, Volume 633, Issue 1, pp 197–211 | Cite as

Assessment of the ecological status of European surface waters: a work in progress

  • Peeter NõgesEmail author
  • Wouter van de Bund
  • Ana Cristina Cardoso
  • Angelo G. Solimini
  • Anna-Stiina Heiskanen


The ‘normative definitions’ of ecological water quality classes given by the Water Framework Directive (WFD) are narrative descriptions of the conditions present in water bodies of different qualities relative to reference conditions found in unimpacted sites. In order to fill these descriptions with a more solid content, the definitions have been a subject of intensive development of quantitative methodologies for ecological status assessment as well as for rules and criteria for setting of reference conditions and ecological status boundaries for classification of water bodies. In this article, we recall the basic principles of the WFD that sometimes have been overlooked and point out some gaps remained and problems arisen during the ongoing implementation of the directive. Defining type-specific reference conditions for water bodies and finding biological metrics that sensitively reflect only the anthropogenic deviations from those conditions are the biggest challenges that the ecological status assessment faces. So far, there is no guarantee that reference conditions are comparable across EU Member States due to a lack of common criteria, which need still to be elaborated. Defining site-specific reference conditions instead of type specific is a novel approach that allows for minimizing uncertainties introduced by applying broad types. Search for new metrics has led to a real boom of multimetric indices which ought to be the adequate tool to measure the multiple human impairments but which should pass a thorough check before being included in monitoring programs. Curiously, some biological indices constructed as surrogates for chemistry (especially nutrients) start ‘living their own life’ and continue indicating the disturbance when the controlling factors change. This shows the obvious advantage of biological indicators against chemical ones. New challenges to WFD implementation are brought about by the need to consider the effects of alien species and climate change in the assessment framework, and by the nonlinear dose—response relationships dominating in biological systems. Attempts to diminish uncertainties in quality assessment have become a new labour-intensive field for researchers.


Water Framework Directive Intercalibration Multimetric index Biological indicators One out/all out principle 


The launch of the Water Framework Directive (WFD; Directive, 2000) to provide a holistic framework for management and protection of all waters in the European Union (EU), was parallel to the development of the Ecosystem Approach (EA), which was adopted as the primary framework for action under the convention of the biological diversity (CBD; UNESCO, 2000) for management and conservation of nature. The WFD follows many principles of the EA which is a holistic framework for protection of biodiversity and for the sustainable management of natural resources in the context of the ecological, socio-economical and cultural aspects. The WFD includes management at the lowest appropriate level (river basin management) in accordance with common principles and environmental objectives, stakeholder involvement in the definition of management objectives, restoration goals and planning of measures. It requires initial characterization of water bodies, and analysis of human pressures impacting the aquatic ecosystems, assessment of ecological and chemical status, and economic analysis of cost-effective protection measures. At present, the recently adopted EU Marine Strategy Framework Directive (Directive, 2008) is following the principles of EA, while providing less detailed definitions on the environmental objectives and criteria for the good environmental status than the WFD. This may be an advantage as it allows more flexible approaches that can be adapted for regional seas that are different in many ecological features, but on the other hand, it may also make diversification of approaches in the setting of ‘good environmental status objectives’ between the regional seas.

The WFD implements the principles of the EA requiring management and protection of the aquatic ecosystems as whole, ensuring sustainable functioning and healthy structure of the ecosystems, as well as acknowledging the critical role of human societies in managing those systems. However, contary to that stated in the CBD definition of EA: The objectives of management of land, water and living resources are a matter of societal choice (UNESCO, 2000), the WFD requires definitions of criteria and standards for good ecological status that is the objective for protection and restoration of water bodies. The interpretation and application of Article 4 in the Directive, which describes the environmental objectives and sets out the so-called exemptions, does not compromise the rule that the criteria for ‘good status’ are based on the scientific knowledge of the aquatic ecosystems and set by the intercalibration exercise between the Member States (CIS, 2009). However, there may be differences in views of the level of ambition how the criteria for good ecological status have been defined and applied in practical implementation of the Directive (e.g., Moss, 2008), and, of course, expert opinions are necessary in all stages of the assessment. These criteria (i.e. the ‘normative definitions’) are described in narrative terms in the Annexes of the Directive, and have been a subject of intensive development of quantitative methodologies for ecological status assessment as well as for rules and criteria for setting of reference conditions and ecological status boundaries for classification of water bodies. In this article, the authors recall the basic principles of the WFD, which sometimes have been overlooked or misinterpreted and point out some gaps remained and problems arisen in the ecological status assessment during the ongoing implementation of the directive.

Ecological status and reference conditions

The Ecological Status assessment of the WFD combines information on several hydromorphological, chemical and biological parameters to acquire a comprehensive picture of the overall status on functioning and structure of the ecosystem. Hydromorphological (e.g. depth, flow rate, salinity, sediment types, etc.) and some chemical parameters (e.g. alkalinity) are used to identify water body types, which are sufficiently similar while chemical parameters (e.g. nutrient and oxygen concentrations) are supporting parameters that are applied to identify reference conditions and good status (CIS, 2003a, b).

Biological parameters are the basis for the ecological status assessment. WFD requires that the status assessment of the water bodies is made based on biological ‘quality elements’, such as phytoplankton, macrophytes and phytobenthos, macroalgae and angiosperms, macroinvertebrates and fish, combination of which is depending on the surface water category (e.g., fish is not a quality element in coastal waters, but required for all other surface waters including transitional waters). The initial status assessment should be based on all biological quality elements required for each specific water category to obtain an overall view of the status of biological communities. Biological quality elements express also the sensitivity to different pressures. Commonly, phytoplankton is reacting first to human impact and provides a fast response to, e.g. eutrophication pressure, while macrophytes and fish may be more sensitive for various degrees of hydromorphological pressures (e.g., Solimini et al., 2006).

The selection and development of indicators representative for the biological quality elements are crucial in the assessment process. For instance, chlorophyll a measurements are widely used as a proxy for phytoplankton biomass. Moreover, also the way of calculation of the metrics is important to provide required robustness to the indicator. For instance, if seasonal monitoring is carried out for phytoplankton chlorophyll, the indicator calculated should make use of all available data instead of reducing datasets for summer period only, unless it is proven that a more precise indicator can be obtained from a subset of data (Carstensen, 2007). In the case of coastal phytoplankton, either the 90th percentile or the summer mean are most widely used as metrics for chlorophyll a (Carletti et al., 2009). The robustness of the selected indicator metrics is crucial, since several biological parameters with a large natural variability cannot be used for ecological status assessment as it is not possible to identify reference conditions for them with sufficient accuracy.

One of the big challenges of the WFD is to find common approaches for defining reference conditions that act as the anchor point for the ecological assessment. A very purist approach—defining reference conditions as the total absence of any anthropogenic influence on the ecosystem—would leave us with no real reference sites at all. This realization could lead to the other extreme where the concept of reference conditions is abandoned and replaced with a focus on ecosystem services and human benefits (Dufour & Piégay, 2009). In practice, however, compromises between these two extreme approaches are usually found, e.g. if reference conditions are based on measurements from reference sites with a minimum level of anthropogenic pressure. The weak point so far is that there is no guarantee that reference conditions are comparable across Member States due to a lack of common criteria; further study will be needed to resolve this issue. Setting reference conditions is especially difficult for those water body types lacking credible natural reference sites, including lowland rivers, large rivers and lakes, and certain types of estuaries and coastal waters.

