call for Bids NL24-CFB01 was issued by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) on April 29, 2024, inviting bids for exploration licenses in the Eastern Newfoundland Region. This call comprised 41 parcels, covering a total of 10,287,196 hectares. Interested parties were required to submit sealed bids by November 6, 2024. On November 6, 2024, the C-NLOPB announced that no bids were received for Call for Bids NL24-CFB01. The beginning is the date is the opening date of the call for bids (cfb_open_dte) and the end is the closing date of the call for bids (cfb_clse_dte).
Sectors - Labrador South NL02-LS The Labrador South Region (Sector NL02-LS) is an offshore area managed by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) as part of its Scheduled Land Tenure System. The beginning is the date is the opening date of the call for bids (cfb_open_dte) and the end is the closing date of the call for bids (cfb_clse_dte).
Sectors - Eastern Newfoundland NL06-EN The Eastern Newfoundland Region (Sector NL06-EN) is an offshore area designated by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) under its Scheduled Land Tenure System. The beginning is the date is the opening date of the call for bids (cfb_open_dte) and the end is the closing date of the call for bids (cfb_clse_dte).
Sectors - North Eastern Newfoundland NL01-NEN The North Eastern Newfoundland Region (Sector NL01-NEN) is an offshore area designated by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) under its Scheduled Land Tenure System. The beginning is the date on which the sector’s licensing or regulatory conditions officially took effect (effective); there is no end date specified.
Sectors - Southern Newfoundland NL01-SN The Southern Newfoundland Region (Sector NL01-SN) is an offshore area designated by the Canada-Newfoundland and Labrador Offshore Petroleum Board (C-NLOPB) under its Scheduled Land Tenure System. The beginning is the date on which the sector’s licensing or regulatory conditions officially took effect (effective); there is no end date specified.
Well Info Summary Contains detailed information on wells drilled within offshore petroleum exploration and production areas. This dataset is designed to offer an inventory of wells, capturing essential data for regulatory oversight, resource management, and operational history. The dataset includes information on well names, operators, drilling dates, statuses, and geographic coordinates, among other fields. The beginning corresponds to the date when drilling for the well began (spud_date) and the end is the date on which drilling or operational activities for the well concluded (well_termination_date).
Table 4.2: Description of licenses, call for bids and sectors for the petroleum industry in Newfoundland
Classification Description
Sectors Defined geographic areas designated for potential exploration and development activities, often used in the regulatory framework to organize land tenure.
Call for Bids Competitive bidding rounds where companies submit proposals to obtain exploration or development rights for specified areas within sectors.
Exploration Initial phase focused on discovering hydrocarbons, involving activities like seismic surveys and drilling exploration wells.
Significant Discovery Identification of a commercially viable hydrocarbon discovery during exploration, associated with a Significant Discovery License (SDL) to preserve rights to the area.
Delineation Appraisal phase to define the size, extent, and quality of a discovered reservoir, refining understanding of the resource’s potential.
Development Preparation phase involving construction of infrastructure, drilling of production wells, and installation of facilities needed to bring hydrocarbons to the surface.
Production Active phase of extracting hydrocarbons, processing them, and transporting them for sale, generating revenue from the resource.
Table 4.3: Simplified Status Classifications with Descriptions
Status Description
Abandoned Wells where activities have been permanently stopped.
Suspended Wells with temporarily halted activities that may resume.
Active Production Wells currently involved in production or injection activities.
Drilling Wells still in the drilling phase.
Disposal Wells used for waste or byproduct disposal.
Closed Wells officially closed, possibly for regulatory reasons.
Off Station Wells or rigs temporarily off their regular location.
NA No information available.

4.2.2 Canada-Nova Scotia Offshore Petroleum Board (CNSOPB)

The Canada-Nova Scotia Offshore Petroleum Board (CNSOPB) (Canada-Nova Scotia Offshore Petroleum Board, 2024) oversees and provides data relevant to offshore petroleum activities in Nova Scotia. This dataset includes multiple files covering significant discovery areas (NA coordinates), significant discovery licenses (NA cooordinates), production licenses (NA cooordinates), and the latest call for bids (NA polygons). The data encompasses information on various petroleum exploration and production activities. Key variables include license IDs, well IDs, geographic coordinates, exploration phases, and parcel identifiers. Data files include both Excel sheets (XLS, XLSX) and shapefiles, and they have been processed to create spatially explicit representations of petroleum activity sites. For the purpose of this project, we only retained the classification of each entry. The classification of petroleum offshore activities follows the same categories as the one presented in Table 4.2. Contrary to the data for Newfoudland and Labrador (see Section 4.2.1), there were no information included on the status and the time period covered.

