Formularios en html
The stock of Blue whiting is on the observation list.
Blue whiting (Micromesistius poutassou) is a pelagic gadoid that is widely distributed in the eastern part of the North Atlantic. The highest concentrations are found along the edge of the continental shelf in areas west of the British Isles and on the Rockall Bank plateau where it occurs in large schools at depths ranging between 300 and 600 meter but is also present in almost all other management areas between the Barents Sea and the Strait of Gibraltar and west to the Irminger Sea. The major spawning takes place in February and March, along the shelf edge and banks west of the British Isles, but in recent years a large fishery has developed between the Rockall-Hatton Plateau. Juveniles are abundant in many areas, with the main nursery area believed to be the Norwegian Sea. Morphological, physiological, and genetic research has suggested that there may be several components of the stock which mix in the spawning area west of the British Isles. Due to the large population size, its considerable migratory capabilities and wide spatial distribution, much remains to be understood regarding the stock composition and dynamics. Accurate estimates of the stock size are difficult to obtain and the management of this species provides therefore a challenge. However, for assessment purposes blue whiting in these areas is treated as a single stock since it has so far not been possible to define an unambiguous border between populations (if separate populations exist).
An almost threefold increase in the spawning stock of blue whiting stock since mid–1990s has raised a series of questions on the biology of blue whiting and possible effects of the environment on the productive capacity of the stock. In last years report (ICES 2006/ACFM:34) it was shown that the increase in temperature and salinity in a confined study area west of the Porcupine Bank in 1997 and onwards coincides with the increase in recruitment of blue whiting.
The main source water masses that enter the Rockall Trough from the south are the Western North Atlantic Water (WNAW) carried by the North Atlantic Current (NAC) and the Eastern North Atlantic Water (ENAW), which is driven from the ‘intergyre region’ between the Azores and the Bay of Biscay. The ENAW is relatively warm and saline compared to the other water masses in the Rockall region. Hátún et al. (2005) studied the hydrographic conditions in the area west of the British Isles and found that after 1996 it was dominated the ENAW, thus giving favourable conditions for spawning over a relatively wide area. However, it remains to be shown whether there is a causal relationship between hydrographic conditions and recruitment of blue whiting, although a tentative hypothesis was suggested by Hátún et al. (2007).
In 2005 ICES stated that fishing within the limits of the existing management plan (F=Fpa) implies catches of less than 1.5 million t in 2006. This will also result in a high probability that the spawning stock biomass in 2007 will be above Bpa. The present fishing level is well above levels defined by the management plan and should be reduced. The Management Plan Point 4 calls for a reduction in the catch of juvenile blue whiting which has not taken place. The primary approach to reducing catch of juveniles is to reduce overall fishing mortality. Catches of juveniles in the last 4 years are much greater than in earlier periods. If an overall reduction of fishing mortality cannot be achieved then specific measures should be taken to protect juveniles.
In 2006 ICES stated that the maximum catch in 2007 corresponding to a new agreed management plan is 1.9 million tonnes, which is expected to leave the spawning stock biomass at 2.86 million t, i.e. above Bpa in 2008, but will lead to an F above Flim in 2007.
The current fishing mortality, estimated at 0.48, is above the fishing mortalities that are expected to lead to high long-term yields and low risk of depletion of production potential.
Fishing at Fpa implies catches of less than 980 thousand t in 2007. This is expected to result in a spawning stock biomass in 2008 well above Bpa.
The newly agreed management plan had been evaluated by ICES and was not considered in accordance with the precautionary approach. ICES concludes that the exploitation boundaries for this stock should be based on the precautionary limits.
1.1.4.1 Coastal States management plan
In October 2006, the coastal states (EU, Norway, Iceland and Faroe Islands) agreed on a sharing arrangement for the blue whiting stock. This arrangement provides for catches in 2007 of 1 700 000 tonnes, allocated as follows: EU 30.5%, Faroe Islands 26.125%, Norway 25.745% and Iceland 17.63%. Russia will be accommodated by transfers from some of the coastal states and additional catches in the NEAFC regulatory area.
1 ) A Delegation of the Faroe Islands, a Delegation of the European Community, a Delegation of Iceland, and a Delegation of Norway met in Tórshavn on 26 and 27 October 2006 to consult on the management of the blue whiting stock in the North-East Atlantic in 2007.
2 ) The Delegations recognised that the basis for management measures in 2007 is the Agreed Record of Conclusions of Fisheries Consultations on the Management of Blue Whiting in the Northeast Atlantic concluded in Oslo on 16 December 2005 (the 2005 Agreed Record), including its Annex I and Annex II.
3 ) In accordance with Annex II, Paragraph 4 of the 2005 Agreed Record, the Delegations agreed to reduce their total allowable catch of blue whiting in 2007 by 300 000 tonnes.
4 ) In accordance with Paragraphs 5 and 6 of the 2005 Agreed Record, the Delegations agreed to recommend to their respective authorities the arrangement for the regulation of the fisheries of blue whiting in 2007 as contained in Annex I to this Agreed Record.
ANNEX_I. ARRANGEMENT FOR THE REGULATION OF THE FISHERIES OF BLUE WHITING IN 2007
1 ) In accordance with the multi-annual management arrangement for the fisheries of blue whiting set out in Annex II to the 2005 Agreed Record, the Parties agree to restrict their fisheries of blue whiting in 2007 to a maximum catch limit of 1 700 000 tonnes on the basis of the following quotas:
1.1 ) European Community 518 500 tonnes
1.2 ) Faroe Islands 444 125 tonnes
1.3 ) Iceland 299 710 tonnes
1.4 ) Norway 437 665 tonnes
2 ) Each Party may transfer unutilised quantities of up to 10% of the quota allocated to it for 2007 to 2008. Such transfer shall be in addition to the quota allocated to the Party concerned for 2008.
3 ) In the event of over-fishing of the allocated quotas by any Party in 2007, the quantity shall be deducted from the quota allocated in 2008 for the Party or Parties concerned.
4 ) The Parties may fish blue whiting within the quotas laid down in Paragraph 1 in their respective zones of fisheries jurisdiction and in international waters.
5 ) Further arrangements by the Parties, including arrangements for access, quota transfers and other conditions for fishing in the respective zones of fisheries jurisdiction, are regulated by bilateral arrangements.
ANNEX II. ARRANGEMENT FOR THE MULTI-ANNUAL MANAGEMENT OF THE BLUE WHITING STOCK
1 ) The Parties agree to implement a multi-annual management arrangement for the fisheries on the blue whiting stock which is consistent with the precautionary approach, aiming at constraining harvest within safe biological limits, protecting juveniles, and designed to provide for sustainable fisheries and a greater potential yield, in accordance with advice from ICES.
2 ) The management targets are to maintain the Spawning Stock Biomass (SSB) of the blue whiting stock at levels above 1.5 million tonnes (Blim) and the fishing mortality rates at levels of no more than 0.32 (Fpa) for appropriate age groups as defined by ICES.
3 ) For 2006, the Parties agree to limit their fisheries of blue whiting to a total allowable catch of no more than 2 million tonnes.
4 ) The Parties recognise that a total outtake by the Parties of 2 million tonnes in 2006 will result in a fishing mortality rate above the target level as defined in Paragraph 2. Until the fishing mortality has reached a level of no more than 0.32, the Parties agree to reduce their total allowable catch of blue whiting by at least 100 000 tonnes annually.
5 ) When the target fishing mortality rate has been reached, the Parties shall limit their allowable catches to levels consistent with a fishing mortality rate of no more than 0.32 for appropriate age groups as defined by ICES.
6 ) Should the SSB fall below a reference point of 2.25 million tonnes (Bpa), either the fishing mortality rate referred to in Paragraph 5 or the tonnage referred to in Paragraph 4 shall be adapted in the light of scientific estimates of the conditions then prevailing. Such adaptation shall ensure a safe and rapid recovery of the SSB to a level in excess of 2.25 million tonnes.
7 ) This multi-annual management arrangement shall be reviewed by the Parties on the basis of ICES advice.
1.1.4.2 North East Atlantic Fisheries Commission regulatory measurements
In addition to the Coastal States management plan, there is a recommendation by the North East Atlantic Fisheries Commission (NEAFC) at its annual meeting in November 2006 to adopt conservation and management measures for blue whiting in the NEAFC area in 2007. This would result in an expected catch of 147,000 tonnes of blue whiting in addition to the Coastal States Agreement of 1.7 million tonnes for 2007.
1 ) NEAFC takes notes of the Agreed Record of Conclusion of Fisheries Consultations between the Faroe Islands, the European Community, Iceland and Norway on the Management of Blue Whiting in the North-east Atlantic in 2007 signed in Tórshavn, 27 October 2006.
2 ) NEAFC further notes that by way of the said Agreed Record, the aforementioned Parties agreed to restrict their fishery on the blue whiting stock in 2007 according to a total catch limitation of 1.7 million tonnes.
3 ) In accordance with Article 5 of the Convention on Future Multilateral Cooperation in North-East Atlantic fisheries, the Contracting Parties recommend the following measure for the blue whiting Stock for 2007.
3.1 ) In order to ensure consistency and compatibility with the said Agreed Record, the Contracting Parties hereby establish an allowable catch limitation of 268 550 tonnes of blue whiting for 2007 in waters beyond the areas under national fisheries jurisdiction of the Contracting Parties.
3.2 ) This allowable catch limitation shall be allocated as follows:
3.2.1 ) European Community 37 400 tonnes (*)
3.2.2 ) Norway 31 450 tonnes (*)
3.2.3 ) Denmark in respect of:
3.2.3.1. Faroe Islands 31 450 tonnes (*)
3.2.4 ) Greenland 10 000 tonnes
3.2.5 ) Iceland 21 250 tonnes (*)
3.2.6 ) Russian Federation 137 000 tonnes
(*) Catches taken under these allocations shall be deducted from quotas allocated to Parties to the Agreed Record referred to in Paragraph 2.
4 ) The national quotas referred to in Annex I of the Agreed Record referred to in Paragraph 2 may be fished in the areas defined in Paragraph 3a.
1.1.5.1 Denmark
The Danish directed fishery blue whiting is mainly conducted by trawlers using a minimum mesh size of 40 mm. The directed fishery blue whiting in the western and northern areas constituted 82% of the total Danish blue whiting fishery (55 000 t) and this fishery mainly was conducted in March and April. The landings from the North Sea and Skagerrak were approximately 5000 tonnes. All landing were for production of fish meal and oil.
