If plankton are the main crop in the ocean, capelin are the main grazers. Their role in the system is to turn much of that food into a form other species can eat. Some large fish, most shellfish and even some whales eat plankton, but most others eat it indirectly, by feeding on smaller plankton-eaters like capelin.

In our waters, capelin is the primary link between zooplankton and predatory fish. Arctic cod and sand lance fill a similar role in the system, but neither comes close to capelin in abundance or importance. However, Arctic cod has been increasing in our northern waters over the last 10 - 15 years, and has replaced capelin as the biggest item in the diet of harp seals. Sand lance is most abundant on the Grand Bank, and is important in the diet of most predators in that area. Its abundance is probably increasing with the decline of major groundfish stocks.

Capelin is a short-lived species. Of those which live to maturity, most die soon after spawning in their third or fourth year. This short lifespan makes capelin subject to large and sudden fluctuations in abundance. With only four ages of fish in the population, the survival rate of each new year-class matters much more than it does with a long-lived species like cod. Two good or two bad years in a row can make an enormous difference to the total number of capelin.

At its peak, the total biomass of the main capelin stock in our waters, in NAFO area 3L, was estimated at 7 million tons. This was in 1990. The lowest biomass, in the late 70's, was probably about one-tenth of that.

These two central facts about capelin - their critical role in our ocean food system and the wide fluctuations in their abundance - affect the wellbeing of all the species that depend on capelin in their diet. How fast these other fish grow, how they survive in hungry periods, and how successfully they reproduce may all depend on the availability of capelin when they need them.


How does a predator respond to huge fluctuations in the abundance of its favourite food? In the case of cod, faced with a highly variable supply of capelin, the responses could include changes in both their wellbeing and their behaviour.

In terms of wellbeing, cod may show a variety of responses when capelin are scarce. They may grow more slowly; approach the rigors of winter with lower reserves of fat; spawn with less success; or survive in fewer numbers from one year to the next.

In terms of behaviour, we know that after the demands of their own spawning, large numbers of cod follow spawning capelin right to shore. The entire inshore cod fishery has been built on this behaviour, but we know little about the factors that influence onshore migration, and what role capelin abundance plays in it. For example, when capelin are scarce, do fewer cod come to shore? When juvenile capelin are abundant offshore, is there less incentive for cod to migrate to the coast?

In studying such relationships, there is no way to conduct large experiments in real life. You cannot, for example, remove capelin from the system and watch how cod respond. The only option is to monitor both species and their environment closely, to follow them through several cycles of abundance and scarcity, and see what relationships emerge.

This is a painstaking approach, yielding essentially one number a year in each of the points of study: cod abundance; capelin abundance; the presence of capelin in cod stomachs; the presence of other prey in cod stomachs; the condition and growth of cod and so forth. In time, a much fuller picture of the link between these two species should emerge, particularly when similar studies monitor the cod-capelin relationship in similar waters, like the Barents Sea and the waters off Iceland.

Through the International Council for the Exploration of the Sea (ICES), researchers here and abroad have worked out a common approach to these questions, so that results from each area can be compared to the others. For example, in the Barents Sea, the growth rate and condition of cod decline when capelin are scarce, but here and in Iceland the evidence on this point is not clear. Here, the abundance of capelin has been highly uncertain in recent years. However, we know that in the 1970's, when capelin were definitely scarce, there was no obvious decline in the condition of cod.

During the life of the Northern Cod Science Program, exceptional circumstances have confounded efforts to understand the dynamics of the relationship between cod and capelin. There has been a disastrous decline in the abundance of cod, as well as distinct and puzzling changes in the distribution of both cod and capelin. These changes are well outside the bounds of previous experience, leaving researchers uncertain how to interpret them. For example, inshore and offshore indications of capelin abundance, which formally tended confirm each other, now point in opposite directions. Offshore surveys have not found much capelin within the normal survey area, yet capelin have been turning up in normal abundance on the spawning beaches.

Uncertainty about the status of capelin has made it very difficult to interpret the data collected on this project so far. In time, however, as the status of capelin becomes more certain, a clearer picture should emerge.

In the meantime, related studies have illuminated some corners of the larger picture, including the food preferences of northern cod of various sizes, the role of cannibalism in cod diet, and the feeding rate of cod on the northern Grand Bank, in NAFO Division 3L. NCSP has funded much of this research.

The Role of Capelin in the Cod Collapse

The dramatic decline in the abundance of northern cod between 1988 and 1992 coincided with a decline in bottom water temperature and an apparent decline in the abundance of capelin in the same period. This makes it tempting to look for a casual relationship between these events.

