Department of Fisheries and Forestry, Fisheries Research Centre
Abbotstown, Dublin, Ireland
Data on elver recruitment and eel stocks in Ireland are reviewed. The annual catch of eels in Lough Neagh of the order of 600 to 800 t indicates an exceptionally high yield per unit area by northern European standards. The corresponding yield in the majority of lakes in Ireland is well below the average. Data on eel populations and on elver recruitment in Ireland are reviewed. It is concluded that the low eel yield results from topographical features of the river systems and from disruption of elver ascent by hydro-electric dams. Data on yield per recruit in the Lough Neagh fishery suggest that at least a tenfold increase in catch elsewhere in Ireland could be achieved by augmenting the existing system of elver transportation.
C'est en 1959 que l'on a commencé en République d'Irlande à prélever des civelles à l'embouchure du Shannon pour les transporter, par voie terrestre, en amont. L'auteur décrit l'évolution des stocks et des captures (quantités et composition) par suite, en premier lieu, de la construction de barrages hydro-électriques et, en second lieu, du transport des civelles. Il passe en revue les travaux consacrés à l'évaluation des stocks d'anguilles et à leur régime alimentaire dans différentes parties de l'Irlande et aboutit à la conclusion que le transport des civelles devrait permettre de multiplier par 10 les rendements sans nuire aux stocks de poissons destinés à la pêche sportive.
The mean annual catch of eels in Ireland (Republic) from 1970 to 1977 was 113 t, while the catch in Northern Ireland was 765 t (FAO, 1979). The eel fishery of Northern Ireland is based on two lakes; Lough Neagh and Lough Erne with by far the greater proportion coming from Lough Neagh (Black, 1981). The total area of lakes in the Republic is considerably greater than the combined area of these two and there are no significant differences in climate between north and south. There is accordingly reason to believe that the yield of eels in the Republic could be enhanced.
Eels have been traditionally caught by long line or by a variety of silver eel traps in rivers. Fyke nets were first used experimentally in 1961 but, in the course of the next 20 years, were adopted by only 15 fishermen in the Republic. The lack of enthusiasm for development by fishermen suggested that stocks were too low to provide a viable fishery in most cases. This suggestion was supported by the results of extensive fyke-netting surveys carried out by the author since 1965.
The major eel fishery in the Republic is based on the River Shannon, which has the most extensive catchment area and the greatest total area of lakes in the country. This fishery is owned by the Electricity Supply Board (ESB), which constructed a major hydro-electric dam at Ardnacrusha in 1928. Even though a flow of water was maintained in the River Shannon, by-passing the dam, it was apparent that the entry of elvers into the lakes had been seriously impeded. An elver pass was constructed at Ardnacrusha in 1958 and overland transport of elvers to Lough Derg began in 1959.
Overland transport of elvers to Lough Neagh has been in progress since 1933, with a break from 1948 to 1959 (Parsons et al., 1977) and to Lough Erne since 1960 (Moriarty, 1981). Elver catch data allow an estimate of yield per recruit to be made. The results of the fyke-netting surveys give evidence of low stocks in most Irish lakes and provide an explanation of the situation. In the light of this explanation and of the yield per recruit data an estimate of the potential yield following a stock enhancement programme is presented.
Comparable data for elver catches are available for Lough Neagh (River Bann system, Fig. 1) and the Shannon Lakes. There is an unknown degree of escapement of elvers in the case of the Bann and the Shannon but it is generally believed that the catches in both cases represent a large proportion of the elver runs. The Bann traps, situated at Coleraine, are operated beside a substantial weir, which is likely to discourage direct ascent by the elvers. The Shannon traps are either situated at weirs or at the hydro dam at Ardnacrusha, which appears to attract the majority of the elvers. The Erne trap, at Ballyshannon, is operated at a hydro dam which controls the entire river. Elvers do not appear to use the salmon pass on this dam.
Mean annual catches for periods between 1960 and 1981 show that similar numbers of elvers enter the rivers Bann and Shannon while recruitment to the Erne is very much less (Table 1).
Details of a number of these surveys have been published (Moriarty, 1979) and the fyke-netting technique has been assessed in Moriarty (1975). The results, summarized in Table 2, are based on a standardized effort of one net fished overnight in water between 2 and 3 m in depth. The net consists of a pair of funnel traps joined mouth to mouth by a single leader.
