Patagonian Toothfish, *Dissostichus eleginoides* Smitt, 1898.

Patagonian Toothfish, Dissostichus eleginoides Smitt, 1898.



Map of the management areas within the CAMLR Convention Area. The region discussed in this report is shaded in green. Throughout this report, “2020” refers to the 2019/20 CCAMLR fishing season (from 1 December 2019 to 30 November 2020).

Map of the management areas within the CAMLR Convention Area. The region discussed in this report is shaded in green. Throughout this report, “2020” refers to the 2019/20 CCAMLR fishing season (from 1 December 2019 to 30 November 2020).



1. Introduction to the fishery

1.1. History

This fishery report describes the licensed fishery for Patagonian toothfish (Dissostichus eleginoides) in the area of the Australian Fishing Zone (AFZ) in Division 58.5.2. The area includes the AFZ surrounding Heard Island and McDonald Islands, is located on the Kerguelen Plateau between 50\(^{\circ}\)–56\(^{\circ}\)S and 67\(^{\circ}\)–79\(^{\circ}\)E.

The fishery began in 1997 as a trawl fishery. Longline fishing was introduced in 2003 and both fishing methods continued to be used, with an increasing proportion of longline fishing in each year. Since 2015 almost the entire catch has been taken by longline.

The fishery is managed by the Australian Fisheries Management Authority (AFMA) in accordance with the Conservation Measures adopted by CCAMLR and Australian law. The annual catch limit is based on the management advice from CCAMLR. The current catch limits on the fishery for Dissostichus spp. in Division 58.5.2 are described in Conservation Measure 41-08.


1.2. Conservation Measures currently in force

The limits on the fishery for D. eleginoides in Division 58.5.2 are defined in Conservation Measure 41-08.

Figure 1: Map of the region discussed in this report.

Figure 1: Map of the region discussed in this report.


1.3. Active vessels

In 2020, 5 vessels participated in this fishery.


2. Reported catch

2.1. Latest reports and limits

Reported catches of Dissostichus eleginoides are shown in Table 1. In this fishery, the catch of D. eleginoides reached a maximum of 4267 tonnes in 2015. In 2020, 3014 tonnes of D. eleginoides were caught.


Table 1. Catch and effort history for Dissostichus eleginoides in this fishery. Source: Fine scale data and past estimates for IUU catch (-: no IUU estimate available).
Season Longline Catch (tonnes) Trawl Catch (tonnes) Pot Catch (tonnes) Total Catch (tonnes) Number of vessels Catch limit (tonnes) Estimated IUU catch (tonnes)
1997 1811 1811 2 3800 7117
1998 2966 2966 3 3700 4150
1999 3341 3341 2 3690 427
2000 3030 3030 2 3585 1154
2001 2599 2599 2 2995 2004
2002 2514 2514 2 2815 3489
2003 286 2468 2754 3 2879 1274
2004 554 2327 2882 4 2873 531
2005 665 2266 2931 3 2787 265
2006 662 1769 72 2503 4 2584 74
2007 624 1714 2338 2 2427 0
2008 835 1445 2280 3 2500 0
2009 1164 1155 13 2332 3 2500 0
2010 1237 1135 31 2404 3 2550 0
2011 1381 1104 32 2517 3 2550
2012 1369 1302 2671 3 2730
2013 2149 563 41 2753 4 2730
2014 2646 107 2754 4 2730
2015 4062 205 4267 7 4410
2016 2624 158 2783 4 3405
2017 3345 24 3369 4 3405
2018 3083 53 3136 4 3525
2019 3334 68 3402 5 3525
2020 2895 119 3014 5 3030



2.2. By-catch

A number of Conservation Measures, which ensure that impacts on the target and other species are minimised, currently apply to this fishery. Conservation Measure 33-02 specifies that there should be no directed fishing other than for the target species, the by-catch limits for incidentally caught species and the move-on rules if the limits for any one haul are exceeded.

Catch limits for by-catch species groups (macrourids, skates (Rajids) and other species) are defined in Conservation Measure 33-02 and provided in Tables 2 and 3.

