The third meeting of the FAO Working Group on the Assessment of Small Pelagic Fish off Northwest Africa was held in Agadir, Morocco, from 31 March to 10 April 2003. The overall objective of the working group was to assess the status of the small pelagic resources in Northwest Africa and recommend on fisheries management and exploitation options aimed at ensuring optimal and sustainable use of small pelagic fish resources for the benefit of coastal countries.
The species assessed by the group were: sardine (Sardina pilchardus), sardinellas (Sardinella aurita and S. maderensis), horse mackerels (Trachurus trecae, T. trachurus and Caranx rhonchus) and mackerel (Scomber japonicus), in the region between the southern border of Senegal and the northern border of Morocco.
The meeting was funded by project GCP/INT/730/NOR: "International cooperation with the Nansen Programme. Fisheries Management and Marine Environment" and the Netherlands Institute for Fisheries Research (RIVO) and organized by FAO and the Institut national des recherches halieutiques (INRH), Morocco.
Altogether 19 scientists from eight countries and the FAO participated. The chairman of the group was Dr Reidar Toresen, Institute of Marine Research (IMR), Norway.
The meeting was the third of its kind following one meeting held in Nouadhibou, Mauritania, 24-31 March 2001 and one in Banjul, Republic of the Gambia, 5-12 April 2002.
The terms of reference for the group were:
|
Marhoum Ahmed |
Morocco |
|
Salah Ben Cherifi |
Morocco |
|
Ana Maria Caramelo |
FAO (Rome) |
|
Hamid Chfiri |
Morocco |
|
Ad Corten |
The Netherlands |
|
Elhabouz Hammou |
Morocco |
|
Remment ter Hofstede |
The Netherlands |
|
Aziza Lakhnigue |
Morocco |
|
Ebaye Ould Mohamed Mahmoud |
Mauritania |
|
Asberr Mendy |
Gambia |
|
Birane Samb |
Senegal |
|
Maria Teresa Garcia Santamaria |
Spain |
|
Vladimir Severin |
Russian Federation |
|
Mahfoudh Ould Taleb Ould Sidi |
Mauritania |
|
Ibrahima Sow |
Senegal |
|
Merete Tandstad |
FAO (Rome) |
|
Nikolay Timoshenko |
Russian Federation |
|
Reidar Toresen (chairman) |
Norway |
|
Abdoulaye Wague |
Mauritania |
The working area for the working group is defined as the waters between the southern border of Senegal and the northern border of Morocco.
The structure of the report was somewhat changed as compared to the reports of the two previous Working Groups (FAO 2000 and FAO 2001). A separate section is devoted to each of the main groups of species (sardine, sardinellas, horse mackerels and chub mackerel). For each of these, standardized information is given on stock identity, the fisheries, abundance indices, sampling, biological data, assessment, management recommendations and future research.
Discussions on the basic data (landings, catch, effort, length distribution and age distribution) collected by each country formed one of the main topics of the 2003 meeting of the FAO Working Group.
The objective of the biological sampling of landings and catches is to obtain data on the exploited part of the stocks. These data are used in analyses of the resources' to elucidate the effects of the exploitation, and to determine sustainable fishing levels.
It is recommended by the working group to take, at least, one biological sample per species per fleet for each 1 000 tonnes caught.
Morocco
Biological sampling for length distribution of small pelagic fish is carried out five times a week on the daily landings of the Moroccan coastal purse-seiners operating in zones A, B and C in the ports of Safi, Essaouira, Agadir, Tan Tan, Laâyoune and Dakhla. The sampling takes into consideration the area where the catch was taken.
Sampling scheme:
Systematic sampling of the purse-seiners at a rate of 1 in 10.
Random sampling of 3 to 4 kg to measure length. For sardine and anchovy total length to the nearest half centimetre below is measured whereas the other species are measured in fork length to the nearest centimetre below.
Random subsampling of more than five individuals per length group for biological studies and removal of the otoliths for age reading.