The catch per unit of effort is taken as a measure of the numbers of eels in the stocks. Within a river system the numbers decrease with increasing distance from the sea. This is shown very clearly in the case of the Corrib system where the three lakes investigated lie on the main river. Eels must travel through the downstream lakes to reach the higher ones.
The decrease in numbers on the Shannon system is not so obvious and the distance from lake to sea in the case of the other three lakes has no apparent effect on numbers. In the case of the two rivers numbers are very much higher than in any of the lakes, with no direct relationship to distance upstream. An increase of 62 percent in numbers with time was recorded in Lough Derg, between 1969 and 1981.
Length frequencies change dramatically from downstream to upstream lakes: small eels are plentiful close to the sea and scarce at increasing distances. Large eels are scarce in the downstream lakes and much more numerous upstream. Age determination confirms that increasing length results from increasing age and shows that the eels in general take a number of years to reach the upper lakes. Small eels are plentiful in Lough Conn at 25 km from the sea and in Lough Gill at 4 km but scarce in Lough Arrow at 24 km while large eels are scarce in Lough Gill and Lough Conn.
The fish eaten are mainly perch, Perca fluviatilis, or cyprinids; salmonids are rarely present in the stomachs. Eels of less than 40 cm rarely eat fish in any case. In the River Blackwater fish become important in the diet of eels of over 40 cm but in the lakes do not appear to any marked extent at lengths of less than 50 cm. Where Asellus is abundant, fish are seldom eaten by eels of any size.
The stomach contents of samples of the eels were analysed with results shown in Table 3.
Compared with the results from the two rivers, the catches in all the lakes are very low. There is no reason to believe that the nets operate any less efficiently in lakes than in rivers and, therefore, it appears that stocks per unit of area in the lakes are much lower than in the rivers. In the case of Lough Corrib and Lough Gill, which are very close to the sea and whose outflowing rivers provide no serious obstruction to the ascent of elvers, it appears that the natural run of elvers is insufficient to populate the available area of lake bed at the maximum possible density. It is possible that similar numbers of elvers enter the two rivers but yield much higher densities of eels since the area of river bed is relatively much smaller than in the lakes.
Even lower numbers of eels are present in the upstream lakes. Comparing the stock density of the rivers with that of the downstream lakes suggests that there is adequate space for many more eels to live in the downstream lakes and therefore no pressure to induce those which have arrived to migrate upstream through the lake system. The downstream lakes thus serve almost as a barrier to the continued migration of the growing eels. There is, in fact, a slow movement through the lakes so that the upstream lakes are not totally devoid of eels.
The low stocks in Lough Derg and Lough Arrow may both be explained by the presence of major obstructions to elver ascent in the firm of the hydro dam at Ardnacrusha and a high waterfall close to the mouth of the Ballysadare River which drains Lough Arrow. Lough Conn is unusual in the Irish context in lying on a tributary of the River Moy which flows directly to the sea without passing through a lake system. All the other downstream lakes lie on the main rivers.
Lough Derg was first sampled in 1969. Very few of the eels in the samples were aged less than ten years. Elver transport had begun in 1959 and would not, therefore, have influenced the numbers in the samples. By 1981 it was clear that the continuing elver transport had brought about a marked increase in the yellow eel numbers and also had altered the length frequency distribution in favour of smaller eels.
Observations on the food of the eels indicated that neither in lakes nor in rivers were salmonids significant prey species. In many lakes fish of any species were rarely present in eel stomachs. In those where the piscivorous diet was normal, perch, cyprinids and eel were the usual food. Since perch and the cyprinids are prolific breeders, it is considered that there is no danger of enhanced eel stocks reducing the existing fish population by predation. In the lakes where Asellus is the dominant food of the eel there appears to be no risk of competition with salmonids and in the other lakes there was little evidence of eels feeding on the principal salmonid food species. Competition for food with cyprinds cannot be completely ruled out but observations to date have not suggested that this is likely to be serious. The most dense lotic water population of eels observed to date, that of the South Sloblands lagoon on the south coast of Ireland, co-exists with an abundant stock of rudd, Erythrophthalmus scardinius (Moriarty, 1972).
