Cyanobacteria biomass indicator

	Information from the Phytoplankton Expert Group (PEG) Authors: Norbert Wasmund1, Susanne Busch1, Sławomira Gromisz2, Susanna Hajdu3, Helena Höglander3, Andres Jaanus4, Marie Johansen5, Iveta Jurgensone6, Chatarina Karlsson7, Janina Kownacka2, Wojciech Kraśniewski8, Irina Olenina9

Key Message  

• Despite of high variability, tendencies of decreasing biomass of the nitrogen-fixing cyanobacteria (genera Aphanizomenon, Nodularia and Anabaena) are visible in the Gulf of Riga, Gulf of Finland, and Arkona Sea for the period studied (1990-2011). An increase seems to occur in the Bothnian Sea (1998-2011).  
• The cyanobacterial biomass indicator confirms the patchy distribution of cyanobacteria seen in satellite images and supports the need for data from different areas of the Baltic Sea. For example some years with low cyanobacteria biomass in the Arkona Sea and Bornholm Sea (2004, 2005) showed high biomass in the Eastern Gotland Sea.
• The Cyanobacteria biomass indicator and the widely used satellite data support each other. Whereas satellite data present spatial distribution patterns of surface accumulations, the Cyanobacteria biomass indicator delivers real biomass data (on genus level) from representative stations of each sea area.

Results and Assessment

Relevance of the indicator for describing developments in the environment

Nitrogen-fixing cyanobacteria are an important component of the ecosystem. In high abundances they can form surface accumulations that are visible even from space. According to Wasmund (1997), cyanobacteria accumulations become visible and may be considered as “blooms” at a biomass concentration of about 200 µg/L in the water. By their ability to fix molecular nitrogen they prevent severe nitrogen shortage and resulting starvation in all trophic levels of the ecosystem in the summer. However, human activity has imported a surplus of nutrients into the water for decades which turned the originally indispensable cyanobacteria into a nuisance because their nitrogen fixation counteracts the measures to reduce eutrophication, as specified in the following section. 

Policy relevance and policy references

The biomass of nitrogen-fixing cyanobacteria seems to have increased at least since the 1960s (Finni et al. 2001). If these cyanobacteria occur in large blooms, they contribute to eutrophication, oxygen depletion in deep waters and intoxication. Already the displeasing outlook of the coloured surface scum may impair the touristic use of the coasts in summer. The changes in cyanobacteria biomass and composition represent changes in the ecosystem with far-reaching consequences. Their trends are of high relevance and interest. This Baltic Sea Environment Fact Sheet serves for long-term documentation of the cyanobacteria biomass development.

The previous series of Baltic Sea Environment Fact Sheets (IFS) on a “Cyanobacteria bloom index” ended with the year 2007 (Kaitala and Hällfors 2008). Information about the spatial extension of the bloom, with help of satellite data, is available during every summer. The HELCOM Phytoplankton Expert Group (PEG) felt, however, that information on the cyanobacteria bloom development, based on biomass measurements, is still needed. For this reason, PEG started a new “Cyanobacteria biomass indicator” in 2011, which is related to the earlier “Cyanobacteria bloom index” but is not its continuation for the following reasons:

• The Cyanobacteria biomass indicator is based on samples taken on discrete stations during monitoring cruises instead of samples taken on highly frequented routes of ships of opportunity.
• The Cyanobacteria biomass indicator uses biomass data instead of an index based on semi-quantitative rankings.
• The Cyanobacteria biomass indicator separates the different sea areas instead of pooling all data of a hydrographically and biologically highly diverse region.
• The Cyanobacteria biomass indicator considers all regions of the Baltic Sea and not only the area between Helsinki and Travemünde.
• The Cyanobacteria biomass indicator uses samples covering the upper 10 m (in the Landsort Deep 20 m) instead of a single depth of approximately 5 m.

Assessment

The first IFS on the “cyanobacteria biomass indicator” appeared in 2011 and covered a period from the years 2000 to 2010. The current version, presented in 2012, is not only an update with data from 2011 but also trace data back to the year 1990. This was not possible in all regions; therefore data series have different lengths.

All quantitative phytoplankton monitoring data, available to PEG members, were included in a first analysis. Stations were pooled for sea areas (Fig. 1) in order to get representative data and fulfil the minimum requirement of at least one sampling per month during summer. This pooling included also stations which are rarely sampled (in the Bornholm Sea and the southern parts of the Eastern Gotland Basin), but which are not specified in Fig.1, in contrast to the IFS from 2011.The data were treated as explained in the technical information below. The seasonal means of the total biomass of the nitrogen-fixing filamentous cyanobacteria in the summer period (mainly June-August, in Bothnian Sea June-October) are presented in the same figure. It has to be noted that the peak values are generally higher than the seasonal means presented here. As shown in an earlier Baltic Sea Environment Fact Sheets of PEG, phytoplankton trends may be even opposite in the different sea areas (Jaanus et al. 2007). Therefore, separate diagrams for the most relevant sea areas were produced. Specific information on the three bloom-forming cyanobacteria genera Aphanizomenon, Nodularia and Anabaena is shown in Fig. 2.

Data of the Bothnian Bay are not presented because cyanobacteria biomass was very low (< 40 µg/L, with only one exception). In the Bothnian Sea, peak values could reach 500 µg/L, but the seasonal averages remained rather low. It seems that cyanobacteria biomass is increasing in this area (cf. Jaanus et al. 2007). It is of special interest that high cyanobacteria biomass was still found in autumn. Therefore, the period from June to October was considered in the Bothnian Sea.

It has to be mentioned that exceptionally late blooms occurred in the Gulf of Riga in September 1996, 1999 and 2011. We avoid, however, including data from September in the general calculation of seasonal averages because low September values occurring in “normal” years would generally reduce the average and makes it hardly comparable with other areas. The inclusion of the September data would change the seasonal mean insignificantly to 98 µg/L in 1996 and 91 µg/L in 1999. Only the value from 2011 would increase substantially to 154 µg/L.

In the other regions, blooms are noticed only from June to August (cf. seasonal pattern presented by Kaitala and Hällfors 2008). The highest blooms occurred in the Gulf of Finland with single peak values in 1998 (2900 µg/L), 1999 (3460 µg/L), 2002 (3670 µg/L), 2004 (7470 µg/L) and 2009 (4410 µg/L).

The cyanobacteria biomass at the Landsort Deep station (BMP H3) appears relatively low for methodological reasons: This was the only station where the upper 20 m were sampled in contrast to 10 m in the other regions. As cyanobacteria prefer the upper water layers, the inclusion of the lower layer of the euphotic zone reduces the depth-integrated average. The cyanobacteria biomass might be more than double, especially for Nodularia, in the 0-10 m water layer.

Data from the Kattegat and from Mecklenburg Bight were not shown because they are generally low and revealed that heavy cyanobacteria blooms did not occur. Only at the end of July 2008, a bloom with peaks of up to 400 µg/L occurred at the two Kattegat stations, but monthly and seasonal means were of course much lower. Monthly means of July 2001 and 2006 in Mecklenburg Bight reached almost 200 µg/L but were still under the cyanobacteria bloom threshold which was suggested by Wasmund (1997) at 200 µg/L. This threshold was exceeded only once (on 27.7.2006) in this area.

Because of the high variability, no clear trend can be noticed in most areas. A decrease in cyanobacteria biomass can be noticed in the Gulf of Riga and probably also in the Gulf of Finland and Arkona Sea. Trend analyses by Wasmund et al. (2011) with data from 1979 to 2005 revealed decreasing trends in summer cyanobacteria in the Bornholm Sea and Arkona Sea but not in the northern and southern parts of the Eastern Gotland Basin. Decreasing trends in the Bornholm Sea, as found by Wasmund et al. (2011), cannot be seen on this dataset for Bornholm Sea because this data series for IFS is shorter (starting 1990 instead of 1979). A visual comparison of the sea areas shows that the long-term patterns are different (Fig 1 and 2). This is not surprising as satellite images revealed already basin-wide differences in bloom distribution (Hansson and Öberg 2011). This stresses the importance to divide the Baltic Sea into all subregions which are different. 

