Canada’s maritime frontier: the science legacy of Canada’s extended continental shelf mapping for UNCLOS

It is a common adage that we know more about the Moon and Mars than we do about the oceans on Earth. As an example, according to the Seabed 2030 initiative, only 23.4% of the global ocean seafloor has been mapped with measured data, as of 2022 (https://seabed2030.org/mapping-progress). On average, there is one depth measurement in the ocean every 5 km. By comparison, 90% of Mars is mapped with elevation data to a scale of one measurement every 100 m (e.g., https://pgda.gsfc.nasa.gov/products/62; Smith et al. 2001). To determine extended continental shelf (ECS) limits as prescribed by Article 76 of the United Nations Convention on the Law of the Sea (UNCLOS), many coastal States needed to conduct detailed analysis of their offshore margins and Canada was no exception. These new data and information are leading to a revolution in our geoscientific understanding of the ocean realm.

Canada ratified the UNCLOS in 2003. With that ratification is an obligation to submit data and information to the U.N. pertaining to the limits of the country’s ECS; the portion of the juridical continental shelf that extends beyond 200 nautical miles (M). Canada submitted documents to the United Nations in 2013 for the Atlantic and 2019 for the Arctic that comprise its proposed outer limits for its ECS and the scientific evidence that justify these limits. These submissions will allow Canada to define with certainty its final maritime boundaries in the Atlantic and the Arctic. According to the criteria of the Convention, Canada could not justify an ECS in the Pacific.

Canada has the longest coastline of any country in the World (243 000 km). The area from the coastline to the outer limit of its exclusive economic zone (EEZ; 200 M limit) is 5.6 million km² (seventh largest in the World). Canada’s submissions to the U.N. for an ECS (the area beyond its 200 M EEZ) and an additional ∼2.4 million km² in area in the Atlantic and Arctic oceans, for which Canada exercises sovereign rights for the purpose of exploring and exploiting its natural resources. In total, therefore, Canada has sovereign jurisdiction for over 8 million km² of the ocean floor. This area is comparable to the size of the Canadian landmass, which is ∼10 million km².

To prepare these submissions, Canada undertook a scientific program of data compilation, acquisition, and interpretation in the Atlantic and in the Arctic. There were significant volumes of legacy geoscience and hydrographic data in the Atlantic, but few data in deep-water regions seaward of the margin. In the Arctic there were few data anywhere. New data were acquired in both regions, therefore. These acquisition programs involved a significant level of international collaboration. The submission documents remain confidential with the United Nations and yet the wealth of information accrued is immensely important for the advancement of scientific knowledge of Canada’s offshore territory.

Modern technologies permit the compilation of complex data sets into single geographic reference frames. With this capability, deep tectonic structures identified from potential field or seismic reflection and refraction data can be integrated with seafloor subbottom profiler and bathymetric data, as an example. Margin-wide, integrated data compilations, as a result, provide new understanding of the complexity of even passive continental margins and lead to challenges to existing paradigms such as location of the continent–ocean transition zone, source-to-sink sediment distribution patterns, and the shelf, slope, and rise classification system. It is the purpose of this paper to show the data sets acquired and compiled in the North Atlantic and Arctic oceans and to illustrate the impact these data sets have had on geoscientific understanding of these regions; from tectonic beginnings to modern sedimentation processes.

This paper introduces Article 76 of the UNCLOS (or the “Convention”) and the prescriptions contained within, to illustrate the data requirements to establish an ECS. The study then shows the data Canada accrued and examples of how these data have significantly amplified the geoscientific knowledge of offshore regions adjacent to Canada and how this science supports Canada’s ECS delineation. There is too much information to include how all of the complexities of Canada’s extensive margins were addressed, so examples will be shown where there has been significant advancement in knowledge and in particular for areas where there were previously few data.

Before proceeding, note that there are mixed meanings to the terminology employed. The Convention uses “continental shelf” and “continental margin” in a juridical (legal) sense and not geological. This paper will use these terms in their juridical sense and will use “geological continental shelf” and “geological continental margin” to discriminate geologic context. Also note that the unit “M” is used to describe nautical miles, not empirical miles. A nautical mile is 1.852 km. The word “State” is used in this text in conformity with the language of UNCLOS, meaning a sovereign country. A coastal State is one that has a coastline on the ocean and is able to establish coastal baselines from which to measure the breadth of its territorial sea.

The UNCLOS, ratified by Canada in 2003, provided Canada with the opportunity to define with certainty the outermost extent of its continental shelf. UNCLOS provides a comprehensive framework for governance of the World’s oceans. It was adopted in 1982 and came into force in 1994 after ratification by 60 States. Most States ratified the Convention by 1999 but the Canadian Government did not formally do so until 2003. Ratification implies that the State indicates its consent to be bound to the Convention. In Canada’s case, ratification of an international treaty generally requires an executive branch of the Federal Government, such as cabinet, to prepare an order in council authorizing the Minister of Foreign Affairs to sign. Since the Convention was already in force, Canada was immediately bound to the terms of the treaty once signed. Presently 168 countries have ratified the Convention.

Part V of the Convention describes the EEZ of a coastal State that extends to 200 M, measured from the territorial sea baseline (i.e., ∼coastline). How this baseline is measured depends on legislation of the State. In Canada’s case, it is based on surveyed benchmarks to which the low-water line along the coast is established. These points were formally established as an act of Parliament (Oceans Act). It identifies the sovereign rights of the coastal State that include the living and nonliving resources of the water column and the seafloor and its subsoil.

Part VI of the Convention and in particular Article 76 defines the juridical continental shelf of a coastal State and prescribes the methods for determination of the limits of the continental shelf. These limits mark the outermost extent of the coastal State’s jurisdiction that may extend beyond the 200 M (370.4 km) EEZ given certain bathymetric and geologic conditions. In these circumstances, the coastal State has sovereign rights of the resources of the seabed and the subsoil (subsurface) for the purpose of exploring and exploiting its natural resources (living and nonliving), according to Article 77. Note that beyond 200 M, these sovereign rights do not extend to the water column.

The Convention established a body of 21 scientific experts, known as the Commission on the Limits of the Continental Shelf (the Commission, or CLCS). These experts are nominated by coastal States and elected for 5 year terms by State Parties (member States) to the Convention. Canada has supported a member of the Commission since 2012. Where a coastal State’s continental shelf entitlement extends beyond 200 M, it is required to submit its proposed outer limits and the body of scientific evidence that supports them to the Commission. The Commission’s mandate is to assess this scientific evidence and the coastal State’s proposed outer limits. Based on this assessment and exchanges with the coastal State in question, the Commission makes recommendations on the State’s outermost limits, ensuring that these limits meet the criteria prescribed in Article 76 of the Convention. The process is an interesting confluence of science, diplomacy, and jurisprudence. While the Commission is not a decision-making body, its recommendations are a form of scientific authentication that is recognized internationally, allowing the coastal State to establish with legal certainty the boundary between seabed areas that fall under its jurisdiction and those of the international seabed area. Since delineation of the ECS under Article 76 is a scientific process, there may naturally be overlapping areas of entitlement between opposite and adjacent States. In these cases, recommendations of the Commission provide for entitlement but the boundary between States must be decided between the States or with the aid of another adjudicating body, such as the International Tribunal on the Law of the Sea.

While Canada’s submission has not yet been considered by the Commission, the rights over this area are inherent without precise delineation, according to Article 77 of the Convention. Article 76 merely defines these limits with certainty. Thus, until recommended otherwise by the Commission, Canada’s outermost limits are as proposed in its two submissions. These limits defined by points of latitude and longitude are published on the U.N. website in Canada’s executive summaries of the submissions: Atlantic, https://www.un.org/depts/los/clcs_new/submissions_files/submission_can_70_2013.htm and Arctic, https://www.un.org/depts/los/clcs_new/submissions_files/submission_can1_84_2019.html.

Article 76 of the Convention prescribes two formula lines for locating the extent of the continental margin of a coastal State and two constraint lines that are meant to limit these extents (Fig. 1; see Mayer and Mosher 2017 for a detailed explanation of the scientific aspects of Article 76). A combination of the formulae and constraints defines the extent of the continental shelf. According to the Convention, the continental margin “comprises the submerged prolongation of the land mass of the coastal State, and consists of the seabed and subsoil of the shelf, the slope and the rise. It does not include the deep ocean floor with its oceanic ridges and the subsoil thereof”.

The two formulae that define the outer edge of the continental margin (OECM) are (1) points at which the thickness of sedimentary rocks is at least 1% of the shortest distance from the foot of the continental slope and (2) points delineated not more than 60 M from the foot of the continental slope (Fig. 1A). Points may be connected by straight lines not more than 60 M in length. To define the OECM, the coastal State may use the seaward-most points identified by these formulae.

There are two constraint lines that apply after the OECM is identified (Fig. 1B): (1) a line not exceeding 350 M from the baseline of the coastal State (the coast). This line is often referred to as the “distance constraint” and (2) a line not exceeding 100 M from the 2500 m isobath (depth contour). This line is referred to as the “depth constraint”. Again, the seaward-most points may be used by the coastal State.

The OECM and the constraints are then combined to identify the outer limits of the continental shelf (OLCS) (Fig. 1C).

• If the OECM does not extend beyond the 200 M limit, then the 200 M limit forms the OLCS, unless of course it abuts another coastal State, in which case the boundary is subject to negotiation or adjudicated settlement.

• If the OECM does not extend beyond the constraint line, then the OECM forms the OLCS.

• If the OECM extends beyond the constraint line, then the constraint line forms the OLCS.

If, in either of the latter two cases, there is overlap with a neighbouring State, then the boundary is subject to negotiation between States or adjudicated settlement. The Commission plays no role in determination of boundaries between States and its recommendations are “without prejudice” to final delimitation between States.

Note that a critical metric cited in the formulae lines is the foot of the continental slope. Article 76 (Paragraph 4B) defines the foot of the continental slope: “In the absence of evidence to the contrary, the foot of the continental slope shall be determined as the point of maximum change in the gradient at its base”. The Convention does not define “base of the continental slope”, but the Commission, in its “Scientific and Technical Guidelines” defines it as “...a region where the lower part of the slope merges into the top of the continental rise, or into the top of the deep ocean floor where a continental rise does not exist”. The Commission considers the base of the continental slope to be identified through geomorphological analysis using bathymetric and geologic data. This rather simple concept is a relatively daunting exercise in practice as there are no simple criteria that can be universally applied to identify the base of the continental slope. Consider, for example, rift, transform, and subduction margins and the complexities inherent within each of these categories. Then consider islands and in particular islands on spreading ridges that have an inherent uplift due to high heat flow. As a final example, consider depositional processes that modify the tectonic morphology of the margin.

So far, the criteria discussed are considered largely morphological, i.e., the Commission interprets such terms as “natural prolongation of its land territory” and “submerged prolongation of the land mass” that appear in Paragraphs 1 and 3 of Article 76 as principally morphologic considerations. There must be a clear morphological continuation of the landmass from the coast to the foot of the continental slope. To complicate matters, however, a ridge may be a morphological continuation of the landmass but the geologic nature of the ridge is important as well.

Paragraph 3 of Article 76 states “The continental margin ... does not include the deep ocean floor with its oceanic ridges or the subsoil thereof”.

And Paragraph 6 of Article 76 refers to submarine ridges:

“… on submarine ridges, the outer limit of the continental shelf shall not exceed 350 nautical miles … This paragraph does not apply to submarine elevations that are natural components of the continental margin, such as plateaux, rises, caps, banks and spurs”.

The Commission interprets “natural components” to mean not just a morphological but also a geological connection. In these cases, the features may extend well beyond the 350 M constraint and in such cases the depth constraint (2500 m depth point plus 100 M) would be the only applicable constraint. A relatively straightforward example is the Grand Banks. In this case, the shallow water platform that is the geological continental shelf extends beyond 200 M and even beyond 350 M on Flemish Cap. There is a clear morphological continuation but also, the Grand Banks and Flemish Cap are underlain with continental rocks (e.g., Jansa and Wade 1975; Keen and Williams 1990; Welford et al. 2010a, efg201b). The foot of the continental slope and therefore the OECM lies well seaward of the 350 M distance constraint and even the depth constraint in some places. Such an assessment, as we shall see, is critical to Canada’s submission here and in the Arctic.