In a review of eel stocking practices, Wickstrom (1979) quotes rates from as low as 6 to as high as 1 500 glass eels/ha. Tesch (1977) gives yield figures for market eels ranging from 0.03 to 40 kg/ha and states that average yields are between 3 and 10 kg/ha. Leopold (1980) provides figures for stocking rates and catch in a number of Polish lakes, with variations in stocking rate and in intensity of fishing. The maximum yields result from a combination of higher than average fishing effort and highest stocking rate, giving a mean yield of 2.39 kg/ha at a stocking rate of 128.8 elvers/ha.
In the case of Lough Neagh, the minimum annual catch is 600 t and a record catch of 1 020 t was made in 1979 (Black, 1981). The usual annual range for this lake of 39 000 ha is 600–800 t. The production of eels has remained high for many years and it is particularly interesting to note that the latest available catch statistics give the greatest catch on record both of silver and of yellow eels. This has been achieved in spite of a period marked by a high level of poaching and serious management difficulties. The mean annual number of elvers transported from 1960 to 1974 was 13.8 million (Parsons et al., 1977).
The numbers of elvers transported to Lough Erne are known with a fair degree of accuracy: the mean annual supply from 1960 to 1981 being 3.3 million. The catch of market eels from Lough Erne is reckoned to be up to 49 t per year (Black, 1981). There is an extensive lake system upstream of Lough Erne. The lakes of the River Shannon have a total surface area similar to that of Lough Neagh and have been supplied with eels at a similar, or possibly higher, rate. However, the yield per hectare of eels is very much lower: at approximately 1.4 kg/ha for the years 1976 to 1980, well below the average for German lakes and of a similar order to figures from Poland. Fishing on the Shannon lakes is exclusively for silver eels.
The situation of Lough Neagh has given it certain advantages for an eel production study. Firstly, the lake is a large one and has no outlet to other potential eel-producing waters. Secondly, elvers have been trapped and known quantities introduced to the lake for a considerable period. The catch figures are those which have been officially reported and therefore indicate a minimum. An unknown number of elvers reach the lake by making an unaided ascent of the River Bann but, as the weir at the trapping station forms a considerable obstacle, it is believed that the catch represents a high proportion of the numbers actually entering the system. Figures for production per unit area of lake surface are, therefore, if anything, an underestimate. The fact that they are well above average for northern European lakes therefore suggests that other lakes are producing eels at a level considerably below the maximum possible.
One factor which might contribute to the success of Lough Neagh is the paucity of its fish fauna. Cyprinids are represented by not more than six species and percids by only one. There could, therefore, be considerably less competition for food and less predation than in continental lakes. On the other hand, the water temperature in Lough Neagh rarely rises above 20°C and the number of species of invertebrates available as food is less than on the continent.
Figures relating directly to stock of elvers and yield of eels are very few. Leopold's (1980) data for Polish lakes indicate that yield varies with stocking rate. In his paper both stocking rate and yield are considerably lower than the Lough Neagh figures.
It is possible that, for reasons unknown, conditions in Lough Neagh are ideal for eels and superior to those in other lakes studied. This is hard to accept, particularly when temperature and sparse fauna are considered. The conclusion, therefore, is that the stocking rate in Lough Neagh, of 350 glass eels/ha is the best known figure to be applied to other lake systems. On this assumption, Table 4 has been compiled as a recommended programme for the larger and richer lakes of Ireland.
Since 1959 the Electricity Supply Board has transported elvers to the Shannon lakes at rates ranging from 29 to 770/ha, with a mean of 340/ha. The mean annual catch of silver eels for the years 1961 to 1970, before transplanting would have had any significant effect, was 19 900 kg. Unpublished age determinations by the author indicate that few female eels in the system become silver in less than 15 years and therefore an increased catch resulting from the transportation might be expected about the year 1975. The mean annual catch since then has been 35 000 kg with a record catch of 44 000 kg in 1980. This yield is still no more than 1.4 kg/ha, less than one tenth of the yield in Lough Neagh.
The transportation programme in the Shannon has not yet, therefore, yielded spectacular results. There are several possible explanations. Although the operation began in 1959, stocking at the rate of 350/ha or over was not achieved until 1966 so that considerable improvements in catches may yet take place. The fishing effort may not have been increased at a rate which would be sufficient to harvest the increased stocks efficiently. Finally, ecological factors in the Shannon lakes may prevent the development of eel populations of similar density to those inhabiting Lough Neagh.