1.1.5.2 Germany
The main fleet targeting pelagic species is based at Bremerhaven and Rostock. The vessels are owned by a Dutch company and operating under the German flag. They consist of three large pelagic freezer-trawlers of lengths between 90 m and 120 m with power ratings between 4200 and 11 000 hp. The crew consists of about 35 to 40 men. The vessels are specially designed for pelagic fisheries. The catch is pumped into large storage tanks filled with cool water to keep the catch fresh until it is processed.
1.1.5.3 Faroe Islands
The Faroese quota for blue whiting was set at about 444 000 tonnes for 2006, of which 312 000 tonnes could be fished by the Faroese fishermen, the remaining amount was traded as part of bilateral negotiations with Russia, EU, and Norway. The Faroese fleet targeting blue whiting consists of nine large vessels and one smaller vessel.
In January the Faroese vessels followed the prespawning blue whiting on their migration southwards in the eastern part of the Faroe zone. Later in January a fishery developed in the spawning area on the Porcupine Bank (VIIc and k). This fishery continued in March, but later in March a large fishery for spawning blue whiting developed west of the Hatton-Rockall Plateau in International waters (VIIc, k and VIb). In April the fishery had moved northwards to the south of the banks on the border between EU and Faroes targeting spawning and postspawning fish (VIb and Vb). In May the postspawning fishery continued in the southern and south-western part of the Faroese EEZ (Vb). This year the postspawning blue whiting migrated northwards past Faroes in the Faroe Bank channel (west of the isles). Later in May and in June the fishery continued north of the Faroes (Vb and IIa) with good catches. There were only scattered catches in the Icelandic, Faroese and International waters during the period from July to October, but the catches started to increase again in late November on the north-eastern continental slope targeting the beginning of the southward migration of blue whiting. In December the fishery gradually moved southwards through the Faroe-Shetland channel targeting prespawning fish.
Only one industrial trawler operating mainly in Norwegian waters (Division IVa) in second quarter, with some catches in Faroese waters.
About 97% of the catches were taken with pelagic trawl the rest with pelagic pair-trawls.
1.1.5.4 Iceland
Iceland and Faroes have a bilateral agreement of mutual fishing rights for blue whiting in each other’s EEZs. Iceland set a total blue whiting catch quota of 352 600 tonnes in 2006 for Icelandic-Faroese and International waters.
The Icelandic directed fishery started in late February in International waters west of the British Isles and continued there through March. In April to June, the fishery was mainly in Faroese waters, but also partly in the Icelandic zone. In July, August and October, the fishery took mainly place on the Dorhnbank, between Iceland and Greenland with a total catch of 21 thousands tonnes. About 15% of the Icelandic catch was taken in the Icelandic zone, 65% in the Faroese zone and the remaining in International waters. Most of the catch was taken in the second quarter of the year (72%). The total Icelandic catch in 2006 was 309 508 tonnes.
A total of 24 trawlers/purse-seiners participated in the Icelandic fishery, as compared to 25 vessels in 2005, using large pelagic trawls with a 40 mm mesh size in the cod-end. The length range of the vessels was 55–105 meters with a mean length of 67 meters. The engine power range of the fleet was 1943–5920 kW (2500–8051 HP) with a mean of 3490 kW (4570 HP). Iceland has set size limitations on landings of blue whiting. If the catch consists of 30% or more of fish smaller than 25 cm, a temporary area closure is imposed.
1.1.5.5 Ireland
The Irish fishery for blue whiting began in late January with the great majority of landings reported from January to March. A total of 21 boats took part and reported landings of 54 900 t. This is a decline from 2005 when the Irish landings peaked at 73 400 t.
The fleet is comprised of 21 pelagic or polyvalent licensed trawlers with RSW tanks and 1 freezer-trawler. In the 2006 fishery, 14 of the largest RSW vessels in the fleet accounted for 90% of the total landings. Blue whiting from the Irish fleet is landed primarily for reduction to fishmeal with smaller but important amounts processed for human consumption. In 2006 landings for human consumption were in the region of 14 000 t with over 9000 t of this coming from the single freezer-trawler in the fleet. The remaining 5000 t were landed from RSW vessels fishing close to the main Irish pelagic port of Killybegs.
In 2007 the freezer-trawler was sold from the Irish fleet after the fishery closed. This is likely to have an impact on the distribution of landings in 2008 for the remaining fleet.
Fishing took place to the west and north of the Porcupine Bank as well as the Rockall trough to the north west of Ireland. Fishing took place on spawning and post spawning aggregations from ICES areas VIa, VIb, VIIb and VIIc. Fishing was concentrated in those rectangles along the shelf-edge and in deeper waters of between 300 and 600 m in depth.
1.1.5.6 Netherlands
The Dutch fleet fishing for pelagic species in European waters consists of 10 freezer trawlers on blue whiting ranging in engine powers from 3500 to 10 000hp. In addition, a number of flag vessels are operating from the Netherlands. In total 41 trips were made. The fishery for blue whiting is carried out with large pelagic trawls and is a directed fishery with almost no bycatch of other species. Catches decreased in 2006 compared to 2005. Most of the catches in 2006 originated from ICES Division VIa and VIIc and were taken in the first half of the year. All catches are landed frozen for human consumption.
1.1.5.7 Norway
After the coastal state agreement in 2005 and quota transfers in other international agreements, the Norwegian TAC for 2006 was set to 637 527 t (of which 472 631 t could be taken in the EU zone and 80 800 t in the Faroese EEZ). The majority (approximately 78%) of the Norwegian catches were taken in a directed pelagic-trawl fishery west of the British Isles and in the Norwegian Sea during the first half of the year. A total of 45 large combined purse-seiners/trawlers took part in this fishery in 2006. The remaining catches (22%) were mainly taken by the industrial trawl fleet (which uses both pelagic and demersal trawls) in the Norwegian deeps and Tampen area (east of 4ºW). This fishery is mainly a directed blue whiting fishery (for the time being with low Norway pout availability) but the bycatch of saithe can be significant (the maximum bycatch proportion of other species is 30% by weight during a trip). 50 industrial trawlers participated in the fishery (defined as landing more than 5 tonnes) in 2006. All the Norwegian TAC was taken in 2006.
Regarding the age and length composition in catches taken by the directed pelagic-trawl fishery west of the British Isles, the proportion of small and young fish seems to have been decreasing since 2004 (Figure 4.1.5.7.1), and the proportion is exceptionally low in 2007. This supports the observation that both the 2005 and 2006 year classes seem weak.
1.1.5.8 Russia
Blue whiting was fished by a few vessels in the eastern part of Faroese fisheries zone in January, but the fishery was interrupted at the beginning of February.
The fishery for blue whiting began on 29 January in the area of 55°00'–55°20'N in the international waters off Porcupine Bank and Rockall to 16°30'W. From 9 February the Russian commercial vessels kept continuously arriving in the area to a total of 29 vessels. The fishery level before the last five days of March was quite often restrained due to the technological constraints on processing catch. In February, aggregations were being formed in the area of 54°00'–56°30'N. In March, when blue whiting was fished by about 45 Russian vessels, the northern boundary of aggregations shifted to 59°N. Main length groups in the catches taken in February were 25–29 cm, followed by 25–27 cm. The spawning took place earlier and was more intensive than in 2005.
In 2004–2006, the fishery ceased approximately at the same time. The total catch in 2006 amounted to 75 thousand tonnes.
Resumption of fishery in the Faroese waters took place up to 4 April. Productivity of that fishery was very high up to 20 May when the main shoals migrated beyond the western border of Faroese zone. The trawlers then shifted to the east of the zone, staying there up to 18 June.
The fishery in the international waters of Norwegian Sea was prosperous after the middle of July. It finished on 20 August. A fishery was attempted in this region in the first half of September, in October and in the first week of December, but with limited success.
Two vessels in October-November and 7–17 vessels in December carrying out blue whiting fishery in the Faroese waters had relatively low productivity, mostly due to weather conditions.
1.1.5.9 Spain
The Spanish blue whiting fishery was carried out mainly by bottom pair trawlers in a directed fishery and by single bottom trawlers in a bycatch fishery. Small quantities were also caught by longliners. These coastal fisheries have trip durations of 1 or 2 days and catches are for human consumption. Thus, coastal landings are driven mainly to market forces, and are rather stable.
This fleet has decreased from 279 vessels in the early 1990s to 135 vessels in 2006 with an average of 28 m length, 444 HP and 141 GRT. 64% of these vessels are operating the whole year as bottom otter trawlers, 28% as pair bottom trawlers and 8% alternate between bottom otter trawls and pair bottom trawls throughout the year.
Pair Bottom Trawl Fishery: The Pair bottom trawl is a traditional fleet that fish mainly blue whiting (above 80%) and other pelagic species in Div. VIIIc and North IXa. In the middle of 1990s, VHVO gear (with 25 m of vertical opening) gradually replaced the traditional one. From 2001 the cod-end mesh size was increased to 55 mm.
Bottom Trawl Mixed Fishery: This métier operates in Divisions VIIIc and IXa North, using a cod-end mesh size of 65 mm and a vertical opening of 1.2–1.5 m. It targets a wide range of species including horse-mackerel, blue whiting, and mackerel (70% of landings together), and also hake, anglerfish, megrims, and Nephrops.
Spanish landings decreased around 14% in 2006 with a landing of 15 173 tones.
This Section describes in sub-sections the available data for assessment.
Catch data from Portugal were revised (a 500 tonnes decrease in total landing weight) after the WG meeting and were not included in this report.
Total catches in 2006 were provided by members of the WG. They were estimated to be about 1.97 million tonnes, 60 thousand tonnes less than in 2005. Time-series with catches by nations and area are given in Tables 4.2.1.1–4.2.1.7.
The spatial and temporal distribution of the catches of blue whiting in 2006 is given by ICES rectangles for the whole year is given in Figure 4.2.1.1 and by quarter in Figure 4.2.1.2. In 2006 the data provided as catch by rectangle represented approximately 98.3% of the total WG catch.
Some details about vessels operated by different nations targeting blue whiting are given in Table 4.2.1.8.
