Pelagic fish

Pelagic fish

In 2020 like in 2019, 0-group fish biomasses were dominated by capelin. In 2020, polar cod and redfish biomasses were also high. Biomasses of most fish species (except redfish and haddock) would be even higher if the whole area had been covered in 2020.

Total biomass

Figure 3.6.1. Biomass of 0-group fish species in the Barents Sea, August–October 1993–2020.

Abundance and biomass estimates were calculated by different software: SAS (for the new 23 fisheries subareas, 1980-2016), MatLab (for the new 15 subareas (Fig. 3.6.2, 1980-2018), StoX (for the new 15 subareas (Fig. 3.6.2, 2016-2020). Due to software upgrading (led to challenges with script running in SAS), personal resource limitation (MatLab), no control of input data and lower estimates (comparing to SAS and MatLab), we decided to developed R-scripts (R is free software) for estimation of abundance and biomass indices. The 2019 and 2020 abundance and biomass indices were calculated by R and presented here, while abundance and biomass indices for other years will be presented later.

Figure 3.6.2. Map showing subdivision of the Barents Sea into 15 subareas (regions) used to calculate estimates of 0-group abundance and fish length based on the BESS (more description in ICES 2018).

Capelin, young herring (for age 1-4), and polar cod constitute the bulk of pelagic fish biomass in the Barents Sea. During some years (e.g., 2004–2007 and 2015–2016), blue whiting (Micromesistius poutassou) also had relatively high biomass in the western Barents Sea (east of the continental slope). Total biomass of the main pelagic species during 1986–2020 fluctuated between 0.5 and 9 million tonnes; largely driven by fluctuations in the capelin stock. During 2017-2018, the cumulative biomass of capelin, herring, polar cod, and blue whiting was close to the long-term average (Fig. 3.6.3). In 2019, the total biomass of pelagic fish in the Barents Sea was at its lowest level over the past 23 years, but the biomass increased considerably from 2019 to 2020 due to strong 2019 year-classes of capelin and polar cod. The significant increase in NSS herring biomass was driven by the growth of the 2016 year class.

Figure 3.6.3. Total biomass of pelagic fish component (excluding 0-group) in the Barents Sea in 1986-2020.

Capelin

Young of the year

In 2019, a record strong year-class of capelin occurred. Estimated abundance of 0-group capelin varied from 2.082 billion in 1993 to 1 911 billion individuals in 2019 with a long-term average of 448 billion individuals for the 1980-2020 period (Fig. 3.6.4). In 2020, the eastern Barents Sea was not covered, where 0-group capelin were often found, and thus abundance and biomass indices were underestimated. Based on the average long-term distribution in 2000-2017, we corrected the 2018 and 2020 abundance indices for lacking coverage. In 2020, the total abundance index for 0-group capelin was well above the long term mean and was 1 265 billion individuals (Fig. 3.6.1). Estimated biomass of 0-group capelin was four times higher than the long term mean and was 697 thousand tonnes. Therefore, the 2020 year-class of capelin seemed to be strong.

Figure 3.6.4. 0-group capelin abundance estimates and fluctuation 1980-2020. Orange dotted line shows the long-term average; the blue columns indicate fluctuating abundance; orange columns indicate corrected indices. Note that estimates were calculated for the new 15 subareas in the Barents Sea 2019 and 2020 in R. Abundance indices for capelin in 2018 and 2020 were underestimated due to lack of coverage in the eastern Barents Sea.

The highest average abundance per strata was found in Hopen Deep (202 billion ind.) and Central bank (108 billion ind.). Most of capelin were relatively large with body length of 5-6.4 cm in 2020 comparing to 4-5.4 cm in 2019. Larger individuals were found mainly in northern areas, while smaller in southwestern areas (South West and Bear Island Trench).

Adult capelin

Due to the delayed start of the survey in the east, the main area of capelin SSB distribution during 2020 was well covered, but the total distribution area (including immature fish) was not covered synoptically. The geographical distribution of capelin recorded acoustically is shown in Fig. 3.6.7. Capelin distribution area in 2020 was the same as in 2019, but capelin concentrations were much larger than in 2019. The main concentrations were found to the southwest of Svalbard (Spitsbergen) between 76°N and 78°N, which is historically the most typical distribution area for feeding capelin at this time of the year. Quite a lot of young capelin were also distributed south of Hopen Island to 75°N. Little capelin was found in the east and in the northern areas, but an aggregation of young capelin was found west of the coast of Novaya Zemlya a little north of 76°N.

Average weight at age was higher than previous years for age groups 2, 3 and 4 and significantly lower for age 1 (Fig. 3.6.8). This could be related to the high abundance of 1year old capelin. The weight at age for 2 and 3 were among the highest on record.

