SESSION IV - FISHERY TECHNOLOGY AS APPLIED TO ARTISANAL INLAND FISHERIES
TECHNOLOGIE DES PECHES TELLE QU'APPLIQUEE AUX PECHES CONTINENTALES ARTISANALES

COLLECTE, TRAITEMENT ET COMMERCIALISATION DU POISSON EN PLAINES INONDABLES
COLLECTION PROCESSING AND MARKETING OF FISH IN FLOODPLAINS

par/by

A. Konare
Directeur, Opération Pêche
Mopti, Mali

1. GENERALITIES

L'économie malienne largement sur le secteur rural, principalement l'agriculture, l'élevage et la pêche, qui occupe près de 90 pour cent de la population malienne. Avec une part égale à 3 pour cent de la production intérieure brute, la pêche occupe une place importante dans l'économie du Mali. Les exportations des produits de la pêche représentent 8 pour cent des exportations totales. La pêche a permis en 1971–72 la distribution au niveau de la production d'environ 4 milliards de francs maliens de revenus monétaires en non monétaires, ce qui représente sur l'ensemble du territoire national un revenu per caput d'environ Mal.F. 904. On estime à Mal.F. 7 milliards environ la valeur brute des produits de la pêche malienne au stade final de la consommation ou de cession en territoire malien. La production de la pêche et les exportations de poisson séché-fumé ont diminué pendant ces dernières années; on a observé cependant un accroissement des exportations en 1975. Le tableau ci-dessous indique les prises totales de poisson et les exportations contrôlées de 1970–75 (en tonnes).

Tableau 1

Cette réduction progressive des quantités pêchées, celle des quantités exportées et à un degré moindre celle des quantités écoulées sur le marché domestique sont dues à 2 raisons:

• les conséquences des effets de la sécheresse
• l'usage de moyens de pêche destructeurs

en particulier l'usage de filets à mailles trop petites dont le résultat est la prise de jeunes sujets dans des proportions démesurées. La diminution des exportations de poisson séché-fumé s'explique par le maintien du niveau de la demande domestique face à des quantités pêchées réduites et aussi par la concurrence accrue de la pêche maritime et un certain développement des pêches continentales dans les principaux pays traditionnellement importateurs des produits maliens.

Il n'en demeure pas moins que la dimunition apparente du stock de poisson a déjà fait l'objet d'une préoccupation sérieuse des pays sahéliens.

2. INTRODUCTION

La pêche se classe parmi les principales activités de l'économie malienne non seulement par son importante commercialisation intérieure mais aussi par le chiffre du tonnage exporté vers les pays voisins. La production annuelle dépasse 100 000 t de poisson frais soit 1/10 de la production africaine, classant le Mali permi les premiers producteurs de poisson de l'Ouest Africain.

Pourquoi donc cette importance de la pêche au Mali? On la doit essentiellement à l'existence d'un vaste delta intérieur avec ses grandes plaines d'inondation d'une productivité considérable comprenant le Niger, son principal affluent le Bani, de nombreux affluents secondaires et la région des lacs que les crues saisonnières transforment chaque année en un système hydrographique complexe.

Ces plaines inondables, qui s'étendent sur plus de 2 000 000 ha pendant la saison des pluies pour former l'une des plus vastes superficies d'alimentation et de reproduction piscicoles de l'Afrique, expliquent l'extraordinaire richesse en poisson du moyen Niger. Dans cette région où la pêche est très importante, on constate que la croissance des espèces est totalement arrêtée durant les basses eaux et qu'elle est, au contraire, rapide pendant les hautes eaux grâce à l'abondance et à la richesse relative des ressources alimentaires que procure l'inondation. Des possibilités d'exploitation sont d'ordre humain materiel et technique.

2.1 Le plan humain

Le Mali dispose de pêcheurs traditionnels spécialisés (Bozos, Somonos) s'adonnant à une méthode de pêche. On a effectué de façon systématique un recensement de ces collectivités pêcheurs et les caractéristiques démographiques sont les suivantes:

• nombre de villages - campement 397
• nombre de familles 8 523
• population totale 80 338
• population active dont 52% masculine 54 112
• nombre moyen d'actifs/famille 6,3
• nombre moyen de personnes/famille 9,4
• population moyenne/base de vulgarisation 2 009
• ethnies dominantes: Bozo et Somono
• population nomade/population totale 21,7%

La famille regroupant plusieurs ménages constitue l'unité d'exploitation.

2.2 Le plan matériel et technique

La pirogue constitue la base de l'activité halieutique traditionnelle. Les pirogues sont motorisées pour se rendre sur les lieux de production et pour assurer le transport du poisson et des marchandises; elles sont non motorisées pendant l'exercice de pêche. Dans ce domaine également un recensement systématique a été fait et les résultats sont les suivants:

8 718 pirogues dont
7 311 de 1 à 2 t pour la pêche
938 de 3 à 5 t pour le transport du poisson
469 de 6 à 15 t pour le transport des marchandises
749 moteurs hors bord
86 sur pirogues de 1 à 2 t
249 sur pirogues de 3 à 5 t
414 sur pirogues de 6 à 15 t
28 088 filets de types

équipement moyen de la famille (unité d'exploitation)
1 pirogue de 1 à 3 t
3 filets et 5 engins de pêche divers

3. COLLECTE ET TRAITEMENT DU POISSON

3.1 Calendrier des pêches

La pêche a un caractère saisonnier au Mali et se caractérise par une période de basses eaux où la pêche est très active et une période de hautes eaux ou l'intensité de la pêche est faible. Le régime hydrologique du fleuve détermine étroitement le calendrier de la pêche.