Most of the catches are taken in the directed pelagic trawl fishery in the spawning and post-spawning areas (Divisions Vb, VIa, b, and VIIb, c). Catches are also taken in the directed and mixed fishery in Subarea IV and Division IIIa, and in the pelagic trawl fishery in the Subareas I and II, in Divisions Va, and XIVa,b. These fisheries in the northern areas have taken 340 000–2 300 000 t per year in the last decade, while catches in the southern areas (Subarea VIII, IX, Divisions VIId, e and g–k) have been stable in the range of 20 000–85 000 t. In Division IXa blue whiting is mainly taken as bycatch in mixed trawl fisheries.
The proportion of landings originating from the Norwegian Sea has increased from 5% in the mid–1990 to around 30% in 2003–2004, after which the proportion has decreased again to around 15% (Figure 4.2.1.3). This might have implications for the stock assessment as much larger proportions of juvenile fish occur in catches from the Norwegian Sea, thus probably changing the exploitation pattern of the fishery as whole.
Discard
Discards of blue whiting are thought to be small. Most of the blue whiting is caught in directed fisheries for reduction purposes. However, some discarding occurs in the fisheries for human consumption and as bycatch in fisheries directed to other species. Discarding in 2006 is not included in the assessment.
Reports on discarding from fisheries which catch blue whiting were available from the Netherlands for the years 2002–2006. A discard sampling programme of the pelagic fleet is carried out in the frame work of the EU Data Collection Regulation. On average about 3% (1%–5%) of the Dutch catch (in numbers) of blue whiting is estimated to be discarded. About 2/3 of the discards comes from the directed fishery and mainly originate from cod-end damage or cleaning of the fish tanks. The other 1/3 is bycatch in fisheries targeting other species. Figure 4.2.1.4 gives a length distribution of landings and discards in the period 2002–2005 from the pelagic fleet in the Netherlands.
Also information of discards was available for Spanish fleets. Blue whiting is a bycatch in several bottom trawl fisheries directed to a mixture of species. The estimates of discard in mixed fisheries in 2006 ranged between 23% and 99% (in weight) as most of the catch is discarded and only last day catch may be retained for marketing fresh. The catch rates of blue whiting in these fisheries are however low. In the directed fishery for blue whiting for human consumption with pair trawls, discards were estimate to be 13% (in weight) in 2006.
Discards information from the Portuguese fisheries (WD Godinho et al., 2007) showed that for the period 2004-2006 the discarded weight of blue whiting was slightly higher than the landed weight (on average 4000 t per year).
1.2.1.1 Review of catch statistics
The Section addresses the ToR(c): “Review the catch statistics of blue whiting especially from 1978 to 1990 and resolve differences between ACFM landings, EOROSTAT and ICES Fishstat data”.
Catch data as recorded in various ICES working group reports (ICES 1985; ICES 1996; ICES 2006) covering the period since 1973 are presented in Table 4.2.1.1.1. These landings are compared with blue whiting, “Area 27”, landings from ICES/Fishstat database. In Figure 4.2.1.1.2 it can be seen that blue whiting working groups has adjusted the landings figures slightly over time, but changes are relative small. The working group landings relative to the FishStat landings show a modest deviance for the period since 1978 where landings exceed 100 000 tonnes annually.
In the table below some significant differences among various sources of information are shown. In the late seventies, the main differences are due to the lack of mixed industrial fishery landings in the officially reported landings. The Working Group was aware that individual vessel landings in some industrial fisheries are recorded as landing from one species only. Dependent on the species composition landings from a fishing trip can be recorded as blue whiting or e.g. Norway pout. This quantity goes into the official catch statistic, however it seems that for some years, the mixed industrial landings were not reported at all. In other years samples from the landings were used to quantify the actual species composition and these data are used to derive the WG estimate of the blue whiting landings.
In the early nineties the misreporting could be due to underreporting of blue whiting.
Examples of differences between ICES WG estimate and ICES FishStat database.
|
|
|
FISHSTAT
|
ICES WORKING GROUP
|
ICES-FISHSTAT
|
|
|
|
Year
|
Country
|
Landings(t)
|
Directed fishery (t)
|
Mixed industrial (t)
|
Total Landings(t)
|
Difference
|
|
1978
|
Norway
|
117 954
|
116 815
|
39 989
|
156 804
|
38 850
|
|
1978
|
Iceland
|
26 377
|
25 293
|
9484
|
34 777
|
8400
|
|
1988
|
Norway
|
209 738
|
208 416
|
24 898
|
233 314
|
23 576
|
|
1988
|
Denmark
|
134 642
|
797
|
18 144
|
18 941
|
-115 701
|
|
1991
|
Norway
|
119 201
|
114 966
|
22 644
|
137 610
|
18 409
|
|
1991
|
Denmark
|
50 368
|
0
|
15 538
|
15 538
|
-34 830
|
|
1991
|
Sweden
|
17 980
|
0
|
1000
|
1000
|
-16 980
|
1.2.1.2 Sampling intensity
In total 1715 samples were collected from the fisheries in 2006. 190 533 fish were measured and 27 014 were aged. Sampled fish were not evenly distributed throughout the fisheries (see text table below).
|
QUARTER
|
FISHERIES
|
DIRECTED
|
MIXED
|
SOUTHERN
|
TOTAL
|
|
1
|
No. of samples
|
498
|
67
|
109
|
674
|
|
|
WG Catch
|
1 013 765
|
5847
|
5439
|
1 025 050
|
|
2
|
No. of samples
|
298
|
127
|
113
|
538
|
|
|
WG Catch
|
627 221
|
42 217
|
5963
|
675 401
|
|
3
|
No. of samples
|
140
|
46
|
88
|
274
|
|
|
WG Catch
|
116 312
|
37 447
|
4636
|
158 395
|
|
4
|
No. of samples
|
95
|
46
|
88
|
229
|
|
|
WG Catch
|
83 086
|
19 729
|
4479
|
107 294
|
|
Total No. of samples
|
1031
|
286
|
398
|
1715
|
|
Total WG Catch
|
1 840 384
|
105 239
|
20 517
|
1 966 140
|
Considering the proportion between catches and sampling, the most intensive sampling took place in the southern fishery of Spain and Portugal. Here one sample was taken for every 52 tonnes, followed by the mixed fishery with one sample for every 368 tonnes, and lastly the directed fishery where there was one sample for every 1785 tonnes caught. In this context it should be noted that implementation of the EU Collection of Fisheries Data, Fisheries Regulation 1639/2001, requires EU Member States to take a minimum of one sample to be taken for every 1000 t landed in their country. Detailed information on the number of samples, number of fish measured, and number of fish aged by country and quarter is given in Tables 4.2.1.2.1 and 4.2.1.2.2 As can be seen, no sampling was carried out by Germany, Sweden and France, all with relatively small landings.
Sampling intensity for age and weight of herring and blue whiting are made in proportion to landings according to CR 1639/2001 and apply to EU member states. For other countries there are no guidelines. Current precision levels of the sampling intensity are unknown and the group recommends reviewing the sampling frequency and intensity on a scientific basis and providing guidelines for sampling intensity.
Data on the combined length composition of the 2006 commercial catch by quarter of the year from the directed fisheries in the Norwegian Sea and from the stock’s main spawning area were provided by the Faroes, Iceland, Ireland, Germany, the Netherlands, Norway, Russia and Scotland. Length composition of blue whiting varied from 12 to 46 cm, with 95% of fish ranging from 21–32 cm in length. The mean length in the fishery was 26.9 cm (Table 4.2.2.1) which is 7 mm larger than the mean length last year. The difference might be due to a decrease in recruitment in the most recent years lowering the proportion of young fish in the population. Length compositions of the blue whiting catch and bycatch from “other fisheries” in the Norwegian Sea and the North Sea and Skagerrak were presented by Norway (Table 4.2.2.2). The catches of blue whiting from the mixed industrial fisheries consisted of fish with lengths of 12–41 cm and a mean of 24 cm. France, Spain and Portugal caught blue whiting in the Southern area. The Spanish and Portuguese data used for length distribution of catches showed a length range from 10–38 cm with a mean length of 23.1 cm (Table 4.2.2.3).
For the directed fisheries in the northern area in 2006, age compositions were provided by Denmark, the Faroe Islands, Iceland, Ireland, Norway, the Netherlands, Russia and Scotland and the sampled catch accounted for 96% of the total catch. Estimates of catch in numbers for unsampled catches were raised according to the knowledge of how, where, and when the catches were taken. The age compositions in the directed fisheries are given in Table 4.2.2.4.
Age compositions for blue whiting bycatches from “other fisheries” in the North Sea and Skagerrak were provided by Norway, Denmark, Faroe Islands and Russia and sampled catch accounted for 96% of catches. These data were used for allocation of the remaining part of the total in that area. The age compositions are given in Table 4.2.2.5.
For the fisheries in the Southern area, age composition representing 81% of the catch were presented by Spain and Portugal. The age compositions in the southern fishery data are given in Table 4.2.2.6.
The combined age composition for the directed fisheries in the Northern area, i.e. the spawning area and the Norwegian Sea, as well as for the bycatch of blue whiting in “other fisheries” and for landings in the Southern area, were assumed to represent the overall age composition of the total landings for the blue whiting stock. The catch numbers-at-age used in the stock assessment are given in Table 4.2.2.7. The SALLOC program (ICES 1998/ACFM:18) was used to calculate the total international catch-at-age, and to document how it was done.
Catch curves made on the basis of the international catch-at-age (Figure 4.2.2.1) indicate a consistent stock-decline and thereby reasonably good quality catch-at-age data, especially for year-classes since 1995.
Mean weight-at-age in the catch data were available from Denmark, the Faroes, Iceland, Ireland, the Netherlands, Norway, Portugal, Russia, Scotland and Spain. Mean weight-at-age for other countries was based on the allocations shown in the Annex II (“ALLOC” files) and was estimated by the SALLOC program for the total international catch. Table 4.2.3.1 shows the mean weight-at-age for the total catch during 1983–2006 used in the stock assessment. There is a general trend towards lower weight-at-age (Figure 4.2.3.1). Although the reasons of this fact have not been investigated, the Working Group identified two possible causes:
· Density dependent effect: From 1996 onwards a series of exceptionally large year classes entered the stock leading to a substantial increase in stock size. The increase in population might have reduced the available food per individual fish leading to the observed decrease in mean weight.
· Shift in fishing pattern: During the same period the fishery extended remarkably, reaching new areas. This may have increased the catches of post-spawners and caused the decreased in the observed mean weights.
During the WG meeting it was not possible to draw conclusions on the changes in mean weight. Intersectional work is necessary and the WG recommends an analysis of the change in mean weight as part of the ToRs for the next WG meeting.