To synthesize this discussion, we can begin to assess some of the geologic and geomorphological evidence needed to identify the OLCS for Canada:

• identification of the base of the continental slope requires first and foremost bathymetric (i.e., measurement of water depth) and morphologic (i.e., shape of the seafloor) data. Geophysical (e.g., seismic reflection, gravity, magnetics) and geological (e.g., samples of rock type, geochemistry, sedimentology) data were submitted to provide evidence that supports the location of the base of the continental slope along Canada’s geological continental margins;

• identification of the foot of the continental slope requires morphologic and bathymetric data along Canada’s geological continental margins;

• application of the sediment thickness formula requires geophysical information, such as seismic reflection and refraction data, as well as drilling information if available to provide seismic velocity control. These data must extend well offshore of Canada from its geological continental margin to neighbouring abyssal plains

• classification of seafloor elevations that form part of, and extend from, Canada’s geological continental margins requires geophysical and geological information, including tectonic formation, rock types, and geological and geophysical structure; and

• note above that Paragraph 4B of Article 76 prefaces definition of the foot of the continental slope with the statement “In the absence of evidence to the contrary...”. The Commission interprets this text to mean that the foot of the continental slope, while still occurring within the base, might not be at the maximum change in gradient. Location of the continent–ocean boundary may supersede the morphologic criteria of maximum change in gradient if this location occurs elsewhere (i.e., seaward). “Evidence to the contrary”, therefore, is largely geophysical evidence that identifies the continent–ocean boundary. Canada has not used evidence to the contrary, so this criterion will not be addressed further.

For the purposes of presentation, this study considers each of Canada’s continental margins separately, starting with the Pacific margin, the Atlantic, and finally the Arctic.

Interaction of the Juan de Fuca, Explorer and Pacific plates and the North American Plate formed Canada’s Pacific margin (Hyndman and Rogers 2010) (Fig. 2). The relative motion of these plates is convergent in the south, seaward of Vancouver Island. To the north, however, seaward of the Queen Charlotte Islands, relative motion between the Pacific Plate and the North American Plate is largely strike-slip. As a result of these plate interactions, the British Columbia margin consists of a relatively narrow continental shelf and a steep continental slope (Fig. 2).

Note that a full ECS assessment of the British Columbia margin has not been conducted and the scenario presented here is for illustration purposes. The narrow shelf and slope of the BC margin results in a presumed base of slope that is relatively close to the subaerial landmass (i.e., <50 M (90 km) from the coast) (Fig. 2). This proximity to the coast and lack of sediment thickness offshore means that application of the formulae to define the OECM results in points that do not extend beyond the 200 M EEZ of Canada (Fig. 2). Canada, therefore, does not have obvious grounds for an ECS along this margin. No submission was made to the Commission for the Pacific region; however, Canada reserved the right to make a submission in future if it feels a case may be made given new information.

Prior to Canada’s ECS program, knowledge of Canada’s adjacent deep-water regions in the Atlantic (i.e., beyond the shelf break) was known mostly from potential field data. A few seismic lines and sediment cores and several scientific boreholes provided limited additional information. In addition to delineation of Canada’s maritime boundaries, a significant outcome of the ECS program, therefore, is an improved knowledge of marine regions of the Atlantic Ocean adjacent to Canada through the acquisition of new data and compilation of existing data into a geographic reference frame. The text below highlights a few significant scientific discoveries to illustrate some of this new knowledge and how it is applied in determination of Canada’s ECS.

Canada delineated an area that is approximately 1.2 million km² as its ECS in the Atlantic (Fig. 3) and submitted this consideration with accompanying scientific and technical material to the Commission. As noted above in reference to Article 77 of the Convention, Canada is able to exercise sovereign rights for exploring and exploiting the natural resources of the seafloor and its subsoil within its ECS. Figure 3 shows the offshore exploration wells. Note that a number of exploration wells are within Canada’s ECS area, in the Flemish Pass and Orphan Basin regions. Some of these wells are considered discoveries for commercial hydrocarbon extraction. If these discoveries are developed, Canada may well be the first country to commercially exploit its ECS.

The shelf and uppermost slope of the eastern Canadian margin had been studied since the 1960s, in large part because of establishment of the federal oceanographic research lab at Bedford Institute of Oceanography in Nova Scotia in 1962 (see Nettleship et al. 2014), and the presence of Memorial University of Newfoundland and Dalhousie and Saint Mary’s universities in Halifax. Earliest publications of the geology in the Atlantic offshore region appeared in the first issues of the Canadian Journal of Earth Sciences, including articles on seismic refraction studies of the eastern seaboard (Barrett et al. 1964) and seismic reflection studies of its crustal structure (Keen and Loncarevic 1966).

In the mid-1970s, the eastern Canadian margin became an area of active hydrocarbon exploration which added significantly to data holdings and provided incentive for numerous direct and indirect studies related to Canada’s offshore geology. As a result, the broad-scale tectonic and stratigraphic frameworks for the margin are well known (e.g., Keen and Williams 1990). Most of this research, however, was confined to the geological continental shelf and uppermost slope regions. Few data existed in the deep-water areas adjacent to the margin and fewer still were specific to the needs of ECS delineation. Canada conducted two major seismic reflection programs in ultradeep water as a result of the need for sediment thickness data (Fig. 4A) (see Verhoef et al. 2011, 2014 for details). In 2007, a total of 6900 km of new multichannel seismic (MCS) reflection data were acquired over the Sohm Abyssal Plain offshore Nova Scotia. In the deep-water portion of the Labrador Sea, an additional 3825 km of MCS data were acquired in 2009. Denmark, on behalf of Greenland, acquired 4692 km of MCS data in 2003 and 2006 in support of their ECS program in the Labrador Sea. In a collaborative agreement, Canada and Denmark exchanged data. The Canadian ECS seismic data holdings are available in Mosher et al. (2016b). An additional expedition was conducted in 2009 to acquire 2700 line-km of seismic refraction velocity information in Orphan Basin and Labrador Sea (see Funck et al. 2010) (Fig. 4A).

Canada also undertook two multibeam echosounder bathymetric surveys (MBES) as part of its ECS mapping to support identification of the base and foot of the continental slope positions and to accurately map the 2500 m isobath (Fig. 4B). In 2006, 16 800 line-km of MBES data were acquired in a single swath line pattern off the southern portion of the Labrador Sea and around the Grand Banks margin. A 7800 line-km of MBES data were acquired offshore Nova Scotia in 2012 through a collaborative effort with the U.S. National Oceanography and Atmospheric Agency. 3825 line-km of additional MBES data were acquired simultaneously with the abovementioned 2009 MCS expedition in the Labrador Sea. These data sets were added to existing Canadian Hydrographic Service (CHS) data holdings and publicly available data, including single-beam bathymetric data. Data are available through the CHS web portal (http://www.charts.gc.ca/index-eng.asp).

Subbottom profiler data, collected using a sonar system that utilizes seismic reflection principles and provides high-resolution images of the upper (<100 m) sediment column beneath the seafloor, proved to be of fundamental importance. These data were used to map the surficial and near-surface geology in support of identification of the base of the continental slope for the Atlantic margin. There is an abundance of these data in shelf and upper slope regions of the Atlantic margin, but few data exist in deep-water regions. A compilation was made of relevant existing data and, in concert with this compilation effort, much of these legacy data were scanned from hardcopy and georeferenced, so they are publicly available and suitable for integration within a Geographic Information System (https://ftp.maps.canada.ca/pub/nrcan_rncan/raster/marine_geoscience). As for all data types discussed above, this compilation is a work in progress as new expeditions and vessels of opportunity acquire additional data.

Compilation of all of these data types in deep water allows for construction of composite profiles (seismic, subbottom, and bathymetric) that extend from shallow water to the deep sea. Such profiles are useful to demonstrate natural prolongation and to locate the base of the continental slope but they also, for the first time, illustrate regional context of the continental margin geology. In context of the following discussion, a few of these profiles will be presented.

The tectonic framework of eastern Canada is the primary factor responsible for the margin’s present-day morphology. This framework is relatively well known for the Atlantic margin (e.g., Keen and Williams 1990; Louden 2002) so will be briefly reviewed here. A few elements of critical importance to the margin in ECS context and specifically to determination of the OECM will be highlighted in the following discussion.

Opening of the Atlantic Ocean began in the central Atlantic in the Triassic and propagated northward (see Louden 2002 for a summary). Rifting of the Scotian margin from north Africa and Portugal occurred in the Late Triassic to Early Jurassic (∼230–190 Ma). Evaporite deposits formed in the epeiric seas of this early ocean (e.g., Jansa and Wade 1975; Welsink et al. 1989; Wade and McLean 1990). Carbonates formed extensively along the margin during the rest of the Jurassic (170–149 Ma) and the resulting carbonate platform even today underpins the location of the shelf break (Weston et al. 2012). Ensuing post-rift subsidence led to development of accommodation space and accumulation of thick sequences of Jurassic to recent sediments in offshore sedimentary basins. Sediment loading led to further subsidence but also triggered movement of evaporites to form diapirs, domes, and canopies (Shimeld 2004; Ings and Shimeld 2006). These features come close and even crop-out at the modern seafloor. Diapirs also created accommodation space and thus account for some of the modern morphology of the margin.

Rifting continued northward to form the Grand Banks that ultimately separated from Iberia and Rockall Bank at about 140 Ma (King et al. 2020). Grand Banks is bound by the Newfoundland fracture zone in the south and the Charlie–Gibbs Fracture Zone in the north (Fig. 3). Rifting in this area is critically important to Canada’s ECS because it accounts for the great seaward extent of continental crust that underpins the Grand Banks of Newfoundland and Flemish Cap (Welford et al. 2010a, efg201b, 2020b). A major volcanic pulse off the Tail of the Banks is believed to have formed J-anomaly and Newfoundland ridges (Sullivan and Keen 1978; Tucholke and Ludwig 1982).

Greenland separated from Labrador at about 92 Ma in what seems to be initially a highly asymmetrical rifting pattern (Chian et al. 1995). Significant hyperextension was also documented (Keen et al. 2018). At about 60 Ma the main spreading axis (mid-Atlantic Ridge) switched to the east side of Greenland and seafloor spreading in the Labrador Sea ceased by about 40 Ma (Roest and Srivastava 1989). The Labrador Sea basin accumulated a large volume of clastic sediment influx during the Late Cretaceous and Tertiary, largely sourced from the Canadian margin as the basin continued to subside. In comparison, the southwest Greenland shelf is narrow and has experienced little or no subsidence south of 63°N (Rolle 1985). Highly folded and faulted Late Cretaceous and Tertiary strata along the Canadian margin suggest a period of tectonic inversion (Japsen and Chalmers 2000). This uplift presumably led to erosion and the shedding of more sediment to the basin. As a result of the vast amounts of sediment deposited in the basin, the sediment thickness formula may be used to identify portions of the OECM in the Labrador Sea.

The Eocene to Miocene interval records significant accumulations of contourites along geological continental margins of the Atlantic (Vahlenkamp et al. 2018); presumably related to major paleoceanographic changes due to opening and closing of North Atlantic tectonic gateways (e.g., Boyle et al. 2017). Major Oligocene hiatuses and contourites are notable along the eastern Canadian margin (Mosher et al. 2017; Mosher and Yanez-Carrizo 2021).

The latest major phase of influence on Canadian geological continental margin morphologies has been Pleistocene glaciations. Glaciers extended to the shelf edge at numerous times during the Pleistocene (Dyke et al. 2002; Piper et al. 2012; Margold et al. 2015). Melting of these glaciers resulted in the shed of vast volumes of sediment directly to deep portions of the margins. Evidence for such events is most apparent seaward of shelf crossing glacial troughs where deposits of glaciogenic debris flows form trough-mouth fans that are significant morphological anomalies (Piper et al. 2017; Mosher et al. 2017). Proximal and distal glacial/deglacial influences not only include direct sediment input, but also ocean circulation and relative sea level change (e.g., ∼120 m in the most recent glacial cycle). Deglaciation also led to isostatic rebound (e.g., Mazzotti and Adams 2005) which is thought to have generated ground accelerations that triggered submarine landslides along Canada’s geological continental margins (Mosher et al. 2004; Schulten et al. 2019a, efg162b; Normandeau et al. 2019b).

Morphology of the continental margin supported by geological data are critically important in identification of the base of the continental slope, according to Article 76. The Commission considers that the base of the continental slope lies where the lower part of the slope merges onto the top of the continental rise, or into the top of the deep ocean floor where a continental rise does not exist. Identification of the rise, if it exists, is paramount, therefore. Large compilations of new and existing data were made and shared by both Canada and the United States along their respective Atlantic margins for this purpose. These compilations offer an unprecedented opportunity to study the effects of geologic processes for the entire northeastern edge of the continent.

