ICES HAWG REPORT 2014 1031

Annex 11 – Stock Annex – Sprat in the Celtic Seas Region Sprat in VI and VII, minus VII d e

Stock Sprat in Subarea VI and Divisions VIIa–c and f–k (Celtic Sea and West of Scotland)

Working Group HAWG

Date February 2013

Revised by WKPELA 2013 / Beatriz Roel

A. General

A.1. Stock definition

Most sprat fisheries in the Celtic Seas area are sporadic and occur in different places at different times. Separate fisheries have taken place in the Minch, and the Firth of Clyde (VIaN); in Donegal Bay (VIaS); Galway Bay and in the Shannon Estuary (VIIb); in var-ious bays in VIIj; in VIIaS and in the Irish Sea (VIIaN):

The stock structure of sprat populations in this ecoregion is not clear. There is insuffi-cient information to conclude that Subareas VI and VII constitute a management unit

1032 ICES HAWG REPORT 2014

for sprat, and further work is required and further work is required to solve this. For the time being, ICES will give advice covering the whole area.

A.2. Fishery

Most sprat in the Celtic Seas ecoregions are caught by small pelagic vessels that also target herring, mainly Irish, English and Scottish vessels. In Ireland, many polyvalent vessels target sprat on an opportunistic basis. At other times these boats target demer-sals and tuna, as well as pelagics. Targeted fishing takes place when there are known sprat abundances. However the availability of herring quota is a confounding factor in the timing of a sprat targeted fishery around Ireland. In Ireland, larger sprats are sold for human consumption whilst smaller ones for fish meal. Other countries mainly land catches for industrial purposes.

Sprat may also be caught in mixed shoals with herring. The level of discarding is un-known.

No TACs or quotas for sprat exist in this ecoregion. Most sprat catches are taken in small-mesh fisheries for either human consumption or reduction to fishmeal and oil. It is not clear whether bycatches of herring in sprat fisheries in Irish and Scottish waters are subtracted from quota.

A.3. Ecosystem aspects

It is difficult to assess predation impacts on sprats in this area. This is particularly true of Area VIa where sprats tend to be aggregated in sealochs where predator populations are likely to differ from those in open sea areas of VIa.

Information on seabird diets is available for part of the area. The very limited data on seabird diets from Irish Sea colonies (for kittiwakes, auks and terns) suggests that sprats are more important in seabird diets in this area than is generally the case else-where.

Chivers et al. (2012) showed that kittiwakes breeding at colonies in the Irish Sea fed predominantly on sprats, at least in the two years of their study. Their data also sug-gested a link between diet composition and breeding success, with kittiwake breeding success in their Irish Sea study colonies being higher when sprats formed a higher pro-portion of the kittiwake diet. Royal Society for the Protection of Birds (RSPB) observers carried out surveys of kittiwake numbers during sprat acoustic surveys in the Irish Sea in 1995 and 1996 and found no evidence of a correlation between sprat densities and kittiwake densities at sea (P-J. Schön, pers. comm.).

B. Data

B.1. Commercial catch

Landings are considered representative of catches; there are no indications of discard-ing or misreporting. There is no information on catches in number /length, or weight in the catch for sprat in this ecoregion. No data on mean length, weight-at-age or ma-turity-at-age in the catch are available.

Division VIa (west of Scotland and northwest of Ireland)

Landings have been dominated by UK-Scotland and Ireland. The Scottish fisheries have taken place in both the Minch and in the Firth of Clyde. The Irish fishery has always been in Donegal Bay. Despite the wide separation of these areas, the trends in

ICES HAWG REPORT 2014 1033

landings between the two countries are similar, though the UK data have always been higher. Irish data may be underestimated, due to difficulties in quantifying the land-ings from vessels of less than 10 m length.

Scottish landings were consistently above 4000 t in the 1970s, declining to about 1000 t in 1979. They fluctuated between 500 t and 4000 t until 1994 and then increased mark-edly to over 8000 t in 1999, after which they have declined substantially.