As the driver or pressure is diminishing, the recovery of ecosystems may not follow a similar pathway and to return linearly to the initial state that prevailed before the pressure increase. In a recent article, Duarte et al. (2009) describe four coastal case studies where nutrient loading reductions did not bring a decline in phytoplankton biomass to about the level before nutrient enrichments. Changes in the species composition, internal ecosystem processes and food web interactions may have caused a shift in the ecosystems and the restoration of the previous stage may not be possible, but after the pressure relief a new alternative steady state will emerge. They raised a question whether static reference conditions are justified or instead shifting base lines should be accepted to provide realistic prospects and public motivation for restoration and ecosystem recovery.

Similarly, the changing and warming climate is affecting many characteristics of surface waters. As a result, some water bodies, especially those located near the type boundaries, may even change their type. Compared to typology characteristics, water quality parameters and biological indicators are even more labile and may be easily affected by climate change (Nõges et al., 2007). The review of the river basin characterization every sixth years, as required by the WFD, needs also encompassing re-evaluation of reference conditions according to the changes observed using potential reference site monitoring networks, or modelling tools that enable distinction of climate change effects. As a consequence, the restoration targets (i.e., the good ecological status) would also need to be evaluated periodically, accepting that some changes (as loss or immigration of some species, or permanently elevated biomass levels of phytoplankton being fed by internal sources of nutrients) have become permanent.

Intercalibration exercise

There is a large variety of indicator metrics, both single and multimetric indices, which are developed for different water categories, ecoregions and surface water types and applied by the different EU Member States. The comparativity of such metrics and classification results based on these indicators is one of the key questions at the EU level. The WFD intercalibration aims to provide a process for comparison of assessment results on quality element or on single parameter level. In many cases the assessment methods were still in development at the time of the intercalibration exercise that had to be carried out based on a limited amount of available data. Results of the intercalibration exercise have been achieved for the most widely used methods (Carletti et al., 2009; Poikane, 2009; van de Bund, 2009), but has not been finalized; some of the more challenging quality elements are yet to be addressed. Because the focus of existing and often more traditional methods with the results that have been laid down in a legal document, there is the risk that intercalibration will limit the development of new and innovative approaches. This leaves a considerable challenge for the scientific community to develop methods that are ecologically sound and to show how they can improve current methodologies. On the positive side, the intercalibration exercise has catalysed an exchange of information in the expert groups, leading to more comparable approaches in ecological quality assessment.

Biological elements

Annex V of the WFD sets out quite clearly the biological elements that need to be monitored for freshwater and coastal ecosystems. However, it allows the elements with high natural variability to be excluded from classification schemes. This exemption together with the reduced probabilities of a correct classification with increasing number of elements used, have been considered responsible for a restriction to the range of biological elements sampled in routine monitoring in comparison with the Annex V shopping list (Irvine, 2004). Indeed, a Commission Decision on Intercalibration (European Commission, 2008) published the classification criteria for only a fraction of the required biological elements and parameters (e.g., taxonomic composition, abundance) within the elements. This may not only reflect difficulties associated to the development of biological assessment methods but also a cautious approach to the inclusion of several biological elements in classification.

Not only the number of biological elements but also the choice of the elements included in the Directive has been the subject to criticism. The omission of zooplankton among the biological elements for the assessment of lakes and coastal areas has been particularly criticized (e.g. Moss, 2007) given the important role of this group in reflecting food web efficiency (e.g., Calbet & Landry, 2004; Jeppesen et al., 2004). In particular, zooplankton plays a key role in shaping shallow lakes’ food webs—whereby in clear water state these lakes are dominated by submerged macrophytes—used by zooplankton as refuge against planktivorous fish (e.g. Schriver et al., 1995; Stansfield et al., 1997), and high zooplankton/phytoplankton ratio suggesting zooplankton control on phytoplankton biomass (e.g. Schriver et al., 1995; Søndergaard & Moss, 1998).

Another important subject to criticism regards the exclusion of fish from the coastal waters assessment. The role of fishes in the food web structure and ecological process in coastal water is well known. The magnitude of the direct pressure on this element (i.e. through fishing) and its consequent cascade effects on food webs (e.g. Daskalov, 2002) is also well documented. Nonetheless, the Directive allows for some flexibility in inclusion of other biological elements, if their importance for the ecological assessment is justified as it allows for exemptions to mandatory biological elements. However, the already significant monitoring efforts may discourage Member States from including elements other than those directly prescribed by the WFD.

Another gap regards the assessment of ecosystem impacts from alien species (Cardoso & Free, 2008). Although these species are not specifically mentioned in the text of the WFD, they represent a significant biological pressure as many of them have an invasive behaviour and can alter the native biological structure and ecological processes of aquatic systems (e.g. Olenin et al., 2007). According to the WFD Annex II, alien species can fall in the category of ‘other pressures’. Whereas, in the Directive’s Annex V, the definition of ‘high status for each biological quality element’ states the taxonomic composition corresponds totally or nearly totally to undisturbed conditions, therefore the departure from naturalness underlines the entire ‘spirit’ of the WFD, and the presence of alien species indicates such a state. Thus, alien species simultaneously indicate high pressure and poor or bad ecological status and should be considered in ecological assessment tools.

In reality, alien species are a widespread problem affecting all biological quality elements but water body management is based on the assumption that ecosystems naturally improve following reductions in human impacts. However, alien species are a self-perpetuating pressure which is very difficult to eradicate once established, and ecological status declines accordingly. This problem is further aggravated because the impacts of alien species are not reliably measured by the classification tools that have been developed for the WFD. For example, invertebrate kick sampling does not adequately collect species such as crayfish and family level identification may fail to detect some alien species.


(i) Type-specific or site-specific approach?: The efficiency of water body assessment depends most crucially on two aspects: (1) the appropriate typology that adequately takes into account regional geographic and climatic differences and guarantees the correct application of reference conditions, and (2) the selection of right metrics that change quantitatively and consistently across a range or gradient of human influence.

As shown by Borja et al. (2009), WFD typology schemes have often been criticized for being oversimplified and not reflecting the high heterogeneity of habitats within the different types and water bodies. Transitional waters represent especially big challenges for typifying and intercalibration on a European scale, due to the heterogeneity of water bodies including wide size spectra of both estuaries and coastal lagoons with a large variety of connections to the open sea and to continental waters. Defining site-specific reference conditions is a novel approach that allows minimizing uncertainties introduced by applying broad types. In large unique water bodies for which long-term historical data exist, this approach has been proved useful (Lyche Solcheim, 2005). With the help of modelling, however, site-specific reference conditions can be determined also for a larger variety of water bodies (e.g., within broad types) as functions of watershed attributes. Carvalho et al. (2009) recommend site-specific reference conditions for chlorophyll in lakes in preference to type-specific reference conditions, as they use the individual characteristics of a site known to influence phytoplankton biomass, rather than adopt standards set to generally represent a large population of lakes of a particular type.

