SEARCH

SEARCH BY CITATION

Keywords:

  • functional diversity;
  • functional traits;
  • life strategies;
  • ordination techniques;
  • phytoplankton

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

1. This is a discussion of the applicability to the phytoplankton of the concepts of ‘plant functional types’ (PFTs) and ‘functional diversity’ (FD), which originated in terrestrial plant ecology.

2. Functional traits driving the performance of phytoplankton species reflect important processes such as growth, sedimentation, grazing losses and nutrient acquisition.

3. This paper presents an objective, mathematical way of assigning PFTs and measuring FD. Ecologists can use this new approach to investigate general hypotheses [e.g. the intermediate disturbance hypothesis (IDH), the insurance hypothesis and synchronicity phenomena] as, for example, in its original formulation the IDH makes its predictions based on FD rather than species diversity.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

The classification of species aids our understanding of the complexity of nature. Such classification is based mainly on the morphology and has only recently been assisted by genetics. At the same time, species have also been classified according to their functional/structural properties but, without using distinctly measurable characters, such classification has had limited application. The classification of the ecological strategies of plants proposed by Grime (1977) has been influential. He assigned species to three groups based on their evolutionary responses to disturbance and stress. This C–R–S concept states that in situations with a low intensity of disturbance and a low intensity of stress, such as nutrient limitation, competitive species dominate, the C-strategists. At low disturbance intensity and in a high-stress environment stress-tolerant species (S-strategists) are expected to outcompete others. So-called ruderals (R-strategists) dominate in low-stress and high disturbance environments. High-stress and high-disturbance environments are too hostile for any species to persist. This concept has been transferred and adapted to the phytoplankton by Reynolds (1988), further modified (Reynolds, 1997) and has provided a framework to explain phytoplankton succession with respect to water column mixing (Reynolds, 1993). It has been applied both to natural phytoplankton communities (e.g. Huszar & Caraco, 1998; Melo & Huszar, 2000; Weithoff, Lorke & Walz, 2001) and to experimental studies (Weithoff, Walz & Gaedke, 2000).

Besides the C–R–S concept, Reynolds (1980) introduced a functional classification by assigning 14 phytoplankton associations to sets of environmental conditions, such as lake size, mixing regime, nutrients, light and carbon availability, etc. Over the past two decades this approach has been refined and an upgraded classification was presented recently by Reynolds et al. (2002). In its present form, 31 associations are described but ‘the new groupings were accommodated on intuitive grounds’ (Reynolds et al., 2002). However, the different algae forming a single group have similar morphological features which are ‘powerful predictors of optimum dynamic performance’ (Reynolds & Irish, 1997), although algae with different ecological strategies might be well adapted to similar environmental conditions. Nevertheless, the overall relationships between the algal associations and their habitats are quite well founded.

Recently, terrestrial plant ecologists have revived the idea of a functional classification in order to predict possible changes in the vegetation as a result of global climate change (Lavorel et al., 1997; Smith, Shugart & Woodward, 1997). The term ‘plant functional types’ (PFTs) was coined (Smith et al., 1993) and defined as ‘sets of species showing similar responses to the environment and similar effects on ecosystem functioning’ (Gitay & Noble, 1997). In terrestrial environments PFTs differ among habitats because habitats, ranging from deserts to marshes or forests, are inhabited by very different kinds of plants, whereas the pelagic phytoplankton communities from different water bodies are relatively similar. Relevant functional traits of species were proposed to be included in the set of functional characteristics (Weiher et al., 1999). A functional trait in this context is a feature or property of an organism which is measurable and influences one or more essential functional processes such as growth, reproduction, nutrient acquisition, etc.

The concept of functional diversity (FD) touches another aspect of the functional characterisation of species and communities. FD reflects the functional multiplicity within a community rather than the multiplicity of species. A simple measure of FD is the number of co-occurring functional types (Martinez, 1996), analogous to species richness at the species level. A more refined quantification of FD includes a ‘distance’ measure. In such a case the distance between species based on their functional traits is calculated, i.e. FD is high when species with widely differing functional traits are present in the same community. Although in many diversity studies a FD is implied, species diversity or species richness is used as a diversity measure, which may lead to misinterpretations. Thus, a new and mathematically quantifiable measure for FD is needed. In the following section I describe the basis upon which functional traits should generally be selected and which traits are useful for investigating PFT and FD in phytoplankton. After that I discuss a number of aspects in general ecology that could be investigated by using the proposed approach.