De juillet à septembre

C'est le début de la crue. On assiste dans le delta central du Niger à un envahissement progressif des plaines d'inondation. Les poissons par une migration latérale quittent le lit mineur du fleuve et se répandent dans les plaines où ils trouvent soit une nourriture abondante (graines, insectes), soit des emplacements favorables à leur reproduction. Cette période est peu propice à la pêche.

Le mois d'octobre

Ce mois marque l'époque des hautes eaux. Le débit du fleuve connaît son maximum.

De novembre à février

C'est l'amorce de la décrue. Les plaines d'inondation commencent à s'assécher. Les poissons regagnent le lit du fleuve par une migration inverse; cette période connaît une pêche active par l'installation des parrages aux abords des marigots ou dans les bras du fleuve.

De mai à juin

C'est la période des basses eaux au cours de laquelle les poissons sont immobilisés dans les bas fonds du lit mineur du fleuve. C'est l'époque favorable à la pêche.

3.2 Les pêcheurs

Les pêcheries sont exploitées par des collectivités familiales qui monopolisent pratiquement la pêche dans le delta central et la région des lacs. Les pêcheurs appartiennent essentiellement à la race Bozo ou Somono.

Le caractère saisonnier des pêches entraîne le nomadisme des pêcheurs qui établissent leurs campements de pêche le long des berges du Niger ou sur une de ses nombreuses émergences pendant les basses eaux de mars à juin, période où la densité du poisson est la plus grande, et rentrent dans leurs villages pendant les hautes eaux. Il se produit une migration saisonnière de dizaines de milliers d'individus qui abandonnent pour plusieurs mois leur village d'origine.

Les pêcheurs constituent une réalité sociale bien structurée qui disposent de connaissances insoupçonnées de leur milieu physique et des moyens traditionnels à leur portée pour maintenir ce milieu de production. Le pêcheur malien est très attaché à sa profession, l'exerce avec une sorte de passion et cette école de la pêche en a fait un être fier de ses acquisitions, doué d'un dynamisme et prêt à s'ouvrir aux techniques nouvelles.

3.3 Moyens de production

Les pêcheurs possèdent une grande variété d'engins de pêche allant de la senne à l'épervier, du filet maillant et de rivant à la ligne. Les grandes pêches saisonnières dans les marigots et les mares se font à l'aide des barrages. Le pêcheur professionnel construit généralement lui-même sa propre pirogue faite de planches clouées de caîlcédrat de 10 m à 12 m de long pour la pêche et de 30 m de long pour le transport. La pirogue constitue la base de l'activité halieutique traditionnelle. Pour longtemps encore la pêche piroguière représentera l'aspect le plus rentable et le plus efficace de la pêche fluviale au Mali. Elle est adaptée à l'économie actuelle et convient au milieu qui se rétrécit pendant la période favorable des captures.

3.4 Traitement du poisson

Dès sa capture le poisson est jeté dans la pirogue, ramené au campement et débarqué par les femmes de la famille du pêcheur. Si la pêche a été collective toutes les femmes se rassemblent pour effectuer cette opération. Les opérations de capture sont effectuées par les hommes, alors que la préparation par contre incombe aux femmes et se déroule en général dans le cadre familial. Environ ⅓ des captures est consommé à l'état frais, soit par les pêcheurs et leurs familles soit dans un rayon de 50 à 100 km à partir des lieux de capture. Les ⅔ des captures sont donc transformées en poisson séché et fumé.

3.4.1 Séchage

Le séchage est le procédé de conservation le plus utilisé par les pêcheurs. Il s'applique à la presque totalité des espèces. Les traitements préliminaires de poisson sont:

• l'écaillage
• l'ouverture
• l'étripage
• la fermentation.

Le poisson est ensuite posé à plat sur des nattes ou la litière de paille et exposé directement aux rayons solaires.

La durée du séchage est fonction de la température extérieure, de l'hygrométrie et de l'espèce de poisson. Pendant le séchage le poisson est soumis à l'attaque de divers insectes: larves de mouches, dermestes et leurs larves, néorobies. Ces insectes consomment une quantité de chair proportionnelle au degré de l'infestation et à la durée du stockage au niveau du campement. Dans tous les cas leur action se traduit par une dépréciation qualitative du produit et de grosses pertes en tonnage.

3.4.2 Fumage

Les pêcheurs maliens fument également le poisson, et les tonnages traités sont supérieurs à ceux du poisson séché. Le poisson destiné au fumage doit être parfaitement frais, donc préparé le plus tôt possible après sa capture. Les opérations préliminaires d'écaillage, d'ouverture, et d'étripage sont faites de la même façon que pour le poisson destiné au séchage.

Le fumage exige beaucoup plus de travail que le séchage et il est également plus onéreux. Le poisson fumé est plus sensible que le poisson séché aux phénomènes de putréfaction et de moisissure.

3.4.3 Congélation

Ce procédé de conservation est pratiqué actuellement dans le cadre de l'Opération Pêche dont l'un des objectifs essentiels est la distribution du poisson frais au niveau des centres urbains. Le poisson livré à la congélation dans un état de fraîcheur parfaite est congélé à -40° et stocké à -20°C.