At first glance, the evidence from southern Labrador, in NAFO Division 2J, is most suggestive. There, studies of the fullness of the stomachs of cod taken in groundfish surveys since 1978 show a decline through the early 1990's to the lowest level ever recorded in 1994. In Division 3K, off Newfoundland's northeast coast, a similar decline came later but was very pronounced in 1994. Superficially, these results suggest that cod experienced a scarcity of food in this period.

However, studies of the condition of cod taken in these same surveys show no appreciable change in their liver size or other measures of their physical wellbeing. Moreover, a closer look at the feeding data makes it clear they should be interpreted cautiously. Most important, the surveys show that cod began to abandon 2J before capelin did, and before the recent period of cold water began. Similarly, the declines in the abundance of cod in 2J and 3K happened before the apparent decline in their feeding success.

During the time of the annual groundfish surveys, most cod are believed to be migrating across the shelf toward wintering areas on the shelf edge. In the same period, capelin are moving southward along the shelf, and water temperatures on the bottom are getting warmer. The results of the trawl survey provide a sort of snapshot of cod abundance and feeding behaviour at one point in these annual cycles. But it may not be the same point in any two years. Because the extent and the timing of cod migration, capelin migration and the warming of the bottom change from year to year, it can be misleading to compare one snapshot to another.

Drawing hasty conclusions from the stomach fullness data can present a similar hazard. For example, capelin have normally left Division 2J before the annual groundfish survey reaches that area, so it is not surprising that cod stomachs are less full by the time the surveys gets to Hamilton Bank. A similar survey a few weeks earlier might find a very different picture.

In summary, there is no evidence that poor feeding had a role in the collapse of northern cod.

Food Preferences Change with Size

What cod eat seems to depend on what is available and what each fish is capable of eating. Because most fish swallow their prey whole, the relative size of predator and prey tends to regulate the choices. Up to about 30 centimetres, cod feed chiefly on small crustaceans like the various relatives of shrimp. From 30 to about 60 cm, they tend to feed on the pelagic fish which travel in schools, like capelin, sand lance and Arctic cod. Larger cod, from about 70 cm up, tend to choose larger prey, like the young of various groundfish species.

The Role of Cannibalism

Large cod will eat nearly anything of the right size, including young cod. Studies of juvenile cod show a preference for habitat the older cod tend to avoid, which probably minimizes the risk of cannibalism. Still, not all young cod avoid the adults and some do get eaten, so one of the questions researchers have examined is whether cod become more cannibalistic when capelin are scarce. This appears to be case in the Barents Sea.

Few cod smaller than 30 cm eat other cod, but the incidence of cannibalism increases as cod grow. Most of the cannibals are more than three times as long as their prey, and most of the prey are younger than age 3. The contribution of cannibalism to the diet increases with age, but does not exceed 9% even with the largest cod.

Here in Newfoundland waters, NCSP funded the analysis of the stomachs of cod sampled in research cruises since 1978. This work, aimed chiefly at a detailed understanding of cod diet, also supplied the data for an assessment of the incidence of cannibalism. The data show that cannibalism varies from one year to the next, but the rate seems to depend more on the abundance of a strong yearclass of young cod than on the scarcity of capelin. That is, when more juvenile cod are available, more will fall prey to older cod, regardless of the status of capelin. However, fewer than 1% of the cod stomachs examined in any year contained the remains of young cod.

Estimating Consumption

To estimate the feeding rate of cod in 3L, researchers relied on two sources of data. One was comprehensive details of the stomach contents of cod taken in groundfish trawl surveys. This allows analysis of the species and size of prey eaten by cod of different sizes. For a full picture of diet, however, scientists have to know not only what cod eat but how often they feed.

Norwegian scientists have developed a computer model that calculates the feeding rate of an individual cod. Given the size of the cod, the temperature of the water in which it was caught, and the kind and quantity of food in its stomach, this model can calculate its feeding rate in the days before its capture.

Given stomach content data and the Norwegian computer model, researchers can now begin to estimate the food required by any given biomass of cod. A biomass of 1 million tons, for example, which is roughly the state of the stock before its recent collapse, would consume about 2.5 million tons of other fish per year.

Ultimately the goal is to describe to flow of energy through the ecosystem of the Newfoundland Shelf, including the productivity of various prey populations and the level of consumption of prey by the major predators in the system.

**back home.

**table of contents.