Atlantic wolffish

Key signals

Biomass indices from stock surveys are high, fishable biomass peaked in 2022-23.
The recruitment index (<40 cm) dropped to low levels in SMB in 2011, but has been increasing since.
Fishing pressure on the stock is below F_msy and F_pa. Spawning-stock is above MGT B_trigger, B_pa and B_lim.
Data from surveys and fisheries indicate that Atlantic wolffish in 5a is currently in good condition. This is also confirmed by the stock assessment.
MFRI recommends a continued closure of the most important spawning area west of Iceland during the spawning and incubation season in autumn and winter.

General information

The Atlantic wolffish is a long-bodied species with a large head, characterized by prominent canine teeth for seizing prey and powerful molars for crushing it. Most individuals caught measure between 50 and 80 cm in length, although the largest specimen recorded in Icelandic waters reached 125 cm. The species is most commonly found on the continental shelf northwest of Iceland. Its feeding grounds are typically located on muddy or sandy bottoms at depths shallower than 100 meters. In contrast, spawning grounds are found on coarser substrates featuring crevices or cavities, generally at depths greater than 100 meters.

Fishery

The primary fishing grounds for Atlantic wolffish are located on the western and northwestern parts of the Icelandic continental shelf. Since 2010, the proportion of the catch taken in the northwest has been increasing, while it has declined in the west. Catches at the main spawning ground, Látragrunn (located west of Iceland), have been decreasing since 2008 (Figure 1 and Figure 2). Notably, part of this area has been closed to trawling during the spawning and incubation period since 2010. Approximately 80% of the Atlantic wolffish catch is taken at depths shallower than 120 meters. The proportion of the catch from the 0–60 m depth range declined between 2003 and 2007 but has been increasing since. In the 61–120 m depth range, the catch proportion has remained relatively stable since 2000. At depths of 121–180 m—which includes the main spawning ground at Látragrunn—the catch proportion increased between 2003 and 2008 but has declined since then (Figure 3).

Atlantic wolffish in mainly caught by longliners, demersal trawlers and seiners (Figure 4). Proportions in each gear have been changing in most recent years with demersal seiners increasing their share, resulting in roughly one third of the Atlantic wolffish catch being caught by that fleet (Figure 4). Since 2001, the number of vessels reporting Atlantic wolffish catches of 10 tonnes/year or more has decreased. Most of the decrease is in the longline fleet, where the number of vessels has dropped from 198 in 2001, down to 39 in 2024. The number of trawlers has been variable but not changed much, and the number in 2024 is the same as in 2002. Number of demersal seiners has increased in the period 2000-2024 (Table 1).

In 1994 and 1995, more than 500 vessels accounted for 95% of the annual catch of Atlantic wolffish in Icelandic waters, but this number had dropped to 200 vessels in 2008 despite higher catches. Since 2010 the number of vessels accounting for 95% of the annual catch has remained relatively constant (about 150-200 vessels), despite catch reductions (Figure 5).

Figure 1: Atlantic wolffish. Geographical distribution of the Icelandic fishery over the past 9 years, based on reported logbook catch data.

Figure 2: Atlantic wolffish. Spatial distribution of the Icelandic fishery by fishing area since 2000 according to logbooks. All gears combined.

Figure 3: Atlantic wolffish. Depth distribution of demersal trawl, longline and demersal seine catches according to logbooks.

Figure 4: Atlantic wolffish. Total catch (landings) by fishing gear since 1994, according to statistics from the Directorate of Fisheries.

Table 1: Atlantic wolffish. Number of Icelandic boats and catches by fleet segment participating in the Atlantic wolffish fishery from logbooks.

