Introduction
The purpose of this paper is to determine whether long-term monitoring of American eel, Anguilla rostrata, recruitment at a fish ladder in Cornwall, Ontario can be used to predict the abundance and commercial catch of eels in the Lake Ontario and upper St. Lawrence River ecosystem. American eel is an economically and ecologically important species for the entire east coast of North America. In Canada, major commercial eel fisheries exist from the east coast, inland to Lake Ontario. American eels are harvested at different life-stages that include elvers (young), juvenile (referred to as yellow), and adults (referred to as silver). Recently, there has been great concern regarding dramatic declines in abundance.
Eels are catadromous species, living in freshwater and spawning in the ocean. Both the American and European eel, Anguilla anguilla, share the same spawning site in the Sargasso Sea. Because spawning adults have never been seen, this spawning site was inferred by the capture of elvers in the area (Schmidt 1925). Spawning of the two species is thought to be separated by a three week period. After spawning, the adults die and their progeny migrate as leptocephali (larvae) to coastal areas; leptocephali of American eel drift with the aid of the Gulf Stream, dispersing from the Gulf of Mexico, north to Labrador and Greenland (Castonguay et al 1994a). Leptocephali undergo metamorphosis and develop into "glass" eels before entering brackish and freshwater, where they grow for many years. Upon maturing, they migrate back to the Sargasso Sea to spawn.
Eels populations are panmictic meaning that random mating occurs at one spawning site resulting in a complete mixing of the gene pool. Hence, while there is "homing" to the spawning area, there is no homing tendency to specific rearing areas in freshwater. Based on our present knowledge, there is no genetic difference among the young entering freshwater systems that are separated by over 4000 km of Atlantic coastline (Avise et al 1986, 1990). The consequence of panmixia is that the species is a single population. The number of adults leaving a specific freshwater system does not necessarily reflect the number of returning young to that system, unless the adults represent a large proportion of the entire population. In this paper, therefore, we refer to the Lake Ontario and upper St. Lawrence River stock (a functional management unit), which includes American eel (hereafter referred to as eel) in waters of Lake Ontario and the upper St. Lawrence River, in the Province of Ontario. This stock is exclusively females and upon maturing, individuals reach 1 m long and weigh 2 to 3 kg.
The St. Lawrence Seaway was built between 1954 and 1958 and included the W. B. Saunders Hydroelectric Dam, at Cornwall, Ontario. It was believed that the dam impaired migration of the Lake Ontario and upper St. Lawrence River stock. In 1974, therefore, an eel ladder was built to aid the passage of eels. Building of the ladder was a cooperative effort between Ontario Hydro and the Ontario Ministry of Natural Resources (OMNR). From 1974 to late- July 1987, total estimates of eels ascending the ladder each year were made from random daily counts. In July 1987, an electronic counter was installed and eels ascending the ladder were counted electronically. Between early 1980s and 1990s, a typical juvenile eel (yellow) ascending the ladder averaged 35 to 45 cm (TL) and weighed 80 to 150 g.
Current Status of the Stock
Between 1975 and 1995, total number of eels ascending the ladder provided an index of recruitment to the Lake Ontario and upper St. Lawrence River stock. From 1975 to 1985, there were relatively large and consistent numbers of eels ascending the ladder, from 600,000 1,300,000 annually. In 1986, however, a substantial decrease occurred as the total number of eels dropped to 200,000. This declining trend continues to date, reaching a low in 1993 when only 8,000 eels ascended the ladder. If this decline is real, there could be severe consequences. These consequences include: loss of the fishery, possible extinction of the Lake Ontario and upper St. Lawrence River stock, a species-wide decline, and the ecological impacts of eliminating eels from the system are unknown.
There has been some question, however, whether the total number of eels ascending the ladder is indicative of eel recruitment to the Lake Ontario and upper St. Lawrence River stock, as eels may be using alternative migration routes (i. e. the canal system). Therefore, these data were re-examined to standardize the counts to a specific interval each year and to provide a sound basis for comparison with indices of abundance. Three sources of long-term data on eel abundance in Lake Ontario and the upper St. Lawrence River were compiled and examined. These sources included: commercial harvest, 95% of which is concentrated in the eastern basin of Lake Ontario, Bay of Quinte, and the upper St. Lawrence River (Province of Ontario); commercial electrofishing in the outlet basin of Lake Ontario and the upper St. Lawrence River (Province of Ontario); and research trawls in the Bay of Quinte. Our specific objectives were to examine these data sets for trends, determine if the number of eels ascending the ladder could be used as an index of recruitment to the commercial fishery, and if the decline in recruitment was real, identify possible causes related to the decline.