Mosher et al. (2017) and Mosher and Yanez-Carrizo (2021) conducted a geomorphological analysis of the entire northwest Atlantic margin to discover the geological processes involved in shaping the margin. In doing so, they challenged the entrenched paradigm of a geological continental margin consisting of a shelf, slope, and rise, and in particular the concept of the “rise”. The “rise” was not identified as a distinct feature of the margin prior to Heezen et al. (1959) who defined it based on strict gradient changes (i.e., between 1.5° and 0.08°). The type-section on which Heezen et al. (1959) based this definition was from the northeast United States. The “rise” in this type-section is the morphology generated by the Chesapeake Drift and Hatteras Outer Ridge—two large contour current generated deposits (contourites). Nowhere else along the north Atlantic margin, nor on most other margins, do such features generate such a massive morphological signature.

Mosher et al. (2017) used margin-crossing seismic profiles and detailed bathymetry to interpret that the shape of the margin resulted from a combination of original tectonic setting, accommodation space, and sedimentary processes. They determined that there is a continuum of margin shapes that reflects the ongoing balance of sedimentary processes between sediment starvation, contour current deposition, sediment mass failure, and turbidity current deposition (Fig. 5). When turbidity current process dominate, they suggest that the shape of the margin is graded (concave); if contour current sedimentation dominates, the shape is above grade linear to convex; and if mass failure dominates, the shape tends to be “stepped” (partly below grade and partly above grade). The latter depends, however, on the scale of mass failure. In the case of little sediment input, the margin retains its tectonic signature, which tends to be below grade.

Mosher and Yanez-Carrizo (2021) made the simplifying assumption that the base-level shape for the slope of a geological continental margin is an exponential decay or “graded” shape. This assumption is found in subaerial stream-bed research—a graded condition that reflects a dynamic equilibrium determined by the balance between inputs and outputs. They modeled a graded shape for the entire margin, using the shelf edge and abyssal depths as boundary conditions. They removed this modeled continental margin shape from measured bathymetry to yield residual bathymetry (Fig. 6). The resulting anomalies (i.e., departures from the graded shape) are highly informative; readily indicating, in three dimensions, areas that are graded, above grade, and below grade. Integrating these results with margin-wide seismic reflection data allows for determination of the causes of these anomalies.

In this geomorphological assessment, Mosher and Yanez-Carrizo (2021) show that more than a third of the margin by area fits the graded curve model (Fig. 6). A similar area is above grade, largely due to deposition of contourites, including those that comprise the massive Chesapeake Drift and Hatteras Outer Ridge. The Laurentian Fan is another large sedimentary deposit that forms an above-grade segment. Below-grade segments are most apparent along uppermost slope portions of the US segment and on the seaward flank of Flemish Cap. The former due to erosion and the latter area due to sediment starvation. These details will be discussed below in assessment of each region of Canada’s ECS.

Note in Fig. 6 that above-grade segments along the Canadian margin are typically above grade from the shelf edge to the abyssal plain, i.e., there is no above-grade anomaly seaward of the slope along any portion of the Canadian margin that might be considered a continental rise such as defined by Heezen et al. (1959). The base of the continental slope zone (BOS) lies consistently seaward of the above-grade portions, illustrating that there is no continental rise.

Much of the margin offshore Nova Scotia is graded to slightly above grade in shape (Fig. 6), meaning a constant curvature to a slightly more linear bathymetric profile shape (Figs. 7 and 8). Identification of the BOS for ECS purposes based on morphology alone was difficult, therefore. Geologic and geophysical support was needed to assist in this regard. Identification of features related to slope processes, such as canyons, channels and sediment mass failure scars and deposits, and certain types of contourite drift deposits was necessary. This analysis was greatly assisted with near 100% multibeam bathymetric coverage as well as an abundance of subbottom and seismic reflection data along this segment of the margin (Fig. 4).

A number of recommendations issued by the Commission have considered mass transport processes in assisting in identification of the BOS (see Mosher et al. 2016a). Initiation of mass failure events are considered processes of continental slopes and not rises (see for example, summary of recommendations for Ireland, http://www.un.org/depts/los/clcs_new/submissions_files/submission_irl.htm). In particular, the downslope termination of mass transport deposits, in certain cases, reflects the downslope limit of the continental slope (see for example, summary of recommendations for Norway, http://www.un.org/depts/los/clcs_new/submissions_files/submission_nor.htm).

To support identification of the base of the continental slope in the Nova Scotia region, therefore, mass failure deposits were identified. The Scotian margin has a long history of study of mass failure processes (e.g., Piper and Sparkes 1987; Mosher et al. 2004). Rarely, however, did such studies consider the downslope extent of such features. ECS multibeam bathymetric and subbottom profiler data were used to study near-surface deep-water sedimentary processes along the Scotian margin and in particular to assess the frequency of event processes such as submarine landslides and turbidity currents (Normandeau et al. 2019,efg139a, efg140b, efg141c; Normandeau and Campbell 2020). One such feature was identified on the western levee of Western Valley of Laurentian Fan (Figs. 9A and 9B). Normandeau et al. (2019b) collected shallow piston cores and additional ultrahigh-resolution seismic reflection data within this deposit. They concluded that this landslide occurred during the Holocene at about 4 ka. They also identified other debris flow deposits that occurred between 4 and 1.5 ka. From these new data, in combination with new insights into the 1929 event, they suggest a recurrence rate of one large submarine landslide per 1000 years along the eastern Canadian margin, two orders of magnitude more frequent than earlier estimates (e.g., Mosher et al. 2004).

Using MBES data acquired in Canada’s ECS program, Normandeau and Campbell (2020) acquired data from additional piston cores to assess the frequency of turbidity current activity in The Gully region of the Scotian Slope. They showed that greatest activity occurred between 24 ka cal BP (last glacial maximum) and 17 ka cal BP, and decreased significantly thereafter. This timing is in-phase with glacial retreat but out-of-phase with relative sea level change. This observation suggests that glacial meltwater produced during de-glaciation was the dominant source for turbidity activity along the margin. Normandeau et al. (2019a) also investigated seafloor bedforms on the Scotian Slope that are associated with turbidity current spillover (Fig. 9C). This research provided insight into the influence of turbidity currents versus contour currents on the morphology, geometry, and distribution of bedforms on the margin.

One of the most prominent morphological features along the Nova Scotia segment of the Canadian margin is the Laurentian Fan. Submarine fans pose some significant issues with respect to identification of the base of the continental slope (Mørk 2016). They tend to have no clear gradient change as they merge with the adjacent abyssal plain.

Laurentian fan includes two 400 km-long valleys with levees up to 700 m high. There have been a few studies of this fan (e.g., Piper et al. 1984; Skene and Piper 2006) but its great areal extent had made it difficult to capture its full scale in any detail. The ECS program provided new multibeam data (Mosher and Piper 2007) and combined with subsequent surveys (Krastel et al. 2016) now includes most of the entirety of the fan and its levee systems (Figs. 10 and 11). The positive residual bathymetric anomaly generated by the fan, as shown in Figs. 6 and 10, is 60 000 km² in size and comprises an estimated 115 000 km³ of sediment (Mosher and Yanez-Carrizo 2021). To put this in perspective, the Great Lakes contain 23% of the World’s freshwater at a volume of 22 671 km³, i.e., a fifth of the volume of the Laurentian Fan. The base of the continental slope appears at the seaward extent of this anomaly and is where the main channels and levees terminate.

These new data precipitated several new studies of the 1929 Grand Banks submarine landslide (Schulten et al. 2019a, efg162b), turbidity current (Stevensen et al. 2018) and ensuing tsunami (Løvholt et al. 2019; Zengaffinen et al. 2020). Schulten et al. (2019a) demonstrated for the first time evidence of deep-seated failure that resulted in a massive slump in the upslope area of the fan (Figs. 10 and 11). The slump reached depths of 500 m below seafloor, had 100 m of vertical displacement and possibly 300 m of down-slope (lateral) displacement. The volume displacement of this sediment mass likely caused the tsunami that impacted the south coast of Newfoundland and killed 28 people. Løvholt et al. (2019) and Zingaffinen et al. (2020) used these landslide metrics to numerically simulate the tsunami (see Fig. 11). Their results demonstrated that both deep-seated and shallow-sediment mass displacement was necessary to generate the tsunami in terms of timing, amplitude, and geographic distribution.

Sediment mass failure rapidly diluted into a high-speed turbidity current that successively broke undersea telegraph cables (Heezen and Ewing 1952). The turbidity current is calculated to have traveled at a maximum of 60 km·h⁻¹ (Heezen and Ewing 1952). The new multibeam data allowed a more accurate assessment of the path of the turbidity current and its thickness, given evidence of its deposits on elevated terraces. Stevenson et al. (2018) showed that the average thickness of the flow was 250 m high and they were able to calculate the average bulk sediment concentration of the flow at 2.7%–5.4% by volume as a result of these metrics. These concentrations are orders of magnitude higher than smaller volume flows in river deltas and submarine canyons that have been measured directly.

The deep-water basin adjacent to the Nova Scotia margin is named Sohm Abyssal Plain. It is the ultimate sink for sediment shed from much of eastern North America. Sediment was supplied along the length of the Scotian Shelf, including the Laurentian Channel and Fan system. The Laurentian Channel is the seaward extension of the St. Lawrence River and estuary, which presently drains the Great Lakes and catchment area. This modern drainage basin totals more than 240 000 km² of eastern North America. The channel was also a prominent ice outlet corridor during past glaciations, through which large volumes of sediment were delivered (e.g., Margold et al. 2015; Leng et al. 2018). Additionally, as will be shown in a later section, sediment was flushed through the Labrador Sea via the Northwest Atlantic Mid-Ocean Channel (NAMOC) and delivered to the Sohm Abyssal Plain just south and east of Newfoundland Ridge.

One of the contributing factors of sediment to the Sohm Abyssal Plain are large mass failure deposits. Deposits that are hundreds of metres thick and extend for hundreds of kilometers beneath the plain have been mapped (e.g., Mosher et al. 2010). This amount of material was presumably supplied by catastrophic collapses of segments of the margin slope. Deptuck and Campbell (2012) utilized the ECS Sohm seismic data set to map out the extent of a mass transport (failure) debris flow deposit that resulted from the Montagnais bolide impact off southwestern Nova Scotia. The impact occurred about 51 million years ago (Jansa and Pe-Piper 1987). Deptuck and Campbell (2012) recognized a large debris field in the seismic data that covers an area ∼93 000 km² in size and that travelled up to 580 km from the impact site (Fig. 12). They interpreted widespread margin collapse and catastrophic failure of the outer shelf and upper slope of southwestern Nova Scotia as a result of the impact.

The stratigraphy of this sedimentary succession beneath the Sohm Abyssal Plain was interpreted with a compilation of ECS and existing seismic reflection data by Desroches and Wade (2019) (Fig. 8). They correlated the stratigraphy to scientific drill sites, to the abundant seismic reflection data on the upper Scotian Slope, and subsequently to industry boreholes on the upper slope. With this correlation, they were able to extend seismic stratigraphic horizons of Wade and Maclean (1990), Wade et al. (1995), Weston et al. (2012), and Campbell et al. (2016) seaward (Fig. 8). Their analysis showed that Jurassic deposits are thickest immediately off the central and western parts of the margin. Thin deposits of this unit extend 700 km seaward of the shelf break in the west of the study area. In the central and eastern regions, however, this unit does not extend as far, implying a lack of sediment source. Early Cretaceous deposition appears to have been largely sourced from the Sable Island Bank area. It is thickest near the shelf edge, but it also forms a thick (∼700 m) deposit further offshore. From the mid-Miocene onward, deposition shifted dramatically to the eastern Sohm Abyssal Plain, as the Laurentian Channel became the main source of sediment for much of the region.

These stratigraphic ties to boreholes in addition to legacy and newly acquired refraction data allowed for determination of sediment thicknesses in the Nova Scotia offshore (Fig. 12). The Scotian margin hosts sediments well in excess of 6 km thickness. The sedimentary package thins in an offshore direction but still sustains sediments more than 2 km thick beneath the abyssal plain out to the 350 M constraint (Fig. 12). Aside from being critical in determination of Canada’s outer limits, such information will prove to be of fundamental importance to guide future hydrocarbon exploration and also for the rapidly developing discipline of exploration for carbon storage reservoirs.

The Grand Banks is an extensive shallow-water platform, generally lying in less than 200 m water depth, which is one of the few places in the World where the geologic continental shelf extends beyond the 200 M limit of the coastal State (Fig. 13). The geological continental margin of the Grand Banks region is complex—tectonically and morphologically. Tectonically, there are transform, rift, and hyperextended segments (e.g., Tucholke et al. 1989; Pe-Piper and Piper 2004; Lau et al. 2006; Sibuet et al. 2007; Welford et al. 2010a, efg201b, 2020b; Gerlings et al. 2011, 2012). The margin morphology is a consequence of these tectonic elements but in some cases modified by sedimentation patterns. Mosher et al. (2017) and Mosher and Yanez-Carrizo (2021) covered in some detail these morphological signatures and their underlying genesis. These will not be repeated but a few of the more critical elements and aspects that are unique to the Grand Banks margin in ECS context are presented.

