3. RESULTS FROM THE TRAWL SURVEYS 1983-84

3.1 The Estimates of the Demersal Resources
3.2 Comparisons with Other Estimates of the Demersal Resources.
3.3 The Catches of Demersal Fish.
3.4 Distribution of Catch and Biomass by Commercial Value.
3.5 Estimates of Yield of the Demersal Resources

3.1 The Estimates of the Demersal Resources

Biomass estimates from the trawl survey programme are shown in Table 8. These estimates concern demersal fish only, as small pelagic fish has been excluded from the calculations when present in the catches. The estimates from the first surveys, given in the cruise reports, have been revised, as some night hauls had erroneously been included in the calculations.

Figure 8. The distribution of demersal fish during the first survey based on registrations by the acoustic system.

The precision of the regional estimates from single surveys is low as few hauls have been carried out in each region in the course of each survey, which gives a high standard error in the data. A random very high or low catch can severely bias the estimates when the hauls are few. This effect is smaller in the total estimates as negative and positive sampling errors tend to balance with a higher number of samples. Therefore total estimates give higher precision and are thus more reliable than the regional ones.

Figure 9. The distribution of demersal fish during the second survey based on registrations by the acoustic system.

The total estimates from the two spring surveys are corresponding, both 335 thousand tonnes, while the Nov-Dec survey gives an estimate about 20% lower, i.e. 260 thousand tonnes. As discussed below, this drop could indicate more a sampling error than an actual decline in the level of the biomass.

Figure 10. The distribution of demersal fish during the third survey based on registrations by the acoustic system.

Table 9 shows the average catches of demersal fish from all random hauls grouped by four depth zones and by the three surveys. These averages can represent indexes of the estimated fish densities in the zones.

Table 8 Trawl survey biomass estimates of demersal fish. Rounded figures thousand tonnes.

	SURVEY I Mar-Apr '83	SURVEY II Nov-Dec '83	SURVEY III May '84	AVERAGE	ADJUSTED AVERAGE	ALL SURVEYS TREATED AS ONE
A North of Ras al Hadd	N.S.	N.S.	75	(75)	(75)
B Ras al Hadd - Masira	125	60	85	90	105
C Masira Bank	50	125	115	95	95
D Sauquara & Kuria Muria Banks	160	75	135	125	145
E Salalah region	N.S	N.S	N.S
Total south of Ras al Hadd	335	260	335	310	345	315 +28% (95 % conf.lim.)

N.S.= Not surveyed

Survey	Period	10-30m	31-60m	61-100m	101-150m
I	Mar-Apr '83	600(4)	710(14)	560(8)	2270(2)
II	Nov-Dec '83	630(9)	690(14)	236(11)	45(2)
III	Apr-May '83	843(8)	630(10)	900(16)	625(2)
I-III		705(21)	681(38)	613(35)	980(6)

A drastic drop in the average catches in the deeper zones was observed during the Nov-Dec survey. This drop should be seen in connection with the presence of oxygen-depleted waters on the shelf. During this survey, the 1 ml/l oxycline was located at around 50 m bottom depth on Masirah and Sauquara banks, while in the spring surveys the same oxycline was found lower than 100 m in March 83 and from 75 to 100 m in May 84 The appearance of oxygen-depleted waters on the shelf is connected to the S.W.-monsoon-induced upwelling, likely to be at its strongest at the peak and at the end of the monsoon period. Unfortunately we have no data for this period.

The lower catches in the deeper zones during the second survey are therefore likely due to lower density of fish in this area as the fish have escaped into shallower and oxygen-rich waters. This is also reflected in the higher average catches in the shallow waters in the same period (Table 9).

Because of this seasonal migration pattern, it is important that sampling is done at all depth strata. As the migration into very shallow waters was brought to light only in later analysis, the sampling programme had not been laid out to study this phenomenon.