Year	nr. longlines	nr. trawlers	nr. demersal seiners	nr. other	Longline	Demersal trawl	Demersal seine	Total catch
2000	161	53	16	0	8 687	3 380	528	12 595
2001	195	59	14	1	11 267	3 391	513	15 182
2002	140	44	13	0	7 773	3 735	601	12 110
2003	138	45	19	0	7 785	5 463	1 066	14 313
2004	103	34	29	0	4 670	4 773	1 609	11 052
2005	91	47	24	1	5 445	6 893	1 140	13 508
2006	120	48	25	0	6 626	6 286	1 149	14 061
2007	105	60	24	0	5 259	7 566	1 338	14 163
2008	87	60	22	3	4 663	6 960	1 427	13 093
2009	114	55	28	1	6 708	5 468	1 205	13 391
2010	74	46	20	3	5 916	4 436	842	11 286
2011	64	37	18	0	5 344	3 565	1 010	9 918
2012	66	24	22	1	5 328	2 827	895	9 091
2013	73	30	18	2	4 652	2 341	647	7 662
2014	70	23	13	1	3 681	1 637	891	6 220
2015	56	34	17	2	3 989	1 905	926	6 848
2016	61	37	18	2	4 848	1 662	1 127	7 661
2017	59	28	18	2	3 829	1 102	1 095	6 049
2018	60	37	27	6	4 923	1 587	2 186	8 770
2019	64	34	21	1	4 595	1 630	2 168	8 404
2020	46	38	24	1	2 491	2 046	2 040	6 588
2021	45	48	22	0	3 343	3 021	2 086	8 451
2022	40	48	23	0	2 706	2 986	2 275	7 967
2023	36	49	20	0	3 154	2 756	2 388	8 298
2024	39	45	24	0	2 612	2 914	2 900	8 426

Figure 5: Atlantic wolffish. Number of vessels (all gear types) accounting for 95% of the total catch annually since 1994. Left: Plotted against year. Right: Plotted against total catch. Data from the Directorate of Fisheries.

Data available

The commercial catch samples taken should be representative of the landings with most number of samples taken in areas of high catch intensity. In general, sampling from commercial catches is considered acceptable from the main gears (demersal seines and trawls). The sampling seems to follow the spatial and seasonal distribution of catches (Figure 7 and Figure 6).

Figure 6: Ratio of samples by month (bars) compared with proportion landings by month (solid black line) split by year and main gear types. Numbers of above the bars indicate number of samples by year, month and gear.

Figure 7: Atlantic wolffish. Spatial distribution of length samples (crosses) from commercial catches in Icelandic waters.

Landings and discards

Landings by Icelandic vessels are obtained from the Icelandic Directorate of Fisheries. Landings of Norwegian and Faroese vessels are given by the Icelandic Coast Guard. Discarding is banned by law in the Icelandic demersal fishery, as well as in Norway. Measures in the Icelandic management system such as converting quota from one species to another are used by the Icelandic fleet to a large extent, and is thought to reduce discards in mixed fisheries.

Sampling from commercial catches

The number of length and age samples from landed Atlantic wolffish was reduced in 2014 (Table 2 and Table 3). Prior to that, approximately 1600–3000 fish were sampled annually for age readings and about 6000–14000 were measured for length. Between 2015 and 2021, otoliths were collected annually from an average of just over 1200 wolffish. In 2024, a total of 10, 29 and 26 samples were collected from longline, bottom trawl and demersal seine catches, respectively (Table 3, Figure 7).

Table 2: Atlantic wolfish. Number of samples and length measured fish from landed catch of Atlantic wolffish.

Year	Bottom Trawl		Demersal Seine		Long Line
Year	Num. samples	Num. lengths	Num. samples	Num. lengths	Num. samples	Num. lengths
2000	22	2 852	4	468	29	3 698
2001	13	1 806	3	376	33	4 147
2002	37	4 912	2	281	44	5 151
2003	36	4 270	7	1 251	42	6 433
2004	33	3 932	12	1 505	23	3 241
2005	57	7 732	16	1 684	33	5 089
2006	43	5 829	7	754	44	5 882
2007	44	5 935	17	1 839	33	3 963
2008	64	9 903	15	1 357	24	3 351
2009	52	8 470	14	1 674	33	4 464
2010	54	9 640	13	1 666	30	3 911
2011	20	3 015	12	1 391	17	2 524
2012	41	9 684	12	1 206	26	3 574
2013	20	3 175	5	671	26	3 711
2014	31	4 296	31	3 623	30	3 748
2015	20	2 358	20	1 984	26	3 168
2016	13	1 596	10	970	26	2 806
2017	9	956	6	633	24	2 973
2018	9	817	17	1 687	22	2 654
2019	12	1 128	23	2 349	23	2 677
2020	18	1 833	17	1 652	9	1 007
2021	25	2 987	15	1 070	14	1 277
2022	23	2 420	17	1 235	3	269
2023	22	2 365	25	2 203	12	1 039
2024	33	3 646	26	2 075	10	1 071

Table 3: Atlantic wolffish. Number of samples and aged fish from landed catch of Atlantic wolffish.

