7. SMALL PELAGIC FISH

7.1 Species composition
7.2 The distribution
7.3 Biological characteristics
7.4 Abundance

7.1 Species composition

This group consists of the smaller schooling fish with a diurnal vertical migration pattern, and includes the following species:

ANCHOVY	- Stolephorus spp.
BARRACUDA	- Sphyreana spp.
DRIFTFISH	- Ariomma spp.
JACK	- Alepes sp., Carangoides spp., Caranx spp.
MACKEREL	- Rastrelliger spp., Scomber sp.
PONYFISH	- Leiognathidae
SARDINE	- Dussumieria sp., Etrumeus sp., Hilsa keelee, Pellona ditchela, Sardinella spp., Thryssa spp.
SCAD	- Decapterus spp., Trachurus sp.

Due to the deviation both in behaviour and biology of the Stolephorus spp. from the rest of the Engraulidae and Clupeidae families, these species have been put into a separate group, the ANCHOVY.

Along the whole coast of Mozambique the recordings of small pelagic fish were mainly confined to the shelf area, i.e. at depths less than 200 m.

To establish the pelagic species composition in an area based on catch data alone is rather dubious, due to the large space and time variations both in abundance and composition. Most of the pelagic species carry out regular vertical migrations. During daytime they occur as schools close to, or a few meters above, the bottom, and are during that time accessible to bottom trawls. At night the pelagic species disperse in the water column and are hence not caught with bottom trawls. This feature has been clearly demonstrated by BIRKETT (1978). By comparing the catches of pelagic fish in bottom trawls during day and night at Sofala Bank he shows that the night catch was only 2-5% of the day catch. The small night-time catches were probably accidental catches made during the few minutes the trawl was in mid-water.

Table 7.1. Pelagic species composition in the bottom trawls of “Aelita” and “Professor Mesyatsev” 20-100 m (% weight).

Vessel	Aelita						Prof. Mesyatsev
Area	Sofala		Bazaruto		Delagoa		Sofala		Bazaruto		Delagoa
Species	Jun-Aug 76	Oct-Dec 76	Apr 77	Jun 77	Dec-Jan 77	Apr-Jun 77	Jan 76	Aug 77	Jan 76	Aug 77	Jan 76	Aug 77
Barracudas - SPHYRAENIDAE			15.6	80.6		36.6	1.8	0.5		66	0.4
Driftfish - ARIOMMIDAE	2.3	34.5		5.3	1.9	35.9	17.7	4.2				59.5
Mackerel - SCOMBRIDAE	26.8	12.2					20.8	14.8			4.5	18.5
Ponyfish - LEIOGNATHIDAE		4.3			26.1	9.7	5.8	3.6
Sardines - CLUPEIDAE/ENGRAULIDAE	29.3	0.4					16.6	14.2				1.2
Scads/Jacks - CARANGIDAE	41.6	48.6	84.4	14.1	72.0	17.8	37.3	62.7	100	34	95.1	20.8

Table 7.1 shows the pelagic species composition in the bottom trawls of the “Aelita” and “Professor Mesyatsev” survey. Table 7.2 gives the same for the Soviet commercial trawlers for some months from which the data seems reliable. As can be seen, the most important pelagic group seems to be the CARANGIDAE, of which the two species round scad (Decapterus maruadsi) and horse mackerel (Trachurus trachurus) are dominant at Sofala Bank and Delagoa Bay respectively. In Delagoa Bay the horse mackerel is most abundant during summer and significantly less so in winter.

Table 7.2. Pelagic species composition in bottom trawl catches from a Soviet commercial trawler, 25-100 m (% weight).