Selection of functional traits

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

The selection of traits is of crucial importance because all subsequent classifications or calculations of ecological distance (see below) depend on them (Gitay & Noble, 1997). The abundance and dynamics of any population are driven by ambient standing stock, growth and loss processes. For phytoplankton, net growth is the sum of intrinsic growth, sedimentation, grazing losses and some other less important loss factors. Therefore, the traits selected for phytoplankton should reflect these three main processes, which are a surrogate for population performance. In general, all traits should be easily measurable (Keddy, 1992; McIntyre et al., 1999; Walker, Kinzig & Langridge, 1999). Thus, for this purpose the capability of a species to acquire a particular nutrient throughN-fixation or phagotrophy, for example, or the demand for another like silica, is the appropriate information, especially as such information is available even for less well-known species. Resource-dependent growth rates are scarce in the literature and more useful for dynamic modelling approaches, e.g. in the PROTECH model (Reynolds et al., 2001).

In the following, I propose six traits as valuable for characterising functional aspects of phytoplankton. Most of them refer to processes which are also reflected in the dynamic PROTECH model (Elliott et al., 1999).

Size

According to allometric theory, size is a determinant of specific physiological activities such as growth, and the size range of phytoplankton covers more than five orders of magnitude, ranging from autotrophic picoplankton (<1 μm3) to large dinoflagellates (>50 000 μm3). Size in combination with shape also affects edibility. For Cladocera, the maximum-sized particle that can be ingested depends directly on the body size of the animal (Burns, 1968). Additionally, size and shape determine the surface to volume ratio that in turn influences nutrient uptake.

Nitrogen fixation

The potential for nitrogen fixation (e.g. in Nostocales, Cyanoprokaryota) gives a competitive advantage under nitrogen-limiting conditions.

Demand for silica

Aside from diatoms, which need silica for their frustules, Chrysophyceae and Synurophyceae form statospores (e.g. Ochromonas), bristles and scales (e.g. Synura or Mallomonas) also made of silica (Lee, 1999). In addition, silica increases the specific weight, which leads to higher sedimentation rates, especially in diatoms.

Phagotrophy

The ability to ingest bacteria serves as an additional source of nutrients and energy for phagotrophs. Particularly under nutrient-poor conditions the uptake of bacteria may contribute significantly to the phosphorus budget of the cell.

Motility

Mobile organisms can migrate into favourable patches and counteract sedimentation. This may be of particular advantage in an environment exhibiting steep gradients, such as the chemocline in stratified lakes (Gervais, 1997). Being non-motile is not a disadvantage per se as, in shallow waters, sedimentation may allow nutrient-depleted diatoms to take up minerals from the sediment surface and begin photosynthesis again soon after resuspension (Sicko-Goad, Stoermer, & Fahnenstiel, 1986). This meroplanktic behaviour can be seen as an efficient life strategy in shallow waters (Carrick, Aldridge & Schelske, 1993). In addition, motility effects nutrient deficiency as the movement of cells minimises the hydrate envelope and, thus, the diffusive boundary layer for nutrients around the cells (Pasciak & Gavis, 1974).

Shape

The shape of a cell or colony is important with respect to their susceptibility to zooplankton grazing. A suitable measure for ingestibility for filter-feeding zooplankton is the longest linear dimension (LLD) of the food item. As mentioned above, the shape together with size also influences the surface to volume ratio, which has been used in the PROTECH model to predict the maximum specific replication rate (Elliott et al., 1999; Reynolds, Irish & Elliott, 2001). In the present study, shape and size are treated separately.

All the above proposed traits are relatively easy to determine or data are available in the literature. Extensive and time-consuming preparations for the taxonomic determination of diatoms or dinoflagellates are not necessary and the proposed procedure is robust against new taxonomic findings that have accelerated recently supported by molecular methods. The set of traits selected may differ from one investigator to another and some variation is reasonable depending on the question under consideration in a particular study.