4. COMMERCIALISATION

Bien qu'une part non négligeable de la production de poisson séché ou fumé soit consommée par les pêcheurs eux-mêmes ou échangée contre des produits vivriers, la plus grande partie de cette production est commercialisée dans le cadre de l'économie monétaire. Dispersée sur 20 000 km² la production de poisson du delta intérieur ne prend sa véritable importance économique qu'après le regroupement des achats des commerçants ramasseurs et la commercialisation des apports directs des pêcheurs. Ce regroupement s'effectue au niveau des grands centres commerciaux spécialisés situés sur le Niger ou le Bani et qui sont reliés d'une manière permanente aux axes routiers. Le principal centre est Mopti qui monopolise pratiquement le commerce d'exportation du poisson séché ou fumé; les autres centres secondaires écoulent la plus grosse partie des apports sur le marché intérieur malien.

4.1 Contraintes de la commercialisation

La commercialisation du poisson séché et fumé dans le delta central du Niger est soumise à un certain nombre de contraintes:

• l'étendue et la dispersion des zones de production
• le nomadisme des pêcheurs qui pour la plupart ne sont pas originaires des circonscriptions administratives dans lesquelles sont situés les lieux de pêche. Ils regagnent leurs villages d'origine la saison de pêche achevée.
• La distance très longue qui sépare les zones de production aux centres de commercialisation.
• Les difficultés de communication pendant les basses eaux qui freinent l'évacuation du poisson.
• La présence d'un grand nombre d'intermédiaires dans le circuit de commercialisation ce qui traduit par des faibles prix d'achat au producteur et des prix de ventre élevés au consommateur.

4.2 Vente sur les campements de pêche

Les pêcheurs en fonction de leurs besoins en argent ou en marchandises vendent la totalité ou une partie des stocks de poisson qu'ils détiennent sur les campements. L'enlèvement des stocks de poisson dépend donc de la fréquence de passage des commerçants et des prix pratiqués. Le désir légitime du pêcheur de vendre le produit de son travail au meilleur prix peut déterminer au stockage prolongé au niveau du campement. Ce stockage réalisé dans des conditions assex précaires est néfaste à la qualité des produits livrés. Les commerçants transportent ensuite le poisson vers les centres de regroupement de la production situés en amont sur le fleuve. Pour cela ils utilisent soit les services des piroguiers transporteurs professionnels, soit les services des pêcheurs eux-mêmes.

4.3 Vente sur les marchés

Il s'agit d'abord des marchés périodiques traditionnels qui se tiennent en principe une fois par semaine. Les pêcheurs s'y rendent dans l'espoir que la présence simultanée de plusieurs acheteurs leur permettra d'obtenir un meilleur prix que sur les campements. Cependant une part très importante de la production du poisson séché ou fumé transite par le centre de Mopti où le commerce se trouve concentré entre les mains de quelques commerçants traditionnels. Les fonctions de Mopti dans le marché du poisson séché ou fumé sont multiples: c'est à la fois le grand centre de regroupement de la production, un marché de gros et de détail, le point de départ d'un trafic interafricain de poisson séché fumé vers la Côte-d'Ivoire, le Ghana, la Haute-Volta.

4.4 Agents économiques

Les agents économiques qui interviennent dans le commerce du poisson sont:

• les pêcheurs qui n'ont pas voulu céder leurs produits aux intermédiaires venus sur les lieux de pêche.
• les courtiers ou coxeurs: ce sont les intermédiaires qui mettent le pêcheur en rapport avec le grossiste acheteur.
• les grossistes magasiniers: la totalité du poisson séché ou fumé commercialisé passe entre les mains des commerçants grossistes. Les principaux grossistes travaillent en liaison avec les intermédiaires à qui ils consentent des avances en argent et qui parcourent les lieux de pêche à la recherche du poisson à acheter.

4.5 Stockage du poisson

Le poisson débarqué sur le marché de Mopti, emballé ou généralement en vrac y est pris en main par les agents économiques qui le trient, le conditionnent et le stockent dans les magasins.

Le conditionnement se fait dans des nattes pour le poisson séché et dans des caisses s'il s'agit du poisson fumé.

4.6 Expéditions

Les expéditions de poisson à partir de Mopti se font d'une part à l'intérieur du Mali et d'autre part dans les pays limitrophes. Le tableau 2 donne les chiffres des expéditions à partir de Mopti pendant ces six dernières années.

Les expéditions sont faites en fonction de la demande. Les exportateurs sont en relations suivies avec leurs propres représentants ou avec des grossistes installés sur les principaux marchés de vente qui les informent régulièrement des tendances du marché extérieur.

Avant les expéditions, l'exportateur procède au conditionnement de ses produits, recherche un transporteur et effectue les démarches auprès des services administratifs chargés de contrôler les transactions.

Le maximum des expéditions se fait entre les mois d'avril et d'août correspondant aux pêches de la grande saison effectuées pendant les basses eaux. Le minimum est atteint durant les mois d'octobre-novembre. Une reprise des expéditions se produit en décembre-janvier correspondant aux pêches de décrue pratiquées d'octobre à décembre.

La commercialisation du poisson donne lieu aux différents stades à la perception des taxes diverses:

Les prix pratiqués aux divers stades de la commercialisation varient selon l'époque de l'année et les disponibilités du marché selon l'espèce concernée et son mode de préparation. L'importance des marges prélevées par les intermédiaires fait que les pêcheurs ne perçoivent finalement qu'une part modeste de la valeur finale de la production.