The weight-at-age for the stock was assumed to be the same as the weight-at-age for the catch.
Maturity-at-age used in the assessment was obtained by combining maturity ogives from the southern and northern areas, weighted by catch in numbers-at-age (ICES 1995/Assess:7). These are the same as those used since 1994. Although the values of maturity-at-age probably are too low, sufficient information for estimating new ogives is not available.
The possible need for revising the current estimate of instantaneous natural mortality rate M for blue whiting was discussed in detail by the 2002 WG. Although it was admitted that the current estimate M>=0.2 yr-1 might be too low, the factual basis for revision was ambiguous. More recent methodological work by WGMG (ICES 2003/D:03) emphasizes that natural mortality rate cannot be estimated reliably with information normally available for stock assessment models. The working group therefore considers that there is no new information that would justify a revision of the current estimate of M.
In the table below, blue whiting natural mortality and proportion of maturation-at-age is shown.
|
AGE
|
0
|
1
|
2
|
3
|
4
|
5
|
6
|
7–10+
|
|
Proportion mature
|
0.00
|
0.11
|
0.40
|
0.82
|
0.86
|
0.91
|
0.94
|
1.00
|
|
Natural mortality
|
0.20
|
0.20
|
0.20
|
0.20
|
0.20
|
0.20
|
0.20
|
0.20
|
1.2.5.1 International Blue Whiting spawning stock survey
Background and status
The International Blue Whiting Spawning Stock Survey (IBWSSS) is carried out on the spawning grounds west of the British Isles in March-April. The survey started in 2004 and is carried out by Norway, Russia and the EU. In 2005 the Faroes joined the survey. This international survey, with broad international participation, allowed for broad spatial coverage of the stock as well as a relatively dense net of trawl and hydrographical stations. The survey is coordinated by PGNAPES (ICES CM 2007/RMC:08).
The International survey directly incorporates both the Norwegian and Russian spawning stock surveys that started in the early 1990s; details of these surveys can be found in previous working group documents (e.g. ICES CM 2006/ACFM:34). The integrity of the Norwegian time-series has been maintained from 1991–2006, and it was used as the major source of survey information in previous assessments. However, in 2007 the Norwegian contribution to the international survey changed, resulting in coverage of a non-standard area, and therefore a break in the time-series. The index from the Norwegian spawning stock survey time-series could therefore not be used this year.
Use of this survey in stock assessment
Both the IBWSSS survey and the Norwegian spawning stock survey were used in the assessment this year.
Quality of the survey
Due to the short time-series from the international survey (2004–2007) there is insufficient data available to fully evaluate its performance. A further problem with the reliability of the joint international survey results is inconsistency between the age readings, due to the increased number of vessels (and readers) participating, (as compared to the Norwegian/Russian time-series). In comparison, the international age readings are probably less consistent than the Norwegian age readings. Between-vessel comparisons have shown significant differences in the past.
For the first time, uncertainty in stock estimates has been assessed. At present, only one source of uncertainty is considered, i.e. spatio-temporal variability in acoustic recordings. Bootstrapping was used to characterize uncertainty in the mean acoustic density. This analysis indicates that confidence limits were stable in 2004–2006 (Figure 4.2.5.1). In 2007, the width of confidence limits was almost doubled because of a few very high acoustic records.
The Norwegian spawning stock survey shows moderate good internal consistency (Figure 4.2.5.2). However, while the international time-series clearly lacks sufficient data points to make a firm conclusion regarding internal consistency, the available data appears inconsistent.
Results:
The distribution of acoustic backscattering densities for blue whiting as recorded by the six vessels is shown in Figure 4.2.5.3 (below). The highest concentrations were generally recorded in the area between the Hebrides, Rockall and the banks southwest of the Faroes, but several large concentrations were found just north of Porcupine Bank. The blue whiting spawning stock estimates based on the international survey are given in Table 4.2.5.1.1.
Figure 4.2.5.3. Blue whiting. Schematic map of blue whiting acoustic density (sA, m2/nm2) found during the spawning survey in spring 2005, 2006 and 2007.
Given the uncertainty in the estimate, no change in blue whiting stock abundance in the spawning area could be detected. Point estimates suggest a slight increase in stock biomass and stable stock numbers. In contrast, the estimates in 2006–2007 are significantly higher than the estimate in 2005.
Abundance estimates from this acoustic survey should generally be interpreted as relative indices rather than absolute measures. In particular, acoustic abundance estimates critically depend on the applied target strength. The target strength currently used for blue whiting is based on cod and considered to be too low, possibly as much as by 40% (see Godø et al., 2002, Heino et al., 2003, 2005, Pedersen et al., 2006). This would imply an overestimation of stock biomass by a similar factor. This bias, however, should be roughly constant from year to year, and does not affect the above conclusions about relative change in abundance of stock.
Mean age has increased from last year’s 2006 and is now the highest on recorded in the four years since the survey started in 2004. Recruitment to spawning stock seems weak with numbers at ages 1–3 years the lowest in the time-series. On the other hand, numbers of “old” blue whiting (ages 6–8 years) are relatively high. However, age distributions seem noisy. In part, this seems to be caused by variability in recruitment, with some cohorts recruiting to the spawning ground earlier than others. However, between-vessel comparisons of mean age at length also suggest that there could be problems in age reading.
The survey area was reduced by about 20% from 2006. Most of the reduction came from areas with low density in 2006. Nevertheless, the estimates would have been expected to be higher if the same coverage were achieved.
Most of the increase in the stock estimate comes from the southern sub-areas (the Porcupine Bank). This area was covered earlier in season this year than in 2006. With later coverage, the biomass would probably have moved to the Hebrides sub-area. In the Hebrides and the Faroes sub-areas biomass was essentially unchanged, whereas biomass decreased in the Rockall sub-area where coverage was also significantly reduced.
Age and length distributions from the 3 last years are shown in Figure 4.2.5.4.
1.2.5.2 International ecosystem survey in the Nordic Seas
Background and status:
The international ecosystem survey in the Nordic Seas (Figure 4.2.5.5 below) is aimed at observing the pelagic ecosystem in the area, with particular focus on herring, blue whiting, mackerel (Norway), zooplankton and hydrography. The observations on herring and blue whiting are done by acoustic observation with main focus on Norwegian spring-spawning herring and blue whiting in the Norwegian Sea. The survey is carried out in May since 1995 by the Faroes, Iceland, Norway, and Russia, and since 1997 (except 2002 and 2003) the EU. The high effort in this survey with such a broad international participation allowed for broad spatial coverage as well as a relatively dense net of trawl and hydrographic stations.
Figure 4.2.5.5. Blue whiting. Areas defined for acoustic estimation of blue whiting and Norwegian spring spawning herring. The dark red box in the middle represents the standard area (8°W–20°E and north of 62°N) of which blue whiting data is used for assessment. The outer green box represents the total survey area.
Estimates in 2000–2007 are available both for the total survey area and for a “standardized” survey area. The latter is more meaningful as the survey coverage has been rather variable in the south where post-spawning blue whiting are entering the Norwegian Sea as well as in the west where large blue whiting occur. As these results in unknown noise that are highly undesirable, the discussion below is therefore based on the estimate for the standard survey area.
Since 2005 this survey has extended into the Barents Sea where the main focus of investigations has been young herring and capelin larvae. The survey is coordinated by PGNAPES (ICES CM 2007/RMC:08).
Use of this survey in stock assessment
The survey has been used in the final assessment from 2005 onwards to estimate recruitment of blue whiting. The performance of this survey in predicting recruitment is not yet well known, as the overlap with the assessment estimate is limited and the latter in general is plagued by uncertainties that reflect scarcity of data on the most recent year classes. However, the result is in line with the recruitment index (age 1) from the Barents Sea where the index in 2006 was the lowest one since 1999.
Internal consistency within the survey’s age composition shows good correlation for the International time-series (Figure 4.2.5.6), however this correlation is mainly driven by one data point.
Results for blue whiting:
The International ecosystem survey in the Nordic Seas shows a strong decline in stock numbers and biomass, and especially a very low index for age 1 and age 2 in both 2006 and 2007. This decline in biomass is far larger than could be explained by acoustic uncertainty. The situation resembles somewhat that in 2000 when what now appear to be too low values were estimated. The reason this is unclear, but could relate to migrations. A well known problem is migration of post-spawning blue whiting from the spawning area to the southern part of the survey area, but this should not affect juvenile blue whiting (for this reason, only indices for ages 1–2 years are used in tuning the assessment). Somewhat higher stock estimate was obtained for similar area in July-August 2007, but even this estimate is considerably (-30%) lower than the estimate in 2006.
Figure 4.2.5.7. Blue whiting. Schematic map of blue whiting acoustic density (sA, m2/nm2) found during the survey in spring 2005, 2006 and 2007.
The decline is particularly dramatic in terms of numbers (-71%), reflecting increasing average size and age of blue whiting in this survey. Mean age of blue whiting fluctuated between 1.3 and 2.1 years in 2000–2005, increased to 2.8 years in 2006, and was estimated to be 3.7 years in 2007 survey. This change reflects strengths of 2005 and 2006 year classes, which are low if not extremely low. Similar signal has been recorded in the Barents Sea February-March, and again in the Norwegian Sea in July-August. There are all reasons to believe that the low numbers of recruits suggested by this survey are real.
The blue whiting stock estimates based on the international survey in both the standard and total survey area are given in Table 4.2.5.1.2.
1.2.5.3 Norwegian Sea summer survey
Background and status:
In 1981–2001 Norway made an acoustic survey in the Norwegian Sea in order to follow the migration of Norwegian spring spawning herring and to measure blue whiting in its feeding areas. This survey used to give the first indication of the incoming year class measured at age 1; in 2004 SGAMHBW recommended using indices from this survey at ages 1–4 years.
This survey was started anew in 2005, but with main focus on mackerel in 2005–2006. From 2007 onwards more focus is/will be put on blue whiting. No estimates have made.
Use of this survey in blue whiting assessment:
As the survey is terminated it provides little information for the latest years in the assessment, and it was decided not to use the survey in the final assessment, but data were used in a SPALY configuration with XSA.
Results:
The stock estimates in numbers at age are given in Table 4.2.5.3.
