Biological interactions in fish stocks: models and reality

Three types of biological interactions; predation, competition and cannibalism are discussed in the context of marine ecosystems. Their effects on stock dynamics are considered and the empirical evidence for biological interactions reviewed. The main predictions from various population models which include some of these interactions are discussed and compared to empirical observations of stock dynamics, in order to see how well the predictions accord with observations. Although some limited evidence exists, showing that interactions - predation in particular - can be important, the fact remains that due to the high level of noise in data from marine ecosystems as well as the confounding effects of environmental varibility, it has proved difficult to find evidence demonstrating conclusively that biological interactions play a significant role in regulating stock dynamics. The purpose of large- scale multispecies models incorporating biological interactions is discussed, a range of such models for boreal ecosystems is reviewed and the necessary and desirable features which such models should have are given.