5. CONCLUSION

L'industrie de la pêche a toujours occupé une place très importante dans le commerce extérieur du Mali. Le poisson représente une part importante quoique limitée du point de vue géographique des ressources alimentaires maliennes. Il assure à l'échelle nationale ⅓ de l'apport des protéines animales. Dans le delta central du Niger, et plus particulièrement à Mopti, la consommation du poisson se situe bien au-dessus de la moyenne nationale qui est de 14,1 kg par an et par tête d'habitant.

Tableau 2

Expéditions de Mopti 1970–1975
(en tonnes)

Le marchê de gros du poisson à Mopti ne présente pas de caractère organisé satisfaisant. Le rassemblement de la production du poisson séché et fumé en vue de l'expédition y est à l'origine de transactions conduites en ordre dispersé entre plusieurs catégories d'agents économiques. Ce marché de gros se présente comme un marché de type traditionnel essentiellement remarquable par le nombre et le volume des affaires qui s'y traitent et la prolifération des intermédiaires favorisés par l'absence d'une organisation rationnelle.

Pour assainer le fonctionnement de ce marché il serait nécessaire de définir sa structure, aussi que le rôle dévolu à chacun des agents économiques. Une organisation se dessine actuellement sur le port de Mopti depuis l'intervention de l'Opération Pêche par un contrôle effectif de la qualité des produits expédiés, la standardisation des emballages et l'uniformisation des poids bruts et nets des colis, un marquage des nattes.

L'amélioration de la situation actuelle de la production et de la commercialisation du poisson séché et fumé se conçoit selon trois formules principales:

• dans le cadre d'un marché libre traditionnel
• dans le cadre d'une action coopérative
• dans le cadre du contrôle et de la régularisation du marché par un organisme étatique de commercialisation.

Le système actuel est celui d'une intervention exclusive du secteur privé dans les transactions. Ce système qui a eu le mérite d'exister et de fonctionner s'est instauré spontanément et dispose d'un personnel bien adapté à un type de négoce voisin du commerce traditionnel. Il évite à l'état toute intervention directe dans le circuit économique. Il laisse cependant le marché entre les mains d'un petit nombre de commerçants qui par des manoeuvres spéculatives amplifient à leur bénéfice les mouvements naturels des prix tels qu'ils résultent de l'offre et de la demande.

THE TRADITIONAL FISHERY OF THE LOWER SHIRE VALLEY, MALAWI L'EXPLOITATION TRADITIONNELLE DES PECHERIES DE LA BASSE SHIRE AU MALAWI

by/par

N.G. Willough by
Fisheries Research Officer
Makhanga Research Unit
P/Bag Chiromo, Malawi

and

R.S. Walker
FAO Fishery Statistician
P.O. Box 593
Lilongwe, Malawi

1. INTRODUCTION

The River Shire is a major tributary of the River Zambesi and flows for approximately 450 km from its source. Lake Malawi, to the River Zambesi, which it enters 150 km from the Indian Ocean.

The perennial marshes and seasonal flood plain of the lower valley of the River Shire cover a maximum area of approximately 650 km² (Fig. 1). These marshes are subject to considerable seasonal variation in area as a result of two factors - the local rains, which occur between December and April, and a barrage at Liwonde which was built to regulate the level of Lake Malawi and consequently also regulates the flow of the River Shire. The marshes are usually relatively dry between August and November and fill during December.

The area produces between 10 000 and 15 000 t of fish per annum, which constitutes approximately 15 percent of the national catch. More than 60 species of fish occur in the lower reaches of the River Shire and its tributaries, but only three of these are of great importance in the commercial fishery. Two species of catfish, Clarias gariepinus (Burchell), C. ngamensis Castenau, and one cichlid Sarotherodon mossambicus Peters make up 90 percent of the catch.

2. METHODS

The valley was divided into four strata, each of which was split into two sub-strata on the basis of topographical or biological features (Fig.1).

Fig. 1 The Shire Valley

The numbers of fishing villages, fishermen and their boats and gears were determined for each sub-stratum by two surveys, one carried out during the high water period (May 1975) and the other during the low water period (October 1975). The composition of the catch and the yields from different gears were assessed by monthly surveys between February and November 1975. The recorder for each sub-stratum questioned between 10 and 20 fishermen each month and recorded their gear and the fish it caught. Fishermen were selected at random, and many were questioned more than once during the monthly assessments.

3. RESULTS

3.1 Size and distribution of the traditional fishery

One hundred and forty-two fishing sites were recorded, of which 42 were in stratum 1, 35 in stratum 2, 19 in stratum 3 and 46 in stratum 4. Nearly 40 percent of these had five or fewer fishing craft operating from them, and 90 percent had 20 or fewer. Strata 1 and 3 contained very few sites with more than 20 canoes, while stratum 2 had 5 of these and stratum 4 had 7. Two of the sites in stratum 4 had more than 60 canoes operating from them during the high water period. A seasonal variation in the size of fishing sites was found, with the number of craft per site increasing during the low water periods, and the number of sites being reduced accordingly.

An interesting feature of the fishery was the temporary nature of many of the fishing sites (with temporary being defined as less than 6 months old). Table 1 shows that there were significant differences in the proportions of temporary sites found in each sub-stratum, indicating that the fishermen in the southern areas of marsh were more mobile than those in the north. The number of temporary sites varied very little between the high and low marsh surveys for any particular sub-stratum.