The comparison with the “Cyanobacteria bloom index” (Kaitala and Hällfors 2008) for the overlapping period from 2000 to 2007 shows only little correspondence. The Cyanobacteria bloom index was especially high in 2007 when the Cyanobacteria biomass indicator was rather low in the areas passed by the ship of opportunity (Gulf of Finland, Eastern Gotland Basin, Bornholm Sea, Arkona Sea). The Cyanobacteria bloom index was rather low in 2004 and 2005, when also the Cyanobacteria biomass indicator was low in the Arkona Sea and Bornholm Sea. It is interesting that the Cyanobacteria biomass indicator was high in the Eastern Gotland Basin, particularly in its northern parts, at the same time. Integration over all these areas, as done by the Cyanobacteria bloom index, would widely level out these differences. The Cyanobacteria biomass indicator wants to retain these naturally occurring differences, which make them better comparable with satellite images.

References

Finni, T., Kononen, K., Olsonen, R., Wallström, K., 2001. The History of Cyanobacterial Blooms in the Baltic Sea. Ambio 30, 172-178.

Hansson, M. & Öberg, J., 2011: Cyanobacterial blooms in the Baltic Sea. HELCOM Baltic Sea Environment Fact Sheet 2011. Online. http://www.helcom.fi/BSAP_assessment/ifs/ifs2011/en_GB/Cyanobacterialblooms2011/

HELCOM Combine Manual: Annex 6: Guidelines concerning phytoplankton species composition, abundance and biomass. http://www.helcom.fi/groups/monas/CombineManual/AnnexesC/en_GB/annex6/

Jaanus, A., Andersson, A., Hajdu, S., Huseby, S., Jurgensone, I., Olenina, I., Wasmund, N. & Toming, K., 2007. Shifts in the Baltic Sea summer phytoplankton communities in 1992-2006. HELCOM Baltic Sea Environment Fact Sheet. Online. http://www.helcom.fi/BSAP_assessment/ifs/archive/ifs2007/en_GB/Phytoplankton/

Kaitala, S. & Hällfors, S., 2008: Cyanobacteria bloom index. HELCOM Baltic Sea Environment Fact Sheets 2008. Online. http://www.helcom.fi/BSAP_assessment/ifs/archive/ifs2008/en_GB/CyanobacteriaBloomIndex/

Olenina, I., Hajdu, S., Andersson, A., Edler, L., Wasmund, N., Busch, S., Göbel, J., Gromisz, S., Huseby, S., Huttunen, M., Jaanus, A., Kokkonen, P., Ledaine, I., Niemkiewicz, E., 2006. Biovolumes and size-classes of phytoplankton in the Baltic Sea. Baltic Sea Environment Proceedings 106, 144pp. Updated Biovolume Table is available at: http://www.ices.dk/env/repfor/index.asp.

Wasmund, N., 1997. Occurrence of cyanobacterial blooms in the Baltic Sea in relation to environmental conditions. Int Rev Gesamten Hydrobiol 82, 169-184.

Wasmund, N., Tuimala, J., Suikkanen, S., Vandepitte, L., Kraberg, A. 2011. Long-term trends in phytoplankton composition in the western and central Baltic Sea. Journal of Marine Systems 87: 145–159. http://dx.doi.org/10.1016/j.jmarsys.2011.03.010

Data 

Figure 1: Map of the regularly sampled stations, containing one graph on cyanobacteria biomass per area; means of the growth period for the years 1990-2011, but lacking data in some areas. 

Figure 2: Mean biomass (wet weight) of the three bloom-forming cyanobacteria genera in the different Baltic Sea areas during their blooming period (Note the different scales). “n” is total number of samples analysed for this region, “n.d.” = no data. Click images to enlarge.

Metadata

Technical information

1. Source: Estonian, German, Latvian, Lithuanian, Polish and Swedish national monitoring data (see list of authors). Main sampling locations are presented in Fig. 1. Original purpose of the data: Phytoplankton monitoring programs in the frame of HELCOM.

2. Description of data: Biomass data (wet weight in µg/L) in integrated samples (0-10 m, less at some shallower coastal stations; 0-20 m at the Landsort Deep). Genera included in index: Nodularia, Aphanizomenon and Anabaena.

