Seafood Watch Seafood Report

Seafood Watch Seafood Report Haddock Melanogrammus aeglefinus B. Guild Gillespie/www.chartingnature.com Atlantic Canada June 17, 2011 Julia K. Baum Jennifer Ford Wendy Norden

About SeaChoice, Seafood Watch and the Seafood Reports This report is a joint product of SeaChoice and the Monterey Bay Aquarium Seafood Watch program. Both organizations evaluate the ecological sustainability of wild-caught and farmed seafood commonly found in the United States marketplace. In doing so, SeaChoice applies the definition of sustainable seafood and the method for its evaluation and presentation developed by the Seafood Watch program at the Monterey Bay Aquarium. Seafood Watch defines sustainable seafood as originating from species, whether wild-caught or farmed that can maintain or increase production into the long-term without jeopardizing the structure or function of affected ecosystems. SeaChoice is a comprehensive seafood markets program with the primary goal of realizing sustainable fisheries in Canada and abroad. SeaChoice uses the best available science, strategic communications and partnerships to mobilize sustainable seafood markets via six main programs: (1) research, (2) industry outreach, (3) public education, (4) retail partnerships, (5) strategic communications, and (6) dialogue with government. More information on SeaChoice can be obtained at www.seachoice.org Seafood Watch makes its science-based recommendations available to the public on our website (www.seafoodwatch.org), print materials, an iphone app and other media. The program s goals are to raise awareness of important ocean conservation issues and empower seafood consumers and businesses to make choices for healthy oceans. Each sustainability recommendation is supported by a Seafood Report. Each report synthesizes and analyzes the most science on a species, then evaluates this information against the program s conservation ethic to arrive at a recommendation of Best Choices, Good Alternatives or Avoid. The detailed evaluation methodology is available upon request. In producing the Seafood Reports, Seafood Watch seeks out research published in academic, peer-reviewed journals whenever possible. Other sources of information include government technical publications, fishery management plans and supporting documents, and other scientific reviews of ecological sustainability. Seafood Watch Research Analysts also communicate regularly with ecologists, fisheries and aquaculture scientists, and members of industry and conservation organizations when evaluating fisheries and aquaculture practices. Capture fisheries and aquaculture practices are highly dynamic; as the scientific information on each species changes, Seafood Watch s sustainability recommendations and the underlying Seafood Reports will be updated to reflect these changes. Parties interested in capture fisheries, aquaculture practices and the sustainability of ocean ecosystems are welcome to use these seafood reports upon receiving permission and instructions for appropriate crediting from the Monterey Bay Aquarium and SeaChoice. For additional information about Seafood Watch, visit www.seafoodwatch.org or call 1-877-229-9990. For more information about SeaChoice, please contact the SeaChoice program via e-mail and telephone information available at www.seachoice.org. Disclaimer SeaChoice and Seafood Watch strive to have all seafood reports reviewed for accuracy by external scientists with expertise in ecology, fishery science and aquaculture. Scientific review, however, does not constitute an endorsement on the part of the reviewing scientists of SeaChoice or the SeaChoice program, or the Seafood Watch program or their recommendations. SeaChoice and Seafood Watch are solely responsible for the conclusions reached in this report. SeaChoice, Seafood Watch and the seafood reports are made possible through grants from the David and Lucile Packard Foundation. 2

Table of Contents I. Executive Summary... 4 II. Introduction... 6 III. Analysis of Seafood Watch Sustainability Criteria for Wild caught Species... 12 Criterion 1: Inherent Vulnerability to Fishing Pressure... 12 Criterion 2: Status of Wild Stocks... 13 Criterion 3: Nature and Extent of Bycatch... 19 Criterion 4: Effect of Fishing Practices on Habitats and Ecosystems... 25 Criterion 5: Effectiveness of the Management Regime... 27 IV. Overall Evaluation and Seafood Recommendation... 32 V. References... 34 3

I. Executive Summary Haddock is a commercially important groundfish species of the North Atlantic Ocean. It is fairly resilient to fishing pressure as it matures quickly and is highly fecund with occasionally very strong recruitment. In Canada, three haddock stocks are currently managed: (1) eastern Scotian Shelf/southern Gulf of St. Lawrence, (2) southern Scotian Shelf/Bay of Fundy, and (3) Georges Bank (co-managed with the US). All three stocks were overfished in the 1990s, but the fishery in the former region has been closed since 1993, and fishing mortality on the latter two stocks has declined substantially. Haddock abundance has increased in the past decade, but the populations are now composed mainly of small, slow growing individuals. Productivity has been impaired in some regions. Haddock are primarily commercially fished using demersal trawls and longlines (although a minimal amount of handline and gillnet fishing does occur) and both gear types are associated with bycatch of several species, including species at risk such cod and cusk, which are assessed as Special Concern and Threatened, respectively, by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC). There is significant concern that historic bycatch of cod in the haddock fishery may have impaired rebuilding of cod populations. Today, cod bycatch is managed through a TAC on cod and/or limits to the haddock fishery that should allow objectives for cod re-building to be met. Bottom trawling can significantly damage benthic marine habitats and several areas in Atlantic Canada and New England have been closed to trawling to reduce this impact, but habitat destruction remains an issue and has had some historic impacts. Demersal longlines can also have negative impacts on benthic habitats, but to a lesser extent in some areas than bottom trawling. The Department of Fisheries and Oceans (DFO) monitors haddock abundance in longterm research surveys and undertakes regular population assessments. The management regime, however, has not prevented trawl-induced habitat damage or significant historic declines in haddock stocks. Until the stocks have fully recovered and there is a plan implemented to mitigate trawl damage, management can be considered only moderately effective for trawl gear, but highly effective for longline fishing due to the lesser degree of habitat damage when fishing with longlines. In Canada, haddock captured using trawl and demersal longline is considered a Good Alternative. The Canada Scotia-Fundy haddock fishery (Southern Scotian Shelf, Bay of Fundy, Gulf of Maine, and Georges Bank) has been certified as sustainable to the Marine Stewardship Council (MSC) standard. The MSC is an independent non-profit organization, which has developed an environmental standard for sustainable and well-managed fisheries. It uses a product label to reward environmentally responsible fishery management and practices (http://www.msc.org/). 4