(ii) Difficulties finding appropriate metrics: In 1997, in an article that has became a classic in the field of biomonitoring of surface waters, Karr & Chu formulated most of the leading principles that form the basis of the nowadays WFD: the concept of good water status (‘biotic integrity’ in their terminology), the need for a water body type-specific approach, the supremacy of biological indicators and the assessment scheme based on comparison of present status with reference conditions. According Karr & Chu (1997), selection of those among the multitude of measurable biological attributes that provide reliable and relevant signals about the biological effects of human activities but remain insensitive to extraneous conditions is a step of extraordinary importance. Only these attributes are called metrics.

The authors warn about attempts of measuring all possible attributes and argue that some attributes are poor candidates for monitoring metrics because of their underlying biology. In particular, they indicate abundance, population size, density and production as attributes that vary too much even in pristine areas to be reliable indicators of human influence. According to Karr & Chu (1997), the belief that biology is too variable to be of practical use in monitoring comes not from a lack of good indicators but from the fact that studies of naturally variable attributes such as population size, density and abundance have dominated ecology for >50 years. They show also that ratios (e.g., of the abundances of two trophic groups), although they may intuitively seem useful, are inherently flawed as they mix independent parameters in ways that make it hard to discern their relative influence. Further, two large numbers and two small numbers may yield the same ratio, although the biological meaning of small and large numbers may be very different. As one of the rare exceptions, relative abundance (number of individuals in a taxonomic or other target group divided by the total number of individuals in the sample) represents ratios that may reveal consistent dose–response relationships.

Karr & Chu (1997) also show that a small number of attributes of biological elements—such as taxa richness and percentages of individuals belonging to tolerant taxa—consistently emerge as reliable indicators of biological condition at sites influenced by different human activities in different geographic areas.

Moss (2008) goes further from such a simple and empirical division of ecological attributes into good and bad metrics for measuring human impact and formulates four primary characteristics of ecological quality which intactness should be measured to decide about impactedness of sites. These primary characteristics include nutrient parsimony (efficiency in recycling scarce materials), characteristic of physical settings and food web structure that ensure this parsimony and maintenance of the intactness of the system as a whole structure in its several aspects, connectedness and size. Moss argues that secondary features such as concentrations of substances or lists of species do not adequately measure these fundamental characteristics, unless these are used in ways to diagnose the state of the more fundamental characteristics.

In the light of the above mentioned arguments by Karr & Chu (1997) and Moss (2008), a question arises as regards how to evaluate the common practice of using chlorophyll concentration as one of the main biological metrics in WFD implementation (Carvalho et al., 2009; Carstensen & Henriksen, 2009; Wolfram et al., 2009). On the one hand, chlorophyll clearly belongs to the abundance/population size or density attributes criticized for their large variability, but, on the other hand, increase of chlorophyll and algal biomass are symptomatic signs of alteration in nutrient parsimony and food chain structure related to eutrophication and as such are included also in the normative definitions of ecological status classes for all the surface water categories. Obviously, Karr & Chu (1997) in their division of biological attributes to suitable and non-suitable metrics for measuring human impact had in mind macroinvertebrates or fish rather than phytoplankton which, as a primary producer, has its own specific features.

New concepts in ecology stemming from the theory of ecosystem resilience and alternative stable states affect also the selection of metrics and their calibration against pressure gradients. As Borja et al. (2009) note, the selection of ecologically relevant attributes to be used within any methodology, should emphasize metrics that respond to both degradative and restorative anthropogenic action. One aspect that complicates this approach is the different lags of biotic metrics to changes in pressure. A long-term study carried out in Lake Mondsee (Dokulil & Teubner, 2005) showed that reductions in phytoplankton biovolume were delayed by about 5 years relative to reduction in total phosphorus (TP) concentration while phytoplankton species differed in timing of their responses to changing nutrient conditions. While Planktothrix rubescens declined concomitantly with the decline in TP concentration, other species indicative of higher phosphorus concentrations, such as Tabellaria flocculosa var. asterionelloides, tended to increase further. In such situations, the efficiency of restoration measures can be evaluated differently depending on how closely the selected metric follows the changes in pressure.

(iii) Are biological indicators surrogates for chemical analyses?: For standing waters (lakes, transitional and coastal waters), eutrophication is a major pan-European problem and the biggest threat for biodiversity (Watt et al., 2007) and reaching good ecological status of surface waters (CIS, 2005). Phytoplankton is the first group of organisms to respond to nutrient enrichment with excessive growth and shifts in community composition resulting in increased turbidity, subsequent loss of submerged aquatic vegetation and oxygen deficiency in bottom waters. A number of trophic state indices have been created based on simple parameters such as TP, Secchi depth and chlorophyll concentration (Carlson, 1977) or scores of algal (LAWA, 1999) or macrophyte species (Schneider & Melzer, 2003) given to them according to phosphorus ranges in which they occur. Even the taxon-specific optima of littoral invertebrates along a TP gradient have been used to develop an ecological classification model (Donohue et al., 2009). The latter type of indices has been criticized for being surrogates for driving pressures (nutrients), against which they have been calibrated (Moss, 2008). Moss points out that such pressures can be independently chemically measured with greater accuracy, but admits also that the biological community has an integrating component that spot analyses do not have. It seems to be rather a rule rather than an exception that biological communities affected by eutrophication show a delayed recovery in the oligotrophication phase (Dokulil & Teubner, 2005; Hajnal & Padisák, 2008; Padisák & Reynolds, 1998) or do not reflect at all the decrease in phosphorus (Kaiblinger et al., 2009; Nõges et al., submitted) even if indices initially calibrated against phosphorus are used. Obviously, these phytoplankton indices are able to capture ecological status in circumstances where factors other than nutrients (e.g. changes in light conditions, food chain or salinity; Borics et al., 2000; Padisák et al., 2006) take over the control over community structure creating a resistance to restoration efforts.

A trophic index although relating community composition to eutrophication pressure, cannot be used for a WFD-compliant assessment of ecological quality unless reference conditions and class boundaries are defined. The species list and scoring system for species may remain the same for all lake types, for example, in the Swedish (Willén, 2007) or Austrian system (Dokulil et al., 2005) or different species lists and sensitivities are applied for different types, for example in German system (Mischke et al., 2008).

WFD represents a transition from chemically to ecologically based assessment of water quality. In principle, the ecological quality assessment must be based on the structure and functioning of aquatic ecosystems. Thus, there is a need to assess the impact of toxic substances on aquatic populations and communities. A probabilistic, community-based approach for defining environmental quality and assessing risk for ecosystems is to calculate the number of species potentially affected by a given concentration of a toxic chemical assuming that the sensitivity of different species toward a given stress factor (e.g. a toxic chemical) follows a normal distribution (see Vighi et al., 2006 for summary). The benefit of this approach is that it is based on the structure of the biological communities present in a given environment and not on toxicological data produced on standard organisms. On the negative side, data for such assessment of toxicity impacts on community or population level are not yet available. This would require deeper knowledge on structure and functioning of natural communities as well as toxicity data on a large number of organisms. At present such data is available only for a very low number of chemicals. On the other hand, toxicity tests on the basic ecotoxicological test organisms like algae, Daphnia, and fish, are no longer sufficient for assessing toxicological stress on the communities (Vighi et al., 2006).