Division VIIa (Irish Sea)

The main historic fishery was by Irish boats, in the 1970s, in the western Irish Sea. This was an industrial fishery and landings were high throughout the 1970s, peaking at over 8000 t in 1978. The fishery came to an end in 1979, due to the closure of the fishmeal factory in the area. It is not known what proportion of the catch was made up of juve-nile herring, though the fishing grounds were in the known herring nursery areas. In the late 1990s and early 2000s, UK vessels landed up to 500 t per year. In recent years a trial fishery for sprat was carried out by the vessels that fish herring in the area. This was carried out to investigate the feasibility of a clean commercially viable sprat fish-ery. The results of the trials were inconclusive and plans to conduct further experi-ments are under discussion. Irish Landings from 1950–1994 may be from VIIaN or VIIaS. Recent Irish landings are mainly from VIIaS, mainly Waterford Harbour.

Divisions VIIbc (west of Ireland)

Sporadic fisheries have taken place, mainly in Galway Bay and the Mouth of the Shan-non. The highest recorded landing was in 1980 and 1981 during the winter of 1980/1981, when over 5000 t were landed by Irish boats. This fishery took place in Gal-way Bay in the winter of 1980/1981 (Department of Fisheries and Forestry (1982). Since the early 1990s landings fluctuated from very low levels to no more than 700 t per year. Irish data may be underestimated, due to difficulties in quantifying the landings from vessels of less than 10 m length.

Divisions VIIg-k (Celtic Sea)

Sprat landings in the Celtic Sea from 1985 onwards are ICES estimates. In the Celtic Sea, Ireland has dominated landings. Patterns of Irish landings in Divisions VIIg and VIIj are similar, though the VIIj landings have been higher. Landings for VIIg and VIIj were aggregated in this report. Landings have increased from low levels in the early 1990s, with catches fluctuating between 0 t in 1993 and just under 4200 t in 2005. Irish data may be underestimated, due to difficulties in quantifying the landings from ves-sels of less than 10 m length.

B.2. Biological

Sampling results of landings (in number length, mean length, weight-at-age or ma-turity, or weight in the landings) are not available for sprat in this ecoregion.

B.3. Surveys

There are many surveys taking place in this area. Only the ones that are considered relevant to describe trends in (part of the) sprat population in the area are described here.

1034 ICES HAWG REPORT 2014

Survey acronym full name

Type of survey / gear Comments years Quarter

2 SWIBTS South West International Bottom Trawl Survey

GOV Survey has been terminated; Index should be split between

VIIa and Celtic (North)

2003–2011 Q4

6 ACVIaLochs Scottish Sea lochs surveys

acoustic/ mid-water trawl

Survey targeted for sprat in six lochs inside the Isle of Skye. No fisheries in this area, doesn’t cover the stock.

2001–2005 Q1Q4

7 SWCGFS6a Scottish Western Coast VIa Groundfish Survey - Quarter 1

GOV Age–length information may be of value in absence of fishery data.

1981–2012 Q1

8 SWCGFS6a Scottish Western Coast VIa Groundfish Survey - Quarter 4

GOV Age–length information may be of value in absence of fishery data.

1990–2012 Q4

10 ACclydeQ3 Clyde acoustic survey for herring and sprat

acoustic/ midwater trawl

Survey-series discontinued but may be informative in combination with new survey (11).

1985–1989 Q3

11 ACclydeQ4 Clyde acoustic survey for herring and sprat

acoustic/ midwater trawl

Survey is informative for sprat, to be continued; no fisheries in the area

2012– Q4

12 AC(VIIaN) Irish Sea Acoustic Survey

acoustic /midwater trawl

Survey designed for herring but surveys sprat all right; no fisheries in the area

1998–2012 Q3

Survey 2) International Bottom Trawl Survey (2003–2011)

In order to derive an index of sprat abundance from this survey the total number of sprat per haul was first standardized to numbers per hour. An average for each rectan-gle was then calculated on a spatial basis as a basis for an annual index constructed as mean sprat catch per hour trawling. A final index can then be calculated as the mean of the mean number per hour from rectangles fished for at least 75% of the time-series.