Justification for Reassessment This report is an addendum to Stevens (2004) Northeastern US haddock assessment. Its purpose is to provide information on Atlantic Canada s haddock stocks, and a joint Sea Choice/Seafood Watch assessment of these stocks. Table of Sustainability Ranks Conservation Concern Sustainability Criteria Low Moderate High Critical Inherent Vulnerability Status of Stocks Nature of Bycatch (Trawl, Longline) Habitat & Ecosystem Effects (Longline) (Trawl) Management Effectiveness (Longline) (Trawl) About the Overall Seafood Recommendation: A seafood product is ranked Best Choice if three or more criteria are of Low Conservation Concern (green) and the remaining criteria are not of High or Critical Conservation Concern. A seafood product is ranked Good Alternative if the five criteria average to yellow (Moderate Conservation Concern) OR if the Status of Stocks and Management Effectiveness criteria are both of Moderate Conservation Concern. A seafood product is ranked Avoid if two or more criteria are of High Conservation Concern (red) OR if one or more criteria are of Critical Conservation Concern (black) in the table above. Overall Seafood Recommendation: Best Choice Good Alternative Avoid 5

II. Introduction Biology Haddock is a demersal (bottom-dwelling) species in the family Gadidae (subfamily Gadinae), the true cods (Nelson 2006). Like many other gadoids (cod, hake, pollock, whiting), it is a commercially important groundfish species. Haddock is distributed on both sides of the North Atlantic Ocean, and in the Northwest Atlantic ranges from Labrador to Cape Charles and Virginia (Brodziak 2005). In Canada, there are several haddock stocks including one in the southern Gulf of St. Lawrence and eastern Scotian Shelf, a major one on the eastern Scotian Shelf and Bay of Fundy, and a major transboundary stock on Georges Bank, which is managed jointly with the United States. Although some haddock stocks undertake seasonal migrations within their areas, there is thought to be relatively little exchange of individuals among the different stocks (Begg 1998; DFO 2005). Haddock are usually found in 45 135m of water and in bottom temperatures between 2 and 10 C (Klein-MacPhee 2002). Adult haddock range in size from about 30 cm to up to 1 m, and most commercially caught haddock weigh from 1 to 3 kg (Brodziak 2005). Haddock are highly fecund, and adult female egg production increases substantially with age, from a few hundred thousand eggs each year in the youngest spawners and up to three million eggs in the oldest spawners (FRCC 2004, Brodziak 2005). Haddock have a broad diet that varies by locality. Among other things, they consume echinoderms (mainly brittle stars), crustaceans, polychaetes, bivalve and gastropod mollusks, and fishes (mostly herring eggs) (Klein-MacPhee 2002). In turn, juvenile haddock are preyed upon by several fish species including other gadoids, and adult haddock are prey for gray seals (Klein- MacPhee 2002). Fishery Atlantic Canada and New England s groundfish fishery has existed since colonial times and focused originally on cod. Fishing methods have changed substantially over time, from hook and line on small sailing dories, to sailboats, to steamships, and finally in the early 1900s to the use of otter trawls and on-board refrigeration (Klein-MacPhee 2002; NRC 1998 cited in Stevens 2004). After World War I, fishers began to target a wider variety of species, including haddock. Groundfish landings decreased during World War II and then increased in the 1960s, largely due to effort by distant-water factory trawlers from Russia and Spain (Stevens 2004). After the creation of a 200-mile Exclusive Economic Zone (EEZ) along Canada's Atlantic coastline in 1977, the domestic fishing fleet grew substantially. Excessive fishing pressure, first by international fleets, then by the domestic fleet, led to severe overfishing of many groundfish stocks, including haddock. Over the past decade, haddock populations have begun to recover and are currently not considered overfished. In Canada, haddock are managed by the Department of Fisheries and Oceans. There are five haddock fishing areas in Atlantic Canada (DFO 2002a), but only two are currently fished: (1) the Southern Scotian Shelf and Bay of Fundy (Stock 4X5Y) and (2) the 6

Eastern Georges Bank (Stock 5Zjm) (Figure 1 and 2). Currently, haddock fishing is permitted only in these two areas to protect poor haddock and/or cod stocks in other regions that have experience declines in productivity. Some commercial haddock fishing is restricted solely by the rebuilding objectives for cod (TRAC 2010). This report will only focus on the stocks that are currently fished (4X5Y and 5Zjm). Haddock is harvested in Canada inshore, nearshore, and offshore by otter trawls, traps, long-lines, and gillnets. Most of Canada s haddock stock is landed in Nova Scotia (DFO 2002a). In 2002, 1,006 Canadian vessels were licensed to harvest haddock and total landings were valued at $27.6 million (DFO 2002a). By 2006, the Maritimes region groundfish fishery included close to 700 active fishers along Nova Scotia s southern shores and in the Bay of Fundy (DFO 2006a). The groundfish fishery targets haddock, cod, pollock, redfish and flounder, and catches a number of other bycatch species (DFO 2006a). Fishers operate throughout the year with various fleets, adjusting the timing of the fishery based on local abundance, weather, market conditions, and the participation of license holders in other fisheries such as lobster (DFO 2006a). The Georges Bank stock is a transboundary stock that is co-managed with the US. In the US, haddock is managed by the National Marine Fisheries Service (NMFS) and regulated under the New England Multispecies Groundfish Fishery Management Plan (FMP) (Stevens 2004). Figure 1. Map of haddock management areas on the southern Scotian Shelf (4X) and Bay of Fundy (5Y) (left panel), and southern Gulf of St. Lawrence (4R) and eastern Scotian Shelf (4VW) (right panel) based on North American Fisheries Organization (NAFO) fishing divisions. Also included in the right panel are the major banks and the closed haddock area. Southern Scotian Shelf & Bay of Fundy Haddock, Stock 4X5Y The southern Scotian Shelf and Bay of Fundy groundfish fishery (Stock 4X5Y), which takes haddock among other species, has typically been dominated by mobile gear (trawlers), except during 1990 1993 when the proportion of landings taken by fixed gear (longline and handline) was greater (DFO 2010a). Reported annual haddock landings in this region since 1960 have been variable, peaking at 43,000 t and with a long-term average of about 18,000 t (DFO 2006b, DFO 2010a; Figure 4). Quotas for haddock were 7