	Sofala		Delagoa
	Dec 77	Jun 77	Nov 77
Barracudas - SPHYRAENIDAE	7.9	0.8	8.9
Driftfish - ARIOMMIDAE	19.9	-	4.8
Mackerel - SCOMBRIDAE	1.7	8.1	0.3
Sardines - CLUPEIDAE/ENGRAULIDAE	0.9	17.4	5.3
Scads/Jacks - CARANGIDAE	69.6	73.7	80.7
Hours of trawling	550	948	620

The sardine group consists mainly of the two species Indian pellona (Pellona ditchela) and orangemouth thryssa (Thryssa vitrirostris). They seem to be more frequent in bottom trawl catches during winter. The same pattern is shown by ponyfish (Leiognathus spp.) and barracuda (Sphyraena spp.). The latter species is especially dominant in the Bazaruto area. Driftfish (Ariomma sp.) contribute significantly to the catches at Sofala Bank in summer and in Delagoa Bay in winter.

Table 7.3. Pelagic species composition in trawl catches from Sofala Bank, 10-50 m (% weight).

Vessel	KATTEGAT (Summer)		F. NANSEN (Summer)		F. NANSEN (Winter)
Gear	Bottom trawl	Pelagic trawl	Bottom trawl	Pelagic trawl	Bottom trawl	Pelagic trawl
Anchovies - STOLEPHORUS SP.				69.3		81.0
Barracudas - SPYRAENIDAE						2.9
Mackerel - SCOMBRIDAE	5.9	43.0	7.2	26.7	4.0	2.6
Sardines - CLUPEIDAE/ENGRAULIDAE	32.7	49.4	92.8		76.6	9.9
Scads - DECAPTERUS SPP.	29.0	0.6		0.8	1.9	3.5
Jacks - CARANGOIDES SP.	32.4	7.0		3.2	17.5	0.1

Table 7.3 shows the pelagic species composition at Sofala Bank in both pelagic and bottom trawls from “Kattegat” and “Dr. Fridtjof Nansen”. As can be seen, the composition of the catch from the two different gears deviates considerably. Some of this discrepancy can be explained by the distribution of day and night hauls. The anchovy (Stolephorus sp.), however, never occurred in bottom trawls. The reason why it did not occur in the pelagic trawl of “Kattegat” is probably due to misidentification, so the species has been put in the sardine group. Anchovies appear to be one of the major pelagic resources off Mozambique. As seen from Tables 7.1 - 7.3, a bottom trawl survey will not include this important species.

In order to reach a better estimate of the pelagic species composition, the catch data and the acoustic abundance data from “Dr. Fridtjof Nansen” were combined. The pelagic species composition in each trawl catch was regarded as representative for a sub-area. The pelagic echo abundance was split according to this composition and summed up for larger areas. The result is shown in Table 7.4. Table 7.5 lists the main pelagic species recorded by “Dr. Fridtjof Nansen”.

Fig. 7.1. Small pelagic fish distribution in September 1977.

Fig. 7.2. Small pelagic fish distribution in November 1977.

Fig. 7.3. Small pelagic fish distribution in February -March 1978.

Fig. 7.4. Small pelagic fish distribution in April -June 1978.

Table 7.4. Pelagic species composition from combined catch and acoustic data of the “Dr. Fridtjof Nansen” survey (% weight).

Area	Sofala			Bazaruto			Delagoa
Species	Sep 77	Oct/Nov 77	Apr/Jun 78	Sep 77	Oct/Nov 77	Apr/Jun 78	Sep 77	Oct/Nov 77	Apr/Jun 78
Anchovies - STOLEPHORUS SP.	70	20	40	33					9
Barracudas - SPHYRAENIDAE				10		5	32
Driftfish - ARIOMMIDAE						5			35
Mackerel - SCOMBRIDAE
Ponyfish - LEIOGNATHIDAE			10			10
Sardines - CLUPEIDAE/ENGRAULIDAE	20	60	40	3		10	10		39
Scads/Jacks - CARANGIDAE	10	20	10	54	100	70	58	100	17

7.2 The distribution

Figs. 7.1-7.4 show the distribution of the most important small pelagic species during the four cruises of “Dr. Fridtjof Nansen”.