Examination of results from the IBTS survey (Figures 6.6.3 and 6.6.4) suggest that the stock has fluctuated during the period of the survey with no indication of stock decline. Regrettably, the IBTS series has now been terminated.

ICES HAWG REPORT 2014 1035

Survey 6) Scottish Sea lochs surveys

Between 2001 and 2005 a series of detailed surveys were carried out in selected main-land Scottish sea lochs, to the east of the Isle of Skye. This survey targeted sprat, but does not cover the stock. There are no fisheries in this area.

This type of survey could inform the assessment in this region but the time-series is not relevant to the current state of the stock.

Survey 7–8) Scottish Western Coast VIa Groundfish Survey; Quarter 1 and Scottish Western Coast VIa Groundfish Survey; Quarter 4

Sprat caught throughout the survey area, even though the survey does not target sprat. Due to gear used and survey design unlikely to produce reliable abundance of sprat.

Age–length information is available and can be used to indicate stock structure. Survey esti-mates need verification before they are used as indicators of biomass because they are based on a bottom-trawl gear.

Survey 10–11) Clyde Acoustic surveys for herring and sprat

A Clyde herring and sprat acoustic survey was carried out in June/July 1985–1990 and then discontinued. Biomass estimates from all years as well as lengths and ages from some years are available from this survey. In 2012 this survey was reinstated as an October/November survey and results from the first survey are being processed at the moment.

Age and length distribution from this survey are available and can be used to indicate stock structure. The survey will need more datapoints to be able to give stock trend information.

Survey 12) Irish Sea Acoustic Survey

This survey uses a stratified design with systematic transects, during the first two weeks of September. Time-series of sprat biomass are available and the associated CVs are generally low which suggests a more continuous distribution than is the case for herring (M. Armstrong, pers comm.).

The vessel currently used for the surveys is the R.V. Corystes (UK(NI)) replacing the R.V. Lough Foyle (UK(NI)). Starting positions are randomized each year. The survey is most intense around the Isle of Man (2 to 4 nautical mile transect spacing) where highest densities of adult herring are expected based on previous surveys and fishery data. Transect spacing of 6 to 10 NM are used elsewhere. A sphere-calibrated EK-500 38kHz sounder is employed, and data are archived and analysed using Echoview (So-narData, Tasmania). Targets are identified by midwater trawling. Acoustic records are manually partitioned to species by scrutinising the echograms and using trawl compo-sitions where appropriate. ICES recommended target strengths are used for herring, sprat, mackerel, horse mackerel and gadoids. The survey design and implementation follows, where possible, the guidelines for ICES herring acoustic surveys in the North Sea and west of Scotland. The survey data are analysed in 15 minute elementary dis-tance sampling units (approximately 2.5 NM). An estimate of density by age class, and spawning–stock biomass, is obtained for each EDSU and a distance-weighted average calculated for each stratum. These are raised by stratum area to give population num-bers and SSB by stratum.

Examination of the time-series of biomass based on this survey suggest that since 2005 the stock has fluctuated at a lower level than in previous years. There are no fisheries for sprat in this area.

1036 ICES HAWG REPORT 2014

B.4. Commercial cpue

Commercial landings per unit of effort were not calculated, because lpue is not consid-ered an indicator relevant to describe the stock. This has to do with the sprat biology (schooling behaviour), the fact that the stock structure is not clear, and indications from the industry that factors unrelated to stock availability determine the landings.