Year	Longline		Bottom trawl		Demersal seine
Year	num. samples	num. otoliths.	num. samples	num. otoliths.	num. samples	num. otoliths.
2000	29	1 395	18	752	4	200
2001	27	1 343	11	509	3	150
2002	25	1 240	16	645	2	100
2003	31	1 525	20	899	6	300
2004	19	950	23	1 060	8	400
2005	15	746	25	1 202	6	292
2006	23	1 110	21	1 029	5	250
2007	18	900	25	1 250	10	451
2008	19	950	25	1 248	4	200
2009	16	800	20	999	4	200
2010	29	1 669	19	1 090	5	285
2011	14	750	15	778	9	550
2012	26	1 300	14	700	7	350
2013	25	1 249	14	691	4	200
2014	30	800	26	675	28	700
2015	25	625	19	479	19	474
2016	25	625	13	325	9	225
2017	23	575	9	220	6	150
2018	22	550	9	225	17	425
2019	22	550	11	276	20	500
2020	9	225	14	350	16	400
2021	14	350	25	625	15	375
2022	3	60	23	465	17	338
2023	12	240	21	420	25	499
2024	10	200	29	580	26	508

Length composition

The length distribution of landed Atlantic wolffish has been relatively stable since 2005 in all gears (Figure 8).

Figure 8: Atlantic wolffish. Length distribution from the Icelandic fishing fleet since 2005

Age composition

Commercial age data is available since 1980s. In samples from commercial landings, the mean age of Atlantic wolffish was around 10.7 years in 1999. Since then, mean age from commercial catches has generally been increasing to around 12 years in recent years. In 2024, 9-14 year old fish are most common in the catches (Figure 9 and Figure 10).

Figure 9: Atlantic wolffish. Catch at age from the commercial fishery in Iceland waters. Bar size is indicative of the catch in numbers and bars are coloured by cohort. Scale of the y-axis varies between plots.

Figure 10: Atlantic wolffish. Catch at age from the commercial fishery in Iceland waters. Biomass caught by year and age; bars are coloured by cohort.

Weight-at-age

Mean weight at age in commercial catches in Icelandic waters are available from 1996. Weight of the oldest year classes has been above average for the past years (Figure 11). Catch by year and age is shown in Figure 12. The contribution of the plus group (16+) is often high.

Figure 12: Atlantic wolffish. Catch by year and age from the commercial fishery in Icelandic waters.

Catch and effort

CPUE estimates of Atlantic wolffish in Icelandic waters are not considered representative of stock abundance, as changes in fleet composition, technical improvements and differences in gear setup among other things have not been accounted for when estimating CPUE. Effort of demersal trawl is defined as the number hours towed, for longline number of hooks and demersal seine number of sets. CPUE is calculated based on sets or tows in which Atlantic wolffish was more than 10% of the total catch and sets where Atlantic wolffish was caught. CPUE index for each year and gear is then caught as the median CPUE for the selected sets for each gear and year.

CPUE in longline was relatively stable prior to 2018, fluctuating around 75 kg/1000 hooks but doubled between 2000 and 2023. CPUE of demersal trawl was at it highest in 2005 and 2015 with approximately 270-300 kg/h, between those years it dropped below 200 kg/h and has been around that level since. CPUE of demersal seine has been increasing since and is at it´s highest (Figure 13). In summary we see contrast between CPUE in bottom trawl and CPUE in demersal seine and long line in last 5 years where bottom trawl CPUE is stable but CPUE in the other gear increasing.