Mauritania
The observer programme on board the pelagic vessels of the European Union, under the joint project Mauritanian Institute of Oceanographic Research and Fisheries-The Netherlands Institute for Fisheries Research (IMROP-RIVO), began in 1999. Under this programme all species caught are sampled for length distribution by the IMROP observers.
The sampling is carried out in such a way as to be able to estimate the species composition of the landings and the discards. For each species landed a basket weighing 20 kg is sampled from each trawl as well as one 20 kg basket for discards
On average, 8 to 12 trips are covered annually, sampling around 50 percent of the trawls per trip.
The biological sampling is carried out on board the Dutch vessels, but also at the institute's laboratory on samples of landings by the artisanal fleet in Nouakchott and from the Dutch industrial vessels in Nouadhibou. The samples studied at the laboratory are of sardinella only whereas the onboard sampling covers all the target species (Sardinella aurita, Sardinella maderensis, Sardina pilchardus, Scomber japonicus, Trachurus trachurus, T. trecae and Caranx rhonchus).
For each species a sample of 25 individuals is examined for fork length, total weight, eviscerated weight,sex, maturity, stomach content and fat content In addition to the above parameters, otoliths are sampled for age determination. The gonad weight is used for studying the effective reproduction potential.
The AtlantNIRO scheme of data collection was established to obtain the age and species composition of the Russian catches. Two observers are onboard one vessel all year long, changing from one vessel to another when possible.
The samples are random, taken from different parts of the trawl and weighted by the total catch of each species.
A minimum of two trawls are sampled per day and 200 fishes are measured per sample. Ten percent of the total number of fish measured are analysed for other biological parameters such as maturity, fat, food, individual weight, etc. and in each length interval a number of otoliths are taken for age readings.
The actual statistics from this fleet are not precise. The landings are for instance reported as a group of species and not by species and the length compositions of the catches split by species are based on the samples from fishing vessels in the region. Reports on catches by species in the region are prepared periodically by the observers.
Catch estimates by the artisanal fishery in the Mauritanian EEZ were carried out in 2002 using the following available sources of information:
Results from a Frame survey carried out by IMROP every six months on the whole coastline (FAO 2003).
Surveys carried out twice a week on the landings at Nouadhibou (since 1988) and at Nouakchott (since 1997).
Senegal
The Centre de Recherches Océanographiques de Dakar-Thiaroye (CRODT) has at least one observer in every main landing site, supported by beach assistants who record the fishing effort and length frequency observations.
The monitoring of the catch composition is carried out directly by interviewing the owners of the vessels when they return from the trips. The proportion of canoes surveyed daily varies, but an effort is always made to obtain a representative level of the day's activity. The sampling is carried out five days a week.
The many boats in the artisanal fleet in Senegal means that it is not possible to cover all the day's trips completely, thus extrapolations are made at the time of processing the data. The extrapolation is carried out on two levels:
(a) At the daily level, the extrapolation factor is obtained by taking the ratio between the total number of trips per day over the number of trips actually sampled. This coefficient is calculated for each gear and applied to all canoes sampled.
(b) At the fortnightly or monthly level a different coefficient is calculated to take into account the days when no survey is carried out (Sundays and holidays). It is weighted by the total number of trips during the period under consideration.
The information collected is: the type of gear used, times of departure and arrival, the meteorological conditions, the number of trawl hauls with or without catch, the number of baskets per haul, the species composition of the catch, the final destination of the product and the discards if applicable.
Twice a year CRODT carries out a national census of the Senegalese artisanal fleet which amounts to 8 574 canoes of which 720 are involved in the exploitation of small pelagics (October 2002).
Biological sampling is carried out during the landings at port or on the beaches. A basket containing 60 kg of fish per canoe is sampled to determine the percentages of each species in the catch and the length frequency distribution. A subsample (a quarter or half) is taken and the fork lengths are measured to the nearest half centimetre below. The weight of the sample is calculated using a weight - length conversion factor (W=aL^b) which is estimated every two years.
