International Bottom Trawl Survey

The IBTS consists of a number of national surveys that aim at improving standardisation and collaboration between surveys. Two areas can be distinguished that differ in terms of the length of time spent on this trajectory and hence the degree to which standardisation was achieved: IBTS North Sea and IBTS Western and Southern areas.

Although there are marked similarities between these two areas and both are represented in the ICES IBTS Working Group it was established at the 1999 meeting of this working group that due to the considerable difficulties in merging the protocols used in the North Sea with those used in the western and southern divisions, two separate manuals are the standard.

IBTS North Sea

In the North Sea the IBTS started in the 1960's as a survey that was directed at juvenile herring and was at that time called the International Young Herring Survey (IYHS).

As it was gradually realised that the survey also yielded valuable information for other fish species, such as cod and haddock, the objectives were broadened and the survey was renamed into the International Young Fish Survey (IYFS). Besides the IYFS, which was carried out in the first quarter, a number of national surveys developed in the 1970's and 1980's that were mainly carried out in the third quarter.

In 1990 ICES decided to combine the international and the national surveys into the IBTS. The IBTS is carried out twice per year (1st and 3rd Quarters) since 1997 and on a quarterly basis in the period 1991-1996.

Prior to 1977 there was no standardisation of gear although all ships used bottom trawls with a small mesh cover. In 1977 ICES recommended that all ships should use a GOV trawl as specified by the Institut des Peches Maritimes, Boulogne. A detailed description of the net is to be found in the manual (Anon. 2004). The GOV trawl has been gradually phased in, e.g. in 1979 only 3 vessels were equipped with the GOV trawl, but by 1983 all 8 nations were using this gear. It should be noted that although the gear is now standard, variations in the rigging exist between the various countries.

The fishing method is also standardized and described in the manual (Anon. 2004). Fishing speed is 4 knots measured as trawl speed over the ground. In 1977 ICES also recommended that the duration of a tow should be reduced from an hour to half an hour with the catch data to be expressed in numbers per hour. All nations accepted this recommendation although it was a number of years before 30 minutes became the standard.

IBTS Western and Southern areas

In 1994, it was suggested to extend the remit of the IBTS Working Group to co-ordinate the surveys in the western and southern areas (i.e. English Channel, Celtic Sea, Bay of Biscay, eastern Atlantic waters from the Shetlands to the strait of Gibraltar) and contacts were established with the national laboratories involved to co-ordinate and standardise these surveys.

The following national bottom trawl surveys were identified in the area:

International coordination of surveys in this region began in 1997 and was based on a previous EU project - SESITS (South-western European Shelf International Trawl Surveys). The surveys covered in this project concentrate on the third quarter and involve France, Spain and Portugal.

Each country conducts surveys in adjacent areas with no overlapping. While some attempts have been made in order to achieve a consensus on the choice of a standard gear, this was not achieved due to the variation in bottom types, and each country uses a different gear (GOV for France, BAKA for Spain and Norwegian Campelen Trawl for Portugal).

With the sampling protocols, however, a significant level of standardisation was achieved, and all three countries are using depth-stratification in their sampling strategy. Building on the success of this project, joint projects with England, Ireland and Scotland were initiated. As a consequence the level of standardisation with these countries is lower; e.g. some countries use depth strata, others ICES rectangles.

Northern Ireland survey

In the Northern Ireland survey, the sampling design is stratified with fixed-position stations. Stratification is by depth and seabed type. Haul duration is 3 nautical miles at 3 knots over the seabed. Stations in the St George's Channel are 1 nautical mile at 3 knots and have only been surveyed since October 2001. Number of stations is 45 in northern Irish Sea and 12 in St George's Channel. Tows are during day time only.

The surveys have been carried out in March and October since 1992. From March 1992 to March 2001, the survey extended from 54° 43' N to 53° 22' N. From October 2001 the survey is extended into the St George's Channel to 52° 18' N. Depth range is from 20 to 120m.

The surveys are carried out on the R.V. Lough Foyle, a 43.5m stern trawler of 880 kw and GRT 547 tonnes. The fishing gear is a rock-hopper otter trawl with a 17m footrope fitted with 250 mm non-rotating rubber discs. The gear has a mean vertical opening of 3 m. The door spread varies from around 25m at 20m depth to 40m at 80m depth. A 20mm (inside mesh) codend is fitted.