Only on the Masirah Bank, which consists of extensive areas of shallow waters, there has been adequate sampling of this depth stratum. In the other two regions, Ras al Hadd - Masirah and Sauquara & Kuria Muria banks, where the shallow water areas are limited, but with seasonally -high fish densities, the sampling programme does not allow to quantify the biomass and evaluate its contribution to the total biomass.

Likely, the biomass estimates in the regions Ras al Hadd - Masirah and Sauquara & Kuria Muria banks have been underestimated during the Nov-Dec survey due to undersampling in the shallower waters. The higher estimates from the two other surveys are considered more representative of the true biomass level.

To conclude, the biomass figures given in Table 7 are not believed to reflect any seasonal variation, neither within regions, nor in the total estimate. The variations shown are expected to be due to high standard error in the sampling and to undersampling in shallow waters where fish concentrate in the high and post monsoon period.

Revised estimates for the regions Ras al Hadd-Masirah and Sauquara & Kuria Muria banks, based only on the figures from the spring surveys, are 105 and 145 thousand tonnes. A total revised estimate for all Oman would then be (in thousand tonnes):

B	C	D	B - E
Ras al Hadd-Masirah	Masirah Bank	Sauquara & Kuria Muria Banks	Total south of Ras al Hadd
105	95	145	345

A total estimate with confidence limits for the whole shelf Ras al Hadd to Ras al Marbat has also been calculated directly treating all random day hauls from the whole survey programme as a single series of data (total 105 hauls). This gives an estimate of 315 thousand tonnes with a precision of + 28% within 95% confidence limit. However, this estimate also suffers from undersampling in shallow waters and is therefore probably slightly negatively biased.

Estimated mean densities of demersal fish within regions, based on the trawl survey data are (thousand tonnes):

A	B	C	D	B - D
North of Ras al Hadd	Ras al Hadd Masirah	Masirah Bank	Sauquara and Kuria Muria Banks	Ras al Hadd-Ras al Marbat
18.8	37.5	27.1	24.7	28.4

3.2 Comparisons with Other Estimates of the Demersal Resources.

The acoustic estimates of the demersal stocks from the 1975-76 “Dr Fidtjof Nansen” surveys are as follows (thousand tonnes):

Survey 1 & 2	Apr-May '75	113
Survey 3	Oct '75	115
Survey 4	Feb '76	40
Survey 5	Apr-May '76	124
Survey 6	Sep '76	127

From Kesteven et al., 1980

By comparing the acoustic and trawl estimates on demersal resources from the latest series of surveys no covariation in the estimates can be found (thousand tonnes):

	SURVEY I	SURVEY II	SURVEY III
Trawl	335	260	335
Acoustic	235	60	70

While the three trawl surveys give fairly consistent estimates, the drastic drop from the first to the second acoustic estimate can only be ascribed to limitations in the acoustic method, itself. When the fish are dispersed in a very scattered pattern, the resolution in the acoustic system is very low, and registrations fall partially below the threshold of the system. This tends to give unprecise and negatively biased estimates.

The 1975-76 surveys of the R.V. “Dr. F. Nansen” have already been critically evaluated in connection with the estimates of small pelagic fish in the previous chapter. The generally low sampling intensity during these surveys applies especially to the shallow waters which probably hold considerable resources during the peak and at the end of the SW-monsoon. This will add to the general tendency of acoustic estimates to underestimate demersal resources.

During the discussion on the estimates of the small pelagic fish, it was suggested that the differences between the pelagic estimates between the 1975-76 surveys and the ones in 1983-84, might indicate a major ecological shift in the ecosystem, giving a higher carrying capacity for the pelagic fish resources. If so, this would also partially explain the increase in the latest estimates of the demersal resources as compared to the earlier ones. However, our data are not sufficient to confirm this hypothesis.