Table 1

Distribution of permanent and temporary fishing sites

Approximately 1 700 fishing craft were seen during the surveys, of which about 95 percent were dugout canoes and the other 5 percent plan canoes. The latter are constructed by the Shire Valley Agricultural Development Project because suitable trees from which to make dugout canoes are becoming increasingly rare. It is probable that the proportion of plank canoes will rise steadily in coming years, although the cost of a new plank canoe on credit terms is Mlwi.K 55 (1 Malawi Kwacha = U.S.$ 1.00 approx.) whereas a dugout canoe might cost between Mlwi.K 10 and Mlwi.K 50 depending on its size and condition.

3.2 Fishing gear usage

Table 2 shows the proportion of fishermen questioned who used a particular type of gear, and the percentage of the total catch which that gear took. Gillnets were manufactured in Blantyre from multi-filament nylon, and locally constructed wooden floats and burnt clay weights were added by the fishermen, who usually mounted their nets by the half. These nets were secured by stakes at each end, and were commonly set in the lagoon systems and along the river banks. Castnets were also constructed of nylon and were used in the shallower areas of lagoons and rivers, while fish traps were made entirely out of local materials and were set along river channels. “Other gears” included seine nets, scoop nets, encircling fish fences, spears and handlines. Many of the fishermen questioned used more than one gear. Little seasonal variation was found in the usage of the four major gears, but seine nets and fish fences were only used extensively during the low water period of August-November.

Table 2

Gear usage in traditional fishery
(February-November 1975)

(2 823 fishermen questioned)

The average quantity of gear used by each fisherman is shown in Table 3. The most commonly used stretched mesh size for gillnets was 76 mm (3 in), although many fishermen also used 63 mm (2½ in). Larger mesh sizes (76 and 89 mm: 3 in and 3½ in) were employed by most fishermen in stratum 4, while many fishermen in other strata used smaller meshed nets as well.

Gillnets were widely used in all areas of the marsh. Longlines were frequently used in every area except Bangula Lagoon (S 3.1) where castnets were of great importance. Traps were set everywhere, although they were most important in Dinde Marsh (S 4.1 and S 4.2), and seine nets were most frequently used in the River Shire (S 3.2).

Knowing the average amount of gear used for each gear type it was possible to estimate the catch/man/day and the catch per unit effort (cpue). Seasonal variations in the cpue for different gears are shown in Table 4. (The mean catch/man/day is not the mean of the cpue for the different gears because many fishermen used more than one gear.). There is remarkable similarity between the cpue form the four major gears; a point which might form a basis for comparison of fishing gears in other systems. The reduction in catches during June and July is probably a result of less fish movement during the cold weather, coupled with a relatively large marsh area at this time. The efficiency of certain gears between August and November is a result of the concentration of the fish into a smaller area due to the low water level at this time. Although the seine nets and fish fences are particularly effective during this period, it should be noted that many people are involved in their operation, so that the catch must be shared among several fishermen.

Table 3

Gear quantities per fishermen
(February-November 1975)

(2 823 fishermen questioned)

Table 4

Seasonal variation in catch per unit effort from different gears (kg/gear/day)
(February-November 1975)

The catch/man/day showed significant variations in different areas of the marsh (Table 5), with fishermen in the east of Dinde Marsh (S 4.2) catching twice as much fish as those in the north eastern part of the Elephant Marsh (S 1.2). However, the southern fishermen did not earn twice as much as those in the north since the beach prices for fish in the south were lower. The average income of a fisherman in the Shire Valley is approximately Mlwi.K 350 per annum.

Table 5

Mean catch/man/day in different strata
(February-November 1975)

The species composition of the catches from different gears varied widely. Clarias spp. accounted for 99 percent of the catch from longlines; 80 percent of that from gillnets; 40 percent of that from traps and only 7 percent of that from castnets. The proportions of the catches for Sarotherodon were 0, 13, 36 and 78 percent respectively.

3.3 Fish marketing and distribution

Fresh fish was usually sold at the landing site for between 10 and 20 tambalas/kg (100 tambalas = Mlwi.K 1 = U.S.$ 1.00). Approximately 30 percent of this was consumed fresh in the immediate vicinity of the fishing village, with the remaining 70 percent of the catch being preserved, usually by smoking. A little sun-drying was also carried out during the hot season. The traditional method of fish smoking was over an open pit. This used large quantities of firewood, required constant tending, was open to the rain and produced a product of very variable quality. Much of the smoking is now done on modified versions of a kiln used in West Africa. These kilns are of brick, 1.3 × 1.0 . 1.0 m in size, open at the top, with a central tunnel for the fire and a baffle plate above it. As many as 7 wire mesh trays containing fish can be placed on top of this kiln, with the top tray covered by a sheet of corrugated iron to allow all-weather usage. The product from these kilns sells at between 70 and 90 tambalas/kg, and will keep for approximately one month if properly stored and protected from insect infestation. Clarias are usually gutted and split before smoking, whereas Sarotherodon are merely gutted.

Most of the fresh fish is sold at the landing site to traders. These dress and smoke the fish, and then either transport and sell the fish themselves or sell it wholesale to more affluent traders with better transport facilities. Important markets for Shire Valley fish are at Nchalo, Ngabu and Makwasa for stratum 2 (see Fig. 1); Sorgin, Mulanje and Thyolo for stratum 2; Bangula and Mulanje for stratum 3; and Nsanje, Blantyre and Luchenza for stratum 4. Fish may be transported from the valley by either road or rail. At least 30 percent of the total catch leaves the valley, of which approximately 70 percent is transported by rail.