  • Source: Canada-Nova Scotia Offshore Petroleum Board (CNSOPB)
  • Accessibility: Open
  • Data Type: XLS, XLSX, Shapefiles
  • Coverage: Offshore regions of Nova Scotia, Canada, 2009–2024
  • Processing Script: prc_offshore_petroleum_ns.R
  • Output File: offshore_petroleum_ns.gpkg

4.3 Offshore wind farms

4.3.1 Wind Regional Assessment - Canada

The Wind Regional Assessment - Canada (Wind Regional Assessment Team, 2024) identifies areas for potential offshore wind development. This dataset includes shapefiles representing:

  • Potential Future Development Areas (PFDA) in Nova Scotia.
  • Preliminary Offshore Wind Licensing Areas (PLA) in Newfoundland.

These areas were identified in the Wind Regional Assessment Interim Report for offshore wind energy. While still in the planning phase, this dataset provides critical spatial data for understanding the geographic distribution of proposed offshore wind development areas. Data from both regions are expected to be made publicly available via the Canadian Impact Assessment Registry and Open Government Portal. The data was transferred manually to the Canadian Wildlife Service and are currently stored on a secure Google Cloud Storage bucket accessible through an authentication key and maintained by inSileco. For this project, data from both regions were integrated into a unified spatial layer. Each entry was given a unique identifier to track its origin and dataset of provenance (e.g., wind_can_ns_XXXX for Nova Scotia and wind_can_nfl_XXXX for Newfoundland).

  • Source: Wind Regional Assessment - Canada
  • Accessibility: Restricted Access
  • Data Type: Shapefiles
  • Coverage: Nova Scotia and Newfoundland, Canada
  • Processing Script: prc_offshore_wind_can.R
  • Output File: offshore_wind_can.gpkg

4.3.2 Focus Area for the Regional Assessment of Offshore Wind Development in Newfoundland and Labrador

The Focus Area for the Regional Assessment of Offshore Wind Development in Newfoundland and Labrador (Impact Assessment Agency of Canada, 2024a) defines priority areas for offshore wind development in Newfoundland and Labrador. The assessment, conducted by the Impact Assessment Agency of Canada, aims to provide comprehensive information, knowledge, and analysis about the potential environmental, social, economic, and health effects of offshore wind development. In November 2023, the Committee identified a Focus Area within a portion of the Study Area, prioritizing this region for future wind energy projects. The dataset includes spatial data representing the Focus Area and is accompanied by a regional assessment agreement document. Key variables include focus_area, wind_energy_assessment, and environmental_impact.

  • Source: Impact Assessment Agency of Canada
  • Accessibility: Open Government Licence - Canada
  • Data Type: Shapefiles, PDF
  • Coverage: Newfoundland and Labrador, Canada
  • Processing Script: prc_offshore_wind_nfl.R
  • Output File: offshore_wind_nfl.gpkg

4.3.3 Study Area for the Regional Assessment of Offshore Wind Development in Nova Scotia

The Study Area for the Regional Assessment of Offshore Wind Development in Nova Scotia (Impact Assessment Agency of Canada, 2024b) outlines the geographic region under evaluation for future offshore wind development. This area was defined by the Impact Assessment Agency of Canada in collaboration with Natural Resources Canada and the province of Nova Scotia. The assessment focuses on evaluating the potential environmental, social, and economic effects of offshore wind energy development within the defined study area. The dataset includes spatial data for the study area and a supporting agreement document.

  • Source: Impact Assessment Agency of Canada
  • Accessibility: Open Government Licence - Canada
  • Data Type: Shapefiles, PDF
  • Coverage: Nova Scotia, Canada
  • Processing Script: prc_offshore_wind_ns.R
  • Output File: offshore_wind_ns.gpkg

4.3.4 Bureau of Ocean Energy Management (BOEM) Offshore Wind Data

The Bureau of Ocean Energy Management (BOEM) (Bureau of Ocean Energy Management, 2024) manages offshore wind energy activities across U.S. waters. This dataset provides detailed geospatial and tabular data on wind energy development, including lease areas, planning areas, marine hydrokinetic leases, project phase areas, and infrastructure components such as turbines, substations, and cable systems (Table 4.4). Each entry in the dataset is accompanied with a lease or protraction number (see Table 4.5). The dataset supports decision-making and analysis for renewable energy development by offering comprehensive coverage of offshore wind projects. Access to the data requires a ArcGIS online membership. Data were therefore harvested and stored on a secure Google Cloud Storage bucket accessible through an authentication key and maintained by inSileco.

