4. FISH DISTRIBUTION AND ABUNDANCE.

The acoustic data from the echo sounders and integrators are processed and classified in a procedure based on experience of the types of echo recordings produced by various forms of organisms and on the information gained from the fishing experiments. Three categories are distinguished:

1. Pelagic fish type I:	Clupeids e.g. sardinellas, sardines, anchovies.
2. Pelagic fish type II:	Other schooling pelagic fish such as horse mackerel, other carangids, scombrids, hairtails, barracudas etc.
3. Demersal fish:	Other than pelagic schooling fish such as hakes, sparids, sciaenids, catfishes, snappers etc.

In addition signals caused by plankton and by mesopelagic fish such as myctophids are recorded, but no further processing is being made of these data.

Assessment of the abundance of fish resources based on acoustic observatons combined with experimental fishing is a method which especially lends itself to fish found in schools or other aggregations in mid water. This is however a type of behaviour which is characteristic for most of Angolas fish species of commercial importance. There are some strictly bottom dwelling fish e.g. rays and flounders which will escape acoustic detection. Also fish in the very surface layer can not be recorded by echo sounder, but schools may be detected by horizontal ranging sonar. For navigational reasons the work with the R/V “Dr. Fridtjof Nansen” is limited to waters deeper than about 15 m. Such very shallow waters form only a small part of the Angolan shelf and are mainly found in the north. The effects of any of these factors on the acoustic assessment will be towards an underestimate.

Figures 7 through 12 show the distribution of the three categories u£ fish mentioned above. The density of fish is indicated by three arbitrary levels of the observations from the echosounder-integrator system.

In the Baia dos Tigres area the occurrence of pelagic I group was limited to a very dense and limited distribution of sardinella south of Porto Alexandre. This aggregation is estimated to 70 thousand tonnes, but should be considered as a tentative figure as the few measurements are statistically unreliable. The pelagic II group is mainly horse mackerel. Due to the rough weather in the area it was difficult to detect fish close to the seabed and the demersal group is hence probably underestimated.

Between Porto Alexandre and Benguela the registrations consisted mainly of horsemackerel. Some few but dense concentrations of sardinella were found close to the shore.

Between Benguela and Luanda sardinella were recorded in partially dense schools. A main area of concentration was found off Lobito. Most of the sardinella in the samples were in the immediate prespawning Phase. In the same area the pelagic II group consisted of Trachinotus ovatus, Trichiurus lepturus, Chloroscombrus chrysurus and Balistes capriscus. Horsemackerels were also represented but mainly small sized and juveniles.

From Luanda north to Ambrizerte the distribution of the pelagic 1 group was limited to shallow waters. The trawl samples indicated that this group consisted mainly of Sardinella maderensis and Sardinella aurita. In the northern area near the Congo River and in Cabinda, Sardinella maderensis dominated offshore while Ilisha africana occurred in the samples taken closer to the shore.

The distribution of the pelagic II group gave only scattered registrations in the area from Luanda to Cabinda. The exeption was dense registrations of a mixture of Chloroscombrus chrysurus and Trachurus trecae in a limited area of shallow water between Luanda and Ambriz. Further offshore, between Luanda and Ambriz, scattered registrations of mostly juvenile horsemackerel were observed. Further offshore north to Cabinda, the hairtail Trichiurus lepturus, was the most numerous species within this group. Selene dorsalis occurred in a few trawl samples off the Congo River.

In the shallow waters from Luanda to Ambriz the bigeye grunt, Brachydeuterus auritus was the most abundant species of the demersal group. The seabreams, Dentex spp. and Pagellus spp., dominated the outer shelf from Luanda to Cabinda. In a limited area just north of the Congo River the trawl samples indicated a small area of croakers (Paeudotholithus typus).

Provisional estimates of the biomass of the various fish categories are made, but they should be treated with some reservations and may be subject to amendments later. Further analyses of the size and species composition are required and more information is needed on the properties of the various species as acoustic targets.

Rounded figures are of the estimated biomass from the two surveys are: (thousand tonnes):

Survey no 2:

	Pelagic 1	Pelagic 2	Demersal	Total
Baia dos Tigres area:	70*	100	5	175
Porto Alexandre- Benguela:	40	30	10	80
Benguela - Luanda:	200	70	45	315
Luanda- Cabinda:	185	50	50	285
Total:	495	250	110	855

Survey no. 1

	Pelagic 1	Pelagic 2	Demersal	Total
Baia dos Tigres area:	50	120	60	230
Porto Alexandre- Benguela:	4	20	8	32
Benguela - Luanda:	200	400	70	670
Luanda- Cabinda:	220		95	315
Total:	474	540	233	1247

The total estimated biomass from the April-May survey of about 850 000 tonnes is about 400 000 tonnes lower than that of the January-February survey. The demersal estimate is about 120 000 tonnes lower, and shows a reduction in nearly all areas. The very low figure for the Baia dos Tigres area is, however, probably caused by difficulties in recording bottom fish due to bad weather at the time of the survey. The greatest difference, however, appears in the estimate of the Pelagic 2 category of fish, horse mackerels, carangids, scombrids etc. in the Benguela-Luanda area. During the January-February survey this category of fish was identified as horse mackerel and a number of other species of Chloroscombrus, Sphyraena, Caranx, Decapterus, smaller tunas and other scombrids. Large size horse mackerel was found in dense concentrations in several localities. Over the middle part of the shelf dense fast swimming schools were found which proved difficult to catch. Small tuna were observed in one locality and a few were caught. Tunas and other large scombrids are known to have seasonal migrations off the shelf. The absence of these dense schools during the second survey and of large horse mackerel could be explained by an offshore migration at this time outside the area covered by the survey. It may be of interest to note that in the upwelling system in the Southeast Pacific off Peru and Chile large size horse mackerel has been found to undertake long oceanic migration, and in later years a fishery of the order 0.5 million tonnes has been conducted on these stocks more than 200 n.m. offshore. Also in the California current system large horse mackerel is known to occur far off the shelf. Whether the horse mackerel of Namibia and Angola have a similar behaviour is unknown, but the large reduction in biomass of the Pelagic II category from Survey 1 to Survey 2 may indicate that such off-shore migrations take place. There is thus a need to consider whether to extend the survey further off the shelf.