Dynamics of changing average weight-at-age reflect capelin feeding conditions during the summer-autumn period. These conditions are determined not only by the stock size, but also by the state of the plankton community in the Barents Sea. It is evident that in 2020 the capelin food base (zooplankton abundance and species composition) was stable.

Figure 3.6.7 Geographic distribution of capelin in 2019 (top panel) and 2020 (bottom panel). Circle size corresponds to SA (area back-scattering coefficient) values per nautical mile. The red dashed line marks the area which had been covered by the time of the preliminary capelin assessment before survey done.

Figure 3.6.8 Biological characteristics of capelin during August-September (1972-2020).

The total stock was estimated to about 1.88 million tons, which is below the long-term average level (2.8 million tons), but 4 times higher than the biomass estimate from 2019 (Fig. 3.6.10). About 28 % (0.53 million tons) of the 2020 stock had length above 14 cm and was therefore considered to be maturing. 1year old capelin (2019 year-class) completely dominated in the capelin stock, and the biomass and numbers of age 1 was the highest since 2000 (Fig. 3.6.9). This agrees well with the results of the 0-group survey in 2019 when capelin generation was estimated as very abundant.

Age 2 capelin (2018year-class) amounted to 31.1 billion ind., only 7.7% of total stock by number, it also shows that in 2019 capelin at the age of 1 (2018 age class) was somewhat underestimated. The 2017year-class (age 3) made up 1.0% of the stock by number.

Thus, in 2020, the maturing stock (MSB) of Barents Sea capelin is at a low level and significantly below long-term average. The total stock (TSB) is slightly below the long-term average, but the largest since 2013 (Fig. 3.6.10).

Figure 3.6.9. Capelin stock age composition (age 1-4) during 1972–2020. (Note: age 5 and older was removed due to negligible numbers in the total stock).

Figure 3.6.10. Capelin biomass based on 1972–2020 acoustic survey data: maturing stock biomass, total stock biomass and long term mean.

Due to near total spawning mortality, the natural mortality of capelin can be estimated indirectly only. Since fishing mortality for ages 1 and 2 is absent or very small, it can be assumed that total mortality for age groups is natural. Fig. 3.6.11 shown natural mortality (M) calculated as the decrease from age 1 to age 2 in the autumn survey. In some years negative mortality values were obtained. It is most likely the consequence of underestimation of age 1 fish in the survey.

Figure 3.6.11. Capelin natural mortality from age 1 to age 2, estimates based on acoustic survey data. (Negative mortality has been removed).

Spatial distribution of capelin in the Barents Sea depends on environmental and stock conditions, primarily: position of the ice edge; distribution of zooplankton; and capelin stock size and structure (Ingvaldsen and Gjøsæter 2013). In years with a large stock, capelin is distributed widely. Juvenile capelin is distributed further south than adults. During the 1972-1979 period, the capelin stock was large and widely distributed. During 1980-1989, the stock decreased, and distribution was more southward. Since the 2000s, capelin began movement north- and eastwards. During 2010-2017, the stock was in good condition and moved significantly northward into ice-free waters (Fig. 3.6.11). This represented a shift northward an average of 60-80 nautical miles further than observed in the 1970s. During 2018-2019, capelin stock size has decreased; the area of distribution has decreased as well (Fig. 3.6.12). In 2020, the capelin stock began to increase due to large recruitment, but the capelin moved further northeast than north. In general, during periods of warming in the Barents Sea, capelin move further north and north-eastward to find feeding grounds with high plankton biomass. However, at low stock levels, capelin have adequate food availability, and temperature does not appear to be a key factor driving northward expansion.

Figure 3.6.12a. Estimated capelin biomass during August-September by decade (1970s, 1980s, 1990s, 2000s, and 2010s). Biomasses presented for World Meteorological Organization (WMO) squares system of geocodes which divide areas into latitude-longitude grids (1° latitude by 2° longitude). One dot is equal to 500 tonnes.

Figure 3.6.12b. Estimated capelin biomass during August-September for recent periods of high temperature condition and cod stock size. Note that cod abundance peaked in 2013 and that temperature decreased in 2018-2019. Time periods are further broken down into sub-periods (2004-2009, 2010-2014 and 2015-2017 and 2018-2020). Biomass is presented for WMO squares. One dot is equal to 500 tonnes

Herring

Young of the year

Estimated abundance of 0-group herring varied from 0.093 billion in 1986 to 940.773 billion individuals in 2004 with a long-term average of 188.65 billion individuals for the 1980-2020 period (Fig. 3.6.13). In 2020, the eastern Barents Sea was not covered, where 0-group herring were also found, and thus abundance and biomass indices were underestimated. In 2020, the total abundance index for 0-group herring was well below the long term mean and was 22.251 billion individuals (Fig. 3.6.13). Based on the average long-term distribution in 2000-2017, we corrected the 2018 and 2020 abundance indices for lacking coverage. The corrected abundance index was somewhat higher than without correction and was 25.015 billion individuals in 2020. Estimated biomass of 0-group herring was half of 2019, less than 1/20 of the long term mean and was 5.25 thousand tonnes. Therefore, the 2020 year-class of herring seemed to be weak.