Biological samples are also processed in the laboratory where the following parameters are measured and recorded: total length, fork length, weight, sex, maturity, gonad weight, viscera weight, stomach contents, fat contents, etc. Otoliths and scales are also taken for age reading.
This sampling is mainly of the two sardinella species (S. aurita and S. maderensis) and the false scad (Caranx rhonchus). However, other species have undergone similar studies in the past.
The Gambia
There are 11 landing sites along the Atlantic Coast and in each one of them, one or two observers are on hand to monitor the activities of the fishers and fish buyers. One of the activities of these observers is to obtain data for the estimation of catch and effort which are conducted on a regular basis as outlined below.
A random sampling scheme called Catch Assessment Survey (CAS) is being used for artisanal catch and effort data collection. CAS is conducted six days a month; three days in the first half of the month and three days in the second half. For each gear type, landings of six canoes are sorted by species and weighed. Raising factors are then applied to account for the canoes and days that have not been sampled.
Due to numerous constraints, biological sampling programmes are not in place for the artisanal and industrial fisheries.
Spain
To estimate the length composition of the sardine, Sardina pilchardus, daily samples of 12-15 kg are taken from five or six vessels landing at Arrecife de Lanzarote. The unit of measurement used is the total length rounded down to nearest half centimetre (FAO 1980). The monthly length sampling data are weighted by the monthly catch to obtain the monthly length compositions. These are subsequently added to obtain the quarterly length composition.
The sample weight with respect to the catch is between three and seven percent, and the number of samples with respect to the total number of trips is between 25 and 30 percent.
To obtain biological information on Sardina pilchardus, one sample per month for each fishing zone is taken at the port of Arrecife de Lanzarote: The first one hundred specimens are chosen at random,and subsequently complemented to obtain at least ten specimens per half centimetre length group. In the laboratory, studies are made of the length (total length in mm) and the weight (to a precision of 0.1 grams) to calculate the quarterly length-weight ratio, sex (for the sex ratio), maturity (for the spawning period) and age determination to obtain the age-length key for each quarter. Applying these age-length keys to the length composition of the catch, the quarterly age composition of the catch is obtained. The Spanish information on sardine in Zone B covers the period 1976-1995 and in Zone C, 1983-1999.
Problems in estimating catches and length distributions for 2002
Several problems were encountered in estimating catches by economic zone. In Mauritania, an official estimate was only available for the total industrial catch and not for the individual fleets (European Union (EU) and non-EU). Moreover, the official catch figures referred to the combined catch of S. aurita and S. maderensis, and to the combined catch of Trachurus trachurus, T. trecae and Caranx rhonchus. In order to convert observer data into numbers of fish per species and length groups for the various fleets, estimates of monthly catches in each fleet have to be made available by Mauritania.
Another problem is that, of the industrial catches in Mauritania, only those made by Dutch and Russian vessels are sampled by scientific observers. However a large part of the industrial catch is from a variety of other fleets, of which no scientific sampling is conducted.
The artisanal fishery in Mauritania has been expanding in recent years but no official statistics were available for this fishery. Working Group members provided a provisional estimate for catches landed in Nouakchott in 2002. Some of the artisanal catches taken in the Mauritanian zone are landed in Senegal and are consequently included in the Senegalese statistics. During the meeting, no precise estimate of the size of these catches was available, but they are assumed to be in the order of some 10 000 tonnes.
For Senegal, very low numbers of fish measured were presented to the WG for 2002. The Senegalese participants indicated that these figures were not complete and that they would be adjusted at a later stage when all sampling data for 2002 had been analysed. The actual weight of the samples was not known and it had to be estimated using a length/weight relationship based on data from previous years. This may have introduced an error in converting the catch into numbers of fish by length group.
In the Gambia, increasing catches of sardinellas were reported in 2002. However, no sampling was conducted in this country, so the species and length composition of the catches remain unknown.