(iv) Is there hope to discover new powerful metrics?: Since the first attempts to use biological indicators in water quality assessment (Kolkwitz & Marsson, 1908, 1909), vast amounts of information on biological indicators have been accumulated and certain traditions established, e.g. that of using macroinvertebrates for river assessments and phytoplankton for lakes. The implementation of the WFD has induced a massive search for new biological methods to measure the anthropogenic impact on water bodies. We may ask: Is there any hope to discover new powerful metrics or develop more sensitive and selective methods? After 15 years’ research in selecting metrics for ecological status assessment of water bodies, Karr & Chu (1997) came to a striking conclusion that the same major biological attributes arose recurrently and served as reliable indicators in diverse circumstances. The recent proliferation of indices is so extensive that the assessment and comparison of their suitability for quality assessment have become a new labour-intensive field for researchers that adds an element of confusion back into what environmental managers had hoped to simplify (Borja et al., 2009). On the other hand, the intensive research, especially in less traditional areas, has already yielded and will yield in the future new metrics, methods and analysing techniques more specific for the widening gamma of pressures.

(v) Multimetric approaches for multivariately determined conditions: One of such fast developing areas is the construction of multimetric indices. The theoretical basis of this approach has reached already a maturity level where ‘cook books’ for designing such indices (Hering et al., 2006) can be produced to guide researchers to the best solutions and helping them to avoid the already known pitfalls. As the relationships between occurrence of particular species, genera or groups and environmental conditions are multivariately determined (Moss, 2007), a multimetric index attempting to represent the multitude of impacts on different biotic levels, ought to be the adequate tool to measure the multiple human impairments. By combining different categories of metrics (e.g. taxa richness, diversity measures, proportion of sensitive and tolerant species, trophic structure) reflecting different environmental conditions and aspects of the community, the multimetric assessment is regarded as a more reliable tool than assessment methods based on single metrics (Hering et al., 2006). The best multimetric indices are more than a community-level assessment because they combine measures of condition in individuals, populations, communities, ecosystems and landscapes (Karr & Chu, 1997). Since the first multimetric index, the Index of Biotic Integrity developed by Karr (1981), which used fish as stream quality indicators, numerous multimetric indices have been developed. In this issue, the Austrian Macrophyte Index (Pall & Moser, 2009) based on metrics like ‘vegetation density’, ‘vegetation limit’, ‘characteristic zonation’, and ‘species composition’ represents a multimetric approach.

A novel approach—an extension to environmental science currently a matter of study—is the use of Fuzzy Logic—a way of using linguistic variables to describe and assess complex systems. It is becoming popular in Spain in studies related to the implementation of the WFD (Ocampo-Duque et al., 2007; see also Borja et al., 2009). One of the main advantages of fuzzy logic is the ability to model expert human knowledge, a necessary feature to be considered in the complex process of environmental management. In a case study where the fuzzy model approach was applied in the Ebro River Basin (Ocampo-Duque et al., 2006), the four components surveyed corresponded rather well to the four primary characteristics of ecological status listed by Moss (2008): total riparian vegetation cover (= size), cover structure (= characteristic structure in its several aspects), cover quality (≈nutrient parsimony), and channel alterations (= connectedness).

(vi) Dominant versus less frequent taxa and the question of taxa sensitivity: Discussing the information that biological samples contain, Reynolds (1998) wrote: ‘An experienced plankton scientist, confronted with an undefined sample of phytoplankton under a moderate quality microscope, would have little difficulty in gleaning sufficient information from the assemblage to be able to deduce a lot about the water body from which it was collected—whether oligotrophic or eutrophic, acidic or alkaline, shallow or deep, mixed or stratified and, in all probability, what month of the year it was taken.’ The same is valid not only for phytoplankton but also for other major biological groups (or ‘elements’ in the WFD vocabulary). The question is, which components of the biological sample contain this information: is it coded in the dominants or in some minor indicator species?

Based on a general assumption that a healthy environment will have the species appropriate to it, a study (Devlin et al., 2009) was carried out in the United Kingdom to identify whether abundance and species richness of 20 most frequently occurring marine phytoplankton species were predictable across seasons and typology, and whether the deviation away from a reference condition in different risk conditions was sufficient to apply phytoplankton measurements into a management tool for assessment of environmental quality. The preliminary study showed that it may be possible to have one or very few generic lists of species that should occur at a particular time of year and could be used for ecological assessment of marine waters.

On the other hand, using the most frequent species with respect to ecological status classification has been criticized (Rawson, 1956), as it appears that the dominant species are often those with rather wide tolerance (eurybionts), and thus, they may be even less indicative of aquatic condition than the less frequent species. In geobotany, the fidelity concept, one of the most important aspects of the Braun-Blanquet (1925) approach, has been widely used to diagnose particular types of natural plant communities. Fidelity is the degree to which a species is concentrated in a given vegetation unit, while the most specialized ones, often occurring with medium or low frequencies, are called diagnostic species that are used to delimit associations. Their diagnostic power, however, has been tested only for natural environments (reference conditions) and cannot tell us anything about human-caused degradation. In order to find out, which species or metrics merit watching for these signs, analysis should focus on species tolerance and community changes across the range from minimal to severe human impact. This approach has yielded several lists of aquatic organisms divided into sensitive, indifferent and tolerant taxa. In some cases, this search for indicators has led to in-depth studies of a single species, applying different metrics combined into a multimetric index (Lopez y Royo et al., 2009).

Information on taxa sensitivity or tolerance has been used in the assessment methods in various ways. The mere presence of very sensitive taxa is a strong indicator of good ecological status while their relative abundance should not necessarily be high. Presence alone of tolerant taxa, on the other hand, says little about ecological status since tolerant groups inhabit a wide range of places and conditions, but as conditions deteriorate, their relative abundance rises (Karr & Chu, 1997). Most community composition indices, however, do not consider this difference and use the abundance-based approach (species score multiplied by abundance) for both sensitive and tolerant species. Results of such approach will be strongly biased towards the tolerance level of the dominant species. Karr & Chu (1997) recommended that only about 10% of taxa in a region should be identified as sensitive or tolerant avoiding the use of the remaining 70–90% taxa with intermediate tolerances, which would swamp the strong signal coming from presence of the most sensitive or most tolerant taxa. A compromise option that takes into account both the abundance and the diagnostic power of species is to add a weighing factor to the index that depends on the stenoecy, i.e. the species tolerance range, as it has been done, for instance, in the German Phytoplankton-Taxa-Lake-Index (Mischke et al., 2008).

One-out–all-out principle. What does the practice to date show?