3. Geographical coverage: Entire Baltic Sea (see Fig. 1).

4. Temporal coverage: Summer 1990-2011 (June-August, in the Bothnian Sea June-October). Note that the years 1992-1993 are missing from the Landsort Deep station and 1994-1996 from the Bornholm Sea area.

5. Methodology and frequency of data collection: Information based on national monitoring samples analyzed and identified by phytoplankton experts, using the mandatory HELCOM methods http://www.helcom.fi/groups/monas/CombineManual/AnnexesC/en_GB/annex6/, specified by Olenina et al. (2006). Sampling frequency was variable in dependency of the national monitoring cruises. At least one sample per month was available. This precondition could also be fulfilled by pooling nearby stations. Total number of samples is indicated in each diagram in Fig. 2.

6. Methodology of data manipulation: The precondition of at least one sample per month could be fulfilled in the representative open sea stations by combining the different national monitoring data. In coastal areas under the responsibility of only one country, many data (e.g. from Lithuania and Poland) had to be rejected because of too low sampling frequency. Other coastal data (e.g. Gulf of Finland and Gulf of Riga, see Fig. 1) could be included, leading to a high number of data (Fig. 2).

From the single data, monthly means were calculated, which served as basis for calculation of seasonal mean values.

Quality information

1. Strength and weakness (at data level): The main problem is the sampling. Samples are taken only at few stations and with a low seasonal coverage. This undersampling problem, occurring generally at ship-based sampling, is dramatic if high patchiness occurs. Especially the buoyant cyanobacteria are inhomogeneous in their horizontal and vertical distribution. The vertical inhomogeneity is tackled by the integrated sampling down to 10 or even 20 m depth. The equipment is however not designed for representative sampling of surface scums. The low sampling frequency by the few national monitoring cruises is improved by combining the different national data taken at the central HELCOM stations.

2. Reliability, accuracy, robustness, uncertainty (at data level): Data on the reliability and precision are not available. A current ring test of HELCOM-PEG, to be conducted in 2012, will give information on the precision of Nodularia countings in dependence of the counting procedure. The sampling problems are discussed above; they have natural reasons. The microscopical counting is a robust method of high accuracy. In contrast to indirect methods (satellites, pigments etc.), the objects can directly be recognized, counted and measured. Moreover, the contribution of the different species, with different environmental requirements, can be recognized and evaluated. The calculation of biomass from the counting results is highly reliable since common biovolume formulas are used (Olenina et al 2007 and its updated biovolume file: http://www.ices.dk/env/repfor/index.asp.).

3. Further work required (for data level and indicator level): The national monitoring programmes must not be shortened. In order to assure a sufficient sampling frequency, the combined efforts of different countries to sample at least the central key station in each sea area have to be maintained or even extended.

FOOTNOTES

¹ Leibniz Institute for Baltic Sea Research, Seestr. 15, D-18119 Warnemünde, Germany

² National Marine Fisheries Research Institute (NMFRI),  ul. Kołłataja 1, 81-332 Gdynia, Poland

³ Department of Systems Ecology, Stockholm University, Svante Arrhenius väg 21 A,

SE-106 91 Stockholm, Sweden.

⁴ Estonian Marine Institute, Tartu University, Mäealuse 14, 12618 Tallinn, Estonia

⁵ Swedish Meteorological and Hydrological Institute, Oceanographic Unit, Sven Källfelts gata 15, SE-426 71 Västra Frölunda, SWEDEN

⁶ Latvian Institute of Aquatic Ecology, Marine Monitoring Center Dangavgrivas str. 8 LV-1048 Riga, Latvia

⁷ Umeå Marine Science Center, Norrbyn SE-91020 Hörnefors, Sweden

⁸ Institute of Meterology and Water Management, Maritime Branch, ul. Waszyngtona 42,

81-342 Gdynia, Poland

⁹ Environmental Protection Agency, Department of Marine Research, Taikos str 26, LT-91149, Klaipeda, Lithuania

For reference purposes, please cite this Baltic Sea Environment Fact Sheet as follows:

[Author’s name(s)], [Year]. [Baltic Sea Environment Fact Sheets title]. HELCOM Baltic Sea Environment Fact Sheet(s) 2012. Online. [Date Viewed], http://www.helcom.fi/environment2/ifs/en_GB/cover/.

Last updated: 11 September 2012