Some tectonic elements of the Grand Banks margin are amongst the most studied in the World (e.g., Tucholke et al. 1989). By contrast, however, the southwest segment of the Grand Banks margin is relatively poorly studied. This component is thought to have been formed through left-lateral strike-slip displacement during seafloor spreading along a structure referred to as the SW Grand Banks Transform (Fig. 13). Pe-Piper and Piper (2004) suggested that the transform fault links with the Cobequid–Chedabucto fault zone onshore, and the Newfoundland Fracture Zone offshore. They suggested there has been displacement on this structure as recently as the Oligocene. While there are few direct observations of structure along this segment, it is a region of enhanced seismicity compared with remaining regions of the passive Atlantic margin (Fig. 13), perhaps indicative of latent motion along the structure even today. The 1929 M7.1 Grand Banks earthquake previously mentioned was a largely strike-slip event (Bent 1995) and is interpreted to have resulted along this structure.

Extending eastward for nearly 1000 km along strike from the SW Grand Bank margin is a feature known as Newfoundland Ridge (Figs. 13 and 14). This ridge has been the target of several scientific drill programs, but it was the sedimentary succession of contourites on top of the ridge that was sampled (e.g., Norris et al. 2012). Little is known about the underlying crustal structure, and there have been no direct studies of the deep structure since Sullivan and Keen (1978).

To demonstrate that the ridge is a natural prolongation and a natural component of the margin, morphologic, tectonic, residual bathymetric and crustal thickness assessments were made (Fig. 14). Morphologically, Newfoundland Ridge lies well above the surrounding seafloor, on average lying at about 3500 m water depth, while the adjacent ocean crust lies at 5200 m (Figs. 14A and 14E). Figures 13, 14A, and 14B show the relationship of the ridge with the Newfoundland Fracture Zone, which is suggestive of a causal relationship. The low Bouguer gravity anomaly suggests a deep moho at this location. Residual bathymetric analysis of Mosher and Yanez-Carrizo (2021) shows that the feature is highly anomalous, being above grade, along the margin (Fig. 14C) and the crustal thickness assessment shows that the feature is much thicker than oceanic crust. It is proposed that these anomalies are accounted for by volcanism related to transform faulting during and shortly after initial rifting. Magma intruded and extruded pre-existing continental crust. The Fogo Seamounts, located westward and along strike, are presumably related to this same volcanic activity.

As a result of the morphology generated by the ridge, the BOS lies where the ridge terminates against oceanic crust in about 5200 m water depth (Fig. 14E). Either constraint could apply since the ridge is considered to be an integral part of the geology of the margin, and thus a natural component of the margin.

Wide-angle reflection and refraction data across Orphan Basin (Line Hud2009019-01) were acquired in a 2009 ECS survey following the track of an earlier MCS reflection line (Line 84-3) (Funck et al. 2010). The new data facilitated refinement of the interpretation of the velocity structure beneath Orphan Basin. Welford et al. (2020b) showed thinned continental crust beneath Orphan Basin and Orphan Knoll that extends over 450 km ocean-ward (Fig. 15). Seaward of Orphan Knoll, the continent–ocean transition zone spans at least 80 km. Crustal velocities remain below 7 km·s⁻¹ and the top of the mantle is more than 20 km deep throughout Orphan Basin (Fig. 15).

As shown in Fig. 6, Orphan Basin is highly anomalous in terms of its residual bathymetry signature. As explained by Mosher and Yanez-Carrizo (2021), in large part this anomaly relates to the elevated nature of the stretched continental crust that underlies the basin. It is also in part related to the fact that Orphan basin is sediment filled with mass transport deposits and turbidites (e.g., Tripsanas et al. 2008; Bartel and Mosher 2009; Li et al. 2012a, efg98b). Figure 16 shows examples of these deposits at a variety of scales, as imaged in industry 3D seismic data, high-resolution seismic data and ultrahigh-resolution seismic data. A render of the basal surface of an MTD from 3D seismic data shows linear scours caused by scraping of the underlying sediment as the debris-flow flowed across the seafloor (Fig. 16D). These channels and mass transport deposits, combined with the continental nature of underlying crust, indicate Orphan Basin is a continental slope environment, even though it has an average seafloor gradient of just 0.4°.

Mass transport deposits of Orphan Basin onlap Orphan Knoll at the seaward extent of the basin (Fig. 17). Orphan Knoll is the seaward-most of a series of rotated continental blocks that underpin the basin (Welford et al. 2020b) (Figs. 13 and 15). DSDP drill Site 111 on the knoll sampled continental rocks. The knoll stands proud of the adjacent seafloor, making a seamount feature of about 35 000 km² in area (Fig. 17A). The seaward flank of Orphan Knoll lies 2400 m above the adjacent oceanic crust. This base of its steep flank, therefore, is the base of the continental slope (Fig. 17A).

More than 200 mounds, some in excess of 600 m high, on Orphan Knoll are apparent, recognized since earliest surveys of the feature, but the nature of these mounds remained enigmatic (e.g., (Enachescu 2004) (see Figs. 15D and 17D). They have been variably hypothesised to represent bioherms, fluid escape features, or mud diapirs but lack of detailed imaging and sampling hindered conclusive interpretation. Multibeam and subbottom data acquired as part of Canada’s ECS program contributed to new detailed imaging of these mounds (Meredyk et al. 2020). While Meredyk et al. (2020) claim there is more than 150 m of sediment drape on top of these mounds, careful processing of subbottom data show less than 50 m (Fig. 17D). Remotely operated vehicle dives showed no evidence of biologic construction of these features (i.e., they are not reef or bioherm structures) (Meredyk et al. 2020). While it has been hypothesized that they may represent fluid escape structures related to deep seated gas or fluid, their origins remain enigmatic.

Flemish Cap is a continental block that rotated away from Orphan Basin during rifting (Welford et al. 2010a). Its northeastern flank is a right lateral shear margin that, combined with seafloor spreading as the Labrador Sea opened, caused rotation of the block away from Orphan Basin. This motion caused hyperextension in Orphan Basin, as was noted above. Flemish Cap remains as a shallow water (<200 m deep) platform that is part of the Newfoundland geological continental shelf with continuous continental crust that extends beneath the Grand Bank, Flemish Pass, and Flemish Cap (Fig. 18).

The Grand Banks margin is not only tectonically complex but is also complicated in terms of sedimentary processes. The extensive geological shelf means that portions of the margin are far from subaerial terrain and are sediment starved (Mosher and Yanez-Carrizo 2021), while other areas were subjected to massive sediment flux from lower sea level stands and glacial input (e.g., Piper 2005; Tripsanas et al. 2008; Bartel and Mosher 2009; Li et al. 2012a, 2012b). While there are canyons and intercanyon areas in which turbidity current and sediment mass failure deposits dominate (e.g., Mosher et al. 2010; Giles et al. 2010; Rashid et al. 2017, 2019; Tang and Piper 2020), much of the deep-water margin was heavily influenced by contour current deposition. The margin’s geographic position between the open Atlantic and the Labrador Sea ensure it is swept with cold ocean geostrophic waters of the Labrador Current and North Atlantic Deep Water and warm surface waters of the North Atlantic Current including the Gulf Stream. Some of the earliest publications on sediments of this region appeared in the Canadian Journal of Earth Sciences (Carter et al. 1979) and pointed to the important role of deep currents in redistribution of sediments around the Grand Banks (Carter and Schafer 1983; Rashid et al. 2017; Boyle et al. 2017). Mosher et al. (2017) and Mosher and Yanez-Carrizo (2021) show how these contourites have influenced the morphology of the margin. This evidence is apparent even in the sediment-starved region around Flemish Cap (Mosher and Yanez-Carrizo 2021) (Figs. 18 and 19). By influencing the morphology of the margin, these contourites play a critical role helping recognize the base of the slope (Fig. 19).

There appears to be little sediment thickness seaward of the Newfoundland margin (Figs. 19 and 20). In large part, this thin sediment section is because of the extensive Grand Banks platform that isolates the margin from the subaerial landmass and sediment eroded from it. Additionally, accommodation space created by Orphan Basin, for example, has captured a significant amount of the sediment destined for the offshore. Figure 20 illustrates that even the 350 M distance constraint lies atop of the shallow-water portion of Flemish Cap, so the depth constraint is the more seaward for much of the margin.

The Labrador Sea is underlain by a semi-enclosed failed rift basin (Peace et al. 2017). Welford et al. (2020a) interpreted an extensive grid of newly acquired long-offset industry seismic reflection data in the Labrador Sea, along with coincident gravity data to produce a map of crustal type and sediment thicknesses. Their interpretations show continental crust extends well out into the basin in the southern Labrador Sea as illustrated in Fig. 21. Additionally, their crustal thickness map shows extensive thickened crust in the northern portion of the Labrador Sea.

Delescluse et al. (2015) analysed data acquired from the 2009 ECS seismic refraction expedition (see Funck et al. 2010) along and across the extinct spreading centre of the Labrador Sea (Fig. 21). Their results, in combination with those of Keen et al. (2018), showed that seafloor spreading was ultra slow in its final stages and that there was extensive continental crust extension and limited development of oceanic crust during formation of the basin. Oceanic crust is thin and serpentinized with the mantle close to the seafloor, suggestive of hyper-extension (Fig. 22).

As a semi-enclosed basin, the Labrador Sea is flanked on its west side by Labrador, Greenland on its east side (Fig. 22) and Hudson Strait, Baffin Island and Davis Strait to the north. With these proximal landmasses and corridors to eastern Arctic drainage systems including glacial drainage, the basin has received significant volumes of sediment throughout its history. In addition, as a northern latitude basin, cold deep ocean currents circulate along its margins to redistribute this sediment input. Dickie et al. (2010, 2011) interpreted new and existing regional seismic reflection data and integrated their interpretation with new biostratigraphic data from exploration wells on the Labrador Shelf. From this analysis, they were able to derive a more precise evaluation of sediment deposition and paleo-environmental history of the region. These results included identification of a number of regional unconformities related to Cretaceous rifting and seafloor spreading, a change in spreading direction, and Paleocene-Eocene episodic volcanism that is presumed to be due to passage of the proto-Iceland hotspot. Peace et al. (2017) contest this latter point however.

Mosher et al. (2017) and Mosher and Yanez-Carrizo (2021) showed that most of the Labrador margin is graded in shape, indicated by its exponential decay shape in profile (i.e., no residual bathymetric anomaly) (Figs. 6 and 23A). They attribute this shape to the dominance of turbidity current processes along the margin. This interpretation is largely based on extensive mapping of the margin by Hesse and others (e.g., Hesse 1992; Wang and Hesse 1996; Hesse et al. 1997, 1999). As a consequence of this graded shape, it is difficult to determine the base of the slope using morphology alone, as gradients gradually diminish in the transition from slope to basin floor (Figs. 21 and 23A, inset). Geological supporting evidence is required, therefore, to identify the base.

As a result of this need for supporting geologic evidence, existing surficial geology maps such as those of Josenhans et al. (1986) and Hesse et al. (1997) were compiled and modified with newer data to produce a surficial geology map of the Labrador margin (Fig. 23B). Geologic elements such as turbidite channels and channel levee systems and mass transport deposits are largely considered to be processes of the slope. The deep ocean floor of the Labrador basin is dominated by the NAMOC system, as will be discussed below. The base of the slope is considered to lie where slope features terminate and deflect southward due to the overall dip of the basin, and features of NAMOC dominate (Fig. 23).

There are at least two regions of the Labrador margin that are exceptional with respect to the shape of the margin. One such region is an area seaward of Makkovik Bank and Hopedale Saddle that shows a dramatic below-grade anomaly (Fig. 23A). This negative residual is the headwall region of several episodes of massive margin collapses that resulted in a series of stacked mass-transport deposits further seaward (Fig. 24) (Deptuck et al. 2007). Note the steep upper slope seaward of Makkovik Bank on Fig. 22, inferred to represent this headwall region, and features further seaward labelled “MTD”. The base of the slope lies at the seaward extent of these mass-transport deposits (Fig. 22).

Another significant departure from the graded slope along the Labrador margin is Hamilton Spur, a detached contourite drift. While a similar structure on the opposite side of the Labrador Sea, south of Greenland, known as Eirik Drift (Fig. 23) has been well studied, including scientific drilling (e.g., Channell et al. 2005; Müller-Michaelis et al. 2013), Hamilton Spur has not been studied at all. The profiles shown in Fig. 25 are amongst the first published from this feature.