  • Source: Bureau of Ocean Energy Management (BOEM)
  • Accessibility: Open Government Licence - United States
  • Data Type: Shapefiles, Geodatabase
  • Coverage: United States Offshore, 2020–2024
  • Processing Script: prc_offshore_wind_usa.R
  • Output File: offshore_wind_usa.gpkg
Table 4.4: Classifications and descriptions of BOEM offshore wind data components.
Classification Description
Leases Represents areas designated for offshore wind energy activities under BOEM’s leasing program. Includes four types of leases: ‘Commercial’ leases for large-scale wind energy production, ‘Easement’ leases for infrastructure such as cables or substations, ‘Right-of-Way Grant’ leases for energy transmission corridors, and ‘Research’ leases for testing and studying renewable energy technologies.
Planning areas Defines geographic areas identified by BOEM for potential offshore wind development. These areas are used for assessing wind energy resources and determining where to initiate leasing activities, based on environmental, economic, and technical considerations.
MHK leases and planning areas Includes Marine Hydrokinetic (MHK) areas for renewable energy from ocean-based resources such as waves, tides, or currents. MHK leases encompass ‘Research’ leases for testing technology, ‘Right-of-Way Grant’ leases for transmission infrastructure, and ‘Easement Area’ leases for supporting installations like substations.
Project phase areas Represents specific geographic areas proposed for different phases of offshore wind energy projects. These areas are identified based on project development stages, including site assessment, construction, and operations.
Substations Provides point data representing substations with their status (either ‘installed’ or ‘proposed’) and location type (either ‘offshore’ or ‘onshore’).
Turbine locations Provides point data representing the locations of wind turbines with their status as either ‘proposed’ or ‘installed’.
Cable interconnections Represents spatial data for cable interconnections linking offshore wind installations to onshore infrastructure or between offshore installations. Includes the status of cables as either ‘proposed’ or ‘installed’.
Ocean observing devices Includes point data for devices deployed to observe ocean conditions in areas relevant to offshore wind development. Observing devices monitor factors such as currents, waves, and water quality, and are categorized by their status as either ‘proposed’ or ‘installed’.
Export cable corridors Represents spatial data for corridors used by export cables connecting offshore wind farms to onshore substations. Includes the status of corridors as either ‘proposed’ or ‘installed’.
Cable landings Provides point data for locations where export cables come ashore to connect offshore wind farms to the terrestrial power grid. Includes the status of cable landings as either ‘proposed’ or ‘installed’.
Project inter-array cables Represents spatial data for inter-array cables within an offshore wind farm. These cables connect turbines to each other and to the offshore substation, with their status categorized as either ‘proposed’ or ‘installed’.
Table 4.5: Comparison of lease numbers and protraction numbers in BOEM offshore wind datasets
Lease numbers Protraction numbers
Purpose Identifies specific lease agreements for offshore activities. Identifies geographic areas on the Outer Continental Shelf (OCS) used for spatial organization.
Format Often alphanumeric (e.g., ‘OCS-A 0501’). Often numeric or grid-based (e.g., ‘NG14-02’).
Focus Administrative and contractual. Geographic and spatial organization.
Context Specific to a lessee’s activities. Applicable to the broader mapping system.

4.4 Automatic Identification System (AIS) Shipping Data

The Automatic Identification System (AIS) (Transports Canada, 2023) collects dynamic positional data from ships to enhance maritime safety, monitor environmental impacts, and facilitate shipping analyses. This dataset, provided by Transport Canada, focuses on 2023 vessel movements across Atlantic Canada. The AIS data provides dynamic vessel information, including position, speed, and course, captured through terrestrial, satellite, and tier-based receivers (Table 4.6). The AIS data shared includes both terrestrial and satellite AIS messages, fused to minimize spatial and temporal gaps and stored as compressed parquet files. The data shared also contains pre-processed tracklines as a geodatabase. However, we used the raw AIS positional data to recreate tracklines for the purposes of this project to allow for more temporal granularity. The data was transferred manually to the Canadian Wildlife Service and are currently stored on a secure Google Cloud Storage bucket accessible through an authentication key and maintained by inSileco.

