Eritrea, a country of some 125,000 km2, is situated in the north-eastern part of the African continent. Few of its 3,230,000 people live along the Red Sea (map, Fig. 1), although most live within 100 to 150 km of it as the crow flies. Half are highlanders living at 1500 metres or more above sea level in the rugged, cool massif, from which roads wind down to the coast on one side and to the warm, arable plains stretching to the Sudan on the other. Most of the other half of the population live on these plains, which are largely watered by seasonal rivers and streams. Another half million Eritreans are currently living in refugee camps in the Sudan. About 15% of the population in Eritrea live in towns, and the overwhelming majority of the remaining 85% are rural dwellers, deriving their livelihood primarily from subsistence farming, pastoralism, and fishing. Asmara, the largest city and economic and political capital of the country located in the central highlands, has a population of some 400,000 inhabitants.
After a protracted struggle extending over three decades, Eritreans are embarking upon a new era of independence. The Eritrean People's Liberation Front (EPLF) took over all the remaining areas of the country not already under its control by ousting the forces of the previous regime and established the Provisional Government of Eritrea (PGE) in May 1991. An internationally supervised referendum on formal independence from Ethiopia is to be held in late April, 1993.
Hardly any hard data exist to provide a basis for detailed characterisation and evaluation of Eritrea's current socio-economic condition. According to World Bank sources, national income accounts, trade figures, and aggregate price or wage statistics are lacking altogether. Banking statistics and revenue and expenditure accounts are partial and rudimentary, and in the absence of any census reliable population figures do not exist. Indeed, the lack of an adequate information on various aspects of national life is itself an indication of the generally disintegrated condition of national life following the long years of war. It is therefore necessary to review the overall situation in fairly indicative terms, using what data and observations are available through reports of PGE agencies and external donor reconnaissance and relief missions.
Before the liberation war period, the Eritrean commercial and subsistence agricultural sectors were extremely active and productive. A wide variety of cereal, vegetable, fruit, and other crops as well as livestock products fed both domestic and export markets. Trade in food products was once the country's foreign exchange earning mainstay. An industrial sector, based mostly on light manufacturing plants for processed food, textiles, and chemicals, also figured importantly in what was generally regarded as a modest but healthy, vigorous, and mostly self-sufficient economy.
Figure 1. Regions and main towns of Eritrea
a Adapted from: “Eritrea 1991 -- A needs assessment study, final report.” Centre for Development Studies, University of Leeds, August 1992.
The situation now is dramatically reversed. Eritrea's economy and infrastructure were critically damaged, and in some parts of the country almost totally destroyed, during the years of warfare and attendant population displacements, and by the acts of maladministration and mismanagement of the former regimes. Damage resulted not only directly from acts of shelling, bombing, and environmental sabotage, but indirectly as well in that the war diverted attention from productive activities and the task of maintaining basic services and infrastructure. Recurrent droughts exacerbated the overall trend of economic and environmental deterioration over this period.
Whereas in the past Eritrean subsistence farmers used to feed themselves, most have come to rely on relief food after years of war compounded by drought. The war itself ruined agriculture by dispersing the farming population, disrupting cropping cycles, reducing animal draught power, and destroying tools and implements. In 1991, another year of limited rainfall, a mere 70,000 mt -- only 10% of the national basic food bill -- was produced on 300,000 ha of the country's 3 million arable hectares. The 1991 output was mainly in the form of cereals and pulses. Oilseed, vegetable, and fruit production was very low.
Prospects for further diversifying production and raising the total food supply increased significantly in 1992 with the first heavy rains in many years. Between 30–40% of Eritrea's basic food needs may be covered by this year's crops. Some agro-industrial production is also being resurrected, currently for domestic consumption.
Although the 1992 rains were heavy, they were also late and caught farmers unprepared and without sufficient seed for planting. Furthermore, they came with such force that road damage was a major consequence, with ensuing disruption of trade and relief food deliveries. Although the harvest is a tremendous improvement over previous years, the situation is still precarious. The depleted livestock population will only be built back up very slowly. Replacement stock is not readily available from neighbouring countries, since their livestock herds have also been affected by war and drought. The lack of stock has profound affects on crop production capabilities because Eritrean farmers keep oxen for ploughing, though they also sell some cattle and smallstock to town residents in order to meet cash expenditures. Recovery in the short- to medium-term will depend heavily on the availability of fertilizers and other essential inputs, new breeding stock, the rehabilitation of water retention, irrigation, and soil conservation systems, and the restoration of effective extension and veterinery services.
Even with good rains, it is likely there will be some need for continued food aid and cereal imports in the years ahead and perhaps until the end of the century. World Bank-reported estimates put the drop in agricultural output at 40% through the 1980s, and the decline of livestock at more than 60% (from 3 million to 1 million animals). Widespread closures and breakdowns in the industrial sector add to the picture of a devastated economy. It is reckoned that those plants still open are operating at only 25 to 30% of their capacity, owing to shortages of energy and foreign exchange needed to buy spares and necessary inputs.
Various social indicators cited in World Bank and other background documents provide further testimony on the bleak situation Eritrea faces as it looks forward to the new era of independence. With a GDP per capita estimated in the range of USD 70 to 150 as compared to USD 330 for Sub-Saharan African countries overall, Eritrea must be ranked amongst the poorest countries in the world. Other indicators are summarised in the following table.
Table 1. Social indicators for Eritrea compared with Sub-Saharan African averages
| INDICATOR | ||||||
| Life expectancy (Years) | Infant mortality/ 1000 live births | Population/ physician | Population/ nurse | Adult illiteracy (%) | Daily calorie intake per capita | |
| ERITREA | 46 | 135 | 48,000 | 1750 | 80 | 1750 |
| SUB-SAHARAN AFRICAN AVG | 52 | 109 | 24,000 | 2450 | 58 | 2096 |
For 1991 it was estimated that over 80% of the Eritrean population was to a greater or lesser extent dependent upon food aid. This estimate fell to 72% for 1992 (WFP, pers.comm. 9/92). Food aid distribution was handled in former EPLF areas by ERRA (Eritrean Relief and Rehabilitation Agency), and the association now coordinates distribution throughout the country. Aid has largely been in the form of wheat and wheat flour, vegetable oil, dried milk, and sugar. This combination should provide basic energy and borderline protein requirements while agricultural capability is built up.
Faced with a devastated economy and infrastructure, a large population of dislocated and marginalised nationals, and severely weakened institutional capacities, the PGE has embarked on an Emergency Recovery and Rehabilitation Programme that aims to “jump-start” the economy through a series of wide-ranging but complementary measures intended to produce rapid restoration of essential economic activities and institutional capabilities. Priority is placed on the agricultural and industrial sectors, where productive capacities must be rebuilt through improved access to necessary inputs in the form of seeds, fertilizer, tools, oxen for draught power and farm animal breeding stock in general, and extension services. Water, soil conservation, and reforestation schemes also require rehabilitation and expansion, as well as the network of rural feeder roads. Substantial restoration efforts are also to be directed towards essential infrastructure, including energy supply systems, roads, bridges, and communications networks. Furthermore, the institutional framework for recovery and economic management, including the banking system, the treasury, and other Government departments, is to be throughly revitalised. Finally, a series of policy initiatives are being developed to facilitate the reconstruction and rehabilitation process, as reflected particularly in the proclamation of new investment, labour, and commercial codes. The overall thrust will be to move away from the restrictive policies of the previous regime by liberalising the commercial environment and encouraging growth of the private sector.
