| CHAPTER IV - State of the marine Environment of the Baltic Sea Regions |
Part 5 : Kattegat and Belt Sea (incl. Mecklenburg Bight, Kiel Bight, Fehmarn Belt, Great Belt, Little Belt and Sound)
J. Albjerg1, G. Behrends, H. Giesenhagen, H.-P. Hansen, J. Wikner
The depth-integrated data from the uppermost 10 m for chlorophyll a, phytoplankton biomass (wet weight) and potential primary production from the stations R1-R7 have been pooled to represent the Kattegat, from Q1 and Q2 to represent the Sound, and data from P1 to represent the Great Belt. For potential production, pooled data from the stations M1, N1, N3 and N4 are used to represent the southern Belt Sea (Kiel Bight, Fehmarn Belt and Mecklenburg Bight).
In the Kattegat/Belt Sea area, the seasons have been defined as follows:
Spring - February to April, summer - May to August, autumn - September to November, and winter - December to January.
Very few data on phytoplankton species composition, abundance and biomass exist from the winter, and this season was therefore not treated separately for these parameters. When the variations/changes of the variables did not differ between the areas, the Kattegat has been chosen as the representative area in the figures shown.
Chlorophyll-a - The average seasonal cycle was calculated from the years 1989-93. The seasonal distribution in chlorophyll a concentrations shows a bimodal pattern with a pronounced maximum during the spring bloom in February/March, and a smaller maximum in autumn (Fig. 4.5.11). The chlorophyll a concentrations were similar in the Kattegat and Great Belt, with maximum values of about 20 mg chlorophyll a m-3, whereas they were lower in the Sound with about 10 mg chlorophyll a m-3.
The chlorophyll-a concentrations showed great interannual variations 1979-93 (Fig. 4.5.12). There was no significant trend in the values during this period, either when considering each season separately or when considering data from all seasons of each year together. Regarding the assessment periods 1979-83, 1984-88 and 1988-93, there may be a tendency to lower chlorophyll a concentrations in spring during the period 1988-93 compared to the earlier ones (Table 4.5.3). However, with only one or at the most a few samplings a month, the probability of sampling during the actual spring peak is low, and the measured concentration may thus not represent the actual peak. A tendency to higher concentrations in summer and autumn during the mid 1980s is seen.
Phytoplankton biomass and species composition - When the phytoplankton biomass is expressed as wet weight, the importance of diatoms is exaggerated due to the large vacuoles in the diatom cells, contributing to the wet weight but not to the carbon biomass. Data from this region consisted of both wet weight and carbon biomass. In order to compare the phytoplankton levels with other areas in the Baltic Sea, wet weight has been used in all figures. Trend analyses were, however, performed on carbon biomass.
The seasonal distribution of the phytoplankton-carbon biomass is characterised by a spring bloom dominated by diatoms, a summer maximum with diatoms and dinoflagellates, contributing equally, and an autumn bloom, mostly consisting of dinoflagellates. Regarding wet weight, the diatoms also dominate the phytoplankton in the autumn (Fig. 4.5.13).
In the summer, the phytoplankon wet weight biomasses reach the same level of about 3,000 mg m-3 in the Kattegat as in the Great Belt. The summer maximum in the Sound is only 1,700 mg m-3. The long-term record of the phytoplankton biomass revealed great inter-annual variations. For example, the maximum values in the Kattegat ranged from >24,000 mg m-3 in 1984 to <4,000 mg m-3 in 1989 (Fig. 4.5.12).
The phytoplankton biomass shows a significant negative trend from 1979 to 1993 in the Kattegat, the Great Belt and the Sound (p=0.008, 0.007 and 0.004, respectively). The negative trend is particularly evident for summer values in the Kattegat (p=0.013), but is also significant in spring and autumn (p=0.025 and 0.054, respectively; Fig. 4.5.14). This is consistent with observations from the Kiel Bight and Fehmarn Belt, where a slight insignificant decrease in both chlorophyll a and phytoplankton biomass was observed from 1986 to 1995. However, the decline was mainly due to rather low phytoplankton standing stocks in 1993, 1994 and 1995 (see REF : 257).
Table 4.5.4 shows the five dominating phytoplankton species or groups in different seasons (spring, summer, autumn) and time periods (1979-83, 1984-88, 1989-93) for the Kattegat, the Sound and the Belt Sea. Dominance due to one or a few mass occurrences has been disregarded. In the Kattegat, this applies to incidents of Gomphosphaeria sp. and Noctiluca scintillans in the periods 1979-83 and 1989-93, respectively. In the Sound, Coscinodiscus sp. have been disregarded in the autumn plankton for the period 1984-88.
