Witch

Glyptocephalus cynoglossus


Technical report
Published by

Marine and Freshwater Research Institute, Iceland

Published

6 June 2025

Key signals

  • Biomass indices from stock surveys remained low until 2003, after which they increased significantly and have remained stable since.

  • Recruitment was poor during the years 2012–2020, with an increase from 2021 in the spring survey and a peak in the autumn survey in 2023.

  • In both surveys and catches, the peak of the length distribution has gradually shifted to the right — toward larger fish — reflecting a period of poor recruitment.

  • Total biomass peaked in 2025 spring survey.

  • Fishing mortality (F) was above FMSY until 2012, but is below that level now.

  • Witch stock in 5a is considered to be in balance.

General information

Witch is found all around Iceland, but the highest concentration is observed in the relatively warm waters south and west of Iceland. It is a demersal flatfish species found at 25-500 m depth but is most common at 50-300 m on a sandy or muddy substrate. Females grow larger than males. Only a small proportion of males become larger than 40 cm, whereas females can get larger than 60 cm. Size at sexual maturity differs between the sexes; about half of the males reached maturity at 25 cm length, but half of the females reached maturity level at 32 cm.

Fishery

The geographical distribution of the witch fisheries has remained more or less unchanged in recent years (Figure 1), with the main fishing grounds in the southwest of Iceland, extending along the south coast in the deeper areas. Very little catch has been reported from the northwest, north and east of Iceland. Witch is common bycatch in the Nephrops fishery. In 2019, one of the main Nephrops fishing areas, Lónsdjúp, was closed for all trawling (for both Nephrops and demersal trawls) to protect young Nephrops. There was also a closure for Nephrops trawling in Jökuldjúp (area already closed for all demersal trawling) and demersal trawling in Breiðamerkurdjúp and Hornafjarðardjúp (MFRI 2021). Those closures can influence the distribution of the witch catch.

Since 2000, the main fishing grounds of witch have been in the southern and western part of the Icelandic shelf (Figure 2) according to logbook entries. Spatial distribution of the Icelandic witch fishery is considerably stable, with around 50% of the witch caught on the south-western part of the shelf.

Of the combined catch in demersal seine and Nephrops trawl, about 80-90% of witch was caught at 100-200 m depth in most years (Figure 3). In 2011-2016, that proportion had declined to about ~60% while the proportion of the catch taken at 50-100 m depth increased. This was solely due to increase in demersal seine effort at that depth range. Most of the catch in demersal seine was taken at 100-150 m and at 150-200 m depth in Nephrops trawl.

Witch on Icelandic fishing grounds is mainly caught in demersal seine and Nephrops trawl, or approximately 95% of all reported landings (Figure 4 and Table 1). This proportion has been a relatively stable throughout the years until Nephrops fishery was closed in 2022. In years 2009-2018 the proportion of witch landed by seiners decreased and reported landings from Nephrops trawlers increased. The drop in proportion of witch landed by Nephrops trawl since 2019 is most likely due to area closures in the SE and W areas mentioned previously.

The number of vessels accounting for 95% of the total catch of witch in Icelandic waters decreased from about 80 vessels in 1996-1997 to about 30 vessels in 2002, despite annual catches being at similar levels (Figure 5). In 2002-2015, the number of vessels accounting for 95% of the catches (ranging from 1200-2300 tonnes) remained relatively constant. In the last three years only about 30 vessels have accounted for 95% of the annual catch of witch.