The Eritrean national fisheries, based almost exclusively on the exploitation of waters off the Red Sea shore and islands, can be expected to play a key role in facilitating the nation's social and economic recovery and rehabilitation. It is a role for which they are uniquely suited. Whereas the potential for mining and petroleum extraction is either rather limited or unproven, and production in the agricultural and industrial sectors will require a considerable amount of time to be fully restored, the stocks of Red Sea fish and other exploitable marine resources remained intact during the war years. Considering that the fisheries, like other sectors of the economy, suffered immensely from population dislocations and loss of productive capacity during the period of war and mismanagement under the previous regime, it may even be the case that these stocks have grown. Considering also that before such external factors precipitated their decline, the fisheries were quite productive and remunerative, the possibility for their full restoration in fairly short order appears all the more likely.
The gains that can be realised through fisheries redevelopment are considerable. Increased fish consumption is seen as one way to build up protein intakes. Also, to the extent that fish can serve as a substitute for meat, a more rapid build-up of herds will be possible. Furthermore, if the foreign exchange earning potential of the Red Sea fishery were to be properly tapped this could contribute indirectly to overall food security by assuring funds to import food when needed.
Such gains will be hard to realise, however, unless the right mix of policy and implementation modalities is brought into play. Part I and Part II of this document are intended to contribute to the identification of such a “right mix” by providing, respectively, a review of various aspects of the Eritrean fisheries sector (resource base, past and present performance, socio-economic characteristics, and institutional framework), and a proposal for a National Fisheries Strategy and Development Programme in the short-and medium-term.
Eritrea has a continental shelf of about 52,000 km2 in the 0–200m depth interval of the Red Sea. The total coastline runs to some 1,720 km, comprising 1,155 km along the continental shore and roughly 565 km around the numerous islands of the continental shelf. About 25% of the shelf is occupied by the Dahlak archipelago plateau, in the 0–35m depth interval. Its average width is variable, being widest (120 km) at the latitude of the Dahlaks themselves and quite narrow to the north (15 km) and south (30 km) of the archipelago (Guidicelli, 1984).
The Eritrean sea is generally recognized as being highly favorable for the development of artisanal fisheries. Important coralline areas, extensive surface and numerous shelters of the Dahlak archipelago plateau and, to the north and south of this plateau, short distances to the continental shelf break, are all conditions conducive to such development. Outside the coralline zones its sedimentary floors, while not particularly attractive for artisanal demersal fishing activities, offer quite favourable conditions for bottom trawling operations.
A sketch map showing past major trawling grounds and fishing areas within Eritrean territorial waters is presented as Figure 2.
Figure 2. Major Past Fishing Areas, Eritrean Red Sea Watersa
No comprehensive quantitative surveys of the resources and the productivity of the Eritrean Red Sea have been carried out, although various short-term investigations of specific stocks were mounted between the mid-1950s and the early 1980s (Sanders and Morgan, 1989). Most hypothetical figures put forward thus far have been based on catch rates, on some experimental fishing, and on extrapolation from other areas. Several attempts have been made to estimate the maximum sustainable yield (MSY) of the Eritrean marine fish resources. These estimates fluctuate between a high of 79,500 and a low of 36,000 tonnes per year, as indicated in Table 2.
Table 2. Estimations of MSY, Eritrean Marine Waters (in '000s mt)
| RESOURCES | MSYa | MSYb | MSYc |
| Demersal fish | 10.0 – 15.0 | 8.5 | 18.0 |
| Shrimp | 0.5 | 0.5 | 0.5 |
| Spiny lobster (whole) | 0.5 | - | 0.5 – 1.0 |
| Coastal pelagics | 50.0 | 25.0 | 25.0 – 50.0 |
| Neritic & oceanic pelagics | - | - | 5.0 |
| Sharks | 5.0 | 2.0 | 2.0 – 5.0 |
| TOTAL | 66.0 – 71.0 | 36.0 | 51.0 – 79.5 |
a Atkins (1956), Ben Yami (1964), Grofit (1971), Atkins and Grofit in Aubray (1975).
b Gaudet (1981).
c Guidicelli (1984)
Sardine (Harengula punctata) and anchovy (Thrissocles baleana and Amentum heteroboloum) resources have been estimated at about 50,000 tonnes per year (Grofit, 1971). In the past they were fished mainly with beach seines, so that only schools approaching close to shore at certain suitable spots were caught. Fishing effort remained concentrated close inshore because no purse seining for small pelagics has ever been introduced in these waters.2
Sardines and anchovies start to appear in the catches in late September or early October. The season ends around March/April but schools are said still to be spotted until May. Offshore schools never came into reach of the beach seines. Larger anchovy were never reported to reach north of Shuma island, so that the main center for this fishery remained the Assab area where many Massawa dhows sailed during the season (Grofit, 1971).
Schools of Scombridae, Indian mackerel, can be observed during most months of the year (Grofit, 1971). These fish may be considered to be a hardly touched resource in the Eritrean waters so far.
Tuna and tuna-like fishes (little tuna, longtail tuna and frigate mackerel) seem to visit inshore waters during winter months (October to May), following schools of sardine and anchovy, and offshore waters from March to June. Although more information on size of this resource is needed, it is considered likely to be quite substantial. At present, strange as it may seem, most tunas are discarded by artisanal fishers because the local market for these species is very poor (consumers seem to dislike red fish meat), and also because they supposedly require too much ice and work on board (bleeding head down) for the benefit derived from their catches. Tunas at present are categorized as “second quality” for the domestic market.
Losse (1975, as cited in Walczak, 1977) gave a rough estimate of 7,000 t/year for the tunas, mackerels and other non-coastal pelagics of Yemen. On this basis, a conservative figure of 5,000 t/year for the Eritrean waters would seem possible (Guidicelli, 1984).
If neritic and oceanic pelagics are little exploited today, it is not for lack of catching knowhow, but simply because there is no market for these products. The situation is the same for the small coastal pelagics. Formerly beach seining operations were producing more than 20,000 t/year of sardines and anchovies. Due to lack of outlet, the fishery for sardines and anchovies is now avoided except for small quantities caught for bait.