The Russian data are provided in the format that was originally specified by the WG:
- monthly proportion of each species in the catch
- quarterly length distributions for the Fishery Committee for the Eastern Central Atlantic (CECAF) areas
- quarterly age-length keys for CECAF areas
- preliminary percentage of annual catch taken in each month by CECAF area and species
The new requirements of the WG for 2004 are:
- quarterly catches by species and by national economic zone
- quarterly length distributions by species and national economic zone
- quarterly age compositions by species and national economic zone
It is recommended that these calculations are made prior to the meeting, and that the results are presented at the start of the meeting, including descriptions of the calculations.
Fork length versus total length measurements
Until now, countries in the region have measured fork length (to the cm or 0.5 cm below) whereas length measurements from acoustic surveys by R/V DR. FRIDTJOF NANSEN were reported as total length. In practice, this causes problems because length distributions measured in fork length are not easily converted into total length.
In order to avoid this problem, it is recommended that in future, length measurements are taken in a uniform way throughout the region. In view of the widespread use of total length in other countries in the world, it is recommended that countries in the West African region also change their length measurements to total length. During the remainder of 2003, countries are asked to record both fork length and total length in order to establish conversion factors. As of 1 January 2004, all measurements should be reported as total length.
Workshop on sardinella age readings
An exchange of otoliths of sardinella was initiated in May 2002. The results of the exchange indicated age determination problems, discrepancies in the assigning of ages and a lack of established criteria. The results also indicated that the preparation technique influences age determination.
A workshop on the age reading of sardinella was therefore organized by the FAO in cooperation with CRODT in Dakar, Senegal from 4-7 February 2003 with funding from RIVO, The Netherlands and Project GCP/INT/730/NOR: "International Support to the Nansen Programme. Fisheries Management and Marine Environment".
The main objective of this workshop was to discuss the age-reading methodologies for the ageing of sardinella using otoliths and to establish age-reading criteria.
Specifically, the workshop aimed to:
(a) Discuss the results of the 2002 sardinella otoliths exchange with the otolith readers and provide suggestions on how to improve their age-reading techniques.
(b) Describe and discuss the otolith preparation methods and the age estimation criteria used by the different research institutes.
(c) Present a synopsis of the biology of Sardinella aurita and Sardinella maderensis focusing on such aspects as stock distribution, spawning period and juvenile distribution, which are relevant to age-reading.
(d) Perform a reading of new otoliths and discuss the results immediately after reading in order to correct the ageing interpretation of each reader.
(e) Prepare guidelines on sardinella age estimation for experienced and inexperienced otolith readers on how to read and interpret the rings in order to achieve high precision.
(f) Provide recommendations on how the quality of the age-readings can be improved.
(g) Stimulate regional collaboration.
The method applied at the Workshop was to invite all the participants to read the same otoliths, reporting their observations of number of rings, nature of edge and age allocated without consultation with the other readers. The readings were followed by a discussion and explanation session where the readers explained their results and their reasoning behind the age allocated. Only Sardinella aurita otoliths were studied at the Workshop. The approach used took advantage of some of the experience gained when determining the age of sardine (Sardina pilchardus).
The workshop focused on familiarizing the participants with some of the terminology and concepts of age determination and in obtaining agreement in terms of age-reading criteria used by the different readers. The Workshop also attempted to look into the influence of the preparation technique on age determination and the use of microscope vs. stereomicroscope.
Due to the low number of otoliths read, no statistical analysis was made of the results. However, as a general impression it can be said that the readers started to read more similarly and that concepts became clearer to the readers during the Workshop. It was also noted that the best results were obtained when reading whole otoliths using a stereomicroscope. Based on the experience gained during the Workshop, a set of preliminary criteria was developed. These criteria are to be further developed and validated. The Workshop made several recommendations concerning sampling programmes. Furthermore, it recommended a new exchange of 50 otoliths from each country (Senegal, Mauritania, Morocco) to start in September 2003, followed by a Workshop to discuss and establish the age reading criteria which will take place in Nouadhibou, Mauritania in the first week of December 2003.
