The opportunity to test the validity of the concept of producing a theoretical map extrapolating from a limited data set arose for a portion of Liberian shelf where theoretical assemblages maps were prepared using method one (pure extrapolation) and compared with the results of a Liberian-Soviet bottom trawling survey carried out in 1981 (Dixon, Overko and Bukatin, 1981).
The map of bottom types for Liberia North is a modification of the maps prepared by Rancurel (1968). The rather imprecise bottom categories used by Rancurel on the whole Liberian shelf were compared to the very accurate ones given by Martin (1973) for the Ivory Coast down to the border with Liberia because the sedimentological features of the Ivorian shelf extend on to the Liberian one - Rancurel's original map was then modifield accordingly. A theoretical assemblages map was drawn using the bottom map, the position of the 20°C isotherm as determined from the Guinean Trawling Survey and an extrapolation of the species distribution/bottom type relationships observed in the western area of the Ivory Coast close to the Liberian border by Caverivière and Champagnat (1980). This “extrapolated” map was then compared to the more documented one prepared from the information collected during the 1981 Trawling Survey (Map 5) which unfortunately did not cover adequately the very shallow waters where the estuarine fascies of the sciaenid community is to be found.
The comparison showed that there are only small differences between the predicted and observed geographical distribution of the sciaenid and sparid assemblages. A difference appears in the slightly deeper position of the boundary between these assemblages as a result of the use of the depth of the 20°C isotherm to define the lower limit of the distribution of the sciaenids. Longhurst (1965) used the 20°C isotherm as it corresponded more or less to the lower limit of the thermocline, but during the Liberian-Soviet survey such a limit was in fact found to be around 18°C. It can therefore be said that the extrapolation procedures gave reasonably good results.
The 1981 bottom trawling survey covered the grounds situated from 22 m down to a depth of about 900 m. The preliminary report on the results of the survey contains information on temperatures at various levels and on individual hauls, but it does not give the type of bottom found at each of them. In order to overcome this limitation use was made of bottom maps prepared from Rancurel ones, modified as described earlier- and navigation charts - to draw a new set of bottom maps (Maps 1.1, 2.1, 3.1).
The Liberian continental shelf is characterized by a shallow inshore area where numerous isolated rocks are found over soft grounds. This rocky area gets progressively wider from Monrovia to the Ivorian border. After the border, it gradually narrows to disappear completely east of Sassandra (Ivory Coast). Such an area constitutes the ideal habitat for the lutjanid community probably mixed with the sciaenid assemblage representatives where suitable bottom conditions are present between patches of rocks. Soft bottom deposits found from Monrovia to the Sierra Leone borderand the presence of some rivers are suitable conditions for the occurrence of the estuarine fascies of the sciaenid assemblage.
Results of the survey show that the sparid assemblage is found from a depth of 50 m to about 100 m all along the Liberian shelf. The most frequent species caught are Dentex angolensis, D. congoensis, Epinephelus aeneus, Priacanthus arenatus and Paracubiceps ledanoisi (Table 1 and Maps 1.2, 2.2, 3.2). By and large, the most abundant species is P. ledanoisi.
A marked reduction of the abundance of the sparid assemblage was recorded during the survey near Cape Palmas. In addition the most frequent accompanying species in this assemblage are not the same in Liberia and the Ivory Coast. Comparable data from catches obtained during the warm season off Grand Bassam (Ivory Coast) shows that Brotula barbata and Pentheroscion m'bizi are important species in the Ivory Coast, whilst in Liberia they were rare during the survey (Table 2). Such heterogeneities in the species communities could also suggest that Liberia, which has a very particular hydroclimate where warm low salinity waters predominate, has its own particular sparid assemblage.
East of Monrovia and down to Sassandra (Ivory Coast) the distribution of the estuarine assemblage could be discontinued due to the presence of an extensive coastal area with isolated rocks.
Assemblages and species distribution maps were prepared using the results of the 1969–70 trawling programme that covered the waters between Axim and Keta (Rijavec, 1980). Most of the fishing was done in the 10–30 fath (18–72 m) interval, probably only covering partially the depth distribution of the sciaenid and sparid assemblages. Fortysix percent of the sampling took place on a transect off Tema.