C. Assessment: data and method

No analytic assessment is put forward for this area. The main indicators for sprat in parts of the area are survey indices that are considered representative for stock trends:

a ) South West International Bottom Trawl Survey (SWIBTS); b ) Scottish Sea lochs surveys (ACVIaLochs); c ) Scottish Western Coast VIa Groundfish Survey - Q 1 and Q4 (SWCGFS6a;

age–length information); d ) Clyde acoustic survey for herring and sprat (ACclydeQ3); e ) Clyde acoustic survey for herring and sprat (ACclydeQ4); f ) Irish Sea Acoustic Survey (AC(VIIaN)).

D. Short-term projection

No short-term projections are put forward.

E. Medium-term projections

No medium-term projections are put forward.

F. Long-term projections

No long-term projections are put forward.

G. Biological reference points

No precautionary reference points are defined for sprat populations in this region.

Molloy and Bhatnagar (1977) estimated F0.1 separately for Irish Sea and Celtic Sea sprat populations. They concluded that the Celtic Sea population could withstand a higher F. The estimates of F0.1 were F = 0.5 and F = 0.8 for Irish Sea and Celtic Sea respectively.

H. Other issues

H.1. Historical overview of previous assessment methods

Sprat in this area was first assessed by ICES in 2011. No analytic assessment has been put forward so far.

I. References Chivers, L.S., Lundy, M.G., Colhoun, K., Newton, S.F. and Reid, N. 2012. Diet of black-legged

kittiwakes (Rissa tridactyla) feeding chicks at two Irish colonies highlights the importance of clupeids. Bird Study 59: 363–367.

ICES. WKSPRAT. 2013. Benchmark Workshop on Sprat Stocks. 11–15 February 2013, ICES HQ, Copenhagen. ICES CM 2013/48.

ICES HAWG REPORT 2014 1037

Molloy, J. and Bhatnagar, K. 1977. "Preliminary Investigations of the Sprat stocks off the South coast of Ireland", Irish Fisheries Leaflet, Department of Agriculture and Fisheries (Fisheries Division) 1977.

1038 ICES HAWG REPORT 2014

Sprat in Divisions VIId,e

Stock Sprat in Divisions VIId,e.

Working Group HAWG

Date February 2013

Revised by WKPELA 2013 / Beatriz Roel

A. General

A.1. Stock definition

Divisions VIId and VIIe comprise a management unit for sprat, with an annual TAC being set by the EC. However it is not clear if sprat populations in this area constitute a unit stock; and if this is an appropriate management unit. Until more information is available, advice will be given for this unit, mainly based on information from the Lyme Bay (ICES statistical rectangles 29 and 30 E6,7).

Most of the catch is taken in Lyme Bay in Subdivision VIIe, where 88% on average of landed sprat are caught.

Figure 5.1.2. Sprat in VIId,e, Lyme Bay. ICES statistical rectangles that constitute Lyme Bay are indicated.

A.2. Fishery

In Lyme Bay the primary gear used for sprat is midwater trawl. Within that gear type three vessels under 15 m actively target sprat and are responsible for the majority of landings (since 2003 they took on average 94% of the total landings). Sprat is also caught by driftnet, fixed nets, lines and pots. Most of the landings are sold for human consumption. The fishery starts in August and runs into the following year into Feb-ruary and sometimes March.

Sprat may also be caught in herring fisheries mixed shoals with herring. The level of discarding is unknown.

ICES HAWG REPORT 2014 1039

A.3. Ecosystem aspects

Fishermen find sprat by sonar search and sometimes the shoals have been too far off-shore for sensible economic exploitation. Skippers then go back to other trawling ac-tivity. This offshore/ near shore shift may be related to environmental changes i. e. temperature and/or salinity.

B. Data

B.1. Commercial catch

Sprat landings prior to 1985 in VIId,e were extracted from FishBase, from 1985 onwards they are ICES estimates. Since 1985 sprat catch has been taken mainly by UK, England and Wales, with some substantial catches taken by Denmark in the late 1980s. Early landings from Denmark are being looked into as there may be some discrepancies be-tween FishBase and ICES data.