  • Source: Transport Canada
  • Accessibility: Restricted
  • Data Type: Parquet, Geodatabase
  • Coverage: Atlantic Canada, 2023
  • Processing Script: prc_shipping_ais_tracklines.R
  • Output File: shipping_ais_tracklines.gpkg
Table 4.6: Columns (attributes) associated with the derived positional AIS parquet files (table from Transports Canada (2023)).
Column_Name Description Information Data_Type
MMSI Maritime Mobile Service Identity 9-digits (MMSIs with valid ship country codes are from 201000000 to 775999999). All ship MMSIs use the format M1I2D3X4X5X6X7X8X9, the first three digits represent the Maritime Identification Digits (MID) and they represent the flag/nationality of the ship. The following digits (X4-X9) can be any numerical value from 0 to 9. Integer
Latitude Latitude of the positional message Latitude recorded in decimal degrees with an accuracy of approximately 10 metres (-90 to 90). Float
Longitude Longitude of the positional message Longitude recorded in decimal degrees with an accuracy of approximately 10 metres (-180 to 180). Float
SOG Speed over ground of vessel The speed of the ship with respect to the ground, recorded in knots (0 to 102.2 are valid; 102.3 is invalid). Float
COG Course over ground of vessel The course of the ship, recorded in degrees (0-359.9 are valid; 360 is invalid). Float
Heading True heading of vessel The true heading of the ship, recorded in degrees (0-359 are valid; 360 is invalid). Integer
YMD_HMS Time of positional message in UTC The UTC time stamp of when the positional AIS message was sent from the ship (YYYY-MM-DD hh:mm:ss). Datetime
Source Data source/message type of AIS message The source (e.g., terrestrial, dynamic, satellite) and class (A or B) of the AIS message sent (e.g., TAIS_A, TAIS_B, SAIS_A). String

4.4.1 Data Processing Overview

To create meaningful tracklines representing vessel movements over space and time, we followed the processing steps described in Veinot et al. (2023):

  1. Data Extraction
    • AIS data was extracted from compressed archive files and loaded into memory using efficient tools such as the arrow (https://CRAN.R-project.org/package=arrow) and vroom (https://CRAN.R-project.org/package=vroom) R packages. This ensured scalability given the dataset size, which often exceeds 10GB annually.
  2. Initial Data Cleaning
    • Raw AIS data was filtered to remove invalid records:
      • Only positions with valid MMSI values (201000000–775999999) were retained.
      • SOG values were restricted to the range 0.1–100 knots to remove implausible speeds.
  3. Day/Night Classification
    • Using the suncalc R package (https://cran.r-project.org/web/packages/suncalc/index.html), each AIS position was classified as daytime or nighttime. The classification was based on local sunrise and sunset times at each position, ensuring alignment with the UTC timestamp.
  4. Trackline Generation
    • Tracklines were created by:
      1. Sorting AIS data by MMSI and timestamp.
      2. Calculating distances between consecutive positions using geodesic measurements.
      3. Splitting movements into separate tracks when:
        • The distance exceeded 50 nautical miles.
        • The time interval between positions exceeded 300 minutes.
      4. Further segmenting tracklines to differentiate between daytime and nighttime navigation.
  5. Trackline Attributes
    • Additional attributes were calculated for each trackline to facilitate further analysis:
      • Average and Maximum Speeds: Derived from AIS speed data.
      • Number of Positions: Total AIS messages contributing to the trackline.
      • Elapsed Hours: Duration of the trackline.
      • Track Length: Computed in kilometers using geodesic measurements.
      • Daytime Classification: Indicates whether the trackline occurred during the day, night, or a mix of both.
  6. Post-Processing
    • Erratic tracklines were flagged and removed using heuristic thresholds:
      • Extremely short tracks (<100 meters).
      • Unrealistic average speeds (<0.3 knots or >100 km/h).
      • Speeds exceeding vessel-type-specific thresholds:
      • 30 km/h for pleasure vessels and special ships.
      • 35 km/h for fishing and tug vessels.
      • 40 km/h for all other vessel types.
    • Tracklines identified as spoofing were excluded from further analyses.
  7. Output
    • The cleaned tracklines were exported as monthly parquet files (.parquet). Each file contains the processed AIS points with vessel information and the necessary information to differentiate individual tracks and whether tracks are during daytime or nighttime. Each point includes relevant attributes, such as speed, duration, and daytime classification, enabling detailed analyses of vessel movements across Canadian waters (Table 4.7).
Table 4.7: Columns (attributes) associated with the AIS positional points processed from the raw parquet files.
Column_Name Description Information Data_Type
mmsi Maritime Mobile Service Identity 9-digit identifier assigned to a vessel Numeric
latitude Latitude of the AIS positional point in decimal degrees Recorded in WGS 84 (EPSG:4326) Numeric
longitude Longitude of the AIS positional point in decimal degrees Recorded in WGS 84 (EPSG:4326) Numeric
sog Speed over ground at the AIS positional point Measured in knots converted to kilometers per hour Numeric
ymd_hms Timestamp of the AIS positional point in UTC Recorded in POSIXct format with timezone UTC POSIXct
date Date corresponding to the AIS positional point Derived from the timestamp for easier grouping and filtering Date
day_or_night Classification of the AIS positional point as day or night Derived using sunrise and sunset times from solar calculations Character
dist_miles Distance (in miles) between consecutive AIS positional points Calculated using geodetic distance between successive points Numeric
time_diff Time difference (in minutes) between consecutive AIS positional points Difference between timestamps of successive points Numeric
new_track Flag indicating whether a new trackline is initiated at the point Determined based on distance or time thresholds for new tracklines Logical
track_id Unique identifier for each trackline within a ship’s movement Generated using the cumulative sum of new_track flags Integer
day_night_segment Unique identifier for segments of tracklines split by day or night Calculated to split tracklines into day and night segments Integer