Potential productivity - The average seasonal cycle of the potential production was calculated from the years 1989-93. As for chlorophyll a, the seasonal distribution of the potential productivity shows a bimodal pattern, with a pronounced maximum in March and a smaller maximum in autumn (Fig. 4.5.11). The level of the potential productivity was almost the same for the Kattegat, the Great Belt and the southern Belt Sea, with about 40-50 mg C m-3 h.-1 in the spring, up to 10 mg C m-3 h.-1 in the summer and 20-30 mg C m-3 h.-1 in the autumn. In the Sound, the spring peak was lower with about 13 mg C m-3 h.-1.
The potential productivity showed great inter-annual variations from 1979 to 1993. For example, in the Kattegat, the maximum values ranged from 84 mg C m-3 h.-1 in 1980 to 12 mg C m-3 h.-1 in 1991. In the Kattegat, there was a significant negative trend in the potential productivity during this period (p=0.05; Fig. 4.5.15). In the Great Belt and in the Sound, there was also a tendency to lower values, although the trends were not significant (p=0.06 and 0.07, respectively). Analyses of the pooled data from the southern Belt Sea did not indicate any change in the potential productivity from 1979 to 1993. However, by separately analysing the potential production from the period 1985/86-95 for different stations and sampling depths in the Kiel Bight and Fehmarn Belt, tendencies of decreasing trends were visible (see REF : 257). Regarding the periods 1979-83, 1984-88 and 1989-93, there was a tendency of decreasing potential productivity in spring.
Depth-integrated zooplankton samples were collected until 1986 by a modified WP-2 net with 100 µm mesh size (see REF : 670) according to the HELCOM BMP Guidelines. After 1986, the Danish zooplankton samples were taken by a submersible pump (400 dm3 min.-1) fitted with a 100 µm net (see REF : 464). Sweden and Germany used the WP-2 net throughout the whole period. Although the WP-2 net may underestimate the zooplankton due to clogging of the net, this is not the case with the pump (see REF : 464). However, the appendicularian Oikopleura dioica is probably damaged by the pump (see REF : 464), and thus underestimated in the Danish samples since 1986.
Kattegat, Sound and Great Belt - The zooplankton data from the stations R1 to R4 have been pooled to represent the Kattegat, Q1 and Q2 to represent the Sound and data from P1 to represent the Great Belt. The seasons have been defined as follows:
Spring - February to March, summer - May to August, and autumn - September to Novem-ber.
There was no zooplankton sampling in the winter from this area. Data exist from 1982 to 1993 for the Kattegat, and from 1980 to 1993 for the Sound and the Great Belt. The southern Belt Sea was treated separately, as the seasonal cycle seems different and the data material are more extensive.
The average year, constructed from all data for the period 1989-93, shows an unimodal seasonal distribution, peaking in June. Copepods dominate the zooplankton throughout the season. Their seasonal variation is more pronounced in terms of biomass than in terms of abundance (Fig. 4.5.16). The most important copepods in terms of biomass are Pseudocalanus spp. and Centropages spp. Acartia spp. also contribute significantly to the copepod biomass in the early season, thereafter Oithona spp. take over in the second half of the year.
Copepod biomass and zooplankton abundance reached the same level in the Kattegat as in the Great Belt, of about 50 mg C m-3 and 25,000 ind. m-3 in summer, whereas the summer maximum in the Sound was lower, i.e., about 30 mg C m-3 and 16,000 ind. m-3. The long-term record of the components of the zooplankton community revealed great interannual variations (Fig. 4.5.17). For example, the maximum values for copepod abundance in the Kattegat ranged from 138,000 ind. m-3 in 1984 to 26,000 ind. m-3 in 1987.
The zooplankton abundances showed a significant negative trend in the Kattegat and Sound (p=0.033 and 0.011, respectively). The negative trend was most pronounced in the Sound in autumn (Fig. 4.5.18), and was due to a decline in copepod abundance and biomass. In terms of biomass, the negative trend was significant in the Kattegat, the Sound and in the Great Belt (p=0.006, 0.005 and 0.009, respectively).
Southern Belt Sea (Kiel Bight, Fehmarn Belt and Mecklenburg Bight) - For the total zooplankton analyses, the following stations were pooled:
Kiel Bight - Boknis Eck, N3 and N4; Fehmarn Belt - N1; Mecklenburg Bight - M2 and M1.
The detailed evaluations on species level were restricted to Kiel Bight data from 1985-93. The data were splitted as follows:
January to March - winter (low abundance), April to June - spring (maximum), July to September - summer (minimum), and October to December - autumn (peak followed by minimum).