Indices of Abundance
The commercial harvest index in the Province of Ontario spans 113 years, from 1884 to 1996. Over this period, harvest fluctuated, but the causes of these fluctuations is often difficult to determine because there was no reliable measure of fishing effort. In 1914, harvest was just under 300,000 pounds, the second highest ever recorded. Afterwards, sharp reductions occurred and a declining trend continued into the early 1940s. Harvest was lowest in 1940s, recording a low in 1942 of only 16,000 pounds. Afterwards, it increased until 1970, when there was a partial closure of the fishery because of mercury contamination. In 1972, the fishery was revived for approved export markets outside of North America (Baldwin et al 1979). Harvest continued to increase to a high in 1978 of just over 500,000 pounds, but in 1982, it plummeted to 64,000 pounds due to the closure of the European market in response to mirex contamination. By the mid- 1980s, the eel market had recovered again but with a decreased fishing effort because some commercial licences were bought out by the OMNR. Harvest remained stable immediately afterwards, but might have been higher without the buyout. Recently, from 1993 to 1996, harvest has declined for four consecutive years. Except for the market closure due to mirex, this is the only four consecutive year decline over the 113- year record of the commercial catch.
From 1945 to 1978, there was a strong relationship between increasing price (Canadian dollars) and harvest (P < 0.001, r2 = 0.846). Similarly, from 1979 to 1984, there was a strong relationship between decreasing price and harvest (P < 0.002, r2 = 0.937), related to mirex and the closure of the European market. Between 1993 and 1996, a new trend emerged. Price increased by 153% and harvest decreased by 51%; this indicates a marked decrease in abundance of harvestable fish.
The Fisheries Research Section of the OMNR developed a trawl index in the Bay of Quinte that spans 25 years, from 1972 to 1996. This trawling program was part of a Fish Community Index that did not specifically target eels, but measures the relative abundance of eels inhabiting the Bay. Statistical analysis of mean annual catches indicate a significant decline from the 1970s to the 1990s. In the 1970s, mean annual catches of eels were > 2 per nautical mile, while most catches in the 1990s were < 1 per nautical mile.
The commercial electrofishing index spans 13 years, from 1984 to 1996. In 1984, the OMNR issued an experimental commercial electrofishing licence to one fisherman who fished the outlet basin of Lake Ontario and the upper St. Lawrence River. This fisherman kept extremely precise records on catch and effort that resulted in very low variance, which made changes in annual catches easy to detect. We calculated mean catch per hour electrofishing from 1984 to 1996. Statistical analysis showed a decline in catch starting in 1991 and 1992, and a significant decline occurring in 1993, which continues to date.
Relationship Between Abundance and Recruitment
To determine if abundance indices from the commercial fishery and research trawls reflect the index of recruitment at the ladder in Cornwall, we first standardized data from the ladder to increase its precision for statistical analysis. Prior to installation of the electronic counter, total counts of eels ascending the ladder were estimated from daily counts, which may have been overestimated in the earlier years. In addition, from 1975 to 1995, the period of annual operation varied from 60 to 125 days. We standardized the ladder data to a 31 consecutive day period when the largest proportion of eels ascended the ladder. This 31 day peak migration period occurs in July and August and provides a more reliable annual index for comparisons among years and to abundance indices. The variance for this peak migration index was the lowest of all indices and permitted detailed quantitative analysis.
The standardized index spanned a 22- year period, from 1974 to 1995 (Fig. 1), and it confirmed the trend of declining numbers of eels ascending the ladder, starting in 1986 and continuing to 1995. The index also demonstrated that the number of eels ascending the ladder were not relatively consistent from the mid- 1970s to 1985, but showed an increasing trend to 1982.
