DISTRIBUTION OF SPECIES
The species identified, except the Diaphus species, from the cruises are listed in Table 1. This table does not give a complete picture of the species taken as only the most numerous species are worked up in some of the samples. The genus Diaphus are only partly worked up and a more complete treatment of this genus must await the revision of Indian Ocean Diaphus being prepared by NAFPAKTITIS (pers. comm.). Some identifications have, however, been carried out based on reference specimens kindly identified by Dr. Nafpaktitis (Table 2).
As expected from the distribution of the sampling, the neretic and surface migrating species are dominating. Most of the species are well known from the area, but many of the records from the Gulf of Aden are new as this area has been little studied previously.
Area A. Gulf of Oman (Fig. 1)
Benthosema pterotum was the only myctophid species caught in the area. Larval studies (NELLEN 1973) gave the same result.
Area B. Coast of Pakistan
Benthosema pterotum was dominating in this area too on all the cuises. Ranging next in abundance were various Diaphus species of which D. thiollierei and D. ashmeadi (?) have been identified. Benthosema fibulatum, Hygophum proximum, Symbolophorus evermanni and Bolinichthys longipes were occasionally caught.
As in area B no sampling of mesopelagic fish has been carried out previously but larval samples (NELLEN 1973, ALI KHAN 1976) supports the impression of a low diversity.
Area C. The Arabian coast and the oceanic area 20° and 24°N.
On cruises 3 and 6, both from autumn, Benthosema pterotum was dominating. During cruise 4 from the early spring B. pterotum and B. fibulatum were about equally abundant. Various Diaphus species and Bolinichthys longipes were also present. These results differ little from those obtained by KOTTHAUS (1972) from the same area.
Area D. Arabian Coast between 15° and 20°N.
On cruise 3, Benthosema fibulatum was the dominating species while B. pterotum and B. fibulatum were about equally abundant on cruise 4. On cruises 5 and 6 various Diaphus species dominated, of which D. regani, D. thiollierei and D. ashmeadi(?) were the most abundant species. D. lütkeni was also identified. Myctophum spinosum, Symbolophorus evermanni, Bolinichthys longipes and Lampanyctus macropterus were also caught.
The nearshore mesopelagic fauna of this area has not been studied previously. A comparison with NAFPAKTITIS & NAFPAKTITIS (1969) data from the offshore regions between the same latitudes suggests that the fauna there is much more diverse than the nearshore fauna. Area E. Gulf of Aden, West of 47°E.
In the inner part of the Gulf of Aden, Benthosema pterotum was the dominating species except on cruises 3 and 6, both carried out during autumn. During cruise 3 Symbolophorus evermanni dominated and during cruise 6 S. evermanni and B. pterotum were equally abundant. Ranging next were Diaphus spp. and Myctophum spinosum. M. nitidulum, M. aurolaternatum and Hygophum proximum were also observed. Neither of the Symbolophorus, Myctophum nor Hygophum species seem to have been reported from the Gulf of Aden previously. For M. nitidulum the records from the Gulf of Aden are the northernmost known from the Indian Ocean.
Area F. Gulf of Aden between 47°E and 51°E.
The myctophid fauna of the outer part of the Gulf of Aden was the most diverse observed during the cruises. Dominating species were Benthosema fibulatum, B. pterotum and Diaphus spp. Of the species identified from this area (see Table 1) Hygophum proximum, the Myctophum species, Symbolophorus evermanni, Lampadena luminosa, Bolinichthys longipes, Lampanyctus tenuiformes and Ceratoscopelus warmingi are not previously reported from the Gulf of Aden. For L. luminosa this record seems to be a northward extension of its known range in the Indian Ocean.
Area C. Somali Coast between 10° and 15°N.
On cruises 3, 4 and 6 Benthosema fibulatum was the dominating species in this area, while Sympolophorus evermanni was most abundant during cruise 1. B. pterotum, Hygophum proximum, Myctophum nitidulum, M. spinosum, and Lampanyctus tenuiformes were also caught. Several Diaphus species were abundant but only D. regani and D. thiollierei are identified. The records are within the known geographical range of these species.
Areas H and I. Coast of Africa between 0 and 10°N.
