The biology and population ecology of the pollan, Coregonous autumnalis pollan Thompson, of Lough Neagh

Abstract

For centuries Lough Neagh pollan have been fished for local consumption and for export. The Irish pollans are the only cisco-type coregonines in Western Europe. The aim of this study is to examine the basic biology of the pollan and to derive possible management strategies for the fishery.

Samples were taken monthly (March 1976 - May 1978) by using the University's trawl, by occasional gillnetting and by subsampling fishermen's catches. Over 3 000 specimens were examined, most of which were taken from the north-eastern part of the Lough.

Ages determined by examination of scales, otoliths and opercula were compared. Scales were the most satisfactory structures for ageing, although age determination is complicated by the formation of an autumn accessory check on the scales of most O+ and l+ pollan and of some older fish. The main annual mark on scales is formed in May, coincident with a marked increase in feeding rate. Von Bertalanffy growth parameters are t = 0.06 yr, Loo= 28.9 cm, k = o.6.47. Total mortality, of which about one-fifth is due to fishing, is high: z = 1.53 in· 1976 and 2.18 in 1977. Few fish more than four years of age are caught.

1+ and older pollan are mainly opportunist feeders on chironomid larvae and Gammarus in winter and on emerging chironomid pupae in late spring. In summer pollan select Iaphnia1 avoiding the abundant copepods in the zooplankton. O+ fish feed -mainly on small chironomids, Bythotrephes longimanus and small Daphnia.

Pollan mature in their third year of life. Spawning takes place on shallow, rocky substrates in late November and December. The time of spawning is similar to that of the American ciscoes but rather earlier than that of the British coregonines. Each female produces 5000 - 12000 eggs. Fecundity was significantly correlated with log length and log weight, but in multiple regression analysis variables such as age and condition factor did not show a significant correlation with egg number. Relative fecundity was higher in 1971 than in 1977.

Covariance analysis revealed that the exponent bin the length weight relationship W = aLb did not significantly vary with year of sample, fish age and sex, or time of year: Wc: L3-152 for all pollan, indicating a relative increase in girth.as the fish grow. The seasonal cycle in relative condition factor W = aLb was studied for male, female and immature fish.

Age composition data for 1976 - 78 showed that the level of recruitment of l+ fish to the fishery was very loil in 1977 and 1978. It is probable that past fluctuations in pollan abundance were due to such recruitment failures rather than to overfishing. The number of recruited fish in the Lough was roughly estimated by extrapolation from rates of catch in experimental trawling at 7 million.

The history of the pollan fishery was documented. In recent years about 200 tons of pollan have been taken annually from the Lough by trawling, draft netting or gillnetting. Fishing mortality, based on 1976 catch figures, was estimated at F = 0.55, giving by subtraction a natural mortality rate M of 0.98. The yield estimation models of Walters and of Beverton & Holt were used to simulate the pollan population. The Walters model was run for four plausible types of stock-recruitment relationships. The results indicated that yield would be maximised by increasing F to about 1 .0, equivalent to an annual rate of exploitation of 51%. As the fishery depends on one or two year- classes and is subject to occasional low levels of recru1tment, the probability of collapse would be reduced if the fished population contained more year-classes. It is accordingly suggested that the pollan fishery could best be regulated as a gillnet fishery with a minimum mesh size of '30 mm bar imposed.

Date of Award	Oct 1979
Original language	English