Until now, the focus of research supporting the implementation of the WFD has been on the development of individual indicators at the level of the Biological Quality Element (BQE) or single parameters within a BQE; hundreds of such studies proposing such indicators have appeared in the scientific literature in recent years, including some articles in this volume (e.g. Kaiblinger et al., 2009; Ptacnik et al., 2009; Pall & Moser, 2009; Donahue et al., 2009; Lopez y Royo et al., 2009). Also, the intercalibration exercise has been carried out at the BQE level. For the final classification of water bodies, it is of crucial importance how classification results for the different BQEs are combined in an assessment of ecological status. The WFD prescribes the ‘one-out–all-out’ approach, where the BQE showing the worst classification determines the ecological status. Moss et al. (2003) showed that the one-out–all-out principle will tend to downgrade sites unjustifiably, depending on the number of metrics included in the assessment. Errors within individual quality elements and metrics tend to show considerable variation (Johnson et al., 2006). If these are combined using the one-out–all-out principle, reliable, precise metrics tend to be overruled by less reliable metrics in a large proportion of water bodies. Alternative approaches to the one-out—all-out principle could be considered, where pressures are more specifically taken into account. Now that WFD monitoring programmes have been established in most EU member states and data will be available, it will be possible to evaluate the strengths and weaknesses of the one-out–all-out and alternative approaches.

Variability in ecological status assessment and resulting classification

Ecological indicators typically exhibit non-linear responses to anthropogenic pressures together with great spatial and temporal variability, which underpin emerging key characteristics of natural systems (Solimini et al., 2009). Predicting the causes and consequences of human activities on freshwater ecosystems requires a quantitative assessment of the level of uncertainty associated with our understanding of ecological structure and processes. The assessment based on Ecological Quality Ratios (EQRs) ultimately results in the assignment of a given site to a certain ecological status class. However, to what extent are we confident that the resulting classification is the true one for that site? Here, the problem is to estimate the level of confidence of the class assignment for a given site or, in other words, the risk of site misclassification which is relevant for decision makers.

The assessment of ecological status and the resulting classification of water bodies are complicated by the inherent variability of biological communities. In the normative terms, the WFD refers to uncertainty as ‘level of confidence’ and/or level of ‘precision’ when dealing with reference conditions (Annex II) and when dealing with monitoring (Annex V). In the first case, the methods and the number of sites used to set spatially derived reference conditions must provide a sufficient level of confidence about the values for the reference conditions to ensure that the conditions so derived are consistent and valid for each surface water body type. In the case of monitoring, estimates of the level of confidence and precision of the results provided by the monitoring programmes must be given in the river management plan. The frequency of monitoring should be chosen so that an acceptable level of confidence and precision are achieved. Accordingly, several recent articles on WFD compliant metrics, whole assessment systems or reference conditions include an estimation of uncertainty basically linked to spatial or temporal factors (Kelly et al., 2009; Carstensen & Henriksen, 2009; Carvalho et al., 2009; Erba et al., 2009; Wolfram et al., 2009). However, the total uncertainty associated with the classification of ecological status reflects also technical, economical and political constraints of the assessment methods (Fig. 1).
Fig. 1

Sources of uncertainty associated with the classification of ecological status of water bodies

Abiotic factors in freshwater ecosystems are highly variable in time and space and often account for a significant proportion of the variation of community patterns in terms of species diversity, abundance, biomass and production. Therefore, most of the variances in EQR-derived classification probably derive from the unspecified sampling variability, resulting from this spatial heterogeneity of abiotic factors at multiple time scale that, in turn, affect the biotic community structure. In addition, other sources of variation include (Fig. 1): the choice of sampling sites within a region or catchment (how many sites?), the sampling effort (how many samples and of what dimension?) and the frequency of sampling (how often?). Imposed over the natural variation of biological metrics, the human activities produce different pressures that interact with abiotic and biotic factors in shaping the structure of communities and dissecting the natural from the human-induced variance can be extremely difficult. A possible strategy to reduce the overall uncertainty is to include significant site-specific features in the models used, e.g. to derive reference conditions (Carstensen & Henriksen, 2009). The inclusion of site-specific features will also provide more realistic boundary values for managing specific water bodies as well as provide more certainty in the boundary estimates (Carstensen & Henriksen, 2009).

A variety of errors and biases can be introduced into the final classification result when considering also the various field and laboratory methods (sampling design and collection, sorting, taxonomy) and the metric used. Often the calculation of metrics involve several intermediate steps, all with associated errors and all potentially adding to the resulting overall uncertainty. It should be emphasized that the size of the various sources of error change depending on the water bodies considered. For example, lowland, deep and turbid rivers require assessment techniques and have an associated variability that is different from those used in wadeable piedmont rivers or streams (see, for example Solimini et al., 2000). Similarly, assessment systems of small lakes might differ from those used for larger systems (Solimini et al., 2008).

A degree of uncertainty is associated with both components of EQR: the expected and the observed value. The reference value of a given metric comes with an associated error because of the difficulties of defining the reference conditions for many water bodies. Therefore, an assessment in the uncertainty associated with the reference condition values should also be included in the overall risk of misclassification of a site. It is clear that metrics should be selected after examining their variability in natural, undisturbed conditions. For example, Erba et al. (2009) supposed that the highest metric variability due to natural factors can be observed at reference sites. Metrics too variable, even at reference sites, are unlikely to be effective for the assessment programs. Although, relevant typology factors of water bodies can be included to diminish variability in reference values for ecological measures, variability may remain high. For example, a recent assessment of TP values in a large dataset of reference European lakes led to coefficient of variation ranging from 1% to over 69% depending on the lake type and sample size (Cardoso et al., 2007). The same authors studied a regression model to predict reference values for TP in European lakes from simple edaphic factors, which gave a variance explained of 51%, similar to other models present in literature (Cardoso et al., 2007). In another recent article, Carvalho et al. (2009) describe the development of a series of regression models for predicting reference chlorophyll concentrations on a site-specific basis. None of the regression models developed has a particularly strong predictive ability, and the authors speculate that uncertainty arises from errors in the data used to develop the models (including natural temporal and spatial variability in data) and also from additional explanatory variables not considered in the models, particularly nutrient concentrations, retention time and grazing (Carvalho et al., 2009). Also Wolfram et al. (2009) found high variability (with r 2 ranging between 0.33 and 0.40) when regressing TP and total biovolume (e.g. algal biomass) even on a larger TP range. Nevertheless, site-specific reference conditions should result in reduced error in ecological status classifications, particularly for lakes close to typology boundaries (Carvalho et al., 2009). Water managers may face the problem of deciding whether the uncertainty associated with the predicted average value is small enough to detect changes with the actual lake conditions.

For decision makers, the importance of the risk of misclassification of a given site depends on the context of the decision to be taken and increases with larger spatial scales and longer time horizons. In the case of ecological data, the complexity of error budget and the magnitude of resulting uncertainty estimates may be very large and so the necessary effort (e.g. the cost) for its estimation. For this reason, many forms of uncertainty are not acknowledged in the models that support decisions or, in some cases, are ignored altogether (Burgman et al., 2005; but see Kelly et al., 2009). Recently, ecologists have become more and more sophisticated in realizing suitable metrics that are stable and robust enough to assess ecological status in a reliable manner (see Ptacnik et al., 2009; Kelly et al., 2009) but it is also necessary that the quantification of EQR related uncertainty is implemented in future assessment programs. Notably, Kelly et al. (2009) found that variability (expressed as the standard deviation of samples through time) could be modelled using a polynomial function, from which the risk of misclassification could be calculated. Moreover, in general, if the EQR assessment outcome can be incorporated into a modelling framework resulting in an overall estimation of the risk of misclassification, then the related uncertainty might be assessed through careful evaluation of model predictions. A possible approach is to account for as many forms of uncertainty as possible (such as error propagation and sensitivity analyses see, for example Blukacz et al., 2005; Lo, 2005) and to use several modelling frames and examine whether the choice makes a difference to management decisions (Burgman et al., 2005). Advanced computational techniques, such as fuzzy logic or neural networks, may be effective in dealing with the sources of uncertainty that affect the process of metric development (Borja et al., 2009).



