Blue shark (Atlantic and Pacific populations) COSEWIC assessment and status report: chapter 10

Population Sizes and Trends

There is no region in the world where blue shark abundance and trends are well understood. Trends in the abundance can only be inferred through catch rate information and limited biological information taken from catches. Internationally there are only a few published accounts of blue shark abundance trends. Due to the highly migratory behaviour of this species, its status in Canadian waters is probably dependent on ocean basin-wide population trends.


Information Sources

Atlantic

There are no surveys specifically designed to estimate abundance of blue sharks. Information on abundance, trends, and biological information in Canada is derived from commercial catch data verified through the Maritime and Newfoundland International Observer Programs (IOP) (1978-2003) and from recreational fisheries in the form of shark derbies. Given that blue sharks found in Canadian waters are part of a much larger Atlantic population, international assessments from NMFS and International Commission for the Conservation of Atlantic Tuna (ICCAT) are also important in determining the status of blue shark in Canada.

Pacific

Although anecdotal information from fishers and other mariners indicates that blue sharks are common in waters off British Columbia, there are no indicators of abundance. There has been only one directed investigation into blue sharks in Canada’s Pacific waters. Catch databases including trawl and hook and line were examined for records as well as all Canadian groundfish surveys from 1979-present. International surveys that occur in Canadian waters, including the IPHC survey and U.S. NMFS Triennial groundfish survey, were examined for Canadian records of blue sharks. International surveys and assessments occurring adjacent to Canadian waters were also examined.

Atlantic Population Abundance and Trends

Atlantic Population Abundance

There are no reliable estimates of the number of mature blue sharks in Canadian waters. Mature females are rarely observed, which is thought to be due to a natural age/sex segregation. There are no defensible methods for estimating absolute abundance in Canadian waters.

Atlantic Population Trends-International Waters

The existence of international high-seas fisheries coupled with the wide-ranging nature of the blue shark makes it difficult to acquire reliable data suitable for determining trends. Specifically, the under-reporting, misreporting, and non-reporting of shark bycatch by several international fleets has hampered the ability to conduct accurate stock assessments (ICCAT 2005). Furthermore, the lack of biological data, including age and length frequencies, sex ratios, breeding periodicity, and estimates of natural mortality further constrain the possibility for accurate assessments. However, there appears to be no disputing that total fishing effort on pelagic species in the North Atlantic has increased approximately tenfold over the last fifty years (Figure 11). Since those fisheries unavoidably capture blue sharks as bycatch, it is highly probable that total fishing mortality of blue shark in the North Atlantic has increased correspondingly.

Large scale abundance indices of blue sharks rely primarily on catch rates (catch per unit of effort or CPUE) reported from a sub-sample of the international fishery. Two international blue shark assessments are presented; one from ICCAT and the other from an analysis published by researchers at Dalhousie University in Halifax (Baum et al. 2003). Both of these assessments use components of the same datasets.

International Commission for Conservation of Atlantic Tunas (ICCAT) Blue Shark Assessment

In 2004 the ICCAT subcommittee on bycatch undertook a review of available blue shark information throughout the Atlantic (north and south) for the purposes of deriving an overall assessment (ICCAT 2005). The incomplete nature of shark catch reporting necessitated the estimation of blue shark catch and mortality using ratios of shark to tuna landings from fleets reporting both to ICCAT. A catch history for blue shark by major gear type was reconstructed using these ratios. The assessments described below use this catch reconstruction. Additional uncertainties include uncertainty on frequency of reproduction (which affects intrinsic rate of increase), and uncertainty about initial carrying capacity of the species. Because of uncertainties in input data the results of the assessments were considered “preliminary and extremely uncertain” by ICCAT, and were used mainly to provide a tentative basis for assessment models which utilize catch, effort and biological information. Models for the North Atlantic were based on what were considered to be the best available data; (1) Japanese longline logbook and (2) the U.S. longline logbook (Figure 14). Three models were explored; a Bayesian surplus production model, an age-structured model, and a ‘catch-free’ model.


Figure 14: Indexed CPUE of blue shark in the North Atlantic from the Japanese longline fleet (JLL) and United States longline fleet (USLL) from 1971-2003

Figure 14: Indexed CPUE of blue shark in the North Atlantic from the Japanese longline fleet (JLL) and United States longline fleet (USLL) from 1971-2003.

Index achieved by identifying the years common to all indices, calculating the mean within each series for those common years, and expressing all values in each series as a proportion of the calculated mean for that series. Figure from ICCAT 2005.


The Bayesian surplus production model involved 10 runs of which the three which converged showed an average current biomass around 85% of the unfished biomass.

