DIRECTORATE GENERAL JRC
JOINT RESEARCH CENTRE
Institute of Environment and Sustainability
WFD Intercalibration Phase 2: Milestone 4 report
Water category/GIG/BQE/ horizontal activity: / Coastal and transitional watersBaltic Sea GIG
Phytoplankton
Information provided by: / Elzbieta Lysiak-Pastuszak, Nijole Remeikaite-Nikiene, Annika Grage, Iveta Jurgensone, Jakob Walve, Helena Höglander, Peter Henriksen, Andres Jaanus, Pirkko Kauppila
with preparation of earlier provided info from Member States by Wendy Bonne (JRC)
1. Organisation
1.1. Responsibilities
Indicate how the work is organised, indicating the lead country/person and the list of involved experts of every country:
Peter Henriksen / DKPirkko Kauppila / FI
Annika Grage / DE
Andres Jaanus / EE
Iveta Jurgensone / LV
Nijole Remeikaite-Nikiene, Irina Olenina, / LT
Elzbieta Lysiak-Pastuszak, Maciej Dubinski / PL
Jakob Walve, Helena Höglander / SE
1.2. Participation
Indicate which countries are participating in your group. Are there any difficulties with the participation of specific Member States? If yes, please specify:
There is currently no leader for the phytoplankton group. The problem has been discussed at several meetings but still without any solution.
1.3. Meetings
List the meetings of the group:
Date / HostNovember 2008 / Latvia
May 2009 / Sweden
September 2009 / Germany
February 2010 / Lithuania
September 2010 / Poland
March 2011 / Denmark
2. Overview of Methods to be intercalibrated
Identify for each MS the national classification method that will be intercalibrated and the status of the method
1. finalized formally agreed national method,
2. intercalibratable finalized method,
3. method under development,
4. no method developed
Member State / Method / StatusDenmark / Chlorophyll a / 1
Biovolume / 3
Community structure / 3
Algae blooms / 4
Estonia / Chlorophyll a / 1
Biovolume / 1
Community structure / 3
Algae blooms / 4
Finland / Chlorophyll a / 1
Biovolume / 3
Community structure / 3
Algae blooms / 4
Germany / Chlorophyll a / 1
Biovolume / 2
Community structure / 3
Algae blooms / 4
Latvia / Chlorophyll a / 1
Biovolume / 2
Community structure / 3
Algae blooms / 4
Lithuania / Chlorophyll a / 1
Biovolume / 2
Community structure / 3
Algae blooms / 4
Poland / Chlorophyll a / 1
Biovolume / 3
Community structure / 3
Algae blooms / 4
Sweden / Chlorophyll a / 1
Biovolume / 1
Community structure / 3
Algae blooms / 4
3. Checking of compliance of national assessment methods with the WFD requirements (April 2010 + update in October 2010)
Do all national assessment methods meet the requirements of the Water Framework Directive? (Question 1 in the IC guidance)
Do the good ecological status boundaries of the national methods comply with the WFD normative definitions? (Question 7 in the IC guidance)
3.1. Methods and required BQE parameters
In the table below it has to be indicated if all relevant parameters indicative of the biological quality element are covered (see Table 1 in the IC Guidance). A combination rule to combine parameter assessment into BQE assessment has to be defined. If parameters are missing, Member States need to demonstrate that the method is sufficiently indicative of the status of the QE as a whole.
Member State / Full BQE method / Taxonomic composition / Abundance (or cover) / Frequency and intensityof algal blooms / Biomass / Combination rule of metrics /
Germany / Yes, for W part
No, for E part / biovolume of Cyanophytes (still under develop-ment) biovolume of Chlorophytes (still under development); / Chlorophyll a (μg/l) -
total biomass (biovolume [mm3/L]) / Weighted average if taxonomic composition can be included; otherwise probably the average of total biovolume and chlorophyll a (not formally agreed)
Estonia / No / Median chlorophyll a conc. - Total median wet weight autotrophic biomass (including autotrophic ciliate Mesodinium rubrum) mg/l
(months VI-IX) / Average of
chl a and biovolume
Finland / No / Mean chlorophyll a – total biomass (mg/l) (months VII-IX) / Total biomass is not yet officially accepted as a national classification metrics. Combination rule will be average of the EQR values of chl and biovolume.
Latvia / No / Still under development and not included into assessment system / Mean chlorophyll a concentration – biovolume (mg/m3, month VI-IX) / Average of
chl a and biovolumeMean chlorophyll a concentration and total biovolume ??
Lithuania / No / Still under development and not included into assessment system / Mean Chlorophyll a – total biomass (mg/l) (months VI-IX) / No official rules for combination. Average of chlorophyll a and total biovolume is considered
Poland / No / Chlorophyll a (mean conc. of summer months (VI-IX) - total biomass, mean of summer months (VI-IX) / Mean of chlorophyll a and biomass
Sweden / No / Biomass of autotrophic and mixotrophic phytoplankton expressed as:
1. Chlorophyll a concentration (µg/L) and
2. Total biovolume (mm3/L) (if available)
June-August (mean) from at least 3 years from the latest 6-year period / Weighted classification
Average (see decscription annex 1). *
As biovolume and chlorophyll data is available, they should be cofactored into one standardised status classification for phytoplankton. If there is no data for any of these parameters, the classification is based on the remaining
Denmark / No / Under development
See text below / x / Summer (May-September) mean Chlorophyll a concentration or 90th percentile of Chl-a conc. from March through September / No combination
* Sweden: From report “Assessment criteria for coastal and transitional waters”:Cofactoring of EQRs for biovolume and chlorophyll a see annex 2
In this case all Member States have to explain why abundance and/or frequency and intensity of algal blooms have not been included in their assessment method and the efforts they made to investigate the usefulness of the parameters !!! All Member States except Germany have to explain why taxonomic composition is not included in their assessment method.
Lithuania - Phytoplankton species abundance is a difficult indicator to assess from monitoring data, as the number of species recognised in a sample highly depends on the taxonomical skills of the person analysing the sample. Moreover, the taxonomy of phytoplankton is constantly developing and the awareness of new types of species is increasing. These factors will impact the use and reliability of species abundance and diversity in the classification of coastal waters. At least, robust and unbiased indicators of the structural changes of phytoplankton communities need to be developed before phytoplankton species composition can be applied for classification of coastal waters of the Baltic Sea."(Carstensen, J., A.-S. Heiskanen, P. Kauppila, T. Neumann, G. Schernewski and Gromisz S. (2005). Developing reference conditions for phytoplankton in the Baltic coastal waters. Part II: Examples of reference conditions developed from the Baltic Sea. Ispra: Inst. Environ. and Sustainability. 35 S. (Technical report/ EU Joint Research Centre; EU 21582/EN/2)).
Latvia – the index of algal blooms is not included in the assessment method due to the lack of data. The taxonomic composition has been considered for inclusion but in later data analysis no meaningful relations between pressure indicators and the taxonomic structure were found. Thus, further development of this parameter is necessary, probably addressing the species level not just taxonomic groups. Species, proposed by Jaanus et al. 2009 (see Estonia comments) have been tested and not been useful for Latvian coastal waters.
Sweden - Efforts during IC-round 1 did not give promising results for taxonomic composition. Five-year Swedish project started in March 2011 to further explore the possibility of using taxonomic composition in the BQE method.
Estonia - Cell abundance in a phytoplankton sample depends greatly on counting strategy. Different counting units can be used for the same taxa (e.g. cells and colonies for some cyanobacteria and green algae, cells and 100 µm filaments for filamentous cyanobacteria etc). It makes the counting results highly variable (up to 2 orders of magnitude). Secondly, abundance/biomass ratio greatly depends on the physiological or developmental state of the cell (e.g. how to count dividing cells?). Thirdly, the number of cells counted depends on the skills of analyst, microscopes used and program requirements (to include picoplankton or not?). Sometimes it is difficult to discriminate between autotrophic and heterotrophic cells and bacteria (epifluorescence technique is needed).
The lack of useful taxonomy-based evaluation systems for Baltic brackish coastal areas is probably caused by the high temporal and spatial variability of hydrological and geochemical parameters. If we choose single indicator taxa, it should be taken into account that most of phytoplankton species appear in moderate or big numbers only for a relatively short period (weeks to 1-2 months). In case of sparse sampling during the assessment season, some potential indicator species can just be eluded or recorded in very low numbers (with biomasses near 0). The sensitive species are rare in abundance in comparison with the omnipotent species and are therefore less suited from the statistical point of view. The species suggested as reliable eutrophication indicators (Jaanus et al. 2009. Potential phytoplankton indicator species for monitoring Baltic coastal waters in the summer period. Hydrobiologia 629: 157-168.) – oscillatorialean cyanobacteria and the diatoms Cyclotella choctawhatcheeana and Cylindrotheca closterium have their maxima toward the end of July or in August-September. For example, C. choctawhatcheeana is among the predominant species in the NE Gulf of Riga in August-September and C. closterium has recurrently formed summer blooms in a eutrophic bay in West-Estonian coastal waters. For these two species, a preliminary assessment system has been developed and testing against available metrics (chl a and total biomass) has given satisfactory results. The indicator taxa could be included in a multimetric indice, where each attribute is calculated from the number of times that the sub-metric exceeds the threshold as a proportion of the total number of sampling times and calculated as a 5-6 year mean.