Northern area

At St. Lazarus Bank schools of banded barracuda (Sphyraena jello) and scad (Decapterus sp.) were observed in September 1977. North of Nacala, round scad (Decapterus maruadsi) were found along the bottom at about 200 m at the same time (Fig. 7.1). In February 1978 round scad and ponyfish (Leiognathus sp.) were found in the same area (Fig. 7.3). “Professor Mesyatsev” in January 1976 had a good catch of round scad from 186 m, just north of Nacala. As also pointed out by BIRKETT (1978), this may indicate that there is a continuous distribution of scad along the whole coast of Eastern Africa. In October 1977 “Kattegat” observed schools of scad in the area between Angoche and Pemba (ANON, 1978 e), which seem to confirm this. In November 1977 and in April 1978, however, there were no similar recordings (Figs. 7.2 and 7.3).

Table 7.5. Small pelagic species from the cruises of “Dr. Fridtjof Nansen”.

*) The most abundant species.

At Sofala Bank the following species were dominant: Buccaneer anchovy (Stolephorus buccanerii), Indian pellona (Pellona ditchela), orangemouth thryssa (Thryssa vitrirostris and round scad (Decapterus maruadsi). Orangemouth thryssa and Indian pellona shared a common distribution area and were not usually found in deeper water than 40 m. The orangemouth thryssa seemed to have a more near-shore distribution, as this species was dominant in catches from shallower depths than 20 m, while Indian pellona contributed most at depths between 20 and 40 m. Scad were mainly observed at depths between 40 and 100 m and anchovies between 20 and 60 m.

There seemed to be a migration to the northern part of Sofala Bank of anchovies from November to February. In April anchovies again had approximately the same distribution area as in September.

Bazaruto area

In this area the pelagic recordings were usually made up by the following species: Round scad, barracuda (Sphyraena japonica, S. obtusata) and round herring (Etrumeus teres). In addition, some driftfish (Ariomma indica) occurred in March. Off Imhambane concentrations of porcupine fish (Diodon maculifer) were observed as schools during all four cruises. This contribution to the acoustic pelagic recordings has been deleted as this species is of no commercial value.

Delagoa Bay

The most abundant Carangidae in this area were horse mackerel (Trachurus trachurus) which were only rarely observed further north. At Boa-Paz Bank schools of layang scad (Decapterus macrosoma) were observed in September 1977. Also barracuda (Sphyraena obtusata, S. japonica) and Indian driftfish (Ariomma indica) were abundant during winter at depths more than 100 m.

Fig. 7.5. Distribution of modal length and range for some small pelagic species. (A)

Fig. 7.5. Distribution of modal length and range for some small pelagic species. (B)

Fig. 7.5. Distribution of modal length and range for some small pelagic species. (C)

Fig. 7.5. Distribution of modal length and range for some small pelagic species. (D)

Fig. 7.5. Distribution of modal length and range for some small pelagic species. (E)

Fig. 7.5. Distribution of modal length and range for some small pelagic species. (F)

Fig. 7.6. Distribution of modal length and range for some small pelagic species. (A)

Fig. 7.6. Distribution of modal length and range for some small pelagic species. (B)

Fig. 7.6. Distribution of modal length and range for some small pelagic species. (C)

Fig. 7.6. Distribution of modal length and range for some small pelagic species. (D)

Inhaca

The only pelagic species observed in this area was round herring, in June 1978.

7.3 Biological characteristics

Figs. 7.5 and 7.6 give the mode and range of the length as well as observations on maturity stages for the most important pelagic species.

Fig. 7.7. Night recording of buccaneer anchovy.

Anchovy

This group was exclusively made up of the buccaneer anchovy (Stolephorus buccaneerii). During night-time the anchovy was dispersed more or less in the whole water column, as seen in Fig. 7.7. During daytime, however, it formed schools in mid-water (Fig. 7.8) and not close to the bottom as did most of the other pelagic species. As seen in Fig. 7.5, the smallest individuals occurred in May and September, thus indicating the main spawning periods to be April and August. Observations from India (ANON., 1976 a) show that some closely related species have very rapid growth and that the adult size is reached after about 6 months. The temperature conditions off the western Indian coast and off Mozambique are approximately the same, and therefore the same growth pattern should be expected along the coast of Mozambique. The apparent migration of this species on Sofala Bank is most likely associated with the main spawning which seemed to take place at the southern part of Sofala Bank at depths of 30-60 m.

