Cusk (Brosme brosme) COSEWIC assessment and status report 2012: chapter 7

• Previous page
• Table of contents
• Next page

Distribution

Global Range

Cusk are a deepwater demersal fish with a depth distribution that varies within its North Atlantic distribution area, but generally ranges from coastal waters to 1000 m on the upper continental slope, on mid-ocean ridges and in deep fjords (Knutsen et al. 2009) (Figure 4.).

Figure 4. Global distribution of Cusk; relative probability of occurrence is based upon combination of modelled predictions and expert opinion using a variety of data sources (from Harris and Hanke, 2010; see AquaMaps for details).

In the northwest Atlantic, the species is distributed along the continental shelf from New Jersey to the Strait of Belle Isle, on the Grand Banks of Newfoundland, and off West Greenland. In the northeast Atlantic, it is found off East Greenland, around Iceland and the Faroe Islands and along the European shelf from southern Ireland to the Kola Peninsula and Spitzbergen, including the deeper parts of the North Sea and Barents Sea. Along the Mid-Atlantic Ridge, Cusk occurs south to the Charlie-Gibbs Fracture Zone (Knutsen et al. 2009).

Canadian Range

An analysis of Canadian and US East Coast of North America trawl survey data collected during 1975 – 1995 indicated that Cuskbelongs to the south-temperate bank/slope assemblage of fishes which ranges from the mid-Atlantic Bight to the Grand Banks off Newfoundland with the centre of distribution being the relatively deep waters of the Gulf of Maine area (Figure 3; Brown et al.1996, Mahon et al. 1998). This is a region encompassing about three degrees of latitude extending from approximately 41° to 44° N and covers NAFOdivisions 4V, 4W, 4X, 5Y and 5Z (Figure 5). Occurrence of Cuskto the north and south of its core distribution is largely confined to the edge and slope of the continental shelf in relatively deep water. As will be seen in the section on Search Effort, Cusk have not been observed in the Arctic and are rare in the Gulf of St. Lawrence and off Newfoundland. It is evident that Cusk,relative to other members of the Gadidae (e.g. cods, hakes) in the northwestern Atlantic, maintains a very restricted core distribution.

Figure 5. Statistical divisions of the Northwest Atlantic Fisheries Organization (NAFO).

The longest time series of Canadian distributional information is the DFO summer bottom trawl survey on the Scotian Shelf and Bay of Fundy, which has been conducted every July – August since 1970. Concerns have been raised (DFO 2004) that the survey results are not representative of the Cusk population. Other surveys (e.g. Halibut industry) are suggested as being more appropriate (DFO 2008). However, the latter is a relatively short time series with a restricted spatial coverage. It is considered that broad indication on long-term distributional changes can be inferred from the DFO summer bottom trawl survey. However, the results may overestimate the extent of distributional change.

During 1970 – 85, while Cusk ranged the extent of the Scotian Shelf, covering about 172,500 km², they were predominantly located off Southwest Nova Scotia (Figure 6). There were also observations along the edge of the continental shelf. By 1995 – 2010, Cusk catches were sparser and more restricted to waters off SW Nova Scotia and in the Gulf of Maine. A similar pattern is evident in the US NMFS fall bottom trawl surveys. During the 1960s-70s, Cusk were abundant throughout the Gulf of Maine while during the 1990s – 2000s, they had become sparser in generally the same area (Figure 7). Cusk are not found on Georges Bank but rather in the Gulf of Maine. As Cusk prefer rocky bottom habitat, it is not surprising that very few are observed on the predominantly sandy bottom Georges Bank.

Figure 6. Comparison of Cusk distribution during 1970 – 85 (top panel) to that during 1995 – 2010 (bottom panel) as observed by DFO summer bottom trawl survey; scale is in thousands of individuals.

Figure 7. Comparison of Cusk distribution during 1963 – 79 (top panel) to that during 1990 – 2010 (bottom panel) as observed by NMFS fall bottom trawl survey; scale is in kg per tow.

Based on the recent DFO summer bottom trawl survey, the COSEWIC extent of occurrence (EO) is estimated to be 302,311 km². The COSEWIC index of the area of occupancy (IAO) was estimated as 3,616 km² (based on all life stages) from both the DFO summer and NMFS fall bottom trawl surveys (Wu 2011). Both the EO and IAO have declined in the past 40 years.

DFO Science routinely estimates a design-weighted area index (DWAI) that represents the area occupied by a species (Smedbol et al.. 2002). It is also expressed as a proportion of the total survey area (NAFODiv. 4VWX). For Cusk, the area and proportion of area occupied have declined considerably from about 40,000 km² in the mid-1970s to about 5,000 km² more recently (Figure 8).

Figure 8. Trend in DFO summer bottom trawl survey area occupied by Cusk; both total (km²) and proportional area are indicated.

As noted above, the bottom trawl surveys are not considered to representatively sample the Cusk population (DFO 2008). Trends described by the DWAI have been called into question elsewhere (e.g. 4X Cod in Worcester et al.. 2008). Also, DFO (2008) indicates that since 1998, there has been no trend in the percentage of Halibut Industry survey stations that have caught Cusk and, since 1991, there has been no change in the range of Cusk in the groundfish longline fishery in 4X (Figure 9). DFO (2008) considers that there has been no reduction in the range of Cusk in Canadian waters.