Data processing and the assignment of functional groups and functional diversity

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

In this section I discuss different mathematical procedures for assigning plants to PFTs and recommend a promising procedure for calculating FD based on the above selected traits. The values for the selected functional traits can be categorised as either binary or other codes. For some traits, such as silica demand, the capability of nitrogen fixation or bacterivory, binary data (0, 1) are recommended. For others a categorisation into size classes (size, LLD) is suggested. Motility can be classified as: 0, non-motile; 0.5, buoyancy regulation (through gas vacuoles); and 1, flagellated species which can move in three-dimensional space. From these data a species-trait matrix is created. As all traits are standardised in a range of 0–1, all traits are regarded to be of equal importance. This is not a prerequisite for this approach, because an a priori weighting is reasonable when some traits are regarded as ecologically more important than others. Common methods for determining functional groups are multivariate statistics, such as clustering techniques and principal component analysis (PCA), correspondence analysis (CA), canonical correspondence analysis (CCA) or principal coordination analysis (PCoA) (e.g. Jaksić & Medel, 1990; Leishman & Westoby, 1992; Pillar, 1999; Usseglio-Polatera et al., 2000; Kruk et al., 2002). CCA is a valuable option when additional environmental data are available. In this case species are ordinated and environmental factors are displayed as vectors within the ordination (e.g. for phytoplankton: Fängström & Willén, 1987; Kruk et al., 2002). Potential pitfalls in CCA with respect to a functional classification may occur when functionally different species co-occur under similar environmental conditions, e.g. small cryptophytes and large cyanobacteria colonies in eutrophic lakes (Sommer et al., 1986).

For cluster analysis, the species-trait matrix is transformed into a distance matrix. On this basis the species are clustered and the outcome of the cluster analysis is a dendrogram, which can be used to assign species into groups. Cutting the dendrogram into sub-dendrograms (branches) leads to a number of functional groups (Petchey & Gaston, 2002). This procedure can be used arbitrarily depending on the resolution in question, i.e. whether many or few groups should be created. The sum of all branch lengths is taken as the FD (Petchey & Gaston, 2002). It may be normalised for species number (s) as FDnorm = FD/s, which reflects the mean branch length and therefore allows comparisons between samples with different s.

Ordination techniques such as PCA, CA or PCoA result in biplots based on species traits. Species with similar ecological traits are located closely together whereas widely differing species are distant from each other. These techniques aim for a reduction of factors and thus inherit a weighting of traits. In PCA, abundant binary data may lead to misinterpretations but categorised variables are suitable and, therefore, PCA is not a suitable technique for the proposed traits. For the generation of biplots in PCA (e.g. PC1 versus PC2 or PC3 versus PC1) the euclidean distance between species is maintained. In contrast to that, the resulting biplots from CA display the χ2 distance between species. Different distance measures, of course, influence the outcome of the functional classification. In PCoA, the species-trait matrix is transformed into a distance matrix prior to ordination. This means the suitable distance measure for a particular investigation can be chosen. This, and robustness even when using binary data, makes PCoA a suitable technique for the approach suggested. The first three axes generated by all three techniques ideally explain a large amount of the variation in the data set. Thus a three-dimensional ordination can be seen as an ecological trait space spanned by the species according to their traits. As in dendrograms, functional groups can be created by dissecting the trait space into different sub-spaces containing the species forming a group. These groups can then be compared with the functional classification proposed by Reynolds et al. (2002). Additionally, each species has distinct coordinates within this ecological trait space, which enables us to calculate the variation around the community mean as a measure of FD. In other words, the centre of gravity of the species present forms the community mean and the size of the space around it is a measure of the FD. This centre of gravity represents the location of a system based on the function of its inhabitants and community shifts can be observed according to the directional movements of this centre. An exemplar comparison of species diversity calculated according to the Shannon index, and FD for the phytoplankton of Lake Constance in southern Germany (Gaedke, 1998) is shown in Fig. 1. The species-trait matrix was generated as described above and was transformed into a distance matrix using the χ2 distance. From the distance matrix a PCoA was run and the species coordinates were extracted. For FD the sum of the biomass weighted squared distances of each species present from the community mean was calculated for each of the first three axes (weighted according to the eigenvalue of each axis). Despite some relation, this figure clearly shows that a high species diversity does not necessarily coincide with a high FD and at intermediate species diversity FD is highly variable. This proposed measure is new and based on distinct functional traits determining the functional distance between species. Therefore, it overcomes existing problems with traditional diversity indices (Hurlbert, 1971) and enables new insights into phytoplankton and general ecology.

image

Figure 1. Comparison of species diversity and FD from 845 phytoplankton samples from Lake Constance over a 21-year period.