4. DISCUSSION

Approximately three-quarters of the animal protein consumed in Malawi is fish, but with the national demand for protein rising steadily, there are pressures on all the fisheries in the country to increase their yields still further. The fishery of the Shire Valley has been self-regulatory for many years, since the population of fishermen in the area is very mobile (as shown by the number of temporary fishing sites) and many fishermen also have small farms to which they can return when the fishing is poor, however, recent pressures on the fishery might have led to over-exploitation of the fish stocks if new laws which regulate the net dimensions and mesh sizes had not been introduced.

The only stocks of fish not being adequately exploited by present methods are the small fish in the marshes (which could be caught in finely constructed traps) and fish in the middle of the main river channel (which could be caught in drifting gillnets). It is unlikely that either of these stocks will contribute significantly to the yield from the marsh in the near future, and 15 000 t per annum is probably as much as can reasonably be expected under the normal flooding regime. If extensive and prolonged flooding of the marsh occurred in one year, this yield would probably be exceeded in the two following years as a result of better survival and growth of fry during the flood year.

The main activities of the Fisheries Department in the Shire Valley are now concerned with increasing the efficiency of preservation and marketing techniques, and with research into the basic biology of the commercially important species. Spoilage of the fish, either by decomposition or insect infestation after preservation, has been significantly reduced by the introduced smoking kilns; and the communications in the valley as a whole have been greatly improved by the road building programme of the Shire Valley Agricultural Development Project. Studies on the basic biology of the commercially important species are now nearing completion, and an account of them will be given in a separate paper.

THE ATALLA FISHERY OF THE LOWER NIGER, NIGERIA LA PECHERIE DE L'ATALLA DANS LE COURS INFERIEUR DU NIGER, NIGERIA

by/par

J.B.E. Awachie and E.C. Walson
Hydrobiology/Fisheries Research Unit
Department of Zoology
University of Nigeria
Nsukka, Nigeria

1. INTRODUCTION

African rivers are rich in fish species and the need to exploit these has led to the development of numerous local types of gear and fishing methods that have made most of the local fisheries possible; however very little basic and detailed information is available on these gear—variations in structure and methods of operation, effectiveness, contribution to and influence on fish production and fisheries, etc. The little information there is has been given by Holden (1961, 1967), Reed (1967, 1970), Obakin (1969) and Welcomme (1971, 1972). Welcomme (1971) has reported that the majority of gear and methods used in catching fish in Africa show little regard to the maintenance of stock from which they draw. This is especially so in Nigeria, where one such gear is the lift net, locally known as ‘atalla’.

Active fishing using the atalla is widespread in Nigeria. No reports on the use of this gear are, however, available from the Ogun and Oshun Rivers to the southwest, and the Imo River to the east of the Niger. Places where the fishery is important include Yola Lau and Makurdi on the Benue River, Lokoja at the confluence of the Niger and Benue Rivers, Lake Kainji area on the Niger, Onitsha on the lower Niger, Otuacha on the Anambra River and Afikpo on the Cross River (Fig. 1). In the lower Niger basin, the atalla fishery forms an important component of the seasonal indigenous fisheries that last from about May to October annually.

Fig. 1 Major Nigerian rivers where the atalla gear is operated

In this paper, the atalla fishery of the lower Niger basin is examined and its importance in the overall fisheries of the area assessed. This study, on which this paper is largely based, was carried out between May 1976 and June 1977 at Onitsha and Otuocha in the Niger-Anambra drainage system.

2. THE FISHING GEAR

Reed (1967) has given a general description of the atalla used in various parts of Nigeria. Essentially, the gear consists of a rectangular framework made from raffia-palm poles to which a nylon net of about 10 mm stretched mesh is attached. To the base of each vertical pole is tied a notched piece of hard wood, which fits over the side of the canoe (Fig. 2), while a flat piece of wood attached about one metre from the base serves as a ‘pedal’, for depressing the gear into the water. Two cords, usually made of local fibres and tied to the outer corners of the framework, are used to lift the gear out of the water during operation. In the Anambra-Niger area, the second pair of cords, described by Reed and used for shaking fish into the canoe, is lacking. Fishermen in the area do not feel that these cords are necessary as the first pair can serve the same purpose. Moreover, they argue that this second pair of cords, usually attached near the centre of the net, tends to shorten the life span of the gear, as it usually initiates a tear of the net at the point of attachment. At this juncture, it is pointed out that up to 20 percent of atalla gear in use in the lower Niger area still makes use of thin slivers of raffia-palm poles tied together with raffiapalm fibres to form a matting, used instead of nylon net (Fig. 3). Four of the 17 gears used by local fishermen, which were closely studied, were made of the raffia-palm matting referred to above and it is interesting that the latter were all found in the Otuocha area.

The details of the size variation of the local atalla surveyed in the Anambra-Niger area are given in Table 1. The overall size of nylon and raffia fibre meshed gear ranged from 12.7–18.4 m² and 8.1–9.6 m² with a mean of 15.4 m² and 8.7 m² respectively.

From the foregoing it is clear that the gear with nylon mesh are generally larger than the raffia-fibre meshed ones - a fact which may be related to the lighter weight of nylon gears. Field observations, however, indicate that the cost of the larger craft (dug-out canoes) appears to be the main factor against the known desire of fishermen to increase the size of nylon-meshed gear in order to increase catches. Also the constant slipping of gear off the canoe is a major source of concern and loss of catches to the fishermen.

A number of modifications, which are cheap to effect, have been put forward to improve the situation.