1.2.5.4 Norwegian bottom trawl survey in the Barents Sea
Background and status:
Norway has conducted bottom trawl surveys targeting cod and other demersal fish in the Barents Sea since late 1970s. From 1981 onwards there have been systematically designed surveys carried out during the winter months (usually late January-early March) by at least two Norwegian vessels; in some years the survey has been conducted in co-operation with Russia. Blue whiting is a regular bycatch species in these surveys, and has in some years been among the numerically dominant species (Heino et al., 2007). This survey is presently giving the first reliable indication of year class strength of blue whiting.
Most of the blue whiting catches (or samples thereof) have been measured for body length, but very few age readings are available (from 2004 onwards otoliths are systematically collected). The existing age readings suggest that virtually all blue whiting less than 19 cm in length belong to 1–group and that while some 1–group blue whiting are larger, the resulting underestimation is not significant. An abundance index of all blue whiting and putative 1–group blue whiting from 1981 onwards is given in Table 4.2.5.4 and follows methods described in Heino et al., (2007). Somewhat different threshold have been used before (21 cm was used in 2004, 20 cm in 2005, and 18 cm in 2006), the latest change being made to consistently accomodate a change in measurement precision in the middle of the time-series.
Results:
Total index has declined substantially from 2006 (-70%), but it is still moderately high in the historic perspective (close to the 3rd quartile of the distribution), and higher than all observations before 1997.
1–group index for 2007 is moderately weak in the historic perspective (being close to the 1st quartile of the distribution). However, it is very weak in comparison to the estimates from this decade, and the lowest one after 1995.
Use of this survey in blue whiting assessment:
The survey is not used in the assessments, but it is used for recruitment estimation in the forecasts.
1.2.5.5 Spanish bottom trawl survey
Background and status:
Bottom trawl surveys have been conducted off the Galician (NW Spain) coast since 1980, following a stratified random sampling design and covering depths down to 500 m. The survey directed to a mixture of species. Since 1983, the area covered in the Spanish survey was extended to completely cover Spanish waters in Division VIIIc. A new stratification has been established since 1997.
Use of this survey in blue whiting assessment:
The survey is not used in the assessments as it is only representative for a small part of the stock area.
Results:
Stratified mean catches and standard errors are shown in Table 4.2.5.5. Larger mean catch rates are observed in the 100–500 m depth range. Since 1988 the highest catch rates in the Spanish survey were observed in 1999 (124 kg/haul). The 2006 estimate is 72 kg/haul (Figure 4.2.5.9).
1.2.5.6 Faroes plateau spring bottom trawl survey
Background and status:
On the Faroe plateau an annual demersal bottom trawl surveys is carried out during spring (March 1996–2007). The survey is aimed at cod, haddock and saithe, but varying amounts of blue whiting are caught as bycatch each year. An updated time-series of the 0–group and 1–group will be provided for next year’s WG.
1.2.5.7 Faroes plateau autumn bottom trawl survey
Background and status:
On the Faroe plateau an annual demersal bottom trawl survey is carried out in autumn (August-September 1994–2007). The survey is aimed at cod, haddock and saithe, but varying amounts of blue whiting are caught as bycatch each year. An updated time-series of the 0–group and 1–group will be provided for next years WG.
The ICES review of the 2006 blue whiting assessment had no serious concerns about the quality of the assessment, but pointed out the need for a statistical analysis of the survey data for their possible inclusion in future assessments. This is initiated in Section 4.2.5.
The review group also pointed out that standard software could be used for sensitivity analysis of the impact of input data for projection on projected biomass and yields. Due to the very uncertain stock size in the terminal year calculated by this year’s working group, it was not possible to make meaningful uncertainty estimates and thus employ the standard software (e.g. the MLA mid-term forecast).
The review group also noted inconsistency in estimating recruits for forecasts. The choice of method for recruitment estimation will be described and justified in more details in Section 4.6.
Catch data
Catch curves made on the basis of the international catch-at-age (Figure 4.2.2.1) indicate a consistent stock-decline and thereby reasonably good quality catch-at-age data, especially for year classes since 1995. Total estimated mortality is higher for the year classes since 1995, as compared to the year-classes from the previous decade.
Quota uptake of blue whiting in 2007, at least 90% by mid-August, does not indicate a strong decline in the stock size, even though catch rates of a pelagic schooling fish are not a good predictor of its abundance.
Survey data
The WGNPBW has previously concluded that the “Norwegian spawning stock survey” (Section 4.2.5.1) and “International ecosystem survey in the Nordic Seas” (Section 4.2.5.2) can be used in the analytical assessment. However, the spatial coverage of the first was too sparse in 2007 to permit an update of the time-series with 2007 data; the available information from this survey is therefore the same as for last year’s assessment.
The “International Blue Whiting Spawning Stock Survey”, includes the 2007 information and provides a good coverage of the spawning grounds, but has a very short time-series. The estimated (spawning) biomass from this survey is around 10 millions tones. This is far higher than the assessment results. Abundance estimates from this acoustic survey should be interpreted as relative indices rather than absolute measures as the abundance estimates critically depend on the applied target strength, which seems too low (see Section 4.2.5.1 for details). The total SSB estimated from the international survey (Figure 4.2.5.1) in 2007 has a wide confidence interval.
The available survey information is presented in Table 4.3.2.1. The indices at age are quite similar for the International Blue Whiting Spawning Stock Survey and the Norwegian survey (Figures 4.3.2.2 and 4.3.2.3), an expected result as the latter is a subset of the former.
The internal consistency is poor for the International Blue Whiting Spawning Stock Survey, except for age groups 4 to 5. The Norwegian survey performs somewhat better (Figure 4.2.5.2).
The “International ecosystem survey in the Nordic Seas” gives a very clear drop in recruitment for the 2005 and 2006 year class. The internal consistency is reasonably good for this survey (Figure 4.2.5.6).
Comparison of the quality of the Norwegian and International Blue whiting spawning stock surveys.
Due to the short time-series from the International Blue Whiting Spawning Stock Survey there are still too few data to fully evaluate its performance. Compared with the Norwegian Spawning Stock Survey, the internal consistency (Figure 4.2.5.2) of the International survey does not perform well, even though the very short time-series does not allow a direct comparison.
The apparent poor-performance of the International spawning stock survey may be linked to inconsistency between the age readings with several vessels (and readers) participating as compared to the Norwegian/Russian time-series. In comparison to the Norwegian survey the international age readings are probably less consistent than the Norwegian and Russian age readings. Between-vessel comparisons have shown such differences in the past.
The spawning stock spatial coverage of the International survey is much more extensive than of the Norwegian survey; Rockall area often contains a substantial part of the stock and has only been covered by the Norwegian time-series from 1998 onwards, although never extensively. Although the international survey gives a better coverage, the difference in biomass estimate is believed to be not more than 1 to 2 million tonnes over the past years.
Problems in survey progression (the chance of double counting fish) are believed to be stronger within the International survey since inter-vessel coordination proves to be difficult. However, both surveys are having large cruise-breaks within in their survey. Results from 2007 show higher sensitivity to the way fish are distributed than expected. Therefore it is believed that the higher cruise track density of the international survey gives a more precise estimate on the total biomass in the core area.
The correct timing of covering the entire spawning stock is believed to be less accurate in the Norwegian survey in comparison to the international survey.
Based on the above discussion the WG concludes that with the present short time-series, the International survey does not give a more precise measurement of the blue whiting stock abundance compared to the Norwegian survey.
Data exploration with survey combinations
Even though the International Blue Whiting Spawning Stock Survey has a very short time-series, it was decided to investigate including the survey in the assessment, as it is the only survey with a good coverage of the spawning stock in 2007.
Below we investigate different scenarios for the inclusion of the available survey information. A number of exploratory runs using the main assessment methods (SMS, AMCI, ICA, and TISVPA) were made using the following survey combinations:
Three surveys:
· Norwegian spawning stock survey, age 3–8 from 1991–2006
· International Blue Whiting Spawning Stock Survey, age 3–8 from 2004–2007
· International ecosystem survey in the Nordic Seas, age 1–2 from 2000–2007
Two surveys:
· Norwegian spawning stock survey, age 3–8 from 1991–2006
· International ecosystem survey in the Nordic Seas, age 1–2 from 2000–2007
No overlap in time for the spawning surveys:
· Norwegian spawning stock survey, age 3–8 from 1991–2003
· International Blue Whiting Spawning Stock Survey, age 3–8 from 2004–2007
· International ecosystem survey in the Nordic Seas, age 1–2 from 2000–2007
Down-weighted international survey (with no temporal overlap on the spawning surveys):
· Norwegian spawning stock survey, age 3–8 from 1991–2003
· International Blue Whiting Spawning Stock Survey, age 3–8 from 2004–2007, down-weighted
· International ecosystem survey in the Nordic Seas, age 1–2 from 2000–2007
The “two surveys” run can be considered as close to the “same procedure as last year” as was possible, using the same survey configuration as the 2006 working group, with the additional inclusion of 2006 data; however, the absence of a 2007 Norwegian survey meant that it was not possible to fully replicate the previous years methodology, and thus should be thought of as an almost-SPALY (“ASPALY”).
By excluding the 2004–2006 data from the Norwegian spawning stock survey there is “no overlap” in the two spawning stock surveys and the same information is not used twice.
The International Blue Whiting Spawning Stock Survey is very short and the internal consistency is rather poor for most ages. The “down-weighted” configuration explores the effect of giving a lower weight to this survey.
The explorative runs were, carried out with the assessment models (AMCI, ICA, ISVPA/TISVPA, SMS and XSA) employed last year. The models AMCI, ICA, ISVPA and SMS use all the assumption of separability in modeling the fishing mortality, whereas XSA is a VPA-type of assessment model. For the family of separable models AMCI, ISVPA and ICA use a fixed input weight for each survey whereas SMS uses the maximum likelihood method and weights the various surveys according to their estimated variance.
The four different runs mentioned above were carried out with the latest version of AMCI, Version 2.4. Apart from tuning fleets, the remaining settings were the same as in the final AMCI run last year. The results from each of the runs are shown in Figure 4.3.3.1. There are only minor differences between the run where all years are used in the tuning data and the run with no overlap between the Norwegian and international spawning stock surveys. Omitting the international spawning stock survey from the tuning leads to higher SSBs and lower Fs in the last years compared to the other three runs. Down-weighting of the International spawning stock survey (50% on all ages and years) gives the highest F and lowest SSB in the last years. All four runs give a steep decline in SSB from 2003 onwards, while the recruitment estimates are quite similar. The estimated SSB in 2007 varies between 3.7 and 5.1 million tonnes, while F in 2006 ranges from 0.37 to 0.46 which is above Fpa. The model residuals (only shown for the final run, see below) (Figure 4.3.3.2) do not show any particularly worrisome features. Some year effects are indicated by the survey residuals, especially in the residuals from the Norwegian and international spawning stock surveys. The catch residuals mostly look nice, but the large positive residual of age 1 in 2006 and the row of negative residuals of age 2 the last 5 years show that surveys and catch data provide conflicting information about the abundance of the youngest age groups.