Figure 13: Atlantic wolffish. Non-standardized estimates of CPUE for demersal trawl (left, kg/h), demersal seine (middle, kg/sets) and longline (right, kg/1000 hooks).

Survey data

The Icelandic spring groundfish survey (hereafter spring survey, IGFS), which has been conducted annually in March since 1985, In addition, the Icelandic autumn groundfish survey (hereafter autumn survey, IAGS) was started in 1996 and expanded in 2000. However, a full autumn survey was not conducted in 2011 due to a labour strike. Both surveys covers the most important distributional areas of Atlantic wolffish in Icelandic waters but the the spring survey has 3 times as many stations in the main distributional areas. Also, the trawl used in the spring survey is heavier and better suited for catching wolffish. Diurnal variations in catches in the surveys are considerable with less catches during day time.

Indices from the spring survey have relatively narrow confidence interval (Figure 15) but have more variability than can be explained by estimated confidence interval or stock size.

Total biomass and harvestable biomass indices decreased from 1985-1995. In 1996, the biomass index increased to 1998, then decreased to a historical low level in 2010-2012, but since then it has been increasing Figure 15. The harvestable biomass shows increasing trend from 1995 with considerable oscillatons. The recruitment index was high in the years 1992-2003, since 1999 it has been decreasing, which coincides with increasing effort and catch of trawlers at the main spawning ground west of Iceland (Látragrunn) during the spawning and incubation time. The recruitment index reached a historical low level in 2011, but has since then been increasing, though with fluctuations. This coincides with that the closed spawning/incubation area on Látragrunn was enlarged from 500 km2 (from 2002) to 1000 km2 in October 2010.

When the spring survey is conducted, Atlantic wolffish are on their feeding grounds which are commonly in relatively shallow waters. In the spring survey, the highest abundance has always been measured in the NW area (Figure 16). Compared to catches (Figure 2) much higher proportion of the stock is in the NE area in the surveys, especially in the late nineties (fig-surveybyarea).

Length distributions from the surveys have remained stable in most recent years with bimodal distribution in the spring survey (Figure 17). Mean lengths have increased since 2000 in both surveys (Figure 17).

Figure 14: Atlantic wolffish. Location and abundance in the most recent spring survey (SMB) and the autumn survey (SMH).

Figure 15: Atlantic wolffish. Total biomass indices (upper left) and harvestable biomass indices (≥60 cm, upper right), large fish biomass indices (≥80 cm, lower left) and juvenile abundance indices (≤40 cm, lower right), from the spring survey (blue) and the autumn survey (red), along with the standard deviation.

Figure 16: Atlantic wolffish. Spatial distribution of biomass index from the spring and autumn survey.

Figure 17: Atlantic wolffish. Length distribution from the spring and autumn survey.

Figure 18: Atlantic wolffish. Age disaggregated indices in the autumn survey and the spring survey. Fill colours indicate cohorts. Note different scales on y-axes.

Stock weight at age

Mean weight at age in the survey is shown in Figure 19. Stock weights are obtained from the groundfish survey in March and are also used as mean weight at age in the spawning stock. Stock weights show considerable positive trend from 1985-2025.

Figure 19: Atlantic wolffish. Stock weights from the spring survey in Icelandic waters. Bars are coloured by cohort.

Stock maturity and natural mortality

Females have more reliable maturity designations than males, a maturation scale for males is unavailable. Therefore, spawning stock is only based on mature females caught in the autumn survey and in commercial catches from June – December. From these data, maturation occurs close to 60 cm and around age 10 but is highly variable and difficult to measure. Proportion mature at age has increased for the past 20 years for most age groups (Figure 20 and Figure 21).

No information is available on natural mortality. For assessment and advisory purposes, the natural mortality is set to 0.15 for all age groups.

Figure 20: Atlantic wolffish. Maturity at age in the autumn survey and commercial catches in June-December. The values are used to calculate the spawning stock.

Figure 21: Atlantic wolffish. Maturity at age in the autumn survey and commercial catches in June-December. The values are used to calculate the spawning stock.

