Annex 6: Guidelines concerning phytoplankton species composition, abundance and biomass

1. Introduction
2. Sampling
2.1 Preservation and storage of samples
2.2 Qualitative determinations
2.3 Quantitative determinations (phytoplankton counting)
3. Biomass determinations
3.1 Introduction
3.2 Biovolume calculation
3.3 Carbon content calculation
4. Semi-quantitative analysis of phytoplankton samples
4.1 Introduction
4.2 Counting procedure
5. References

1. Introduction

Phytoplankton serves as an indicator of the effects of eutrophication. Nutrient enrichment may give rise to shifts in phytoplankton species composition and biomass. Furthermore, an increase in the frequency, magnitude and duration of harmful phytoplankton blooms may occur.

Analysis of phytoplankton species composition, abundance and biomass is carried out for the following purposes:

• to describe temporal trends in phytoplankton species composition, their abundance, biomass and intensity of abundance of blooms

• to describe spatial distribution of phytoplankton species, their abundance, biomass and blooms

• to identify key phytoplankton species (e.g. dominating, harmful and indicator species)

2. Sampling

Phytoplankton species composition, their abundance and biomass in the euphotic zone form the basis for the determination of temporal trends of phytoplankton. High frequency sampling at a number of stations covering all basins in the Baltic Sea area is needed to reveal trends. As phytoplankton shows a substantial seasonal variation, sampling needs to cover the entire growth season, which in parts of the Baltic Sea extends over the entire year. In addition to the sampling at fixed sampling stations, ships-of-opportunity transects, satellite images and aerial surveillance help to identify variability in temporal and spatial extent of phytoplankton. Synoptic surveys are necessary for the study of the extension of phytoplankton blooms.

For the purpose of quantitative studies in the open sea, the minimum requirement is to take an integrated sample from 0-10 m depth using a hose (Lindahl, 1986). In Coastal Monitoring Programme (CMP), the sample from 0-1 m or an integrated sample (0-10 m) could be analysed.

The same integrated sample should be used for chlorophyll determination and, if desired, primary production.

An additional sample, 10 – 20 m, is recommended.

The 0 – 10 m integrated sample may be approximated by pooling equal amounts of water from the depths 0 – 1 m, 2.5 m, 5 m, 7.5 m and 10 m.

The integrated sample should be thoroughly mixed in a bucket. One subsample of 200 cm³ is drawn from the well-mixed sample for quantitative phytoplankton counts.

For ship-of-opportunity and helicopter sampling a single sample from the mixed surface layer can be taken.

If there is a subsurface chlorophyll a maximum an additional sample may be taken at this depth, using a water sampler. 200 cm³ of this sample is drawn for at least qualitative analysis of phytoplankton species composition. Another subsample is taken for chlorophyll a determination.

It is recommended to take a net sample from the 0 – 20 m water column in order to obtain a concentrated plankton sample. This sample serves as a support for species identification. Observing living material facilitates identification. A plankton net with a 10 µm mesh-size is recommended. In case of higher concentration it is advisable to use a net with 25 µm mesh-size.

2.1 Preservation and storage of samples

2.1.1 Preservatives

Acid Lugol's solution (Willén 1962):

200 cm³            distilled or deionized water
20 g  potassium iodide (KI)
10 g  resublimated iodine (I₂)
20 cm³              glacial acetic acid (conc. CH₃COOH)

Mix in the order listed. Make sure the previous ingredient has dissolved completely before adding the next. Store in a tightly locked glass bottle cooled.

Alkaline Lugol’s solution (modified after Utermöhl 1958):

Replace the acetic acid of the acid solution by 50 g sodium acetate (CH₃COONa). Use a small part of the water to dissolve the acetate.

Neutralized formaldehyde gives incomparable results to Lugol’s solution and should not be used, except at a few coastal stations where long time series are already established using formaldehyde.

2.1.2 Preservation

Net samples to be studied alive can be kept fresh for a few hours in an open container in a refrigerator.

For preservation of water samples, 0.25 – 0.5 cm³/ 100 cm³ sample of acid Lugol's solution have to be added immediately. If coccolithophorids need to be preserved with the coccoliths intact, a parallel subsample should be fixed with 0.25 – 0.5 cm³ alkaline Lugol’s solution / 100cm³ sample. Clear, colourless iodine-proof bottles with tightly fitting screw caps should be used for iodine-preserved material. With such bottles it is easy to see when the iodine becomes depleted and more preservative needs to be added.