A quantitative risk assessment of the Caml grenadier (Macrourus caml) was undertaken in 2015 and WG-FSA-15 recommended a catch limit of 409 tonnes for M. caml and Whitson’s grenadier (M. whitsoni) combined based on the risk assessment in WG-FSA-15/63, and a catch limit of 360 tonnes for bigeye grenadier (M. holotrachys) and ridge-scaled grenadier (M. carinatus) combined based on the previous assessment from 2003. These by catch limits were introduced in 2016 and are reflected in Table 2.


Table 2. Reported catch and catch limits in tonnes for by-catch of Macrourids in this fishery (see Conservation Measure 33-02 for details). Source: fine-scale data.
Macrouridae
M. caml and M. whitsoni
M. holotrachys and M. carinatus
Season Catch Limit Longline Catch Trawl Catch Total Catch Catch Limit Longline Catch Trawl Catch Total Catch Catch Limit Longline Catch Trawl Catch Total Catch
1997
<1 <1
1998
<1 <1
1999
<1 <1
2000
4 4
2001
1 1
2002 50 3 3
2003 465 3 1 5
2004 360 42 3 45
2005 360 72 2 74
2006 360 26 <1 27
2007 360 61 5 66
2008 360 81 5 86
2009 360 110 2 112
2010 360 100 3 102
2011 360 147 4 151
2012 360 89 3 92
2013 360 154 3 157
2014 360 175 1 176
2015 360 299 4 303
2016
409 78 1 80 360 220 220
2017
409 89 <1 90 360 235 <1 235
2018
409 100 4 104 360 253 <1 253
2019
409 101 4 105 360 250 <1 250
2020
409 48 <1 48 360 59 59

An analysis of the by-catch species unicorn icefish (Channichthys rhinoceratus) and grey rockcod (Lepidonotothen squamifrons) indicated that both species are widespread over the plateau in depths of <1,000m (WG-FSA-15/50). Up to 2015, the catch limits of C. rhinoceratus and L. squamifrons, 150 tonnes and 80 tonnes respectively, were based on assessments carried out in 1998 (SC CAMLR-XVII, Annex 5). Catches of each of these species since 2004 have been well below the limits set by CCAMLR (Table 3). A quantitative risk assessment of C. rhinoceratus was undertaken in 2015 and WG-FSA-15 recommended a by-catch limit of 1,663 tonnes for C. rhinoceratus.

Table 3. Reported catch and catch limits in tonnes for by-catch (Rajids, C. rhinoceratus, L. squamifrons and other species) in this fishery (see Conservation Measure 33-02 for details). Source: fine-scale data.
Rajids
C. rhinoceratus
L. squamifrons
Other species
Season Catch Limit Longline Catch Trawl Catch Total Catch Number Released Catch Limit Longline Catch Trawl Catch Total Catch Catch Limit Longline Catch Trawl Catch Total Catch Catch Limit Longline Catch Trawl Catch Total Catch
1997
2 2 0
0 <1 <1
<1 <1
7 7
1998 120 2 2 0
0 <1 <1
<1 <1
31 31
1999
2 2 0
0
<1 <1
5 5
2000
6 6 0
0 <1 <1
<1 <1
12 12
2001 50 4 4 0
0 <1 <1
3 3
113 113
2002 50 3 3 0
0 1 1
1 1
55 55
2003 120 5 7 13 0
0 <1 <1
<1 <1 <1
9 13 21
2004 120 62 11 73 155 150 0 1 1 80 3 3 50 107 59 166
2005 120 70 3 73 8412 150 0 2 2 80 2 2 50 144 9 153
2006 120 19 12 31 3814 150 0 3 3 80 <1 5 5 50 46 19 65
2007 120 8 10 18 7882 150 0 12 12 80 <1 10 10 50 70 18 87
2008 120 13 8 21 9155 150 0 29 29 80 20 20 50 94 21 116
2009 120 15 9 24 10290 150 0 46 46 80 26 26 50 130 14 145
2010 120 11 6 17 10382 150 0 26 26 80 48 48 50 114 10 124
2011 120 11 3 14 6838 150 0 23 23 80 26 26 50 163 8 172
2012 120 7 3 9 8484 150 0 42 42 80 34 34 50 99 12 111
2013 120 13 11 24 12602 150 0 25 25 80 <1 44 44 50 172 72 244
2014 120 16 <1 16 19565 150 0 <1 <1 80 <1 2 2 50 196 2 198
2015 120 19 5 24 37863 150 0 1 1 80 2 2 50 344 10 354
2016 120 20 1 22 32287 1663 0 9 9 80 <1 3 3 50 331 18 349
2017 120 30 2 31 43848 1663 0 2 2 80 <1 2 2 50 371 18 389
2018 120 21 1 23 31187 1663 0 2 2 80 <1 4 4 50 387 7 394
2019 120 25 <1 25 47657 1663 0 2 2 80 <1 <1 1 50 390 7 398
2020 120 6 <1 6 20769 1663 0 <1 <1 80 <1 4 4 50 121 4 125