2.1 Suggested modifications

The suggested modifications, some of which are already employed in the two experimental gear presently in use at Onitsha and Otuocha, include:

1. Increase in the overall size of the gear. The experimental gear are 25.4 m² each compared to a mean size of 12.0 m² currently used by local fishermen

2. Use of Stroke 2 nylon net instead of Stroke 3 used by the fishermen, in order to reduce weight. This is necessary to compensate for increase in weight due to increase in size in (i) above

3. The use of a modified pivoting device to eliminate the use of ‘pedals’ for depressing the gear into water (Fig. 4). Pedals tend to reduce the life of raffia-palm poles as the fisherman normally steps directly on them. The pivoting device ensures that the gear does not ‘roll’ during operation.

4. The use of wire cable instead of the natural fibres commonly employed by local fishermen for lifting the gear. The wire cable lasts much longer than the natural fibre and will in the long run pay for itself many times over.

5. Use of detachable horizontal poles to facilitate folding and storage of the gear under shelter, instead of open areas where the nylon net is adversely affected by sunlight. It has been established that nylon nets deteriorate under sunlight and this is readily observed in the humid tropics.

Fig. 2 Locally made atalla

Fig. 3 Atalla fitted with raffia-palm mat

Table 1

Table showing dimensions of different types of atalla used in Anambra-Niger basin

*EG = Experimental gear used on the Niger and Anambra Rivers

Fig. 4 Device for pivoting atalla on the side of the canoe

Fig. 5 Fishermen operating atalla on the River Niger at Onitsha

3. ASSOCIATED CRAFT

This is usually the dug-out canoe. A survey of canoes employed in the fishery show that they are mostly of medium size with their length ranging from 4.9–7.2 metres with a mean of 6.2 metres. Generally the small canoes are used in operating the usually smaller atalla made of raffia-palm matting, while the larger craft are employed in the operation of larger atalla. The correlation between the size of craft and gear is noteworthy. In practice, however, it is the size of the canoe that determines the size of the gear used and it was interesting to note that most fishermen are able/prefer to buy only canoes of moderate size. The canoes used in operating the larger experimental gears average about 8 m.

Table 2

Length of canoes employed in atalla fishery of the lower Niger and Anambra basins

Fig. 6 Fishermen operating atalla fitted with raffia-palm mat; dragging the gear shorewards

Fig. 7 Fishermen operating atalla fitted with raffia-palm mat; lifting the net manually

4. OPERATION OF THE GEAR

Reed (1967) described two common methods of operating the atalla: the fixed canoe and the drifting canoe methods. In the lower Niger basin, the drifting canoe method is predominant. Fishermen may either allow the canoe to drift with the current or they may paddle shorewards while depressing the gear into the water with their feet. The gear remains submerged for about one minute (Fig. 5) before it is lifted. The latter method is more effective than the drifting method. The fixed canoe method is not commonly practised in the lower Niger, however, at the peak of the flood season when most bank vegetation is submerged, fishermen, especially older folk, may be seen using this method.

A third method involves the use of the gear fitted with raffia-palm matting. It is commonly employed by women and young people operating along shallow banks and on the flood-plain, and is quite productive. With the base of the gear hooked on to a canoe, the women and children standing in water and holding the free ends under water, drag the gear shorewards. On reaching the edge of the bank, the gear is lifted manually and fish shaken into the canoe (Figs. 6 and 7). This operational procedure endangers the reconstitution of the fish stock on the floodplain.

5. SAMPLING PROGRAMME

Collections were made fortnightly with experimental atalla gear from May 1976 to June 1977. The actual collecting/sampling periods each day were from 4–6 a.m. and 4–6 p.m. and these times were chosen to fall within the normal period of operation by local fishermen. Of course the areas fished were not as extensive as those of the professional fishermen.

6. THE CATCH

The species composition of fish taken by atalla in the lower Niger-Anambra basin is given in Table 3. A total of 38 species belonging to 13 families was taken in experimental fishing carried out over a 12-month period. Thirty-three out of the 38 species occur in Anambra River around Otuocha while only 30 of the species were taken in the Niger (Onitsha area) (Table 4). The differences in the species catch composition of the two areas may be due to the likelihood that the extensive floodplain of the Anambra River may provide the main nursery grounds for the fish of the Niger and Anambra drainage complex.

In Table 5, the three dominant taxa by both number and weight are shown to be the Clupeidae, Schilbeidae and Characidae. The Mormyridae and the Citharinidae are next in importance and, what is more, are represented by their fry and fingerlings.

Figs. 8 and 9 show seasonal trends in the occurrence of the major groups, with Clupeidae dominating both in numbers and weight, followed by Schibeidae and Characidae. The almost complete absence of the cichlids in the fishery is probably an indication of the unsuitability of the gear for catching them (Awachie and Walson, 1976).

The above finding indicates that along river banks, the atalla fishery is essentially a small species fishery which in season takes fry and fingerlings of larger species e.g. those of Citharinidae, Distichodontidae and Characidae in the lower Niger. It is note-worthy that when used on floodplains, pools and lagoons, which serve as nursery ground for most of the economically important fish species, the fry/fingerling component predominate with consequent damage to the fishery. Happily, however, most atalla fishing is done on or along the river sides. This apparently is primarily because of frequent damage to gear by wood and tree stumps when used on most parts of the floodplain during the flood phase.