introduced in this area in 1970 and landings increased substantially until the early 1980s (FRCC 2004; Figure 3). The haddock TAC has been low since around 1990 and over the past decade has averaged 7,000 t (DFO 2010a). Figure 2. Map of Georges Bank (5Z) haddock management areas based on NAFO subdivisions, and showing the closed area and Canada US international boundary etc. Source: Van Eeckhaute and Brodziak 2005. In 2005, total haddock landings of 5,946 t on the southern Scotian shelf and Bay of Fundy were well below the 8,000 t TAC, and the trawl fleet reported difficulty locating concentrations of large haddock in the Bay of Fundy (DFO 2006ba). Recently, the proportion of catches from the 4Xn and 4Xp region has been increasing due to larger haddock in deeper waters (DFO 2010a; Figure 1). However, low fish prices and high fuel, insurance and bait costs have contributed to the shortfalls. Industry has expressed that these factors, combined with low catch rates and small fish size, have left little economic incentive to fish haddock on portions (excluding 4X) of the Scotian Shelf (DFO 2005a). The total allowable catch for stock 4X5Y has been set at 7,000 tons since 2006 (DFO 2006a; DFO 2010a). In 2010, the total allowable catch was reduced to 6,000 mt (Anonymous 2010). The population of fish in 4X5Y is primarily small fish that are less than 43 centimeters in length. 8

Figure 3. Haddock landings and total allowable catch (TAC) on the southern Scotian Shelf and Bay of Fundy (4X5Y). Source: DFO 2010a. The distribution of the Scotian Shelf groundfish fishery has also changed over time, and fishing is now more concentrated in the Gulf of Maine and Bay of Fundy (DFO 2006b). Since 2000, the proportion of haddock landed from January to March has increased (much of it from 4Xn), reportedly because it is possible to fish for haddock with a minimal bycatch of cod (DFO 2006b). Increased catches in 4Xp (Figure 3) reflect targeting for larger haddock in deeper water, which generally has higher market value, and also has relatively low cod bycatch (DFO 2006b). The most recent assessment of recruitment indices, age structure and F for the Scotian shelf fishery suggests that exploitation rates are moderate and can allow for some rebuilding of haddock populations (DFO 2010a). In addition, spawning stock biomass has increased over the past ten years and is above the historic long-term average (DFO 2010a). Eastern Georges Bank, Stock 5Zjm The eastern Georges Bank haddock stock has been jointly managed by Canada and the US since the implementation of the Hague Line in 1984 (Figure 2). In May 2004, a formal quota sharing agreement between Canada and the US was implemented to share the harvest of the transboundary eastern Georges Bank haddock management unit (Brodziak et al. 2006). This agreement includes total allowable catch quotas for each country as well as in-season monitoring of the catch of haddock on eastern Georges Bank (Brodziak et al. 2006). The eastern Georges Bank haddock stock has been commercially exploited since the late 1880s. Total average catch during the 1930s and 1950s was 25,000 t. Catch statistics are unavailable for the 1960s; however, average yearly catches for this period are estimated to be approximately 60,000t. Catches from the 1970s until present have been substantially lower. Catches declined from 6,504 mt in 1991 to a low of 2,150 mt in 1995, but have since increased and in 2009 catches were 19,707 mt (TRAC 2010; Figure 4). As in the other haddock stocks, overfishing led to significant long-term declines in spawning biomass and recruitment, and by 1993 adult haddock biomass reached a 9

historical low of only 9,000 t (Van Eeckhaute and Brodziak 2005). Since 1994, fishery closures in parts of the eastern Georges Bank, decreases in fishing effort, and larger mesh sizes for trawl gear have allowed the stock to largely recover (Brodziak et al. 2006). The estimate of total adult (ages 3+) biomass for eastern Georges Bank haddock has fluctuated in recent years ranging from 69,500 mt in 2003 to 58,600 mt in 2005 with a large increase in biomass (157,000 mt) in 2009 followed by a decrease to 125,200 mt in 2010 (TRAC 2007; TRAC 2010). An exceptional year-class in 2003 (largest in the time series, 293 million fish age 1) resulted in some of the highest adult fish biomass records in recent years (TRAC 2010). The 2005 years-class was near the time series average (26.5 million fish) and the preliminary estimate for the 2009 year-class is expected to be below average (5 million fish) (TRAC 2010). Stock size is expected to decrease in the near future as the 2003 year-class has reached maximum biomass (TRAC 2010). Scope of the analysis and the ensuing recommendation This report focuses on Canadian managed haddock stocks in the Scotian Shelf, Bay of Fundy, and the Canadian and US managed eastern Georges Bank stock. Imported haddock is not analyzed in this report. Figure 4. Haddock catch in 5Zjm from 1969 to 2009. Figure from TRAC 2010. Availability of Science The scientific information required to carry out this seafood assessment is relatively complete and accessible. The southern Scotian Shelf/Bay of Fundy and the Georges Bank haddock stocks are assessed regularly, with updates almost every year (TRAC 2010; 10