Length-weight relationships, length-at-maturity data and estimates of abundance from survey data for rajids were presented in WG-FSA-05/70. An analysis of the skate tagging program (WG-FSA-13/22) indicated a recapture rate of <1% and an average distance between release and recapture of 4 nautical miles. An analysis of catch rates from 1997 to 2014 of the three skate species (Nowara et al., 2017) shows a decrease in the average total length of Eaton’s skate (Bathyraja eatonii), but little evidence of depletion on the main trawl grounds. One of the skate species Kerguelen sandpaper skate (B. irrasa), shows a slight decline in catch rates in the deeper waters around Heard Island and McDonald Islands where the longline fishery operates. This study also calculated a growth rate of ca. 20mm per year, and a maximum age >20 years for B. eatonii, as estimated from tag returns.

2.3. Vulnerable marine ecosystems (VMEs)

Fishing gear deployed on the seabed can have negative effects on sensitive benthic communities. The potential impacts of fishing gear on the benthic communities in Division 58.5.2 are limited by the small size and number of commercial trawl grounds and the protection of large representative areas of sensitive benthic habitats from direct effects of fishing within the Heard Island and McDonald Islands Marine Reserve, an IUCN Category 1a reserve where fishing is prohibited (SC-CAMLR-XXI/BG/18). The marine reserve covers a total area of 71,000 km2.

By-catch of benthos has been monitored by observers since the early stages of the development of the fishery and the rate of benthos by-catch is generally lower in areas that have subsequently become the main fishing grounds as opposed to locations sampled in the Random Stratified Trawl Survey. As CM 22-06 does not apply to this subarea there are no CCAMLR VMEs or VME Risk Areas designated in Division 58.5.2.


2.4. Incidental mortality of seabirds and marine mammals

Seabird mortality rates during longline operations in this fishery remains low (WG-FSA-19/31); The three most common species injured or killed in the fishery were Cape petrel (Daption capense), white-chinned petrel (Procellaria aequinoctialis) and grey petrel (P. cinerea) (Table 4).

The level of risk of incidental mortality of birds in Division 58.5.2 is category 4 (average-to-high) (SC-CAMLR-XXX, Annex 8, paragraph 8.1). Longline fishing is conducted in accordance with Conservation Measures 24-02 and 25-02 for the protection of birds so that hook lines sink beyond the reach of birds as soon as possible after being put in the water. Between them, these measures specify the weight requirements for different longline configurations and the use of streamer lines and a bird exclusion device to discourage birds from accessing the bait during setting and hauling. A core fishing season and season extensions are specified in Conservation Measure 41-08. If three seabirds are caught during the season extension by a given vessel, fishing during the season extension is to cease immediately for that vessel.

