The ecology of macrozoobenthos in Arhus Bay, Denmark
The aim of this thesis has been to: 1) assess the state of pollution in the two study areas and relate them to recent changes found in the Kattegat - Belt Sea area; 2) examine the observed spatial and temporal variability in species composition, abundance and biomass in Arhus Bay and the Formes area and relate the variability to antropogenic and natural causes; 3) discuss and assess methods, particularly for the estimation of secondary production and the use of multivariate analyses as methods for examining changes in macrozoobenthic communities. Macrozoobenthos were sampled at 15 stations in Arhus Bay, Denmark from 1985 to 1991 while data from Formes (reference area) included 55 sampling stations from 1986 to 1990. Monthly sampling took place at one station in Arhus Bay in 1990 and 1991. The two study areas are both situated on the eastcoast of Jutland in the Kattegat - Belt Sea area at 13-17 m depth and both receive waste water from long sea outfalls. Although both areas are Subjected to salinity stratification for most of the year, the exposed position of the Formes area on the open Kattegat coast prevents it from suffering from severe oxygen deficiencies, unlike the Arhus Bay which is a sheltered, semi-enclosed sedimentation area where oxygen concentrations in the bottom water can be very low. At Formes the sediment is sandy while it is silty in Arhus Bay. The spatial and temporal variability in the benthos in Arhus Bay could to a great extent be explained by the variation in 7 important species: Abra alba, Corbula gibba, Mysella bidentata, Nepthys hombergii, N. ciliata, Ophiura albida and Echinocardium cordatum. The fluctuations in the number and biomass of A. alba had a pronounced effect on the total abundance and biomass in Arhus Bay. The severe winter of 1986/87 with low temperatures and oxygen depletion under the ice cover practically eliminated A. alba from the bay. A. alba quickly recolonized the area and was found in high numbers in 1988. Studies of growth of A. alba in 1990 and 1991 showed that by the end of 1990 the population had reached an average length of 10 mm while the average shell length was only 5 mm by the end of 1991. The difference between the two years could be attributed to the difference in sedimentation of phytoplankton from the water column. As in other parts of the Kattegat - Belt Sea area, Arhus Bay has experienced low oxygen concentrations in the bottom water in late summer early autumn throughout the 1980s. Only the oxygen depletion under the ice cover in early spring 1987 and the local oxygen deficiencies south of the outlet in 1989 and 1990 actually killed parts of the benthic fauna. Apart from 1981, the oxygen deficiencies have thus been less severe in Arhus Bay than in other parts of the southern Kattegat in the 1980s. The number of species, abundance and biomass decreased at Fornres from 1980 to 1985 while the discharge of BOD was fairly constant during the same period. From 1986 there was a slight decrease in the discharge of BOD but a considerable increase in the number of species, abundance and biomass. At least for the second half of the 1980s there was no straightforward relation between the organic enrichment from the outlet and species composition, abundance and biomass and suggests that other factors are also important influencing the fluctuations in the benthic fauna. Estimates of total secondary community production were found to be very dependent on the method used. The method described by Brey (1990) was found acceptable for estimating secondary production in Arhus Bay but care should be exercised when comparisons are made with other areas where different methods have been used to estimate production. Secondary production was estimated more accurately for some of the abundant species in Arhus Bay on the basis of monthly samplings by the method described by Crisp (1984). Among the multivariate analyses the Detrended Correspondence Analysis (DCA) and nonmetric Multidimensional Scaling (MDS) proved to be the most successful with the Arhus Bay and Fornes data sets. Two Way INdicator SPecies ANalysis (TWINSPAN) did not work well with the Fornes data because it imposed discontinuities on data sets with continous variation in distribution of species among samples. As community types existed to a certain degree in Arhus Bay TWINSPAN worked well with these data. DCA and MDS were found to be useful techniques for analysing large data sets because they can summarize the data matrices to a manageable form and find possible patterns in the data sets. The results of the analyses can then be used as starting point for more detailed investigations of single species/samples or groups of species/samples. By using different transformations of the raw data the role of dominant or rare species can be assessed. A major 'problem in the assessment of multivariate techniques is the lack of external standards to compare with. The results of multivariate analyses must therefore be assessed critically on the basis of a careful examination of the species list combined with the knowledge and experience of the investigator. The methods used all had their advantages and limitations but each of the different methods added some important information to the picture of the benthic community in Arhus Bay and Fornes. It was thus an considerable advantage to use several different methods to analyse the spatial and temporal variability in the benthic fauna in relation to antropogenic and natural causes.