In order to identify possible seasonal differences in the main species distribution, maps were prepared with the results of surveys made in May during the warm period and in August during the cold season in 1970.
Bottom maps were prepared by combining information from a British navigation chart-covering Cape Three Points to Cape Formosa - as well as the Guinean Trawling Survey and Rijavec's reports. Such maps (Maps 4.1, 5.1) are less detailed than the maps prepared for the other countries due to the fact that both reports presented the various bottom types grouped under three main broad categories: soft, mixed and hard. In addition to that, the likely uncertainty on the exact position of most of the fishing stations carried out during the 1967–70 survey prevents the full use of the information about bottom types to more precisely delineate the bottom types strata.
Lack of geographic continuity observed in the distribution of some species could be attributed to limitations in the sampling coverage of the 1969–70 trawling programme.
Ghanaian waters are characterized by strong seasonal upwellings; there is a predominance of hard bottom areas and a reduced extension of soft grounds which are only found around Cape Three Points and Cape Saint Paul (Maps 4.1, 5.1) and in the coastal area down to the 15 fath (27 m) isobath. As a consequence, particularities of the Ghanaian waters as compared to neighbouring ones are the reduced relative importance of the warm water/soft bottom sciaenid assemblage and the increased importance of the cool water/hard bottom sparid assemblage.
The sciaenid assemblage is identified here by two most abundant species, Brachydeuterus auritus and Galeoides decadactylus. The most frequent species in the sparid community are Pagellus bellottii and Sparus caeruleostictus (Table 3).
The seasonal upwelling of cold and saline waters, over the Ghanaian shelf provokes changes in the geographical distribution of the species. During the upwelling season, the bathymetric extension of the sciaenid assemblage is reduced to a minimum, while species of the deep water sparid group are found near the coast and the abundance of some components of the sparid assemblage in trawl catches greatly changes. Epinephelus aeneus, a relatively common species during the warm season, disappears from catches, while Sepia spp., rarely caught during the warm season, is fished frequently in relatively high abundance. The westward reduction in the frequency of appearance of the most common sparids that is observed around Cape Three Points during all the year is probably linked, on the one hand, to the near disappearance of this assemblage east of the Cape where sandy mud bottoms predominate and possibly, on the other hand, to a change in its dominant species in the Ivory Coast. Information about the most abundant species found in the Grand Bassam transect gives support to this idea (Table 4). On the contrary, the similarity of the most frequent species around Cape Saint Paul in the eastern side of Ghana reflects the fact that such species extend well into Togolese waters.
Information on bottom types and species composition by individual trawling operations presented by Crosnier and Berrit (1966) was used to prepare maps of demersal species assemblages.
The basic information was recorded during two exploratory trawl surveys that took place in October 1963 and July 1964. These surveys covered the waters of the continental shelf from 9m down to 100 m depth. Surveys included acoustic work to determine depths and bottom types, the collection of bottom samples, the measure of oceanographic data. Fifty-four fishing operations were carried out during daytime and six during the night.
As a result of the small size of the trawler used during the surveys, the waters from 61 m to 100 m were inadequately explored; 96 percent of the hauls took place in the 9–60 m depth interval. As a consequence, the deep part of the sparid assemblage was seldom sampled.
Sediments collected during the surveys were classified according to their granulometric characteristics. The definition of mud given by Crosnier and Berrit for Togo-Benin and Martin (1973) for the Ivory Coast corresponded approximately.
The Togo-Benin continental shelf is characterized by its narrowness (maximum width = 18 nautical miles). In its shallower portion, soft deposits and rocky areas are found, where estuarine sciaenids and lutjanids find suitable living conditions. The sandy area found between 15 and 40 m depth records the lowest catch rate and constitutes a transition zone between the sparid and sciaenid assemblages. The most frequently found species among the sciaenid assemblages were, in decreasing order: B. auritus, G. decadactylus and Arius spp. Amongst the sparid group P. bellottii, S. caerulostictus and Raia miraletus (Table 5) were noted.