Fig. 7.8. Day recording of buccaneer anchovy.

Barracuda

This family includes the four species listed in Table 7.5 of which obtuse barracuda (Sphyraena obtusata) were the most abundant. Barracuda seemed to form commercial concentrations at 200-250 m depths, during April-July in the Bazaruto area. Fig. 7.9 shows an echo recording of Sphyraena japonica off Bazaruto in March at 260 m depth, resulting in a catch rate of about 2 tonnes/hr. The modal length distribution in Fig. 7.6, together with the maturity observations from “Aelita” (BUDNICHENKO et al., 1977), suggests a main spawning period in June-August for the obtuse barracuda.

Driftfish

The Indian driftfish (Ariomma indica) was found at depths between 30 to 350 m but mainly at 50-100 m. There seemed to be a migration of driftfish from Delagoa Bay to Sofala Bank in September-October and back again in February-March. The species composition in Tables 7.1 and 7.2 supports this view. The modal length distribution in Fig. 7.6 gives no indications as to any preferred spawning period. However, spent females were observed in the Bazaruto area in January and maturity observations from “Aelita” (BUDNICHENKO et al., 1977) indicate another spawning in July-August.

Fig. 7.9. Echo recording of barracuda at the bottom.

Ponyfish

This group consists of the following species: Secutor insidiator and S. ruconius, Gazza minuta, Leiognathus elongatus and L. equulus, of which the latter is the most abundant. The ponyfish showed typical pelagic fish behaviour and can occur in large schools at the bottom during daytime, as can be seen in Fig. 7.10, catch rates of up to 3.5 tonnes/hour can be achieved. The modal length distribution in Fig. 7.5 gives some support for a spawning in April for L. elongatus. The observation of small juvenile Leiognathus sp. 1-5 cm long in April might mean that others of the ponyfish group were also spawning at the same time.

Sardine

Spawning females of the Indian pellona (Pellona ditchela) were observed in April, the smallest individuals being noted in April-May (Fig. 7.5). The main spawning season therefore seemed to be in March-May. Most likely there was also a secondary peak in the spawning activity from August to November. During daytime the pellona formed schools close to the bottom, as seen in Fig. 7.11.

Fig. 7.10. Echo recording of ponyfish at the bottom during day time.

Fig. 7.11. Day recording of Indian pellona.

The Indian pellona was usually found together with orangemouth thryssa (Thryssa vitrirostris) which seemed to have a peak spawning period in March-May (Fig. 7.5).

The round herring (Etrumeus teres) was only observed south of Sofala Bank. Juveniles 4-7 cm long were observed in September and a few spawning females in May.

A few catches of kelee shad (Hilsa kelee) were made at Sofala Bank, but this species had probably a very near-shore distribution and was therefore rarely observed by the research vessels. A fishery on this species is carried out in shallow water in the bays of Maputo and Beira.

Scad/Jack

The most abundant species of this group was round scad (Decapterus maruadsi). It was often recorded together with layang scad (Decapterus macrosoma). As can be seen in Fig. 7.5 the smallest individuals of round scad were found in May and September. This, in addition to the observation of spawning females in April by “Aelita”, (BUDNICHENKO et al., 1977) indicates peak spawning periods in April and September. Round scad probably migrate to deeper water for spawning. This agrees with the observations on the Californian Jack mackerel (Trachurus symetricus) which spawns in offshore deeper grounds (FREY, 1971).

Support to this theory is also given by the reduction of catch in the 0-50 m layer, with a corresponding increase from 50-100 m during March-May by the Soviet commercial trawler. “Professor Mesyatsev” observed the same tendency in August 1977 compared to January 1976 (BIRKETT, 1978). In October 1977 “Professor Mesyatsev” recorded substantial concentrations of round scad at the Mahé plateau (BIRKETT, 1978), but in July 1978 mainly juveniles were observed (ANON, 1978 d). Of the few recordings of adults, about 20% were in maturity stages 5 and Fig. 7.12 shows a typical day recording of round scad. As can be seen, they were observed as schools at the bottom while the sardine schools were usually found a few meters above the bottom. During night-time they were dispersed in the upper part of the water column.