The study was supported by EU FP7 grant 226273 (WISER). Special thanks go to Dr. Judit Padisák for her valuable comments and suggestions.


  1. Blukacz, E. A., W. G. Sprules & J. Brunner, 2005. Use of bootstrap for error propagation in estimating zooplankton production. Ecology 86: 2223–2231.CrossRefGoogle Scholar
  2. Borics, G., I. Grigorszky, S. Szabó & J. Padisák, 2000. Phytoplankton associations under changing pattern of bottom-up vs. top-down control in a small hypertrophic fishpond in East Hungary. Hydrobiologia 424: 79–90.CrossRefGoogle Scholar
  3. Borja, A., A. Miles, A. Occhipinti-Ambrogi & T. Berg, 2009. Current status of macroinvertebrate methods used for assessing the quality of European marine waters: implementing the Water Framework Directive. Hydrobiologia. doi: 10.1007/s10750-009-9881-y.
  4. Braun-Blanquet, J., 1925. Zur Wertung der Gesellschaftstreue in der Pflanzensoziologie. Vierteljahrsschrift der Naturforschenden Gesellschaft in Zürich 70: 122–149.Google Scholar
  5. Burgman, M. A., D. B. Lindenmeyer & J. Elith, 2005. Managing landscapes for conservation under uncertainty. Ecology 86: 2007–2017.CrossRefGoogle Scholar
  6. Calbet, A. & M. R. Landry, 2004. Phytoplankton growth, microzooplankton grazing, and carbon cycling in marine systems. Limnology and Oceanography 49: 51–57.Google Scholar
  7. Cardoso, A. C. & G. Free, 2008. Incorporating invasive alien species into ecological assessment in the context of the Water Framework Directive. Aquatic Invasions 3: 361–366.CrossRefGoogle Scholar
  8. Cardoso, A. C., A. Solimini, G. Premazzi, L. Carvalho, A. Lyche & S. Relolainen, 2007. Phosphorus reference concentrations in European lakes. Hydrobiologia 584: 3–12.CrossRefGoogle Scholar
  9. Carletti, A., Jowett, D. & A.-S. Heiskanen (eds), 2009. Water Framework Directive Intercalibration Technical Report. Part 3: Coastal and Transitional Waters. JRC Scientific and Technical Reports, EUR 23838 EN/3: 240 pp.Google Scholar
  10. Carlson, R. E., 1977. A trophic state index for lakes. Limnology & Oceanography 22: 361–369.CrossRefGoogle Scholar
  11. Carstensen, J., 2007. Statistical principles for ecological status classification of Water Framework Directive monitoring data. Marine Pollution Bulletin 55: 3–15.PubMedCrossRefGoogle Scholar
  12. Carstensen, L. & P. Henriksen, 2009. Phytoplankton biomass response to nitrogen inputs: a method for WFD boundary setting applied to Danish coastal waters. Hydrobiologia. doi: 10.1007/s10750-009-9867-9.
  13. Carvalho, L., A. Solimini, G. Phillips, O-P. Pietiläinen, J. Moe, A. C. Cardoso, A. L. Solheim, I. Ott, M. Søndergaard, G. Tartari & S. Rekolainen, 2009. Site-specific chlorophyll reference conditions for lakes in Northern and Western Europe. Hydrobiologia. doi: 10.1007/s10750-009-9876-8.
  14. CIS, 2003a. River and lakes – typology, reference conditions and classification systems. Common Implementation Strategy for the Water Framework Directive (2000/60/EC), Guidance document 10, European Commission: 86 pp.Google Scholar
  15. CIS, 2003b. Transitional and Coastal Waters – Typology, Reference Conditions and Classification Systems. Common Implementation Strategy for the Water Framework Directive (2000/60/EC), Guidance document 5, Working Group 2.4 – COAST, European Commission: 119 pp.Google Scholar
  16. CIS, 2005. Towards a guidance document on eutrophication assessment in the context of European water policies. Common Implementation Strategy for the Water Framework Directive (2000/60/EC). Interim document. European Commission: 133 pp.Google Scholar
  17. CIS, 2009. Guidance document on exemptions to the environmental objectives. Common Implementation Strategy for the Water Framework Directive (2000/60/EC), Guidance document 20. European Commission: 46 pp.Google Scholar
  18. Daskalov, G. M., 2002. Overfishing drives a trophic cascade in the Black Sea. Marine Ecology Progress Series 225: 53–63.CrossRefGoogle Scholar
  19. Devlin, M., J. Barry, S. Painting & M. Best, 2009. Extending the phytoplankton tool kit for the UK Water Framework Directive: indicators of phytoplankton community structure. Hydrobiologia. doi: 10.1007/s10750-009-9879-5.
  20. Directive, 2000. Directive 2000/60/EC of the European Parliament and of the council of 23 October 2000 establishing a framework for community action in the field of water policy. Official Journal of the European Communities L 327: 1–72.Google Scholar
  21. Directive, 2008. Directive 2008/56/EC of the European Parliament and of the council of 17 June 2008 establishing a framework for community action in the field of marine environmental policy (Marine Strategy Framework Directive). Official Journal of the European Communities L 164: 19–40.Google Scholar
  22. Dokulil, M. T. & K. Teubner, 2005. Do phytoplankton communities correctly track trophic changes? An assessment using directly measured and palaeolimnological data. Freshwater Biology 50: 1589–1593.CrossRefGoogle Scholar
  23. Dokulil, M. T., K. Teubner & J. Greisberger, 2005. Typenspezifische Referenzbedingungen für die integrierende Bewertung des ökologischen Zustandes stehender Gewässer Österreichs Gemäß der EU-Wasserrahmenrichtlinie. Modul 1: Die Bewertung der Phytoplankton Struktur nach dem Brettum-Index. Projektstudie Phase 3, Abschlussbericht. Im Auftrag des Bundesministeriums für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Wien.Google Scholar
  24. Donohue, I., L. A. Donohue, B. Ní Ainín & K. Irvine, 2009. Assessment of eutrophication pressure on lakes using littoral invertebrates. Hydrobiologia. doi: 10.1007/s10750-009-9868-8
  25. Duarte, C. M., D. J. Conley, J. Carstensen & M. Sánchez-Camacho, 2009. Return to Neverland: shifting baselines affect eutrophication restoration targets. Estuaries and Coasts 32: 29–36.CrossRefGoogle Scholar
  26. Dufour, S. & S. Piégay, 2009. From the myth of a lost paradise to targeted river restoration: forget natural references and focus on human benefits. River Research and Applications 25: 568–581.CrossRefGoogle Scholar
  27. Erba, S., M. T. Furse, R. Balestrini, A. Christodoulides, T. Ofenböck, W. van de Bund, J.-G. Wasson & A. Buffagni, 2009. The validation of common European class boundaries for river benthic macroinvertebrates to facilitate the intercalibration process of the Water Framework Directive. Hydrobiologia. doi: 10.1007/s10750-009-9873-y