Download figure to PowerPoint

The ‘intermediate disturbance hypothesis’ and ecosystem function

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

Over the past few decades ecologists have searched for the main factors driving diversity. Among others, the frequency and intensity of disturbances (the intermediate disturbance hypothesis (IDH); Connell, 1978), the productivity of a system (e.g. Rosenzweig, 1995) and predation (including herbivory) have been shown to influence diversity directly. The IDH states that diversity peaks at an intermediate frequency or intensity of disturbance as a result of the co-existence of pioneers, stress-tolerants and ruderals. Thus, the IDH relies on the concept of FD, but empirical studies testing this hypothesis have concentrated almost entirely on species diversity (but see Weithoff et al., 2001). This same inconsistency appears when species richness (species number per unit area/volume) is considered. Calculation of FD overcomes this inconsistency and ecologists are encouraged to perform studies on the IDH in its original sense using FD.

In recent years the role of diversity has been investigated with respect to ecosystem function (e.g. Schulze & Mooney, 1993; Díaz & Cabido, 2001; Kinzig, Pacala & Tilman, 2001), but no defined measure of FD was used. I suggest the use of FD, in the way proposed in this paper, to compare different studies and provide a more objective way of characterising FD. In aquatic sciences, FD has often been neglected and the consequences for ecosystem processes of FD in the phytoplankton should be investigated in the future.

The ‘insurance hypothesis’ and functional redundancy

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

Important questions in community ecology are ‘Why are there so many species?’ (Hutchinson, 1961) and ‘what role do the rare species play in a given community?’ (Gaston, 1994, 1996). In many habitats, a few species dominate the biomass of a community whilst many others share a very limited amount. The question arises, therefore, as to whether the rare species are specialised and have found their ecological niche or whether they are in a transient state of being outcompeted or having recently colonised the habitat. Occupying a distinct niche makes a species less susceptible to competitive exclusion. The insurance hypothesis states that rare species form a functional backup for dominant species, and that they can respond quickly to disturbance thus increasing the resilience of ecosystem processes. There is theoretical and experimental evidence from plankton and other communities supporting this hypothesis (McGrady-Steed, Harris & Morin 1997; Naeem & Li, 1997; Yachi & Loreau, 1999; Fonseca & Ganade, 2001). Studies on the insurance hypothesis should be based on quantified relative functional redundancy among species, as is possible if the proposed procedure is applied.

Ecosystem shifts

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

Ecosystems continuously undergo changes caused by season, climate change or multiple anthropogenic impacts, such as eutrophication, chemical pollution or disturbances in the cycling of nutrients or water. Such changes may cause gradual or rapid shifts from one state to another (Scheffer et al., 2001). In each case, the community exhibits a directional change which can be traced by ordinating the community or its centre of gravity at any given time in an ecological space based on the abundance of the functional entities present. A functional analysis of phytoplankton long-term data may reveal such directional shifts, and inter-lake comparisons are facilitated by the functional ataxonomical approach. Such shifts in communities may then be related to climate change, eutrophication or other factors on a broader geographical scale. This will increase our understanding of perturbed systems and provide a guide to better management.

Conclusions

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

The aim here is to encourage phytoplankton ecologists to adopt functional concepts and to apply them in phytoplankton research. The selection of functional traits is crucial and requires broad discussion among researchers working in different types of lakes. The traits proposed in this contribution are also open to debate and different traits or different modalities may in fact prove more useful. One of the trait selection criteria was practicability. Many phytoplankton data sets already exist and the proposed traits could offer new insights into phytoplankton and general ecology, without laboriously collecting new data. Compared with higher plants, generation times in algae are very short and, even within a season, true succession takes place (Sommer et al., 1986). This should allow plankton ecologists to investigate hypotheses of general ecological interest and lead to a better perception of limnological studies within the general ecological community (Reynolds, 1998). The application of functional classifications and FD is by no means an argument for neglecting a careful species determination and/or autecological research, although it is seen as a very valuable addition to traditional taxonomic approaches and to the list of phytoplankton associations proposed by Reynolds et al. (2002).