Table 4. Number of reported birds caught (killed or with injuries likely to substantially reduce long-term survival) in this fishery in each fishing season.
Season Daption capense Procellaria aequinoctialis Procellaria cinerea Other
1998 2
2003 5
2004 2
2005 1
2009 1
2010 2
2012 2
2013 1
2014 1
2015 1
2016 1 2
2017 1
2018 1 1
2019 3
2020 3 1


Mammal mortalities reported in the longline fishery in Division 58.5.2 (Table 5) mainly consist of southern elephant seal (Mirounga leonina).

Low levels of sperm whale depredation have been observed in Division 58.5.2 since 2011 (WG-FSA-15/53). Sperm whale sightings occur exclusively in the April-June period.

Conservation Measure 25-03 is in force to minimise the incidental mortality of birds and mammals during trawl fishing. Measures include developing gear configurations which minimise the chance of birds encountering the net, and the prohibition of discharge of offal and discards during the shooting and hauling of trawl gear.

Table 5. Number of reported mammals killed in this fishery in each fishing season.
Season Arctocephalus gazella Mirounga leonina Otaria byronia Otariidae, Phocidae Phocidae
1998 2
2003 1 3 1
2004 2 1
2005 1 1
2006 1 2
2007 1 1
2008 1 1
2009 2
2012 1
2013 5
2014 1 1
2015 2 2
2016 7 2
2017 4 2
2018 5
2019 3
2020 4

3. Illegal, Unreported and Unregulated (IUU) fishing

No illegal, unreported and unregulated (IUU)-listed vessels were sighted in Division 58.5.2 inside the Heard Island and McDonald Islands exclusive economic zone (EEZ) since 2006. However, surveillance reports indicate that IUU fishing activities did occur in Division 58.5.2 outside the Heard Island and McDonald Islands EEZ, and therefore brief fishing forays into the EEZ cannot be discounted. IUU fishing gear was also recovered in 2006 and 2011, indicating IUU fishing activities have potentially occurred in the region. Information from satellite surveillance trials indicated the presence of unidentified vessels in this division outside the Heard Island and McDonald Islands EEZ in 2016. In May 2017, a section of gillnet was recovered during fishing operations in Division 58.5.2. Following the recognition of methodological issues in its assessment, no estimates of the IUU catch of Dissostichus spp. have been provided since 2011 (SC CAMLR-XXIX, paragraph 6.5).


4. Data collection

4.1. Data collection requirements

The collection of biological data as part of the CCAMLR Scheme of International Scientific Observation (SISO) includes representative samples of length, weight, sex and maturity stage, as well as collection of otoliths for age determination of the target and most frequently taken by-catch species. Data are collected during commercial fishing trips and during Random Stratified Trawl Surveys (RSTS). The surveys cover a geographic area over the whole of the plateau shallower than 1,000 m in Division 58.5.2 to determine abundance of D. eleginoides. These surveys have been conducted since 1990 with survey designs described in detail in WG-FSA-06/44 Rev. 1 and in WG-FSA-19/03 for the 2019 survey.


4.2. Length frequency distributions

Dissostichus eleginoides occurs throughout the Heard Island and McDonald Islands area of the Kerguelen Plateau in Division 58.5.2, from shallow depths near Heard Island to at least 3,000 m depth around the periphery of the plateau. Fish smaller than 60cm total length (TL) are predominantly distributed on the plateau in depths less than 500m, where a small number of areas of persistently high local abundance have been discovered. As fish grow, they move to deeper waters and are recruited to the fishery on the plateau slopes in depths of 450 to 800m where they are vulnerable to trawling. Some areas of high local abundance comprise the main trawling grounds where the majority of fish caught are between 50 and 75cm Total Length. Larger fish are seldom caught by trawling and there is evidence from tag recaptures and size distribution of the catch by depth that fish, as they grow, move into deeper water (>1,000m depth) where they are caught by longline.

The length frequency distributions of D. eleginoides caught by trawl and by longline in Division 58.5.2 are shown in Figures 2 and 3 respectively. Since the start of the fishery >500,000 fish have been measured in this division.