  28. European Commission, 2008. Commission decision of 30 October 2008 establishing, pursuant to Directive 2000/60/EC of the European Parliament and of the Council, the values of the Member State monitoring system classifications as a result of the intercalibration exercise Official Journal of the European Union L 332: 20–44.Google Scholar
  29. Hajnal, É. & J. Padisák, 2008. Analysis of long-term ecological status of Lake Balaton based on the ALMOBAL phytoplankton database. Hydrobiologia 599: 227–237.CrossRefGoogle Scholar
  30. Hering, D., C. K. Feld, O. Moog & T. Ofenböck, 2006. Cook book for the development of a multimetric index for biological condition of aquatic ecosystems: experiences from the European AQEM and STAR projects and related initiatives. Hydrobiologia 566: 311–324.CrossRefGoogle Scholar
  31. Irvine, K., 2004. Classifying ecological status under the European Water Framework Directive: the need for monitoring to account for natural variability. Aquatic Conservation: Marine and Freshwater Ecosystems 14: 107–112.CrossRefGoogle Scholar
  32. Jeppesen, E., J. P. Jensen, M. Søndergaard, M. Fenger-Grøn, S. Sandby, P. Hald Møller & H. U. Rasmussen, 2004. Does fish predation influence zooplankton community structure and grazing during winter in north-temperate lakes? Freshwater Biology 49: 432–447.CrossRefGoogle Scholar
  33. Johnson, R. K., D. Hering, M. T. Furse & R. T. Clarke, 2006. Detection of ecological change using multiple organism groups: metrics and uncertainty. Hydrobiologia 566: 115–137.CrossRefGoogle Scholar
  34. Kaiblinger, C., O. Anneville, R. Tadonleke, F. Rimet, J. C. Druart, J. Guillard & M. T. Dokulil, 2009. Central European water quality indices applied to long-term data from peri-alpine lakes: test and possible improvements. Hydrobiologia. doi: 10.1007/s10750-009-9877-7.
  35. Karr, J. R., 1981. Assessment of biotic integrity using fish communities. Fisheries 6: 21–27.CrossRefGoogle Scholar
  36. Karr, J. R. & E. W. Chu, 1997. Biological monitoring and assessment: using multimetric indexes effectively. EPA 235–R07-001. University of Washington, Seattle, WA: 149 pp.Google Scholar
  37. Kelly, M., H. Bennion, A. Burgess, J. Ellis, S. Juggins, R. Guthrie, J. Jamieson, V. Adriaenssens & M. Yallop, 2009. Uncertainty in ecological status assessments of lakes and rivers using diatoms. Hydrobiologia. doi: 10.1007/s10750-009-9872-z
  38. Kolkwitz, R. & M. Marsson, 1908. Ökologie der pflanzlichen Saprobien. Berichte der Deutschen Botanischen Gesellschaft 26a: 505–519.Google Scholar
  39. Kolkwitz, R. & M. Marsson, 1909. Ökologie der tierischen Saprobien. Beiträge zur Lehre von der biologischen Gewässerbeurteilung. Internationale Revue der gesamten Hydrobiologie und Hydrographie 2: 126–152.CrossRefGoogle Scholar
  40. LAWA, 1999. Gewässerbewertung – Stehende Gewässer. Vorläufige Richtlinie für eine Erstbewertung von natürlich entstandenen Seen nach trophischen Kriterien. Kulturbuch Verlag, Berlin: 1–74.Google Scholar
  41. Lo, E., 2005. Gaussian error propagation applied to ecological data: post-ice-storm-downed woody biomass. Ecological Monographs 75: 451–466.CrossRefGoogle Scholar
  42. Lopez y Royo, C., C. Silvestri, M. Salivas-Decaux, G. Pergent & G. Casazza, 2009. Application of an angiosperm based classification system (BiPo) to Mediterranean coastal waters: using spatial analysis and data on metal contamination of plants in identifying sources of pressure. Hydrobiologia. doi: 10.1007/s10750-009-9880-z
  43. Lyche Solcheim, A. (ed.), 2005. Reference conditions of European lakes. Indicators and methods for the Water Framework Directive assessment of reference conditions, REBECCA Project, deliverable 7: 105 pp.Google Scholar
  44. Mischke, U., U. Riedmüller, E. Hoehn, I. Schönfelder & B. Nixdorf, 2008. Description of the German system for phytoplankton-based assessment of lakes for implementation of the EU Water Framework Directive (WFD). In Mischke, U. & B. Nixdorf (eds), Brandenburg Technical University of Cottbus, ISBN 978-3-940471-06-2, BTUC-AR 2/2008: 117–146.Google Scholar
  45. Moss, B., 2007. Shallow lakes, the water framework directive and life. What should it all be about? Hydrobiologia 584: 381–394.CrossRefGoogle Scholar
  46. Moss, B., 2008. The Water Framework Directive: total environment or political compromise? Science of the Total Environment 400: 32–41.PubMedCrossRefGoogle Scholar
  47. Moss, B., D. Stephen, C. Alvarez, E. Becares, W. Van de Bund, S. E. Collings, E. Van Donk, E. De Eyto, T. Feldmann, C. Fernandez-Alaez, M. Fernandez Alaez, R. J. M. Franken, F. Garcia-Criado, E. Gross, M. Gyllstrom, L.-A. Hansson, K. Irvine, A. Järvalt, J. P. Jensen, E. Jeppesen, T. Kairesalo, R. Kornijow, T. Krause, H. Künnap, A. Laas, E. Lill, B. Lorens, H. Luup, M. R. Miracle, P. Nõges, T. Nõges, M. Nykanen, I. Ott, W. Peczula, E. T. H. M. Peeters, G. Phillips, S. Romo, V. Russell, J. Salujõe, M. Scheffer, K. Siewertsen, H. Smal, T. Virro, E. Vicente & D. Wilson, 2003. The determination of ecological status in shallow lakes – a tested system (ECOFRAME) for implementation of the European Water Framework Directive. Aquatic Conservation: Marine & Freshwater Ecosystems 13: 507–549.CrossRefGoogle Scholar
  48. Nõges, P., W. van de Bund, A. C. Cardoso & A. S. Heiskanen, 2007. Impact of climatic variability on parameters used in typology and ecological quality assessment of surface waters – implications on the Water Framework Directive. Hydrobiologia 584: 373–379.CrossRefGoogle Scholar
  49. Nõges, P., U. Mischke, R. Laugaste & A. G. Solimini, submitted. Analysis of 44-year changes in phytoplankton of Lake Võrtsjärv (Estonia): the effect of nutrients, climate, and the investigator on phytoplankton based water quality indices. Hydrobiologia.Google Scholar
  50. Ocampo-Duque, W., N. Ferré-Huguet, J. L. Domingo & M. Schuhmacher, 2006. Assessing water quality in rivers with fuzzy inference systems: a case study. Environment International 32: 733–742.PubMedCrossRefGoogle Scholar