Figure 1: Witch in 5a. Spatial distribution of catches by all gears according to Icelandic logbooks.
Figure 2: Witch in 5a. Changes in spatial distribution of witch catches as recorded in Icelandic logbooks.
Figure 3: Witch in 5a. Depth distribution of witch from Nephrops and demersal trawl and seine according to Icelandic logbooks.
Figure 4: Witch in 5a. Landings in tonnes and percent of total by gear and year.
Table 1: Witch in 5a. Number of Icelandic vessels landing witch, and all landed catch divided by gear type.
Year Nr. Demersal seine Nr. Bottom trawl Nr. Nephrops trawl Nr. Other Demersal seine Bottom trawl Nephrops trawl Other Total Catch
2000 57 76 27 26 870 160 70 6 1 106
2001 45 66 32 42 920 66 150 4 1 140
2002 41 52 33 26 876 34 240 1 1 151
2003 48 40 29 21 1 690 30 228 0 1 948
2004 55 48 28 16 1 732 57 333 0 2 122
2005 52 51 29 18 1 967 111 246 1 2 324
2006 50 44 25 13 1 738 130 162 0 2 030
2007 51 38 18 22 1 531 113 159 2 1 805
2008 47 41 18 19 1 165 102 158 1 1 427
2009 49 41 16 24 1 233 121 437 2 1 792
2010 46 36 17 15 733 75 514 3 1 325
2011 41 35 17 28 621 98 601 1 1 321
2012 44 35 17 33 686 52 570 6 1 315
2013 41 28 15 34 653 39 456 15 1 163
2014 34 26 16 29 650 74 443 14 1 181
2015 31 30 13 25 650 107 564 11 1 332
2016 26 29 11 13 507 118 289 10 924
2017 29 27 9 10 640 62 308 1 1 012
2018 30 33 9 10 502 61 304 1 867
2019 27 31 8 13 584 92 204 0 881
2020 28 38 8 11 638 92 216 0 946
2021 26 38 7 14 412 95 146 0 654
2022 31 38 3 8 592 73 5 0 670
2023 28 40 0 13 599 51 0 1 651
2024 29 37 1 11 451 65 3 0 519
Figure 5: Witch in 5a. Number of vessels (all gear types) accounting for 95% of the total catch annually since 1994. Left: Plotted against year from 1994. Right: Plotted against total catch. Data from the Directorate of Fisheries.

Catch per unit effort (CPUE) and effort

CPUE estimates of witch 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.

CPUE in demersal seine sets (kg/set) in where witch was more than 10% of the catch have fluctuated between 200 and 550 kg/set, with highest value in 2022. CPUE for witch in Nephrops trawl (kg/h), in hauls where witch is more than 10% of the catch, fluctuated from 50 to 75 kg/hour in 2009-2021, but no fishing for Nephrops was allowed since 2022 (Figure 6).

Figure 6: Witch in 5a. Catch per unit of effort in the most important gear types. The dashed lines are based on locations where more than 10% of the catch is witch and solid lines on all records where witch is caught.

Natural mortality

No information is available on natural mortality.

Sampling overview

Sampling of biological data from main gears (demersal seine and bottom trawl) in commercial catches is considered generally acceptable. The sampling cover the spatial and temporal distribution of catches, from main gears, to a satisfactory extent. The sampling coverage by gear in 2024 is shown in Figure 7 and overview of the number of samples is shown in Figure 8.

Table 3 and Table 2 show length and otolith sampling from the commercial witch catch from 2000-2024.

Figure 7: Witch in 5a. Fishing grounds in 2024 as reported in Icelandic logbooks (colours) and positions of samples taken from landings (asterisks) by main gear types.
Figure 8: Witch in 5a. Ratio of samples by month (bars) compared with landings by month (solid black line) split by year and main gear types. Numbers above the bars indicate sample size.

Length distribution of landed catch

Over the past eight years, there has been a shift towards larger fish in the relative length distribution of landed catch (Figure 9): the average length in the samples taken from commercial catch has increased from 37 cm to 42 cm in the past two decades. Few smaller fish have been seen entering the fishery over the past 5 years.

The bycatch of witch in the Nephrops fishery was poorly reported early on. Before 2003, less than 50% of witch landings from Nephrops trawl were reported in logbooks. Comparison of the length composition of the witch catch in the Nephrops survey to the catch from the Nephrops trawlers indicated some discarding of the smaller witch in earlier years.

Table 2: Witch in 5a. Number of samples and length measurements from landed catch.
Year
Bottom Trawl
Demersal Seine
Nephrops Trawl
Num. samples Num. lengths Num. samples Num. lengths Num. samples Num. lengths
2000 31 4 307 36 5 204 10 1 377
2001 6 893 39 5 284 7 914
2002 3 451 38 5 978 25 3 513
2003 1 150 65 9 402 13 1 926
2004 5 749 101 14 993 29 4 287
2005 12 1 630 94 12 233 16 2 276
2006 7 1 002 131 18 426 13 1 870
2007 3 326 102 12 337 11 1 537
2008 4 600 84 10 732 13 2 131
2009 11 1 641 100 13 382 37 5 495
2010 7 1 039 50 7 193 54 8 264
2011 5 765 39 5 346 58 8 633
2012 1 150 49 7 193 53 8 187
2013 4 469 40 5 451 32 5 118
2014 3 375 29 3 780 19 2 495
2015 2 249 37 4 641 24 2 990
2016 3 375 21 2 691 10 1 242
2017 5 622 31 3 848 12 1 500
2018 4 500 21 2 570 8 1 000
2019 3 374 22 2 831 8 1 000
2020 4 500 15 1 875 5 611
2021 4 500 10 1 210 6 1 112
2022 3 336 21 2 311 0 0
2023 3 360 19 2 065 0 0
2024 1 120 11 1 140 0 0
Figure 9: Witch in 5a. Commercial length distributions by gear and year