4.5 Visible Infrared Imaging Radiometer Suite

The Visible Infrared Imaging Radiometer Suite (VIIRS) is a key instrument aboard the National Oceanic and Atmospheric Administration’s (NOAA), Suomi National Polar-orbiting Partnership (Suomi NPP) and NOAA-20 satellites. VIIRS collects visible and infrared imagery and radiometric measurements, supporting applications such as environmental monitoring, weather forecasting, and maritime surveillance, including boat detection.

4.5.1 VIIRS Boat Detection (VBD) Dataset

The VIIRS Boat Detection (VBD) Dataset (Earth Observation Group, Payne Institute for Public Policy, 2024a) provides nightly global data on boat detections based on radiance observations captured by the Visible Infrared Imaging Radiometer Suite (VIIRS). This dataset is widely used for maritime surveillance, tracking vessel activity, and monitoring illegal fishing. The raw data includes detections with associated radiance, confidence metrics, and spatial coordinates (Table 4.8). The data was acquired for the years 2020–2023 and pre-processed to facilitate further analysis of vessel patterns within the Canadian Exclusive Economic Zone (EEZ). The processed data is stored as a Geopackage and includes filtered, high-confidence detections with key attributes for radiance intensity and spatial location.

  • Source: Earth Observation Group, Payne Institute for Public Policy
  • Accessibility: Public domain with registration
  • Data Type: CSV files
  • Coverage: Global, 2020–2023
  • Processing Script: prc_viirs_boat_detection.R
  • Output File: viirs_boat_detection.gpkg
Table 4.8: Columns (attributes) associated with the VIIRS Boat Detection dataset used for this project.
Column_Name Description Data_Type
Date_Mscan Date and time of the observation in UTC POSIXct
Lat_DNB Latitude of the detected boat in decimal degrees Numeric
Lon_DNB Longitude of the detected boat in decimal degrees Numeric
Rad_DNB Radiance intensity detected at the location (nW/cm²/sr) Numeric
QF_Detect Quality flag indicating confidence in the detection (1 = high confidence) Integer

4.5.1.1 Data Processing Overview

To prepare the VBD dataset for further analyses, the following pre-processing steps were applied:

  • Raw .csv.gz files were downloaded and processed using the arrow package, which allows for efficient reading of compressed CSV files.
  • High-confidence detections were retained by filtering for rows with:
    • QF_Detect = 1: Ensures only high-confidence detections are included.
    • Rad_DNB > 0: Removes detections with non-positive radiance values.
  • Optionally, filters for SMI (Signal-to-Mean Index) and SHI (Signal-to-High Index) can be applied for stricter data quality criteria (e.g., SMI > 0.1, SHI > 0.75).
  • The filtered dataset was converted to a spatial object with spatial projection EPSG:4326 using the sf package, enabling geospatial analyses and integration with other datasets.
  • Observation timestamps were standardized to UTC using the lubridate package to ensure consistent temporal referencing across datasets.
  • The pre-processed dataset was saved as a Geopackage (viirs_boat_detection.gpkg)

4.5.1.2 Signal-to-Mean Index (SMI) and Signal-to-High Index (SHI) in VIIRS Boat Detection Data

The Signal-to-Mean Index (SMI) and Signal-to-High Index (SHI) are metrics included in the VBD dataset that provide additional quality indicators for radiance detections. While these indices were not directly applied in the current processing pipeline, they could be used for further refining data quality and reducing false positives:

  • SMI: Measures the signal strength relative to the mean radiance value in the surrounding area. Higher values indicate stronger, more distinct detections.
  • SHI: Measures the signal strength relative to a high radiance threshold, helping to exclude detections influenced by bright land-based lights or other high-radiance sources.