DFO 2010a). Assessments are based on long-term standardized trawl surveys, catch data, and other biological information. Because haddock is a valuable commercial species, biological and abundance information are fairly well known. In addition to government reports, there are also peer-reviewed scientific papers (Van Leewen et al. 2008). However, there are still uncertainties in stock assessments with estimating biomass, reference points and harvest control rules (DFO 2010a; TRAC 2010). Market Availability Common and market names: Haddock. It may also be called scrod, although this term typically refers to cod. Seasonal availability: Haddock is readily available in Canada year round (frequently sold frozen). Product forms: Haddock is available in Atlantic Canada in many forms including fresh or frozen, as bait, boneless, cooked, dressed, fillets, salted, smoked, steaks, or whole (Anon. 2006). Import and export sources and statistics: In recent years, haddock has accounted for only a minor proportion of Canada s total fish and seafood exports (1.4% in 2004, 1.5% in 2005) (Anon. 2007a), reflecting the overall decline in groundfish. Among groundfish species, however, haddock is relatively important. In 2007, Canada exported 8,018 tonnes of haddock, while in 2006 8,482 tonnes were exported (Anon. 2007a). Canada imports almost as much haddock as it exports. Canada imported 7,426 tonnes of haddock in 2007, and 7,955 tonnes in 2006 (Anon 2007a). In 2010, the United States imported approximately 950 tonnes of Canadian Haddock valued at $26,824,000, which represents 28% of the Haddock imported to the United States (NMFS 2011). The United States fishery landed 5,813 tonnes of haddock in 2009 (NMFS 2011). 11

III. Analysis of Seafood Watch Sustainability Criteria for Wild-caught Species Criterion 1: Inherent Vulnerability to Fishing Pressure The life history characteristics of haddock make it resilient to fishing pressure (Table 1). The age and size at maturity of haddock vary slightly among stocks, but in all stocks 50% of females are mature by age three (Mohn & Simon 2002, DFO 2005a), considered a low age at first maturity. There is some evidence that selective fishing pressure may have reduced the size and age at maturity of haddock in the Southern Scotian Shelf/ Bay of Fundy stock and size at maturity in Georges Bank haddock (Brodziak 2005; Mohn and Simon 2002). Haddock growth (both size-at-age and weight-at-age) is also depressed in these haddock stocks. The growth rate or von Bertalanffy growth coefficient (K) is a measure of the rate at which the asymptotic length is approached and it is often used as an indicator of a species resilience to fishing pressure. According to Froese and Pauly (2006), the von Bertalanffy growth coefficient, K, for haddock is 0.12 0.23. The maximum known age for haddock is 14 years (Stevens 2004), but only a small proportion of haddock survive past age 9 (Brodziak 2005). Table 1. Life history characteristics of haddock (from Froese and Pauly 2006). Age at Maturity Median age at maturity (years for female/male): SSSBF: ~3 Georges Bank: 1.5/1.3 Growth Rate/Max. Size K=0.12 0.23 Max. size: 110cm, 16kg, 35-60cm Maximu m Age ~14; ages 9+ rare SSSBF=Southern Scotian Shelf and Bay of Fundy Fecundity Range: 12,000-3,000,000 eggs depending on female size Species Range NW Atlantic: Labrador to Cape Charles and Virginia the Bay of Biscay to Spitzbergen; in the Barents Sea to Novaya Zemlya; around Iceland; and southern Greenland in the northeast Atlantic Special Behaviors Spawning aggregations Populatio n Variabilit y Recruitment highly variable Haddock has a moderate distribution, ranging from Labrador to Cape Charles and Virginia in the northwest Atlantic, and the Bay of Biscay to Spitzbergen; in the Barents Sea to Novaya Zemlya; around Iceland; and southern Greenland in the northeast Atlantic (Froese and Pauly 2006). The fecundity of haddock is high, with females producing between 12,000 and 3,000,000 eggs depending on their size (Table 1). They aggregate for spawning, which may increase 12

ease of capture in fisheries. Major spawning grounds for haddock include Georges Bank, Browns Bank on the southern Scotian Shelf, and Emerald, Western and Sable Island Banks on the eastern Scotian Shelf (DFO 2002b, 2005, Brodziak 2005; DFO 2010a). Haddock form spawning aggregations in these areas at various times of the year, with peak spawning occurring on Georges Bank in late-march through April, on Browns Bank from late April to early May, and on the eastern Scotian Shelf in spring (Brodziak 2005, DFO 2002b, 2005a). Spawning occurs near the ocean bottom over various substrates including rocks, gravel, smooth sand and mud (Klein-MacPhee 2002). The quality of their required habitat remains robust. The retention of haddock larvae in suitable nursery areas is highly dependent on oceanographic conditions and is an important determinant of the strength of haddock recruitment (the number of individuals surviving until the size of entry into the fishery) (Brodziak 2005). Following the larval stage, juvenile haddock gradually move from the water column to the seafloor (thus the name groundfish) where they spend the remainder of their lives (Klein-MacPhee 2002). Haddock recruitment is highly variable, characterized by periods of low recruitment that are punctuated by exceptionally high years. For example, low recruitment on Georges Bank for the past 40 years has been punctuated by extremely high years in1963, 2000 and 2003 (Figure 4; Brodziak et al. 2006; TRAC 2010). Synthesis Haddock is an early maturing, highly fecund species that is inherently resilient to fishing pressure. Heavy fishing pressure, however, has acted as a strong selective force to decrease size and age at maturity, which may adversely impact recruitment and natural mortality. Inherent Vulnerability Rank: Resilient Moderately Vulnerable Highly Vulnerable Criterion 2: Status of Wild Stocks Factors 1 & 2: Management classification status & abundance thresholds Historically, Canadian haddock stocks have been significantly overexploited and can be considered to be recovering from overfishing. It is unclear what the biomass is relative to B msy as DFO assessments do not usually report this information. Mohn & Simon (2002) have noted that there are no established biomass reference points for Canadian haddock stocks. These two factors receive a yellow ranking; details for each stock follow. On the southern Scotian Shelf and Bay of Fundy (stock 4X5Y), haddock spawning biomass declined from a peak in 1979 and reached a low of 22,000 t in 1990 (Figure 5; DFO 2005a; DFO 2010a). However, spawning stock biomass for the 4X5Y stock has increased over the past ten years, and in 2008, was above the long-term average of 39,317 13