Seismic profiles down axis and across Hamilton Spur show its structure and form, including sediment waves, large-scale cross-bedding, and truncation surfaces (Fig. 25). The spur forms a near-linear seafloor profile that extends from the shelf edge to the deep sea (Fig. 25A) and creates an above-grade bathymetric anomaly (Fig. 23A). At the seaward end of the profile is the NAMOC. The channel’s western levee onlaps Hamilton Spur (Fig. 24C), suggesting deposits of NAMOC are younger than those of the spur. This onlap creates a change in gradient and this morphological and sedimentological change from the spur to the levee is the only logical place to search for the foot of the slope.

The Labrador Sea is not a typical ocean basin from either tectonic or sedimentologic perspectives. It has sustained high rates of sediment input, largely as a result of adjacent continental glaciations, which has resulted in its general graded shape of the margin. A fascinating feature of the basin is the NAMOC. This channel wends its way for 3800 km through the Labrador Sea, around the Grand Banks of Newfoundland and terminates in the Sohm Abyssal Plain (Fig. 26A). Such long submarine channels are more typical of large submarine fan systems.

One of the first publications that described the channel appeared in Canadian Journal of Earth Sciences in 1969 (Heezen et al. 1969). As described above, elements of the channel are important considerations in identification of the base of the continental slope region. In compiling evidence for this analysis, it became apparent that there were few modern data that show details of the channel architecture. Canada’s submission relied heavily on scans of paper records dating from the early 1970s. There were few crossings of the channel with modern multibeam bathymetry, and there were no systematic surveys since the 1990s.

Partly as a result of this paucity of data, a program was carried out in 2021 to survey 1800 km of the northern part of the channel with modern multibeam bathymetric, subbottom profiler, and multi-channel seismic reflection systems (Krastel et al. 2021; Krastel and Mosher 2022). These new data show a channel that is remarkably consistent in its form over its entire length, at about 5 km in width and 100 m deep (Fig. 26). It was formed by turbidity currents related to glacial outwash during melting of the Laurentide ice sheet and possibly earlier North American deglacial episodes. The average gradient down-axis of the channel is just 0.05° (Fig. 26D). Suspended sediment spillover during turbidity current events generated large levee systems, the western levee being consistently higher than the eastern levee due to the influences of the Coriolis effect (Figs. 26E). It is this levee spillover and its truncation against the Canadian margin that helps identify a logical base of slope position in much of the Labrador Sea for ECS mapping purposes.

A significant amount of sediment infills the Labrador Sea basin; particularly given that the basin is relatively young (i.e., <92 Ma). This volume of sediment accounts, to some degree, for the fact that the basin is shallower than the isostatic position of a fully developed ocean basin. This infill is likely due to proximity of landmasses on its three sides. Glacial processes no doubt had a significant influence in this regard, as evident by shelf-crossing glacial troughs (Margold et al. 2015) with trough-mouth fans at their seaward extents. There is significant evidence too that the Labrador Sea is a highly dynamic environment for such a small basin, given the presence of the NAMOC, large contourite drifts, mass–transport deposits, and dissected slopes that are dominated by turbidite channels. This sediment thickness allows for construction of the OECM using the sediment thickness formula, as well as the distance formula (i.e., FOS + 60). Figure 27 shows that the OLCS lie landward of the 350 M constraint for much of the margin.

Prior to Canada’s ECS program in the Arctic, which commenced with active data acquisition in 2006, there were few geoscience data and very few in deep-water regions. For example, it is estimated that there were less than 5000 km of seismic reflection data within Canada Basin. Most data had been acquired on the periphery of the margin (Verhoef et al. 2011) where summer ice conditions permitted, or via drifting ice camps where the tracks were dictated by winds and currents. To achieve the data necessary to support an ECS submission, Canada and neighbouring countries embarked on aggressive data acquisition campaigns involving ice camps and icebreaker expeditions. For logistical and cost reasons, these programs involved a high degree of international collaboration.

Advances in knowledge of the Arctic Ocean geology over the past two decades as a result of these programs have been extensive and too much to review in this publication (see, for example, Piskarev et al. 2019 and Nikishin et al. 2021a, efg136b, efg137c). This paper focuses more specifically on scientific advancements that are in support of the Canadian ECS submission.

Canada delineated an area that is approximately 1.2 million km² as its ECS in the Arctic (Fig. 28) and submitted this delimitation with accompanying scientific and technical material to the Commission in 2019. Submissions have also been made by the other Arctic coastal States, including Russia, Denmark on behalf of Greenland, and Norway (Spitzbergen).

As mentioned, prior to ∼2006 there were few data in the Arctic and fewer still that were relevant to establishment of the ECS in the Arctic Ocean for any of the coastal States. Historical perennial sea ice in the Arctic Ocean, particularly thick along the Canadian margin, made conventional marine data acquisition extremely difficult to impossible. Work from drifting ice camps was the typical mode of operation until the beginning of the 21st century. The first non-nuclear icebreakers (and the first science research ships) to reach the North Pole was only in 1991 (e.g., Thiede et al. 1991). Given dramatic reductions in the extent and thickness of summer sea ice that has taken place since 2006, Canada employed a combination of ice camp and icebreaker operations. In total, between 2006 and 2016, Canada undertook five ice camp and eight icebreaker expeditions and participated in three Danish-led icebreaker missions and one shared aeromagnetic/gravity survey.

Canada and Denmark occupied an ice camp in 2006 to acquire seismic reflection, refraction, and bathymetry data across the geological continental shelf and on to LR (Fig. 29). A similar program was run in 2008 to Alpha Ridge (Fig. 29). In this region of the interface between the landmass and Alpha and Lomonosov Ridges, Canada and Denmark also collaborated on an aeromagnetic–gravity program in 2009, covering an area nearly 439 000 km² (Fig. 29) (Matzka et al. 2010). Canada conducted a modified seismic reflection program to Canada Basin, seaward of the Beaufort Shelf, in 2007 from its flagship icebreaker, CCGS Louis S. St-Laurent. This survey in 2007 demonstrated that, aside from the need for numerous modifications to the seismic system, a single icebreaker cannot reasonably acquire seismic reflection data and concurrently break ice. As a result, Canada and the United States collaborated on programs from 2008 to 2011 with each country providing an icebreaker to acquire bathymetric and seismic reflection and refraction data (Figs. 29–31) (Armstrong et al. 2012; Mosher et al. 2013). One icebreaker would break ice ahead of the other while the following vessel acquired data.

Canada further used two of its own icebreakers for acquisition in 2014 and 2015 and collaborated with a Swedish icebreaker in 2016. In addition to these shared programs, Canada struck formal data sharing agreements with the United States, Denmark, Germany, and Sweden to augment data holdings and the five Arctic coastal States held annual meetings to share information concerning their respective field programs and progress on their submissions.

In total, Canada and collaborators acquired in excess of 32 000 line-km of MCS reflection data (Shimeld et al. 2021) with 297 expendable sonobuoy and 391 on-ice seismometer deployments for wide-angle seismic reflection and refraction analysis (e.g., Chian and Lebedeva-Ivanova 2015) (Fig. 29). A total of 117 287 line-km of new multibeam bathymetric and subbottom profiler data were acquired from icebreaker expeditions (Fig. 30). In addition, a number of shallow sediment cores and dredge samples of bedrock were taken during these ECS expeditions. This record of data acquisition is no small achievement, given that areas seaward of the Canadian margin sustain 100% ice cover year round (Fig. 31).

To acquire data in ice-covered seas and freezing temperatures, a number of technological innovations were implemented. In large part, these innovations consisted of modifications to conventional technologies including increasing equipment tow depths to >11 m so that it remained below the base of sea ice. Expedition reports and two publications in particular capture details of these operations (see Armstrong et al. 2012 and Mosher et al. 2013, for example). One piece of technology, however, was developed specifically to permit data acquisition in areas of challenging ice conditions.

Canada’s ECS program envisaged a need to acquire multibeam bathymetric data in areas where even icebreakers could not travel due to heavy ice cover. An autonomous underwater vehicle (AUV) system was designed and built by International Submarine Engineering Ltd. of Vancouver (Crees et al. 2010) in collaboration with Defense Research Development Canada (DRDC of Department of National Defense) to operate beneath the ice. Requirements of the system were to (1) acquire multibeam bathymetric data, (2) operate under ice to 5000 m water depth, (3) operate from a drifting ice camp, (4) home-in to a dynamic position (as the ice camp drifts), (5) recharge and be re-missioned while still in the water, and (6) be modular to allow transportation in small aircraft. Two AUVs were built. The unique aspects of these vehicles in particular are points 4 and 5 above. Since an ice camp drifts, the home position could not be pre-programmed, so the AUV had to be able to echo-locate to a position that could not be known prior to launch. Furthermore, it was not possible to take the vehicle out of the water at a remote camp, so it had to be feasible to recharge and interrogate the system while in the water.

The AUV had a successful run from an ice camp to a remote ice camp in the Arctic in 2010, collecting single-beam bathymetric data (Fig. 32A). It spent 10 days under ice with a total of close to 1000 km of under-ice data acquisition (McFarlane and Murphy 2013). This was the first time an AUV was deployed and recovered on ice and operated under 100% ice cover. The problem, however, was that since 2007 the degradation of ice extent and loss of multiyear ice in the Arctic has meant that there are few stable platforms that could sustain a camp for human occupation for any period of time. As a result, in 2011, the AUV was deployed from an icebreaker and multibeam data were successfully acquired on two separate deployments under 100% ice cover (Mosher et al. 2015; Fig. 32).

The Arctic Ocean comprises two major basins, Amerasia and Eurasia basins, separated by the LR (Figs. 28 and 33) (Jakobsson et al. 2003, 2020). Eurasia Basin is surrounded by Europe and Asia (Greenland, Norway, and Russia). Its formation is relatively well understood. Seafloor spreading reached from the North Atlantic into the Arctic in the early Eocene (Funck et al. 2022) and separated a segment of the Barents Shelf from the Eurasia continent. This separation created the LR that remained part of the North American Plate, with a rift margin along its flank (see Fig. 34). Conjugate to LR therefore, are the margins of Barents and Kara seas of the Eurasia Plate. Gakkel Ridge formed as the spreading centre and Nansen and Amundsen basins formed as a result of creation of new oceanic crust via this seafloor spreading (Figs. 33 and 34). There is a clear pattern of magnetic reversals in the adjacent Nansen and Amundsen basins (Nikishin et al. 2018) that calibrate this scenario and provide some age control.

Amerasia Basin is surrounded by the continents of North America and Asia (United States and Canada, and Russia) (Fig. 33) and its genesis is poorly known, in contrast to Eurasia Basin. Lack of data prior to the programs mentioned herein made most theories for its creation purely speculative (see Lawver and Scotese 1990). Amerasia Basin is believed to have formed during the Mesozoic but there is no apparent spreading ridge, no well-established magnetic reversal pattern (Gaina et al. 2011) and no obvious conjugate margin pairs. A major subbasin of Amerasia Basin, known as Canada Basin (Figs. 33 and 34), is too shallow for a fully developed oceanic basin. Furthermore, emplacement of Alpha Ridge and Mendeleev Rise, commonly interpreted as an offshore component of a large igneous province (LIP; Tarduno 1998; Maher 2001), masks features that might provide evidence as to the basin’s origin.

The interconnected seafloor elevations of the central Arctic basin, including Alpha, Mendeleev, and Lomonosov ridges, are collectively referred to as the Central Arctic Plateau. For establishment of the OECM and the OLCS along the Central Arctic Plateau, Canada needed to know not only the morphology of these ridges that comprise the plateau but also their tectonic and geologic characteristics. As pointed out briefly above, the CLCS interprets three forms of seafloor highs relevant to Article 76 (see Brekke and Symonds 2011 for a detailed discussion on the matter of ridges): (1) oceanic ridges that are features of the deep ocean floor and therefore cannot be considered to be part of the continental margin, (2) submarine ridges that are morphologically adjoined to the landmass of the coastal State and therefore can be considered a “natural prolongation” of the continental margin, and (3) submarine elevations that are both a natural prolongation and a natural component of the continental margin. In the latter case, these ridges are morphologically and geologically connected based on tectonic, structural, and (or) geologic affinities with the adjacent land mass. The applicable constraint for submarine ridges is the 350 M distance constraint. The applicable constraint for submarine elevations is either the distance constraint or the depth constraint (2500 m isobath + 100 M), whichever is more seaward. It was important, therefore, to determine if Alpha Ridge, Mendeleev Rise, and LR are submarine elevations of the Canadian continental margin. If so, then Canada’s continental margin may extend across the entire Arctic Ocean basin to the East Siberian Shelf, constrained only by the 200 M limit of Russia.