Benthosema fibulatum, Myctophum brachygnathum and Diaphus spp. dominated the catches. Of other species Electrona rissoi, Hygophum proximum, B. pterotum, Symbolophorus evermanni, Bolinichthys longipes, Lampanyctus tenuiformes, L. nobilis, Diaphus garmani and D. thoillierei were caught. The catch of B. pterotum at 3 17’N is a southward extension of the known range of this species in African waters.
BEHAVIOUR
To study the diurnal variation in the behaviour of the fish and its influence on the echo recorings, a diurnal station was conducted in the Gulf of Oman (24° 35’N 57° 11’E) from 5 to 6 March 1976. Benthosema pterotum was the only myctophid fish found in the area and during day-time this was distributed in two layers. The upper one (layer A) generally lay between 130 and 200 m depth (Figs. 2, 3). During the first day, its mean contribution to the integrated echo intensity was 292 mm per nautical mile. This layer consisted of very dence aggregations and often descrete schools. The lower layer (B), which was more diffuse, generally lay between 220 and 300 m, sometimes extending down to about 350 m. Its contribution to the integrated echo intencity was 200 and 348 mm per n.mile during the first and the second day respectively.
Both layers were sampled, but no difference in length, maturity of gonads, fullness of stomachs or of digestion of the stomach contents could be observed between fishes from the two layers. The migration towards the surface started about 30 minutes before sunset, and the two layers joined at depths between 10 and 100 m within half an hour after sunset. During night the most dense concentrations were observed between 10 and 50 m depth but more diffuse recordings were obtained down to about 200 m. About 30 minutes before sunrise the layers separated and migrated down to their daytime depths. The integrated echo intensity was raising from about 300 mm per n.mile to 500 mm per n.mile during the night, probably due to fish drifting or swimming into the area.
During daytime the depth of the A layer approximately corresponded to the O2 - minimum where less than 1.5 ml/l of O2 were present (Fig. 5). The salinity and temperature also had minima at this depth. Comparison of hydrographical data and fish distribution from other areas showed that the myctophids were often found in water with less than 0.5 ml O2 per liter (Fig. 6). The migration pattern observed during the diurnal station is rather typical of the neretic areas where Benthosema pterotum and B. fibulatum dominated.
In offshore areas a DSL in depth varying between about 250 - 350 m was the most general feature. This layer was similar to layer B at the diurnal station and it migrated towards the surface during night. In some areas an additional layer was observed between about 350 and 500 m depth. This layer gave much more diffuse recordings on the 38 kHz echo sounder and it contributed little to the integrated echo intensities, as compared to the other layers. This layer or parts of it was sometimes found at the same depth also during night time.
The echo recordings obtained at the diurnal station (Fig. 4) may suggest a decrease in echo abundance during sunrise and an increase during sunset. The variation due to other sources makes it, however, impossible to draw any conclusions.
To further study whether there is a diurnal variation in the echo recordings from mesopelagic fish in the upper 450 meters, data from some sections (Nos. 1-6 on Fig. 1) were analysed (Fig. 7). To test whether there was a difference in mean abundance during day and night, the recordings were transformed using ln(M+1) where M is five miles average of integrator deflection per n.mile. The values obtained during sunrise or sunset were not included in the analysis. The hypotheses that the mean of the 400 recordings of M made during day were similar to those 345 made during night, could not be rejected (t = 1.19 p < 0.05). It is therefore concluded that although part of the fish stock may occasionally stay above the upper limit of integration (8 m) during the night or below the lowest limit (450 m) during the day, this does not give a serious difference between day and night recordings. These results contrast, however, with those obtained off Pakistan during the summer 1977 by MYRSETH (in prep.). The data from Section 1 from the Pakistanean area were therefore analysed separately. The result (t = 0.74 p < 0.05) was consistent with that based on the whole material.
The integrated echo intensities along six sections (see Fig. 1) extending between 40 and 200 miles offshore were recorded (Fig. 7). To test whether there was a consistent relation between integrated intensities and distance from the shore, a regression of the transformed integrator outputs ln(M+1) on the distance from the shore was carried out. The significance of the regression coefficients was tested using analysis of variance tests (e.g. ZAR 1974). Of the 26 tests carried out, 16 were significantly different from zero (Table 3). Of these, 12 were positive, i.e. the echo intensities increased with distance from shore, and four were negative.
