In long term monitoring programmes it is important to distinguish between natural and anthropogenic changes. In the Baltic Sea, observed long term changes of the plant and animal communities have traditionally been explained by anthropogenic influence, while the natural fluctuations due to reoccurring climatic cycles was ignored. However, in recent years also the role of large scale climatic fluctuations for species distribution is discussed.
In this study four aquatic species with different life-history and distribution were modelled, using data from the years 1994 to 2007. The data were collected annually within the Swedish national monitoring programme of the northern Baltic proper. We analysed the occurrence of the annual or pseudoannual algae species Cladophora glomerata and Ceramium tenuicorne, the perennial algae Fucus vesiculosus and the phanerogame Potamogeton pectinatus to see if there was a temporal or spatial pattern that could be explained by climate or anthropogenic influence. The climate indices the North Atlantic Oscillation (NAO) and the Baltic Sea Index (BSI) was used as environmental variables in the analysis, together with water temperature, nutrient concentrations and chlorophyll-a content. As the effect of the climate variables might not show until the next growing season or later, the species distribution data were correlated to the environmental factors with a time-lag of 0 to 3 years. To find the large-scale variable that best explained the temporal variation of each species, Akaikes Information Criterion (AIC) was used to select the best model.
The annual alga Cladophora glomerata was best modelled by the concentration in the pelagic system of total nitrogen in April the same year. The correlation between Cladophora glomerata and total nitrogen was negative in areas where the predicted probability of species occurrence was high. The phanerogam Potamogeton pectinatus was best modelled by the temperature in November two years prior to sampling. The correlation between Potamogeton pectinatus and temperature in November two years prior to sampling was negative on sandy substrates, which was also where the predicted probability of species occurrence was high. For the pseudoannual alga Ceramium tenuicorne the variation in occurrence was best explained by the chlorophyll-a concentration in September two years prior to sampling, but the results were inconclusive. The inter-annual variation of the bladder-wrack Fucus vesiculosus was not well explained by any of the parameters chosen.
Our results indicated that on the time-scale of 15 years, single environmental variables are insufficient for explaining the spatial variability of phytobenthic species. The results also indicate that the phytobenthic system is roubust to the regular climatic variations as manifested in the NAO- and the BSI-cycles, since they only to a limited extent influence the species occurrence in the area.