Prime reef fish
Handlining for prime fish takes place on rock and reef banks. When fishing has been carried out intensively for a few days running, grounds are often depleted and several years may pass before they are replenished. Under certain regulatory conditions (e.g. restrictions on hook sizes permitted on heavily exploited banks), Grofit (1971) estimated the potential of prime fish to be in the order of 3,500 t/year. Prime fish include following species and families of species:
Red snappers: Lutjanus gibbus, L. argentimaculatus, L. janthinuropterus, etc.;
Blood snapper: L. sanguinus;
Rock cods: Cephalopholis argus, Epinephelus tauvina, E. latifasciatus, etc.; and
Groupers: Epinephelus summama, E. aureolatus, Plectopomus maculatus.
Other demersals
Trawl fishing (demersal or bottom fish) was in the past carried out over about some 5,700 km2 (2,200 miles2) of the Eritrean Red Sea bed. Catches by trawlers during 1965–67 showed a composition of around 40% each of lizzardfishes (Saurida spp.) and threadfin breams (Nemipterus spp.), with the remaining 20% made up of barracuda (Sphyraena spp.), snappers (Lutjanus spp.), and groupers (Epinephelus spp.). Average commercial sized catches for 240 HP steel stern trawlers were about 2.2 to 3.3 tonnes per fishing day. Additional quantities of “trash” fish which amounted to between 30% and 50% of the gross hauls were discarded at sea. Grofit (1971) estimated potential resources of bottom fish in the Eritrean waters at around 10,000 tonnes per year.
In the case of the demersal fish it would seem that the total MSY of 8,000–15,000 t/year could represent a rather low estimate (Guidicelli, 1984). With a continental shelf surface of 52,500 km2 in the 0–200 m interval, these figures imply an MSY of 0.16 to 0.29 t/km2. This is far below Gulland's estimate of 1 t/km2/year for the groundfish crop potential of the Red Sea (as cited in Walczak, 1977), and also below Walczak's figures (1977) for the potential yield of demersal marketable fish in Yemen: 1.9 t/km2 in shallow waters and 1 t/km2 in deeper waters. It would thus not seem unreasonable to assume that the marketable demersal species of the Eritrean continental shelf could support an MSY of one third of Gulland's estimate (0.35 t/km2) and consequently offer a potential of some 18,000 t/year.
Only about 30 tonnes of shrimp where landed annually in the sixties. This quantity was originating from Zula Bay, Hergigo Bay, and from the Gurgusum grounds. The fishing season seems to coincide with the cooler months of the year. During the warmer season (June to September), shrimp become scarce. Previous surveys have shown that large quantities of shrimp exist in several areas along the Eritrean shore (Laor, 1969; Porat, 1970; Grofit, 1970). A potential of about 500 tonnes per year has been anticipated.
Spiny lobsters have been found, presumably in commercial quantities, at Difnei Island, Harwil island, Ras Kasar, Raheita, Barasole, and Assab Bay (Grofit, 1971). It has been estimated that by introducing lobster pots, at least 140 tonnes of lobster tails could be processed annually (Atkins, 1965). An estimated MSY of 500 t/year (whole lobster) does not seem exaggerated especially since the bulk of the stock is found within the 40–60 m depth interval.
The demersal resources of the continental slope, an area that covers 17,000 km2 (about one-third of the continental shelf) in the 200–550 m interval north of the Dahlak archipelago seem not to have been investigated. Exploratory surveys and commercial fishing efforts (Sanders and Morgan, 1989) have demonstrated that valuable deepwater shrimp of various species and lobster resources were present in the Gulf of Aden, including Djibouti waters. In the long term, the Eritrean continental slope could quite possibly prove an important focus of exploitation.
Trochus shells, used for the manufacture of “pearl-buttons” were landed at a rate of around 220–250 tonnes annually in the late sixties, mostly at Massawa. “Snail nails” (opercula of certain conches) have also been heavily exploited in the past around the Massawa area and in the Dahlak archipelago. Even today, the exploitation of this resource is carried out at low tide by many foot-fishers, especially women and children. After the “nails” are extracted, the conches are often returned to the water as it is believed that the opercula can be regenerated.
Ornamental fishes and corals have been exploited near the islands off Massawa (Madote, Disei, Duhr-Gaham, etc.) and in the Hergigo Bay. The most easily accessible localities for Massawa-based operators, but such fishes are to be found around practically every island and reef. Most sought-after fish belong to the families of Pomacanthidae, Platacidae (angels), Ostraciontidae (boxfishes), Chaetodontidae (butterflyfishes), and Scaridae (parrotfishes).
The Eritrean marine fisheries of thirty to forty years ago have been described in some detail by earlier observers. The fisheries of the past can generally be characterised as very active. Catches of well over 25,000 tonnes per year were reported in 1954, prior to the withdrawal of Yemeni fishing units (FAO, 1983). Over 80% of this production consisted of small coastal pelagics -- sardines and anchovy -- which were processed into fishmeal or sun-dried in Massawa for export to European and Far East markets (Aubray, 1975). Reported harvests were in the order of 19,600 mt for fishmeal, 1,250 mt for processed sharks, 150 mt for shells, and 1,300 mt for foodfishes. Fisheries in those days were essentially oriented towards exports.
Table 3. Evolution of marine resource landings in Eritrea from 1954/55 to 1983/84. (in '000 mt)a
| Species groups | 1954/55 | 1962/63 | 1963/64 | 1964/65 | 1965/66 | 1966/67 | 1967/68 | 1968/69 | 1969/70 |
| Pelagic | 25.0 | 6.7* | 7.0* | 16.3* | 19.7 | 17.9 | 6.8 | 9.0 | 13.0 |
| Demersal | n/a | 1.6 | 1.3 | 1.2 | 1.3 | 1.2 | 1.2 | 1.5 | 0.1 |
| Sharks | n/a | n/a | n/a | n/a | 1.1 | 1.3 | 5.5 | 1.9 | 1.5 |
| Lobster/Shrimps | n/a | n/a | n/a | n/a | 0.01 | - | - | 0.1 | 0.08 |
| Shells | 0.3 | 0.2 | 0.1 | 0.2 | 0.15 | 0.2 | 0.2 | 1.2 | 0.4 |
| Totals | 25.3 | 8.5 | 8.4 | 17.7 | 22.3 | 20.6 | 13.7 | 13.7 | 15.0 |
| Species groups | 1970/71 | 1971/72 | 1972/73 | 1973/74 | 1974/75 | 1975/76 | 1976/77 | 1977/78 | 1983/84 |
| Pelagic | 19.0 | 11.0 | 2.0 | 3.5 | 2.0 | 0.25 | - | - | n/a |
| Demersal | 1.2 | 1.2 | 1.0 | 0.6 | 0.4 | 0.6 | 0.15 | 0.02 | n/a |
| Sharks | 2.3 | 1.1 | 0.4 | 0.5 | 0.03 | 0.1 | 0.01 | - | n/a |
| Lobster/Shrimps | 0.03 | 0.03 | - | - | - | 0.02 | - | - | n/a |
| Shells | 0.4 | 0.5 | 0.3 | 0.3 | - | - | 0.01 | - | n/a |
| Totals | 22.9 | 13.8 | 3.7 | 4.9 | 2.4 | 0.97 | 0.17 | 0.02 | 2.0 |
Production, standing at 21,000 mt in 1966, dropped to 14,000 mt in 1967 (Table 3). This was due not to any decline in the productivity of the fishing grounds but because of the closure of the Suez Canal and consequently of the major cheap fishmeal export route. Fish exports in 1966/67 only amounted to some 5700 tonnes of processed products (Table 4). By 1972, growing internal warfare and subsequent instability provoked a further decrease in fishing activity and landings fell to 4,000 mt. This tendency continued during the following years. Numerous fishing craft were destroyed and, without outlet for their production, local fishers increasingly turned to other activities or other countries. As a result, by the end of the seventies, the fisheries had almost completely collapsed. According to various sources, total production only amounted to about 328 mt in 1980.