The proposed final AMCI run is the one where the international spawning stock survey is down-weighted. Estimated fishing mortality, stock numbers and stock summary can be found in Tables 4.3.3.1 to 4.3.3.3. Plots of the catch and survey residuals are shown in Figure 4.3.3.2. In Figure 4.3.3.3 the results from this year’s and last year’s final AMCI assessments are compared. Compared to last year, this year’s assessment gives a downward revision of recruitment in 2005, a downward revision of SSB in recent years and an upward revision of F in the recent time period.
As in the previous assessment (2006), the “triple-separable” version of the ISVPA model (TISVPA) was used for exploratory runs. This version allows to taking into account possible cohort-dependent peculiarities in the selection pattern. Such effects can arise from differing interactions of cohorts with the fishing fleet, by possible aging difficulties and errors in a particular cohort or by some other unrevealed reasons.
Bearing in mind that the present situation with the blue whiting stock can be characterized as rather uncertain (it is certain that the stock is declining, but uncertain how rapidly), a wide range of possible settings of the model were used for analysis. In choosing the best settings the guiding ideas were to get non-contradicting signals from all available data (catch-at-age data and 3 surveys: Norwegian acoustic spawning stock survey 1991–2006 (survey 1); International ecosystem survey in the Nordic seas 2000–2007 (survey 2), and International blue whiting spawning stock survey 2004–2007 (survey 3)) in order to retain the meaningful input into the assessment method from all of them.
As a whole, two intrinsically non-controversial possible solutions were found. One of them was in line with the TISVPA result obtained in 2006 (hereafter referred to as the “pessimistic” scenario). Another one showed an almost halved rate of decline in final years-let us name it the “optimistic” scenario.
In the so-called pessimistic run, the model settings were used as follows: “triple-separable” case with the window for estimation of cohort-factors-from age 1 to age 8; the “catch-controlled” version (catch-at-age is assumed as true and all residuals in catch-at-age are attributed to violations of selection pattern stability); the condition of unbiased separable representation of fishing mortalities (more correctly-of exploitation rates); the measure of closeness of fit for catch-at-age; the sum of squared residuals in logarithmic catch-at-age, and the absolute median deviation (AMD) of residuals in logarithmic abundance-at-age for all surveys; catchability-at-age are estimated for all surveys.
The so-called “optimistic” solution came from the TISVPA run with somewhat ideologically different setting: the catch-controlled version was substituted by the “mixed” version (assuming equal possibility for errors in catch-at-age and in assumption of stability of the selection pattern) and the AMD of residuals in logarithmic catch-at-age was used as a measure of closeness of the model fit to catch-at-age data. For Survey 1 the measure of closeness of fit was changed to the median (MDN) of the distribution of squared residuals in logarithmic abundance at age as giving more sharp minimum. For Survey 2, as the data set containing the data only for 2 age groups, it was decided not to estimate catchability coefficients and the measure of closeness was also changed to the median of the distribution of squared residuals in logarithmic abundance-at-age as giving more sharp minima. For Survey 3, bearing in mind that there could be strong year-to-year variations of the stock coverage by the survey (and hence in the effective survey catchability), not logarithmic abundance-at-age values were fitted, but logarithmic age proportions. Moreover, age proportions were weighted by stock abundance (by years) in order to give more weight to the data for years with more abundant stock (as probably being more representative), and the measure of closeness was the median of the distribution of squared residuals in logarithmic weighted (by stock abundance) age proportions as giving a more distinct minimum.
First, let us present the results of these two runs in comparison to the result of the TISVPA assessment in 2006 (Figure 4.3.4.1). As it can be seen, the results of the “pessimistic” run are much more similar to the results of the previous assessment. It is also can be seen that both solutions are equally pessimistic with respect to recruitment in 2005 and 2006 and probably the main difference in solutions is in estimates of the 2002–2004 year-class strength.
Figure 4.3.4.2 proves that both results are rather well supported by signals from all of the available data sources, but these signals were produced using different statistic assumptions and restrictions on the solution, and different degrees of robustness of the measures of closeness of fit used. Generally speaking, all of them are logically and statistically meaningful. However, which of them are more pertinent to the data under consideration is a difficult question requiring deep investigation and is beyond the scope of the WG meeting. Instead we examine some simple tests.
Figure 4.3.4.3 represents the residuals for pessimistic (first column) and optimistic (second column) results for catch-at-age and surveys information. It is difficult to find any apparent difference in the quality of residuals for catch-at-age, as well as for survey 2. For Surveys 1 and 3 residuals in terminal years are somewhat smaller for the optimistic solution, but it is necessary to remember that for Survey 3 in the “optimistic” run we measure residuals in age proportions-not in abundances.
Figure 4.3.4.4 compares the results of bootstrap (conditional parametric with respect to catch-at-age, surveys are noised by lognormal noise with sigma=0.3). It is difficult to draw any firm conclusion from the bootstrap results: confidence intervals for SSB are wide in terminal years for both pessimistic and optimistic cases. However, while for the pessimistic case the bootstrap-median SSB curve in final years generally reflects the shape of SSB curve of the basic run, for the optimistic case the bootstrap-median SSB reveals an unexpected peak in 2005 up to 9 million tones. This may be evidence that the model settings used in the “optimistic” run are less stable with respect to overestimation of the stock. This conclusion is very conditional due to the oversimplified error model used in the bootstrapping exercise.
It appears difficult to choose between the pessimistic and optimistic solution. The results for each scenario are given in Tables 4.3.4.1–4.3.4.6.
The ICA (Integrated Catch-at-age Analysis) model was used to explore blue whiting data. In previous years, various ICA settings were investigated. In 2007 it was decided to concentrate on the tuning fleets used in the assessment. The same settings as the final ICA run in 2006 were used for each run in 2007.
The survey combinations used in each run are described in Section 4.3.2. In order to be comparable the same survey options were used for each model.
A downward trend in recruitment can be seen in each of the ICA runs (Figure 4.3.5.1). The International Norwegian ecosystem survey is the main recruitment index used in the assessment. The 2007 index is the lowest in the current time-series. SSB from 2003 shows a declining trend in all runs with the lowest value obtained from the run with no overlap. Mean F values are decreasing and varying between 0.34 and 0.4 depending on the tuning fleet used. The highest F, equal to 0.4 is also produced from the run where there is no overlap between surveys.
The final run chosen has the most recent data from the International survey. The Norwegian spawning stock survey is used only as far as 2003, to avoid “double counting” and overlap with the International survey which runs from 2004–2007. This survey has a short time-series but has increased coverage when compared to the Norwegian survey. The residual patterns from this final run are presented in Figure 4.3.5.3. There is no consistent pattern in the catch residuals. Year effects can be seen in the Norwegian acoustic survey and International survey residuals. The stock summary from this final run is shown in Table 4.3.5.1.
When compared to other models, SMS, AMCI, XSA and TISVPA, ICA produces a similar low value for recruitment. ICA shows a high SSB and the low Mean F in the 2006. A comparison between the final run in 2006 and 2007 is shown in Figure 4.3.5.2. The 2007 runs show a higher mean F, a lower recruitment and SSB.
|
SETTINGS USED FOR ICA FINAL RUN
|
2006
|
2007
|
|
Number of age structured tuning series
|
2
|
3
|
|
International survey included
|
No
|
Yes
|
|
Number of biomass tuning series
|
0
|
0
|
|
Number of years for separable constraint
|
8
|
8
|
|
Reference age for separable constraint
|
3
|
3
|
|
Constant exploitation pattern
|
Yes
|
Yes
|
|
S to be fixed on last age
|
1.5
|
1.5
|
|
Catchability model for tuning fleets
|
Linear
|
Linear
|
|
Age range for the analysis
|
1–10
|
1–10
|
|
Survey weights for all fleets
|
100%
|
100%
|
|
Shrinkage
|
No
|
No
|
|
Manual down weighting
|
Yes
|
Yes
|
|
Weighting of age 1 catch numbers
|
50%
|
50%
|
Data exploration runs were performed using the Stochastic Multi-species (SMS) model (Lewy and Vinther, 2004) to examine the impact of adding and removing the available surveys (Figure 4.3.6.1), as described above in Section 4.3.2. The temporal overlap between the Norwegian acoustic survey and the International Blue Whiting Spawning Stock Survey (IBWSSS), and the corresponding “double counting” data from the one source, does not appear to create an appreciable discrepancy; the “Three surveys” and “No overlap” runs agree very closely with each other. This result suggests that the information contained in the international survey is similar to that in the Norwegian survey and that there is no contradiction between the surveys in the most recent (overlap) years.
Omitting the IBWSSS survey and simply using the Norwegian acoustic survey (the “two surveys” scenario) caused a large change in the results, both quantitatively (SSB in the terminal year halved) and qualitatively (F3–8 increasing in recent years rather than decreasing); the IBWSSS appears to have a very strong influence on the results in the terminal year. All runs showed similar results in recruitment, both in terms of general trends and in estimates of the terminal recruitment. The 2006 assessment agreed closely with the “Three surveys” and “No overlap” runs in terms of mean F, and was midway between the two sets of SSB trajectories.
The SMS uses maximum likelihood to weight the various data sources, giving a higher weight to observations associated with a low uncertainty. Both the results of the described scenarios, and detailed examination of the diagnostic output showed that the model is fitting the IBWSSS with a low standard deviation (0.2) for most ages, and thereby placing a high weight upon it. However, this may be scientifically unwise, due to the very short time-series (four years) and lack of contrast in the data. Reducing the weight given to this survey would thus produce a more robust assessment.
The effect of down-weighting the IBWSS survey was examined by varying the a priori weights applied to all survey information in the SMS model (Figure 4.3.6.2). The a priori weight on catch observations was kept constant at the default value (i.e. 1.0). If catch and survey observations give the same information of the stock development a different weighing of the data should have no effect. However, for this assessment, decreasing the weight on the survey information, and thereby increasing the relative weight on catch information, reduced the terminal SSB and increased the terminal F estimates. All runs showed similar results in both the trend and magnitude of recruitment. This indicates that the catch data and the survey indices show the same signal for the recruitment in the terminal year, such that down-weighting one of them has a very limited effect.