All three indices of abundance (commercial electrofishing; commercial harvest; research trawls) were highly correlated with the ladder index. When lag times between eels ascending the ladder and catch were applied, the best statistical correlation for the electrofishing index was based on a five year lag (Fig. 2). This means that eels ascending the ladder between 1979 and 1991 were reflected in the electrofishing catch between 1984 and 1996, five years later. The commercial harvest index showed the best statistical correlation with the ladder index with an eight year lag (harvest -- 1982 to 1996, ladder -- 1974 to 1988) (Fig. 3). The research trawl index and the ladder index were correlated when there was a four year lag (trawls -- 1978 to 1996, ladder -- 1974 to 1992) (Fig. 4). The reason for the different lag times is likely the different types of capture gear that catch different sizes of eels. These correlations, however, confirm that the ladder index of recruitment can predict subsequent catches of eel by various gears in the upper St. Lawrence River and Lake Ontario.
The ladder provided a reliable index of eel recruitment to Lake Ontario and the upper St. Lawrence River and demonstrated dramatic declines over the past decade, which began in 1986. The commercial harvest in the Province of Ontario followed a similar trend as the ladder index, lagging by seven to nine years; the commercial electrofishing index indicated a decline in catch starting in 1991 and 1992, and a significant decrease commenced in 1993.
Causes
Four possible causes of declining recruitment have been reviewed by Castonguay et al (1994b). The first cause is habitat change. From 1954 to 1958, the St. Lawrence Seaway was built, which consists of a series of dams and locks. The dams did not completely obstruct upstream migration of the Lake Ontario and upper St. Lawrence River stock, but they may delayed migration. More importantly, turbines at the hydroelectric dams could effect downward migration of adult "silver" eels returning to their spawning site in the Sargasso Sea. A second cause is overfishing. There is general concern that some portions of the stock are being overfished, but no specific data presently exists. A third cause of declining recruitment is pollution. Recent studies, however, suggest that chemical contamination by persistent chlorinated compounds is lower now than in the 1970s. Because eel numbers are declining now and were high in the 1970s, this cause does not appear related. However, there are few data on other types of pollution. The fourth cause of declining recruitment is ecological change. Some researchers believed that changes in ocean climate may have slowed the Gulf Stream, which was important for migrating young, but the data are inconclusive.
We believe that if escapement of adult "silver" eels is not increased, the commercial fishery in the upper St. Lawrence River and Lake Ontario may not be viable in another generation's time, and eels may even become rare. Decline of the Lake Ontario and upper St. Lawrence River stock could have consequences for the whole population. This stock has been estimated to represent 5% of the entire population, and 20% of all females. This research demonstrates that we now have a precise tool to predict future abundance of eels from recruitment and to help determine factors controlling eel population dynamics.
Acknowledgements
We thank the many field technicians that worked for the Ontario Ministry of Natural Resources (formerly the Department of Lands and Forest) who collected and maintained the long-term data sets. We thank the people from the Ministry of Natural Resources, Ontario Hydro, and Hydro-Quebec who aided in the acquisition of data and reports that covered an extensive time period. Special thanks go to J. Rorabeck who voluntarily collected data on commercial electrofishing in cooperation with J. Casselman. Some of this research was funded by Environment Canada as part of the Ecological Monitoring and Assessment Network (EMAN).
References
Avise, J. C., G. S. Helfman, N. C. Saunders and L. S. Hales. 1986. Mitochondrial DNA differentiation in North Atlantic eels: population genetics consequences of an unusual life history pattern. Proceedings of the National Academy of Sciences, USA 83: 4350-4354.
Avise, J. C., W. S. Nelson, J. Arnold, R. K. Koehn, G. C. Williams and V. Thorsteinsson. 1990. The evolutionary genetic status of Icelandic eels. Evolution 44: 1254-1262.
Baldwin, N. S., R. W. Saalfeld, M. A. Ross and H. J. Buettner. 1979. Commercial fish production in the Great Lakes 1867-1977, p. 20. In Great Lakes Fishery Commission, Technical Report No. 3. Ann Arbour, Michigan.
Castonguay, M., P. V. Hodson, C. Moriarty, K. F. Drinkwater and B. M. Jessop. 1994a. Is there a role of ocean environment in American and European eel decline? Fisheries Oceanography 3:3, 197-203.
Castonguay, M., P. V. Hodson, C. M. Couillard, M. J. Eckersley, J.-D. Dutil and G. Verreault. 1994b. Why is recruitment of the American Eel , Anguilla rostrata, declining in the St. Lawrence River and Gulf? Canadian Journal of Fisheries and Aquatic Sciences 51: 479-488.
Schmidt, J. 1925. The breeding places of the eel. Smithsonian Institute Annual Report 1924: 279-316.