UK landings data are available by gear type from 1981 to date. For trawlers, associated effort was recorded both as number of days and hours fishing. In the case of driftnets effort corresponds to number of hauls times the total number of nets and for gillnet the length of the gear (m) times the number of days fishing. Technological improvements in the fishery such as high technology sonars (such as CH 32), were put in place in the early 1980s.

There is a TAC for sprat for VIId,e, English Channel.

B.2. Biological

Catch sampling information was not available for ICES.. Biological information was collected in the acoustic surveys. Age composition data suggested a majority of age 2 in the survey area. Ages 0, 1, 3 and 4 were also represented. Percentages at age by survey are compared in the plot below.

B.3. Surveys

Acoustic surveys covering the area where the fishery operates were conducted in 2011 and 2012. The surveys are carried out in October, coinciding with the early months of the sprat fishing season, which runs from September to February. The survey included a series of pre-designed parallel, equidistant (10 nautical mile, nmi) transects perpen-dicular to the coast, covering the ICES rectangles where most of the annual sprat catches in the past decade (Roel et al., 2011) have been made, with particular focus on the western part of Lyme Bay. The pre-designed transects covered a larger area than

%

2011

1040 ICES HAWG REPORT 2014

was feasible, but based on previous experience, it was anticipated that the main sprat concentrations were likely to be found in a relatively small part of the whole area. Given that the location and extent of the sprat distribution was not known in advance of the survey, an adaptive design was adopted, with transects being dropped as the biomass dropped progressively from earlier transects.

A local sprat fishing vessel was chartered for the survey: the 11.98 m FV Mary Anne. Although its size restricted its range and speed to some extent, it was imperative to the programme that the survey be conducted using a local fishing vessel experienced in fishing for the target species, i.e. sprat. Also, the aim of the project was to quantify the sprat population targeted by the inshore fishery, and we were confident that the vessel could cover the area of interest adequately.

Acoustic surveying took place during daylight, because sprat disperse into loose ag-gregations at night (Plirú et al., 2011), so would be more difficult to detect acoustically then. Where possible, the transects were completed from east to west because anecdo-tal information suggested that sprat move in from the west. Surveying the transects in the opposite direction, therefore, would have increased the likelihood of double count-ing.

Fisheries acoustic data

Scientific quality acoustic data were collected using a portable EK60 Simrad echo-sounder operating at 120 kHz, connected to a Furuno GPS. Ping rate was set to 0.4 s–1 and pulse duration to 0.256 µs, to collect high-resolution data. The transducer was at-tached to an over-the-side mount on the port side of the vessel. The mount consisted of a vertically orientated aluminium pole that, when deployed, protruded 2 m below the surface, so that the transducer remained clear of the bubbles formed by the hull during steaming. A 5 inch fin was attached to the aft side of the pole to prevent vortices developing as a result of the drag through the water; they would cause the pole to vibrate during steaming (see van der Kooij et al., 2011, for more detail).

Prior to the survey, the echosounder was successfully calibrated outside Torquay har-bour using a 23 mm copper sphere following standard methods (Foote et al., 1987). Depending on weather, the vessel speed during the acoustic transects was a constant 7 knots. Faster than that and in adverse weather, noise would have reduced the quality of the acoustic data.

Acoustic data were recorded continuously. A scanning sonar, traditionally used by the fishery to search for sprat schools, was switched off while running the acoustic tran-sects because it interfered with the acoustic data. When marks were encountered on-transect and a decision made to fish, the vessel would come off-transect and use the sonar to track the schools that had been seen on the echosounder. After completion of the haul, the sonar was switched off again, and the transect resumed where it had been interrupted.