The eel fishery of Ireland, therefore, presents two aspects of stock enhancement. In the case of the major fishery in the Shannon transportation of elvers at a rate known to yield excellent results has been in progress for 16 years. The catch has doubled in this period, but appears to be far below the potential value. In the case of the other lakes, a transportation scheme has been planned but has to-date been implemented only at a pilot stage. However, considerable details of the present stock situation in these lakes are available and will enable a very valuable monitoring of future developments to be made.
Black, R.C., 1981 Report of the Committee of Inquiry into angling in Northern Ireland. Belfast, Government of Northern Ireland, 65 p.
FAO, 1979 Yearbook of fishery statistics. Annuaire statistique des peches. Anuario estadistico de pesca. Catches and landings. Captures et quantites debarquees. Capturas y desembarques. FAO Yearb.Fish.Stat./Annu.Stat.Peches/Anu.Estad.Pesca, (46):372 p.
Flanagan, P.J. and P.F. Toner, 1975 A preliminary survey of Irish lakes. Dublin, An Foras Forbartha, 164 p.
Leopold, M., 1980 Some remarks on eel management in lakes. Aquaculture, 19:97–101
Moriarty, C., 1972 The growth and nutrition of fish in Ireland with particular reference to the eel, Anguilla anguilla (L.). Ph.D. Thesis, Trinity College, Dublin
Moriarty, C., 1975 The small fyke net as a sampling instrument in eel research. EIFAC Tech.Pap., (23) Suppl.1, vol. 2:507–18
Moriarty, C., 1979 Biological studies of yellow eels in Ireland. Rapp.P.V.Reun.CIEM, 174:16–21
Moriarty, C. and D. O'Leary, 1981 Glass-eel catches and flow data from River Erne at Ballyshannon, Ireland. Paper presented to the EIFAC Working Party on Eel (mimeo)
Parsons, J., K.U. Vickers and Y. Warden, 1977 Relationship between elver recruitment and changes in the sex ratio of silver eels Anguilla anguilla L. migrating from Lough Neagh, Northern Ireland. J.Fish.Biol., 10:211–29
Tesch, F.–W., 1977 The eel. London, Chapman and Hall, 434 p.
Wickstrom, H., 1979 Preliminary recommendations for stocking with eels. Inf.Sotvattens-Lab.Drottningholm, (5):1–24
Table 1 Mean annual quantities of elvers caught for transport (River Bann data from Parsons et al., 1977)
|River Shannon||1972–81||16.6||5 529||1 053|
|River Bann||1960–74||13.8||4 613||546|
Table 2 Numbers and length frequency of eels caught by fyke net
|Distance from sea (km)||Sampling effort|
(nets × nights)
|Catch/unit effort||%Length Frequency|
Table 3 Dominant food organisms
|Asellus spp.||Lakes Derg, Key, Mask, Arrow, River Barrow|
|Chironomidae and fish||Lough Corrib|
|Trichoptera and fish||Lough Gill|
|Gastropoda and fish||Lakes Carra, Conn|
|Ephemeroptera and fish||River Blackwater|
Table 4 Eel yield potential at 20 kg/ha and stocking requirements at 350 elvers/ha in Irish lakes of area 500 ha and conductivity 125 us/cm (Physical data from Flanagan and Toner, 1975)
(kg at 3 000 elvers/kg)
|Derg||11 635||233 100||1 396|
|Derravaragh||1 100||22 000||132|
|Ennell||1 400||28 000||168|
|Gara||1 100||22 000||132|
|Ree||10 500||210 000||1 260|
|Sheelin||1 900||38 000||228|
|Carra||1 500||30 000||180|
|Corrib||17 000||340 000||2 040|
|Mask||8 000||160 000||960|
|Arrow||1 250||25 000||150|
|Conn||5 000||100 000||600|
|Cullin||1 100||22 000||132|
|Gill||1 400||28 000||168|
|Gowna||1 100||22 000||132|
|Oughter||1 300||26 000||156|
|Total including smaller lakes||72 223||1 440 000||8 666|
Fig. 1 Ireland, showing eel-producing lakes and sampling stations on rivers