t (Figure 5). Model estimates show recruitment was above average for the 2003, 2005 and 2006 year classes (DFO 2010a). Recruitment is typically stronger on the Scotian Shelf than in the Bay of Fundy (DFO 2006b; DFO 2010a). However, surveys in this region indicate that the proportion of haddock biomass in the Bay of Fundy is declining while the proportion of catches from the area is increasing (DFO 2006b). The haddock catch also is comprised of much smaller fish than in the past, and handliners have noted the absence of haddock throughout most of the traditional inshore fishing grounds (e.g., off Cape Sable Island) (DFO 2005b). Recruitment indices, age structure and F suggest that exploitation is moderate and can allow for some rebuilding (DFO 2010a). Figure 5. Spawning stock biomass (ages 4+ shown as a line) and age 1 recruitment (10 3 ), shown as bars) in the subsequent year for southern Scotian Shelf & Bay of Fundy (4X5Y) haddock. Figure from DFO 2010a. Eastern Georges Bank haddock (adult biomass, >3 years old) declined to low levels in the early 1970s, recovered slightly in the early 1980s and fell to a historical low in 1993. Adult biomass remained low throughout the mid-1980s and 1990s but has significantly increased since 2001. Adult biomass decreased slightly in 2005 to 58,600 mt; however, an extremely strong recruitment year from 2003 has increased adult biomass since 2006 (Figure 6; Van Eeckhaute and Brodziak 2005; TRAC 2010). The strong 2003 year-class has reached its maximum biomass and is expected to decline (TRAC 2010). In 2009, the adult biomass increased to over 157,000 mt, but declined in 2010 to 125,100 mt and is expected to be 68,000 mt in 2012 (TRAC 2010). 14

Figure 6. Haddock adult (age 3+) biomass and number of age 1 recruits in area 5Zjm of Georges Bank during 1931 1955 and 1969 2010. Figure from TRAC 2010. Factor 3: Occurrence of overfishing Haddock are no longer being overfished, thus, this factor receives an overall green ranking. The 5Zjm and 4X5Y stocks are well below historic exploitation rates and the F 0.1 reference point (Figure 4 and Figure 7). Figure 7. Exploitation Rate (ages 5 7) for 4X5Y Haddock stock. Figure from DFO 2010a. 15

Low exploitation (and high recruitment) in the early 1990s allowed the southern Scotian Shelf/Bay of Fundy haddock population to start to rebuild (FRCC 2004). A fishing mortality reference point of F 0.1 =0.25 has been used as a harvest strategy for this stock. The exploitation rate on age 5 7 haddock increased from the 1970s to ~50% in the early 1980s but then declined to close to F 0.1 (20%, currently used value of F 0.1 =0.25) in the late 1980s (FRCC 2004). By 1994, the exploitation rate had dropped below F 0.1 and has remained low since (DFO 2005a). If catches are maintained at the current level (around 5,300 t in 2009 and 5,400 t in 2010), fishing mortality will remain below F 0.1 (DFO 2010a). A reduction in fishing mortality below the reference threshold of 0.26 has been achieved for the eastern Georges Bank stock since 1995. In addition, an increase in the minimum trawl mesh sizes has reduced the amount of juvenile bycatch and led to a greater abundance of older fish. Catch levels for 2005 2007 are considered to have had a low risk (<25%) of fishing mortality exceeding the fishing reference threshold (Van Eeckhaute and Brodziak 2005). In 2009, fishing mortality was below reference levels at 0.13, and in 2010, assuming a catch of no more than 29,600 tons, will result in a neutral risk of the 2011 fishing mortality exceeding the 0.26 reference point. Factor 4: Overall degree of uncertainty in status of stock Being a commercially important species, haddock receives considerable research attention. Haddock stocks are monitored in a long-term research vessel survey in addition to surveys initiated together by DFO and industry in the 1990s. Evaluation of stock status is derived from these surveys together with reported landings, shore and sea-based samples taken from commercial landings for size and age composition (DFO 2002). With respect to the Scotia-Fundy region, although the southern Scotian Shelf and the Bay of Fundy are managed as a single stock, they are thought to be separate stocks (DFO 2005a). Different population dynamics between these areas increases the uncertainty of analyses (DFO 2005a). Past assessments for haddock in this region have a general tendency to over-estimate population abundance in the most recent year, particularly when strong year classes occur (DFO 2005a). In addition, there are large uncertainties in estimating reference points and harvest control rules, as well as in estimating current biomass (TRAC 2010; DFO 2010a). For example, the 2003 year-class in 5Zjm has reached maximum biomass and as a consequence, age 3+ stock biomass has declined from a peak of 157,000 mt in 2009 and is expected to be 68,000 mt in 2012 (TRAC 2010). Catches in 5Zjm for 2010 and 2011 will continue to be dependent on the 2003 year-class but could decrease quickly if the partial recruitment on ages 9+ remains low (TRAC 2010). All stocks for this factor therefore receive a yellow ranking. Factors 5 & 6: Long-term and short-term trends in abundance Haddock stocks have experienced highly variable trends in abundance (Figure 5 and 6; DFO 2010a; TRAC 2010). In general, the stocks reached low points in the early 1970s, and then largely recovered before crashing again in the early 1990s (Figure 5 and 6; DFO 2010a; TRAC 2010). The short term trends (i.e., over the past decade) for the Scotian Shelf and Georges Bank show rebuilding abundance (Figure 5 and 6; DFO 2010a; TRAC 16