In the Kiel Bight, the annual mean abundance of the mesozooplankton community varied between 12,470 ind. m-3 (1986) and 73,561 ind. m-3 (1989), and in the Mecklenburg Bight between 7,615 ind. m-3 (1982) and 38,606 ind. m-3 (1989). The highest intra-annual variability was observed at the coastal station Boknis Eck (Fig. 4.5.19a,b,c). The seasonality of the total mesozooplankton community is shown in Figure 4.5.20.
The mesozooplankton community of the Kiel and Mecklenburg Bights was clearly dominated by copepods (Fig. 4.5.21). Regarding the abundances, in certain months other groups can occur in high densities. These are often either rotifers (April/May) or meroplankters, especially bivalve larvae (June). Among the copepods, Oithona similis is the most abundant species in the area. In spring and summer, Centropages hamatus and Acartia spp. can be more important in some years. The second rank is taken by Pseudo- and Paracalanus spp. in winter and autumn. Temora longicornis is a less abundant copepod species in this area. However, there is a strong interannual variability of the percentage of the species composition.
The water inflow event in spring 1993 did not cause any recognisable reaction in the meso-zooplankton community in this area. Most probably, the transition area is not as much influenced by these events as more central parts of the Baltic Sea.
Comparing the monitoring periods 1984-88 and 1989-93, it is obvious, that the earlier period was characterised by very low abundances, while the latter showed several years with high abundances. Figures 4.5.19d,e,f show the deviations of the total mesozooplankton abundances from the long-term monthly median values. The differences between mesozooplankton-rich and -poor years are more pronounced at the Kiel Bight stations than in the Mecklenburg Bight, but such differences are not shown at the station Fehmarn Belt. This might be due to the fast currents at this station, which lead to an adjustment of the differences, while in the shallower and more stable coastal stations, differences can be more developed.
In spring and summer, there was an increase of mesozooplankton abundances from 1985/86 to 1989/91. Afterwards, the standing stocks decreased (Fig. 4.5.19, Fig. 4.5.20). This difference in abundances was caused mainly by the copepods Oithona similis and Pseudocalanus minutus elongatus, as well as by the meroplanktonic mollusc larvae, and in spring in the Mecklenburg Bight by rotifers. Both copepod species showed a minimum in 1986. Thereafter, the concentrations increased until 1992. In 1993, the abundances decreased again. The larvae of molluscs showed a low concentration during the period 1985-87. From 1988 onwards, the abundances doubled or even increased threefold.
An important factor seems to be the winter temperature. The years 1985-87 were characterised by low winter temperatures. The stock of mesozooplankton was extremely low in these years. The warmer period during 1989-92 showed higher amounts of mesozooplankton. This was particularly true for Oithona similis and Pseudocalanus minutus elongatus, which both survive the winter as copepodide stages. However, the concentration of meso-zooplankton was again lower in 1993, although this year was not very cold. Consequently, the statistical analyses did not show a significant correlation between temperature and mesozooplankton abundance.
An increase was observed in Acartia tonsa. This species is relatively new in the Baltic Sea, in that it migrated in the 1930s (see REF : 106). As it is a warm water species, it had favourable conditions during the last assessment period. Since 1990, the jellyfish, especially the Aurelia aurita populations, have been monitored in the Kiel Bight together with ichthyoplankton. To assess the influences of predation, the zooplankton data were evaluated, and showed a significant inverse relationship between jellyfish and mesozooplankton abundances (see REF : 60). The ichthyoplankton stocks seem also to be influenced by the jellyfish stocks, but it is not yet clear, whether this is a direct effect or through diminishing of the food source meso-zooplankton.
Bacterioplankton has been monitored in the Kiel Bight, Fehmarn Belt and Mecklenburg Bight at monthly intervals at 4 stations, Boknis Eck, N3, N1 and M2, which represent a gradient from the more eutrophic fjords to offshore areas (see REF : 177). Only data from Boknis Eck and Fehmarn Belt (N1) will be presented here to allow a direct comparison of integrated values at equal water depths. They are regarded as representative of the general temporal and spatial developments in the area.