Prior to the ECS programs, there were few data from the Central Arctic Plateau to make this assessment on the nature of these ridges. Magnetic and seismic data acquired from a Canadian ice camp over LR in the late 1970s (LOREX) reinforced theories on the continental nature of this ridge (e.g., Weber 1979; Sweeney et al. 1982). In 1991, the first icebreaker expedition to cross LR acquired a seismic reflection profile to show sedimentary layers that further illustrated the connection of the ridge with the Eurasia shelf (Jokat et al. 1995), and this sediment sequence was sampled in 2004 with a scientific drilling expedition (ACEX, see location on Fig. 29) (Backman et al. 2004).

Alpha Ridge and Mendeleev Rise are generally considered a single entity based on potential field data (e.g., Vogt and Avery 1974; Gaina et al. 2011). There is no direct evidence for a continuous structural element, however, and there is a bathymetric low between the two. There have been many hypotheses as to their origin, including (1) subsided continental crust (Eardley 1948; Saks et al. 1955; King and Zietz 1966), (2) rafted continental fragment (Sweeney et al. 1900), (3) extinct spreading centre (Vogt and Ostenso 1970), (4) extinct island arc/subduction zone (Herron et al. 1974; Scotese 2011), (5) trace of a hot spot/mantle plume (Vogt et al. 1979, 1984; Irving and Sweeney 1982; Morgan 1983; Maher 2001), (6) aseismic oceanic ridge or plateau (Vogt et al. 1979; Jackson et al. 1986), and (7) volcanism along a transform margin (Cochran et al. 2006).

A Russian seismic line acquired in 2000 across Mendeleev Rise provided information on the velocity structure of this feature (Lebedeva-Ivanova et al. 2006), arguing that it is continental in origin based on velocities and corresponding gravity solutions (density model). Bruvoll et al. (2012), based on seismic data acquired in 2005, compared the structure of Alpha Ridge and Mendeleev Rise to an LIP such as the Ontong Java Plateau, and inferred that it was clearly not oceanic in origin. Artyushkov (2010) argued that Mendeleev Rise, Alpha Ridge, and LR are elevations that did not subside to the degree that adjacent seafloor components did precisely because they are continental in origin.

Weber (1986), based largely on gravity, magnetic, and bathymetric data from the 1983 Canadian Expedition to Study the Alpha Ridge (CESAR), published that Alpha Ridge consisted of mafic rocks and is not structurally connected to the continent. While he drew some comparison of Alpha Ridge to Iceland-Faroe Ridge, he admittedly suggested that his model of Alpha Ridge is unique on the planet. He later added that, based on organic fragments found in a sediment core retrieved from the eastern Alpha Ridge during CESAR, Alpha Ridge was an archipelago of forested islands which formed above sea level about 95 Ma ago (Weber 1990). A few samples from Alpha Ridge were obtained during CESAR and included basalt and Cretaceous diatomaceous oozes (e.g., Van Wagoner and Robinson 1985; Aksu 1985; Mudie 1985). Dredged volcanic rock samples were interpreted to be erupted subaqueously although likely in shallow water and probably in an intraplate tectonic environment (Van Wagoner and Robinson 1985). Magnetic signatures suggested a volcanic origin but otherwise are inconclusive; gravity anomalies show a thick rock mass but provide little clue as to its origin and limited seismic data showed no internal structure of bedrock.

Alpha Ridge is a critical feature of the Canadian continental margin in terms of establishment of the ECS, yet, as indicated above, the ridge’s origins and nature remain enigmatic. A significant effort went into studying Alpha Ridge as part of Canada’s ECS program as a result, but the region continues to sustain ice conditions that make data acquisition difficult and slow.

Alpha Ridge as a distinct morphologic entity separate from the abyssal plains was first recognized from ice station Alpha in 1957–1958 (Weber 1983). Figure 35 illustrates the morphologic character of the ridge and its relationship to the Canadian landmass. It clearly shows that the ridge lies well above the deep ocean floor of the adjacent Canada Basin and is continuous with the land mass of Canada, so is considered a submerged prolongation of the land mass in the sense of Article 76.

Oakey and Saltus (2016) studied the potential field signatures of the Alpha–Mendeleev Ridge complex (AMR) utilizing regional compilations of magnetic and gravity anomalies for the Arctic basin (e.g., Fig. 36). Residual marine Bouguer gravity anomalies over AMR have low amplitudes implying that the structure has a deep crustal root (Moho lies at depth). Oakey and Saltus (2016) showed that high amplitude “chaotic” magnetic anomalies (termed the High Arctic Magnetic High Domain or HAMH) are associated with the AMR. A closed polygon around the HAMH identifies the extent of the ridge complex and shows that it is ∼1.3 × 10⁶ km² in area (Fig. 36C). This area extends beyond the bathymetric feature of AMR beneath the sediment cover of the adjacent Canada Basin, as shown in Figs. 36C and 37. In fact, Oakey and Saltus (2016) show that this characteristic signature extends on land, across the Queen Elizabeth Shelf, to a portion of Ellesmere Island as well (Figs. 36C and 37C).

The crustal structures of Mendeleev Rise and Alpha Ridge have been studied as separate features (e.g., Asudeh et al. 1988; Lebedeva-Ivanova et al. 2006; Funck et al. 2011; Poselov et al. 2012; Evangelatos et al. 2017; Butsenko et al. 2019). Figure 38 summarizes these analyses in a single figure to show a continuous structural feature with 30 km thick crust that spans the Arctic basin from the East Siberian Shelf of Russia to the Queen Elizabeth Shelf of Canada. There is some discrepancy in identification of upper crust versus lowermost sedimentary layers on velocity alone as these layers appear to be a mix of volcanic and volcaniclastic sediments, but the Moho is identified at depths between 26 and 32 km throughout the AMR. Crustal compressional and shear wave velocities suggest the rock types are likely continental or intermediate between continental and oceanic, but not basaltic as would be expected for a mafic LIP (Jackson and Chian 2019). This evidence is consistent with an interpretation that a continental landmass was intruded with volcanics to form the AMR.

AMR is similar to other LIP and was termed the High Arctic Large Igneous Province (HALIP) (Tarduno 1998; Funck et al. 2011; Oakey and Saltus 2016; Evangelatos et al. 2017). These authors suggested that Kerguelen Plateau is a close analog to AMR, as it too is considered a continental plateau that was intensively modified by plume-related volcanism.

In terms of tectonics and the origin of Alpha Ridge, Døssing et al. (2013) interpreted aeromagnetic data acquired in the joint Canada–Denmark expedition of 2009, and identified a landward Early Cretaceous (ca. 138–125 Ma) giant dyke swarm (minimum 350 × 800 km²) that permits reassembly of the continents prior to formation of Eurasia Basin. This reconstruction demonstrates that Franz Josef Land of the Barents Shelf, Alpha Ridge, and the Queen Elizabeth Islands were once adjoined. The swarm also points towards a 250 km wide donut-shaped anomaly on the southern Alpha Ridge, which Døssing et al. (2013) proposed was the centre of the HALIP mantle plume (Figs. 37C and 37D). They implied from this evidence that intrusive activity associated with this mantle plume caused continental break up in the northern Amerasia Basin, and is consistent with a rotational opening scenario for Amerasia Basin.

Samples from Alpha Ridge are historically few. A dredge sample from the CESAR expedition in 1983 provided first evidence of the composition of Alpha Ridge, comprising a highly altered fragmental volcanic rock (Van Wagoner and Robinson 1985). Geochemical discriminators suggest the rocks formed in an intraplate environment. Possible samples of bedrock were recovered in shallow piston cores during a Polarstern expedition in 1998 (Mühe and Jokat 1999; Jokat et al. 2013). In a 2008 joint United States/Canada icebreaker expedition, a dredge site on Nautilus Spur of Alpha Ridge yielded silicic volcaniclastic sedimentary rocks interpreted to have been deposited in shallow water during a phreatomagmatic eruption (Brumley 2009) (Fig. 39). In 2016, a second site on Nautilus Spur was dredged. The majority of these samples were described in the field as volcaniclastics, tuff, hyaloclastic basalt, and rhyolite (Fig. 39) (Mayer et al. 2016). Also in 2016, in a joint mission between Canada and Sweden, a dredge on Alpha Ridge near Makarov Basin recovered volcaniclastic rocks but also limestones (Fig. 39). The general consensus from these samples is that Alpha Ridge was emergent or shallowly submerged. The in situ samples consist predominantly of continental flood basalts derived from melting in the subcontinental lithospheric mantle and volcaniclastic sedimentary rock formed subaerially and shallow marine (e.g., Brumley 2009; Mukasa et al. 2020) which also accounts for the limestone.

A compilation of the ages for mafic magmatism/volcanism results indicates that the majority of the voluminous basaltic magmatism in the onshore Canadian continental margin was emplaced over 50 Myr, which is longer than most LIP, as noted by Dockman et al. (2018).

The distribution of age dates define three temporally distinct volcanic pulses:

• Phase-1: (130–120 Ma) consisted exclusively tholeiitic basalts.

• Phase-2: (106–90 Ma) included both tholeiitic and alkali basalts.

• Phase-3: (85–73 Ma) included alkali basalts and less frequent tholeiitic basalts.

An Ar–Ar age of 90.4 ± 0.26 Ma has been obtained for Alpha Ridge (Williamson et al. 2019) which is identical (within analytical error) to the Jokat et al. (2013) sample (89 ± 1 Ma). These results show that magmatism on Alpha Ridge occurred at the tail end of the phase-2 (106–90 Ma), coincident with episode of HALIP magmatism on Ellesmere and Axel Heiberg Islands. Naber et al. (2021) concluded that the deeper parts of the Alpha Ridge could easily include on-land equivalents, as suggested by geophysical signatures (Oakey and Saltus 2016).

From the insight gained through these studies, it would seem that Alpha Ridge and Mendeleev Rise were part of the continental margin along the Barents/Kara Sea margin prior to Eocene rifting and commencement of the formation of Eurasia Basin. Alpha Ridge formed as an LIP by hot spot volcanism that intruded into, underplated, and extruded on top of a pre-existing segment of continental crust. Whether or not this hot spot volcanism is related to break up and formation of the Amerasia Basin is unknown, but Døssing et al. (2013) and Evangelatos et al. (2017) contend that it is presumably related to transform and transtensional tectonic motion during the basin’s formation. From what is known from the few rock samples, it was emergent or only shallowly submerged and presumably remained that way until sufficiently rifted from Eurasia that it underwent a degree of subsidence.

Lomonosov Ridge (LR) was first recognized in 1954 by Soviet scientists, discovered from their drifting ice camps (Weber 1983). The origin of LR is better known and less contentious than that of the AMR (see Jokat et al. 1992). In fact, the only drill site in the entire high Arctic Ocean sampled the geology of the ridge (Backman et al. 2004). Figure 40 illustrates that LR is morphologically continuous with the landmass of Canada, lies well above the deep ocean floor of the adjacent Amundsen Basin and is geologically distinct from the Amundsen Basin.

To assess a continuous geologic prolongation from the Canadian landmass to LR (i.e., natural component), a seismic reflection and refraction program was run from an ice camp in 2006 (see Fig. 41A, labelled LORITA) (Jackson and Dahl-Hensen 2007), followed by three Canadian icebreaker expeditions that cross the ridge (e.g., Shimeld and Boggild 2017; Mosher et al. 2018). Jackson et al. (2010) found that the Moho depth varies from 25 km beneath the Queen Elizabeth Shelf to 20 km beneath a shallow trough just seaward and then more than 26 km thick beneath LR. They found continuous layer velocities that extend across the Canadian margin, and correlate with sedimentary basins on land including with the Mesozoic to Palaeozoic Sverdrup Basin. A layer with a velocity of 5.5–5.9 km/s underlies the inferred sedimentary strata on the continental shelf and traces seaward until it pinches out at a basement high on LR.

Morphologically, the northern flank of LR, facing Amundsen Basin, is rugged and appears to comprise normal-faulted rift blocks (Fig. 41B). This evidence is consistent with rifting of LR from Barents Shelf as seafloor spreading migrated from the Atlantic into the Arctic during the Eocene, as shown by the seafloor magnetic reversal pattern in Eurasia Basin (e.g., Brozena et al. 2003; Nikishin et al. 2018).

Funck et al. (2022), in a follow-up study, used newly acquired seismic reflection and refraction data to assess the crustal structure of LR and this rifted segment of its northern flank. They found the ridge consisted of two layers of continental crust and a Moho depth of 21 km (Fig. 41C). This crust is overlain by 1 km thick layer they associate with extrusive volcanics related to formation of the HALIP. Within Amundsen Basin, they interpret a 40 km wide transition zone of thinned continental crust followed by normal ocean crust (Fig. 41C). They also find evidence of magmatic underplating, again related to formation of the HALIP. Thus, while LR is clearly a component of continental crust rifted from the Barents margin, suggested by many lines of evidence, it also was affected by emplacement of the HALIP.