Figure 3.6.13. 0-group herring abundance estimates and fluctuation 1980-2020. Orange line shows the long-term average; the blue columns indicate fluctuating abundance. Note that estimates were calculated for the new 15 subareas in the Barents Sea 2019 and 2020 in R. Abundance indices for herring in 2018 and 2020 were underestimated due to lack of coverage in the eastern Barents Sea.

Most of herring (77%) were distributed in the south western areas (Bear Island Trench and South West) of Barents Sea. Most of 0-group herring were relatively small with body length of 5.5 - 7.5 cm, comparing to slightly larger individuals in 2019 (4.5- 6.5 cm). Larger individuals were observed west of Svalbard (Spitsbergen) archipelago, while in the southern, central, and northern Barents Sea fish length varied and were most likely depended on feeing condition in the area.

Herring age 1-2

Figure 3.6.16 shows biomass estimates of age 1 and 2 herring combined in the Barents Sea based on the last ICES assessment for age 2+ herring, assuming M=0.9 for age 1. During 2013–2017, abundance of young herring in the Barents Sea was relatively stable. It increased from 2017 to 2018 mainly due to contribution of the strong 2016year-class, and then decreased to a lower level in 2019-2020. Figure 3.6.17 shows herring distribution in 2019-2020 with highest amounts in the south-west (in particular in 2020) and south-eastern parts of the Barents Sea.

It should be noted that in the herring surveys in the Barents Sea in 2020 age 4 herring dominated. Neither the abundance of age 1-2 as shown in Fig. 3.5.16 nor the abundance estimates from the surveys (June survey and BESS) carried out on young herring give a coherent picture. The estimates from the assessment depend heavily on the assumption of M and also it varies between years whether or not also age 3 herring is present in the Barents Sea. On the other hand, there are not survey data for all years, and they are often inconsistent between years (e.g., the abundance of the 2016- and 2017year-classes in 2020 was higher than the estimates of those year-classes in 2019). Thus, an analysis to determine a time series for young herring abundance in the Barents Sea, taking all data sources into account, should be given high priority.

Figure 3.6.16. Estimated biomass of Norwegian Spring Spawning herring Age 1 and 2 in the Barents Sea – based on Working Group on Widely Distributed Stocks (WGWIDE) VPA estimates (ICES 2020b).

Figure 3.6.17. Estimated distribution of herring, August-October in 2019 (top panel) and 2020 (bottom panel). Circle sizes correspond to SA (area back-scattering coefficient) averaged over 1 nautical mile.

Polar cod

Polar cod is an Arctic species with a circumpolar distribution. Historically, the world’s largest population of this species has been observed in the Barents Sea. In recent years, there have been significant fluctuation of the polar cod stock size.

Young of the year

Estimated abundance of 0-group polar cod varied from 0.519 billion in 1995 to 2 428 billion individuals in 1994 with a long-term average of 440 billion individuals for the 1980-2020 period (Fig. 3.6.4). In 2018 and 2020, the eastern Barents Sea was not covered, where 0-group polar cod were often found, and thus abundance and biomass indices were underestimated. However, in 2020, estimated biomass of 0-group polar cod was high and was 2.5 time higher than the long term mean, although only the western component has been covered. The eastern component usually has been substantially dominated in abundance and biomass and therefore it will be difficult to use long term mean distribution for correction of indices and shown below correction can be more uncertainty then for other species. In 2020, corrected abundance index was highest since 2002.

Figure 3.6.18. 0-group polar cod abundance estimates and fluctuation 1980-2020. Orange line shows the long-term average; the blue columns indicate fluctuating abundance; orange columns indicate corrected indices. Note that estimates were calculated for the new 15 subareas in the Barents Sea 2019 and 2020 in R. Abundance indices for polar cod in 2018 and 2020 were underestimated due to lack of coverage in the eastern Barents Sea.

In 2020, the distribution area of 0-group polar cod increased significantly in the north western areas compared to previous years. Polar cod biomasses were largest in Svalbard North subarea. Most of 0-group polar cod were relatively large with body length of 5.5 - 6.5 cm, and slightly larger than in 2019 (4.5- 5.5 cm).

