Invasion trajectory of Pacific oysters in the northern Wadden Sea.

Invasion trajectories of introduced alien species usually begin with a long establishment phase of low abundance, often followed by exponential expansion and subsequent adjustment phases. We review the first 26 years of feral Pacific oysters Crassostrea gigas around the island of Sylt in the Wadden Sea (North Sea, NE Atlantic), and reveal causal conditions for the invasion phases. Sea-based oyster farming with repeated introductions made establishment of feral oysters almost inevitable. Beds of mussels Mytilus edulis on mud flats offered firm substrate for attachment and ideal growth conditions around low tide level. C. gigas mapped on to the spatial pattern of mussel beds. During the 1990s, cold summers often hampered recruitment and abundances remained low but oyster longevity secured persistence. Since the 2000s, summers were often warmer and recruitment more regular. Young oysters attached to adult oysters and abundances of >1000 m-2 were achieved. However, peak abundance was followed by recruitment failure. The population declined and then was also struck by ice winters causing high mortality. Recovery was fast (>2000 m-2) but then recruitment failed again. We expect adjustment phase will proceed with mean abundance of about 1000 m-2 but density-dependent (e.g., diseases) and density-independent (e.g., weather anomalies) events causing strong fluctuations. With continued global warming, feral C. gigas at the current invasion fronts in British estuaries and Scandinavian fjords may show similar adjustment trajectories as observed in the northern Wadden Sea, and also other marine introductions may follow the invasion trajectory of Pacific oysters.

Parasites affect food web structure primarily through increased diversity and complexity.

Comparative research on food web structure has revealed generalities in trophic organization, produced simple models, and allowed assessment of robustness to species loss. These studies have mostly focused on free-living species. Recent research has suggested that inclusion of parasites alters structure. We assess whether such changes in network structure result from unique roles and traits of parasites or from changes to diversity and complexity. We analyzed seven highly resolved food webs that include metazoan parasite data. Our analyses show that adding parasites usually increases link density and connectance (simple measures of complexity), particularly when including concomitant links (links from predators to parasites of their prey). However, we clarify prior claims that parasites "dominate" food web links. Although parasites can be involved in a majority of links, in most cases classic predation links outnumber classic parasitism links. Regarding network structure, observed changes in degree distributions, 14 commonly studied metrics, and link probabilities are consistent with scale-dependent changes in structure associated with changes in diversity and complexity. Parasite and free-living species thus have similar effects on these aspects of structure. However, two changes point to unique roles of parasites. First, adding parasites and concomitant links strongly alters the frequency of most motifs of interactions among three taxa, reflecting parasites' roles as resources for predators of their hosts, driven by trophic intimacy with their hosts. Second, compared to free-living consumers, many parasites' feeding niches appear broader and less contiguous, which may reflect complex life cycles and small body sizes. This study provides new insights about generic versus unique impacts of parasites on food web structure, extends the generality of food web theory, gives a more rigorous framework for assessing the impact of any species on trophic organization, identifies limitations of current food web models, and provides direction for future structural and dynamical models.

An evaluation of small-scale genetic diversity and the mating system in Zostera noltii on an intertidal sandflat in the Wadden Sea.

BACKGROUND AND AIMS: The dwarf eelgrass, Zostera noltii, is a predominant inhabitant of soft-bottom intertidal regions along the coasts of northern Europe. It is a monoecious, protogynous angiosperm in which the potential for self-fertilization and inbreeding are high, especially if clone sizes exceed pollen dispersal distances. The aim of the present study was to determine the relationship between mating system and clonal structure, examine the relative roles of geitonogamous selfing and biparental inbreeding, measure pollen availability (multiple paternities) and estimate pollen dispersal. METHODS: A 100-m(2) plot was established in a large, intertidal Z. noltii meadow on the island of Sylt in the German Wadden Sea. A total of 256 adult shoots was sampled: one from the centre of 100 fixed 1-m(2) quadrats (large scale resolution) and an additional 156 from within eight randomly selected 1-m(2) sub-quadrats (small-scale resolution). DNA was extracted from seeds and leaf tissue of all samples and genotyped with nine microsatellite loci. KEY RESULTS: Mating system analysis revealed high multilocus and single locus outcrossing rates. Average pollen dispersal distance was nearly the same as the average genet (clone) size. Multiple paternity was common and 20-30 % of mature seeds originated from matings within the plot. Among inbred seeds, most resulted from geitonogamy rather than biparental inbreeding. CONCLUSIONS: Moderate disturbances intrinsic to the intertidal habitat appear to facilitate seed recruitment by gap formation. Pollen dispersal distances are sufficient to maintain outcrossing and high clonal diversity.