In a study on the fishery resources of Oman, Vidal-Junemann (1981) have estimated the demersal fish resources in the region Ras al Hadd to Ras al Marbat to 260 thousand tonnes. This figure is obtained by linking together information from a shallow water trawl survey (< 40m bottom depth), with catch data from commercial trawlers on the best fishing grounds and with the acoustic estimates of demersal fish by the 1975-76 R/V “Dr. Fridtjof Nansen” surveys. If one keeps in mind that the acoustic estimates of demersal fish probably are underestimated, and that reported catches from the commercial fishery mainly concerns marketable species, this estimate does not seem to conflict with the 345 thousand tonnes estimated from our surveys. The last estimate includes all demersal fish and is corrected for the seasonal drop in average density due to migration into shallow waters during the period of upwelling.

We do therefore not consider our estimate of 345 thousand tonnes to be in contradiction with earlier estimates.

Absolute estimates from trawl surveys are dependent upon the gear's catchability coefficient (q), applied in the calculations. While we have used q=1 in our calculations, values in the range 0.5 - 1.0 can be found from other studies. The absolute estimates are inversely related to the size of q, and applying q=0.5 in our estimates will increase them to the double. In lack of detailed knowledge about the catch properties of the bottom trawl we have used a catchability coefficient which gives estimates that falls in the lower side of the range likely possible. In other words, our figures may underestimate the resources, but will then represent minimum assesments from which safe management actions can be taken.

3.3 The Catches of Demersal Fish.

3.3.1 Distribution of Abundances by Species Based on Catch Data.

Figure 11 shows the distribution of the catches grouped in approximate logaritmic increasing classes. The distribution is based on 106 random trawl stations.

The random hauls give an average catch of 784 kg/hour in a trawl with approximately 20m between the wings (for more details about gear, see Appendix II). Figure 11 shows that 39 % (41 hauls) fall between 300 and 1000 kg/hour and 80 % (85 hauls) between 100 and 3000 kg/hour. Only 3% (3 hauls) exceed 3 tonnes/hour. Catches from commercial trawlers with aimed trawling and concentrated effort in areas of high densities would likely give a distribution shifted to the right in Figure 11.

Figure 11. Catch distribution of demersal fish in the random bottom trawl hauls.

Tables 10-13 show the catch distribution and importance in the catches of the dominating demersal species in four regions: 1) North of Ras al Hadd 2) Ras al Hadd - Masirah Island, 3) Masirah Bank and 4) Sauquara and Kuria Muria Banks respectively. Table 14 shows a similar analysis for all hauls carried but in Oman. From our observations, the dominating species seem to be Nemipterus japonicus, closely followed by Argyrosomus hololepidotus, Arius thalassinus and Argyrops spinifer. The importance of Argyrosomus hololepidotus may have been overestimated as its rank in the table is based on one extremely high and incidental catch in very shallow waters (9200 kg/hour, st. 174).

Tables 10-14 include all species contributing down to 2 % each of the total catch. The remaining species, grouped as other fish are those contributing 1 % or less each to the total catch.

Table 10 Catch distribution by demersal species in bottom trawl. Region A, North of Ras al Hadd, all surveys. Random trawling. Total number of hauls: 17.

catch Species	Number of catches in catch groups					% of total	Mean
	1-9 kg/h	10-49 kg/h	50-199 kg/h	200-499 kg/h	>500 kg/h
(kg/hour)
Saurida tumbil	8	1	4	0	1	15	72.6
Upeneus sulphureus	0	2	1	0	1	13	59.2
Sphyraena barracuda	5	4	0	1	1	8	36.0
Nemipterus japonicus	3	2	3	1	0	6	29.5
Leiognathus fasciatus	0	3	0	1	0	5	24.6
Carangoides malabaricus	5	7	1	0	0	5	21.7
Argyrops spinifer	3	6	2	0	0	3	15.2
Gnathanodon speciosus	0	2	1	0	0	2	11.8
Sepia sp.	1	4	2	0	0	2	11.3
Trichiurus lepturus	5	3	1	0	0	2	9.9
Nemipterus sp.	0	1	1	0	0	2	9.3
Sharks	1	0	1	0	0	2	9.3
Carcharhinidae	1	5	0	0	0	2	8.6
Lutjanus malabaricus	0	3	1	0	0	2	8.4
Psettodes erumei	3	5	0	0	0	2	7.2

Table 11 Catch distribution by demersal species in bottom trawl. Region B, Ras al Hadd - Masira Island, all surveys.