The age structured model was run using the complete CPUE series from the U.S. longline logbook CPUE series and the CPUE series from the Japanese longline fleet without the first three values (1971-1973) (Figure 15). The model was run using two different weightings of the CPUE series; equal weighting (Run 1) and catch dependent weighting (Run 2). The model was also run assuming a biannual or annual reproduction cycle (Run 3). The probability density functions for population depletion of all three runs indicated depletion of about 50% but also suggested that the ratio of the current stock size to virgin stock size is close to 1 (i.e., no depletion).

The catch-free model consisted of three scenarios using the averaged CPUE index (Figure 14). The results from the original assessment produced implausible results and were corrected in a later ICCAT document (Brooks 2005). The results of the re-assessment are the same as those made during the assessment meeting, namely that blue shark do not appear to be overfished. 

ICCAT (2005) and a subsequent peer-review (Simpfendorfer 2005) concluded that the assessments showed that North Atlantic blue shark population is above the biomass at maximum sustained yield, but stated that results should be considered “preliminary and extremely uncertain”. One critical assumption for all of the assessments described above is the initial blue shark carrying capacity: if the assumed value was correct, then current biomass is well above MSY levels; if underestimated, then the current biomass could be below MSY levels. Overall, the ICCAT document provides a very useful data summary but does not offer any strong conclusions one way or the other about the status of North Atlantic blue shark.


Figure 15: Fit of the Age Structured Model for North Atlantic Blue Shark for Each of the Runs Considered Using Japanese (above) and U.S. CPUE Data (below)

Figure 15: Fit of the age structured model for North Atlantic blue shark for each of the runs considered using Japanese and U.S. CPUE data.

Figure from ICCAT 2005.


The abundance index for the years 1971-2003 based on catch per unit effort in the US and Japanese longline fleets used by ICCAT (Figure 14) can be used to explore abundance trends. This index shows no long-term change since 1971, although there was an increase to the mid-1980s followed by a decrease.

Dalhousie University Blue Shark Assessment

Analyses by Baum (2002) and Baum et al. (2003) use the same U.S. longline logbook data as ICCAT but restrict their analysis to catch rates in the period 1986-2001 when species identification was considered relatively reliable. The author/s examined data from nine assessment regions covering the western Atlantic from South America to the Grand Banks of Canada (Figure 16). The dataset analyzed consists of 214 234 longline sets between 1986 and 2000 with a mean of 550 hooks per set. A total of 1 044 788 records of blue sharks are found in the database.


Figure 16: Map of the Northwest Atlantic Showing the Distribution of Effort in the U.S. Pelagic Longline Fishery Between 1986 and 2000

Figure 16: Map of the Northwest Atlantic showing the distribution of effort in the U.S. pelagic longline fishery between 1986 and 2000.

Categorized by number of sets (0 to 800+), within the nine areas assessed: 1, Caribbean; 2, Gulf of Mexico; 3, Florida East Coast; 4, South Atlantic Bight; 5, Mid Atlantic Bight; 6, Northeast Coastal; 7, Northeast Distant; 8, Sargasso/North Central Atlantic; 9, Tuna North/Tuna South. Areas were modified from the U.S. National Marine Fisheries Service classification for longline fisheries. The 1000-m coastal isobath (dotted line) is given for reference. From Baum et al. 2003. Note, it is difficult to discern the 100 hexagon from the 700 hexagon in the grey scale presented. With the exception of a few hexagons in close proximity to high density areas (darkest grey), the majority are low density (0-100).


For international waters adjacent to the Scotian Shelf (Area 6) and the Grand Banks (Area 7) declines in CPUE of 63.8% and 9.6% respectively were estimated for the years between 1986-2000 (Figure 17, Baum 2002). The overall trend in blue shark CPUE in the entire northwest Atlantic reported by Baum et al. (2003) showed a gradual and constant decline of 60% (95% CI: 58 to 63%) over the period 1986-2000, equivalent to a decline rate of 6.4% per year (Figure 18).


Figure 17: Catch Rate of Blue Sharks in Three Assessment Regions in the northwest Atlantic

Figure 17.Catch rate of blue sharks in three assessment regions in the northwest Atlanticexpressed as a mean (full circles) and median (solid line)

Expressed as a mean (full circles) and median (solid line). Figures reprinted from Baum 2002. Note these areas are the same as those found in Figure 15.