Table 4. Mid-1960s export level of Eritrean marine commoditiesa
| Commodity | Quantity (t) | Values (E.$) |
| Fishmeal | 1 430 | 459 585 |
| Dry fish | 3 422 | 2 080 408 |
| Fresh fish | 702 | 237 073 |
| Shell products | 213 | 125 286 |
| Other products | 37 829 | |
| Total | 5 767 | 2 940 181 |
Some improvement in the situation was noted in 1984, when it was estimated that production from the Massawa and Assab areas was about 2000 mt. Recovery was slight and short-lived. The disruption of further warfare coupled with the absence of any significant and functional marketing network which could purchase catches at reasonable prices and transport and distribute them in the densely populated central highlands, brought ever heavier pressures on local fishers to export informally to the other side of the Red Sea, into countries where active markets existed and where all necessary supplies and services were available.
The composition of the local fishing fleet changed rather markedly during the decade of the 1960s (Table 5), with an increase in the number of dhows and the level of motorisation, and a decrease in the number of canoes. Prior to 1962, there where only very few motorised dhows in use. The first planned attempts towards the introduction of marine engines were made in 1963.
In 1970, the artisanal fishing fleet strength was estimated at some 500 houris (6 to 15 m open long boats), 70 of which were equipped with outboard engines, and roughly 300 dhows (8 to 17 m open or decked boats), 80 of which had 10–45 HP inboard engines (Aubray, 1975). By 1981, however, the fleet size was assessed at only some 130 craft, less than half of which were operational. In 1984, the situation seemed to have improved a little in the Assab area (Dankil region) where some 56 operational and well-maintained craft were counted (Guidicelli, 1984).
Table 5. Level of marine fishing craft in the 1960s
| Type/year | 1962/63 | 1966/67 |
| canoes | 433 | 174 |
| canoes+outboard | - | 78 |
| sailing dhows | 184 | 275 |
| motorized dhows | 21 | 52 |
| stern trawlers | 9 | 11 |
| fish carriers | 3 | 2 |
| unregistered rowing boats | n/a | estim. 214 |
The former artisanal fishery was rather simple, but financially and economically efficient. In spite of some structural weaknesses (due more to the frequent lack of good building material than any absence of local carpentry skills), the houris and dhows were seaworthy and appropriate to the marine and human context. Fishing techniques, essentialy beach seining and hand-lining and, to a lesser extent, troll-lining, shark gillnetting, and long-lining, proved to be efficient, adequate, and well-managed by the operators.
From a level of about 23,000 in the fifties, the size of the local fisher force dropped steadily with the decrease of activity. During the sixties, Grofit estimated the number of fishers to be around 7,000 (Table 6). By 1970 the force was estimated to have declined to 5000 individuals, 80% of whom were part-time operators (Aubray, 1975). In 1981 their number was estimated at some 3,500 mostly occasional operators. In 1984, in the Assab area, some 250 full-time professionals and active fishers were observed (Guidicelli, 1984).
Table 6. Number of Eritrean coastal fishers in the 1960s.a
| Type/year | 1962/63 | 1966/67 | 1967/68 |
| Full-time | 1 270 | 2 195 | 1 456 |
| Part-time | 3 295* | 2 369 | 1 769 |
| Occasional | 2 500* | 3 150 | 2 726 |
| Total | 7 065 | 7 714 | 5 951 |
a Source: Grofit (1971).
* = estimated number
Observers calculated that well-operated fishing boats in the 1960s were profitable enterprises yielding annual incomes per fisher ranging between USD 150 to 350, according to the type of fishery (Grofit, 1971). In 1984 the artisanal fishery was still a financially efficient operation even though domestic (then Ethiopian) markets were not strong. To obtain a sufficient financial return, fisherfolk had to commercialise nearly all their production in Yemen (Guidicelli, 1984).
The industrial fishing fleet strength in the sixties involved up to 4 inshore trawlers (50–120 HP), 9 off-shore trawlers (150–400 HP) and about 3 handliners (Ben Yami, 1975; Aubray, 1975). The trawler fleet was mostly commanded by expatriates and crewed by Eritrean deck-hands. Its financial productivity was satisfactory for the well-operated units. For several years, the average catch of a properly run 240 HP trawler, using Mediterranean-type gear, was between 200 and 280 kgs of commercial fish per trawling hour. Complementary catches of penaeid shrimp were also obtained in amounts averaging from 20 to 30 kg/trawling hour during the good shrimping season, approximately from November to April. Today this fishery has totally disappeared. It remains the case, however, that about 35,000 km2 of good trawling grounds are well known, with a probability that other trawlable areas exist (Guidicelli, 1984).
According to Ben Yami (1964), about six fishmeal plants were working in Massawa during the sixties. These factories processed sardine and anchovy captured and sun-dried by about 2,000 -3,000 beach seining fishers from October to April, all along the coast from Massawa to Assab, and transported by small 80–200 HP carriers. Average annual production of this activity was of 1,000 t of fishmeal for European markets and about 3,000 t of dried products for the Far East. In the Massawa port, an enterprise commercialised first quality fish on the national market, transporting an average of 200 t/year by truck to Asmara and Addis Abeba. Two firms operating with five trawlers were also exporting about 1,500 t/year of frozen table fish to Israel, and one enterprise traded shells abroad.
All the fish companies possessed their own facilities for ice, processing, freezing, storage and transport. In those years Massawa had a jetty, two boatyards, and a slipway. Assab had a modern refrigeration storehouse totaling 1,300 m3 at -18 to -20 ° C. Today all these activities have collapsed and nearly all the facilities have disappeared or have been destroyed.
Results of the 1992 Marine Frame Survey (FS)3 (Bellemans and Reynolds, 1992) and field observations made in the course of Project TCP/ERT/2251 confirm that the present situation of the Eritrean Red Sea fishery has drastically deteriorated from what it was some decades back. Catches are exceedingly low, the fleet largely dispersed or in disrepair, the fishing population substantially dislocated or marginalised, and supporting facilities and services mostly in a state of collapse.
The Eritrean marine artisanal and industrial fisheries that were flourishing some three to four decades back produced relatively high tonnages, on the order of of 20 – 26,000 t/year principally comprising sardine and anchovy (ca. 80%), demersal fish (15%), and shark (5%). These fisheries have now all but disappeared owing to events of warfare and attendant dispersal and disintegration of the national fisheries workforce and fleet.