Random trawling. Total number of hauls: 28

catch Species	Number of catches in catch groups					% of total	Mean
	1-9 kg/h	10-49 kg/h	50-199 kg/h	200-499 kg/h	>500 kg/h
(kg/hour)
Argyrops spinifer	0	5	6	5	3	28	220.0
Nemipterus japonicus	4	4	6	0	2	13	101.1
Pomadasys stridens	2	3	4	3	0	8	61.8
Saurida sp.	0	1	0	0	1	5	39.8
Arius sp.	1	1	1	0	1	4	35.5
Pagellus affinis	5	2	8	0	0	4	35.1
Arius thalassinus	3	3	0	0	1	4	32.0
Cheimerius nufar	3	5	3	1	0	3	27.0
Lethrinus elongatus	0	0	1	1	0	3	21.5
Trichiurus lepturus	1	2	2	1	0	3	20.4
Carangoides chrysophrys	2	3	2	1	0	2	16.9
Saurida undosquamis	3	4	3	0	0	2	16.6
Lethrinus nebulosus	1	3	3	0	0	2	12.9

Table 12 Catch distribution by demersal species in bottom trawl. Region C Masira Bank, all surveys.

Random trawling. Total number of hauls: 34

catch Species	Number of catches in catch groups					% of total	Mean
	1-9 kg/h	10-49 kg/h	50-199 kg/h	200-499 kg/h	>500 kg/h
(kg/hour)
Argyrosomus hololepidotus	0	0	1	0	1	29	275.8
Arius thalasinus	3	6	2	1	2	19	191.8
Pomadasys stridens	2	0	0	0	2	9	86.6
Nemipterus japonicus	5	3	1	1	1	9	82.2
Lethrinus nebulosus	2	2	3	0	1	4	37.8
Lepidotrigla betuviae	0	0	0	1	1	3	30.2
Cheimerius nufar	2	1	3	2	0	3	26.5
Rhinobathidae	0	0	1	0	1	2	20.6
Rays	1	0	0	2	0	2	17.9

Table 13 Catch distribution by demersal species in bottom trawl. Region D, Sauquara and Kuria Muria Banks, all surveys.

Random trawling. Total number of hauls: 39

catch Species	Number of catches in catch groups					% of total	Mean
	1-9 kg/h	10-49 kg/h	50-199 kg/h	200-499 kg/h	>500 kg/h
(kg/hour)
Nemipterus japonicus	1	2	2	3	1	16	99.6
Lethrinus nebulosus	3	2	9	7	1	14	90.7
Pagellus affinis	5	1	1	1	2	10	64.7
Cheimerius nufar	3	6	13	1	0	8	48.8
Lepidotrigla bentuviae	1	1	0	1	1	7	42.4
Charybdis edwardsi	0	0	0	0	1	6	35.7
Saurida sp.	2	1	0	0	1	5	29.6
Saurida undosquamis	0	1	2	1	1	4	25.9
Epinephelus sp.	4	4	3	1	0	3	18.0
Sepia sp.	3	0	4	0	0	2	13.4
Carangoides chrysophrys	4	7	3	0	0	2	11.4

Table 14 Catch distribution by demersal species in bottom trawl All Oman, all surveys. Random trawling.