The report from the workshop will be published as an FAO Fisheries Report.
Sardine
A new exchange of sardine otoliths with otoliths from the Nansen surveys and from the commercial catches should be initiated in July 2003.
Horse mackerel and mackerel
It was noted that none of the research institutes in the region have experience with the reading of otoliths from horse mackerel and mackerel. It was recommended to not initiate any age reading activities for these species this year, but to concentrate efforts on sardine and sardinella. An exchange of horse mackerel and chub mackerel otoliths should be carried out in early 2004.
A planning group for acoustic surveys of pelagic fish off Northwest Africa was held in Dakar, Senegal from the 26 to 28 October 2002 to initiate the planning for the continuation of the acoustic survey activities in the region in the future. Scientists and technicians from Senegal, the Gambia, Mauritania, Morocco, the Netherlands, Norway and FAO participated.
It was agreed to make this planning group permanent and that it should meet once a year. The terms of reference for this group, where agreed upon, and include:
Review the results from last year regional acoustic surveys and to evaluate how the surveys were carried out.
Plan the following year's regional acoustic surveys including a detailed programme for the schedule of each vessel and coordination of vessel time.
Plan the calibration of acoustic equipment and intercalibration of the research vessels during the regional survey.
Standardize methods, such as age reading, scrutinizing and sampling from trawl hauls on acoustic surveys.
Discuss and seek solutions to problems that have occurred.
Discuss and initiate research related to acoustic abundance estimation.
Present the report of the planning group to the small pelagic assessment working group.
The planning group agreed that the chairmanship of the group should alternate and that the representatives from Morocco should chair the group for three years starting from 2003.
Over the last five years the landings of small pelagic fish in the region have fluctuated around 1.5 million tonnes. In 1999 the total landings were somewhat more than 1.3 million tonnes, whilst in 2001, the total landings reached a high level of nearly 1.7 million tonnes (see Figure 1.8.1a).
Sardine (S. pilchardus) is by far the most important of the small pelagic species, constituting about 50 percent of the total landings. The level of the sardine catches has been 530 000-780 000 tonnes over the last five years. In the period 1993-2000 the total annual landings of sardine were around 600 000 tonnes.
Round sardinella (S. aurita) is the second most important species in the region and over the last five years the total annual catch has fluctuated between 275 000 and 460 000 tonnes with an average of about 340 000 tonnes. The annual total landings of flat sardinella (S. maderensis) has been significantly lower than those of round sardinella fluctuating between about 100 000 tonnes to some 140 000 tonnes with an average of about 120 000 tonnes.
Of the horse mackerels, the cunene horse mackerel (T. trecae) is the most important with an average annual catch over the last five years of 150 000 tonnes. The corresponding figure for Atlantic horse mackerel (T. trachurus) is approximately 70 000 tonnes, whilst the mean catch over the same period for false scad (Caranx rhonchus) was 25 000 tonnes.
The chub mackerel (S. japonicus) catches fluctuated between 100 000 and 200 000 tonnes, and the average catch during the last five year period is estimated at some 160 000 tonnes.
Morocco
The exploitation of small pelagics off the Atlantic coast witnessed two important events in 2002: the decrease in sardine catch by the coastal purse seiners in Zone B and the absence of fishing for other pelagic species by the pelagic trawlers in Zone C, north of Cape Blanc.
In terms of volume of catch of small pelagics, sardine represents on average more than 70 percent of the total tonnage. In 1990, the catch reached more than 1 100 000 tonnes. The observed decrease between 1990 and 1992 can be explained by the progressive withdrawal of the ex-soviet fleet from Zone C north of Cape Blanc and by access agreements to the zone (fishing agreements between Morocco and the Russian Federation and Morocco and the EU) (Figure 1.8.1b). From Figure 1.8.1b it can be seen that the sardine catches off the Moroccan coast was relatively stable for the period 1992 to 2000, fluctuating around 600 000 tonnes. As for the increase in catch in 2001 (total catch of around 800 000 tonnes), this is probably due to the development of the coastal fishing fleet in Zone B and to the availability of sardine in the area around Laâyoune (27º30'N) which is the main sardine port in Zone B.