Contemporary Eritrean marine fisheries are almost entirely of an artisanal nature, consisting mostly of hand-lining or gillnetting operations using small inboard or outboard propelled dhow-like vessels. A considerable amount of “foot-fishing” -- angling and collecting along the tidelines and shallow reefs -- continues to be prosecuted, though at a much reduced level in comparison to the past.
Limited trawling operations for demersal species were recommenced on a provisional license basis in late 1992 around the Massawa area, with one foreign company running two units (ca. 30m and 20m) under Government supervision. Results of these operations are not known in detail, but it appears that the immediate consequence is a plentiful supply of fish for the local markets in Massawa and Asmara.
Most of the artisanal catch of fresh fish now marketed domestically is landed at Massawa; far smaller quantities are recorded at the southern port of Assab. Statistical records of landings from the larger artisanal craft at Massawa and Assab fish receiving stations started only in January 1992, so the time series data available is of very shallow depth. Based on available returns, it can be projected that annual landings of prime food fish are running close to 450 tonnes at Massawa, and about 200 tonnes at Assab (figures in terms of rough whole weight equivalents of gutted weights recorded). A considerable amount of dried marine products including especially salt-dried shark meat, shark fins, mullet, trepang, and snail nails are also collected through the artisanal fishery and mostly are channelled into the formal and informal export markets. The unclear manner in which statistical information on all such products is assembled does not allow a very exact picture of what quantities are involved. An FAO Investment Centre mission that visited some of the fishing areas in December 1991 estimated actual (as distinct from officially recorded) total production at around 1950 t/yr in and around the Massawa area (FAO/UNCDF, 1992).
Table 7. Recorded landings at Assab and Massawa, 1992 (kgs gutted fish)
| MONTH | ASSAB | MASSAWA | TOTAL |
| January | 11 310 | 34 748 | 46 058 |
| February | 11 856 | 44 631 | 56 487 |
| March | 11 106 | 44 632 | 55 738 |
| April | 6 311 | 26 240 | 32 551 |
| May | 5 069 | 25 603 | 30 672 |
| June | 7 824 | 16 662 | 24 486 |
| July | 6 781 | 28 603 | 35 384 |
| August | 9 292 | 21 403 | 30 695 |
| September | ? | 31 798 | 31 798 (*) |
| October | ? | 25 448 | 25 448 (*) |
| November | ? | 31 079 | 31 079 (*) |
| TOTAL | 69 549 (*) | 330 847 | 400 396 (*) |
Note: (*) = Incomplete information.
Current major species caught
The species groups represented in current catches at the 57 coastal and island sites covered by the 1992 FS are shown in Table 8, broken down by region. It can be seen that globally, shark and mullet are the most commonly recorded fish, with snappers, groupers, reef-fishes, mullets, and jack-fishes figuring to a far lesser but still notable extent. Sardines are recorded as being commonly caught in only 3 villages, although direct observation in the field indicates that they are widely captured for baiting purposes.
Shark meat is usually salt-dried, and shark fins and mullet sun-dried on islands or beaches adjacent to fishing grounds. Whenever possible prime reeffishes are gutted and stored in ice boxes for sale at Massawa. Formerly such “refrigerator” fishing was also practised out of Assab, but machinery failure has choked ice supply for the last year or so.
The collection of “snail nails” (the opercula of certain conches) and trepang (sea cucumber) in the Dahlak and Dankalia regions clearly constitutes an important commercial activity. Both products are harvested exclusively for sale or trade, so there is no subsistence (auto-consumption) dimension to such collection.
In the Dahlak region, it was reported that catches of turtles occurred at several sites. From what could be learned through interviews, these catches are opportunistic rather than systematic, and are more of a subsistence-oriented than commercial character. The collection of coral, as registered in the Dahalks, is apparently done mainly for domestic construction purposes.
It should be pointed out that some of the species noted in Table 8 are not necessarily “target” fish as such. The fact they are being caught at all may have more to do with gear availability than the intentions or wishes of artisanal operators. If substantial improvements in gear availability were to occur, the “targets” of fishing effort could very possibly change quite markedly.
Table 8. Major species caught, Eritrean marine fisheriesa
| REGION | SEMHAR | AKELE GUZAI | DANKALIA | GLOBAL | |
| SPECIES | COAST | DAHLAK | |||
| SARDINES | - | 1 | 1 | 1 | 3 |
| SNAPPERS | 1 | 11 | 1 | 6 | 19 |
| GROUPERS | - | 11 | 1 | 4 | 16 |
| SHARKS | 1 | 11 | 1 | 33 | 46 |
| REEFFISHES | - | 1 | 1 | 3 | 5 |
| MULLETS | 1 | 13 | 2 | 34 | 50 |
| MACKERELS | 2 | 7 | 1 | 8 | 18 |
| JACKFISHES | 1 | 10 | 3 | 7 | 21 |
| TUNAS | - | 3 | 2 | 4 | 9 |
| SCAVENGERS | 1 | 4 | - | 4 | 9 |
| BARRACUDAS | - | - | - | - | - |
| LOBSTERS | - | - | - | - | - |
| SHELLS | - | 1 | - | 1 | 2 |
| SNAIL NAILS | 1 | 11 | - | 20 | 32 |
| TREPANG | 1 | 11 | - | 15 | 27 |
| CORALS | - | 2 | - | - | 2 |
| TURTLES | - | 5 | - | 4 | 9 |
| ORNAMENTAL | - | - | - | - | - |
a Figures indicate the number of villages reporting the species-group.
Producer prices
Current Government policy is to keep the price of fresh fish low enough to make it accessible to poorer consumers who cannot afford meat, and prices accordingly are controlled by DMRIF particularly at the two major domestic landing sites of Massawa and Assab. Off-loaded fish is graded into two rather arbitrary commercial categories: Grade I fish include barracudas, snappers, seabreams, groupers, rock cods, scavengers, emperors, seabasses, etc.; Grade II fish comprise kingfishes, jacks, mackerels, tunas, and fish from the first grade of poorer quality or not gutted. Iced shark meat is now also included in the second grade. Before October 1992 tuna and shark were categorised as “third grade” fish. Regulated ex-vessel prices for 1992 varied somewhat but generally were about Br 2.00/kg for Grade I fish, Br 1.75/kg for Grade II fish, and Br 1.35 for grade III fish. At the beginning of December producer and retail prices were both adjusted upward. Fishers now receive Br 3.75/kg for Grade I fish, and 3.25/kg for Grade II.
Prices for dried or salt-dried processed commodities including shark meat and sharkfin, mullet, jackfish, trepang (sea cucumber), and snail nails (opercula of certain conches) have also been fixed within certain limits by DMRIF. Fixed unit prices differ somewhat according to the location (vide Section 4.2), but all tend to fluctuate within certain limits according to the Yemen market. However, the underlying adjustment mechanisms are difficult to understand on the basis of information available.