Rock samples from LR, recovered by dredge during ECS programs, include a mixed assemblage of lithologies of sedimentary and meta-sedimentary rocks (Fig. 42). These assemblages are identical to those of the Barents Shelf and support the continental nature of the ridge.

Canada Basin lies seaward of the Queen Elizabeth Shelf of the Canadian landmass in the Arctic (Fig. 29). While the general morphology of the basin was known prior to these ECS programs, mostly from spot soundings by helicopter, there were few continuous profiles of either bathymetric or seismic reflection data. To establish an ECS in this region, Canada needed to know much more about the morphology of the margin and of the thickness of sediments within the basin.

There are a diversity of tectonic models within the published literature to account for formation of Amerasia Basin (Lawver and Scotese 1990), but these models are largely without constraining evidence. One of the favoured concepts, known as the rotational model, interprets a linear low gravity anomaly through central Canada Basin as a paleospreading ridge and the basin opened in a counterclockwise fashion with the pole of rotation centered in the Mackenzie–Beaufort region (Grantz et al. 1979, 2011). One of the more significant advancements in support of this model was identification of oceanic crust with flanking transitional crust in central Canada Basin based on refraction velocity information collected for the ECS program (Chian et al. 2016). These basement characteristics flank the central Canada Basin gravity low and a possible axial valley as identified on seismic reflection profiles (Figs. 37 and 43).

Figure 43C was generated to visualize the basement surface without the effects of sediment loading. The sediment load was removed and basement elevation adjusted assuming Airy isostasy, following the specific “backstripping” methodology of Steckler and Watts (1978). Seismic data in the time domain were converted to depth using velocities of Shimeld et al. (2016). To calculate sediment load, sediment thickness was derived from the depth-converted seismic profile and sediment densities were derived from a generic curve for clastic sediments of Kominz et al. (2011). For the Airy isostasy correction, mantle density is assumed to be 3.4, grain density of the sediments is 2.65, and sea water is 1.013 g·cm⁻³. All adjustments were conducted within the SEGY seismic file. The “corrected” basement surface falls at an elevation that is expected for cold oceanic crust (i.e., 5000–5500 m below sea level) and the axial valley of the spreading center is clearly anomalously deep (∼1000 m) relative to the average rugosity of the basement surface. Other features, such as en echelon faults are clearly notable.

Further support for this rotational model is provided by studies of Makarov Basin and the Amerasia flank of LR. Evangelatos and Mosher (2016) interpreted the seismic sequence within a profile acquired across Makarov Basin that ties to the ACEX scientific drill site on LR, as well as surrounding morphological data. They argued that the steep flank of LR and an adjacent deep basin within Makarov Basin, as well as horsetail-splay fault structures in the region, are evidence of transform to transtensional tectonics. The basin formed as a pull-apart basin, therefore, which is an interpretation supported by Kazmin et al. (2016) and Nikishin et al. (2021b). Evangelatos and Mosher (2016) suggested that rotational opening of Amerasia Basin terminated at the flank of LR, where dextral strike-slip motion that became extensional, accommodated rotation (Fig. 44). Evangelatos et al. (2017) showed a rapid transition in crustal velocities and crustal thicknesses from Makarov Basin to LR, further supporting a transform margin argument. They interpreted a small area of oceanic crust in Makarov Basin, supported by the fact that it lies in full ocean depths, where surrounding terrain is above 3000 m water depth.

A problem remains with the rotational model in spite of this supporting evidence. In this model, Grantz et al. (2011) suggest ∼66° of counterclockwise rotation of Arctic Alaska away from the Canadian Arctic Archipelago. The amount of oceanic and transitional crust in Canada Basin and the dextral displacements along LR do not support such extension. Additionally, Grantz et al. (2011) invoke clockwise rotation of Chukchi Borderland to restore it to its present location. This restoration requires >1000 km offset along a left-lateral strike-slip fault, but there is no evidence to support such displacement (Coakley et al. 2016).

Close inspection of modern potential field data compilations, in combination with the velocity data discussed above and geologic and geomorphologic data acquired as part of Canadian and American programs indicates a northeast-trending structural fabric in Eurasia Basin and specifically Canada Basin (Hutchinson et al. 2017) (Fig. 36). These features are explained by initial Jurassic–Early Cretaceous strike-slip deformation (phase-1) followed in the Early Cretaceous (134–124 Ma) by rotational opening via seafloor spreading (phase-2) which completed the formation of oceanic crust and the configuration of crustal blocks as they exist today. In other words, a significant amount of displacement occurred as a result of strike-slip/transform deformation prior to actual seafloor spreading. This scenario requires the Alaska polar margin to be an extensional margin and the Canadian polar margin and the Northwind Ridge to initially be transform margins, followed by extension/rifting. Døssing et al. (2020) used 3D gravity inversion and vertical gravity gradient from satellite gravity models to identify oblique segmentation of the buried spreading ridge in a right-stepping en echelon pattern. This pattern may be represented in a series of short, apparent normal-offset faults in seismic profile (see Fig. 43C (inset)).

While tectonics is the principal control of margin morphologies, sediment stratigraphy, and sedimentology play a modifying role, as shown in the Atlantic case. In this way, they are important elements to consider in the establishment of the OLCS under Article 76. More directly, sediment thickness can play a critical role in establishment of the OECM.

The regional nature of seismic reflection, subbottom profiler, and multibeam bathymetric data acquired by the ECS field programs allowed for the first time an assessment of the stratigraphy and sedimentology of Canada and Makarov Basins and surrounding margins. Since there were few previous data in these regions, these new data have led to exciting discoveries. Figure 45, for example, shows a seamount newly discovered in 2009 within Canada Basin. By definition, a seamount must be at least 1000 m in elevation (IHO 2019), thus it is astounding that such a large feature was heretofore unknown.

Seismic reflection data are acquired in the time domain, so one of the first steps to map sediment thickness is to convert these data to the depth domain, which requires velocity information. ECS programs, therefore, utilized expendable sonobuoys to collect wide-angle reflection and refraction data concurrent with acquisition of seismic reflection data (see Fig. 29) (Mosher et al. 2013; Chian and Lebedeva-Ivanova 2015).

Shimeld et al. (2016) were able to show regional variability of the velocity–depth function in Canada Basin utilizing these data. They showed five distinct subregions: the Mackenzie Fan, the continental slopes beyond the Mackenzie Fan, the abyssal plain, southwestern Canada Basin, and the Alpha–Mendeleev magnetic domain. They interpret that these differences are largely related to proximal to distal, source-to-sink parameters. Lithological factors, however, do not fully account for the elevated velocity–depth trends associated with the HAMH domain. Shimeld et al. (2016) suggest accelerated porosity reduction due to elevated paleoheat flow may be responsible in this region. Evidence for such high heat flow is shown in a recent compilation of heat flow data by Rupple et al. (2019) and in a sediment core collected from atop the newly discovered seamount (Fig. 45).

The seismic stratigraphy of Canada Basin is shown in Figs. 37B and 46, with four main horizons mapped throughout the data set (Coakley et al. 2016; Mosher and Hutchinson 2019). The sedimentary succession is thickest in the Beaufort Sea region beneath the Mackenzie Fan, reaching more than 12 km. The entire sequence generally thins to the north and west, largely reflecting southward tilting basement (Fig. 46). The seafloor of much of Canada Basin is remarkably flat, with little bathymetric change over most of its extent. Reflections that correlate over tens to hundreds of kilometres comprise most of the succession and on-lap bathymetric and basement highs.

Stratigraphic correlations reflect changing sediment source directions during the basin’s history (Mosher and Hutchinson 2019). The lowest stratigraphic sequence is the R40 horizon to acoustic basement (Figs. 37B and 46). This interval is probably Late Cretaceous to Paleocene in age and is interpreted as a synrift sequence. Assuming thinning of the sequence in a distal direction from its source, then sediments likely derived from the Alaska and Mackenzie–Beaufort margins as a result of erosion from uplift of the Brooks Range during early onset of rifting and formation of the basin. This unit shows a series of onlap unconformities that progress northwards into the basin; perhaps in response to contemporaneous subsidence as ocean crust formed and cooled. Sediment source direction appears to have shifted to the Canadian Arctic Archipelago margin for the Eocene and Oligocene, as shown by the R10–R40 sequence (Figs. 37B and 46). This change in source direction was likely due to uplift of this region during the Eurekan Orogeny.

The final stage of sedimentation, represented by the seafloor to R10 interval, appears to be largely sourced from the Mackenzie–Beaufort region, possibly commencing in the Oligocene. Oligocene to Miocene was a period of convergence between the northeastern Brooks Range and Richardson Mountains of Alaska and the Yukon Territory and the southeastern Canada Basin. This convergence created the Beaufort fold belt beneath the eastern Beaufort Shelf and upper slopes of the Mackenzie Fan (Fig. 46, see features labelled “folds”). Regional uplift in the Yukon area resulted in a shift in river drainage patterns toward the east. Sediment shed from tectonic uplifts were funneled offshore via Mackenzie River to the Mackenzie delta and fan and out into Canada Basin. Today, the Mackenzie River is the 18th largest river in the World by volume of water discharge (Shiklomanov et al. 2021). This youngest stratigraphic sequence onlaps the Alaska margin, Northwind Ridge, and Alpha Ridge, so post-dates these features and any sediment derived from these features. Included in this shallowest sequence is sediment input to the basin from Pleistocene glaciations, as evidenced by glacigenic debris flows seaward of shelf-crossing troughs (e.g., Fig. 47C).

Mosher and Hutchinson (2019) showed the important role of submarine landslides, including these glacigenic debris flow deposits, in supplying sediment to the basin, both from the Mackenzie Fan and along the Canadian Arctic Archipelago margin (Figs. 46 and 47). Important for ECS purposes, however, is that these mass transport products assist in identification of the base of the continental slope; particularly when morphology alone is not sufficient (see Mosher et al. 2016a).

In addition to mass transport deposits along the margin, Mosher and Hutchinson (2019) interpreted that flat-lying reflections of the abyssal plain largely represent turbidites (see Fig. 37B). They conjectured that these turbidites were generated by mass failure of sediments along the margins that evolved into turbid density flows, and from suspended sediment injected to the water column as hyperpycnal flows. The region of termination of submarine mass transport deposits and their evolution into turbidites, as well as onlap of turbidites elsewhere along the margins of Canada Basin, helped to identify the base of the continental slope (e.g., Figs. 46 and 47).

This interpretation that sediments within Canada Basin principally comprise turbidites had prevailed since earliest sampling of the basin (e.g., Clark et al. 1980; Grantz et al. 1996). New multibeam bathymetric data, however, resolve seafloor features previously not observed, including bedforms indicative of along slope currents (contour currents) (Fig. 48A) (Mosher and Boggild 2021). Additionally, the fact that new high-resolution seismic and subbottom profiler data are digital make it possible to analyse these data at a variety of aspect ratios. Extreme vertical exaggeration illuminates mounded deposits along the margin, interpreted by Mosher and Boggild (2021) to be drift deposits indicative of bottom current activity (Fig. 48B).

There are significant volumes of sediment within Canada Basin, generally greater than 12 km thick beneath the Mackenzie Fan and thinning in a more-or-less concentric pattern away from this region (Fig. 49). This sediment distribution pattern is largely due to the vast amounts of sediment shed into Canada Basin from the Mackenzie River. Canada Basin represents a relatively small catchment area for these sediments, allowing this thick accumulation. These sediments appear to have been largely supplied within turbidity currents and the result is a remarkable flat seafloor within Canada Basin. Reflection sequences truncate against morphologic highs in most regions of the basin resulting in clear change in gradient and distinct geologic change from the slopes to the basin floor.

This volume of sediment within Canada Basin permits use of the sediment thickness formula to determine the OECM beyond the applicable constraints.

Makarov Basin lies north of Canada Basin and is isolated from it by Alpha Ridge (see Figs. 28, 30, 31, and 44). Stratigraphy within Makarov Basin is distinct from Canada Basin as a result (Evangelatos and Mosher 2016) (Figs. 44 and 50). The lowermost stratigraphic sequence of Makarov Basin is inferred to be derived from off-shelf sediment transport from the Mesozoic Barents Shelf prior to opening of Eurasia Basin (Unit 1, Fig. 50). Emplacement of the LIP that includes Alpha Ridge resulted in interbedded volcanic and volcaniclastic sediments within this unit. Unit 2 consists of base of slope sediments that include slump deposits. As LR rifted from the Barents Shelf during formation of the Eurasia Basin, Makarov Basin became increasingly isolated from sources of sediment, except from the Siberian Shelf via Podvodnikov Basin and some material eroded from LR. As LR subsided, sediments appear to be increasingly hemipelagic to pelagic (Unit 3), as underlying topography is draped. The infilling characteristic of Unit 4 suggests it comprises interbedded turbidites and hemipelagic/pelagic sediments (Fig. 50). Correlation to the ACEX drill site on LR shows that the 26.2 m.y. hiatus (44.4–18.2 Ma) in the drill core (Backman et al. 2008) does not appear to correlate to the basin; i.e., sedimentation appears to be continuous in the basin.