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References

This manuscript benefited greatly from the comments by F. Gervais, G. Fussmann, U. Gaedke, V. Bissinger and E.M. Bell. A. Hildrew helped with clarity. Th. Kumke gave valuable statistical advice. The data in Fig. 1 were acquired within the Integrated Research Project (SFB) 248 ‘The cycling of matter in Lake Constance’ and successor projects.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Selection of functional traits
  5. Data processing and the assignment of functional groups and functional diversity
  6. Applications of functional classifications and FD
  7. The ‘intermediate disturbance hypothesis’ and ecosystem function
  8. The ‘insurance hypothesis’ and functional redundancy
  9. Ecosystem shifts
  10. Conclusions
  11. Acknowledgments
  12. References
  • Burns C.W. (1968) The relationship between body size of filter-feeding Cladocera and the maximum size of particles ingested. Limnology and Oceanography, 33, 675678.
  • Carrick H.J., Aldridge F.J. & Schelske C.L. (1993) Wind influences phytoplankton biomass and composition in a shallow, productive lake. Limnology and Oceanography, 38, 11791192.
  • Connell J.H. (1978) Diversity in tropical rain forests and coral reefs. Science, 199, 13021310.
  • Díaz S. & Cabido M. (2001) Vive la différence: plant functional diversity matters to ecosystem processes. Trends in Ecology and Evolution, 16, 646655.
  • Elliott J.A., Reynolds C.S., Irish A.E. & Tett P. (1999) Exploring the potential of the PROTECH model to investigate phytoplankton community theory. Hydrobiologia, 414, 3743.
  • Fängström I. & Willén E. (1987) Clustering and canonical correspondence analysis of phytoplankton and environmental variables in Swedish lakes. Vegetatio, 71, 8795.
  • Fonseca C.R. & Ganade G. (2001) Species functional redundancy, random extinctions and the stability of ecosystems. Journal of Ecology, 89, 118125.
  • Gaedke U. (1998) Functional and taxonomical properties of the phytoplankton community: interannual variability and response to re-oligotrophication. Archiv für Hydrobiologie Special Issues: Advances in Hydrobiology, 53, 119141.
  • Gaston K.J. (1994) Rarity. Chapman & Hall, London.
  • Gaston K.J. (1996) Biodiversity – A Biology of Numbers and Differences. Blackwell, Oxford.
  • Gervais F. (1997) Diel vertical migration of Cryptomonas and Chromatium in the deep chlorophyll maximum of a eutrophic lake. Journal of Plankton Research, 19, 533550.
  • Gitay H. & Noble I.R. (1997) What are functional types and how should we seek them? In: Plant Functional Types – Their Relevance to Ecosystem Properties and Global Change (Eds T.M.Smith, H.H.Shugart & F.I.Woodward), pp. 319. Cambridge University Press, Cambridge.
  • Grime J.P. (1977) Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist, 111, 11691194.
  • Hurlbert S.H. (1971) The nonconcept of species diversity: a critique and alternative parameters. Ecology, 52, 577586.
  • Huszar V.L.M. & Caraco N. (1998) The relationship between phytoplankton composition and physical–chemical variables: a comparison of taxonomic and morphological–functional approaches in six temperate lakes. Freshwater Biology, 40, 118.
  • Hutchinson G.E. (1961) The paradox of the plankton. American Naturalist, 95, 137147.
  • Jaksić F.M. & Medel R.G. (1990) Objective recognition of guilds: testing for statistically significant species clusters. Oecologia, 82, 8792.
  • Keddy P.A. (1992) A pragmatic approach to functional ecology. Functional Ecology, 6, 621626.
  • Kinzig A.P., Pacala S.W. & Tilman D. (2001) The Functional Consequences of Biodiversity. Princeton University Press, Princeton.
  • Kruk C., Mazzeo N., Lacerot G. & Reynolds C.S. (2002) Classification schemes for phytoplankton: a local validation of a functional approach to the analysis of species temporal replacement. Journal of Plankton Research, 24, 901912.
  • Lavorel S., McIntyre S., Landsberg J. & Forbes T.D.A. (1997) Plant functional classifications: from general groups to specific groups based on response to disturbance. Trends in Ecology and Evolution, 12, 474478.
  • Lee R.E. (1999) Phycology. Cambridge University Press, Cambridge.
  • Leishman M.R. & Westoby M. (1992) Classifying plants into groups on the basis of associations of individual traits – evidence from Australian semi-arid woodlands. Journal of Ecology, 80, 417424.
  • Martinez N.D. (1996) Defining and measuring functional aspects of biodiversity. In: Biodiversity: A Biology of Numbers and Differences (Ed. K.J.Gaston), pp. 114148. Blackwell, Oxford.