Total number of hauls: 123

catch Species	Number of catches in catch groups					% of total	Mean
	1-9 kg/h	10-49 kg/h	50-199 kg/h	200-499 kg/h	>500 kg/h
(kg/hour)
Nemipterus japonicus	12	10	12	5	4	11	81.3
Argyrosomus hololepidotus	0	0	1	0	1	10	76.2
Arius thalasinus	15	21	6	1	4	9	69.9
Argyrops spinifer	14	20	14	6	3	8	61.0
Lethrinus nebulosus	7	10	16	8	2	6	45.8
Pomadasys stridens	6	6	5	3	2	5	39.3
Cheimerius nufar	8	13	20	5	0	4	33.6
Pagellus affinis	12	7	10	1	2	4	29.9
Lepidotrigla bentuviae	5	3	0	2	2	3	22.4
Saurida sp.	4	4	0	0	2	3	18.9
Saurida undosquamis	7	6	7	1	1	2	13.9
Saurida tumbil	12	3	6	0	1	2	11.9
Charybdis edwardsi	0	0	0	0	1	2	11.3

3.3.1 Distribution of Abundances by Species Based on Catch Data.

Distribution of the total estimated biomass of demersal species can tentatively be assessed by applying the relative distribution of the species in the catches (the % distribution in Table 14) to the total estimate. Table 15 shows these estimates.

Table 15. Distribution of total estimated demersal biomass by species.

Species	% in catch	Biomass thousand tonnes
Nemipterus	11	38
Argyrosomus hololepidotus	10	35
Arius thalassinus	9	31
Argyrops spinifer	8	28
Saurida	7	24
Lethrinus nebulosus	6	21
Pomadadys stridens	5	17
Cheimerius nufar	4	14
Pagellus affinis	4	14
Lepidotrigla bentuviae	3	10
Other fish, not more than '1% each	33	113
Total	100	345

3.4 Distribution of Catch and Biomass by Commercial Value.

In order to have an idea of how the demersal resources are distributed according to economic groupings, the catches from the bottom trawl have been distributed into four economic classes according to the species market value (see list below). As basis for this groupings, fish market prices in the United Arab Emirates, for the period May-June 1978, have been used (FAO 1981).

Division of the most common species into classes according to their market value. Based on fish market prices UAE May-June 1978 (From FAO 1981). Species in brackets have been added by similarity to the existing groups.

Class 1	(<5 Dirham/Kg): Megalaspis, Arius, Thunnus, Lutjanus sanguineus (Trachurus, sharks, rays, Ariomma)
Class 2	(5-9 Dirham/Kg): Sardinella, Plectorhynchus, Scolopsis, Epinephelus, Rhabdosargus, Rachycentrum, Scarus, Chirocentrus, Scomberoides, Sphyraena, Euthynnus, Chanos, Sciaenidae, Selar, Decapterus, Caranx, Argyrops, Pomadasys, Scomberomorus (Drepane, Atule, Mullidae)
Class 3	(10-15 Dirham/Kg): Apolectus, Carangoides, Sillaginidae, Acanthopagrus, Trachinotus, Sparidae, Elops, crabs, Seriola, Seriolina, Alepes, Pampus, Gnathanodon, Nemipterus, Mugil, Lefchrinus, Siganus, Rastrelliger
Class 4	>15 Dirham/Kg): Siganus, shrimp, Sepia, Loligo.

Table 16 shows the distribution of the total catch according to commercial value, according to the given classification. Two distributions are given, one for demersal species only, and one including the small pelagic fish.

Table 16 Classification in economical classes of catches from bottom trawling (% of total catch).

	Demersal fish only %	Small pelagic fish included %
Class 1 (<5 Dirham/Kg)	30	61
Class 2 (5-9 Dirham/Kg)	35	26
Class 3 (10-15 Dirham/Kg)	33	12
Class 4 (>15 Dirham/Kg)	2	0.8

By distributing the estimated total demersal biomass according to the commercial composition of the total catch, a rough absolute estimate of the quality composition of the demersal fish resources of Oman can be made. This is shown in Table 17. The most valuable class (4) is estimated to only 7 thousand tonnes, while the remaining three classes constitute roughly equal amounts of the remaining biomass, 104, 120 and 120 thousand tonnes in Class 1, 2 and 3 respectively.