Fig. 7.12. Day recording of round scad.

Layang scad seemed to have a different spawning time to the round scad. Observations from “Aelita” (BUDNICHENKO et al., 1977) showed that more than 50% of the females were in maturity stage 5 in December 1976. In April the majority of the females were in stages 1-3, and in June 100% were in stage 3. Though perhaps somewhat dubious, this could indicate only one main spawning period in January-February, which also seems to fit the distribution of modal length in Fig. 7.5.

The malabar cavalla (Carangoides malabaricus) (Fig. 7.6) was found along most of the coast at depths between 20 and 100 m. No spawning or near-spawning females were observed on any of the cruises.

The horse mackerel (Trachurus trachurus) is not usually observed north of Bazaruto, and is most abundant at Boa-Paz Bank. It occurred at depths between 30 and 150 m, with the greatest concentrations at about 50 m. A main spawning season seemed to take place in September-October.

7.4 Abundance

Table 7.6 shows the acoustic abundance estimates of the main pelagic stocks. As can be seen, the most important species was the buccaneer anchovy, amounting to about 300 000 tonnes, followed by scad, Indian pellona and orangemouth thryssa.

Table 7.6. Acoustic estimates of the main pelagic stocks from “Dr. Fridtjof Nansen” (thousand tonnes.)

Time	Species	St. Lazarus-Delgado	Sofala Bay	Bazaruto	Delagoa	Inhaca	Total
Sep 1977	Decapterus spp.	32	41	32	5	-	110
	Etrumeus sp.			2	2		4
	Sphyreana spp.	5		6	6		17
	Stolephorus sp.		290	20			310
	Pellona sp.		62				62
	Thryssa sp.		20				20
	Trachurus sp.				6		6
	Total	37	413	60	19	-	529
Oct-Nov 1977	Decapterus spp.		30	1	5		36
	Stolephorus sp.		30				30
	Pellona sp.		50				50
	Thryssa sp.		50				50
	Trachurus sp.				4		4
	Total		160	1	9		170
Apr-Jun 1978	Decapterus spp.		25	26			51
	Stolephorus sp.		95		5		100
	Sphyreana spp.			2			2
	Etrumeus sp.			4	5	5	14
	Pellona sp.		45				45
	Thryssa sp.		45		16		61
	Leiognathus sp.		25	4			29
	Ariomma sp.			2	17		19
	Trachurus sp.				5		5
	Megalaspis sp.				5		5
	Total		235	38	53	5	331

BIRKETT (1978) presents abundance estimates for pelagic fish from the “Professor Mesyatsev” investigation (Table 1.5). For Sofala Bank he gives 94 500 and 42 400 tonnes for January and August respectively. For Delagoa Bay the corresponding figures are 40 200 and 4600 tonnes. These estimates are much lower than those presented in Table 7.6. Birkett’s values were obtained by “the swept area” method. Only day hauls were used, and the efficiency coefficient of the trawl was taken as 1.

The main sources of error in such calculations are the efficiency coefficient and the necessary assumption that the trawl hauls are representative of the general fish density within the surveyed area. The supposition of 1 as the efficiency coefficient will for pelagic species result in a underestimate of the fish density, as echo recordings have shown that even during daytime a proportion of the pelagic population is situated well above the headline height. It is also likely that some of the pelagic and fast-swimming species ahead of the trawl are able to avoid the net.

If the trawl hauls are going to be representative for the whole surveyed area, random chosen fishing localities should be used. The trawl hauls of “Professor Mesyatsev” did not fully satisfy this condition for the pelagic species. However, the distribution of trawl stations at Sofala Bank (BIRKETT, 1978) indicates that this source of error is of minor importance off Mozambique. The assumption of 1 as the efficiency coefficient is probably more important, and thus the estimates of BIRKETT (1978) are most likely undervalued.

By removing from Table 7.6 the Stolephorus sp. at Sofala Bank which were not recorded by “Professor Mesyatsev”, the total pelagic stock estimate of “Dr. Fridtjof Nansen” in this area varies between 123 000 and 140 000 tonnes. This means that one has to multiply Birkett’s estimate by a factor between 1.3 and 3.3 in order to reach the pelagic stock estimates made by “Dr. Fridtjof Nansen”.