  • McGrady-Steed J., Harris P.M. & Morin P.J. (1997) Biodiversity regulates ecosystem predictability. Nature, 390, 162165.
  • McIntyre S., Díaz M., Lavorel S. & Cramer W. (1999) Plant functional types and disturbance dynamics – introduction. Journal of Vegetation Sciences, 10, 604608.
  • Melo S. & Huszar V.L.M. (2000) Phytoplankton in an Amazonian flood-plain lake (Lago Batata, Brasil): diel variation and species strategies. Journal of Plankton Research, 22, 6376.
  • Naeem S. & Li S. (1997) Biodiversity enhances ecosystem reliability. Nature, 390, 507509.
  • Pasciak W.J. & Gavis J. (1974) Transport limitation of nutrient uptake in phytoplankton. Limnology & Oceanography, 19, 881888.
  • Petchey O.L. & Gaston K.J. (2002) Functional diversity (FD), species richness and community composition. Ecology Letters, 5, 402411.
  • Pillar V.D. (1999) On the identification of optimal plant functional types. Journal of Vegetation Sciences, 10, 631640.
  • Reynolds C.S. (1980) Phytoplankton assemblages and their periodicity in stratifying lake systems. Holarctic Ecology, 3, 141159.
  • Reynolds C.S. (1988) Functional morphology and the adaptive strategies of freshwater phytoplankton. In: Growth and Reproductive Strategies of Freshwater Phytoplankton (Ed. C.D.Sandgren), pp. 388433. Cambridge University Press, Cambridge.
  • Reynolds C.S. (1993) Scales of disturbance and their role in plankton ecology. Hydrobiologia, 249, 157171.
  • Reynolds C.S. (1997) Vegetation Processes in the Pelagic. A Model for Ecosystem Theory. ECI, Oldendorf.
  • Reynolds C.S. (1998) The state of freshwater ecology. Freshwater Biology, 39, 741753.
  • Reynolds C.S. & Irish A.E. (1997) Modelling phytoplankton dynamics in lakes and reservoirs: the problem of in-situ growth rates. Hydrobiologia, 349, 517.
  • Reynolds C.S., Irish A.E. & Elliott J.A. (2001) The ecological basis for simulating phytoplankton responses to environmental change (PROTECH). Ecological Modelling, 140, 271291.
  • Reynolds C.S., Huszar V., Kruk C., Naselli-Flores L. & Melo S. (2002) Towards a functional classification of the freshwater phytoplankton. Journal of Plankton Research, 24, 417428.
  • Rosenzweig M.L. (1995) Species Diversity in Space and Time. Cambridge University Press, Cambridge.
  • Scheffer M., Carpenter S.R., Foley J.A., Folke K. & Walker B. (2001) Catastrophic shifts in ecosystems. Nature, 413, 591596.
  • Schulze E.D. & Mooney H.A. (1993) Biodiversity and Ecosystem Function. Springer-Verlag, New York.
  • Sicko-Goad L., Stoermer E.F. & Fahnenstiel G. (1986) Rejuvenation of Melosira granulata (Bacillariophyceae) resting cells from anoxic sediments of Douglas Lake, Michigan. 1. Light microscopy and 14C uptake. Journal of Phycology, 22, 2228.
  • Smith T.M., Shugart H.H., Woodward F.I. & Burton P.J. (1993) Plant functional types. In: Vegetation Dynamics and Global Change (Eds A.M.Solomon & H.H.Shugart), pp. 272292. Chapman & Hall, New York.
  • Smith T.M., Shugart H.H. & Woodward F.I. (1997) Plant Functional Types – Their Relevance to Ecosystem Properties and Global Change. Cambridge University Press, Cambridge.
  • Sommer U., Gliwicz Z.M., Lampert W. & Duncan A. (1986) The PEG-model of seasonal succession of planktonic events in fresh waters. Archiv für Hydrobiologie, 106, 433471.
  • Usseglio-Polatera P., Bournaud P., Richoux P. & Tachet H. (2000) Biological and ecological traits of benthic freshwater macroinvertebrates: relationships and definition of groups with similar traits. Freshwater Biology, 43, 175205.
  • Walker B., Kinzig A. & Langridge J. (1999) Plant attribute diversity, resilience, and ecosystem function: the nature and significance of dominant and minor species. Ecosystems, 2, 95113.
  • Weiher E., van der Werf A., Thompson K., Roderick M., Garnier E. & Eriksson O. (1999) Challenging Theophrastus: a common core list of plant traits for functional ecology. Journal of Vegetation Sciences, 10, 609620.
  • Weithoff G., Lorke A. & Walz N. (2000) Effects of water-column mixing on bacteria, phytoplankton, and rotifers under different levels of herbivory in a shallow eutrophic lake. Oecologia, 125, 91100.
  • Weithoff G., Walz N. & Gaedke U. (2001) The intermediate disturbance hypothesis – species diversity or functional diversity? Journal of Plankton Research, 23, 11471155.
  • Yachi S. & Loreau M. (1999) Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. Proceedings of the National Academy of Sciences USA, 96, 14631468.