According to mid-1992 official figures for the Massawa-Dahlak area, producer prices for snail nails are the highest (450.00 Br/kg in Massawa/Dahlak for the small quality). White-tipped sharkfin was selling for 80.00 Br/kg, and large and small regular sharkfin for 46.50 and 18.00 Br/kg respectively. Sea cucumber was fetching 18 Br/kg, small shark meat 3.30, large shark meat 1.00, large mullet 3.00, medium mullet 2.00, small mullet 1.00, and jackfish 0.50.
Fleet characteristics
A total of 636 fishing craft were recorded in the baseline survey, though nearly one-third of this number is made up of non-active (out of service) units. Another rather disturbing finding is that the fleet as a whole is quite aged -- 8 years old as a global average. This suggests that little new investment or reinvestment has taken place for a considerable period of time.
Three major types of fishing units were distinguished during the survey, namely: (i) canoes, or non-motorised, double-ended smallcraft; (ii) houris, or mid-sized units generally with open decks and fitted with outboard engines; and (iii) sambuks, or larger units generally decked and equipped with inboard diesel engines. No clear definitions of fishing craft, i.e. definitions based on the structural characteristics of boats rather than on the presence or the absence of an engine, have yet been introduced by the fisheries administration. This made the correct identification of fishing craft less reliable, especially for those craft which could not actually be observed by the field teams.
Figures on fleet composition indicate that most craft are of the houri type. The smaller canoes and larger sambuks account for 24% and 10% of the active Eritrean national fleet respectively. The level of motorisation corresponds to the craft proportions, since only the houris and sambuks are powered by engine.
Figure 3 depicts the distribution of active fishing units by region according to type. It appears that over three-quarters of all active canoes are concentrated in the Semhar-Dahlak region. For houris, some 60% are to be found along the Dankalia coast, 33% in Semhar-Dahlak, and most of the remainder in Dankalia islands. The distribution of sambuks is about 50% along the Dankalia coast and 43% in Semhar-Dahlak, with the rest scattered between the Semhar coast and dankalia islands.
The regions of Akele Guzai and Semhar-coast are relatively poor in fishing craft, and most of what is found comprises smaller non-motorised canoes.
It is obvious from length frequency data that the population of canoes can rather easily be separated from the two other types of craft (Table A1, Annex 1). The houris and the sambuks show a much wider range of overlap -- perhaps a result of basing definitions on engine type (inboard/outboard) rather than on the structural characteristics of the craft themselves. Some “houris” (in the sense of mid-sized open-decked boats with narrower beams and square sterns) with inboard engines may thus become categorized as “sambuks” while “sambuks” (in the sense of larger double-ended dhows with decks and wide beams) equipped with outboard auxiliary engines may be categorized as “houris”. The fisheries administration should move to sort out such confusion so that standardised usages apply. It is suggested that the technical distinctions noted here be adopted as standard for future surveys as well as routine statistical collection activities.
Craft purchase and replacement costs
Information on average purchase and replacement costs for the various length classes of fishing craft, as reported by local fishers, indicates that canoes and houris of overlapping length classes have quite different values attributed to them (Table A2). This is to be expected given their fundamentally different structural characteristics. A similar observation can be made for the houris and the sambuks of comparable length classes. The influence of wood types used for craft construction may also be considered an important factor in cost reckoning.
The most expensive craft are the sambuks of around 28 m. Their purchase price is put at about 82,000 E. Birr at a baseline of some 13 years ago, whereas their current replacement costs are estimated to be around an average of 142,000 E. Birr. In general it would appear from the data that present estimated average replacement costs for various fishing craft have almost doubled from their original purchase prices effective some 6 to 13 years ago. It is difficult to determine the real impact of price changes because the extent to which inflation/devaluation factors have to be taken into account is not readily known. Also, as already noted, the average replacement costs cited represent only estimates given by the fishers. Actual costs could be even higher, especially since most of the craft are manufactured outside the country.
As indicated by FS results, the vast majority of Eritrean-based fishing craft, about 85%, reportedly have been purchased from across the Red Sea, in Saudi Arabia (60%) and Yemen (25%). Reports of local acquisition of craft in Eritrea (the remaining 15%) may well refer to second-hand purchases in many cases. The fact that nearly all of the larger craft are purchased abroad (93% for sambuks and 88% for houris) is quite indicative of the present poor state of boat construction capability in Eritrea. Whether this situation results from lack of materials or expertise or a combination of the two is a question that bears further investigation.
Equipment and gear kit
A breakdown (Table A3) of FS data on average horsepower for motorised fishing craft by type of engine and region shows that the global average size for outboard engines, the common form of houri propulsion, is 33 HP. For those sambuks equipped with outboards (mostly as standby engines), the global average is 54 HP. For inboard engines, found only on sambuks, the global average is around 54 HP. The motorised category by definition does not include canoes.
Reconfiguration of these data in terms of distribution of size classes according to the engine type (inboard/outboard) for both the sambuks and houris confirms that the majority of outboards on the houris have a power range of between 31 and 40 HP (Table A4). All outboards on sambuks are of a much larger size, ranging from 40 to 80 HP.
As in the case of similar records for fishing craft, anticipated replacement costs for engines as reported by FS respondents have probably been estimated on the low side (Table A5). It is likely that many respondents, without firm knowledge of current shelf-prices for items that cannot be purchased locally, can only offer rough guesses for replacement costs. Local purchase of engines is in fact quite rare (Table A6). Most outboard engines have been bought in Saudi Arabia (62%) and Yemen (31%). For the few inboard engines on which complete information was collected, it seems that the only place of purchase was Saudi Arabia.
With regard to the age of engines powering the Eritrean fishing fleet, outboards fitted on houris average 6 years and those carried as auxiliaries on sambuks 10 years (Table A7). For inboards (sambuks only) the average is 9 years. It is apparent that some very old engines of more than 15 years age are still reported to be part of the marine fleet's equipment inventory (Table A8). The bulk of the outboard engines seem to cluster around an age class of 7 to 8 years, with a second modal point of 4 years. On the whole there would appear to be cause for concern that the national inventory of marine outboards -- engines that cannot be expected to have long working lives under the difficult operating and maintenance conditions of local artisanal use -- may be tending towards obsolescence.
As for the diesel inboards that power the sambuk fleet, the majority are reported as belonging in the 7 to 10 year age classes. Diesel inboards can be expected to have a longer working life than benzine outboards even under the less than ideal service and maintenance conditions that currently prevail in the Eritrean marine fisheries sector. The inboard situation may thus not be quite so critical as with the outboards. Nevertheless, just as with the fishing craft situation reviewed above, the age frequency distribution for mechanical components of the national fleet reveals inherent weaknesses with regard to the absolutely vital processes of reinvestment and renewal.