An alternative and more useful approach to this problem is to down-weight the IBWSSS itself directly, rather than down-weighting all survey information collectively, as was done in Figure 4.3.6.2; while we may question the use of the current, short IBWSSS time-series, we have no a priori justification for doing so with the Norwegian acoustic survey or the Norwegian Sea survey, both of which have much longer time-series. By setting a lower level for the uncertainty on the estimates for catchability for this survey, it is possible to define constraints for how closely the method can fit the survey observations, thereby preventing “over-fitting” and excessive weighting of the dataset. Down-weighting the IBWSSS, via this parameter, had the effect of reducing the estimated SSB and increasing the estimated F in the terminal years (Figure 4.3.6.3). Again, the recruitment showed similar results in both the trend and magnitude of recruitment.
The three sets of exploratory runs, viewed together, show that inclusion of the IBWSSS increases the estimate of the SSB and decrease the fishing mortality. The precise reasons for this are unclear, but are clearly related to the most recent value of the survey (i.e. the 2007 value). Over-reliance on this survey is scientifically unwise as the performance of this survey is still uncertain and probably not better than the Norwegian survey on the spawning ground (Section 4.3.2). The final SMS run was chosen so that the survey variances of the IBWSSS survey were approximately the same as the Norwegian survey; this was found to correspond to a “minimum CV on cpue observations” parameter of 0.4.
Examination of the catch residuals from the final SMS run (Figure 4.3.6.4) showed no appreciable patterns. The residuals from the survey observations (Figure 4.3.6.5) showed significant year effects in the IBWSS and Norwegian acoustic surveys, a well-known phenomenon with acoustic surveys. The observed indices from the IBWSSS were higher than the predicted model values for all ages in 2007 and in general lower for the preceding three years. No age or cohort effects were apparent.
Examination of the diagnostic output from the final SMS run (Table 4.3.6.1) does not show any major causes for concern, although there is an unusual effect in the values of the survey catchabilities-at-age. The catchability in the Norwegian Spawning Stock Survey increases with age, and reaches at maximum at age 4. This is an unusual result, and tends to contradict the trend seen in the IBWSSS, where the catchability increases monotonically with age, even though these two surveys are quite similar. A similar phenomenon was observed Norwegian Survey in the final SMS run in the 2006 working group. There is no good explanation for the result, but could simply be due to a lower (trawl) catchability of the oldest fish on the Norwegian survey.
The final SMS run (Figure 4.3.6.6, Tables 4.3.6.2–4.3.6.4) shows a small decrease in fishing mortality in the terminal years. SSB is rapidly decreasing associated with a strong decreasing recruitment from 2001 onwards. The 2006 recruitment level is at historic low levels.
The trends in SSB and recruitment estimated by the current assessment using the SMS method extend those estimated during the 2006 working group (Figure 4.3.6.7). The fishing mortality estimated in the terminal years has decreased slightly and SSB increased slightly over the 2006 assessment. The ASPALY run (using the same two fleets used last year, but no update of the Norwegian spawning ground survey) shows now a marked increase in F and a much lower SSB compared to the 2006 assessment.
Comparison of the observed and fitted catches from the SMS run (Figure 4.3.6.8) did not provide strong evidence that the separability assumption has been violated; there is close agreement between the two time-series.
The stock-recruitment relationship derived from the SMS run (Figure 4.3.6.9) clearly shows the stock as having had two distinct regimes since the early 1980s; the first corresponding to seemingly random moderate-low recruitment and biomass levels, followed by a large circular trajectory corresponding the recent boom-bust cycle. The role of recruitment in driving the dynamics of this stock is clearly apparent from this figure.
The uncertainties on SSB and mean F are large (Figure 4.3.6.10). The CV of SSB increases from 5% in 2001 to 15% in 2006. CV on F increases from 8% to 20% in the same period. Uncertainties on stock numbers (1. January 2006) varies between 17% and 47%; largest on the youngest and oldest ages. All the estimated uncertainties are clearly higher than the values estimated last year.
1.3.7 Data exploration in XSA
The SPALY run was carried out using following fleets:
· Norwegian spawning ground survey (NSSS) 1981–2006, ages 2–8
· Russian spawning stock survey 1982–1996, ages 3–8
· Norwegian Sea ecosystem survey 1989–2001, ages 1–7
These data did not provide sufficient basis for the assessment of young fish abundance. Therefore this year a new set of data sources was applied:
· Norwegian spawning ground survey 1996–2006, ages 2–8
· Norwegian Sea summer survey 1989–2001, ages 1–7
· International ecosystem survey in the Nordic Seas, 2000–2006, ages 1–2
The last of the surveys contains indices in 1st and 2nd age groups. As result, the estimation of recruits in the terminal was determined from surveys and the number of years with year-effect in log-catch residuals decreased. Investigating the different tuning configurations, the following combination was found to give the best fit to the data:
· q plateau set at age 6
· catchability depends on stock size for ages less than 3
· SE at survey estimates set as 0.3
· Regression type P
The results are presented in the Figures 4.3.7.1 and 4.3.7.2. Retrospective analysis suggests that the assessment is not liable to bias. According to those outcomes, F seems to be excessive and a drop in recruitment level is evident. Nevertheless, a significant decrease in SSB is not yet visible.
The effect on F and SSB of including the “International blue whiting spawning stock survey” in the assessment was not consistent for the various methods. Compared to the “Two surveys” configuration used in last year’s assessment, AMCI gave a higher F, ICA almost an identical F and SMS a much lower F.
Compared to last year, all the final exploratory runs estimated an increase in F2005. The models show mainly a small drop in F from 2005 to 2006.
Comparing the results of the assessment methods with each other (Figure 4.3.8.1) shows some differences in the terminal year with ISVPA-pessimistic as the most extreme with a steep increase in F and a very clear drop in SSB. The other final explorative runs estimate an F2006 in the narrow range 0.40-0.46 (and XSA F at 0.53). The recruitment in 2006 (age 1) is estimated very low by all models, whit the highest estimate from XSA Compared to the 2006 assessment the 2004 year class is now estimated lower, and the 2002 and 2003 year classes higher.
There are several important points on which all assessment methods agree. All methods predicted a severe decreasing trend in recruitment since 2001, with the 2006 recruitment being at a historic low. All methods agree that fishing pressure is well above FPA, and the TISVPA “pessimistic” estimate F is above Flim. SSB clearly shows a decrease from the peak in 2003, with the steepest decrease from ISVPA “pessimistic”/ bringing SSB below BPA.
The explorative runs showed that surveys indicate a rather stable SSB in the most recent years. Catch data and the assumption of a separable F model indicate however an increase in F, which in combination with a decrease in recruitment produces a rapid decline in estimated SSB. The relative weightings given to each of these information sources varies between the assessment methods and most likely accounts for the diverging estimates in both SSB magnitude and the fishing mortality trends.
The WG decided to bring the SMS assessments into the forecast. SMS has been used for the last two years as the final assessment method and SMS in its final configuration gave results similar to the AMCI, ICA and TISVPA/optimistic methods.
Input data are catch-at-age numbers (Table 4.2.2.7), mean weight-at-age in the sea and in the catch (Table 4.2.3.1.) and natural mortality and proportion mature in section 4.2.4. Survey data are presented in Table 4.3.2.1.
The key settings and data for the final blue whiting assessment in 2006 and 2007 are shown in the table below.
|
Settings/options for the final assessment
|
2006
|
2007
|
|
Software
|
SMS
|
SMS
|
|
Age range for the analysis
|
1–10+
|
1–10+
|
|
Last age a plus-group?
|
Yes
|
Yes
|
|
Recruitment in the terminal year
|
Estimated
|
Estimated
|
|
Catch data
|
|
|
|
Constant selection pattern for the catch fleet?
|
2 periods:
1981-1992,1993-2005
|
2 periods:
1981-1992,1993-2006
|
|
First age with age independent catchability
|
8
|
8
|
|
Age groups with the same variance
|
1, 2, 3-6, 7-10
|
1, 2, 3-6, 7-10
|
|
Age-structured tuning time-series
|
|
|
|
Norwegian spawning ground survey, ages 3–8,
|
1993-2006
|
1993-2003
|
|
First age with age independent catchability
|
7
|
5
|
|
Age groups with the same variance
|
3-4, 5-6, 7-8
|
3-4, 5-6, 7-8
|
|
International ecosystem survey in the Nordic Seas, ages 1-2
|
2000–2006
|
2000–2007
|
|
First age with age independent catchability
|
2
|
2
|
|
ages 1–2
|
2000–2006
|
2000–2007
|
|
Age groups with the same variance
|
1, 2
|
1, 2
|
|
International blue whiting spawning stock ground survey , ages 3–8
|
Not used
|
2004-2007
|
|
First age with age independent catchability
|
|
5
|
|
Age groups with the same variance
|
|
3-8, min std 0.4
|
Due to the short time-series for the “International blue whiting spawning stock ground survey” is was not possible to present a meaningful retrospective analysis of the assessments.
The model was run until 2006. The SSB January 1st in 2007 is estimated from survivors without taking the contribution from recruits into account. With the (expected) low recruitment this omission has practically no implications. The key results are presented in Tables 4.3.6.2–4.3.6.4 and summarized in Figure 4.3.6.6. Residuals of the model fit are shown in Figure 4.3.6.4 and Figure 4.3.6.5 and discussed in section 4.3.6.
The assessment results are presented in Figure 4.3.6.6.
Recruitment (age 1) has been at a lower level around 10 billion in the period 1981–1995 after which it increased to a higher level of around 35 billion in period 1996–2004. The highest recruitment at around 60 billions was reached in 2001 after which it declines almost linearly to a historic low level in 2006. Preliminary estimates for the 1–group in 2007 indicate also a very weak recruitment.
SSB was rather stable around Bpa in the period 1981-1995 after which it steadily increased to the highest observed level in 2003 at 7 million tonnes. SSB declined afterwards to around 4 million tonnes in 2007.
Fishing mortality increased from around 0.25 in the start of the 1980s to around Blim five years later. This was followed by historical low F level in start of the 1990s after which F steadily increased to a level of around Flim since 2000.