Future iterations of the pipeline could incorporate these metrics for scenarios requiring stricter detection thresholds, such as high-confidence studies of vessel activity in heavily lit areas.

4.5.2 VIIRS Night Fire (VNF) Dataset

The VIIRS Night Fire (VNF) Dataset (Earth Observation Group, Payne Institute for Public Policy, 2024b; Elvidge et al., 2013b) provides nightly and annual global thermal anomaly detections captured by the Visible Infrared Imaging Radiometer Suite (VIIRS). This dataset is commonly used for monitoring gas flares, wildfires, industrial emissions, and other heat sources. The raw data includes processed radiance-derived fields (e.g., temperature and radiant heat indices), quality flags for detection confidence, spatial coordinates, and emitter-specific information. Table 4.9 presents a subset of available attributes for the nightly data that were selected for their relevance to this project.

In addition to the nightly VNF dataset, the annual VNF data from 2017 to 2023 was harvested. This dataset provides aggregated yearly thermal anomaly detections, including gas flares, with data specific to Canada. The processed dataset contains fields for average temperature, detection frequency, and estimated burnable carbon mass (BCM), among others (see Table 4.10).

  • Source: Earth Observation Group, Payne Institute for Public Policy
  • Accessibility: Restricted with credentials
  • Data Type: Nightly CSV files and annual GeoPackage files
  • Coverage: Global
  • Processing Scripts: prc_viirs_night_fire.R, prc_viirs_night_fire_annual.R
  • Output Files: viirs_night_fire.gpkg, viirs_night_fire_annual.gpkg
Table 4.9: Columns (attributes) associated with the VIIRS Night Fire dataset used for this project.
Column_Name Description Data_Type
Date_Mscan Date and time of the observation in UTC POSIXct
Lat_GMTCO Latitude of the detected thermal anomaly in decimal degrees Numeric
Lon_GMTCO Longitude of the detected thermal anomaly in decimal degrees Numeric
Temp_primary Primary temperature of the emitting source (Kelvin) Numeric
ESF_primary Emission scaling factor for the primary source Numeric
RHI_primary Radiant heat index for the primary source Numeric
Area_primary Estimated area of the primary emitting source (square meters) Numeric
Temp_secondary Secondary temperature of the emitting source (Kelvin) Numeric
ESF_secondary Emission scaling factor for the secondary source Numeric
RHI_secondary Radiant heat index for the secondary source Numeric
id_iremitter Unique identifier for the potential emitter Character
Type_iremitter Type of the emitter (e.g., industrial, flaring) Character
Category_iremitter Category of the emitter (e.g., petroleum, thermal anomaly) Character
Table 4.10: Columns (attributes) associated with the annual VIIRS Night Fire dataset used for this project.
Column_Name Description Data_Type
avg_temp_k Average temperature of the emitting source (Kelvin) Numeric
detection_frequency Detection frequency as a proportion of clear-sky observations Numeric
clear_obs Number of clear-sky observations available for the source Numeric
type Type of the thermal anomaly (e.g., flare, fire, etc.) Character
bcm Estimated burnable carbon mass (BCM) Numeric
year Year of the aggregated data Integer
geom Spatial point geometry (latitude and longitude) sfc_POINT

4.5.2.1 Data Processing Overview

To prepare the VNF dataset for further analyses, the following pre-processing steps were applied, with information on flares coming from Elvidge et al. (2016) and Zhizhin et al. (2021):