The gillnet is by far the most commonly found fishing gear within the national marine fishery. Gillnets are used by fully 74% of all units in the sector, with some variation showing from region to region. Hand-lines (hook-and-line) are the second most frequent type of gear in use, being recorded for around 19% of all marine units. Incidence of use of other gear like beach seines, castnets, stake nets, traps, spears, and simple diving apparatus is apparently quite low, though it is likely that the gear count for foot-fishers was under-recorded by the survey teams.
Average purchase and replacement costs for major fishing gear gathered through the 1992 FS should be regarded as very indicative only since relatively little gear can be found on the local markets (Table A9). The foreign provenance of the national gear kit is clearly apparent (Table A10).
A total of 1414 full-time embarking (boat-using or craft-based) fishers were recorded during the 1992 FS. In addition, 94 “part-time” (10–15 days/month) and 44 “occasional” (<10 days/month) embarking fishers were registered, along with 1,063 “foot-fishers” (land-based collectors, anglers, divers, net and spear users).
A breakdown of the total fisher population by region (Fig. 4) indicates that in Semhar around 64% are engaged aboard fishing craft (63% on a full-time basis and 1% on a part-time basis), and about 35% are shore-based operators. In Akele Guzai some 32% of the fishing workforce are craft-based and about 68% shore-based. For Dankalia the data indicate that 57% of all fishers are engaged on the water (50% full-time, 5% part-time, and 2% occasional). About 43% of the Dankalia workforce operates along the shore as foot-fishers.
Additional survey findings indicate that some 57% of the full-time embarking fishers can be found in the Dankalia region, while Semhar accounts for around 43% of their number. About 81% of the parttime and 89% of the occasional workforce can be found in the Dankalia coast region.
Table 9. Distribution of foot-fishers
| REGION | FULL-TIME | PART-TIME | OCCASIONAL | ||||||
| MEN | WOMEN | CHILD. | MEN | WOMEN | CHILD. | MEN | WOMEN | CHILD. | |
| SEMHAR-COAST | - | 10 | 24 | - | - | - | - | 100 | 100 |
| SEMHAR-DAHLAK | 12 | 49 | 53 | 4 | - | - | - | - | - |
| AKELE GUZAI | 11 | - | - | - | - | 4 | - | - | - |
| DANKALIA-COAST | 26 | 177 | 277 | 103 | 23 | 20 | - | - | - |
| DANKALIA-ISL. | 20 | 27 | 23 | - | - | - | - | - | - |
| TOTALS | 69 | 263 | 377 | 107 | 23 | 24 | 0 | 100 | 100 |
| 709 | 154 | 200 | |||||||
Data on various characteristics of the marine foot-fisher population are arrayed in Table 9, including its distribution by region, extent of participation in work, gender, and age. What is particularly striking is the high level of participation of women and children (≈ 15 years or younger) in the foot-fishing workforce, 36% and 47% respectively. Since embarking fishers are almost exclusively adult men, these figures can be recalculated in terms of the global picture to show that approximately 15% of the total Eritrean marine fisheries workforce is now comprised of women, and 19% of children. Taken together, this represents over one-third of the entire fisherfolk population. A regional breakdown of the footfisher population by gender is presented in Figure 5.
Except for Ras Dumeira, just north of the border with Djibouti, all the villages surveyed are occupied on a continuous basis.
Infrastructural features of the coastal settlements are almost uniformly poor (Table A11). Although most villages are served by a water supply (usually wells) lying within a radius of 3 kms (the distance that defined convenient water availability for survey purposes), the incidence of any other form of service or amenity is low to nil. Primary schools are found in only 33% and some form of health care facility (health centre or hospital) in only 21% of the sites. Other items register hardly or not at all. Only the large towns of Massawa and Assab offer anything approaching the full complement of services required for equipping and provisioning of local fishing operations. Yet even in the major centres vital inputs may be hard to come by. In Massawa only a few small shops could be found which offered at least some fishing inputs for sale. Their stocks were very minimal, not even including twines for net repair. It was said that these latter were only available through Dahlak fishers.
Field visits and discussions in local communities made it clear that wood, spare parts, nets and other fishing equipment and supplies (fuel and ice reservoirs, engines, etc.) appear to be regularly purchased and imported from Yemen and Saudi Arabia. Dahlak and Dankalia operators traditionally have strong commercial relations with neighboring countries and there is little doubt that they export a part of their production to outside markets in order to pay for basic inputs presently lacking along the Eritrean Red Sea. It is also likely that some fishers live and operate for extended periods of time in the Yemen before returning to their native communities.
Fishing operators interviewed at the Massawa landing centre and in the Dahlak Islands defined two major fishing seasons, namely: the winter season, extending from October/November to April, and the hot season, extending from May to September. These seasons seem to affect the artisanal fishery's activity to a large extent. During the hot season, fishing grounds lie further away, fish resources remain in deeper waters, changes in fishing techniques are required, and trips must be of longer duration. During the “good” cool fishing season, fishing takes place around banks and reefs, 5–20 fathoms deep. During the “bad” hot season, fishing will generally take place at depths of around 50–80 fathoms mostly towards the south.
Fishing activity of one form or another is pursued the whole year round. As far as seasonal migration patterns are concerned, no major movements are registered for the Semhar and Akele Guzai regions. In Dankalia some migration is recorded for several of the fishing sites lying towards the south, involving two principal routes: one towards the fishing grounds located around the Dahlak Archipelago and the other towards Assab. Migrations reportedly take place mainly during the cooler season that runs from October/November to April, as operators follow fish movements. In a few cases it was noted that operators prefer working outside their home grounds at this time because of rough weather conditions.
Rather surprisingly, not a single village covered by the FS teams reported regular visits by fish traders. Apparently in the vast majority of cases fishers have to make their own arrangements for channelling their production into trading networks. Fifty-one or 89% of the fishing sites reported no direct or regular access to marketing outlets for the products. Only in a few places, such as in or around the large urban areas of Massawa and Assab where there is a ready demand for fresh fish, or very small localities where production activity is minimal, does it seem that no special catch-disposal problems arise.
Out of the 51 fishing sites where operators must directly seek outside markets for their products, reliance on boat transport in nearly all (47) cases. In a few instances fishers carry out products on foot or by camel.
With regard to the main destinations for marine products, the villages of the Semhar coast and Dahlak Islands report an orientation mainly towards the Massawa market, as could be expected. In the Akele Guzai region, it appears that fisherfolk do not market their catches outside of their local areas. In the northern reaches of Dankalia, from Ganferor to Akelo, villagers bring their products mainly to Massawa or Harena. Further south, the flow of products is either towards Tiyo or Assab.
The reported frequency of visits to market outlets is broken down in Table 20. It appears that at the majority of sites (57%), trips are made once per month.