Distance decay of similarity among parasite communities of three marine invertebrate hosts.

The similarity in species composition between two communities generally decays as a function of increasing distance between them. Parasite communities in vertebrate definitive hosts follow this pattern but the respective relationship in intermediate invertebrate hosts of parasites with complex life cycles is unknown. In intermediate hosts, parasite communities are affected not only by the varying vagility of their definitive hosts (dispersing infective propagules) but also by the necessary coincidence of all their hosts in environmentally suitable localities. As intermediate hosts often hardly move they do not contribute to parasite dispersal. Hence, their parasite assemblages may decrease faster in similarity with increasing distance than those in highly mobile vertebrate definitive hosts. We use published field survey data to investigate distance decay of similarity in trematode communities from three prominent coastal molluscs of the Eastern North-Atlantic: the gastropods Littorina littorea and Hydrobia ulvae, and the bivalve Cerastoderma edule. We found that the similarity of trematode communities in all three hosts decayed with distance, independently of local sampling effort, and whether or not the parasites used the mollusc as first or second intermediate host in their life cycle. In H. ulvae, the halving distance (i.e. the distance that halves the similarity from its initial similarity at 1 km distance) for the trematode species using birds as definitive hosts was approximately two to three times larger than for species using fish. The initial similarities (estimated at 1 km distance) among trematode communities were relatively higher, whereas mean halving distances were lower, compared to published values for parasite communities in vertebrate hosts. We conclude that the vagility of definitive hosts accounts for a high similarity at the local scale, while the strong decay of similarity across regions is a consequence of the low probability that all necessary hosts and suitable environmental conditions coincide on a large scale.

Clonal architecture in an intertidal bed of the dwarf eelgrass Zostera noltii in the Northern Wadden Sea: persistence through extreme physical perturbation and the importance of a seed bank.

Genotypic structure and temporal dynamics of the dwarf seagrass, Zostera noltii, were studied in an intertidal meadow that has persisted since prior to 1936 near the Wadden Sea island of Sylt. Samples were collected from two 10 × 10 m plots separated by 250 m from May 2002 to June 2005 and from four 1 × 1 m plots from June 2003 to September 2004. All the samples were genotyped with nine microsatellite loci. No genotypes were shared between the plots separated by 250 m. Genetic diversity was higher in the Wadden Sea than in the other regions of its geographic range. The average clone size (genets) (SD) in the two plots was 1.38 (0.26) and 1.46 (0.4) m², respectively, with a range up to 9 m² and <20% persisted for >4 years. A high genetic and genotypic diversity was maintained by annual recruitment of seedlings despite a dramatic decrease in ramet density that coincided with the severe heat stress event of 2003. Fine-scale (1 m²) analysis suggested that extensive loss of seagrass cover precluded space competition among the genets, while a persistent seed bank prevented local extinction. Long-term persistence of Z. noltii meadows in the intertidal Wadden Sea was achieved by high genet turnover and frequent seedling recruitment from a seed bank, in contrast to the low diversity observed in large and long-living clones of Z. noltii and other seagrasses in subtidal habitats.

Spatial heterogeneity in parasite infections at different spatial scales in an intertidal bivalve.

Spatial heterogeneities in the abundance of free-living organisms as well as in infection levels of their parasites are a common phenomenon, but knowledge on parasitism in invertebrate intermediate hosts in this respect is scarce. We investigated the spatial pattern of four dominant trematode species which utilize a common intertidal bivalve, the cockle Cerastoderma edule, as second intermediate host in their life cycles. Sampling of cockles from the same cohort at 15 sites in the northern Wadden Sea (North Sea) over a distance of 50 km revealed a conspicuous spatial heterogeneity in infection levels in all four species over the total sample as well as among and within sampling sites. Whereas multiple regression analyses indicated the density of first intermediate upstream hosts to be the strongest determinant of infection levels in cockles, the situation within sites was more complex with no single strong predictor variable. However, host size was positively and host density negatively correlated with infection levels and there was an indication of differential susceptibility of cockle hosts. Small-scale differences in physical properties of the habitat in the form of residual water at low tide resulted in increased infection levels of cockles which we experimentally transferred into pools. A complex interplay of these factors may be responsible for within-site heterogeneities. At larger spatial scales, these factors may be overridden by the strong effect of upstream hosts. In contrast to first intermediate trematode hosts, there was no indication for inter-specific interactions. In other terms, the recruitment of trematodes in second intermediate hosts seems to be largely controlled by pre-settlement processes both among and within host populations.