The “front” between the cool, saline upwelled waters (present off Ghana as well) and the warm, less saline waters of Biafra Bay is found in the Togo-Benin area. According to the information provided by Crosnier and Berrit, such a boundary is situated between Lomé and Cotonou.
Differences in oceanographic conditions between Togo and Benin are also reflected in the geographical distribution of the assemblages. During July 1964, when cold season conditions prevailed, sparid and sciaenid assemblages were found nearest to the shore, west of Grand Popo (Map 6.4). Also, the estuarine group was rarely caught in that area. During the warm season (October 1963) assemblages distribution west and east of Grand Popo (Map 6.5) were similar.
The effect of upwelling waters is more marked when the distribution of the species components of the sciaenid group is analyzed. During the cold season sciaenids are found more frequently and are more widely distributed over the shelf to the east of Grand Popo where warm and less saline waters are found (Map 6.2). When the upwelling stops, members of the sciaenid community are found all over the coast of Togo and Benin (Map 6.3).
The almost continuous distribution of P. bellottii and S. caeruleostictus from Cape Coast to Grand Popo could suggest the existence of a common sparid assemblage off Ghana-Benin.
Information about bottom types (Crosnier, 1964) and species caught (Crosnier, 1964; Robertson, 1977) was used to prepare bottom and assemblages maps. Crosnier carried out two fishing surveys off Cameroon (December 1962 to January 1963 and August 1963) where he collected data about bottom quality and fish abundance. He defined some types of bottom according to the granulometric characteristics of collected samples. His category “mud” corresponds approximately to the same used by Martin (1973). The R/V FIOLENT also carried out two surveys (April-May 1976 and September 1976), but the only published information is on species compositions for each individual haul.
Forty hauls were carried out during daylight hours and ten during the night hours during the Crosnier survey. FIOLENT carried out 15 day and 11 night hauls.
As no significant differences were observed between the Crosnier and FIOLENT surveys in respect of the geographical position of the thermocline or species composition in the catches, the results of fishing operations carried out in both periods were grouped together. Due to the limited engine power of the vessel used by Crosnier, the waters inside 10–16 m were fished more intensively than deeper ones. Only about 9 percent of the hauls were made between 61 and 110 m. The poor catch results obtained by Crosnier in this area may also be related to the vessel's limited engine power. The FIOLENT explored the waters from 21 to 108 m but did not properly cover the coastal area where the estuarine fascies of the sciaenids assemblage is to be found. Eighty percent of her hauls took place at depths greater than 50 m.
One factor likely to influence fish distribution in a very high freshwater runoff area like Cameroon is the seasonal variation in salinities: they are higher from June to September and lower in December. However, the available data are insufficient to detect any seasonal change in species distribution to be attributed to changes in salinities.
Most of the continental shelf is covered by soft deposits. The coral belt that is found near the deeper edge of the shelf in other regions of West Africa , in Cameroon is reduced to small areas south of Londji to the border of Equatorial Guinea.
Due to bottom (predominance of soft grounds) and oceanographic characteristics (high temperatures, low salinities and deep permanent superficial mixed layer) of the region, the sciaenid assemblage is well represented by a high variety of species, the ones characteristic of the estuarine fascies being widely distributed (Map 7.2). The most frequent species found were Pseudotolithus typus, Drepane africana and B. auritus and G. decadactylus. In addition, Dentex angolensis, D. congoensis and Epinephelus aeneus from the sparid assemblages were also found frequently (Table 6). The interruption of P.typus, Ilisha africana and Arius heudeloti distribution and their replacement by D. africana as the main “estuarine” type of species south of Cameroon River was one major change in the species composition of the sciaenid group off Cameroon (Map 7.3).
The most frequent species caught by the trawling surveys off Cameroon and Togo-Benin were different: whilst in Cameroon A. heudeloti, D. africana, I. africana and Pentanemus quinquarius were relatively frequent, they were rarely found (especially the latter species) off Togo-Benin. Such differences and the smallness of the individuals caught in Cameroon (Crosnier, 1964) support the idea that the populations exploited off Cameroon are rather isolated from the other areas studied.