Table 10. Frequency of market disposal of products
| Frequency of disposal | # of villages |
| less than 1 ×/week | 5 |
| at least 1 ×/week | 6 |
| at least 1 ×/2 weeks | 3 |
| at least 1 ×/month | 29 |
| no set routine/depends on catches | 8 |
| Total | 51 |
In fine, there seems to be a major constraint on the disposal of marine catches that are in excess of local consumption needs. It would thus not be unreasonable to assume that at certain fishing centres at least production is deliberately kept under its potential level because of marketing difficulties. Catch disposal for prime species is also made more problematical by lack of ice supplies. Ice is only available on a regular basis at the port of Massawa, and even here the existing plant, operating with antiquated equipment and with its capacity severely restricted because of the extensive damage incurred by shelling and bombing during 1990, is totally unable to meet demand. Assab ice making facilities are in serious need of repair and there has been no production at all for over a year.
It should be re-emphasised here that a substantial quantity of processed commodities, especially dried shark (meat and fins), mullets, and jacks, find their way through long established trading channels across the Red Sea to Saudi Arabia and Yemen. This fact is well known both on the basis of informal interviews at fishing camps and villages up and down the coast, and the indirect evidence that shows these countries to be major sources of fishing inputs and other supplies. It is hardly surprising that villagers did not lay particular stress on such information in the context of formal FS interviews, nor were the survey teams asked to probe in depth when touching upon this subject.
DMRIF has registered 121 mechanised boats for the Semhar Region (Massawa-Dahlak area), according to records current to November 1992.4 Not all of the Semhar units are active exclusively as fishing boats, however. Many are also engaged in transport of goods and people across the Red Sea. Indeed, it seems from the records that most boats do not regularly off-load catches at the Massawa receiving station. It is possible that some units alternately land fish at points across the Red Sea and secure supplies of ice, fuel, spare parts, etc., in those places. Units may also switch between the gillnet and handline fishery for prime food-fish, sometimes referred to as the “refrigerator” fishery because once the catch is removed from the water it is immediately gutted and put on ice in large insulated boxes or holds, and the gillnet fishery for sharks.
On the basis of available data on fish landed and purchased by the two marketing cooperatives (Red Sea and Merat Bahri) at Massawa for the first 11 months of 1992 (Table A12), it can be calculated that the global average landing of fresh food-fish was about 150 kgs per sea day. Assuming that these records sufficiently represent landing tendencies during the entire year, a 1992 total landing of around 300 tonnes of prime (Grade I & II) gutted food-fish (perhaps 350–375 tonnes whole weight) can be projected for the Massawa fish receiving station. Total landings of all fish can be projected to be in the range of 350 tonnes (perhaps 420–440 tonnes whole wet weight). With an estimated 65% of the total 1992 catch composed of Grade I fish, 22% of Grade II, and 13% of Grade III (up to November), the global value of the Semhar Region “refrigerator” fishery can be roughly estimated at Br 0.65 million.5
Comparison of the number of landings per fishing boat for the two Massawa-based cooperatives over a period of 7 months in 1992 (February to August -- the period for which complete information was available at the time of review) shows that a total of only 48 operational fishing craft can be recorded for the period (Table A13).
According to former estimates in the region (Walczak, 1977), the production per craft could be put at around 10–15 t/boat/year of first quality fish. Walczak estimated the annual production for large mechanised houris in Yemen to be around 14 t/year. Information on catch rates in Djibouti showed the same order of magnitude (15 t/year for the more efficient units). This production mostly consists of snapper and grouper and of some sharks. From the 48 active fishing boats operating in the Semhar region (Massawa-Dahlak area), all else being equal, it would not be unreasonable to expect an annual production of 650+ tonnes for 1992 -- a figure far above the 350–375 tonnes (ungutted wet weight) projected on the basis of official records.
When the number of distinct fishing boats landing their catches on a monthly basis is examined, it becomes obvious that relatively few boats are frequent callers at the Massawa centre and that marked changes are apparent over time (Table A14). The activity coefficient (AC)6 of the Semhar fishing fleet, derived from frequencies of landings per fishing craft (Table A15), displays very low levels during the hot season months of June to August (25%, 27%, and 33% respectively), and much higher levels during the cooler months of February to May (63%, 46%, 44%, and 44% respectively).
In order to acquire a more thorough picture of Semhar unit performance, a sample of 20 different craft landing catches during 1992 at Massawa fish receiving station was traced in closer detail. The results of this “Fishing Craft Tracer Study” reported more fully in Annex 2, indicate that Eritrean (Semhar) fishers have surprisingly low monthly sea-times, averaging 11 days for sambuks, 6.8 for houris, and 8.6 overall. This contrasts with their rather active performance in the past, when monthly sea-time was reckoned to average between 15 and 20 days. The present retarded activity levels may well be due to the fact that many operators have moved more or less permanently to bases across the Red Sea, landing their products in Massawa only from time to time. Also, the generally poor state of the remaining units permanently based in Eritrea strongly affects their operational capabilities.
The craft tracer study also provided information on crew recruitment and supply requirements (fuel and ice) as well as catch compostition per type of craft and type of fishing method (handlines, gillnets, or the two combined). These findings were used to construct tentative financial analyses of both houri and sambuk operations, as reviewed in Annex 3. In each case it was found that the financial internal rate of return was negative at pre-December 1992 levels of running, maintenance, and replacement costs versus earnings from catches sold at the Massawa landing centre, but that positive rates of return are achieved under the new (post-December 1992) producer price regime.
The analyses suggest that fishers, especially those using mainly gillnets, were finding it extremely difficult to survive financially and socially by selling their production exclusively at the Massawa station at the former ex-vessel price levels. Even with the improved financial performance of artisanal units that is now apparent under the new price adjustments it is possible that margins are not yet high enough to cover debt servicing obligations. Further escalation of input and supply prices and any other necessary but unforeseen expenditures would quickly erode the rather modest returns that now seem possible to achieve.
One aspect the financial performance of artisanal units which remains unclear so far deals with the additional catches of sharks that are salted-dried on island and mainland sites close to the fishing grounds but which are not landed at the Massawa receiving station. Some additional income must be generated through this activity but it is open to question just how much this improves the financial balance of the whole operation.
Though the situation has now been ameliorated to a substantial degree, there would still appear to be a strong incentive to commercialise at least a proportion of local catches across the Red Sea, where producer prices are up to three times higher. It is possible that attractive returns could thus be realised, in spite of the steaming costs involved and the fishing time lost. These latter problems could be circumvented by selling processed commodities to local traders-exporters collecting on the spot.
Squeezed between a floor of basic production costs and a ceiling of fixed selling prices, local operators who wish to continue their fishing activities appear to have few other options available to them. It is noteworthy that handlines have become the most important fishing method in the Massawa-Dahlak area while gillnets are on the wane (they mainly produce only second and third grade fish), and that the fishing effort of the artisanal fleet is being concentrated around banks and coral reefs while open water areas and their respective resources are being neglected. These perhaps are indications that a basic transformation is taking place in the sector as a result of the pressures operators may be feeling.
The high producers price paid for “snail nails” and “sea cucumber” must also play an important role as source of additional income for women and the fisher families in general. Quantifying these additional inputs could provide better understanding of the whole fishing activity.