2010). These two factors receive rankings of yellow because of the long-term repeated pattern of overexploiting haddock stocks, and because the short-term trend, although generally positive due to high-but-variable recruitment, has not yet showed full and sustained recovery. Factor 7: Current age, size, or sex distribution Haddock has experienced marked reductions in growth rate, size-at-age is at a historical low, and productivity has been impaired in some regions. This factor therefore receives a red ranking. Haddock on the southern Scotian Shelf and Bay of Fundy have experienced declines in growth rates. The average weight-at-age of haddock has been decreasing since the mid- 1970s with a more severe decline for fish that are five to seven years old (Figure 8; DFO 2010a). The population is dominated by small haddock (< 43 cm), and in 2006, 83% of haddock on the southern Scotian Shelf research survey were less than 43 cm (DFO 2006b; DFO 2010a). Under the current growth regime, productivity is only about two thirds of what it was in the 1970s, mainly because of the decrease in growth (DFO 2005a). Although a yield-per-recruit analysis indicated that there would be no increase in yield by delaying fishing of haddock to older ages, catching the small fish does have implications for future spawning stock biomass and other aspects of production (DFO 2003a, 2005a). The age composition of haddock shows an expanding age range in recent years on the southern Scotian Shelf, but the age range for the Bay of Fundy remains truncated (DFO 2005b; DFO 2010a; Figure 8). Although weight at age for haddock remains low and the mean weight per tow in the Bay of Fundy is below the long-term average, the most recent data from the Scotian Shelf show a slightly positive trend over the past five years (DFO 2010a). Figure 8. Research vessel data showing the mean weight-at-age (kg) (left panel) for the Scotian Shelf and mean weight-at-age (kg) (right panel) for Bay of Fundy haddock. Figure from DFO 2010a. Mean weights at age of eastern Georges Bank haddock have been in decline since the 1990s, although these declines have mostly halted recently (Table 2; TRAC 2010). Consequently, the productivity of the stock was diminished due to reductions in average 17

fish size (Van Eeckhaute and Brodziak 2005). In addition to fishing mortality, environmental change and/or a response to increased haddock density are suggested as possible reasons for this decline (Brodziak et al. 2006). The current population age structure of the eastern Georges Bank shows full representation of all age classes (TRAC 2010). However, weight at length has exhibited declines since the late 1990s and declined in 2010 to below average for each length (TRAC 2010). There are some positive trends for the 5Zjm region, including expanding age structure, improvements in size at age for younger fish and a broad spatial distribution; however, weight at age and size structure remains skewed and have not recovered from historic declines. Table 2. Average weight at age (kg) of haddock from the Canadian commercial fishery on the eastern Georges Bank during 1969 2004. Source: Van Eeckhaute and Brodziak 2005. Age Group 1 2 3 4 5 6 7 8 Minimum 0.394 0.589 0.812 1.272 1.643 1.631 2.002 2.032 Maximum 0.797 1.215 1.724 2.235 2.639 3.760 4.114 4.009 Median 0.600 0.993 1.399 1.791 2.144 2.507 2.868 3.108 Average 0.583 0.99 1.376 1.797 2.142 2.471 2.814 3.074 2000 04 0.484 0.912 1.368 1.708 1.957 2.116 2.337 2.567 Status of Wild Stocks Rank: Healthy Moderate/Rebuilding Poor Critical 18

Criterion 3: Nature and Extent of Bycatch Factor 1: Quantity & composition of bycatch 1 The Atlantic haddock fishery is primarily conducted with mobile (trawl) and fixed gear (bottom longline) gear. In 2007, 68% of the commercial catch was caught by trawl and 32% was caught by longline. Bycatch of cod, the most common bycatch for both gear types, is currently managed by quotas limiting the total allowable take of cod. Longline vessels are limited by the amount of cod bycatch the fleet is allowed, while trawlers have sufficient cod bycatch quotas. In addition, trawlers fishing Georges Bank are required to use trawl separator gear that significantly reduces the number of cod caught while fishing for haddock (TRAC 2010; Anonymous 2011). Current haddock fishing practices and catch ratios are limited to meet the rebuilding objectives for cod (TRAC 2010). Cod has been assessed as endangered by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC 2011). The quantity and composition of bycatch, including cod, differ between gear types. Cod is landed and accounted for and haddock catch is constrained by cod quotas, bycatch limits, or mitigation gear (separator panels); however, cod is addressed in this report because it is a species of concern caught in the haddock fishery. Otter trawls are traditionally a fairly indiscriminate gear type. Consequently, the trawl fishery is considered a mixed species fishery and bycatch and subsequent discarding of unmarketable (e.g., undersized) or otherwise prohibited species occurs frequently. In particular, haddock, cod and pollock are often caught and landed together in groundfish fisheries (DFO 2005a). As a condition of license, since 2007, the trawl fleet in eastern Georges Bank has been required to use separator panels with bottom escape chutes to reduce bycatch of cod. Estimates of overall discard levels have been made through an extrapolation of observed discard rates for otter trawls and the bottom longline fisheries. In general, bycatch estimates have a number of limitations, including (1) they have not been analyzed at fine temporal (e.g., seasonal) or spatial scales, and (2) in the case of Area 4X5Y, they are based on low observer coverage levels (mean of 2.1% for trawl and 0.4% for longline) (MSC 2010). Observer data from trawlers using separator panels reported that the majority of fish caught were landed (99%), with only 1% of catch discarded at sea. Landed catch consisted mostly of haddock (92%) with the remainder consisting of equal amounts of cod and pollock (4% each). Discarded species may also include species considered at risk by COSEWIC, including porbeagle and shortfin mako sharks (endangered and threatened, respectively); winter skate (special concern) and the northern (Anarhichas denticulatus) and spotted (Anarhichas minor) wolfish (listed as endangered by SARA). Nevertheless, the combined discards of these species are extremely low (< 0.4% of the catch)(tables 3, 4, 6 and 7; MSC 2010). Habitat and nest 1 Percentages are reported based on percent weight of catch for a given year; catch data is based on DFO quota reports (Anon. 2007b); bycatch information is from DFO Observer data. 19