Average recruitment at age 1, 1981–2005 is estimated by SMS as 20.8 billions (arithmetic mean) or 14.8 billions (geometric mean). For the most recent period, 1996–2005, with high recruitment the values are 37.8 (AM) or 35.6 (GM) billions.
Last year the potential recruitment signals in different survey time-series were explored, and recruitment (age 1 in the assessment year) was predicted based on three of these. The same prediction models that were used last year are this year fitted to new recruitment estimates from the final assessment.
These models are as follows:
1 ) Linear regression on log-log scale with the Norwegian bottom trawl survey in the Barents Sea winter survey index (Section 4.2.5.4.)
2 ) Regression on natural scale forced through the origin with the International ecosystem survey in the Nordic Seas (Section 4.2.5.2), standard area.
3 ) Regression on natural scale forced through the origin with the International ecosystem survey in the Nordic Seas, full coverage.
All three regressions are illustrated in Figure 4.6.1. The Barents Sea index is catch per nautical mile of blue whiting smaller than 19 cm and the time-series is shown in Table 4.2.5.4. The two time-series estimates of (acoustic) age 1 from the International ecosystem survey in the Nordic Seas are shown in Table 4.2.5.1.2. The full coverage includes areas outside the standard area (which is shown in Fig. 4.2.5.5), and these areas may be important nursery areas for blue whiting. The time-series from the surveys with full coverage are therefore believed to give the best recruitment signal of the two.
The text table below shows estimated recruitment in 2006 and 2007 based on the three prediction models mentioned above. The variation between the different estimates is high for both years, but the general picture is that both the 2005 and 2006 year classes are weak. Information from commercial catches in 2006 (catch-at-age 1) and 2007 (age and length distributions in Figure 4.1.5.7.1.) support this impression.
|
Data Source
|
Time period (used in the regressions)
|
Year class 2005 (Recruitment 2006
at age 1 in billions)
|
Year class 2006 (Recruitment 2007
at age 1in billions)
|
|
WG07 assessment (SMS)
|
|
2.90
|
n.a.
|
|
Barents sea winter survey
|
1981–2006
|
17.83
|
8.38
|
|
International ecosystem survey in the Nordic Seas, standard area
|
2000–2006
|
0.435
|
0.016
|
|
International ecosystem survey in the Nordic Seas, full coverage
|
2001–2006
|
5.50
|
1.29
|
Short term forecasts were conducted with MFDP (Multi Fleet Deterministic Projection) Version 1a.
Input
A deterministic short term projection is presented based on final assessments, catch weights, stock weights and exploitation pattern are averaged over the previous three years (2004–2006). Population numbers and fishing mortalities were taken from the assessment outputs. Catches in 2007 is assumed at 1.8 million tonnes, which is the agreed rounded TAC (section 4.1.4).
Recruitment at age 1 in 2006 was assumed at 11.67 billion. Recruitment at age 1 in 2007 was assumed at 4.84 billion. Both of these estimates are obtained by taking the mean of two survey-based estimates: one from the International ecosystem survey in the Nordic Seas with full coverage and one from the Barents Sea winter survey. The reason for not using the final assessment estimate of recruitment at age 1 in 2006 is that this is unrealistically low and appears as an extreme outlier (Figure 4.6.1, upper panel). Both estimates are still quite low. Geometric mean recruitment for the period 1981–2005 (14.8 billion) was used for recruitment in 2008.
Table 4.7.1.1 gives an overview of the input data.
Output
The predicted catch and SSB from the forecasts are presented in Table 4.7.1.2.
Fishing at Fpa = 0.32, is associated with expected landings of 845 thousand tonnes in 2008 and an SSB of 2.60 million tonnes in 2009, slightly above the precautionary limit, Bpa = 2.25 million tonnes.
No medium forecast was done this year.
Last year ICES made a review of the present blue whiting management plan (Section 4.1.4). ICES considered that given the high recruitment level observed for the period 1996–2005, the management plan is robust to uncertainties in both assessment and implementation. For low recruitment scenarios, the management plan is not robust to these uncertainties, unless there are unrealistically low levels of noise and bias in both stock assessment estimates and implementation of the TAC. ICES concludes that the management plan is not precautionary as the lower recruitment scenario is plausible (given that it was the case in the past) and, under this scenario for range of realistic assessment and implementation bias, there is higher than 5% probability that the spawning stock biomass will fall below Blim.
The assessment this year highlights that the blue whiting assessment is very uncertain and that the 2005 and probably also the 2006 year-classes are low and definitely outside the range of recruitment referred to by the ICES review as high recruitment. This emphasis that the present management plan with a reduction of landing of (minimum) 100 000 tonnes per year until Fpa is reached is far from sufficient to maintain the stock within safe biological limits.
The present precautionary reference points have been introduced in the advice of ACFM in 1998. The values and their technical basis are:
|
Reference point
|
Bim
|
Bpa
|
Flim
|
Fpa
|
|
Value
|
1.5 mill t
|
2.25 mill. t
|
0.51 yr-1
|
0.32 yr-1
|
|
Basis
|
Bloss
|
Blim* exp(1.645*σ), with σ= 0.25.
|
Floss
|
Fmed
|
Although problems have been identified with these reference points they have remained unchanged since then. A major problem is that fishing at Fpa implies a high probability of bringing the stock below Bpa, in other words the present combination of Fpa and Bpa is inconsistent.
The Workshop on Limit and Target Reference Points (WKREF) considered the biological reference points for Blue Whiting at a meeting in Gdynia, Poland in January this year (ICES CM 2007/ACFM:05). The original reference points for this stock were set in 1998, before the era of high productivity became apparent. The group examined the consequences of these new observations on the reference points by first splitting the time-series into two productivity regimes (low productivity from 1981–1994, and high productivity from 1995–2005). Standard methods (i.e. using the guidelines from the Study Group on Precautionary Reference points, SGPRP (ICES CM 2003/ACFM:15)) were then used to re-estimate the reference points, which were found to be comparable to the current values. A new probabilistic approach for estimating Blim was also employed, but again, the result was found to be comparable with the current values. The group concluded that there was no basis for revising the current reference points. WKREF also noted that there may be no need for different Blim values in different productivity regimes.
The assessments presented this year should be considered as very uncertain. As may be seen in Figure 4.3.8.1 the estimated spawning stock and fishing mortality vary considerably between the different models that were run. All the models should be using the same data so this difference between models lies mostly in different weighting of data. A statistical approach (SMS), where the data sources are weighted according to their uncertainty and methods (AMCI, ICA) based on fixed weighting of the data sources gave a stable F in the most recent years and moderate decline in SSB. TISVPA can produce both a stable F and a steep increase in F alternative with likely configurations. The problem for all types of models seems to be conflicting information from the catch and survey data in combination with a relatively high uncertainty of the acoustic estimates of the stock size.
The difference in the model results is only compared from the mean values of the SSB, F and recruitment. The confidence intervals of those are however very large. Figure 4.3.4.4 presents uncertainties of SSB and stock numbers estimates from the two alternative TISVPA runs. The 90% confidence interval of SSB ranges from around 2 to 10 million tonnes for the “pessimistic” alternative. The uncertainty of the output from the final assessment (Figure 4.3.6.10) is high and has increased compared to the assessment done last year, even though the methods used are the same. Such uncertainties are usually an underestimate of the “real uncertainty” as uncertainties in the structural model are not included.
The assessment is consistent with the results presented last year. In 2006 ICES stated that the maximum catch in 2007 corresponding to the existing management plan is 1.9 million tonnes, is expected to leave the spawning stock biomass at 2.86 million tonnes in 2008. The short term projection made this year estimates an SSB of 3.01 million tonnes for 2008, with an assumed landing of 1.8 million tonnes in 2007.
While there is substantial uncertainty about the precise value of both F and SSB in the most recent year, the recruitment estimates for the terminal year are more consistent between methods. However, the 2005 year class derived from the surveys alone is estimated 4 times higher than the estimate from the analytical assessments using both survey and catches. Both estimates of the 2005 year class are in the low end of the historical time-series. The estimate of the 2006 year class is mainly coming from surveys and the estimated low level has not fully been confirmed by the fishery as catch data from 2007 on the immature stock component are not yet available.
Based on the most recent estimates of fishing mortality and SSB, ICES classifies the stock as having full reproductive capacity, but being harvested unsustainably.
The 2005 and probably also the 2006 year classes are in the low end of the historical time-series. It is impossible to say if this is a shift towards the low recruitment regime, as observed in the period before the mid 1990s, or just an anomaly. SSB has declined since its historical peak in 2003 at around 7 million tonnes to 4.3 million tonnes in the beginning of 2007. Landings of just below 2 million tonnes in 2007 in combination with the small 2005 year class will reduce SSB further.
Blue whiting fisheries have entered a new era as agreement on a management plan was reached in the end of year 2005. This will guard against excessively high catches, but is no guarantee on sustainability if productivity of the stock declines. The ICES review of the plan pointed out that that the management plan is not precautionary in a situation with a (continued) low recruitment, as observed before 1995, when a realistic level of assessment uncertainty is taken into account.
The current estimate of the size of the blue whiting stock is uncertain because commercial catch data and data from scientific surveys give conflicting signals. However, all models estimate a considerable decline in SSB since 2003 and a fishing mortality which currently is above the precautionary level. Fishing at Fpa in 2008 is predicted to give a TAC at 845 thousand tonnes and a SSB at 2600 thousand tonnes, which is just above Bpa.
Understanding of the factors which drive blue whiting recruitment is poor. In the past decade, recruitment of blue whiting has been high compared to the period before. Presently there are however strong indications that the 2005 year class is weak. The available information indicates that the 2006 year class is weak as well. These two year classes might be an order of magnitude lower than the preceding 10 year classes. The consequence of this is that the biomass of blue whiting will continue to decrease rapidly in the near future. We will therefore be facing a collapse in the blue whiting stock if the present fishing pressure is not substantially reduced.
Sampling intensity for age and weight of herring and blue whiting are made in proportion to landings according to CR 1639/2001 and apply to EU member states. For other countries there are no guidelines. Current precision levels of the sampling intensity are unknown. The group recommends, as part of the ToRs for the next WG meeting a review of the sampling scheme on a scientific basis and provide guidelines for sampling intensity.
WG recommends an analysis of the change in mean weight as part of the ToRs for the next WG meeting.
![[FSF Associate Member]](http://mmeixide.pbworks.com/f/FSF.png){kind=small}
Comments (0)
You don't have permission to comment on this page.