Age composition of landed catch

In 2002-2008, most of the witch catch was 4-7 years old (Figure 10). The proportion of these age classes has since decreased and shift towards 8-10 year old was noticeable in 2016-2018. Thus, witch in the catch has become older, and there are little signs of recruitment of younger fish into the fishery.

Table 3: Witch in 5a. Number of samples and otoliths collected from landed catch.
Year Sampling_BMT n_BMT Sampling_DSE n_DSE
2000 31 1 541 36 1 800
2001 6 300 38 1 848
2002 3 150 38 2 050
2003 1 50 59 2 924
2004 5 250 93 4 651
2005 7 350 68 3 400
2006 6 299 104 5 198
2007 1 50 83 4 132
2008 4 200 74 3 700
2009 11 550 80 4 000
2010 7 350 45 2 239
2011 3 150 38 1 900
2012 1 50 46 2 300
2013 3 150 39 1 950
2014 3 75 26 650
2015 1 25 35 875
2016 3 75 20 500
2017 5 123 30 750
2018 4 100 19 475
2019 3 75 18 450
2020 4 100 15 375
2021 4 100 8 200
2022 3 65 19 380
2023 3 60 19 380
2024 1 20 11 220
Figure 10: Witch in 5a. Catch at age from the commercial fishery in Iceland waters. Bar size is indicative of the catch in numbers and bars are colored by cohort.

Survey data

The Icelandic spring groundfish survey (hereafter spring survey, SMB), which has been conducted annually in March since 1985, covers the most important distribution area of the witch fishery. In addition, the Icelandic autumn groundfish survey (hereafter autumn survey, SMH) was commenced in 1996. The autumn survey was not conducted in 2011. The spring survey is considered to measure changes in abundance/biomass better than the autumn survey. It does not, however, adequately cover the main recruitment grounds for witch that are poorly known.

Figure 11 shows trends in various biomass indices and a recruitment index based on abundance of witch smaller than 30 cm in the spring and autumn surveys. Abundance and changes in spatial distribution are shown in Figure 12, Figure 13 and survey length-disaggregated abundance indices in Figure 14.

In both surveys, total biomass indices and biomass indices for witch larger than 30 cm (harvestable part of the stock) increased rapidly in 2004 (Figure 11) and have remained relatively high and stable since. Biomass of large fish (43 cm and larger) increased rapidly from 2010-2015 and remained at that level since. Recruitment indices (numbers below 20 cm) have decreased rapidly from quite high numbers in the beginning of the spring survey 1985 and stayed at that level through the period with small recruitment pulses occasionally. In 2011-2019, the recruitment indices from both surveys were at an all-time low. Recruitment increased in the spring survey in 2021 and is at a similar level this year.

Witch is caught all around Iceland in the spring survey, but only in very small quantities off the east coast ( Figure 12; Figure 13). Abundance of witch is highest and quite stable in the SW and W areas. Abundance in the SE area is quite low except for years 2002-2012. In the northern areas abundance has increased in 2008 and remained stable since. The autumn survey shows a similar trend (Figure 12;Figure 13).

Relative length distribution of witch in the spring survey has shifted towards larger fish (Figure 14). The average length of witch increased from 31 cm in 1988 to 38-39 cm in 2016-2023. Data from the autumn survey supports these trends, with a marked increase in average size of witch 31 cm in 2000 to 36.5 in 2024.

Figure 11: Witch in 5a. Indices in the spring survey (SMB) 1985 and onwards (line shaded area) and the autumn survey (SMH) (point ranges).
Figure 12: Witch in 5a. Location of witch in the most recent spring (SMB) and autumn (SMH) surveys, bubble sizes are relative to catch sizes.
Figure 13: Witch in 5a. Changes in geographical distribution of the survey biomass.