Figure 9. Percentage of commercial groundfish longline trips in 4Xnopq where Cusk were caught (prevalence – commercial fishery), percentage of Halibut industry survey stations which were sampled in all years where Cusk were caught (prevalence – Halibut industry) and percentage of 5 x 5 minute geographic blocks with groundfish longline effort in 4Xnopq where Cusk were caught (range – commercial fishery) (from DFO, 2008).

The magnitude of decline in the DWAI may overestimate the true change in Cusk distribution because the trawl survey is restricted to relatively smooth and sandy bottom types. On the other hand, the commercial fishery and the Halibut survey may be more restricted to the preferred habitat of the species that have remained occupied despite reductions in overall abundance.

Search Effort

A number of surveys have been conducted in the core distributional area (Scotian Shelf – Gulf of Maine) of Cusk. These consist of DFO Maritimes Science spring, summer, fall and winter surveys, NMFS spring, summer, fall and winter surveys and a suite of industry surveys. Elsewhere in the Canadian zone, a variety of bottom trawl surveys have been conducted with similar sampling designs to that of the Scotian Shelf and US surveys (Doubleday 1981). As well, a number of DFO – industry collaborative surveys have also been conducted outside the Cusk core distributional area (Gillis 2002).

Observations of Cusk in all these survey activities have been rare. In the southern Gulf of St. Lawrence, 40 September bottom trawl surveys (1971-2010), six seasonal surveys (1987-1991) and four January surveys (Cabot Strait, 1994-1997) found no Cusk (Morin 2011). In the northern Gulf, there have been no observations of Cusk in the groundfish and shrimp multidisciplinary survey conducted every August since 1984, only two occurrences of Cusk in the RV Gadus Atlantica survey conducted in January during 1978 – 1994 and two occurrences in the October sentinel surveys conducted since 1995 (Bourdages et al. 2010, Bourdages 2011). There have been more observations off Newfoundland but even there, Cusk are rare. Of the 95,681 trawl sets in the DFO Newfoundland Science database, which includes all surveys for all seasons and programs since 1947, a total of 254 sets encountered Cusk, most of these having 1 – 2 individuals (Power 2011). There are no reports of Cusk from survey activities in DFO’s Central and Arctic Region (Martin 2011, Ratynski 2011, Treble 2011).

The longest time series of consistent observations of Cusk is the summer bottom trawl survey of the Scotian Shelf and Bay of Fundy. It has been conducted every July – August since 1970 using a stratified – random design with standardized data collection and processing protocols (Halliday and Kohler 1971, Doubleday 1981). The survey samples about 50 strata on the Scotian Shelf with an average of 173 trawl sets. Spring and fall bottom trawl surveys covering NAFO Divisions 4VWX5Z in a stratified random design were conducted during 1978/79 – 1984. The Georges Bank February bottom trawl survey has been conducted annually since 1986. Its coverage is restricted to NAFO Subarea 5 with a focus of 5Z. Cusk observations have been limited (Table 1).

The NMFSspring and fall bottom trawl surveys are conducted using a stratified random design with standardized protocol (Grosslein 1974), sampling about 76 strata in the Gulf of Maine area (Table 2). The proportion of sets in the fall survey on which Cusk have been observed has declined over the length of the time series (Figure 10).

Figure 10. Percent of NMFS fall bottom trawl survey sets that caught Cusk.

A number of DFOdeepwater surveys have also been conducted on the Scotian Slope. Observations of Cusk in these surveys have been rare. Ten were caught in a deepwater survey conducted in 360 – 600m depth of water off the Scotian Shelf in 1984 (Markle et al. 1988). None were caught in a 1994 trawl survey of 900 – 1800m depth water, a 1991 fixed gear survey of 500 – 2800m depth water as well as a number of mesopelagic surveys conducted in the 1980s (Halliday 2011).

Since 1994, a number of DFO – fishing industry collaborative surveys have been conducted in the Scotian Shelf – Gulf of Maine area (Table 3). Of these, the only surveys that have caught Cusk to any degree are the 4VsW sentinel and Atlantic Halibut surveys. Both of these surveys use longline gear and have been relatively more successful than the DFO and NMFS bottom trawl surveys in catching Cusk. The protocol of the Halibut survey is discussed by Trzcinski et al.. (2011) while that of the 4VsW sentinel survey is discussed in O’Boyle et al.. (1995). Sampling intensity in these surveys has however been more limited than the DFO surveys, both in station number (Table 4) and spatial extent (Figure 11). For instance, while the 4VsW sentinel survey covered an extensive part of the Eastern Scotian Shelf prior to 2004 (using the same stratified – random design as the DFO summer bottom trawl survey), station intensity was reduced from about 200 to 53 stations in 2004. Sampling is now concentrated in the Emerald / Western Bank area and along the coast of Nova Scotia both marginal to the overall Cusk distribution. Station coverage by the Halibut survey has been more consistent and is focused on specific fixed stations (Figure 11b). Note that only the most consistently sampled fixed stations (which are in 4VWX) are displayed as these are used to develop the abundance index. Harris and Hanke (2010) consider that a significant feature of the Halibut survey is that it fishes in deepwater areas along the Shelf Break that are not well sampled by the DFO summer trawl survey (see comparison of station locations of two surveys in Figure 11b). Harris and Hanke (2010) indicate that the majority of the Halibut survey Cusk catches are in the deep water along the edge of the Scotian Shelf.

Figure 11. Station distribution of 4VsW sentinel longline survey (a: top panel; + indicates pre-2004 stations and circle indicates post-2003 stations) and Atlantic Halibut industry longline survey (b: bottom panel; open circles indicate Halibut survey sets and + indicates DFO summer survey sets for 2000 – 2010).