The acoustic data were cleaned and processed after the survey using the processing software Echoview version 5.3 (Myriax). Acoustic data collected during fishing opera-tions and the steam to and from the transect were discarded, retaining only on-transect data. Surface aeration caused by bad weather was removed, setting a surface exclusion line, and acoustic data from closer to the seabed than 0.5 m were also removed, to ex-clude the strong signals from the seabed. Owing to the presence of occasional noise, interference and weaker scatterings caused by other organisms, several algorithms were applied so that only sprat schools were extracted from the raw data. This included a filter to remove all non-clupeid backscatter and a backscatter threshold (of -60 dB;

ICES HAWG REPORT 2014 1041

Figure 2). Sprat schools were identified based on a combination of expert knowledge and trawl catches.

As small numbers of other species were caught in the trawl, acoustic energy was par-titioned by species, based on the weight ratios obtained from the trawl catches. Macke-rel, however, were not considered because they only give a weak signal at 120 kHz and were automatically filtered out using the algorithm mentioned above.

Trawling

Trawling is conducted using the vessel’s standard commercial midwater gear designed to catch sprat. As sprat biomass was calculated from the acoustic data, trawl catches were used only to establish the species composition of the acoustic marks, and to collect biological material on the pelagic fish community, in particular length frequency, and age and maturity information. This meant that only relatively small catches were re-quired, so the skipper ensured this by carefully monitoring the trawl procedure using a combination of the sonar and the netsonde.

Once on board, the catch was sorted by species. Fish were counted and measured to the nearest 0.5 cm or 1.0 cm, depending on species. When catches of a species were very large, a subsample of that species was taken. At every station, however, five sprat from each length category were collected and retained on ice, then once ashore, were frozen and taken on board the RV Cefas Endeavour for further analysis: length, weight, sex and maturity were recorded, and otoliths were extracted for age determination.

Biomass calculation

The acoustic density attributed to sprat (sA or Nautical Area Scattering Coefficient) was converted into numbers according to standard procedures. First, the TS was calculated for each sprat length group:

TS = 20LogL + b20, (1)

where b20 was -74.2 dB at 120 kHz (Saunders et al., 2012), and L was the fish length group. This was converted into the backscattering cross section for each length group:

σ = 4π10(TS/10), (2)

This in turn provided the weighted average backscattering cross section per individual fish. Dividing the sA or the NASC (mean acoustic energy attributed to sprat) per nmi by this number yielded the mean number of sprat per nmi, which was converted into biomass by multiplying by the mean weight of sprat. Because fish weight could not be determined accurately on board the commercial fishing vessel, the mean weight was derived as follows: a length–weight relationship was calculated based on trawl sam-ples in the area obtained from the international bottom trawl survey which takes place in November. The calculated weight of a sprat at a mean length of 13.27 cm was 16.57 g. The biomass was calculated separately for each of the four ICES rectangles covered in the survey.

B.4. Commercial lpue

A midwater trawl landings per unit of effort (lpue) series were constructed based on the three vessels that take most of the catch around Lyme Bay. Lpue is calculated based on the landings per hour away from port. Annual lpue is presented per hour and by fishing season, which runs from August to February–March depending on the year but referred to by the year when the season started.

1042 ICES HAWG REPORT 2014

Communication with fishermen that target sprat in the Channel has indicated that the fish may not be found on occasions so, if lpue was to be used as an indicator of stock abundance it may be preferable to include all effort spent which would include search-ing time. If there were no landings in August or March the effort in those months was excluded when computing lpue.

C. Assessment: data and method

For Lyme Bay sprat there is information that can be put forward to inform advice. For the rest of Divisions VIId,e there is insufficient information to assess the state of the stock. The majority of sprat caught in VIId, e is taken in the Lyme Bay area. The pro-portion of landings between Lyme Bay and the rest of the area should be monitored to ensure that the assessment covers the main part of the fisheries.

Lyme Bay

Relevant information to inform the advice for the Lyme Bay area are:

• trends in the lpue since 1988; • trends in the acoustic survey index since 2011.