As far as the other small pelagics are concerned, these species make up the by-catch of the sardine fleet operating in Zones A and B. However, depending on the season, a particular species may be targeted when it is available in the zone fished by the fleet.
As for Zone C, the coastal pelagic fleet has adopted a strategy of specialisation according to the availability of one or more species. In fact, because of the migration of sardinella from the south up to the north of Cape Blanc, the pelagic fleet devoted part of its effort to this species between 1995 and 2001.
Looking at the trends in catches of species other than sardine, illustrated in Figure 1.8.1b, reveals that these species reached a maximum between 1996 and 1999 before decreasing from 2000. The drop is due mainly to the decrease in activity by the pelagic trawler fleet in Zone C.
Mauritania
Figure 1.8.1c shows the trend in industrial catches of the main coastal pelagic species in the Mauritanian EEZ from 1990 to 2002. The trends in catches of the different species show notable fluctuations.
For the period between 1990 to 1995 the average catch was 200 000 tonnes, whereas for the period from 1996 to 2002 the catch increased considerably, reaching on average 420 000 tonnes. Between 1995 and 1996, sardinella catches increased threefold going from 82 000 tonnes to 247 000 tonnes. Over the last four years, the catches dropped to a level of about 180 000 tonnes.
The horse mackerel catch in 2002 was similar to that in 2001, which is the highest level for the series 1990-2002 corresponding to twice the catch compared with the average for the period 1990-1999.
The statistical series of sardine catches began in 1996 with the entrance of the European Union small pelagics fleet into the fishery. This fleet mainly targets sardinella, but seasonally sardine as well. The catch of 37 573 tonnes in the year 2002 is the highest for the series.
The chub mackerel catches show annual fluctuations with small catches of an average of 17 000 tonnes between 1990 and 1994. From 1994 to 2002, the average catch tripled compared to the previous level, reaching its highest levels in 1996 and 2002 and its lowest in 1999.
Senegal
Figure 1.8.1d shows the annual trend in coastal pelagic fish landings for both the artisanal and the industrial fishery. During the period under consideration, the round sardinella represents 54 percent of the catch on average, the flat sardinella 38 percent, the cunene horse mackerel and the false scad 3 percent each and chub mackerel 2 percent. The two sardinella species are therefore the most important over the whole series.
The trends in sardinella catch show large annual fluctuations with an increase in 1992 notably due to the round sardinella, followed by a decrease until 1995. There is a significant increase in catch for the two species in 1996 followed by a further increase in catch of round sardinella alone in 1997. After a relative decrease in catch, mainly concerning round sardinella in 1997 and 1998, relative stability can be observed in the catches at the level between 200 000 and 225 000 tonnes over the last three years.
It should be noted that only in 1999 is the catch of flat sardinella greater that that of the round sardinella.
The Gambia
Since the gradual demise of the Ghanaian fishing company, Seagull Coldstores beginning in 1987 and finally closing in 1990, the fisheries for the species of interest, sardinellas (Sardinella aurita and S. maderensis), horse mackerels (Trachurus trecae and T. trachurus) and mackerel (Scomber japonicus) have been virtually nonexistent. The landings of these species by the artisanal and industrial fisheries are mainly considered as by-catch. Artisanal fishers operating in the Gambia target mainly Ethmalosa fimbriata, the bonga shad. No licence is currently issued for small pelagic fisheries.
From the plot of the total landings of the species as shown in Figure 1.8.1e it can be observed that sardinella landings have been very low since 1992, but following an upward trend and suddenly tripling in 2002. Total landings of horse mackerels and mackerel have been erratic over the years. The tendency is that of a steady increase over 2001/2002 as more small pelagics have appeared in the catches of both the artisanal and industrial fisheries.