  1. Data Extraction
    • Raw CSV files were processed using arrow::read_csv_arrow() for efficient import.
    • Only necessary columns were selected during import to minimize memory usage.
  2. Quality Filtering
    • Data quality was ensured by applying bitwise filtering to the QF_Detect field.
    • Flags corresponding to high-confidence detections and specific radiance thresholds were retained:
      • High Radiance Thresholds: Retained flags ensuring significant thermal anomalies in bands M07–M16.
      • Thermal Anomalies: Retained detections flagged as local maxima or cluster detections.
  3. Spatial Filtering
    • The dataset was filtered to detections within a bounding box approximating the Canadian Exclusive Economic Zone (EEZ):
      • Longitude: -141° to -50°.
      • Latitude: 40° to 85°.
  4. Processed Radiance Fields
    • Radiance-derived fields, including Temp_primary, RHI_primary, and ESF_primary, were used for analyzing thermal anomalies.
    • Secondary radiance-derived fields (Temp_secondary, RHI_secondary, ESF_secondary) were retained for comprehensive heat calculations.
  5. Total Radiant Heat Calculation
    • A total_heat field was computed as the sum of RHI_primary and RHI_secondary, representing the combined radiant heat output of both primary and secondary sources.
  6. Flare Identification
    • A temperature threshold of 1450 K was applied to Temp_primary to flag likely gas flares.
    • This threshold is based on established studies identifying gas flares as high-temperature sources.
  7. Temporal Standardization
    • Timestamps were standardized to UTC using the lubridate package to ensure temporal consistency across datasets.
  8. Output
    • The pre-processed dataset was saved as a GeoPackage (viirs_night_fire.gpkg).
    • The GeoPackage contains spatial coordinates in the EPSG:4326 projection, radiance fields, emitter information, and retained quality flags.
Table 4.11: Retained Quality Flags for VIIRS Nightfire Data Processing.
Flag Description Rationale
1 M10 radiance above threshold Ensures M10 radiance is significant for thermal analysis.
2 M10 local maximum (bow-tie corrected) Captures prominent local maxima in M10, reducing noise from bow-tie duplication.
4 M7 radiance above threshold Identifies potential detections based on M7 radiance thresholds.
8 M8 radiance above threshold Ensures inclusion of detections with high M8 radiance, often indicative of thermal events.
16 M12 radiance outside scatterplot limits Captures M12 detections outside scatterplot-defined radiance limits, indicating anomalies.
32 M13 radiance outside scatterplot limits Captures M13 detections outside scatterplot-defined radiance limits, indicating anomalies.
64 DNB local maximum within 3 km of M10 Identifies significant DNB maxima close to thermal detections for contextual analysis.
128 M10 cluster detection Highlights clusters of M10 detections, often related to persistent sources.
65536 M11 cluster detection Highlights clusters of M11 detections for thermal anomaly identification.
2048 M10 tophat-filtered Filters M10 detections using a tophat approach to remove spurious signals.
4096 M12 tophat-filtered Filters M12 detections using a tophat approach to remove spurious signals.
8192 M13 tophat-filtered Filters M13 detections using a tophat approach to remove spurious signals.
262144 M11 tophat-filtered Filters M11 detections using a tophat approach to remove spurious signals.
524288 M14 local thermal anomaly Identifies strong thermal anomalies in M14, essential for flare detection.

4.5.3 VIIRS Nighttime Light (VNL) Dataset

The VIIRS Nighttime Light (VNL) Dataset (Earth Observation Group, Payne Institute for Public Policy, 2024c; Elvidge et al., 2013a) provides global observations of nighttime light intensity captured by te Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the Suomi NPP (npp) and NOAA-20 (j01) satellites. This dataset is widely used for analyzing human activity, urbanization, and environmental monitoring. The monthly VNL data includes radiance measures aggregated over time, with options for stray-light corrected data and various file types, including cloud-free composites and coverage metrics.

For this analysis, the file type avg_rade9h was selected, which represents the average radiance values over the specified temporal period, expressed in nanowatts per square centimeter per steradian (nW/cm²/sr). This file type is particularly suited for analyzing overall light intensity while excluding temporary anomalies. The data was obtained using the vcm configuration, which excludes stray light, ensuring higher data accuracy in regions with minimal artificial light. Additionally, the data was sourced from the j01 platform, representing the NOAA-20 satellite, part of the VIIRS suite, which provides enhanced coverage and resolution compared to earlier satellites.

The raw raster files were downloaded in GeoTIFF format and processed to focus on the Canadian Exclusive Economic Zone (EEZ). Processing involved loading the rasters, cropping them to a bounding box defined by the coordinates (-141, 40, -50, 85), and exporting each monthly raster as a Cloud Optimized GeoTIFF (COG). This format allows for efficient storage, faster data access, and better interoperability with modern geospatial tools, particularly for web-based or cloud-hosted applications. The processed COGs retain the original resolution and include internal overviews for optimized rendering.