Figure 2. Annual length frequency distributions of *D. eleginoides* caught by trawl in this fishery. The number of hauls from which fish were measured (N) and the number of fish measured (n) in each year are indicated. Note: length frequency distributions are only shown where more than 150 fish were measured.

Figure 2. Annual length frequency distributions of D. eleginoides caught by trawl in this fishery. The number of hauls from which fish were measured (N) and the number of fish measured (n) in each year are indicated. Note: length frequency distributions are only shown where more than 150 fish were measured.


Figure 3. Annual length frequency distributions of *D. eleginoides* caught by longline in this fishery. The number of hauls from which fish were measured (N) and the number of fish measured (n) in each year are indicated. Note: length frequency distributions are only shown where more than 150 fish were measured.

Figure 3. Annual length frequency distributions of D. eleginoides caught by longline in this fishery. The number of hauls from which fish were measured (N) and the number of fish measured (n) in each year are indicated. Note: length frequency distributions are only shown where more than 150 fish were measured.


The majority of D. eleginoides caught by trawl measured between 25 and 100cm with a mode around 50-60cm, while those caught by longline measured between 50 and 125cm with a mode around 75cm. The length frequency distribution for the longline fishery includes larger fish because of gear selectivity and because the longline fishery occurs in deeper water where larger toothfish occur. These length frequency distributions are unweighted; they have not been adjusted for factors such as the size of the catches from which they were collected. The interannual variability exhibited in the figure may reflect changes in the fished population but is also likely to reflect changes in the gear used, the number of vessels in the fishery and the spatial and temporal distributions of fishing.

4.3. Tagging

A tagging study has been undertaken in Division 58.5.2 since the start of the commercial fishery in 1998.

To date, 70495 D. eleginoides have been tagged and released (12367 have been recaptured, 11434 of which were released in this area; Table 6).

Table 6. Recent numbers of Dissostichus eleginoides tagged and recaptured in the area for each fishing Season.
Recaptured
Season Tagged 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Total
1998 1073 72 66 24 10 10 4 2 1 1 2 1 193
1999 757 56 71 19 2 1 1 1 1 1 153
2000 1777 125 101 66 12 8 2 1 1 1 1 318
2001 1599 199 94 48 14 2 1 1 1 1 1 362
2002 1534 255 149 41 12 4 1 2 1 1 466
2003 1576 169 124 24 18 2 6 2 2 3 1 3 1 1 1 357
2004 1562 287 135 25 10 8 7 2 5 2 3 7 3 494
2005 1701 266 88 16 5 9 8 4 3 5 6 11 2 1 1 425
2006 2430 220 179 51 26 13 11 12 19 9 11 4 7 1 1 1 565
2007 1841 200 120 35 21 13 6 12 10 13 9 3 2 1 445
2008 1741 50 61 25 14 9 31 20 25 10 6 5 1 2 259
2009 2423 89 100 52 15 28 40 51 14 28 5 8 7 437
2010 1768 55 65 14 18 55 37 10 20 14 13 8 309
2011 2398 124 150 54 46 46 32 36 37 23 9 557
2012 2986 161 124 53 48 43 62 46 28 17 582
2013 2002 31 58 99 46 54 57 23 22 390
2014 2126 12 87 64 95 50 50 36 394
2015 8345 85 285 386 311 250 171 1488
2016 5955 54 254 319 262 154 1043
2017 6916 67 367 396 316 1146
2018 6168 58 423 201 682
2019 6798 114 218 332
2020 5019 37 37
Total 70495 11434

Historically, the tagging program had been largely restricted to releases and recaptures of fish caught by trawl on the main trawl ground (WG-FSA-14/43). Tagging data from the main trawl ground were used to estimate natural mortality independently of the CASAL assessment as described in Candy et al. (2011), while the limited spatial extent of the program and mixing of the population to other areas initially restricted the ability to include tagging data as an unbiased index of abundance in the stock assessment. With the start of longlining in 2003, tagging and recapturing of fish has become more widespread. However, the spatial distribution of longline fishing and tagging of fish has been highly variable between years and the level of fish movement and the period of complete mixing is still unknown. Tagging data have been included into the stock assessment since 2014 to inform stock abundance.