The biomass of bacterioplankton (Fig. 4.5.22, Fig. 4.5.23) declined in a linear manner during 1988-95 at both Boknis Eck and Fehmarn Belt. This could be shown by non-parametric Mann-Kendall analysis (p=0.003 and 0.01, respectively), and also by linear regression on the integrated annual average biomass (p=0.001 and 0.005, respectively). The latter analysis indicated a decline rate of 8.2 and 7.5 mmol C m-2 yr-1, respectively, at these stations. In contrast, the regression analysis on integrated annual average bacterial numbers did not show any significant trend for the whole investigation period at either station. Instead, there was a significant increase in bacterial abundance during the years 1990-95 at Boknis Eck and Fehmarn Belt (p= 0.01 and 0.03, respectively, by linear regression; Fig. 4.5.24), the opposite of the development in bacterial biomass. This lack of coherence between the two variables indicates the difficulty of accurate cell volume measurements. Usually, biomass and cell number show similar trends (see REF : 371), since the biomass calculation is based on cell counts and individual cell volumes, of which the latter is usually less variable. Discrepancies may arise from the fact that the reproducibility of cell-number estimates between different personnel are within 10 % variation, while biovolume estimates often show variations of about 50 % (see REF : 277). Since there was a change in counting personnel between 1991 and 1992, artifacts with respect to cell volume measurements cannot be excluded. Lower and more stable biomass values after 1991 support this assumption. An independent study at station Boknis Eck from August 1988 to April 1990 with a comparable sampling schedule presented an integrated annual average biomass value of 58 mol C m-2 yr-1 in 1989, and an average biovolume of about 0.045 µm3 (see REF : 176). The combination of this biomass value with data from the monitoring period 1992-95 revealed no significant trend during the time of investigation. This is in accordance with the results from the Bothnian Bay (cf. Chapter 4.1).
Because of possible overestimation of the bacterial cell volume during the first four years of monitoring, bacterial production values for this period were calculated by assuming a constant biovolume of 0.045 µm3, a carbon conversion factor of 0.35 pg C µm-3, and a conversion factor from thymidine incorporation to cell production of 1.1 1018 cells per mole of thymidine. The data from 1992-95 were calculated with the actual determined biovolumes.
No clear trend in bacterial growth (Fig. 4.5.23, Fig. 4.5.25) could be demonstrated at station Fehmarn Belt, whereas for station Boknis Eck a decline rate in growth of 0.18 mol C m-2 yr-1 (p=0.02) was obtained. The average annual growth of bacterioplankton during the whole period at Boknis Eck and Fehmarn Belt corresponded to 2.83 and 2.61 mol C m-2 yr-1, respectively (Table 4.5.5). The production values for station Boknis Eck in 1989 are in good agreement with those from the independent study (see REF : 176), which justifies the applied correction and supports the slightly decreasing trend. Nonetheless, there was an increase in the thymidine incorporation rates (data not shown), which appeared in the same period when the cell numbers increased.
The specific bacterial growth rate, i.e., the P/B ratio, did not show any long-term development (Fig. 4.5.26). The average growth rate was 0.13± 0.04 day-1 at both stations. But, the likely overestimation of bacterioplankton biomass during the first four years would imply an underestimation of the specific growth rate, and indicates a possible decline of this parameter for the whole investigation period and a stabilisation of the specific growth rate during the last four years. The corrected scenario shows stable bacterioplankton biomass and production, while the cell numbers and the thymidine incorporation rates increased, and the specific growth rate decreased, during 1988-95.
The increase in bacterial abundance and thymidine incorporation rates at both stations, without a corresponding increase in biomass, has several possible explanations:
Changes in micro-nutrient concentrations, e.g., by decreasing phosphorus loads, and composition, i.e., by increased N/P ratios, and changes in the grazing pressure may result in the occurrence of strong substrate control of the bacterioplankton without coupling between phytoplankton and bacteria. Although a linear relationship between phytoplankton carbon and bacterioplankton biomass at the two stations is lacking, both for the 1988-91 period with uncertain bacterial biomass data and for 1992-95 (Fig. 4.5.27), an increasing tendency towards bottom-up control of the bacterial community during the summer seasons of the last four years can be demonstrated at station Boknis Eck (Fig. 4.5.28). The linearity of values points towards bottom-up control, whereas data clouds may indicate a grazing-control of the bacterial community (see REF : 84).
The uncoupling between phytoplankton and bacterioplankton parameters indicates that the observed substrate control at both stations is not closely connected to the primary producers. The exceptional warm summers in 1992, 1994 and 1995, and the low precipitation, especially during 1994/95, provide evidence that a possible decline in allochthonous inputs via precipitation and run-off from land ‘dried out’ the bacterial community. Support for this hypothesis may be derived from increasing glucose-turnover rates during 1990-94 and low values in 1995 (Fig. 4.5.29). During 1995, the summer temperatures were similar in height and duration compared to 1992 and 1994, but precipitation was almost absent. Long stagnation periods presumably changed the general nutritional state, and in turn modified the interactions between highly active organisms of the microbial food web.
Although pronounced changes in the system can be detected, which point towards increasing bottom-up control of the bacterioplankton, the lack of coupling between phytoplankton and bacteria identifies the Kiel/Mecklenburg Bight area as still eutrophic. The increase of colony-forming units during 1990-95 (data not shown) implies a favourable supply of organic substrates, and therefore leaves increasing water temperatures and competition for inorganic nutrients as the most likely factors to explain the observed changes.