Irish surveys

The Irish West Coast Groundfish Survey started in 1990 and for the first two years consisted of circa 25 stations concentrated around the Irish coast in ICES Areas VIa South and VIIb. Adverse weather in 1992 limited station coverage to only 4 stations which effectively broke the time series.

The survey was re-established in 1993 and has consisted of circa 70 stations, for Parts A and B combined, since then. Spatial coverage was extended west out to the 200m contour, but remains as VIa South and VIIb.

Due to the restrictions of the current and previous research vessel commercial trawlers have been contracted to carry out the survey work. Wherever possible continuity of vessel and gear has been maintained and standard IBTS methodology applied. However, due to reduced staffing on commercial vessels it has rarely been possible to completely sort the catch.

Until 2000 all cod were sorted from the catch and then a sub-sample of two baskets was taken and completely sorted. From 2001 onwards, in response to an overall review of survey sampling undertaken during the IPROSTS Project (Anon 2001), the catch is completely sorted for all target species and a qualitative assessment made of the residual catch as a minimum.

The Irish surveys use an ICES rectangle based sampling strategy. The sampling design attempts to allocate at least two stations per rectangle (where the sea area is appropriate). Stations are selected randomly within each rectangle from known clear tow positions.

On the WCGS and ISCGS circa 70 fishing stations are planned on each survey every year. The number of hauls is adjusted according to the ship time available at sea.

The Irish West Coast Groundfish Survey is carried out in two parts: Part A covers ICES Division VIa (south) and VIIb (north); Part B covers ICES Division VIIb and VIIj. The survey is conducted from 15 to 300 m depths during the fourth quarter (October - November).

The Irish West Coast Groundfish Survey is carried out on chartered commercial fishing vessels. Whilst the same vessel (MFV Marliona, 224 gross tonnage and 30 m LOA) has normally been used each year for Part A and Sionann for Part B, in 2001 Part B was conducted from the MFV Regina Ponti (34.5 m LOA).

Both vessels use a Rockhopper net with 12 inch discs and 11 inch Thyboron doors. The nets are fitted with a 20 mm codend liner. Gear performance throughout the survey is monitored using Furuno Ch24 (Headline Monitor).

The Irish Sea Celtic Ground Survey commenced in 1997 and evolved from an earlier Irish Sea Juvenile Fish Survey. As a consequence early survey stations concentrated largely, though not exclusively, around a number of shallow spawning areas along the Irish east coast in VIIa. These positions were expanded in combination with clear tows provided by the industry and CEFAS as well as some exploratory tows.

Spatial coverage therefore extended into the western Irish Sea from 2001 into the area around the Isle of Man, Liverpool and Cardigan Bays and the Welsh coast. The survey is carried out on the Irish research vessel the R.V. Celtic Voyager.

The sampling procedure on board conforms to the IBTS standard protocols and as such all cod are sampled and aged, the entire catch is sorted and then sub-sampled as and where appropriate.

The Irish survey in the Irish Sea and Celtic Sea covers ICES Division VIIa and VIIg. The survey is conducted from 10 to 150 m depths during the fourth quarter (November-December) by the RV Celtic Voyager. This vessel is 32 m in length with gross tonnage of 340 t. The fishing gear used is a GOV 28.9/37.1 Trawl with Morgere Kite (0.85 by 0.85m). Mean vertical opening is 6 m and door spread 48 m. Morgere Polyvalent doors (Type AA4.5) are used and gear performance is monitored throughout the survey using the SCANMAR (RX400) net monitoring system (Headline height, Door spread).

Trials on the new 65m research vessel, the R.V. Celtic Explorer, are due to commence in late 2002. Therefore, from 2003 onwards all Irish Surveys will be conducted on this new vessel, starting around mid October through to late November.

English surveys

During the late 1970's the western mackerel stock fishery was expanding and concern for over-exploitation increasing. The Celtic Sea and Western Approaches Groundfish Survey was started in 1981, with the aim of investigating the distribution, biology and pre-recruit abundance of this mackerel stock. These objectives were almost immediately extended to all species that could be adequately sampled with a bottom trawl.

While mackerel was the primary target the survey covered all or part of the western continental shelf from the northern North Sea to the north coast of Spain. Later, as the objectives changed, the area shrunk in stages to its present boundaries: 47° 30' N to 52° 30' N and 3° W to 12° W. This has been the standard area since 1987.

In the early years a March/April and December survey was carried out each year but since 1989 only the spring (Quarter 1) survey has been conducted.

The survey fishes fixed station positions allocated by area (division lines at 48° 45'N and 50° 15'N) and depth strata (40-89, 90-114, 115-139, 140-179, 180-299 and 300-600m and is carried out on the RV Cirolana, a stern trawler 74m in length with a gross tonnage of 1731 t.

The fishing gear used is a modified Portuguese High-Headline trawl (PHHT) with 350mm rubber bobbins, a bunt tickler chain and a 20mm codend liner. Since 2004, this survey has been conducted from RV CEFAS Endeavour.

A fourth quarter survey is currently being established, as part of the internationally coordinated survey for southern and western waters. The first survey was conducted in 2002 on RV Cirolana, and used a Baca trawl. Due to major gear damage, the survey switched to a PHHT for the remainder of that year. During 2003, a modified GOV was used with rockhopper groundgear, deployed from RV CEFAS Endeavour.

French surveys

The French demersal survey began in 1987. The survey area was first limited to the Bay of Biscay (ICES Divisions VIIh, VIIIa,b,c and d) and in 1990, the survey area was extended towards the north to cover the grounds of Celtic sea deeper than 100 meters (ICES Divisions VIIe,f,g,h and j).

For the 1987 to 1996 period, the survey was conducted in the Bay of Biscay on an annual basis with the exception of the years 1993 and 1996. Most years it was conducted in the third or fourth quarter (October-November) except in 1991 when it took place in May. In 1988 two surveys were conducted, one in May the other in October.

The Celtic Sea was surveyed from 1990 to 1994 but the sampling was restricted to a small geographical area. The duration is between 40 to 45 days depending on year and availability of ship. Since 1997, the survey covered all the Celtic Sea and Bay of Biscay during the 4th Quarter.

Prior to 1997, the sampling designs were as follows: In the Bay of Biscay (ICES Divisions VIIh, VIIIa,b,c and d) a stratified sampling scheme was originally used.

The area was divided according to latitude into 3 blocks and the hauls were distributed in seven depth zones (15-30, 31-80, 81-120, 121-160, 161-200, 201-400, 401-600 m). 100 hauls were made at fixed locations and 35 at changeable stations from year to year. Since 1989, all the hauls (mean number by survey 135) are made at the same locations.

In the Celtic Sea (ICES Divisions VIIe,f,g,h and j) the sampling design was systematic, stations were located at the intersection points of a grid of lines 25 nautical miles apart both in latitude and in longitude. The mean number of sets was 56.

From 1997 onwards the whole area has been separated in 5 geographical strata or sectors: southern Bay of Biscay (GS) and northern Bay of Biscay (GN), southern Celtic Sea (CS), central Celtic sea (CC) and northern Celtic sea (CN). In each sector a depth-stratified sampling strategy has been adopted with 7 depth ranges: 0-30, 31-80, 81-120, 121-160, 161-200, 201-400 and 401-600 m.

The sampling design is a stratified random allocation. The number of hauls per stratum is optimised by a Neyman allocation taking into account the most important commercial species in the area (hake, monkfish and megrim). Minimum of two stations per stratum is sampled and 140 fishing stations are planned every year. This number of hauls is adjusted according to the ship time available at sea.

Catch weight and catch numbers were recorded for all species, only selected finfish and shellfish species were measured until 1990. Since 1991, all finfish and a selection of shellfish (mainly nephrops and squids) are measured.

Since 1997, the French survey has been carried out on the R/V Thalassa, a stern trawler of 73.7 m length by 14.9 m wide, gross tonnage of 3022 t. The fishing gear used is a GOV 36/47 without exocet Kite which is replaced by 6 additional floats. In average, the gear has a horizontal opening of 20 m and a vertical opening of 4 m. The doors are plane-oval with 1350 Kg.

Spanish surveys

The Spanish survey in the Porcupine Bank began in 2001 and covers ICES Divisions VIIb-k corresponding to the Porcupine Bank and adjacent area in western Irish waters from longitude 12° W to 15° W and from latitude 51° N to 54° N. The survey takes place in the third quarter (September) and covers depths between 170 and 800 m. The survey area has been separated in two geographical strata and three depth strata (170-300 m, 301-450, 451-800 m), resulting in 5 strata, given that there are no grounds shallower than 300 m in Southern geographical strata. The sampling design is random stratified with proportional allocation and a minimum of two stations per stratum with a total of 80 fishing stations.

The survey is carried on the R/V Vizconde de Eza. This vessel is a stern trawler of 53 m length and 13.5 m wide with gross tonnage of 1400 t. Fishing gear used is a Porcupine baca 40/52 with 39.46 m footrope and a 51.96 headline. Doors are oval with 800 kg and 4.5 m² surface. Diameter of warp used is 20 mm, of sweeps is 55 mm and the groundrope 98 mm with a double synthetic coat. Mean vertical opening is around 3.5 m and door spread 120 m. Codend mesh size is 20 mm.

Regarding the surveys on the Northern Spanish Shelf, the IEO has performed bottom trawl surveys in the Atlantic continental shelf waters of the Iberian Peninsula since 1974 (SESITS 1999). From 1980 the fishing resources of Divisions VIIIc and IXa of ICES were monitored through surveys, with the objective of following variations in the abundance of demersal and benthic species of commercial interest by means of indices independent of fishing activity. At the same time estimations were obtained of the strength of recruitment of diverse species (principally hake) during the autumn.

The evaluations were made according to a stratified sampling protocol, maintaining other factors constant, such as time of year, ship, fishing gear, speed, trawl time, etc. Tows were of one hour duration in all surveys before 1984, and were reduced to 30 minutes thereafter. Since 1990, gear geometry is monitored using Scanmar equipment.

Two series of surveys have been conducted, one at spring (April-May), starting in 1984, and the other in the autumn (September-October) starting in 1980. The spring series ended in 1988 and the autumn one has continued up to the present.

Two Spanish groundfish surveys are distinguished: The northern Spanish groundfish survey covers ICES Division VIIIc and the northern part of IXa corresponding to the Cantabrian Sea and off Galicia waters. This survey is conducted during the third and the fourth quarter (September-October) and covers a depth range of 35 to 700 m. Stratification was redefined in 1997, and is based on three depth strata (70-120, 121-200, 201-500 m) and 5 geographic sectors. Additional hauls both in deeper water (500-700 m) and shallower waters (30-80 m) are conducted yearly depending on the ship time available at sea. The coverage is approximately 5.4 hauls for every 1000 Km² (120 hauls per survey).

The Southern Spanish groundfish survey is conducted on the southern part of ICES Division IXa, the Gulf of Cádiz, twice a year: i) during late winter from 1995 and ii) in autumn from 1997. The area covered extends from 15 m to 700 m depth. In the southern surveys five depth strata have been used (15-30, 31-100, 101-200, 201-500 and 501-700 m).

All Spanish surveys in Spanish waters were carried out with R/V Cornide de Saavedra except in 1989 when another research vessel (N/V F. de P. Navarro) was used to conduct the survey. Several modifications were applied over time to this stern trawler: the engine power increased in 1983 (from 1700 Kw to 2651 Kw), in 1984 it was modified from its original 56 m (LL) and 990 GRT to 67 m and 1133 GRT at present, and a new bridge was used in 1990 (GPS, colour Echosounder, Plotter, Doppler log, etc.).

The gear used is a Baka trawl 44/60 with a 43.6 m footrope and a 60.1 headline. The traditional trawl doors used are rectangular, weighting 650 Kg and 3.6 m² of surface (2.67*1.34 m). The diameter of warp used is 22 mm (1.9 Kg/m). The mean vertical opening is 1.8 m and the horizontal opening is 21 m. Up to 1985, a codend cover of 20 mm mesh was used, and since then, a 20 mm mesh codend liner has been adopted.

Portuguese surveys

The Portuguese groundfish surveys have been conducted along the Portuguese continental waters since 1979 with the RV “Noruega”. The area extends from latitude 41°20' N to 36°30' N (ICES Div. IXa) and from 20 to 500 m depth. Surveys took place three times a year, in winter, summer and autumn. The winter survey took place in 1992 and 1993, then from 2005 to 2008 when it was interrupted. The summer survey series has started in 1979 and stopped in 2002. The autumn survey is conducted since 1979 with the exception of years 1984 and 2012 when the vessel was in reparation and since 2009 it is the only Portuguese groundfish survey to be performed.

Initially (1979) the main objectives of the surveys were to estimate the abundance and study the distribution of the most important commercial species in the Portuguese trawl fishery: hake, horse mackerel, blue whiting, seabream and Norway lobster. Recruitment abundance and distribution for hake and horse mackerel were monitored in the autumn surveys. Additionally, trawl selectivity experiments for hake and horse mackerel with 40 mm mesh size, were also conducted during 1981 surveys using the covered codend method.