 


Figure 18: (A) Relative Abundance of Blue Sharks in the Entire West Atlantic Indicated by an Analysis of U.S. Commercial Longline Fishery Logbook (60% Decline) from 1986-2000; (B) Estimated Annual Rate of Change for Each Assessment Region and Total

Figure 18: (A) Relative abundance of blue sharks in the entire West Atlantic indicated by an analysis of U.S. commercial longline fishery logbook (60% decline) from 1986-2000; (B) estimated annual rate of change for each assessment region and total

Note the areas in (B) are shown in Figure 16. Source: Baum et al. 2003.


The findings by Baum et al. (2003) are supported by similar findings by Brooks et al. (2005) who analyzed standardized catch rates for blue sharks using U.S. pelagic logbook data (1986-2003) from the entire North Atlantic. Brooks et al. (2005) found that the overall relative decline in catch rates between 1989 and 2003 dropped from a relative value of about 0.9 to 0.3. Similarly, Simpfendorfer et al. (2002), using catch and effort data from a fishery-independent longline survey in the western North Atlantic found that male blue sharks showed an approximately 80% decline between the mid-1980s and the early 1990s, while a significant change in female catch rates could not be demonstrated.

On the other hand, Burgess et al. (2005) challenged some aspects of the analysis by Baum et al. (2003) and argued that the trend observed for the blue shark should be considered preliminary without the full benefit of data from multiple international sources and a complete stock assessment.


Atlantic Population Trends-Canadian Waters

Population trends observed in Canadian waters have been reported in Campana et al. (2004) based on catch rates in the commercial fishery and fishing derbies and inferences from biological data. Campana et al. (2004) used a general linear model (GLM) to analyze standardized commercial catch rates from Canadian and Japanese vessels with year, region, season, target species and vessel (CFV) as factors. The analysis was restricted to fall and winter and the regions Newfoundland and Scotian Shelf, for the period after 1994. The GLM trends in swordfish and bigyeye tuna fisheries were similar and were analyzed together whereas the trend for bluefin tuna fisheries necessitated a separate analysis. The GLM of blue shark catch rate based on bigeye tuna and swordfish fisheries data indicated that all factors but season and target species were significant. The marginal catch rate based on significant factors indicated that catch rates have declined significantly since 1995 (Figure 19). The decline is estimated to be 53% (6.6%/yr) from 1995-2003 with the year term in the model being highly significant (P<0.001) (Figure 19). The GLM based on bluefin tuna fisheries was significant with respect to all factors. However, the significant interaction terms necessitated that the marginal trends be plotted separately by region (Figure 20). The trend based on the Scotian Shelf fishery showed a significant decline for the first three years of the time series, but with relative stability in recent years. The trend based on the Newfoundland fishery suggested a modest increase since 1995, although there were few significant differences among years.

Catch rates from the recreational derby fisheries were also used as an index of blue shark abundance (Campana et al. 2004). Estimation is complicated by the presence of multiple fishers per boat, discarding of undersized sharks, and difficulties in assigning catches to specific fishers on the boat. A coarser approximation of fishing success (percentage of fishers catching sharks) at each derby suggests that catch rates across derbies (i.e., different regions) within a year are often synchronous and have declined slightly between 1998 and 2003.


Figure 19: Standardized Commercial Catch Rate (ln-transformed kg/hook ± 95% CI) of Blue Shark in Canadian and Japanese Large Pelagic Fisheries Targeting Bigeye Tuna and Swordfish

Figure 19: Standardized commercial catch rate (ln-transformed kg/hook ± 95% CI) of blue shark in Canadian and Japanese large pelagic fisheries targeting bigeye tuna and swordfish.

From Campanaet al. 2004.


Analysis of SST (sea surface temperature) in relation to recreational catch rate suggests a relationship with temperature, since the year with the highest catch rate (1999) was also among the warmest, while the year with the poorest catch rate (2001) was among the coldest. Anecdotal comments by fishers also supported the view that there was a correspondence between warm water conditions and higher catch rates for blue sharks. Records maintained by a charter operator indicate that blue sharks were caught at temperatures between 10-20°C; however, most were caught between 14-18°C (Campana et al. 2004). A more rigorous statistical analysis of these data was not carried out.

A standardized catch rate for the recreational fishery was prepared by Campana et al. (2004) using a binomial dependent variable (successful/unsuccessful) and derby location as a fixed factor in a GLM. The model was not statistically significant. A final GLM was based on the overall fishing success at the 5 shark fishing derbies carried out annually since 1998. Individual catch rates were not available, so an index based on the percentage of fishers successful in catching a shark at each derby was used. This model was less than ideal, since the derbies represented fixed factors, and thus year X derby interaction terms could not be assessed. With these deficiencies in mind, the model suggested a significant decline since 1999 of 27% (Figure 21). When scaled to the same scale as the standardized bigeye/swordfish model (Figure 19), the trend across years was similar in the two models. These results suggest that the derbies and the offshore commercial fishery are samples from the same population, and that the catch rate from 1999 to 2003 was significantly lower than in the period 1995 to 1998.