Table 3

List of fish species taken by atalla in the Anambra-Niger basin

Table 4

Distribution of fish species taken from Anambra and Niger rivers by atalla

Table 5

Summary of the composition of fish groups taken by Atalla fishery in the Lower Niger
(May 1976-June 1977)

Fig. 8 Seasonal trends in the occurrence by number of fish taken by atalla in the lower Niger (fishing 4 h daily at 2-week intervals)	Fig. 9 Seasonal trends in the occurrence rence by weight of fish taken by atalla in the lower Niger (fishing 4 h daily at 2-week intervals)

7. PRODUCTIVITY OF THE FISHERY

The fishery is seasonal, lasting for about six months, with peak yields about September-November (Fig. 10). This corresponds to the peak flood season when river banks are completely flooded and most vegetation submerged, making a lot of insect larvae available as food for fish. The poor productivity during flood recession could be accounted for by the rapid draw down which suddenly makes less feeding grounds available over a very short period.

It is estimated that an average atalla fishing unit, which comprises two men operating a gear for a period of 4 hours daily, catches between 6 and 8 kg. daily at peak season and 3–5 kg during the less productive period. These are sold on landing to local market women at N2.5 per 2.4 kg. The fish are measured in small standard baskets weighing 2.4 kg when full. Thus average earnings may vary from N108 – N210 per month during the season. If, from available data, there are three months of peak production, then each fishing unit may earn up to N1 000 during the six months, based on an input of approximately N60. This makes atalla fishing a profitable artisanal fishery. Considering that market women are known to retail the fish at about twice the landing price, the actual market value of the catches of each fishing unit would be approximately N2 000 per season.

It is estimated that over 500 fishing units are actively involved in the atalla fishery of the lower Niger basin, producing over 600 t of fish each season in the area under consideration. It is to be noted, however, that most of these fishermen are involved in other gainful activities such as crop farming and trading. During the out-of-season period, most of them, apart from farming and trading, change to other forms of fishing such as the lucrative Alestes fishery using beach seines or longlining for catfishes.

8. EFFECT OF THE ATALLA FISHERY ON THE OVERALL FISHERIES PRODUCTIVITY OF THE LOWER NIGER

As shown above, the three dominant groups in the catches belong to the families Clupeidae, Schilbeidae and Characidae. Of these major groups, only the Characidae contain species that grow into large fish such as Alestes Macrolepidotus, A. nurse, Hydrocynus forshali, H. vittatus and Hepsetus odoe, and these form less than 20% of the total catch (Table 6). It has also been shown that the fry/fingerlings of large species such as Mormyridae, Citharinidae, Distichodontidae and Bagridae are taken. Other species belonging to other families form an insignificant proportion of the catches. Thus in season, normal atalla fishing along the bank does not seem to cause much damage to the stock and recruitment cycle of the fish/fisheries of the lower Niger, since mainly small species which are not easily and effectively harvested by other gear are taken, however, the effect of the gear on the fingerlings and therefore the recruitment cycle of Citharinidae, Distichodontidae, Bagridae and Mormyridae may be considered of some importance but damage may be minimal to their stock reconstitution. On the other hand, it is established that the fishery provides cheap protein very essential in the diet of the people of the area and what is more serves as a seasonally fairly reliable source of income for many.

9. CONCLUSION

The atalla fishery is widespread in Nigeria and forms and important component of existing indigenous fisheries. In the lower Niger, from Lokoja to Yenagoa, over 500 fishing units are involved in this essentially seasonal floodphase fishery. The gear is adaptable and three operational procedures are adopted in the lower Niger to achieve maximum production. The method of operation adopted depends to some extent on bank morphology and available man-power.

The fishery yields 600 t of fresh fish in each season of six months. The catch is made up of about 38 species of fish belonging to 13 families. The dominant families are Clupeidae, Schilbeidae and Characidae. Available data and field observations show that most of the catch comprises the small species elements of the fish fauna and it would seem that the fishery may only be marginally harmful to the fish stock and therefore fisheries production of the lower Niger. Indeed by ensuring that the rather rich small species in tropical inland waters are cheaply cropped, the fishery provides, for a period of six months, cheap animal protein for the lower income group in a protein-poor area of the country.

Table 6

Showing percentage composition by number and weight of the family Characidae

Fig. 10 Trends in monthly catches by weight, using experimental gear

REFERENCES

Awachie, J.B.E. & Walson, E.C. 1977 Preliminary observations on the composition of fish species taken by lift net (Atalla) fishery in the Anambra-Niger drainage system. Proc.Abstr. 18th Annual Conference of the Sciences Association of Nigeria

Holden, M.J. 1961 Fishing methods in Sokoto Province, Northern Nigeria. Nigerian Field 26, 147 – 158

Holden, M.J. 1967 Further notes on fishing methods in Northern Nigeria. Nigerian Field 32, 27 – 32

Obakin, M.A., 1969 Fishing for the pink prawn, Penaeus duorarum (Burkenroad) in the Lagos, Lagoon. Federal Fisheries Annual Report, Lagos

Reed, W., 1967 Fish and Fisheries of Northern Nigeria. Ministry of Agriculture Publication, Publication, Northern Nigeria

Welcomme, R.L., 1971 A description of certain indigenous fishing methods from Southern Dahomey. Afri. J. Trop. Hydrobio. Fish. 1, 129 - 40

Welcomme, R.L. 1972 An evaluation of the cadja method of fishing as practised in the coastal lagoons of Dahomey (West Africa). J. Fish Biol. 4, 39 – 55