Off the Pakistanean coast (section 1) four of the five tests were nonsignificant. In the Gulf of Oman (section 2) intensity increased offshore on cruises 2 and 4, but decreased on cruise 6. For the other cruises the tests were nonsignificant. Off the Kuria Muria Islands (section 3) echo intensities decreased offshore on cruises 4 and 6 and increased on cruise 5. In the Gulf of Aden (section 4) and off Al Arar (section 5) most cruises showed a significant increase in echo intensity with increasing distance from the shore.
Although not obvious from the sections, an increased echo intensity was often observed from the 200 m depth contour and about 1-2 miles offshore or less. It seems, however, safe to conclude that generally the survey design in relation to the shore is of minor importance compared to other sources of variance.
The sections off Kuria Muria Islands were run twice in April 1975. The mean and standard deviations of the integrated intensities were 114.5 ± 48.9 and 106.8 ± 53.2 respectively. A test carried out on the transformed data [ln(M + 1)] showed that the two means were not significantly different (t = 0.8 p < 0.05).
ABUNDANCE ESTIMATES
To estimate the abundance of mesopelagic fishes the mean integrator reading for each of the areas A - I (Fig. 1) was multiplied by the size of the area outside the 200 m depth contour. For each of these areas an average fish length was calculated (Table 4), so that the lengths were approximately weighted by the numerical abundance in the layers where they were caught. Fishes that were only caught in bottom trawl, e.g. Diaphus coeruleus, are not included in the mean. Table 4 also shows the species mainly contributing to the recordings. When no samples were available for an area, the lengths at stations close to the border of the area or the length in the same area at another cruise conducted at the same time of the year was used.
Based on these mean lengths an integration constant CW was calculated for each area and each survey. The more accurate method recommended by FORBES & NAKKEN (1972) when several species or length groups contribute to the recordings was not used, as the extra accuracy is probably not justified by the data. The mean integrator readings referred to mesopelagic fish in each area and for each cruise are shown in Fig. 9 and the corresponding abundance estimates in Table 5.
The total abundance recorded during one survey varied between 56 million tons (summer 1977) and 148 million tons (spring 1976) with a mean of 102 million tons. Both the spring surveys (cruises 1, 2 and 4) gave higher abundances than the summer (cruise 5) and the autumn cruises (cruises 3 and 6).
As indicated by the sections (Fig. 7), the differences in abundance between the cruises were the same both near shore and offshore. A difference in vertical distribution between seasons may give a difference, as observed, if a larger part of the fish is found below 450 m during summer and autumn than during spring. This should, however, lead to a diurnal variation in abundance as most of the myctophids is supposed to rise above 450 m during nighttime.
The highest densities were usually recorded in the Gulf of Oman where the mean recordings varied between 374 and 118 mm deflection, corresponding to approximately 88 and 35 fish or 63 g and 25 g per m2 surface area. The lowest density, 9 mm deflection, was recorded off north-western Somalia during summer and corresponds to about 2 fish or 5 g per m2 surface area.
CATCH RATES
Trawling was carried out to identify sound scatterers and to get biological samples and in a few cases to compare a krill trawl and a pelagic fish trawl. Although it was no specific goal to get large catches, high catch rates were obtained at some stations. 26 stations gave catch rates higher than 400 kg per hour of trawling (Table 6). Of these, 18 stations gave catch rates of 1000 kg/hour or more and six of 5000 kg/hour or more. The highest catch rate recorded was 20,000 kg/hour (st. 427).
Eleven of the stations listed in Table 6 were from the Gulf of Oman (Fig. 9), six from Pakistanean waters, six from the coast of Arabia, two from the Gulf of Aden and one from south-west of Socotra. One of the stations was taken with bottom trawl and the others with pelagic trawl. Thirteen of the stations were taken during daytime, twelve during nighttime and one at dawn.
The species composition was studied in 24 of the catches. Benthosema pterotum was the only one or the dominating species at 21 of these stations and B. fibulatum at two of them. At one station Diaphus spp. were most abundant.
A comparison between the 1360 mesh pelagic trawl and the krill trawl showed that although the opening of the krill trawl was only a quarter of the opening of the other one, the two gears caught equal quantities of fish (ANON. 1976c). This indicates that a large part of the fish entering the pelagic trawl is filtered off through the meshes, while the much smaller meshes of the krill trawl retain a larger part of the fish.