BURCZYNSKI (1976) presents a pelagic fish stock evaluation based on the echo survey of “Professor Mesyatsev” in January 1976 (Table 1.4). The results are based on calibration of the echo integrator by means of sample hauls. At Sofala Bank he arrived at a pelagic stock size of 65 000 tonnes and in Delagoa Bay 11 000 tonnes. Burczynski believs the last figure to be an underestimate due to bad weather conditions which dispersed the stock and an unsatisfactorily grid pattern.

Some additional information on the acoustic survey of “Professor Mesyatsev” was brought forward during the FAO/IOP Workshop on the Fishery Resources of the Western Indian Ocean south of the Equator held in the Seychelles 23 October - 4 November 1978 (ANON, 1979). According to the Soviet calculations the pelagic stock at Sofala Bank was 90 000 and 62 000 tonnes in January and February 1976 respectively, and only 5000 tonnes in August 1977. In Delagoa Bay the pelagic stock both in January 1976 and in August 1977 amounted to 7000 tonnes.

As can be seen, the echo survey results from “Professor Mesyatsev” give lower values for the pelagic stocks off Mozambique than those from “Dr. Fridtjof Nansen”. This is believed to be due to the following reasons:

- The anchovy (Stolephorus sp.) is not included in the estimates from “Professor Mesyatsev”.

- The tendency of “Professor Mesyatsev” to only count concentrations of pelagic fish and not to attempt to include also the dispersed distribution.

- The calibration procedure of the echo integrator aboard “Professor Mesyatsev” by using the trawl catches as a measure of absolute density.

Also the final report of the “Aelita” investigation (BUDNICHENKO et al., 1977) includes some abundance estimates based on the “swept area method” (Table 1.6). The catches have not been split into demersal and pelagic species, and the calculation was carried out only for a proportion of the total fishable areas. The results presented therefore give the total fish abundance and are quite obviously an underestimate.

To estimate the potential annual yields of the pelagic stocks in the area, the formula of GULLAND (1970), Y_max = 0.5·M·B_O, was applied. M is the natural mortality rate and B_O the virgin or unexploited stock. If fishing is significant some modification is needed. In this case the equation Y_max = 0.5 (C + MB₁) max was used, where C is the present catch and B₁, the exploited stock. For the sardine, ponyfish and scad, which are caught as by-catch in the shrimp fishery at Sofala Bank, the last formula was applied.

The natural mortality coefficient for the fish in the region is not satisfactorly known. Estimates for similar species in other regions make it reasonable to use M = 2 for the anchovy and M = 1 for the rest of the pelagic species (PAULY, 1978). The results of these calculations appear in Table 7.7. For B_O or B₁, the maximum stock size recorded was used. The results of Y_max in Table 7.7 therefore presuppose that the most optimum fishing pattern is used.

If the anchovy is subtracted, the maximum potential annual yield in Table 7.7 amounts to 162 000 tonnes. The pelagic stock excluding the anchovy varied between 140 000 and 231 000 tonnes, as seen in Table 7.6. If one estimates the average stock to be 200 000 tonnes, and the present catch to be 40 000 tonnes as a by-catch in the shrimp fishery, this will give a potential annual yield of 120 000 tonnes.

Table 7.7. Maximum potential annual yield of pelagic stocks (thousand tonnes).

Species	Bmax	CB	CS	M	Ymax
Anchovies	310	-	-	2	310
Barracudas	17	-	0.5	1	8
Driftfish	19	-	0.5	1	9
Mackerel	-	2	0.5	1	-
Ponyfish	29	10	-	1	20
Sardines	120	10	0.5	1	65
Scads	116	4	4.5	1	60

B_max = Maximum stock size
C_B = Estimated by-catch in shrimp fishery
C_S = Estimated Soviet catches Sep. 77 - Jun. 1978
M = Instantaneous mortality rate
Y_max = Maximum potential yield