Table 17 Classification in economical classes of estimated biomass based on species composition in the catches and their market value.

	Thousand tonnes
Class 1 (<5 Dirham/Kg)	104
Class 2 (5-9 Dirham/Kg)	120
Class 3 (10-15 Dirham/Kg)	114
Class 4 (>15 Dirham/Kg)	7
Total	345

3.5 Estimates of Yield of the Demersal Resources

The method of Beddington and Cooke applied in 2.3.3 to assess the yield of the small pelagic fishes, cannot easily be applied to the demersal stock. The demersal biomass is composed of a variety of species with different sizes, growth patterns, natural mortalities and ages at recruitment. Nevertheless, as the much used Gulland's formula (Y=0.5M B₀) seems to overestimate the yield in most instances, the influence of the growth and recruitment parameters should be taken into account, even though only rough general estimates of these parameters are available.

Vidal- Junemann (1981) classified the demersal fishes of Oman in three main groupings; a) small demersal fishes, including breams, porgies, grunts, goat fishes, lizardfishes, flat fishes and parrotfishes - all fish generally less than 50 cm maximum length, and b) larger demersal fish including scavengers, croakers, groupers, barracudas and sharks - fish with a maximum size generally exceeding 50 cm. For the small fish he suggested a natural mortality (M) within the range 0.4-0.8 and for the larger species M=0.2-0.4.

Pauly (1980) has listed growth parameters and natural mortalities for 175 fish species. Of these, 34 species are from waters with a higher temperature than 20 C and represent families which are common in Omani waters. Table 18 shows the range and the mean of the natural mortality (M) of these species grouped into two classes; a) less than 50cm maximum length, and b) more than 50 cm maximum length. The means of the M are 1.18 and 0.74 respectively, and this is considerably higher than the figures used by Vidal-Junemann. The lower limit used by him seems more to apply to fishes in temperate waters, while the upper limit seems more appropriate for perhaps a conservative estimate of M in tropical waters. We therefore suggest M=0.8 for the small fish and M=0.4 for the big fish as a first approximation of the natural mortality in Oman waters.

Table 18. Ranges and means of natural mortalities of selected fish species. (Raw data taken from Pauly 1980).

		No of species	Range of M	Mean of M
Big demersal fish >50cm max length
	Serranidae Lutjanidae Sciaenidae Lethrinidae	17	0.1 - 2.24	0.74
Small demersal fish <50cm max length
	Serranidae Nemipteridae Sciaenidae Mullidae	17	0.34 - 1.88	1.18

Cushing (in press) supports the critics to the equation Y=0.5MB₀ and agrees that it tends to overestimate the yield. He suggests the constant 0.5 be reduced to 0.4 for demersal fish and to 0.3 for small pelagics. In lack of the necessary information to use the Beddington and Cooke's method, we adopt Cushing's procedure and apply it to our calculations. The yield equation used on the demersal stock is therefore:

Y= 0.4 M*B₀

Table 19. Estimate of yield by species and by total biomass of demersal fish in Oman waters.

Species	Estimated biomass (1000 t)	Best guess of M	Equivalent estimated yield (1000 t)
Nemipterus	38	0.8	12
Argyrosomus	35	0.4	5.5
Arius	31	0.4	5
Argyrops	28	0.4	5
Saurida	24	0.4	4
Lethrinus	21	0.4	3.5
Pomadasys	17	0.8	5.5
Cheimerius	14	0.4	2
Pagellus	14	0.8	4.5
Lepidotrigla	10	0.8	3
Other fish (not more than 1% each species)	113	0.6	27
Total	345		77

As we have not been able to get information on the present level of fishing on the demersal stocks, the above calculations are made on the assumption that we are working on a biomass with a very low fishing mortality.