Since 2015, the proportion of 0-group polar cod in the southeast of the Barents Sea (Pechora region) has been decreasing (Eriksen et al. 2017). Low abundance of 0-group cod in the traditional core area, the Pechora Sea, most likely due to redistribution of spawning sites out of the Barents Sea and into the western part of Kara Sea. This is indirectly confirmed by 2019-2020 studies in the Kara Sea, where a significant amount of the mature polar cod were found.

Adult polar cod

From 2012 total abundance and biomass of polar cod in the Barents Sea has decreased significantly. Only the strong year-class of 2015 gave a short-term increase the polar cod stock in 2016 (Fig. 3.6.21), then the stock quickly decreased again. In 2020, the area of polar cod distribution was covered much better than in previous years. The assessment of the polar cod in 2020 in the Barents Sea was complete and synoptic. The main distribution of polar cod was found in the north-eastern parts of the survey area around Franz Josef bank which is typical (Fig. 3.6.22), but densities were much higher than recorded in recent years. Polar cod were also abundant west of 35°E, to south-east of Svalbard (Spitsbergen), which has not been observed in recent years. Overall, the polar cod distribution was much wider than in 2019.

The total stock was estimated to be 1721 thousand tonnes. The 1-year-olds completely dominated and constituted 80% of the estimated total abundance by numbers and 58% by biomass. The abundance of 1-year-olds was the highest on record, and also the abundance of the other age groups was above average. In 2020, the polar cod biomass estimate was the highest since 2006 and one of the highest on record. However, it should be noted that the survey area in the northeast part of Barents Sea has not been well surveyed in 2018-2019 and in several previous years.

The issue of the situation with stocks of polar cod in the Barents Sea and adjacent areas remains open. The polar cod stock variation may be the result of natural mortality due to cod consumption. Assuming that a significant part of the polar cod stock migrated outside the Barents Sea, why did some of the polar cod remain in the Barents Sea? It is obvious that polar cod populations of the Barents and Kara Seas are related. The distribution of the 0-group polar cod in the Barents Sea shows that it is very likely that the 0-group polar cod from spawning grounds in the south-west part of Kara Sea is brought into the Barents Sea through the Kara strait and then lives in the Barents Sea. Thus, the polar cod population in the Barents Sea may be replenished by “Kara Sea recruits”. The last investigation in the Kara Sea (Fig. 3.6.23) shows that the TSB of polar cod increased and was estimated at 304 thousand tonnes (Anon, 2019) and the total number of fish aged 3years and more was 78%.

Figure 3.6.21. Total abundance in billions (coloured bars / left axis) and biomass in millions of tonnes (green line / right axis) of polar cod in the Barents Sea (acoustic survey and BESS data) collected August-September during the 1986–2020 period. (2003 values based on VPA due to poor survey coverage. A reliable estimate is not available for 2018).

Figure 3.6.22. Estimated distribution of polar cod during August–October in 2019 (top panel) and 2020 (bottom panel). Circle size corresponds to SA (area back-scattering coefficient) values per nautical mile.

Figure 3.6.23. Estimated distribution of polar cod in the Kara sea. Survey RV “Professor Levanidov” 15.09-29.09.2019. Background area colour correspond to SA (area back-scattering coefficient) averaged over 1 nautical mile.

Blue whiting

Acoustic estimates for the proportion of the blue whiting stock present in the Barents Sea have been made since 2004. In 2017, the BESS data time-series were recalculated using a newer target strength equation (Pedersen et al., 2011), and a standardized area. The revised estimates were on average about one third of previous estimates. During 2004–2007, estimated biomass of blue whiting in the Barents Sea was >200 000 tonnes (Fig. 3.6.24) but decreased abruptly in 2008 and remained low until 2012. In 2012 and 2013 the strong 2011 year-class contributed to an observed increased abundance of blue whiting and in 2015 and 2016 the even stronger 2014 year-class contributed largely to the total estimated biomasses >150 000 tons in 2015 and 2016 (Fig. 3.6.25). With strong year-classes the young blue whiting is abundant along the shelf break to the Norwegian Sea and partly distribute into the Barents Sea. In 2018-2020 the blue whiting abundance in the Barents Sea was very low.

Figure 3.6.24. Total abundance in billions (coloured bars / left axis) and biomass in millions of tonnes (green line / right axis) of blue whiting in the Barents Sea (BESS data revised in 2017) collected August–September during the 2004–2020 period.

Figure 3.6.25. Estimated distribution of blue whiting during August-October 2019 (top panel) and 2020 (bottom panel). Circle size corresponds to SA (area back-scattering coefficient) averaged over 1 nautical mile.