The seafloor of Makarov Basin is extremely flat and is in stark contrast to the surrounding elevations of Alpha and Lomonosov ridges (Figs. 44 and 50). As a result, the BOS for ECS purposes is readily determined where the steep morphology of the ridge flanks intersect the basin floor. The inset map of Fig. 50 shows a series of 60 M arcs (circles) generated from hypothetical FOS positions. In this case, the FOS + 60 formula is applied. This figure shows that, with overlapping arcs, the entire basin is enclosed and, therefore, is considered within the continental margin; i.e., no OECM is generated within Makarov Basin.

Amundsen Basin (once known as Fram Basin) lies between LR and Gakkel Ridge and spans the Eurasia Basin from North America to Russia. It is relatively isolated from subaerial landmasses and thus from sources of sediment. Only at the basin’s extreme southern limits does it have a direct connection with subaerial terrain via the Laptev Sea of Russia and the Lincoln Sea of Canada and Greenland (Fig. 28). The Russian side includes input from the Lena River; however, which has one of the World’s largest volume discharges (ranked sixth globally (Shiklomanov et al. 2021).

On the North American side of Amundsen Basin, the source of sediment input is less clear; although new data in this ice-covered region provide some insight. Kristoffersen et al. (2004), based on observations from only three seismic crossings, proposed the presence of a turbidity current channel and submarine fan seaward of the Canadian and Greenland margins. The channel, which they termed the NP-28 Channel, runs lengthwise along the basin toward the North Pole and parallels the lower flank of LR. It is the World’s northernmost deep-sea channel. They theorized that the source region for the channel and fan is Klenova Valley in the Lincoln Sea, seaward of Nares Strait (Fig. 51).

Newly acquired ECS data, including multibeam bathymetry, subbottom profiler and seismic reflection, permitted refined mapping of the channel (Boggild and Mosher 2021; Fig. 51) and investigation of the effects of the Coriolis force at this extreme latitude. A higher right-hand levee is one of the indications of the influence of Coriolis on sediment distribution (Fig. 51C). A nearby sediment core shows turbidite sequences (Fig. 51D) that are likely deposits of channel spillover (levee deposits) as turbidity currents swept down the channel. The area of this channel, and in particular its potential source region, hosts some of the most difficult ice conditions in the Arctic.

The first continuous seismic reflection profile acquired from a surface vessel across Amundsen Basin was in 1991 (Jokat et al. 1995) and few additional data were acquired between then and 2008. To establish better distribution of sediment thickness data, particularly closer to the Canadian and Greenland margins, there was a need for additional data acquisition and a number of Canadian/Danish collaborative programs were instituted as a result (as presented above).

Despite these efforts at seismic acquisition, data density was still too sparse to provide reliable sediment thickness information. Døssing et al. (2014) utilized the aero-gravity data acquired in the joint Canada/Denmark LOMGRAV-09 program (see Fig. 29) to generate a sediment thickness grid with 3D gravity inversion (Fig. 52), calibrated with the existing few seismic reflection lines (e.g., Fig. 41). Seismic reflection data acquired subsequent to this analysis is largely in agreement with these inversion results, although there are clear discrepancies close to LR as one might expect (Fig. 52). Long wavelength aero-gravity data are unlikely to resolve details along the steep flank of LR.

The flank of LR appears to be largely unsedimented and original tectonic rift elements such as rotated blocks are evident (Fig. 41). Little sediment in the basin appears to be sourced from LR although some slumping is apparent from the ridge to Amundsen Basin. The result is that there is a clear change in morphology and geology at the base of the ridge. Flat-lying sediments of the basin abut the flank of the ridge creating a significant gradient change. Despite there being no evident proximal source of sediment, there appears to be significant volumes of sediment within the basin (Fig. 52). This volume of sediment allows the use of either formula within Amundsen Basin to determine the OECM.

A major compilation of existing and new data was assembled by Canada in the Atlantic and Arctic oceans to provide scientific evidence for its proposed OLCS in these regions. These compilations provide the foundation for an exciting new wave of scientific understanding of Canada’s offshore lands. In the Atlantic, for example, this data compilation led to recognition that some morphological segments of the margin are related to original tectonic elements, while other segments are a function of the interplay of three primary sedimentary processes driven by turbidity currents, contour currents, and sediment mass failure. In particular, the prominent role of contour current deposits in shaping the margin was newly recognized. This element challenges the source-to-sink paradigm that underpins sequence stratigraphy concepts, since these deposits form by lateral transport of sediment. This paradigm is also challenged by recognition that the deep sea “sink” is far more dynamic than previously considered. For example, the NAMOC that wends its way for 3800 km through the Labrador Sea and around the Grand Banks provides a conduit for removal of sediment from the basins; thus, the ultimate “sink” is far removed from the source of sediment. As a result, the entrenched notion that a continental margin includes a continental rise that lies at the base of the continental slope and above the abyssal plain seems particularly difficult to justify when considering a continental margin with all these complexities. No such feature as a continental rise according to its conceptualization by Heezen et al. (1959) is distinguishable along the Canadian margin; as even recognized by Heezen et al. in this 1959 publication.

The Arctic is where the ECS program and acquisition of new data has had the greatest impact. However, before the Canadian ECS program, few geophysical or geological data existed in regions seaward of the Canadian landmass. New multibeam bathymetric, subbottom profiler, seismic reflection, and refraction data and new sample data have revolutionized our knowledge of the Arctic Ocean. Features such as a previously unknown seamount were discovered. Foremost amongst this knowledge is demonstration that Canada Basin is indeed a fully developed ocean basin, albeit significantly infilled with sediment. A buried relict seafloor spreading ridge paired with crust of oceanic velocities supports this understanding. Isostatic compensation after correction for sediment loads illustrates that basement is indeed at full ocean depth. Based on this knowledge and identification of related structures, new scenarios for opening of the Amerasia Basin are proposed that include significant transform tectonic displacement combined with extensive transitional (stretched continental) crust. Conventional seafloor spreading accounts for only a portion of the entire Canada Basin. Once thought to be relatively stagnant due to perennial ice cover, sedimentary processes such as found in many ocean basins were discovered in the Arctic Ocean. Evidence of geostrophic currents, sediment mass failures, and deep-sea turbidity current channels were found to be ubiquitous; even in the deepest parts of the basins. Major features of the Arctic Ocean include three prominent ridges that span the width of the basin. With seismic velocity and rock sample information, the continental nature of Alpha, Mendeleev, and Lomonosov ridges has been substantiated. This fact also is key to understanding the complex tectonic formation of the basin.

It will be many years before Canada’s submissions to the Commission will come to the front of the queue. Canada’s submissions are #70 for the Atlantic and #84 for the Arctic. The Commission is currently at submission #48, after 25 years of considerations. In the mean time, it is critical to keep the data that support Canada’s submission available and to keep the scientific knowledge that underpins the submissions alive. It is also imperative to continue to acquire additional data to support Canada’s submissions, particularly as technologies evolve, and to continue to answer scientific questions about our margin and the deep ocean seafloor, on which the submissions are based.

The implications of such massive compilations are many-fold. Despite limited data in deep-water regions of the the Atlantic and the Arctic, the compilations presented herein show the power of spatial and temporal integration of tectonic, crustal, stratigraphic and morphologic data and analysis. These compilations underscore the inter-connectedness of all earth processes and highlight the complexity of continental margins—even presumed simple passive margins. One of the paramount revelations of these compilations that applies to all continental margins, is that, despite many years of data acquisition in the Atlantic, for example, very little of deep-water regions have been mapped to any resolution. There are few bathymetric data in deep water, and fewer still seismic profiles and samples. Seabed 2030 estimates that only 25% of the World’s oceans depths have been measured. There are only 20 scientific drill sites in the study region of the Northwest Atlantic Ocean and only one in the Arctic. And yet, inherent in UNCLOS are requirements of coastal States to manage the regions for which they are sovereign. How can a State manage a region for which they have no or little information?

Many coastal States have undertaken in-depth studies of their margins to address the criteria prescribed in Article 76 of the UNCLOS. In most cases, coastal States feel obliged to keep such data confidential to prevent other States from utilizing these data or to prevent others from publishing arguments that may be counterproductive to their use in Article 76. Yet these data are critical to scientific understanding of the marine world. Release and scientific publishing, in fact, strengthens scientific arguments for their use in Article 76 and lend scientific credibility to these arguments. A few States have released their data and information and, with this publication, it is hoped that other States will be encouraged to do so as well.

The authorship on this paper reflects the individuals who made a significant contribution to the science that is reflected in this publication. A project of this scale involves many additional people and organizations to thank—too many to name individually. Of particular note, however, the authors want to express their deep appreciation for the leadership of Dr. Jacob Verhoef, Mr. Richard MacDougal, Mr. Steve Forbes, and Mr. Louis Simard. These individuals provided the vision and spearheaded the initiation of Canada’s ECS program and led the program through its initial phases that included submission of the Atlantic component and much of the Arctic data acquisition. Additional staff from organizations involved in data acquisition and technical aspects of submission preparation include Global Affairs Canada, Natural Resources Canada, Canadian Hydrographic Service, Canadian Coast Guard, Environment Canada, and Department of National Defense (DRDC). International partnerships include formal collaborations with the United States of America, Denmark, Sweden, and Germany. These colleagues have been stellar advocates for science regardless of borders. Finally, we would like to thank the reviewers, M. Duchesne, H. Brekke, and J. Hernandez-Molina for the helpful comments and corrections. NRCan contribution number/Numéro de contribution de RNCan: 20220146.

Data used in this study are publicly available and are cited in the text. Aside from the cited expedition reports, navigation and other information for all Natural Resources Canada expeditions are available in the Natural Resources Canada expedition database; https://ed.marine-geo.canada.ca/.

• Seismic reflection data are available through the Canadian National Marine Seismic Data Repository in the Open Canada web portal; https://search.open.canada.ca/.

• Legacy (scanned) seismic data are available for ftp download from https://ftp.maps.canada.ca/pub/nrcan_rncan/raster/marine_geoscience/.

• Seismic reflection data acquired specifically for Canada’s ECS program are available in GSC Open File Reports 8850, (Shimeld et al. 2021, doi.org/10.4095/329248) and 7938 (Mosher et al. 2016b, doi.org/10.4095/297590). Arctic refraction results are available in Chian and Lebedeva-Ivanova 2015, Geological Survey of Canada, Open File 7661, 2015, 55 pages, doi.org/10.4095/295857. Open File reports are available through the NRCan GEOSCAN web portal: https://geoscan.nrcan.gc.ca/.

Hydrographic (bathymetric) data are available through:

• CHS web portal; https://www.charts.gc.ca/data-gestion/resourcemaps-cartesressources-eng.html.

• National Centers for Environmental Information; https://www.ngdc.noaa.gov/mgg/bathymetry/relief.html.

• Center for Coastal and Ocean Mapping at the University of New Hampshire;http://ccom.unh.edu/theme/law-sea.

• Global bathymetric grids are available through the GEBCO portal; www.gebco.net.

David C. Mosher: conceptualization, formal analysis, funding acquisition, investigation, methodology, project administration, software, validation, visualization, writing – original draft, writing – review & editing; Mary-Lynn Dickson: funding acquisition, investigation, project administration, supervision; John Shimeld: data curation, formal analysis, investigation, methodology, software, writing – review & editing; H. Ruth Jackson: formal analysis, investigation, methodology; Gordon N. Oakey: data curation, formal analysis, investigation, validation, visualization; Kai Boggild: data curation, formal analysis, investigation, software, validation, writing – review & editing; D. Calvin Campbell: formal analysis, investigation, visualization, writing – review & editing; Paola Travaglini: data curation, formal analysis, investigation, methodology; Walta-Anne Rainey: data curation, investigation, visualization; Alain Murphy: formal analysis, validation, visualization, writing – review & editing; Sonya Dehler: formal analysis, investigation, project administration; John Ells: data curation, formal analysis, investigation, methodology.

All funding for this work was provided by the Federal Government of Canada, principally through the UNCOS program provided to the three responsible departments of Global Affairs Canada, Natural Resources Canada, and Fisheries and Oceans Canada.