Empirical field studies in seagrass have revealed that overgrowth by filamentous algae which reduces seagrass growth can be explained by a top-down cascading effect caused by declines in top predators, which is enforced by eutrophication. On the Swedish west coast, 60% of the seagrass has disappeared since the 1980s. We hypothesised that overfishing, responsible for a > 90% decline in the cod stock, and the 4 to 8 times increase in nutrient load since the 1930s have altered the seagrass structure and function during recent decades. We used quantitative samples from the 1980s and 2000s and analysed the trends in abundance of the 4 feeding guilds: top predatory fish, intermediate predatory fish, crustacean omnivores and mesoherbivores. Since the 1980s, the commercial catch of gadoids on the Swedish west coast has decreased by > 90%, and here we found that the biomass of top predators (gadoids and trout) that forage in seagrass has decreased by approximately 80%. In contrast, the biomass of intermediate predatory fish (gobids and sticklebacks) has increased 8 times during summer and 11 times during autumn, while mesoherbivores (idoteids and gammarids > 7 mm) have more or less disappeared from the seagrass bed. We thus found clear evidence that a shift in seagrass food web structure has taken place over the last 3 decades. Combining these findings with our recent empirical results from field cage experiments in the Skagerrak seagrass, where we manipulate top-down and bottom-up regulation, we conclude that lack of grazers in concert with eutrophication most likely contributed to the overgrowth by filamentous algae and disappearance of the seagrass on the Swedish west coast. KEY WORDS: Food web structure · Overfishing · Eutrophication · SeagrassResale or republication not permitted without written consent of the publisher

A global trend of a warming climate may seriously affect species dependent on sea ice. We investigated the impact of climate on the Baltic ringed seals (Phoca hispida botnica), using historical and future climatological time series. Availability of suitable breeding ice is known to affect pup survival. We used detailed information on how winter temperatures affect the extent of breeding ice and a climatological model (RCA3) to project the expected effects on the Baltic ringed seal population. The population comprises of three sub-populations, and our simulations suggest that all of them will experience severely hampered growth rates during the coming 90 years. The projected 30 730 seals at the end of the twenty-first century constitutes only 16 % of the historical population size, and thus reduced ice cover alone will severely limit their growth rate. This adds burden to a species already haunted by other anthropogenic impacts.

The relative importance of recruitment and post-recruitment processes in determining adult density among marine species varies considerably between species and populations. This study aimed to investigate the relative effects of variation in recruitment and survival rates on the density dynamics of 3 spatially separated natural populations of the barnacle Semibalanus balanoides (Isle of Man, SW Ireland and west Sweden). We analysed 2 different data-based stochastic matrix models of open populations with space-limited recruitment. Assumption testing supported the application of matrix-model theory for studying and comparing population variables of open populations. Recruitment was found to be partly dependent on free space, and mortality was size-specific and independent of other vital rates. Model simulations showed that the relative importance of recruitment and survival in shaping adult density depends on the existing variation of these vital rates at each location. In the Isle of Man the variation in density was primarily caused by variability in recruitment, while in SW Ireland the variability in survival induced the most variation. In west Sweden variation was high due to variability in both recruitment and survival. A life-cycle elasticity analysis across locations indicated that the population dynamics were generally more sensitive to changes in survival than to changes in recruitment. Also, locations with high variability in vital rates seemed better able to buffer sudden changes in vital rates. In addition, differences in survival and growth produced population structures that varied between populations. In the Isle of Man and west Sweden the populations consisted of mainly Size Category 1 individuals, while the population in SW Ireland had a linear relationship between size category and the proportion of individuals.

Isolated populations or those at the edge of their distribution are usually more sensitive to changes in the environment, such as climate change. For the barnacle Semibalanus balanoides (L.), one possible effect of climate change is that unpredictable spring weather could lead to the mismatching of larval release with spring phytoplankton bloom, hence reducing the recruitment. In this paper, model simulations of a variable open population with space limited recruitment were used to investigate the effects of low and zero recruitment on population abundance in S. balanoides. Data for model parameters was taken from an isolated population in the Isle of Man, British Isles. Model simulations with observed frequencies of years with low recruitment showed only small changes in population dynamics. Increased frequencies of low recruitment had large effects on the variation in population growth rate and free space and on population structure. Furthermore, populations with intermediate to high frequencies of low recruitment appeared more sensitive to additional changes in recruitment. Exchanging low recruitment with zero recruitment severely increased the risk of local extinctions. Simulations with consecutive years of low recruitment showed a substantial increase in free space and an increase in the time taken to recover to normal densities. In conclusion, model simulations indicate that variable populations can be well buffered to changes in the demography caused by introduced environmental noise, but also, that intermediate to high frequencies of disturbance can lead to a swift change in population dynamics, which in turn, may affect the dynamics of whole communities.

No abstract