Analytical assessment

Atlantic wolffish in 5a became part of the ICES assessment process after an MoU between Iceland and ICES was signed in December 1st 2019. Assessment method for Atlantic wolffish was established during benchmark in april 2022 and SAM model (state-space stock assessment model) was agreed upon as an assessment tool.

The new assessment model is a statistical catch at age model based on:

commercial catch-at-age and landings data from 1979 onwards, In the period 1981-1997 the age disaggregated catches are not available but total landings are uses (Figure 9)
the Icelandic spring groundfish survey from 1985 (Figure 18).
the Icelandic autumn groundfish survey in Iceland from 2000 (Figure 18).

The maximum age of the model is 16, which is considered a plus group. The assessment showed that SSB has been rather stable over the time period, while fishing mortality has gradually decreased, and recruitment has slightly decreased after 2001 but remained stable.

Natural mortality of 0.15 was chosen for all age groups. During the workshop, a wide range of estimates for natural mortality were tested and none showed a significant improvement in terms of model fit. It was therefore decided to use a M of 0.15.

Model fit

Figure 22 shows the overall fit to the survey indices and catch. In general, the model appears to follow the stock trends historically but the model fits better to catch and SMB compared to SMH. Furthermore, the terminal estimate is not seen to deviate substantially from the observed value for most age groups. Neither observation nor process residuals show obvious trends (Figure 23 and Figure 24).

Figure 23: Atlantic wolffish. Observation error residuals of the SAM model.

Figure 24: Atlantic wolffish. Process error residuals of the SAM model.

Model results

Model results show that Atlantic wolffish Spawning stock biomass has also shown a steady increase since 1992, part of it due to higher weight and maturiy at age. Indices of biomass >60cm and >80cm do also show increase (Figure 15) for the last 30 years so the increase in SSB in not only caused by increased proportion mature and stocck weights. Recruitment levels have also increased after being at the lowest level in 2011. Larger spawning stock and closures of the main spawning areas contributed to this increase. Landings have decreased from high levels in 1990-2009 and have been relatively stable since 2014 (Figure 25).

Retrospective analysis

The results of an analytical retrospective analysis are presented in Figure 26. The analysis indicates relatively stable estimate, except in the earliest peel. Mohn’s rho was estimated to be -0.0601 for SSB, 0.097 for F, and -0.0454 for recruitment.

Figure 26: Atlantic wolffish. Retrospective plots illustrating stability in model estimates over a 5-year “peel” in data. Results of spawning stock biomass, fishing mortality F, and recruitment (age 4) are shown.

Reference points

As part of the WKICEMP 2022 HCR evaluations (ICES 2022c), the following reference points were defined.

Table 4: Atlantic wolffish. Reference points adopted from ICES WKICEMP 2022

Framework	Reference_point	Value	Technical_basis
MSY Approach	MSY Btrigger	21000	Bpa
	FMSY	0.2	F that produces MSY in the long term

Precautionary Approach	Blim	18500	Bloss (SSB in 2002)
	Bpa	21000	Blim x e1.645 * σB
	Flim	0.33	Fishing mortality that in stochastic equilibrium will result in median SSB at Blim.
	Fpa	0.2	Maximum F at which the probability of SSB falling below Blim is <5%
Management plan	MGT Btrigger	21000	According to the harvest control rule
	FMGT	0.2	According to the harvest control rule

Current Advisory Framework

Reference points were calculated for the stock. This resulted in B\(_{pa}\) of 21 000 t, based on the lowest estimate of SSB observed after the 2001 shift in recruitment had been observed (2002), and B\(_{lim}\) = B\(_{pa} e^{-1.645\sigma_B}\) of 18 500 t, with \(\sigma_B\) being set to the ICES default of 0.2. The fishing pressure estimates, defined in terms of fishing mortality applied to ages from 10 to 15, were estimated in accordance to the ICES guidelines. This resulted in an estimate of F\(_{lim}\) of 0.33, F\(_{p05}\) of 0.20 and F\(_{msy}\) of 0.20. The MSY B\(_{trigger}\) was set as B\(_{pa}\).

