We evaluate the hypothesis that Atlantic cod larvae are passively transported by sea currents from off-shore spawning areas to settle in coastal waters, a hypothesis which has recently gained support from genetic analysis of cod in the North Sea-Skagerrak area. Such larval transport has been suggested to be an important mechanism behind the commonly observed low spatial genetic differentiation in many marine organisms. Here, we apply an ARMAX(2,2) model for juvenile abundance and use long-term monitoring data from the Skagerrak coast, constituting 54 continuous annual series from 1945 to 1997. Analysing the model, we find that the product of the size of the North Sea breeding stock and the strength of the net inflow of North Sea waters had a significant, positive effect on the abundance of coastal juvenile cod. The peak effect occurs during the month of March, just after spawning, when eggs and larvae remain pelagic and sensitive to currents. In contrast, we find no evidence of any direct effect of the North Sea spawning stock alone. Our analyses indicate that 15-20 000 0-group larvae from the North Sea reach each fjord per year, on average. This corresponds to about 1-10% of the total 0-group population in each fjord on average. These findings clearly demonstrate a direct link between larval drift and gene flow in the marine environment.

As with most teleosts, the life cycle of the cod (Gadus morhua) is characterized by high productivity of eggs leading to high concentration of larvae. These stages are known to be associated with high mortality rates. Mortality of older stages (juveniles and adults) has been relatively poorly studied. We use capture-mark-recapture (CMR) methodology to estimate time and age variation of natural mortality and fishing mortality in a coastal cod population within the Norwegian Skagerrak coast. A total of 36 728 reared 6-mo-old, and 2415 wild-caught mostly 18-mo-old, individually tagged cod were released in four consecutive years, leading to 4155 recaptures by fishermen. Tag loss and mortality due to tagging appeared negligible for reared young fish, but up to 60% of the wild-caught tagged fish disappeared just after release. Tag return by fishermen was ϳ50-60%. Natural mortality was found to be high in the 6-12 mo range, but subsequently it decreased rapidly and was no longer age-dependent. Fishing mortality was negligible for fish younger than 1-yr-old, intermediate for 1-yr-old fish, and high for older fish. The seasonal pattern of fishing varied much between age classes, with 2-yr-old fish mostly taken by recreational fishermen in summer, and older fish mostly taken by professional fishermen in winter. Because it is directed toward relatively young fish, recreational fishing may have a strong impact on the population; we specifically point out the coincidence between the increase of tourism and the decrease of the cod stock in the Risør area during the last 20 yr. Total mortality varied little between years, except during ϳ5 mo after a major algae bloom in May 1988, when fish disappeared at a high rate whatever their age (no emigration was observed in connection with the algae bloom). We conclude that mortality of adult fish may be an important determinant of the stock size.

The long-term effects of the 1988 algal bloom (Chrysochromulina polylepis Manton et Parke) along the Norwegian Skagerrak coast are evaluated and discussed on the basis of several monitoring programmes. Effects on individual growth and survival of coastal cod and its population dynamics are analysed. Cod suffered a high mortality from June until November, and the 1988 yearclass was strongly reduced. Growth was only slightly affected. Furthermore, the effects at the community level are evaluated for the coastal fish community and the benthic communities. These communities were strongly affected on a short time scale, but recovered surprisingly fast. Populations of most organisms had recovered within months, and after 1 yr few traces of the toxic bloom could be observed; after 4 to 5 yr all communities had essentially recovered. As part of the review we also discuss to what extent harmful blooms are likely to reoccur, and conclude that blooms have reoccurred and will continue to do so. However, nothing can be concluded about the toxicity of such blooms. We expect that even large perturbations are unlikely to leave any profound long-lasting effects. The effects of the 1988 bloom are discussed within a theoretical framework including stability, resilience and inertia. In conclusion we emphasise the importance of long-term monitoring data; without such data the analyses reported in this paper would have been impossible.

Beginning in the 1880s, management of marine fisheries by hatching and releasing yolk‐sac‐stage larvae was advocated in both the United States and Norway. Major cod hatchery programs were popular in both countries until the mid‐20th century, despite lack of evidence that cod abundance increases with release of hatchery‐reared fish larvae; the potential value for such management procedures was repeatedly advocated throughout the 20th century. In Norway, a beach‐seine monitoring program was begun in the early 1900s to collect data on fall abundance of 6‐mo‐old demersal fish in 21 fjords along the Norwegian Skagerrak coast and is still going on. We used these data in conjunction with hatchery data on numbers of yolk‐sac larval cod released each spring in several fjords to test for an effect of the releases on the abundance of fjord cod populations. Using both a permutation test and a statistically derived time‐series model for the cod's population dynamics, we found a slight, but statistically significant, dependence of 6‐mo‐old cod abundance on the number of yolk‐sac larvae released in four of the 16 fjords (for which we had adequate release and beach‐seine data needed for carrying out the testing). However, using the time‐series model, we did not find evidence of long‐term increases in the abundance of mature cod in any of the fjords. We discuss our findings on the basis of the literature on marine fish population enhancement programs worldwide.

As with most teleosts, the life cycle of the cod (Gadus morhua) is characterized by high productivity of eggs leading to high concentration of larvae. These stages are known to be associated with high mortality rates. Mortality of older stages (juveniles and adults) has been relatively poorly studied. We use capture-mark-recapture (CMR) methodology to estimate time and age variation of natural mortality and fishing mortality in a coastal cod population within the Norwegian Skagerrak coast. A total of 36 728 reared 6-mo-old, and 2415 wild-caught mostly 18-mo-old, individually tagged cod were released in four consecutive years, leading to 4155 recaptures by fishermen. Tag loss and mortality due to tagging appeared negligible for reared young fish, but up to 60% of the wild-caught tagged fish disappeared just after release. Tag return by fishermen was ϳ50-60%. Natural mortality was found to be high in the 6-12 mo range, but subsequently it decreased rapidly and was no longer age-dependent. Fishing mortality was negligible for fish younger than 1-yr-old, intermediate for 1-yr-old fish, and high for older fish. The seasonal pattern of fishing varied much between age classes, with 2-yr-old fish mostly taken by recreational fishermen in summer, and older fish mostly taken by professional fishermen in winter. Because it is directed toward relatively young fish, recreational fishing may have a strong impact on the population; we specifically point out the coincidence between the increase of tourism and the decrease of the cod stock in the Risør area during the last 20 yr. Total mortality varied little between years, except during ϳ5 mo after a major algae bloom in May 1988, when fish disappeared at a high rate whatever their age (no emigration was observed in connection with the algae bloom). We conclude that mortality of adult fish may be an important determinant of the stock size.