destruction for wolfish species by trawling gear may have a larger impact on the recovery of wolfish than bycatch (MSC 2010). This will be covered in the following section. Observer data from longliners on eastern Georges Bank in 2006 reported that 91% of the fish caught were landed, with the remaining 9% of the catch discarded at sea. Landed catch comprised 63% haddock, 17% cod, 5% cusk, 3% white hake and small amounts of other species (<1% each). Discarded catch comprised mainly spiny dogfish (56%), barndoor skate (30%) and winter skate (5%), with the remaining 9% of the discarded catch consisting of nine different species ( 2% each). The longline fishery catches several non-target species that are considered at risk by COSEWIC, including cusk (threatened), winter skate, blue shark and Atlantic wolfish (all special concern). Despite the fact that for some species bycatch values are small in terms of catch by weight, for species consisting of relatively small individuals, bycatch values for cusk may translate to higher impacts on a per capita basis than for species made up of large individuals. Discards are shown in Tables 5 and 6. Currently, discards are not routinely recorded in logbooks and are only estimated by the independent observer program (MSC 2010). This factor receives a red ranking for both the trawl and the longline fishery because bycatch and discards regularly include species of special concern (species assessed as endangered, threatened or of special concern), including winter skate and cusk (in the longline fishery only) (Tables 3 8), and because these fisheries also catch cod, a species of special concern. However, it should be noted that trawl fisheries in some areas (Georges Bank fishery) have mandated the use of separator gear that has significantly reduced bycatch rates, especially for cod and winter skate, which overlap with the Georges Bank haddock fishery (Table 7). In addition, other species of concern, such as the porbeagle population, are threatened more by longline fleets targeting swordfish or tuna than by haddock trawls (Tables 3 and 4; SARA 2011). No cusk is discarded as bycatch in the trawl fishery. Table 3. Trawl discards in the Southern Scotian Shelf/ Bay of Fundy Haddock fishery 2004 2008 (from MSC 2010) 1 estimate based on one shark and is considered a statistical anomaly (MSC 2010) 20

Table 4. Trawl discards in the Eastern Georges Bank Haddock fishery 2004 2008 (from MSC 2010). Table 5. Longline discards in the Southern Scotian Shelf/Bay of Fundy Haddock fishery 2004 2008 (from MSC 2010). Table 6. Longline discards in the Eastern Georges Bank Haddock fishery 2004 2008 (from MSC 2010). 21

Table 7. Total discards (2002 2008) of skate species and their proportion by species from observations averaged from 2004 2008. OTB = Bottom trawl and LL = Longline. Table from MSC 2010. Table 8. Discarded skate species and their SARA and COSEWIC listings. Skate Species SARA COSEWIC Barndoor Skate Not Listed Not at Risk Brier Skate Not Listed Not Listed Jensen s Skate Not Listed Not Listed Little Skate Not Listed Not Listed Round Skate Not Listed Not Listed Smooth Skate Not Listed Not Listed Soft Skate Not Listed Not Listed Spiny-tail Skate Not Listed Not Listed Thorny Skate Not Listed Not Listed White Skate Not Listed Not Listed Winter Skate No Schedule Southern Gulf of St Lawrence- Endangered Eastern Scotian Shelf- Threatened Georges Bank and Western Scotian Shelf Special Concern 22

Factor 2: Population consequence of the bycatch The bycatch of cod in the haddock fishery has been a particular concern even though haddock are explicitly managed to protect cod (FRCC 2004, DFO 2005a, RAC 2010). With regard to current fishing practices for the southern Scotian Shelf and Bay of Fundy haddock stock, DFO (2006b) and TRAC (2010) have stated that the achievement of rebuilding objectives for cod and pollock may constrain the harvesting of haddock. According to both the Department of Fisheries and Oceans and the Fisheries Resource Conservation Council, an imbalance in quotas between haddock and cod creates the potential for discards and requires improved monitoring (FRCC 2004, DFO 2005a). Threats to the persistence of cod in Atlantic Canada include directed fishing, bycatch in other fisheries, natural predation, and natural and fishing-induced changes to the ecosystem (COSEWIC 2003a). The population consequences of the haddock longline and trawl fisheries on species of special concern, including cod, cusk and winter skate, are not well understood. However, management has taken actions to limit catch of these species. Although there is limited directed fishing for cusk, it is mainly taken as bycatch on longlines that target Atlantic halibut, cod, haddock and pollock (SARA 2011). The Scotia-Fundy region experienced a 93% decline in cusk populations from 1970 to 2001 driven by over-exploitation from fisheries, in particular longline bycatch, according to the COSEWIC assessment last updated in 2003 (SARA 2011). At that time, therefore, the levels of bycatch of cusk probably were contributing to declines in this species. In addition, a DFO publication from 2004 suggests that the trajectory in abundance indices for cusk at that time appeared flat, suggesting that a decrease in human-induced mortality (from bottom longline fisheries such as haddock) was required for an increase in abundance (DFO 2004). However, more recently, DFO has reduced the allowable catch of cusk. The allowable catch was set based on projections that a catch of 600 mt in area 4X (equivalent to a catch of 858 mt in all areas) would yield a 54% chance that biomass of cusk could increase by 50% in 15 years (DFO 2008). The TAC for cusk was lowered accordingly to 750 mt for all areas (DFO 2010b) with actual catch at about 609 mt in 2009, compared to landings in previous years ranging from 790 mt to 1,490 mt (MSC 2010). DFO was not able to set a recovery target for cusk given the high uncertainty in its abundance measures (DFO 2008). Given the uncertainties in the status of cusk and whether the current quota will be sufficiently low to allow cusk to recover, it is unknown whether or not fishing is currently limiting the recovery of cusk. Similarly, cod populations have not recovered as hoped in recent years, but DFO has responded by lowering the quota. The cod quota for area 4X5Y was reduced to 6000 mt in 2000 and 5000 mt in 2006. In 2009, an assessment showed that a catch of 3000 mt would result in a 90% chance of a 10% increase in biomass. Accordingly, the TAC in 2009 was set to 3000 mt for the area. However, cod in 5Zjm were not projected to increase, even with no catch. Management therefore set a policy to achieve a low to neutral risk of overfishing the stock. In 5Zjm (Eastern Georges Bank), bycatch of cod in the haddock trawl fishery is fairly low due to the success of separator trawl gear. Given the concerns and uncertainty around the potential of cod to recover, maintaining cod 23