The oscillatorean cyanobacterium Planktothrix agardhii has been proposed as one more potential eutrophication indicator species in the northern part of the Baltic Sea, since it responds positively to increased TN levels (Carstensen & Heiskanen, 2007. Phytoplankton responses to nutrient status: application of a screening method to the northern Baltic Sea. Marine Ecology Progress Series 336: 29-42). This species has been found during summer throughout the 20th century, when temperature conditions were stable until the 1990s in the coastal waters surrounding the cities of Stockholm and Helsinki (Johansson & Wallström 2001. Urban impact in the history of water quality in the Stockholm Archipelago. Ambio 30: 277–281; Finni et al. 2001. The history of cyanobacterial blooms in the Baltic Sea. Ambio 30: 172–178), but also in Kuressaare Bay (Trei & Piirsoo, 1996. Short-term effect of the sewage treatment plant on the phytoplankton in Kuressaare Bay. Proceedings of the Estonian Academy of Science, Ecology 6: 154–166). These authors attributed the decrease in total biomass and change in phytoplankton dominance from P. agardhii to a more species-rich community to an effective reduction in nutrient load. And vice versa, P. agardhii has replaced Aphanizomenon flos-aquae as the most abundant cyanobacterium during the late 1980s in Neva Bay and during the 2000s in the Curonian Lagoon (Basova & Lange, 1998. Trends in late summer phytoplankton in the Neva Bay and eastern Gulf of Finland during 1978 to 1990. Memoranda Societatis pro Fauna et Flora Fennica 74 (1); Jaanus et al., 2011. Changes in phytoplankton communities along a north-south gradient in the Baltic Sea between 1990 and 2008. Boreal Environment Research 16 (Suppl. A): 191-208.). Both localities have suffered from gradual deterioration in environmental quality during the last few decades.
Finland - Finland – Explanations why phytoplankton species abundances are not a good indicator are already clarified above (see Lithuania and Estonia). The spring bloom intensity index, developed by Fleming and Kaitala (2006) for the status assessments of HELCOM, is based on high-frequency monitoring data of chlorophyll a (HELCOM 2009). The index is under further development to be possibly used in MSD assessments. Until now the spring bloom index lacks of the definition of reference conditions, of which reason it does not fulfill the requirements of the WFD. A statistical approach to define algal blooms based on chlorophyll a (Carstensen et al. 2006) was tested during the CHARM project using the intensive monitoring data of Finnish coastal waters. This approach requires at least biweekly sampling data (Carstensen et al. 2006).
Phytoplankton total biomass will be included in Finnish classification system, not only to describe other aspects of biomass than chlorophyll a, but also as a kind of proxy to phytoplankton composition. Biomasses of different phytoplankton groups, their ratios and chosen phytoplankton species were tested using Finnish data of the period 1990-2010 during the Baltic GIG work. The links of phytoplankton species and groups with nutrient concentrations were weak (R2 0.2). The results is in accordance with the study by Olli et al. (2011), where eutrophication-related parameters (total and mineral nutrients) revealed low association with the phytoplankton community composition in all Baltic Sea sub-basins (R2 0.2). Even so, there are clear evidences that changes in eutrophication level affect phytoplankton community structure. For example, in Helsinki sea area the cyanobacterial species like Anabaenopsis spp., Planktothrix agardhii and Oscillatoriales (narrow filaments) showed positive correlation with total nitrogen, phosphate phosphorus and total phosphorus (Pellikka et al. 2007). Additionally, in the northern and eastern Baltic Sea, phytoplankton composition has changed along with proceeding eutrophication (e.g. Kauppila and Lepistö 2001, Suikkanen et al. 2007).
Denmark - In Denmark the use of biovolume as a measure of phytoplankton biomass and the use of composition of phytoplankton in classification has been examined. Total biovolume of phytoplankton correlated with TN when including all stations with sufficient phytoplankton data (Figure 1). However within a single station no relationship was found and even the relationship covering all stations was characterised by so much scatter that it is difficult to establish relevant assessment classes. The relationship between individual phytoplankton classes and TN was also examined. A few classes correlated with TN but, as for total biovolume, the scatter within stations was very large and made a division into distinct classes difficult as exemplified for euglenophytes in Figure 2.