  51. Ocampo-Duque, W., M. Schuhmacher & J. L. Domingo, 2007. A neural-fuzzy approach to classify the ecological status in surface waters. Environmental Pollution 148: 634–641.PubMedCrossRefGoogle Scholar
  52. Olenin, S., D. Minchin & D. Daunys, 2007. Assessment of biopollution in aquatic ecosystems. Marine Pollution Bulletin 55: 379–394.PubMedCrossRefGoogle Scholar
  53. Padisák, J. & C. S. Reynolds, 1998. Selection of phytoplankton associations in Lake Balaton, Hungary, in response to eutrophication and restoration measures, with special reference to the cyanoprokaryotes. Hydrobiologia 384: 41–53.CrossRefGoogle Scholar
  54. Padisák, J., I. Grigorszky, G. Borics & É. Soróczki-Pintér, 2006. Use of phytoplankton assemblages for monitoring ecological status of lakes within the Water Framework Directive: the assemblage index. Hydrobiologia 553: 1–14.CrossRefGoogle Scholar
  55. Pall, K. & V. Moser, 2009. Austrian Index Macrophytes (AIM-Module 1) for lakes: a Water Framework Directive (WFD) compliant assessment system for lakes using aquatic macrophytes. Hydrobiologia. doi: 10.1007/s10750-009-9871-0
  56. Poikane, S. (ed.), 2009. Water Framework Directive intercalibration technical report. Part 2: Lakes. JRC Scientific and Technical Reports, EUR 23838 EN/2: 176 pp.Google Scholar
  57. Ptacnik, R., A. G. Solimini & P. Brettum, 2009. Performance of a new phytoplankton composition metric along a eutrophication gradient in Nordic lakes. Hydrobiologia. doi: 10.1007/s10750-009-9870-1
  58. Rawson, D. S., 1956. Algal indicators of trophic lakes types. Limnology & Oceanography 1: 18–25.Google Scholar
  59. Reynolds, C. S., 1998. What factors influence the species composition of phytoplankton in lakes of different trophic status? Hydrobiologia 369(370): 11–26.CrossRefGoogle Scholar
  60. Schneider, S. & A. Melzer, 2003. The Trophic Index of Macrophytes (TIM) – a new tool for indicating the trophic state of running waters. International Review of Hydrobiology 88: 49–67.CrossRefGoogle Scholar
  61. Schriver, P., J. Bøgestrand, E. Jeppesen & M. Søndergaard, 1995. Impact of submerged macrophytes on the interactions between fish, zooplankton and phytoplankton: large-scale enclosure experiments in a shallow lake. Freshwater Biology 33: 255–270.CrossRefGoogle Scholar
  62. Solimini, A. G., P. Gulia, M. Monfrinotti & G. Carchini, 2000. Performance of different biotic indices and sampling methods in assessing water quality in the lowland stretch of the Tiber River. Hydrobiologia 422(423): 197–208.CrossRefGoogle Scholar
  63. Solimini, A., A. C. Cardoso, A.-S. Heiskanen (eds), 2006. Indicators and methods for the Ecological Status Assessment under the Water Framework Directive. Linkages between chemical and biological quality of surface waters. EUR 22314 EN. European Commission: 248 pp.Google Scholar
  64. Solimini, A. G., M. Bazzanti, A. Ruggiero & G. Carchini, 2008. Developing a multimetric index of ecological integrity based on macroinvertebrates of mountain ponds in central Italy. Hydrobiologia 597: 109–123.CrossRefGoogle Scholar
  65. Solimini, A. G., R. Ptacnik & A. C. Cardoso, 2009. Toward a holistic assessment of ecosystem functioning: the relationships between anthropogenic pressures, chemical and ecological status under the Water Framework Directive. Trends in Analytical Chemistry 28: 143–147.CrossRefGoogle Scholar
  66. Søndergaard, M. & B. Moss, 1998. Impact of submerged macrophytes on phytoplankton in shallow freshwater lakes. In Jeppesen, E., Ma. Søndergaard, Mo. Søndergaard & K. Christoffersen (eds), The Structuring Role of Submerged Macrophytes in Lakes. Ecological Studies Series 131. Springer, New York: 115–132.Google Scholar
  67. Stansfield, J. H., M. R. Perrow, L. D. Tench, A. J. D. Jowitt & A. A. L. Taylor, 1997. Submerged macrophytes as refuges for grazing Cladocera against fish predation: observations on seasonal changes in relation to macrophyte cover and predation pressure. Hydrobiologia 342(343): 229–240.CrossRefGoogle Scholar
  68. UNESCO, 2000. Solving the Puzzle: The Ecosystem Approach and Biosphere Reserves. UNESCO, Paris.Google Scholar
  69. van de Bund, W. (ed.), 2009. Water Framework Directive intercalibration tecnical report. Part 1: Rivers. JRC Scientific and Technical Reports, EUR 23838 EN/1: 136 pp.Google Scholar
  70. Vighi, M., A. Finizio & S. Villa, 2006. The evolution of the environmental quality concept: from the US EPA Red Book to the European Water Framework Directive. Environmental Science and Pollution Research 13: 9–14.PubMedCrossRefGoogle Scholar
  71. Watt, A. D., R. H. W. Bradshaw, J. Young, D. Alard, T. Bolger, D. Chamberlain, F. Fernández-González, R. Fuller, P. Gurrea, K. Henle, R. Johnson, Z. Korsós, P. Lavelle, J. Niemelä, P. Nowicki, M. Rebane, C. Scheidegger, J. P. Sousa, C. van Swaay & A. Vanbergen, 2007. Trends in biodiversity in Europe and the impact of land-use change. In Harrison, R. M. & R. E. Hester (eds), Biodiversity under Threat. Issues in Environmental Science and Technology 25. Royal Society of Chemistry, Cambridge: 135–160.Google Scholar
  72. Willén, E., 2007. Växtplankton i sjöar. Bedömningsgrunder. SLU, Institutionen för Miljöanalys, Rapport 2007: 5, 33 pp.Google Scholar
  73. Wolfram, G., C. Argillier, J. de Bortoli, F. Buzzi, A. Dalmiglio, M. T. Dokulil, E. Hoehn, A. Marchetto, P.-J. Martinez, G. Morabito, M. Reichmann, Š. Remec-Rekar, U. Riedmüller, C. Rioury, J. Schaumburg, L. Schulz & G. Urbanič, 2009. Reference conditions and WFD compliant class boundaries for phytoplankton biomass and chlorophyll-a in Alpine lakes. Hydrobiologia. doi: 10.1007/s10750-009-9875-9

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  • Peeter Nõges
    • 1
    • 2
    Email author
  • Wouter van de Bund
    • 1
  • Ana Cristina Cardoso
    • 1
  • Angelo G. Solimini
    • 1
    • 3
  • Anna-Stiina Heiskanen
    • 1
    • 4
  1. 1.Institute for Environment and Sustainability, Joint Research CentreIspraItaly
  2. 2.Centre for Limnology, Institute of Agricultural and Environmental SciencesEstonian University of Life SciencesRannuEstonia
  3. 3.Department of Experimental MedicineSapienza University of RomeRomeItaly
  4. 4.Marine Research CentreFinnish Environment Institute (SYKE), Research Programme for the Protection of the Baltic SeaHelsinkiFinland

Personalised recommendations