Exploratory (Schaefer, Bayesian) assessments will be further developed before they can be benchmarked.

CATCH ADVICE

Catch advice was based on category 3 (WKLIFE 2012) according to the data and anal-ysis that were available. This category includes stocks for which survey indices (or other indicators of stock size such as reliable fishery-dependent indices are available that provide reliable indications of trends in stock metrics such as mortality, recruit-ment and biomass. Those are a time-series of lpue (1988- 2012) and two acoustic sur-veys (2011 and 2012) performed in the area where the fishery takes place in the vicinity of Lyme Bay.

A Depletion-Corrected Average Catch (DCAC) procedure was implemented but based on the data available it was not considered appropriate.

Data and computations

Catch and lpue data and, predicted lpue based on the surplus production model for 2008 – 2012 were used. The period chosen was based on common practice.

ICES HAWG REPORT 2014 1043

Catch advice in 2014 is computed according to the following equation:

𝐶𝐶𝑦𝑦+1 = 𝐶𝐶𝑦𝑦−1 �∑ 𝐼𝐼𝑖𝑖𝑦𝑦−1𝑖𝑖=𝑦𝑦−𝑥𝑥 𝑥𝑥⁄

∑ 𝐼𝐼𝑖𝑖𝑦𝑦−𝑥𝑥−1𝑖𝑖=𝑦𝑦−𝑧𝑧 (𝑧𝑧 − 𝑥𝑥)�

�

Where I is the survey index, x is the number of years in the survey average, and z > x. For example, x = 2 would be a two year survey average, and x = 2 z = 5, which is anal-ogous to the five steps in the ICES MSY transition from 2010 to 2015 (ICES Introduction 1.2);

An Uncertainty Cap and the Precautionary buffer were also applied. The catch advice was based on the observed lpue as the surplus production model is exploratory and not an accepted assessment at present.

D. Short-term projection

No short-term projections are put forward.

E. Medium-term projections

No medium-term projections are put forward.

F. Long-term projections

No long-term projections are put forward.

G. Biological reference points

No precautionary reference points are defined for sprat populations in this region.

H. Other issues

The advice for sprat in Lyme Bay can be better informed if in time the acoustic survey is long enough to be able to tune this with other tuning data such as the lpue and land-ing statistics such as number or weight in the landings, mean weight-at-age or ma-turity-at-age. When this is will be the case depends on the length of the time-series (at least five years) and the consistency with other information.

Catch lpue pred lpue2008 1029.26 938.0082009 773.19 896.9422010 4404.294 1526.95 898.3972011 3136 1047.41 827.9812012 4434 1988.61 835.466

C(2010-2012) 3991.431Unc Cap + 4789.718Unc cap - 3193.145lpue(2011-12) 1518.01 831.7235lpue(2008-10) 1109.8 911.1157

1044 ICES HAWG REPORT 2014

H.1. Historical overview of previous assessment methods

ICES has started to give quantitative advice for this datalimited stock in 2012.

I. References Foote KG, Knudsen HP, Vestnes G, MacLennan DN, Simmonds E.J. 1987. Calibration of acoustic

instruments for fish density estimation: a practical guide. Report No. 144. ICES, Copenha-gen.

Plirú A, van der Kooij J, Engelhard GH, Fox CJ, Milligan SP, Hunter E. 2011. Is recruitment of plaice in the Irish Sea constrained by predation of eggs by sprat? Report No. ICES CM 2011/H: 33.

Roel, B. A., Readdy, L., and van der Kooij, J. 2011. Review of UK data for sprat in the English Channel (VIId, e). Working Document presented to the Herring Assessment Working Group 2011, Copenhagen.

Van der Kooij, J., Brown, D., and Roel, B. A. 2011. Western Channel sprat (Sprattus sprattus L.) population assessment. FSP Programme Report, 32. 16 pp. Available at: http://www.cefas.defra.gov.uk/media/585780/mf050_report2011_vfinal2.pdf.