They should be stored dark and cool and counted as soon as possible, and within a year. Samples stored for more than one year are of little use.

2.2 Qualitative determinations

Net samples are recommended to be studied for the identification of sparsely occurring microplankton species with a standard research microscope. The advantages include potentially higher resolution, thinner preparations and the possibility to turn the cells around by tapping the cover glass. This is especially helpful when examining the plate structure of dinoflagellates. Dinoflagellate plates are also well studied using the epifluorescence method with Calcofluor (Andersen & Throndsen, 2003).

2.3 Quantitative determinations (phytoplankton counting)

The recommendation is based on the counting technique with an inverted microscope as described by Utermöhl (1958).

2.3.1 Settling procedure

Before sedimentation the sample should be adapted to room temperature to avoid excessive formation of gas bubbles in the sedimentation chambers. Gas bubbles will adversely affect sedimentation, the distribution of cells in the bottom-plate chamber, and microscopy.

Immediately before the sample is poured into the sedimentation chamber, the bottles should be shaken firmly but gently in irregular jerks to homogenize the contents. Too violent shaking will produce a lot of small bubbles, which may be difficult to eliminate. A rule of thumb is to gently turn the bottle upside-down at least 50 times. If the sample must be shaken vigorously in order to disperse tenacious clumps, this should not be done later than one hour before starting sedimentation.

The chambers should be placed on a horizontal surface and should not be exposed to temperature changes or direct sunlight. Covering the settling chamber(s) with an overturned plastic box will provide a fairly safe and uniform environment for sedimentation. If moistened tissue paper is included under the hood, problems caused by evaporation will be reduced considerably. It is essential that the supporting surface is vibration free, since vibrations will cause the cells to collect in ridges.

Settling time is dependent on the height of the chamber and the preservative (e.g. Hasle, 1978 and Rott, 1981). The times given in Table 1 are recommended as minimum. If vibration is a problem, the minimum time should not be significantly exceeded. Otherwise it is suggested that counting be performed within four days. Sedimented samples not counted within a week should be discarded. Separated bottom chambers not counted immediately should be kept in an atmosphere saturated with humidity.

Table 1: Settling time in phytoplankton samples, preserved with Lugol solution, in dependence of the volume of the sedimentation chamber

Sedimentation chambers of 100 cm³ (height 20 cm and settling time 48 h for Lugol) should be used with caution since convection currents are reported to interfere with the settling of plankton in chambers taller than five times their diameter (Nauwerck, 1963 and Hasle, 1978). Such chambers can be used only when phytoplankton is very sparse, as in late autumn and winter. For such samples it is recommended scanning the whole chamber bottom.

If the cells are too strongly stained by iodine for comfortable identification, surplus iodine can be chemically reduced to iodide by dissolving a small amount of sodium thiosulphate (Na₂S₂0₃ . 5 H₂0) in the aliquot to be sedimented.

2.3.2 Counting procedure

In order to save time and to achieve a reasonable accuracy in counting, the sedimented sample should first be examined for general distribution of cells on the chamber bottom, and the abundance and size distribution of the organisms. The settled sample should be discarded if the distribution is visually uneven, one-sided or in ridges, indicating convection or vibration, respectively. If this occurs consistently, measures should be taken to eliminate the sources of disturbance.

Counting begins at the lowest magnification, followed by analysis at successively higher magnification. For the sake of adequate comparison between samples, regions and seasons, it is important to always count the specific species at the same magnification. In special situations, such as bloom conditions, however, this may not be possible. Large species, easy to identify (e.g. Ceratium spp.), which are usually also relatively sparse are counted at the lowest magnification and over the entire chamber bottom. Smaller species are counted at higher magnification and possibly on only a part of the chamber bottom.

Small microplankton species can preferably be counted together with the nanoplankton when they occur in abundance, or they can be counted using an objective with intermediate magnification, 20 – 25x. A grid of 5 x 5 squares in one of the oculars is very helpful when counting dense fields of small cells.

The recommended magnifications for different phytoplankton sizes are listed in Table 2.

Table 2: Recommended magnification for counting of different size classes of phytoplankton (From Edler and Elbrächter, in prep.).

  * picoplankton cannot be properly analysed using the Utermöhl method

Counting the whole chamber bottom is done by traversing back and forth across the chamber bottom. The parallel eyepiece threads delimit the transect, where the phytoplankton are counted (Fig. 1.). Phytoplankton cells crossing the upper thread are counted, but not those crossing the lower thread.

Figure 1. Traversing the whole chamber bottom with the parallel eyepiece threads to indicate the counted area. (From Edler and Elbrächter, in prep.)

Counting part of the chamber bottom can be done in different ways. If half the chamber bottom shall be analysed every second transect of the whole chamber counting method is counted. If a smaller part shall be analysed, one, two, three or more diameter transects are counted. After each transect is counted the chamber is rotated 30-45^o.

How much of the chamber area should be counted and the magnification to be used is dependent on the size of the organisms and their abundance, and on the kind of counting units used. The common counting unit is the cell. This applies also to colonies with irregular numbers of cells. Estimation of cell numbers in small-celled and densely-packed colonies may be realised by visual dividing of the colony into smaller areas, counting cell numbers in one area and multiplying with the number of “small areas”.

Colonial algae which occur regularly as groups of four cells or a multiple are most conveniently counted and reported as colonies, e.g.:

Choricystis

Crucigenia

Crucigeniella

Desmodesmus

Dictyosphaerium

Elekatothrix

Gonium

Merismopedia

Pandorina

Pediastrum

Scenedesmus

Tetrastrum

Willea

Filamentous cyanobacteria are to be counted in lengths of 100 µm. Numbers of 100 µm pieces per litre are reported. Diatom with any plasma inside the cell should be counted as a living cell.

While counting, the species/individuals have to be allocated to size classes according to the scheme of Olenina et al. (2006). This is important for a reliable biovolume calculation.

At least 50 counting units of each dominating taxon should be counted, and the total count should exceed 500. All cells should be counted and reported even if fewer counted units progressively will decrease the precision of the count and increase the statistical error of the population estimate. The approximate 95 % confidence limits of a selected number of counted units are given in Table 3. They have been calculated according to the formula:
 

where n is the number of units counted. Actually the error is not symmetrical, but increasingly asymmetrical with lower counts. Thus, for four units counted the theoretical limits are -73 to +156 % (Lund et al., 1958, Kozova and Melnik, 1978).

Table 3: The approximate 95 % confidence limits of a selected number of counted units.

It should be recognized that these are not maximum errors. The statistics assume perfectly random-distribution of cells on the bottom of the sedimentation chamber, a condition which is probably never realized. The several subsampling steps involved also tend to increase the variance (cf. Venrick, 1978a; Venrick, 1978b). In order to check the precision of the method it is recommended to count one dominating species in low and one in high magnification in a new subsample in every 20^th sample.

With species for which the counting unit is smaller than the individual, e.g. some colonial forms, chain forming diatoms, and filamentous species with average filament length in excess of 100 µm, the distribution of the counting units will be aggregated even in perfectly sedimented samples. The variance will be higher, and the precision accordingly lower. If it is necessary to keep the error within the same limits as for "randomly" distributed units, the number of counted units should be increased in the ratio average size of individual/size of counting unit.

The number of counting units per dm³ sea water is calculated by multiplying the number of units counted with the coefficient C, which is obtained from the following formulas:

where:

A = cross-section area of the top cylinder of the combined sedimentation chamber

the usual inner diameter is 25.0 mm, giving A = 491 mm² (the inner diameter of the bottom-plate being irrelevant)

N = number of counted fields or transects

a₁ = area of single field or transect

a₂ = total counted area

V = volume (cm³) of sedimented aliquot

Reliable quantitative counting of the picoplankton fraction requires fluorescence microscopy.

When counting phytoplankton in a sedimentation chamber, it is suitable to count protozooplankton (e.g. ciliates and colourless flagellates). This recommendation is also valid for these forms. However, it must be stressed that the protozooplankton are a separate group and must not be mixed with the phytoplankton. Thus, they must not be included in abundance or biomass values of phytoplankton. The exceptions are the autotrophic ciliates Mesodinium rubrum and the genus Laboea that should be counted and included in abundance and biomass values of phytoplankton.

While counting, the species/individuals have to be allocated to size classes according to the scheme of Olenina et al. (2006) and the latest update of its appendix.

2.3.3 Cleaning of the sedimentation chambers

After use no part of the combined sedimentation chamber should be allowed to dry out before it is carefully cleaned. Dried phytoplankton or formalin preservative may be quite difficult to remove. The separate parts are first rinsed under running tap water, and then soaked for a few minutes in Lukewarm water with some nonabrasive detergent added, thereafter cleaned with a soft brush or soft tissue paper, and rinsed with tap water. The sedimentation chamber may also be cleaned with 95% ethanol. Finally, they are given a rinse with deionised or distilled water, and are put away to dry. Special care should be taken not to scratch either end of the top cylinder and the entire upper surface of the bottom plate.

3. Biomass determinations

3.1 Introduction

Biomass data are a much better descriptor of phytoplankton than abundance, especially because the latter is strongly influenced by the highly abundant picoplankton and small nanoplankton, which can be analysed only with limited certainty. Thus, biomass data are preferred for characterising spatial and temporal phytoplankton patterns and modelling.   Depending on the purpose of the investigation, phytoplankton biomass can be expressed as cell volume (or weight) or carbon. The transformations to cell volume are based on measurements of the size of the species and the adaptation of the shapes to geometrical figures. The mandatory geometric formulas, size groups and the resulting biovolumes per counting unit are compiled in the paper of Olenina et al. (2006)and its updated appendix. In a further step, the carbon content should be calculated because organic carbon is the universal component of organisms and is the energy source transported along the food chain.

3.2 Biovolume calculation

As specified in Section C.4.3. 2.4, the species/individuals have to be allocated to size classes according to the scheme of Olenina et al. (2006) during the counting process. The individual biovolumes of the different counting units have to be multiplied with their abundance to get the biovolume per dm³.

Biovolume _taxon [mm³ dm^-3] = abundance [dm^-3] x VCU x 10^-9

VCU                 = volume of counting unit (in µm³)

From the bio volume data, the biomass (wet weight) can simply be derived by a rough assumption of a plasma density of 1 g cm^-3.

3.3 Carbon content calculation

In the previous guidelines (HELCOM 1988) it was recommended to calculate the carbon content from the plasma volume by a constant factor. Since the calculation of the plasma volume of diatoms bears a lot of uncertainties and, moreover, the conversion factor is not constant in reality, the calculation of carbon has been suspended for some years.

New formulas by Menden-Deuer and Lessard (2000) take the decrease in specific carbon content with cell size into account and calculate the carbon content of diatoms directly from the cellular biovolume without plasmavolume calculation.

The general formula for phytoplankton is:

          Carbon [pg C cell^-1] = 0.216 x CV ^0.939

Diatoms require a particular formula because of their lower specific carbon content:

          Carbon [pg C cell^-1] = 0.288 x CV^0.811

If cell aggregates are the counting unit (CU), their carbon content has to be calculated via the cells carbon by the following formulas. It has to be differentiated between counting of cell packages (e.g. 100 cells of Microcystis as a CU) and filaments (e.g. 100 µm of Nodularia as a CU). In filaments, the cell length has to be known.

For multi-cell colonies:

Carbon [pg C CU^-1] = 0.216 x CPU x (VCU/CPU) ^0.939

For filaments:

Carbon [pg C CU^-1] = 0.216 x LCU/CL x (VCU*CL/LCU) ^0.939

CU    = counting unit

VCU                 = volume of counting unit (in µm³)

          CPU        = number of cells per counting unit

          CL           = cell length (in µm)

          LCU        = length of counting unit (mostly 100 µm)

The calculation of the carbon content is non-obligatory, but if executed it has to be done according to the given formulas.

4. Semi-quantitative analysis of phytoplankton samples

4.1 Introduction

Microscopic determination is the only method to get information on the species composition of phytoplankton samples. This information is needed in order to reveal changes in the phytoplankton communities in time and space and e.g. to estimate the potential toxicity of a bloom. The counting of cell number is time consuming, and when, mainly information on phytoplankton species composition is needed (ship-of-opportunity transects, additional vertical samples), a semi quantitative counting method can be used instead of the quantitative one. In this method, all the taxa will be identified and listed, but their abundance is estimated using a semi-quantitative ranking (Leppänen et al., 1995).

4.2 Counting procedure

For the analysis, the inverted microscope technique is used. At least half of the chamber bottom should be analysed using small magnification (10x-objective) and two bottom transects with larger magnification (40x-objective). All the species found should be listed using the HELCOM counting software with the net option, so when recorded they get the smallest ranking 1 automatically. The semi quantitative ranking should be done using a scale from one to five. The ranking is sample specific, and several species can also get the same ranking, even the highest one.

1. very sparse, one or a few (less than five of the >20 µm fraction) cells or units in the analysed area = in the sedimented sample

2. sparse, slightly more cells or units in the analysed area

3. scattered, irrespective of the magnification several cells or units in many fields of view

4. abundant, irrespective of the magnification several cells or units in most the fields of view

5. dominant, irrespective of the magnification many cells or units in every field of view

When the accurate abundance of a species (e.g. a potentially toxic one) should be counted, at least 20 fields (with 40 x objective), or one transect (with 10 x objective) should be analysed.

The sedimentation chambers etc. should be cleaned as for the quantitative analysis.

5. References

Andersen, P. and Throndsen, J., 2003. Estimating cell numbers. In Manual on Harmful Marine Microalgae. EDS.: Hallegraeff, G.M, D.M. Anderson and A.D. Cembella. Unesco Publishing, Paris, p. 99-129.

Hasle, G.R., 1978. The inverted-microscope method. In: Sournia, A. (ed.): Phytoplankton manual. UNESCO Monogr. Oceanogr. Method. 6: 88-96.

HELCOM, 1988. Guidelines for Baltic Monitoring Programme for the third stage. Part D. Biological determinands. Baltic Sea Environment Proceedings No. 27 D.

Kozova, O.M. and Melnik, N.G. 1978. Instruction for plankton samples treatment by counting methods. Eastern Siberia Pravda, Irkutsk, 52 pp. (In Russian).

Leppänen, J.-M.; Rantajärvi, E.; Hällfors, S.; Kruskopf, M. and Laine, V., 1995. Unattended monitoring of potentially toxic phytoplankton species in the Baltic Sea. ­Journal of Plankton Research 17: 891-902.

Lindahl, O., 1986. A dividable hose for phytoplankton sampling. In Report of the ICES Working Group on Exceptional Algal Blooms, Hirtshals, Denmark, 17-19 March 1986. ICES, C.M. 1986/L:26.

Lohmann, H., 1908. Untersuchungen zur Feststellung des vollständigen Gehaltes des Meeres an Plankton. Wiss. Abt. Kiel NF. 10.

Lund, J.W.G., Kipling, C. and Le Cren, E.D., 1958. The inverted microscope method of estimating algal numbers and the statistical basis of estimations by counting. -Hydrobiologia 11:2, pp. 143-170.

Menden-Deuer, S. and Lessard, E.J., 2000. Carbon to volume relationsships for dinoflagellates, diatoms and other protist plankton. Limnol. Oceanogr. 45: 569-579.

Nauwerck, A., 1963. Die Beziehungen zwischen Zooplankton und Phytoplankton im See Erken. Symb. Bot. Ups. 17(5): 1-163.

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 No.106, 144pp. Printed Paper is available: http://www.helcom.fi/stc/files/Publications/Proceedings/bsep106.pdf

Updated Biovolume Table (Annex 1=HELCOM PEG Biovolume) is available at: http://www.ices.dk/env/repfor/index.asp.

Rott, E., 1981. Some results from phytoplankton counting intercalibrations. Schweiz. S. Hydrol. 43: 34-62.

Sicko-Goad, L., Stoemer, E.F. and Ladewski, B.G., 1977. A morphometric method for correcting phytoplankton cell volume estimates. Protoplasma 93.

Smayda, T.J., 1965. A quantitative analysis of the phytoplankton of the Gulf of Panama II. On the relationship between 14C assimilation and the diatom standing crop. Inter-Amer. Trop. Tuna. Comm. 9:7.

Strathmann, R.R., 1967. Estimating the organic carbon content of phytoplankton from cell volume or plasma volume. Limnol. Oceanogr. 12.

Utermöhl, H., 1958. Zur Vervollkommnung der quantitativen Phytoplankton-Methodik. Mitt int. Verein. theor. angew. Limnol. 9: 1-38.

Veldre, S.R., 1961. Statistical verification of counting methods used for quantitative analysis of plankton samples. In: Application of mathematical methods in biology. Collected papers, Leningrad State University, Vol. 2: 124-131. (In Russian)

Venrick, E.L., 1978a. The implications of subsampling. In: Sournia, A. (ed.): Phytoplankton manual. UNESCO Monogr. Oceanogr. Method. 6: 75-87.

Venrick, E.L., 1978b. How many cells to count? In: Sournia, A. (ed.): Phytoplankton manual. UNESCO Monogr. Oceanogr. Method. 6: 167-180.

Willén, T., 1962. Studies on the phytoplankton of some lakes connected with or recently isolated from the Baltic. Oikos. 13: 169-199.

Last updated 26.10.2011