Figure 20: Standardized Commercial Catch Rate (ln-transformed kg/hook ± 95% CI) of Blue Shark by Region in Canadian and Japanese Large Pelagic Fisheries Targeting Bluefin Tuna on (A) the Scotian Shelf and (B) Waters off Newfoundland

Figure 20: Standardized commercial catch rate (ln-transformed kg/hook ± 95% CI) of blue shark by region in Canadian and Japanese large pelagic fisheries targeting bluefin tuna on the Scotian Shelf and waters off Newfoundland.

Source: Campana et al. 2004.

 


Figure 21: Standardized Catch Rate (sharks/fisher, ± 95% CI) of Blue Sharks at Recreational Shark Derbies

Figure 21: Standardized catch rate (sharks/fisher, ± 95% CI) of blue sharks at recreational shark derbies.

From Campana et al. 2004.


Biological data from long-term changes in mean length can be used as an indicator of exploitation intensity. Since 1986 there has been a significant decline in the mean lengths of commercially caught blue sharks in both the Canadian and Japanese fisheries (Figure 22). The analysis was restricted to the fall and winter seasons in order to minimize seasonal differences. The difference between the Canadian and Japanese fisheries is most likely due to differences in depth fished (Campana et al. 2002). There was no apparent trend from the recreational fishery, presumably due to targeting of the largest individuals.


Figure 22: Trend in Mean Fork Length (± 95% CI) of Blue Sharks Caught in Fall and Winter in Japanese (open square-top series) and Canadian (closed circle-bottom series) Pelagic Longline Fisheries, as Observed by the International Observer Program

Figure 22: Trend in mean fork length (± 95% CI) of blue sharks caught in fall and winter in Japanese and Canadian pelagic longline fisheries, as observed by the International Observer Program.

From Campana et al. 2004.


Summary of Atlantic Trends

Overall there appears to be a decline in blue shark abundance in the northwest Atlantic based on several indices (see Table 1 for summary). All of the indices examined have shortcomings. The Canadian Atlantic population of blue sharks is part of a population occupying the entire North Atlantic. Studies that cover large geographic areas are necessary for understanding the overall status of this population. Results from regional studies may be influenced by local environmental conditions and demographic partitioning of the hemisphere-wide population. With respect to geographical coverage, the ICCAT population assessments cover the entire North Atlantic range of the population. However, the input data are considered to be highly uncertain and the results are inconclusive. The ICCAT CPUE series is based on uncertain identification to species in the earlier years. The Baum et al. (2003) analysis covers only the northwest Atlantic and a single fleet, which probably represents the best large-scale dataset presently available, but the data quality has been called into question due to the fact that it was derived from fishermen’s logbooks. The indices presented by Campana et al. (2004) cover a smaller area but are in or near Canadian waters and use the most reliable data (observer data). The abundance indices presented by both Campana et al. (2004) and Baum et al. (2003) were consistent with one another despite coverage of only the northwest Atlantic. Overall, the Baum et al. (2003) and Campana et al. (2004) indices are the most representative of the status in Canadian waters. 

 

Table 1: Population Trend Indices for Blue Shark in Canadian and Northwest Atlantic Waters
Study Years Area Overall Trend Comments
ICCAT 2005 1971-2003 (32 yrs) NW Atlantic No overall trend 1971-2003 (increase then decrease)

Up to 60% decline 1987-2003
US and Japanese longline logbook CPUE data. Early years uncertain because of identification issues (Figure 14)
Baum et al. 2003 1986-2000 NW Atlantic 60% decline CPUE from US longline logbook data (Figures 17,18B)
Baum et al. 2003 1986-2000 (14 yrs) Area 6, adjacent to Scotian shelf 64% decline (Figure 18A)
Baum et al. 2003 1986-2000 Area 7, adjacent to Grand Banks No overall trend (increase, then decrease - 9.6% decline start to end) (Figures 17,18B)
Campana et al. (2004) 1995-2003 (8 yrs) Newfoundland and Scotian shelf 53% decline Commercial catch rates (Figure 19)
Campana et al. (2004) 1998-2003 (5 yrs) Canadian waters 27% decline Recreational catch rate (Figure 21)


Pacific Population Abundance and Trends

There has been little research on blue sharks in Canada’s Pacific waters. There are two Canadian-based studies; one which involved assessing bycatch of blue sharks in an experimental Canadian high seas squid driftnet fishery (1985-1987) (McKinnell and Seki 1998) and the other consisting of a single four-day research trip done through a private company in 1991 (IEC Collaborative Marine Research and Development Limited 1992). Neither of these studies provided information on trends or abundance, but some limited biological information was obtained. All other blue shark data in Canada’s Pacific waters are from catch databases and from observations recorded during various research surveys. The low encounter rate found in fishery operations and research surveys does not provide sufficient data to allow for analysis of population trends.