A Management Strategy Evaluation (MSE) was conducted for Atlantic wolffish in 5a. The operating model, which generates the “true” future populations in the simulations, was based on equilibrium simulations (eqsim). Selection, maturity and stock weights were based on the resampling of estimates by age from previous 10 years. Recruitment was projected using a log-normal distribution based on the distribution of CVs and autocorrelations estimated by the assessment model with MCMC resampling with a break point in B\(_{lim}\). Advice error in the simulations was implemented as auto-correlated log-normal variations in F, with a CV of 0.212 and \(\rho\) of 0.423.

The proposed HCR for the Icelandic Atlantic wolffish fishery, which sets a TAC for the fishing year y/y+1 (September 1 of year y to August 31 of year y+1) based on a fishing mortality \(F_{mgt}\) of 0.20 applied to ages 10 to 15, modified by the ratio SSB\(_{y}\)/MGT B\(_{trigger}\) when SSB\(_y\) < MGT B\(_{trigger}\), maintains a high yield while being precautionary as it results in lower than 5% probability of SSB < B\(_{lim}\) in the medium and long term.

Management considerations

A reduction in fishing mortality has led to harvestable biomass and SSB slowly increasing. Atlantic wolffish is a slow-growing late-maturing species, therefore closures of known spawning areas should be maintained and expanded if needed. Similarly, closed areas fishing where there is high juvenile abundance should also be maintained and expanded if needed. All available indicators from commercial catch and survey data suggest that the Atlantic wolffish stock in Icelandic waters is currently in good condition. This conclusion is supported by the results of the analytical assessment.

Ecosystem considerations

Most fishing for Atlantic wolffish occurs in the northwest and west of Iceland, where the fastest growing Atlantic wolffish are found. A likely cause for differences in growth is environmental differences between the relatively warm southwestern waters versus colder northeaster waters. The surveys indicate that 25-50% of the stock is in the NE area while less than 10% of the catch is taken there (Figure 16) and (Figure 1)

However, Atlantic wolffish are also highly sedentary, especially while guarding nests during spawning and rearing season, and therefore additional metapopulation structure cannot be excluded. Therefore, it is possible that local depletion may occur in more heavily fished areas despite a stable overall biomass level.

Management

The Ministry of Industries and Innovation is responsible for management of the Icelandic fisheries and implementation of legislation. Atlantic wolffish was included in the ITQ system in the 1996/1997 quota year and as such subjected to TAC limitations. From that time to the fishing year 2004/2005, the catch was on average 5% more than recommended by the MRI, although in some years it was lower than advised TAC. In the fishing years 2005/2006 to 2011/2012, the catch was on average around 34% above the advised TAC (Figure 27). The main reasons were that national TAC was set higher than the advised TAC, and quota of other species were being transferred to Atlantic wolffish quota (Figure 28). Net transfer of Atlantic wolffish quota for each fishing year is usually less than 15%.

Figure 27: Atlantic wolffish. Comparison of advice, quota, and total catch.

Figure 28: Atlantic wolffish. Net transfer of quota to and from Atlantic wolffish in the Icelandic ITQ system by fishing year. Between species (upper): Positive values indicate a transfer of other species to Atlantic wolffish, but negative values indicate a transfer of Atlantic wolffish quota to other species. Between years (lower): Transfer of quota from given quota year to the next quota year.

References

Gunnarsson, Á., Hjörleifsson, E., Thórarinsson, K., Marteinsdóttir, G., 2006. Growth, maturity and fecundity of wolffish Anarhichas lupus L. in Icelandic waters. Journal of Fish Biology, 68, 1158-1176. doi: 10.1111/j.1095- 8649.2006.00990.

Gunnarsson, Á., Sólmundsson, J., Björnsson, H., Sigurðsson, G., Pampoulie, C., 2019. Migration pattern and evidence of homing in Atlantic wolffish (Anarhichas lupus). Fisheries Research, 215. https://doi.org/10.1016/j.fishres.2019.03.001

ICES. 2022. Workshop on the evaluation of assessments and management plans for ling, tusk, plaice and Atlantic wolffish in Icelandic waters (WKICEMP). ICES Scientific Reports. 4:37. 271 pp. http://doi.org/10.17895/ices.pub.19663971