5. Research

During late March to mid April 2020, the annual random stratified trawl survey (RSTS) around Heard Island and McDonald Islands (HIMI) was conducted in CCAMLR Division 58.5.2 (SC-CAMLR-39/BG/35), with the completion of 151 stations. The survey was conducted on the FV Atlas Cove. Sampling protocols such as the design and the duration of the hauls were similar to recent surveys, but with a new set of randomly selected station points. However, only 18 of the 30 stations allocated in Plateau Deep East could be sampled due to damage to the trawl warps which prevented fishing on deeper stations.

The calculated biomass for 2020 of the target species D. eleginoides and C. gunnari in the survey area were the highest estimates for the past 10 years. Biomass estimates for the managed by-catch species C. rhinoceratus and Macrourus spp. remained at a high level and the estimate for L. squamifrons showed the first substantial increase since 2014. Among the three species of skate, biomass estimates show an upward trend over the last few years with of Bathyraja murrayi also being at the highest levels for the past 10 years.

Length measurements and sex were taken for nearly 17,000 fish and for more than half of those, biological measurements were also recorded. Otoliths were collected from D. eleginoides (763) and a number of other species, and 645 toothfish were tagged and released.

In 2019, catch removals due to killer and sperm whale interactions across subantarctic fisheries were estimated (WG-FSA-19/33).

6. Stock status

6.1. Summary of current status

The 2019 assessment model (WG-FSA-19/32) lead to a smaller estimate of the virgin spawning stock biomass B0 than that obtained in 2017, with an MCMC estimate of 70,519 tonnes (95% CI: 65,634-76,626 tonnes). The estimated SSB status at the end of 2019 was 0.51 (95% CI: 0.49-0.53).


6.2. Assessment method

The assessment model in 2019 was a single-sex, single-area, age-structured CASAL integrated stock assessment model (WG-FSA-19/32).


6.3. Year of last assessment, year of next assessment

Assessments are reviewed biennially, the last assessment was in 2019.


7. Climate Change and environmental variability

A recent summary of the potential impacts of climate change on Southern Ocean fisheries (FAO 2018) highlights the following key points:

The Antarctic region is characterized by complex interaction of natural climate variability and anthropogenic climate change that produce high levels of variability in both physical and biological systems, including impacts on key fishery taxa such as Antarctic krill.

The impact of anthropogenic climate change in the short-term could be expected to be related to changes in sea ice and physical access to fishing grounds, whereas longer-term implications are likely to include changes in ecosystem productivity affecting target stocks.

There are no resident human populations or fishery-dependent livelihoods in the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) Area, therefore climate change will have limited direct implications for regional food security.

The institutional and management approach taken by CCAMLR, including the ecosystem-based approach, the establishment of large marine protected areas, and scientific monitoring programmes, provides measures of resilience to climate change.

There is no formal evaluation of the impacts of climate change and environmental variability available for this particular fishery.


Additional Resources

References

Candy, S.G., D.C. Welsford, T. Lamb, J.J. Verdouw and J.J. Hutchins. 2011. Estimation of natural mortality for the Patagonian toothfish at Heard and McDonald Islands using catch-at-age and aged mark-recapture data from the main trawl ground. CCAMLR Science, 18: 29-45.

Francis, R.I.C.C. 2011a. Data weighting in statistical fisheries stock assessment models. Can. J. Fish. Aquat. Sci., 68: 1124-1138.

Francis, R.I.C.C. 2011b. Corrigendum: Data weighting in statistical fisheries stock assessment models. Can. J. Fish. Aquat. Sci., 68: 2228.

Nowara, G.B., P. Burch, N. Gasco, D.C. Welsford, T.D. Lamb, C. Chazeau, G. Duhamel, P. Pruvost, S. Wotherspoon and S. Candy. 2017. Distribution and abundance of skates (Bathyraja spp.) on the Kerguelen Plateau through the lens of the toothfish fisheries. Fish. Res., 186: 65-81.