At present, the main objective of the Autumn survey is to monitor the abundance and distribution of hake and horse mackerel recruitment.Additionally the survey also estimate (i) abundance indices and biomass of the most important commercial species (ii) biological parameters, e.g. maturity, ages, sex-ratio, weight, food habits. (iii) biodiversity indicators. The primary species are hake, horse mackerel, blue whiting, mackerel, Spanish mackerel, anglerfish, megrim and Norway lobster.

Sampling design was progressively modified in order to improve the precision of the abundance and biomass estimates. During 1979-1980 a stratified random sampling design was adopted with a total of 15 strata. Each stratum was divided into units of approximately 25 nm^2. The stratification was based on 3 depth intervals and 5 geographical areas, being the depth ranges 20–100 m, 101–200 m and 201–500 m. During 1979–1980, the number of hauls per stratum performed was based on the previous information of the relative abundance of the target species in each geographical area and on the ship time available. During 1981–1989, the number of strata was increased to 36, (12 sectors and 3 depth ranges) and two random units were sampled by stratum, whenever possible, to allow an estimate of the standard error of the stratified mean by stratum.

In 1989 a fixed stations sampling design was adopted, comprising 97 positions spread over 12 sectors, which one divided into 4 depth ranges: 20–100 m, 101–200 m, 201–500 m and the new 501–750 m, making a total of 48 strata. The positions of the 97 fixed stations were based on common stations made during 1981–1989 surveys and taking into account that at least two stations per stratum could be sampled. A maximum of 30 supplementary stations were planned to be carried out if ship time was available or to replace positions that due to particular factors were not possible to trawl.

A new sampling scheme was implemented in 2005. The design is a mixture of a fixed sampling scheme with trawl positions distributed over a fixed grid with 5’ per 5’ miles and random trawl positions. The new sampling scheme allows performing future calculations with the former strata and also allows to perform spatial analysis by using geostatistics methods. The 500-750 m depth stratum was removed due to small trawlable and low relevance for the survey purposes. The Autumn groundfish survey plan comprises 96 fishing stations, 66 at fixed (grid) positions and 30 at random. The tow duration is 30 min, with a trawl speed of 3.5 knots, during day light.

The Netherlands

The Netherlands BTS was initiated in 1985 to estimate the abundance of the dominant age groups of plaice and sole including pre-recruits. Initially the survey was only carried out in the south-eastern North Sea (ICES area IV) using RV Isis equipped with a pair of 8 m beam trawls rigged with nets of 120 mm and 80 mm stretched mesh in the body and 40 mm stretched mesh cod-ends. A total of 8 tickler chains are used, 4 mounted between the shoes and 4 from the ground-rope.

The survey was designed to take between one and three hauls per ICES rectangle (boxes of 0.5° latitude by 1° longitude). The stations are allocated over the fishable area of the rectangle on a "pseudo-random" basis to ensure that there is a reasonable spread within each rectangle. No attempt is made to return to the same tow positions each year. Towing speed is 4 knots for a tow duration of 30 minutes and fishing occurs during daylight only.

In 1995, the survey was expanded into the central and northern part of the North Sea using RV Tridens. This vessel uses the same gear but is equipped with a flip-up rope as it covers rougher grounds. Sampling strategy is also similar but only one haul per rectangle is taken, preferably close to the centre of the rectangle.

England

An English beam trawl survey has been carried out annually in July-August since 1989 using a commercial 4m beam trawl. The primary aim was to assess the relative abundance of pre-recruit plaice and sole in ICES Division VIId. Consequently, most of the sampling was carried out in areas known to be nursery grounds for these species. In 1995 the survey was extended to include the southern North Sea in order to sample the whole population of plaice and sole.

The standard survey had a total of 107 tows of 30 minutes duration, of which 79 were in the eastern English Channel. Since 1999 the number of tows worked has been reduced to 91 of which 75 are in the eastern English Channel. The sampling gear consists of commercially rigged (1989 styl 4m beam trawl (measured between inside edges of shoes) fitted with a chain mat, flip-up ropes, and a 40mm cod-end liner. The gear is towed at 4 knots (ground speed) for 30 minutes on a warp length appropriate to the depth of water. Fishing is only carried out in daylight on this survey, as catch rates can be very different in the dark under certain conditions.

Germany

The German beam trawl survey started in 1991 and covers much of the south eastern North Sea. It takes about 60 hauls each year using a pair of 7m beam trawls rigged with tickler chains. The cod-end mesh size is 75 mm.

Belgium