Pacific Population Trends-International Waters

Blue sharks are commonly caught by international pelagic fisheries throughout the North Pacific, but there have been few attempts to assess trends. Assessments in the North Pacific are constrained by under-reporting of catches and/or grouping of all sharks together as a single category. Nakano (1996) used standardized shark CPUE (~70% assumed to be blue shark) from Japanese commercial logbooks (1971-1993) and concluded that there was a 20% decline of CPUE in the North Pacific. The most comprehensive assessment for blue shark in the North Pacific was undertaken by Klieber et al. (2001) who reported that the MSY for blue sharks in the Pacific was 1.8-4 times greater than current estimated fishing mortality in the North Pacific and concluded that the blue shark population appears to be in no danger of collapse.


Pacific Population Trends-Canadian Waters

There are numerous ongoing research surveys undertaken in Canadian waters. These surveys are not designed for sampling pelagic species and consequently only provide limited data. Since 1993, the International Pacific Halibut Commission (IPHC) has been recording non-halibut species in their annual surveys along the west coast of North America. Data are available for blue sharks from 1998 to 2004 as prior to this all sharks were aggregated into a single category. The IPHC changed their survey protocol beginning in 2003 and therefore inter-annual comparisons for rarely caught species require some interpretation. From 1998-2004 a total of 170 blue sharks were recorded in the Canadian portion of the IPHC survey. Of the 171 survey stations in Canadian waters, 68 have a record of a blue shark since 1998 (Figure 7). These surveys clearly show that most of the encounters take place in the southern portion of Hecate Strait. Blue shark catch rates since 1998 have varied between 0.06-0.93 sharks/1 000 hooks (Figure 23). Catch rates in the 2004 survey were the highest on record which also coincides with anecdotal reports from tourism and other marine operators of more encounters of blue sharks possibly in response to an El Niño event.


Figure 23: Catch Rate of Blue Shark in IPHC Set Surveys in Area 2B (Canada) from 1998-2004

Figure 23: Catch rate of blue shark in IPHC set surveys in Area 2B from 1998-2004

Numbers above bars represent the number of blue sharks caught. Data source: IPHC set survey database.


The NMFS Triennial trawl survey has conducted tows off southwest Vancouver Island on a triennial basis since 1981. There are no records of blue shark from Canadian waters and only four records in total from U.S. waters (database accessed by authors).

A variety of groundfish and shrimp survey databases undertaken by DFO from 1979 to present were examined for blue shark records. In total 12 records consisting of 14 blue sharks were found (Appendix 3).

The Canadian experimental high seas flying squid driftnet fishery occurred between 1979-1987 but blue shark data were only collected for three years (1985-1987) and biological data for only 1987 (McKinnell and Seki 1998). It should be noted that most of the fishery took place in the eastern North Pacific outside of Canada’s Exclusive Economic Zone (EEZ). An important finding from this study was that the CPUE was nearly tenfold greater than that observed in the Japanese central North Pacific driftnet fishery (McKinnell and Seki 1998) indicating that blue sharks are relatively abundant in high seas waters near Canada.

In summary, there is insufficient information to estimate abundance or trends in Canada’s Pacific waters. The hook and line fleet appears to be the largest source of blue shark mortality on Canada’s Pacific coast although catch mortality is low (see Fishery section above). The distribution of the catch and the IPHC survey records indicate that blue sharks appear to be found throughout the continental shelf with a concentration closer to the slope break (Figures 6 & 7). A single research trip conducted in 1991 caught 134 sharks in a period of four days which suggests that blue sharks can be caught quite readily in Canadian waters if they are targeted (IEC Collaborative Marine Research and Development Limited 1992).


Rescue Effect

Canadian waters represent a small portion of blue shark habitat in both the Pacific and Atlantic oceans, reproduction does not occur in Canadian waters and there is good evidence that blue sharks are highly migratory. Thus, the population status of blue shark and the potential for rescue in Canadian waters is completely dependent on the status of the hemispheric populations. If the ocean basin-wide population increases, so should the relative abundance in Canadian waters.

Page details

Date modified: