This paper considers the structure of freshwater phytoplankton assemblages and promotes a scheme of ‘vegetation recognition', based upon the functional associations of species represented in the plankton. These groups are often polyphyletic, recognizing commonly shared adaptive features, rather than common phylogeny, to be the key ecological driver. Thirty-one such associations are outlined and the basic pattern of their distinctive ecologies is outlined. An invitation to other plankton scientists to assist in the development of this scheme is issued.
No less than other scientific disciplines, biology is founded on a vast amount of verified but miscellaneous information that, necessarily, must be classified and ordered. Linnaeus may not have been the first to name and group organisms together but the essence of his binomial system of species names, strictly within higher hierarchical groupings based upon morphological affinities, has survived intact through over two centuries of continuous expansion and refinement. The new age of molecular biology and genetic verification has not made Linnaean taxonomy redundant but, rather, has greatly enriched it, at once furnishing an independent test of suspected evolutionary affinities and a means of verifying the mechanisms of gene expression. Modern molecular techniques consolidate the strength of a sound phylogenetic base to taxonomy and, more often than not, lend weight to many past judgements about the interrelationships of individual species.
Ecology is a predominantly biological discipline concerned with the distributions of organisms and their interrelationships with each other and their environments. Ecologists also recognize that unless they can name species or, at least, have a close idea of their affinities, the point of their work is very largely lost. At the same time, however, whether they are trying to work out how a system is organized and its functions are allocated, or todistinguish the differences in energy flow through the community structures of old meadow or arable cropland, various types of forest or coral reefs, ecologists will soon resort to additional schemes of classification (blLavorel et al., 1997). These will be concerned with the functional roles and structural adaptations of the main species, whether they are producers or consumers, whether they areinvasive or tolerant of site maturation, whether they share tolerances to acidity or waterlogging or other environmental constraints, and so on. Terrestrial ecologists recognize types of system and predominant pathways of energy flow based wholly on their ability to distinguish (say) sandy heath from ombrogenous bog, or Carex paniculata swamp from Salix carr or from Quercus woodland, simply on the grounds of which kinds of species are present. Communities are better, more reliable indicators of habitat conditions than are the presence or absence of component species. The work of the great phytosociologists [(Tansley, 1935; Tüxen, 1955; Braun-Blanquet, 1964); see also (Shimwell, 1971)] has bequeathed a system of diagnosing and naming the very distinct associations of plant species that constitute vegetation. In essence, the associations are the basic functional units. Each is named after one or two species that are characteristically represented in that particular community-type, using a distinctive binomial construction based on the name of one of them: Lemnetum minoris is the association of free-floating mats of duckweed, typically including Lemna minor (Haslam et al., 1975). The system works because the overlapping requirements of individual species can often be satisfied simultaneously in particular locations, so long as the adaptations of each allow them to tolerate the conditions obtaining. Most interestingly, such tolerances may require behavioural or physiological adaptations that are not necessarily constrained by phylogenetic affinity: plant associations can be polyphyletic.
When it comes to describing the structure of pelagic communities however, most aquatic ecologists might be content to know that phytoplankton was present (perhaps judged by its content of extractable chlorophyll). In contrast, specialists may well delight in being able to provide a full list of species, perhaps separated according to criteria distinguishable only with an electron microscope or verifiable using polymerase chain reaction. At a more general level, the distinction of a diatom- or flagellate- or even a Cyanoprokaryotedominated phytoplankton may convey a suitable summary of what is there. Yet the question as to why one phylogenetic category of organisms does better than another under sketchily defined environmental conditions remains largely unanswered, while the ability to explain, much less predict, which species dominate continues to puzzle (Huisman and Weissing, 2001).
In this paper, we seek to consolidate a view that plankton ecology will benefit from the adoption of an alternative scheme of ‘vegetation recognition', based upon the representation of functional associations of species. These may be selected or excluded on the basis of major adaptive features but which are not specific to one or a few phylogenetic groups. Similarly, selection or exclusion of these very features by the conditions obtaining in given systems will not predict species dominance but they may well narrow down the probabilities to a particular functional group well-represented in a given habitat or some given variation in the conditions.
Before elaborating the notion that microscopic algae might form sociological groupings much as do terrestrial plants, it is necessary to say a word about our choice of group terms, none of which has universal understanding or acceptance. ‘Communities' is rejected for, however fortuitous is their composition, the species present interact and some may perhaps benefit each other in some way. We should also recognize the community role of heterotrophs and phagotrophs, but this is not a feature of the scheme we describe here. ‘Assemblages' is better in recognizing the fortuity of species presence. However, material collected in a single sample of plankton may well comprise several dozens of taxa, few of them numerous, some being there only because they have been introduced by advection or entrainment (for example) from the benthos or littoral. Some species will have been increasing in representation in conditions they had found favourable, others will have been in decline because the same conditions were not altogether suitable for them. Our scheme seeks to distinguish these behaviours, at least among the relatively more numerous species.
‘Associations' is the word used by terrestrial plant ecologists. It has the merit of conveying the notion of a group of species responding similarly to a single set of environmental conditions, even when the positions in a fluid environment are more difficult to deal with. Yet this too suffers a drawback in failing to acknowledge the fact that some species may show analogous adaptations to similar conditions but have yet to be found simultaneously ‘in association' at the same localities—thus, Cylindrospermopsis and Anabaena minutissima show similar antennal properties and nitrogen-fixation capacities that suit them to turbid, nitrogen-deficient water columns but, to our knowledge, they have never been found in mutual association in the same location.
The collective term closest to encompassing our proposal to bracket together species with similar morphological and physiological traits and with similar ecologies is perhaps, ‘narrowly defined functional groups'. Accepting that the expression ‘functional group' does not have a unique or universal interpretation either, we emphasize that the terminology seeks to differentiate among phytoplankton on the basis of specialist adaptations and requirements (such as having a high affinity for phosphorus or carbon dioxide at low external concentrations, or of requiring skeletal silicon, or of being a good light antenna). The separation then permits two useful deductions to be made. One is that a functionally well-adapted alga will be likely to tolerate the constraining conditions of factor deficiency more successfully than individuals of a less well-adapted species. The other is that a habitat shown typically to be constrained by light, or C or N or whatever, is more likely to be populated by species with the appropriate adaptations to be able to function there (but NOT that they WILL be there). Thus, our application of the term ‘functional group' is sensitive to the sets of appropriate adaptive specialisms and the clusters of species that have them. Until such time as it is possible to forecast the presence of individual species in given locations, this scheme seems to us to offer the clearest way into understanding and predicting the distributions and dynamics of natural populations of phytoplankton.
Adopting this concept of ‘associations' and of ‘functional groups', we set out here to demonstrate from our individual experiences the measure of consensus we have attained about the general applicability of the model. We seek to encourage others to use it and, because the method is in its infancy, we invite them to share in the means of its development.
THE VALIDITY OF THE FUNCTIONAL GROUPING
The idea of developing ecological categories of phytoplankton and the desire to be able to use them to describe variations in composition among the natural lakes has a long history (Hutchinson, 1967). The inspiration for Reynolds' first attempt (Reynolds, 1980) to devise a system of classification of planktonic algae to be sensitive to environmental change (essentially to eutrophication but also to shorter seasonal fluctuations in stratification and in the accessibility of adequate nutrient supplies) came from the sorting methods of Tüxen and Braun-Blanquet (Tüxen, 1955; Braun-Blanquet, 1964): the relevé, the small arbitrary unit of vegetation in which plant presence is scored for abundance, was substituted by a list of the species counted (in excess of a cut-off point) in a particular sample from a particular lake on a particular date. Species frequently found to co-exist and to increase or decrease in number simultaneously were delimited and given association identities. Although these were claimed to be analogous to plant associations, to accord them formal phytosociological names seemed premature and the alphanumeric devices have been retained provisionally.
Fourteen groups of phytoplankton were identified in Reynolds's original study (Reynolds, 1980). Some have been subdivided since, although the biggest change has been to re-label them (Reynolds, 1984). Such validity as the sorting method carries continues to underpin these associations. Today, the list of associations (Table I) has more than doubled, with most of the later groups having been added, as it were, ‘by eye'. Whilst still retaining the ideal that the subdivisions reflect the simultaneity of responses of individual species to environmental variability, it has to be emphasized that the new groupings have been accommodated on intuitive grounds, without the benefit of even the sociological separation. This does not make them wrong but an invalid principle gives no grounding for robust advocacy of the functional grouping.
However, two developments enhance the validity of the functional grouping. In the analysis of the phytoplankton in a small urban lake in Montevideo, Kruk et al. (Kruk et al., 2002), applied canonical variate analysis (ter'Braak and Smilauer, 1998) to discriminate phases of differing species composition with several classification approaches. They found that the eigenvalue for the first CCA axis for the associations was over 0.7: 78% of the cumulative variance was explained by the scheme as it had evolved by 1997, despite having been designed primarily for European lakes. This was a statistically better performance than any of the other schemes considered. The functional-group classification seems to capture much of the statistical variation. Analogous findings have been reported by Fabbro and Duivenvoorden (Fabbro and Duivenvoorden, 2000), although they used a different sorting method.
The second enhancement comes from the finding that algae forming a single functional group also have similar morphologies, as quantified by the dimensions of the algal ‘units' (cells or colonies, as appropriate, together with any peripheral mucilage): surface area (s), volume (v) and maximum linear dimension (m) are powerful predictors of optimum dynamic performance (Reynolds and Irish, 1997). Typical representatives of each main association strike mutually similar co-ordinates on a plot of ms/v against sv. The plot pulls apart rounded and stout cylindrical units of differing sizes but separates those larger units that preserve a high s/v by being attenuated in one plane (sometimes two, but not three). Separation of the algae on the basis of their morphologies coincides substantially with the distributions of the same species among different types of habitat distinguished on the basis of accessibility to light and all nutrient resources (Reynolds, 1984, 1997).
The converse of this is that it is now easier to fit hitherto functionally unclassified species into existing functional categories, at least on a provisional basis. What to do about those that do not readily fit the classification is to become one of the recommendations of our essay.
THE RECOGNITION AND NOMENCLATURE OF THE FUNCTIONAL GROUPS
The latest list of trait-differentiated functional groups (Table I) has 31 entries. Alphanumeric terms are retained but some groups that have stood robustly for the last 20 years are named provisionally for the first time. In this section, their distinguishing properties and ecologies are briefly reviewed. At the beginning, the alphanumeric coda were allocated in blocks to reflect seasonal shifts (A–D for vernal blooms, E–H for associations at the start of summer stratification, and so on) and within each block, the trophic ‘preferences’ were also distinguished (so B → E → L → N might represent the seasonal progression in a mesotrophic temperate lake and C → G → M → P would summarize a more eutrophic system). As the system has become adapted to accommodate other types of water and especially waters at other, non-temperate, latitudes, so the intrinsic significance has been lost. In the text below and in the Tables, we have found it unhelpful to run through the coda in alphabetical order, preferring instead to develop the distinctions among near-related functional groups.
As indicated, the groups A to C were devised originally to apply to the vernally growing diatoms in temperate lakes of differing trophic levels. Group A is characterized by a number of centric diatoms of the genus Cyclotella(C. glomerata, C. comensis) and, especially, of Urosolenia (formerly Rhizosolenia) that are prominent in the plankton of many medium-to-large high-latitude lakes that are typically clear, dilute in solutes and deficient in phosphorus. The algae in this group are mostly unicellular (103–104 μm3 in volume), apparently have high affinity for nutrients, though not, apparently, for carbon. Its distinctiveness may earn the title ‘Urosolenietum'. The A association is also represented in the clear, low-phosphorus Amazonian flood-plain lake, Lago Batata, where it occurs together with Merismopedia and Peridinium umbonatum (Group LO, q.v.) during the filling and high-water stages (Melo and Huszar, 2000). Group B diatoms belong to more enriched lakes, merging into Group C without precise boundaries. The large-celled Stephanodiscus (rotula, neoastraea) are indicative of calcareous, often very phosphorus-rich, systems where they may co-dominate with Asterionella and such species of Aulacoseira as A. ambigua. Asterionella and other Aulacoseira spp. (A. subarctica, A. islandica) and Cyclotella (C. meneghiniana, C. stelligera) co-exist in softer-water lakes. Diatoms of groups B and C form larger units (104–105 μm3 in volume) but shape contributes to the maintenance of high s/v and efficient light-harvesting. Population development is often subject to the availability of silicon and dependence upon turbulence for suspension leaves them relatively sensitive to mixed depth and the seasonal onset of near-surface density stratification.
Diatoms of Group D are mostly found in shallow, nutrient-enriched, well-ventilated waters, liable to be turbid. Small-celled (≤ 103 μm3 in volume) and fast growing, such diatoms include planktonic Nitzschia spp., Synedra acus, Stephanodiscus hantzschii, Cyclotella ocellata and, possibly, Cyclotella pseudostelligera.
Groups N and P also involve diatoms but both are associated either with lower latitudes or with the summer period in temperate lakes. The dependence upon physical mixing is strongly apparent, requiring a continuous or semi-continuous mixed layer of 2–3 m in thickness. Both associations can be represented in shallow lakes where the mean depth is of this order or greater, as well as in theepilimnia of stratified lakes when the mixing criterion is satisfied. Species of Group P (which include Aulacoseira granulata and Fragilaria crotonensis) are apparently more tolerant of carbon dioxide depletion than those of Group N (Tabellaria spp.) and tend to be present in the more eutrophic waters, subject to their other requirements being satisfied. Both groups are characteristically associated with a presence of desmids, though these associations have still to be investigated in detail. Some of the slender-celled Closterium spp. and hard-water desmids (Staurastum pingue) are P members, whereas the affinities of some common planktonic Cosmarium, Staurodesmus, Pleurotaenium and Xanthidium spp. are with N. Desmids may be more or less abundant than diatoms in established assemblages dominated by Groups N or P. Strong dominance by desmids has also been noted in some tropical lakes as a consequence of atelomixis, where the surface mixed layer is alternately attenuated and deepened on a diel basis (Barbosa and Padisák, 2002).
In more persistently mixed layers, in which light is increasingly the limiting constraint, filamentous algae show preferential adaptations. Some of these (Binuclearia, Geminella, Mougeotia, Tribonema) were grouped relatively recently (Reynolds, 1997), in spite of their disparate phylogenetic affinities, into Association T. This is one of the groups that merges into another, the S group of photoadapting solitary filamentous Cyanoprokaryotes (Cyanobacteria). The most familiar of these is Planktothrix agardhii, which is common among enriched, exposed and generally shallow lakes at most latitudes, where it can aspire to monocultural populations persisting throughout the year, constituting ‘the third stable state' to which shallow lakes may gravitate (Scheffer, 1998). Other potentially dominant species from the group include Limnothrix redekei, Pseudanabaena limnetica and Planktolyngbya contorta. The original, temperate-lake Group (S1) reasonably aspires to a ‘Planktotrichetum'. The subdivision S2, comprising the group of genera, Spirulina, Arthrospira and Raphidiopsis that inhabit warm, shallow and often very alkaline waters, was proposed by Reynolds (Reynolds, 2000). The solitary habits of Cylindrospermopsis and Anabaena minutissima often in relatively phosphorus-rich lakes and their tolerance of vertical mixing set them apart from other dinitrogen-fixing members of the Nostocales. They had been allocated earlier to Group SN (Padisák and Reynolds, 1998).
At the other extreme of warm-water systems are the clear, low-biomass oligotrophic and ultra-oligotrophic water columns in which relatively the largest biomass resides in the picophytoplankton. Group Z was an ad hoc creation to accommodate the organisms and what, at first, seemed to be their habitats. The significance of autotrophic picoplankton [photoautotrophs in the size range 0.2–2 μm: (Sieburth et al., 1979)] to the carbon flow in the sea and in large lakes has been recognized for some time (Fogg, 1986; Stockner, 1988; Pick, 2000). Picophytoplankton were originally understood to include single-celled synechococcoid and Cyanobacteria and, certainly in lakes, a number of chlorococcal species of Chlorella and Choricystis. Prochlorophyte Cyanobacteria also figure in the oceanic picoplankton (Chisholm et al., 1988). The emphasis in the designation is the small unit sizes (unicells, < 10 μm3 in volume), which have a tendency not to sink out rapidly but to be persistent, at least for so long as they remain substantially ungrazed. Colonies comprising small cells of Cyanobacteria (such as Aphanocapsa) or of chlorophytes (such as Dictyosphaerium or Coenochloris) belong in other groups (K and F respectively). Neither is it the case that picophytoplankton is confined to oligotrophic systems. The high biomasses of Synechocystis aquatalis,Synechococcus and Chlorella minutissima observed by Huszar et al. in certain coastal lagoons suggested to them a more eutrophic assignation (X1) than is implied by the use of Z for the synechococcoid picoplankton (Huszar et al., 2000). Provided that the energy-, carbon- and nutrient-requirements of these algae can be met, they are not obliged to live at low population densities (Vörös et al., 1991). Some of the highest recorded densities of synechococcoid picoplankton come from shallow, eutrophic ponds in Antarctica where grazing on the autotrophs was also apparently restricted (Izaguirre et al., 2001). In addition, there is now recognized a functional difference between the deep-water prokaryotes that contain phycoerythrin pigment and those living in surface waters and do not (or contain rather less; Padisák et al., 1997). Pending further investigation, this distinction has not been formalized.
It is apparent that the Z association requires further consideration. Huszar et al. ((Huszar et al., 2000) have gone as far as a proposal to transfer all eukaryotic members (such as Chloromonas and Chlorella minutissima) to X, on the grounds that, functionally, these are simply the smallest members of the nanoplankton. For the present, we retain the Z category to refer to the picocyanobacteria that dominate in the photic layers of oligotrophic lakes, either in the upper mixed layers or forming deep chlorophyll maxima in the upper hypolimnion, as they do in the Stechlinsee (Padisák et al., 1997). Elsewhere, Olrik's (Olrik, 1997) discrimination among the smaller nanoplankters (mostly unicells, < 103 μm3 in volume) is sufficient to justify the ascription of oligotrophic (X3), meso-eutrophic (X2) and eutrophic-hypertrophic (X1) associations.
The Y association was set aside for larger nanoplanktic flagellates (103–104 μm3 in volume), although it has been used more or less synonymously with the larger Cryptomonad species. Well-adapted to live in a wide-range of habitats, their dynamics are, nevertheless, vulnerable to the feeding activities of a wide range of crustacean zooplankton and even of some rotifers and protistans.
In the earliest attempt to separate phytoplankton associations, the sparse, oligotrophic–mesotrophic early summer plankton, often dominated by chrysophytes and mucilaginous colonial green algae, provided difficulties. With more information, however, the groups E and F are known to be represented in the plankton of a wide spectrum of lakes but not for the same reasons. Some of the chrysophyte genera are demonstrably capable of strong growth in the well-insolated water, where they can supplement nutrient uptake from dilute solution by the phagotrophic ingestion of bacteria. They are, however, obligate consumers of carbon dioxide gas (i.e. they cannot resort to bicarbonate as many eutrophic algae do). The non-motile but near-neutrally-buoyant colonial green algae of Group F seem to have an elevated light threshold: they function best in clear water and are otherwisetolerant of deep mixing. Thus, algae of both groups E and F have a strong representation among mesotrophic lakes but both are sensitive to nutrient enrichment and the additional demands that high biomass may place on the carbon and light fluxes.
In contrast, the strongly motile colonies of Volvox, Eudorina and their allies are able to regulate their position in relation to the light field. They respond to nutrient-rich conditions in stagnating water columns and are, therefore, most familiar in small eutrophic lakes and during very stable phases in larger river-fed basins and storage reservoirs. The distinctive Group G could be proposed as a ‘Volvocetum'.
Another distinctive group (J) of mainly non-gelatinous, non-motile Chlorococcales is prominent in shallow, highly enriched systems (including many low-gradient rivers), represented by Scenedesmus, Pediastrum and Coelastrum. At one end, that of barely colonial, mucilage-free chlorococcals (such as Golenkinia, Treubaria), it merges into X1; at the other, where light becomes critical, it merges into S.
Close to J is another group of (mostly) shallow-water small-celled colonial Cyanoprokaryotes of the genera Aphanocapsa, Aphanothece and other non-vacuolate types. We are uncertain of the precise significance of abundant populations of these Group-K species, save that they seem to survive high pH particularly well and may thus represent an association in transition towards LM or M.
The Group H is another of long standing, having been created originally for dinitrogen-fixing Cyanobacteria of the bloom-forming Order Nostocales. However, the group has been subject to recent reviews: the solitary forms were recognized to be more Oscillatoria-like than the buoyancy-regulating types in their ecology (Padisák and Reynolds, 1998; Huszar et al., 2000) and were distinguished as SN (for nitrogen-fixing S types). Now, we feel moved to separate the classical eutrophic, low nitrogen group (H1), built around Aphanizomenon flos-aquae, Anabaena flos-aquae, An. circinalis and An. spiroides, from the group of species (H2), such as An. lemmermanni, An. solitaria and Gloeotrichia echinulata, that are tolerant of mixing conditions in rather larger but typically less eutrophic (hence, clearer) lakes (Reynolds and Lund, 1988; Garcia de Emiliani, 1993).
Among well-stratified lakes, the activities of planktonic primary producers leads to further differentiation of the habitat, so that the upper waters, though well-insolated, become severely deficient, either or severally, in phos-phorus or nitrogen or carbon, whereas beneath them is water where nutrients are less depleted, or may even accumulate through biological transport, but where there is too little light for massive photoautotrophic exploitation (Margalef, 1978). In this situation, one of the more advantageous adaptive strategies is to be able to make substantial vertical migrations between the two compartments. The essential adaptation is the combination of efficient motility with large size: this is evident among the algae of the next four categories, all of which generate unit sizes that approach 105 μm3 or are rather greater. U is for Uroglena, possibly the unique member of the group of large, motile, colony-forming chrysophytes that may be observed in stratifying oligotrophic and mesotrophic lakes. LO refers originally to the Peridinium–Woronichinia association of stratified mesotrophic lakes, though some usage has tried to accommodate more acidic assemblages involving P. inconspicuum and such Cyanoprokaryotes as Merismopedia. Ceratium hirundinella and similar species are actually rather eurytrophic and may be associated equally with Cyanoprokaryote associations based on Microcystis aeruginosa as easily as with Woronichinia; indeed, all three may co-exist within the confines of LM.
While the L assemblages may need careful review and recasting, the near-relative M seems better characterized, referring to the almost monocultures of large colonies of such as Microcystis aeruginosa, with many of its overlapping subspecies and ecotypes, Microcystis wesenbergii and Sphaerocavum brasiliensis. Each may be well over 106 μm3 in volume and its buoyancy control copes not just with segregation per se but accommodates quite striking diel fluctuations in stratification and mixing in low-latitude lakes.
In many of the deep, clear alpine lakes, the metalimnia are sufficiently stable to become populated by plate-like developments of photoautotrophs that remain centred within narrow depth ranges for weeks to months on end, moving upwards or downwards in response to small changes in the light field or in the availability of other essential resources. The behaviour is typified by that of the Cyanoprokaryote, Planktothrix (formerly Oscillatoria) rubescens [(Zimmermann, 1969); see also (Bright and Walsby, 2000)] which is recognized in the group R. Besides having a remarkable capacity for sensitive buoyancy regulation, the species has a powerful facility for chromatic adaptation during stratification deep in the light gradient. However, the alga also tolerates holomictic entrainment and may even grow under these circumstances. In smaller stratifying lakes, other species of Planktothrix (P. limosa, P. mougeotii) and Planktolyngbya (P. subtilis) showing analogous physiological and behavioural adaptations, may dominate. These are also understood to fit in Group R. However, dominance of the stable, optically deep-water layers in the stable gradients of small, density-stratified lakes at all latitudes (tropical forest lakes, ice-covered polar lakes, solution hollows) may be equally open to dominance by chrysophyte (E) or cryptomonad (Y) flagellates (Reynolds, 1997). Where there is also a pronounced gradient in redox, the key autotrophs may be species of the purple and green sulphur bacteria that make up Group V.
Small ponds, well supplied by organic matter (either from farm stock or vegetal decomposition, including leaf fall), support algal flora which have not been well characterized. There may well be present elements of the D–J–X1 group of shallow enriched systems, possibly also Y-type cryptomonads, though one or several of the following components may also be represented: euglenoids (e.g. Euglena, Phacus, Lepocinclis), small colonial volvocines (Gonium), some chrysophytes, provided the pH is not too high (Synura spp.), and the smaller dinoflagellates of the genera Peridinium and Glenodinium. Grouping these in a single association (W, now W1) is provisional and open to further separation. Euglenoids, represented particularly by Trachelomonas, are also found in the distinctive bottom-dwelling community of shallow, aerated lakes which appears in open water on occasions. It is tentatively distinguished as W2. Finally, a new codon is introduced here, Q, to refer to the populations of Gonyostomum that dominate in humic but generally productive forest lakes, mostly at high latitudes, usually having a low calcium content and a pH on the acid side of neutrality [see especially (Korneva, 2001)].
THE APPLICATION OF FUNCTIONAL GROUPS
In this section, we seek to show some of the uses of the functional-group model. These include the aid to understanding why certain species of phytoplankton should be more favoured than others in the assembly of communities and why planktic species composition should vary in space and time. Conversely, the similarity in plankton assemblages in similar lakes and at similar times becomes easier to explain, as do the differences in assemblage structure represented in lakes of different types. Such understanding is the basis of making valid predictions or, at least, of extrapolating probabilities, about the structure of the phytoplankton. This must be of value to managers who now, more than ever, require decision support in determining their policies, priorities and reactions to developing situations.
Typical problems for managers are to do with the capacities of the waters in their charge to support phytoplankton and, once certain conditions are triggered, how quickly the capacity may be filled. These are not questions that the functional classification is designed to meet, although appropriate simple models are now available (Reynolds, 1992, 2002; Reynolds and Maberly, 2002). However, the next question is about the identity of the dominant algae and whether it is likely to cause nuisance because of its latent toxicity, taste/odour implications, or the ease of treatment for potable supplies (taken to be the converse of any propensity to block filters or to penetrate floc blankets). In the case of storage reservoirs, helixors, circulating pumps, multiple depth inflows and draw-off points might be operated in different ways to deliver a more favourable raw water for treatment.
Limnologically and hydraulically, reservoirs differ extensively from lakes but there is no difference in the principles that eventually influence the composition of the phytoplankton (Reynolds, 1999). In reservoirs, as in lakes, predicting dominant species, peak populations and their dates is almost impossible. Even with the benefit of hindsight, to recount the forces, their intensities and duration, to say nothing of the stochasticity of the contributory events, that led precisely to the observed condition, is defied by the chaotic nature of variability.