4.5.3.1 Data Overview

  • Source: Earth Observation Group, Payne Institute for Public Policy
  • Accessibility: Restricted access, credentials required
  • Data Type: GeoTIFF files (.tif.gz)
  • Coverage: Global
  • Temporal Resolution: Monthly
  • Processing Script: prc_viirs_night_light.R
  • Output File: viirs_night_light.tif

4.6 References

Bureau of Ocean Energy Management (2024). United states offshore wind data - BOEM. Bureau of Ocean Energy Management Available at: https://www.boem.gov/renewable-energy/boem-renewable-energy-geodatabase.
Canada-Newfoundland and Labrador Offshore Petroleum Board (2024). Offshore petroleum data for newfoundland. C-NLOPB.
Canada-Nova Scotia Offshore Petroleum Board (2024). Offshore petroleum data for nova scotia. CNSOPB.
Earth Observation Group, Payne Institute for Public Policy (2024a). VIIRS Boat Detection (VBD) Dataset. Available at: https://eogdata.mines.edu/products/vbd/.
Earth Observation Group, Payne Institute for Public Policy (2024b). VIIRS Night Fire (VNF) Dataset. Available at: https://eogdata.mines.edu/products/vnf/.
Earth Observation Group, Payne Institute for Public Policy (2024c). VIIRS Night Light (VNL) Dataset. Available at: https://eogdata.mines.edu/products/vnl/.
Elvidge, C. D., Baugh, K. E., Zhizhin, M., and Hsu, F.-C. (2013a). Why VIIRS Data Are Superior to DMSP for Mapping Nighttime Lights. Asia-Pacific Advanced Network 35, 62.
Elvidge, C. D., Zhizhin, M., Baugh, K., Hsu, F.-C., and Ghosh, T. (2016). Methods for Global Survey of Natural Gas Flaring from Visible Infrared Imaging Radiometer Suite Data. Energies 9, 14. doi:10.3390/en9010014.
Elvidge, C. D., Zhizhin, M., Ghosh, T., and Hsu, F.-C. Annual Time Series of Global VIIRS Nighttime Lights Derived from Monthly Averages: 2012 to 2019. Remote Sensing.
Elvidge, C. D., Zhizhin, M., Hsu, F.-C., and Baugh, K. E. (2013b). VIIRS Nightfire: Satellite Pyrometry at Night. Remote Sensing 5, 4423–4449. doi:10.3390/rs5094423.
Environment and Climate Change Canada (2024). Integrated satellite tracking of pollution (ISTOP) program. Environment and Climate Change Canada.
Impact Assessment Agency of Canada (2024a). Focus area for the regional assessment of offshore wind development in newfoundland and labrador. Government of Canada Available at: https://open.canada.ca/data/en/dataset/81560d1e-8394-4b51-8212-2bbe4d8a5ea2.
Impact Assessment Agency of Canada (2024b). Study area for the regional assessment of offshore wind development in nova scotia. Government of Canada Available at: https://open.canada.ca/data/en/dataset/051c7034-c13b-49a9-9d4a-99b0b594d157.
National Environmental Emergency Centre (2024). Atlantic (including Quebec) notifications impacting or potentially impacting a waterbody with select substances, 2016-2023. Extracted by Philippe Muise 2024-01-29. Environment and Climate Change Canada.
Transport Canada (2024). National Aerial Surveillance Program (NASP). Data extraction for period between April 2012 to March 2023. Transport Canada.
Transports Canada (2023). Satellite navigation data from automatic identification system (AIS) in atlantic canada, 2023. Data provided by transports canada. Transports Canada.
Veinot, T., Nicoll, A., Rozalska, K., and Coffen-Smout, S. (2023). Vessel Density Mapping of 2019 Automatic Identification System (AIS) Data in the Northwest Atlantic. Can. Tech. Rep. Fish. Aquat. Sci. 3520: Vi + 29 p. Department of Fisheries and Oceans Available at: https://waves-vagues.dfo-mpo.gc.ca/library-bibliotheque/41105163.pdf [Accessed February 23, 2024].
Wind Regional Assessment Team (2024). Offshore wind development areas - canada. Environment and Climate Change Canada.
Zhizhin, M., Matveev, A., Ghosh, T., Hsu, F.-C., Howells, M., and Elvidge, C. (2021). Measuring Gas Flaring in Russia with Multispectral VIIRS Nightfire. Remote Sensing 13, 3078. doi:10.3390/rs13163078.