catch at an even lower level than current quotas in both 5Zjm and 4X5Y (as low as possible) would be preferable from a precautionary standpoint (MSC 2010). The trawl and longline fisheries receive a yellow ranking because bycatch of some species, such as cod and cusk, were likely contributing to declines or limiting recovery of these species in past years, but bycatch has since been constrained by reduced quotas and reduced bycatch (particularly for winter skate and cod) due to the use of separator panels that are mandated on Georges Bank. Currently, the population consequences for bycatch species remain unknown in the trawl and longline fisheries, but management is addressing the issue of catch of species of special concern. Factor 3: Trends in interaction bycatch rates Fishermen in both the fixed and mobile gear fleets of the southern Scotian Shelf and Bay of Fundy reported to the Fisheries Resource Conservation Council (FRCC) in 2004 that they had made significant progress in fishing for haddock with minimal bycatch of cod in what was previously predominantly a mixed cod and haddock fishery (FRCC 2004). However, the FRCC recognizes that more effort is required by fishers and DFO management to fish for haddock within specific areas and times so as to minimize cod bycatch, as well as to develop innovative ideas in gear technology that will accomplish the same end (FRCC 2004). DFO (2005a) has similarly noted that impacts on cod may be mitigated through enhanced monitoring and modifications to fishing gear. Currently, haddock fishing is regulated to meet the objectives of rebuilding cod populations (TRAC 2010). Mandatory gear modifications (separator panels) for the 2007 Georges Bank trawl fishery resulted in relatively low bycatch rates, particularly reducing bycatch of cod and winter skate. However, it is unclear whether these low numbers were the result of gear modification or low availability of non-target species in this area. This mandatory gear modification did not extend to the entire trawl fleet, and thus the remaining 41% of trawlers fishing the southern Scotian Shelf and Bay of Fundy did not consistently adopt these measures. In area 4X, trawlers may have not adopted the separator panels because cod bycatch is managed using a quota, and they were not mandated to use the panels (Anon. 2011). The longline fishery reports moderate bycatch levels, including species at risk. To improve this trend, the longlining fishery needs to adopt new management measures to reduce the amount of non-target species it captures. However, total catch of major at-risk species like cod and cusk has been further limited due to reduced quotas designed to allow these species to recover. This factor receives a yellow ranking for the trawl fishery because of the use of separator panels in some areas, but further improvements (and evidence of these improvements) are needed to demonstrate declines in bycatch interaction rates for this fishery as a whole. The longline fishery also receives a yellow ranking because bycatch and discard rates have not been increasing; for some species rates are decreasing due to reduced quotas. 24

Secondary Factors: Ecosystem impacts In the western north Atlantic, intense fishing pressure in the past fifty years has resulted in a marked shift from a marine community dominated largely by benthic fishes to one dominated largely by smaller pelagic forage fish (Frank et al. 2005, Fogarty and Murawski 1998). Recent evidence suggests that the collapse of cod and other commercially exploited groundfish, including haddock, triggered this change and that the new ecosystem state may be impairing the recovery of cod (Van Leeuwen et al. 2008, Frank et al. 2005). Continued harvest of cod, including cod caught incidentally, will slow the recovery of this stock and thus contribute to the persistence of this altered ecosystem. This factor therefore receives a red ranking. Nature of Bycatch Rank: Low Moderate (Trawl, Longline) High Critical Criterion 4: Effect of Fishing Practices on Habitats and Ecosystems Factor 1: Impacts of fishing gear on habitat In 2007, trawlers and longliners accounted for 68% and 32% of the fishery (by weight), respectively (Anon. 2007b). In 2008, area 4X5Y landings were 78% from bottom trawl and 22% from longline while area 5Zjm landings were 85% from trawl and 15% from longline (MSC 2010). Trawlers in Canada and the US employ various types of demersal trawl gear (mainly otter trawls) tailored to the habitats of target species or species assemblages (Stevens 2004). The impact of trawling on benthic (bottom) ecosystems has been the subject of much research over the past decade. Trawling impacts sea-floor communities by scraping the ocean bottom and causing: 1) sediment re-suspension (turbidity) and smoothing, 2) removal and/or damage of non-target species, and 3) destruction of abiotic and biotic three-dimensional habitat (Watling and Norse 1998, Auster and Langton 1999; Figure 9). This factor is given a red ranking for the trawl fishery. Based on these results and those of the large body of research on this topic, it is evident that otter trawling and other bottom trawling gear can alter the surrounding ecosystem by degrading or removing available fish as well as invertebrates and benthic habitats. Longline capture methods cause far fewer impacts on benthic habitats in comparison to bottom trawling; however, moderate damage can occur from the terminal anchors and by contacting the bottom when hauling in gear (Fuller et al. 2008). This factor is given a yellow ranking for the longline fishery. Factors 2: Resilience of the habitat Among the areas where haddock are commercially fished in Canada, fishers encounter a mix of habitats that differ in sediment type. The seabed on Georges Bank is predominantly sandy whereas southern Scotian Shelf/Bay of Fundy represent a variety of bottom types including sand, gravel, clay and silt. Some sediment types (and their 25