Yet, as has often been pointed out, certain aspects of plankton dynamics are broadly predictable. As an example, spring blooms in temperate lakes, in response to lengthening days more than warming waters, may be predicted with confidence. The dominant species may also be predicted on the basis of standing inocula furnished by past spring blooms, while the alternative dominants will come from a robustly predictable group of species. These groups correspond to the functional associations: there is merit in being able to describe in these terms the sequence of phytoplankton dominance across temporal and spatial sequences of water bodies. One of the most comprehensive of such comparisons was that of the phytoplankton of deep alpine lakes made by Sommer (Sommer, 1986): he demonstrated a similarity of year-to-year behaviours in individual lakes, sequential stages (spring bloom, summer stratification, the summer–autumn phase of mixed-layer deepening) common to all of them, and of different kinds of dominant association reflecting trophic status and limiting nutrient availability. However, the patterns are reported in terms of a few key dominants and of the phylogenetic groupings of the main species. Setting the same information in terms of the functional associations provides a striking summary of the patterns detected (Table II). The classification also holds for the large lakes south of the Alps (Salmaso, 2000). Huszar and Caraco have made similarly convincing deductions on the basis of their study of a series of lakes in the northeastern United States (Huszar and Caraco, 1998). In the meantime, the temporal and spatial variabilities in phytoplankton composition of several tropical and sub-tropical systems have been readily described in terms of association representation (Huszar et al., 2000).
In order to apply the approach to the prediction of the consequences of changed management practices—such as the occurrence of toxic Cyanoprokaryota—it is helpful to carry forward an understanding of the concepts of tolerance and sensitivity shown in Table I and which underpin the ‘+' and ‘–' response signs in Table III. Without this, it is difficult to determine whether an alga is in a particular functional group because it is in a particular kind of lake, or it is in a particular kind of lake because of its functional affinity. A much more pragmatic view is that most species have a cosmopolitan or even ubiquitous distribution but they may be more sensitive to certain properties and environmental circumstances that weaken their dynamic performances in comparison with those that are less sensitive or positively tolerant of the same properties or circumstances. The longer the conditions obtain, the poorer is the survival of the first and the relatively better is the survival of the second. In this way the standing stock becomes biased towards the survivorship of the tolerant species. In less exacting environments, more species will, potentially operate successfully, but as the limiting conditions become more severe and the operational criteria discount against less tolerant species, the survivors become more predictable. This is the principle of community assembly and it underpins the generalized application of the original model concept that we advocate. In this way, we are able to assert that we have determined a system that is sensitive across latitudes, morphometries and trophic states, by virtue only of the algal preferences and sensitivities and not because of where and when the algae are found. Thus, we can anticipate, for example, that Group S1 is not excluded from tropical lakes (correctly so) but the warm-water preference of Group S2 means that we should not expect it to be represented at high latitude (hypothesis not disproved). Similarly, tolerance of nutrient deficiency or an inability to operate at low concentrations of available carbon does not prevent populations of species from the E, LO, X3 or Z categories from attaining high concentrations in eutrophic environments, provided that their light and carbon requirements can be met (all are verifiable phenomena).
Hence, there is no longer any paradox about whether the algal assemblage or the kind of lake determines the floristic group that may be favoured. Indeed, the supposed circularity of the argument becomes a positive strength when algae from the same functional association are found under analogous conditions in different lakes of similar types, as in Table II. Other examples include the sensitivities of the various functional groups of Cyanoprokaryota to the depth, stability and nutrient-content of water columns, shown in a small number of studies (Padisák and Reynolds, 1998; Beyruth, 2000; Huszar et al., 2000). Thus, it should be possible to decide the best way of managing a lake or reservoir to make it less suitable for the growth of toxic Cyanoprokaryotes (reduce nutrients? destratify? biomanipulate?). The existing knowledge to be able to do this is summarized in the semiquantitative check-list (albeit, still lacking in some details) presented in Table III.
DEVELOPMENT OF THE FUNCTIONAL CLASSIFICATION
The remaining objective of this presentation is to advocate its wider adoption and to secure a wider constituency for its development. We are certain that not all valid groupings have been described, while the functional significance of many algal species has not been accorded to functional groups and it is not clear that it will be simple to do so. The system as we have presented it has numerous weaknesses, including the prediction of assemblages of species with overlapping habitat preferences. As yet, we have made no attempt to deal with tychoplanktonic entries from the periphyton or from the benthos, although to develop a scheme that included these life-forms in their own right is likely to be a worthwhile objective. On the other hand, the increasing availability of statistical methods for sorting and discriminating among numerical and graphical information (Ten Brink et al., 1991; Seip and Reynolds, 1995; Kruk et al., 2002) invites additional contributions to consolidating the functional-group concept. The application of statistics based on Bayes' theorem on the inferences of probability distributions from data sets relating to the definition and distribution of the functional categories has particular appeal, as does the introduction of Artificial Neural Networks to hypothesis analysis. What began as qualitative shorthand for comparing separate plankton assemblages, in space and time, now aspires to a verifiable quantitative method of describing community structure and change.
We hope that this article will stimulate interest and the further contributions of other plankton scientists. There are sufficiently tempting analogies to propose that the approach is suited to the marine phytoplankton (Smayda and Reynolds, 2001
One of the keys to writing a descriptive essay is to create a picture in your reading audience’s mind by engaging all five of their senses – smell, sight, touch, taste and sound. If you can do this, then your essay is a success, if not, then you have a lot of work to do. The first steps in writing a descriptive essay will lay the groundwork for the entire piece.
Step 1: Choose a topic
A descriptive essay will usually focus on a single event, a person, a location or an item. When you write your essay, it is your job to convey your idea about that topic through your description of that topic and the way that you lay things out for your reader. You need to show your reader (not tell them) what you are trying to describe by illustrating a picture in their mind’s eye very carefully.
Your essay needs to be structured in a manner that helps your topic to make sense. If you are describing an event, you will need to write your paragraphs in chronological order. If you are writing about a person or a place you need to order the paragraphs so that you start off in a general manner and then write more specific details later. Your introductory paragraph sets the tone for the rest of the essay, so it needs to set out all of the main ideas that you are going to cover in your essay.
Step 2: Create a statement
The next step is to create a thesis statement. This is a single idea that will be prominent throughout your essay. It not only sets out the purpose of the essay, but regulates the way that the information is conveyed in the writing of that essay. This is an introductory paragraph that sets out your topic framework.
Step 3: Get the senses right
Next, create five labelled columns on a sheet of paper, each one having a different of the five senses. This labelled list will help you to sort out your thoughts as you describe your topic – the taste, sight, touch, smell and sound of your topic can be sketched out among the columns. List out in the columns any sensation or feeling that you associate with the topic that you are writing about. You need to provide full sensory details that help to support the thesis. You can utilize literary tools such as metaphors, similes, personification and descriptive adjectives.
Once you have the columns laid out you can start to fill them with details that help to support your thesis. These should be the most interesting items that you have noted in your columns and will the details that you flesh out into the paragraphs of the body of your essay. Topics are set out in each separate paragraph and a topic sentence begins that paragraph and need to relate to your introductory paragraph and your thesis.
Step 4: Create an outline
The next step is to create an outline listing the details of the discussion of each paragraph. Students in high school are generally asked to write a five paragraph essay while college students are given more freedom with the length of their piece. The standard five paragraph essay has a particular structure including the introductory paragraph with the inclusion of a thesis statement, followed by three body paragraphs which prove that statement.
Step 5: Write the conclusion
Finally, the conclusion paragraph makes a summary of the entirety of your essay. This conclusion also needs to reaffirm your thesis (if necessary). Your conclusion needs to be well written because it is the final thing to be read by your reader and will remain on their mind the longest after they have read the remainder of your essay.
Step 6: Review your essay
It is important to take a break from your writing once you have completed the work. By stepping away from the work for a short time you can clear your mind and take a short rest. You can then take a look at the essay with fresh eyes and view it in much the same way that a person reading it will when they first see the piece.
After you have taken a short break or a walk (or whatever the case may be), read the entire essay again thinking about your reader. You should ask yourself if you were the reader, would the essay make sense to you? Is it easy to read so that anyone can understand what the topic of the essay is? Do any of the paragraphs need to be rewritten because they are confusing and need to be better written to be descriptive?
Your choice of words and language need to convey what you are trying to describe when you talk about a particular topic. The details that you have provided should give your reader enough information that they can form a complete picture. Any details in the essay should help a reader to understand the meaning of the topic from the writer’s point of view.
Read your entire essay over again, out loud this time. Sometimes reading something out loud can help to identify any issues that should be worked out. Read the essay again to a friend or family member and have them give you any criticisms that they might have. Have someone else ready your essay and then ask them if anything needs to be clarified or if they received a clear picture from the details given in the essay.
Step 7: Finish it up
Finally, read your essay again very carefully and check for any grammar, punctuation or spelling errors that are obvious within the essay. If you find any clichés, be sure to delete them, they certainly do not belong in your essay. If there are any parts that are not completely descriptive or don’t make as much sense as you would like them to, rewrite them once again and then follow the proof reading and reading aloud process again to ensure that the final product is exactly as expected. You can never be too thorough when it comes to reading the essay over again and checking for any areas that need to be reworked.
Enjoyed this post? Here’s what you can do next: