S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ loca te / sc i to tenv

The Water Framework Directive: Total environment orpolitical compromise?

Brian Moss⁎

School of Biological Sciences, University of Liverpool, Liverpool L69 3BX, UK

A R T I C L E I N F O

⁎ Tel.: +44 1517954390.E-mail address: brmoss@liverpool.ac.uk.

0048-9697/$ – see front matter © 2008 Elsevidoi:10.1016/j.scitotenv.2008.04.029

A B S T R A C T

Article history:Received 18 March 2008Received in revised form 4 April 2008Accepted 10 April 2008Available online 27 May 2008

The European Water Framework Directive (2000/60/EC) is potentially ground-breakinglegislation. It seeks to bring about improvement of aquatic habitats in Europe to ‘goodecological status’, defined as slightly different from ‘high ecological status’, with no orminimal human impact. The characteristics of pristine ecological status include nutrientparsimony, a defined characteristic structure of the system (including geomorphologicalstructure and hydrology, biological and food web structure) and the connectivity and extentof the system that are essential for resilience to change. This modern ecologicalunderstanding is being ignored by government agencies charged with enacting theDirective. Schemes are being devised that measure secondary characteristics of habitatsusing approaches drawn from traditional water quality management. Typologies, indicatedby the Directive to give a geographical basis within which to determine ecological status, arealso being corrupted with different typologies used for different determinands. Theecological reality of reasonably distinctive, integrated systems (an erosive upland riverversus a floodplain system, for example) is being avoided. Emphasis is being placed onprecision of measurement of specific determinands rather than accuracy in what is beingmeasured and proposed schemes are complex and expensive when accurate assessmentcould be carried out much more cheaply. Many are also likely to become redundant aseffects of climate change take hold. The current approach will lead to some improvement inwater quality but not to the fundamental change in ecological quality intended by theDirective and has partly been encouraged by lack of definition and contradictions within theDirective itself. Documented details currently available from the UK agencies are used toillustrate how the intentions of the Directive are being undermined for ostensibly politicalconvenience through processes of redefinition and limitation of characteristics measured.There appears to be a parallel concern among official and non-governmental Europeanbodies.

© 2008 Elsevier B.V. All rights reserved.

Keywords:Water Framework DirectiveEcological statusEcosystemEuropean legislationEarth systems science

1. Introduction

No natural scientist would maintain that the mechanisms bywhich the biosphere functions are at the whim of humanopinion. But currently it seems fashionable to enfold all areas ofknowledge into the social sciences, those concerned with theanalysis of human compromise. Problems of the planet, for

er B.V. All rights reserved

example,must be solved fromthe triple viewpoints of the social,economic and environmental (Daly, 1999). The ‘EcosystemApproach’, widely adopted for the Convention on BiologicalDiversity (UNESCO, 2000) purports that ‘The objectives of manage-ment of land, water and living resources are amatter of societal choice’.It seems that the advance of the 18th century Enlightenmentthat there are absolute phenomena independent of human

.

33S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

view, has been sidelined. What seems to take precedence overnatural phenomena, in the issues of environmental manage-ment, is human policy for human convenience.

Some time ago, I referred (Moss, 2001) to the recently passedEuropeanWater Framework Directive (WFD, European Commu-nity, 2000) as ‘red hot and revolutionary’ because it recognised thataquatic ecosystems had ultimate properties, reflected in undis-turbed reference states that could be used as standards forrestorationof theecological statusof Europeanwaters. It centredon thenatural sciences rather than thesocial thoughwithstrongsocial elements (Steyaert andOllivier, 2007). Itmayyet still proverevolutionarybut, eightyearson, indicationsare that itwill bringabout a small evolution in attitudes and practices for themanagement of aquatic habitats but will fall well short ofrevolution. My thesis is that the Directive is becoming mired inpolitical compromises, through the conservatism of watermanagement bodies that have been unable to change theirapproach from practices that the Directive was intended todisplace. Such conservatism has been helped, however, bycontradictions and lack of definition within the Directive itself,and is to the advantage of those bodies who wish to minimisethe inconvenience the Directive would otherwise bring to theirexploitative activities. The WFD will improve the environmentto a degree but, more importantly will not form the pilot legaltemplate for solving the greater problems of ecosystemdestruc-tion and climate change that it might have done.

What follows is an analysis of the current situation, seenparticularly from the viewpoint of activities in the UK. This isnot so parochial as it might seem, for the UK chairs or co-chairs a number of influential working groups (on ecologicalstatus, hydromorphology, and agriculture (Surridge andHarris, 2007)) concerned with common implementation ofthe Directive across member states and its general approachappears to be becoming representative. The Directive stronglyemphasises ecology as opposed to chemical water quality, theprevious basis for water quality management in Europe andelsewhere. The key starting point for the Directive is the ‘highecological status’ (pristine state) against which change mustbe measured and eventually rectified. Fundamental thereforeto any discussion of the Directive is first to establishwhat highecological status means in modern ecological terms.

2. A worldview of ecological status

There are logical arguments for determining the nature of highecological status, defined by the Directive as a state with no ornegligible human impact. These arguments may be sum-marised as:

1. The biosphere is a non-equilibrium system maintained byliving organisms in a state equable to their evolvedbiochemistry;

2. The evolved natural ecosystems are the units for thismaintenance and their extent and integrity are essentialfor sustained maintenance;

3. These systems are highly efficient because their compo-nents are continually tested and modified through naturalselection; they therefore represent optimal systems thatcan be damaged but not improved;

4. Measures of this efficiency are summed up in fourfundamental characteristics that characterise ecologicalquality. These are (a) efficiency in recycling scarce materi-als (nutrient parsimony), (b) characteristic physical andfood web structure that ensures this parsimony andmaintenance of the intactness of the system as a whole,(c) connectivity with other systems that also maintainsintactness, and (d) a large enough size to allow resilience tochange;

5. Impairment of these features represents damage anddecline in ecological quality;

6. Secondary features such as particular concentrations ofsubstances or lists of species do not adequately measurethese fundamental characteristics;

7. Ecological quality is accurately measured by attentionprimarily to the intactness of the fundamental character-istics rather than thedetails of the secondary characteristics.

These arguments begin with (step 1) the discovery that thebiosphere ismaintained in a chemically non-equilibrium state(Lovelock, 1988, 2003; Lovelock and Margulis, 1974; Margulisand Lovelock, 1974) by systems of living organisms that arevery diverse, but whose members share a common biochem-istry. The latter requires about twenty elements and a sourceof energy, solar radiation, to maintain its systems against thethermodynamic tendency for entropy to increase and forcomplex systems to disperse into amorphousness. Theprecise, non-equilibrium compositions of the atmosphereand oceans are thus not random but have meaning. Theyfall within the equable range that permits the persistence ofliving organisms characterised by such biochemistry.

It follows (step 2) that interconnected natural sub-systems,or ecosystems, of the biosphere are needed for themaintenanceof the equable range and that there is a limit to which theirintactness can be compromised without conditions movingoutside that range. Pressure on this intactness is currently beingdemonstrated by the consequences of anthropogenic climatechange (Intergovernmental Panel on Climate Change, 2007) anderosion of ecosystem services (MillenniumEcosystemCommis-sion, 2004). We have little idea of the extent to which thesystems can be damaged, or replaced by agricultural systems,without serious consequences. Current problems suggest thatwe have already pushed them too far.

Step 3 is recognition that these ecosystems have developedthrough a ruthless and continual process of natural selection oftheir component species, in which less fit individuals areeliminated through differential reproduction (Darwin, 1859;Dawkins, 1986). Natural ecosystems, whose components arecontinually tested to the point of destruction, thereforerepresent optimal systems for maintenance of equable condi-tions. The selection mechanism, plus an interconnectednessthat allows movement of organisms across environmentalgradients, allows adjustment to degrees and cycles of naturalchange provided that the ecosystems remain big enough.

The fourth step in the argument is recognition that theecosystems have basic common characteristics, developedthrough natural selection, that are essential to maintain theirintactnessand independent functioning (i.e.without anyneed forexternal management) in the mechanisms that maintain globalconditions within the equable range. These characteristics are

34 S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

fourfold. They include (a) an efficient recycling (nutrient parsi-mony) of elements (Furch,1984; Likens et al., 1971; Miyasaka andHabti, 2001; Raven et al., 2005) that are essential but inherentlyscarce in available forms for reasons of elemental properties andplanetary accident.

Next (b) they include a characteristic structure in a givenlocation that reflects adaptation of organisms to environ-mental drivers such as local climate, geology and hydrology,the activities of the organisms, and the structure imparted bydrivers themselves. There is a physicochemical componentderived from the geomorphology and hydrology and theactivities of organisms (trees, coral reefs, vegetation types,soil, woody debris) and a biological food web structurereflecting the need for regulation of populations within thatstructure, manifested by food webs culminating in toppredators (Calman et al. 2002; Fitkau, 1970; Leopold, 1949;Rao et al., 2001; Ripple and Beschta, 2004a,b; Terborgh, 1988;Terborgh et al., 2006) and required if the overall functioning isnot to be compromised.

The third characteristic (c) allows adjustment to continuingnatural change (resilience) and is the connectivity amongecosystems that allows global cycling of water and scarceelements, such as nitrogen (in available forms), through theatmosphere and rivers, the migrations of many animals andpassive movement and dispersal of other organisms (micro-organisms, seeds) as well as many local interactions (Baxteret al., 2005). Linkedwith connectivity and resilience is (d) scale,the last attribute. All independent ecosystems are on a largescale, adequate to compensate for disturbances within themthrough provision of refuges against natural disturbance andto maintain large enough gene pools of each organism toguarantee the raw material for natural selection.

None of these four fundamental characteristics demandsfixed concentrations of substances nor fixed lists of species.Always there is a very wide range with continuous, some-times cyclic but, frequently directional change to maintainthe most fundamental feature of resilience to change.Description of the nature of an ecosystem in terms of a listof components (a snapshot inventory) at a given time is notonly an ephemeral exercise but seriously misunderstandsthe fundamental importance of variability in maintainingecosystem resilience. It follows (steps 5–7) that establish-ment of criteria for ecological quality for the WFD shouldinclude primary measures of nutrient parsimony, character-istic structure in its several aspects, connectedness and sizebut not (step 6) of the shifting secondary details of speciescomposition and discontinuous chemistry, unless these areused in ways to diagnose the state of the more fundamentalcharacteristics.

At the heart of current attempts to establish standards forenactment of theWFD is reliance on secondary details ratherthan measures of fundamental value. There is an importantcontradiction within the Directive between its definition ofhigh ecological status and the instructions it gives (in itsAnnex 5) on the way that this is to be established. Thedefinition embraces the fundamental; the instructionsencourage concentration on secondary details and hencemay have undermined the possibility of fundamentalimprovement of aquatic habitats that was intended. Tounderstand this, some details of the Directive are needed.

3. The Water Framework Directive

The WFD has several main aims that represent changes ineither the national or pan-European legislation that hadpreceded it. It is concerned with protecting all waters (fresh,estuarine, coastal, ground) and its basis is management of thecatchments (river basins). The Directive combines the need foremission controls (formerly the usual approach for point sourcepollutants) with requirements to guarantee ultimate quality ofthe receiving habitats and it phases out licence to acceptconcentrations of particularly hazardous substances at thecurrent limits of detectability. Economic analysis and economicinstruments topromotewiseuseofwater are required, asalso ispublic participation in decisions. But the heart of the Directivelies in determination (classification) of ecological status ofaquatichabitats ona scale fromhigh (effectively pristine) to bad,including intermediate steps of good, moderate and poor withdue respect to the individual natures (typologies) of differentsorts of rivers, lakes, estuaries and coastal waters. The WFDthen requires restoration of all habitats (subject to certainderogations for heavily modified and man-made habitats) togoodecological statusby2015.Goodstatus isdefined in theWFDas ‘slightly’ different from high status.

An unusual feature of the Directive is that there are copiousannexes to themain text, with rathermore detail in them thanis usual for European Directives. The legislative approach ofthe European Community is to express the spirit and generaldirection and to leave the details to enabling legislationpassed nationally by member states. A strong intention oftheWFD has been to achieve comparable standards across theEuropean Community and Annex 5 is explicit in howtypologies and classification systems should be designed.Typologies are the means of recognising different sorts ofwaters (for example, high mountain Arctic lakes in igneouscatchments versus shallow lakes in fertile glacial plains versusMediterranean lakes that dry out seasonally).Within each unitof the typology, the characteristics of the pristine state havethen to be established by finding current examples, palaeoe-cological reconstruction or use of expert opinion and then ascale of decreasing quality established with grades of good,moderate, poor and bad must be devised to classify ecologicalstatus.

Typology should pose the fewest problems. The Directiveoffers a prescribed system (A) and a more flexible system (B)that can be designed bymember states so long as it achieves acomparable resolution of sites. Features to be used includelatitude, area, depth, and local geology and are largely fixturesof geography that are unlikely to affected by human impact.This is essential to avoid confounding them with character-istics that are affected by human activities, which must beused for determining ecological status. There is a dangeroushint that nutrients might be used that suggests that thiscrucial distinction might not have been fully appreciated bythe drafters of the legislation. The rapid changes in climatenow occurring to some extent themselves confound thetypology for although latitude is fixed and was intended tobe a surrogate for climate, the link is changing. Nonetheless itis clear that Annex 5 envisages definition of different sorts ofsites that have fixed common characteristics.

35S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

Not since the 1960s have ecologists thought in terms of‘types’ of habitat (Elster, 1958), though it was a majorpreoccupation of early twentieth century limnologists. Theperception has moved to recognition of continuous variationalong many axes such that every site is effectively unique(Moss et al., 1994; Sandin and Johnson, 2000), but for practicalpurposes, typologies based on broad enough geographical andtopographical criteria are workable. A general disappointmentis that although aquatic ecology is independent of nationalboundaries, eachmember state appears to be devising its owntypology (e.g. Wasson et al., 2004; UKTAG, 2007a). TheDirective suggests that for a System A typology, Illies' (1978)biogeographic regions, a system for dividing geographicalareas based on the biogeography of lake macroinvertebrates,might be used and this has perhaps promoted a nationalapproach, for the regions do often approximate to nationalboundaries, especially where islands are concerned. Conclu-sions of large scale research projects (Moss et al, 2003; Sandinand Verdonschot, 2006; Verdonschot, 2006) however, clearlyshow that international typologies are easily devisable andwould help cross-nation comparisons of standards.

In determining ecological status, Annex 5 separatelyrecognises biological status and chemical status and saysthat biological status is pre-eminent and chemical statusmustbe appropriate to it. There is potential confusion however inthat ‘chemical status’ refers to toxic pollutants, not to thechemical features of waters familiarly intertwined withecological processes, such as oxygen and nutrient concentra-tions, which are rightly embraced in ecological status. Annex 5specifies that particular elements of the biological communitymust be used to determine ecological status (e.g. for rivers, themacroinvertebrate, fish and macrophyte/phytobenthos com-munities) but does not say how they should be used, merelyhow often minimally they should be measured. This fre-quency sometimes bears little correspondence to how rapidlysuch communities change seasonally (Carstensen, 2007;Dworak et al., 2005).

There are two much greater flaws, however, that under-mine the potential effectiveness of the Directive. The first is inthe paucity of precise definition. The reference high status isdefined as having no or minimal human influence, withoutcomment on the meaning of ‘minimal’, though the intentionis clearly close to ‘no’. Good status is defined as being slightlydifferent from high, without definition of the meaning of‘slightly’ outside that of a conventional dictionary, where itmeans little difference. There are no usable guidelines to themeaning of moderate, poor and bad, yet it is the boundarybetween moderate and good that is crucial, for that is thestandard to which member states will aspire as the leastexpensive objective in restoring their habitats to good status.The extent to which the term ‘slightly’ can be stretched isapparently quite large.

The second flaw (Hatton-Ellis, 2008; Moss, 2007) is in aperception that ecological quality can be measured in terms ofsimple ratios, called ecological quality ratios (EQR). Given x fishin the high quality state, there should be 0.9 x (or some suchnumber) in the good, 0.5 in the moderate and so forth. As withthe concept of typologies this suggests that the drafters of theDirective were unaware of advances in understanding ofecology. Such ratios might be appropriate in determining

standards for alien toxic substances, in laboratory test systems,where the concept arose (Chapman, 1995). They are not even inthe conceptual frame of professional ecologists. They embodythe secondary characteristics referred to in theprevious section.

The WFD represented (without that being either inten-tional or widely realised) a new approach in confrontingplanetary problems. Simply in taking a river basin approach, itacknowledged that what happens in waters depends substan-tially on what happens on land in the catchments. It isimpossible to manage a waterway without sympatheticmanagement of the land, though almost all previous legisla-tion had ignored this and was directed largely within thebounds of the wetted perimeter. It was a first approach torealisation of the importance of at least three (nutrientparsimony, connectedness and scale) of the four fundamentalcharacteristics of natural systems and could have reflectedcharacteristic structure also in that themeasures prescribed inAnnex 5 could potentially be used diagnostically to establishdeviations from high status (Moss, 2007).

Alas, fundamental understanding has not been reflected inthe moves now being made to accommodate the require-ments of the WFD. As a detailed analysis of UK performancewill show, competent authorities (those agencies chargedwithaction on the Directive) have paid some attention to nutrientparsimony, but much less to characteristic structure, con-nectance or size and resilience in designing systems fordetermining ecological status. They are thinking in terms ofspecific species (indicator species) at a time when climatechange is rendering doubly redundant any concept of fixedspecies lists (Noges et al., 2007). This is not to deny that thereare relationships between occurrence of particular species,genera or groups and environmental conditions, but it is toargue that these relationships are multivariately determined(Moss, 2007) and not simply related to individual drivers in away that would validate the use of EQR. They have also failedto grasp that the four characteristics of self-maintaining, highquality states are not modules that can be adopted or ignoredat will but are themselves mutually dependent. There is nopoint in reducing nutrient concentrations, for example, if alack of characteristic structure or connectivity means that thiscan only be maintained by continual, expensive intervention.

4. Current literature trends

Since 2001, about 1000 papers have been published in ISI listedjournals that have “Water Framework Directive” in their titleor key words. It has become almost mandatory to refer to theDirective in any paper concerned with applied aspects ofaquatic ecology in Europe. There are also some thousands ofweb sites concerned with the Directive and containing officialand commentary documents on it. Many of the publishedpapers are detailed studies of components of freshwaterhabitats (e.g. fish, macroinvertebrates and aquatic plants)that defer to the Directive as requiring measurements orassessments of these components, whilst illustrating thecomplexity of the ecological issues with large sets of data andgenerally coming to a covert conclusion that simple assess-ment systemsarebeyondachievement (Baatrup-Pedersenetal.,2006; Noble and Cowx, 2007; Verdonschot, 2006).

36 S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

Many papers purport to provide information that might beused in establishing criteria for ecological quality under theDirective but prominently they concern only one componentand do not examine the ecosystem as a whole. Ecologicalquality can only validly be a property of a whole system. A fewsuch papers have been published in response to demands bycompetent authorities to develop indices that assess quality(e.g. Kelly et al., 2007) and again these are highly reductionist,using biological indices as surrogates for driving pressures likenutrients, against which they have been calibrated. Suchpressures can be independently chemically measured withgreater accuracy, though the biological community has anintegrating component that spot analyses do not. Some ofthese approaches, however, reveal deviations from therequirements of the Directive, for example establishment ofchemical criteria for high status that are not even remotelysimilar to values determined for equivalent, relatively undis-turbed systems (for example, compare Kelly et al., 2007 withSmith et al., 2003 and Wade et al., 2004).

A few papers (Carter, 2007; Kidd and Shaw, 2007) examinethe policy implications of the Directive, realising that it hasimplications for all aspects of the use of catchments andlamenting that current planning arrangements are ill-fitted tocope. Almost none (Moss, 2007) examine the fundamentalmeaning of terms within the Directive. There has been littleintrospection, little discussion about ecological status byfundamental ecologists. The impetus has been taken fromthe details of Annex V to measure individual variables, ratherthan from the main text where ecological status is the keyterm. It bears repetition that ecosystems are not merelycollections of independent variables; they are interactingsystems with emergent properties that are not estimatableby the sum of measurements of single variables.

The competent authorities seem to have given free rein totheir predominant chemists to convert biology into a surro-gate chemistry. What should have been the first step once theDirective was passed was to establish just what ecologicalstatus means and then to consider how Annex 5 details mighthave been incorporated. It still needs to be done because theliterature reveals a plethora of difficulties in designingclassification schemes for ecological status (e.g. Hering et al.,2006). Furthermore, the major changes being imposed byglobal warming are rendering redundant the details of suchschemes as have been devised (Noges et al., 2007). Macro-invertebrate communities are proposed in Annex 5 as a keybiological element, but there is already ample evidence thatadult insects whose juveniles are aquatic are steadily movingtheir ranges northwards and that the phenology of suchanimals is changing significantly (Hassall et al., 2007; Hicklinget al., 2005; Parmesan, 2006).

5. UK approach to typology and classification

Thegeneralities that are emerging inapproachcanbe illustratedwith the details emerging of typology and classification systemsin the UK. The UK has designated, as the competent authoritiesfor action on the Directive, its environment agencies. There areseparate ones for England and Wales, Scotland and NorthernIreland, but they have set up a committee of their representa-

tives and some from statutory conservation agencies calledUKTAG (UK Technical Advisory Group) to devise the typologiesand classification schemes to be used in the UK. A previous riskanalysis, required by the Directive to have been completed by2004, had used a limited number of existing data and demon-strated that most British waters would fail to meet thennominally (and liberally) determined criteria for good status.This was also true for virtually all central andwestern Europeanmember states (Commission of the European Communities,2007; Quevauviller, 2006).

There are many documents that have emerged fromUKTAG, three of which, that summarise policy and give finaldetails, have been analysed for the current purposes. The firstis UK Environmental Standards and Conditions (Phase 1) 2006,Updated Report November 2007, (UKTAG, 2007a) which gave thefirst indications of the approach being taken for typology andsome details of standards, largely for the water chemistryassociated with ecological features. The updated reportfollowed a prior version that had been commented on by thepublic, though comments largely came from water-relatedorganisations and affected industries and almost no profes-sional ecologists. The second is the Phase 2 Report of June 2007(UK standards and conditions (Phase 2) SR1-2007, (UKTAG, 2007b)yet to be updated, which gives details of some determinands,including morphological criteria that add to the Phase 1 reportbutmaintain the same approaches. These two reports have analmost miscellaneous flavour to them and seem to representcollections of views from members of the group separatelyassociated with rivers, lakes and estuarine and coastal waters,respectively, rather than an integrated understanding andcommon approach.

The third report (UKTAG, 2007c) (UK Technical AdvisoryGroup on the Water Framework Directive Recommendations onSurface Water Classification Schemes for the purposes of the WaterFramework Directive. December 2007) is very different and amoreformal statement. UKTAG (2007c) exists in two versions,unfortunately both given the same date and title. The firstwas published on the UKTAG web site in December 2007, thesecond in March 2008. The second version is exactly similar tothe first up to page 50, then adds a fifth Appendix on biologicalclassification tools, with electronic links to more details ofthese. It is referred to here as UKTAG (2007d, updated). It hasclearly been through inspection by legal experts to make surethat it conforms in terminology and aspiration with therequirements of the Directive, but it excludes the details ofthe Phase 1 and 2 reports that give more insight into thethinking. Were UKTAG (2007c) the only source available, theimpression would be given of a straightforward attempt tofollow the provisions of the Directive and this may be theultimate intention. Careful examination, however reveals anumber of places where the ground is being obfuscated forreduced aspirations.

The document has six sections 1. The Directive's classifica-tion schemes; 2. Classification and river basin managementand planning; 3. Classification and monitoring data; 4.Presenting and using information on confidence class; 5.Spatial and temporal issues and classification; 6. Mapping andreporting the results of classification. There are also twoAnnexes concerning issues that had been raised by consulteesto the former documents, including connectivity and fish

37S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

access up rivers, damage to the headwaters of river systems(which have not previously been included as ‘controlledwaters’, for which responsibility is taken by the environmentagencies in the UK), and alien species. The list given for thelatter, evidence of damage by which might lead to down-grading of a water, is fairly comprehensive and includesinvertebrates, fish and plants and represents the firstacknowledgement of problems caused by alien animals(Garcia-Berthou et al., 2005; Jeschke and Strayer, 2005),including the common carp (Cyprinus carpio) (Cowx, 1997), amajor angling species.

Section 1 of UKTAG (2007c) is key to understanding theapproaches to be taken. The remaining, collectively muchlengthier, sections are essentially operational but the pre-occupation with the risk of misclassification is an interestingindicator of attitude. The worry is that a site might beclassified as worse than it really is and that therefore fundsmight be invested unnecessarily. UK civil servants commonlyuse the term ‘no gold plating’ to emphasise that provisions ofEU Directives should not be exceeded.

Section 1 (UKTAG, 2007c) has an early statement that ‘Theclass given to a particular water body will represent an estimate ofthe degree to which the structure and functioning of the aquaticecosystem supported by the surface water body have been altered byall the different pressures to which that body is subject.’ There ishowever, little evidence that either characteristic structure(the abundance of woody debris in upland rivers or theintactness of floodplains in the lowlands, for example(Vaughan et al., 2007)), has been considered or that ‘function’has much meaning. There are hints that function might beconsidered, for example in the criteria for connectivity in riversystems to allow fish migrations, but the classificationapproach is largely static and restricted in the determinandsto be used.

There are problems with typologies. The idea of habitattypes based on a number of features repeating themselves togive a general pattern is quite straightforward, yet UKTAGwasunable to embrace this concept and proposes different, andvery simple typologies for different determinands. There is atypology for oxygen and ammonia, a different one for fish(which is an invalid one as it uses an element required forclassification) in rivers, one based on geology and depth forlakes, yet a separate one for hydromorphology. The rivertypology (UKTAG 2007a,b) embraces only four types based oncombinations of two alkalinities and two altitudes, yetmentions 43 types (end groups) based on the invertebratetool to be used (UKTAG, 2007d updated), the lakes on 45potential combinations of depth, altitude and geology. Thelatter is defensible; the river typology is very confused. Therewas even a concept, in the draft version of UKTAG, 2007a, ofsalmon versus non-salmon rivers that ignores the historic factthat almost all UK rivers gave access to salmon and wereeffectively salmon rivers (Regan, 1911; Varley, 1967). Thisstatement is omitted in the updated version (UKTAG, 2007b)but the background thinking is still reflected in the typologybased on fish that appears in UKTAG (2007c). UKTAG iseffectively using a typology that mitigates against theachievement of good status in many UK rivers.

Within the chemical quality elements, temperature, dis-solved oxygen, pH and soluble reactive phosphorus concen-

tration are proposed as the sole determinands for rivers, anddissolved oxygen, conductivity, acid-neutralising capacity andtotal phosphorus concentration for lakes. Dissolved oxygenand dissolved inorganic nitrogen alone are to be used forestuarine and coastal waters. ‘For the remaining elements,UKTAG considers that the data currently available or the naturalvariability do not allow the derivation of a reliable standard for use inclassifying aquatic ecosystems.’ (UKTAG, 2007c). This is ques-tionable and the basis of the standards so far proposed for thesubstances to be used is sometimes equally doubtful. Whyshould available soluble reactive phosphorus (SRP) be used forrivers, with concentrations that are very high (20–40 µg P L−1

for high status, 40–120 µg P L−1 for good status) when lowervalues of total P are stipulated for the lakes which these riverssupply? Surveys of pristine (high quality) catchments in theUSA show values for total P in rivers for the closest region(Upper mid-west and North-east) of just over 10 µg P L−1. Thevalues quoted by UKTAG for SRP even for high status are thoseassociated with substantial contamination by waste watereffluent. The lake values, which use total phosphorus, aremuchmore in linewith experience of limnologists with valuesfor high status averaging between 5 µg P L−1 and 35 µg P L −1

and good status 8 µg P L−1 and 49 µg P L−1.Nitrate provides another example of confused thinking

mixed with political compromise. UKTAG (2007a) claims that‘Although nitrogen may have a role in the eutrophication in sometypes of freshwaters, we consider the general understanding of this tobe insufficient at present for it to be used as a basis for settingstandards or conditions. The possibility is too strong that thestatistical associations produced by these methods would representcorrelation between nitrogen and phosphorus (and other factors), andnot the standards for nitrogen that are truly needed to protect thebiology. For these reasons no standards for nitrogen are proposed inthis report.’

This statement is misleading. There is a substantialliterature on nitrate and eutrophication, with ample evidencethat values above about 1–2 mg N L−1 are associated withecological problems like reduction in diversity of plants andamphibians (Boar et al., 1989; Maberly et al., 2002; James et al.,2003, 2005; Bergstrom and Jansson, 2006; Camargo and Alonso,2006; Smith et al., 2006). Moreover, the risk assessment using astandard linked to that for nitrate vulnerable zones of 11.3 mgN L−1 in 2004 showed that even this standardwould not bemetin amajority of rivers in lowland UK so that failure to establishany nitrate standard for freshwaters can only be seen asmotivated by issues outside scientific knowledge. Agricultureis the major source of nitrate and acknowledged to be asubstantial political lobby. Strangely, it was possible forUKTAG to set a standard for estuaries (good, 0.28–0.42 mg NL−1 depending on type (UKTAG, 2007b)). The oddity is that itseems not to have occurred to UKTAG that amajor componentof estuarine water was river water a short time before and ashort distance away.

There are further issues concerning reference states.UKTAG (2007c) recommends that reference conditions shouldreflect ‘a state in the present or in the past corresponding to very lowpressure, without the effects of major industrialisation, urbanisationand intensification of agriculture, and with only very minormodification of physico-chemistry, hydromorphology and biology’.The wording follows the Directive but the issue of how ‘minor’

38 S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

is defined has, as always, been avoided and the concept isclearly to be defined ‘ad hoc’ depending on circumstances:‘Consequently, reference values have been derived from sites atwhich the quality element concerned is estimated to be in its referencecondition but other elements at the sites may not be so. These sitesmay also be in water bodies within which there are other sites atwhich the quality element may not be in its reference condition’. Thistoo suggests that the concept of types, with an ultimatereference ecology has been misunderstood.

In fact, UKTAG has abandoned the concept of typology ofhabitat sites, which is clearly the intention of the Directive `(Heiskanen et al., 2004) and expressed in the two systems Aand B which guide their designation, in favour of setting ofarbitrary standards for individual determinands with nogeographical reference whatsoever. This approach also con-flicts with the principle of ‘one-out all-out’. This is a widelyheld legal opinion that all the chemical and biological qualityelements (fish, invertebrates, plants) must comply with aparticular grade for a site to be classified by that grade. It is anecologically questionable principle, given the natural varia-bility of ecosystems where use of any one group can bemisleading (Everard, 2007; Moss et al., 2003), but if it isaccepted, a site that has only some conditions in the referencecondition, cannot be a reference site!

There is also an attempt (UKTAG, 2007c) to define goodstatus differently from the Directive. ‘At good ecological status,the Directive requires that the general physicochemical qualityelements comply with standards established by the Member Stateto protect the functioning of the ecosystem’. This is a crucialstatement. It subtlymoves away from theDirective's intentionthat good status should be only slightly different from highstatus to an arbitrary prescription by the member state. Fordifferent biological determinands (UKTAG, 2007d updated),reference states are frequently defined as ‘the best available’and lesser categories on a simple percentile basis of the dataset. The percentages and the proposed ecological qualityratios vary from one determinand to another and are eitherarbitrary or based on notions of the number of tolerant/intolerant species for a particular impact. It is not clear howthese have been determined.

A lack of understanding of how aquatic ecosystems areintimately linkedwith the ecology of their catchments is plain.‘UKTAG recommends that, where there is no evidence frommonitoring or risk assessments to the contrary, the status of awater body should be classed as being at high ecological status andgood chemical status. The confidence in such classifications (SeeSection 4) will depend on the quality of the information on pressuresand the reliability of the methods of risk analysis. For example, thereare places where water bodies are so free of significant pressures thata simple risk analysis will provide high confidence that the waterbodies are at high status’. The UK comprises a set of countrieswhere all of the landscape has been modified for humanactivity by substantial conversion of vegetation from foresttypes to grassland types (Godwin, 1956). A considerableliterature relates stream hydrology (a major component ofthe ‘structure’ of river systems) to catchment vegetation(Leopold et al., 1995). A separate literature shows that atmo-spheric pollution has influences through acidification on eventhe remoter areas of the north and west, even of the remoterhighlands of Scotland (Battarbee et al., 2005), a finding echoed

elsewhere ( Hayden, 1998; Bergstrom and Jansson, 2006). Onecan have high confidence on these two issues alone that thatUKTAG has indulged in wishful thinking rather than analysisof evidence.

Perhaps even more worrying is an attempt to movecompletely away from the intended comprehensive approachof the Directive in determining ecological quality: ‘The effect ofany errors in monitoring results is more dramatic when allied to theone-out all-out principle. This is because monitoring results for onlyone quality element or chemical need to wrongly suggest an adverseimpact in order for the water body to be assigned a lower class thanits true class. The probability of placing water bodies wrongly in alower class than their true class increases as more quality elementsare brought into the classification process. It is important that thisrisk is managed in order to reduce the risk of wasting resources onunneeded improvements in status. One way of helping with this, forexample, is to keep the number of quality elements to a minimum,consistent with assessing all the important pressures. UKTAGrecommends that only data on the condition of the quality elementsmost sensitive to the pressures placing a water body at risk are usedin classifying that water body’.

There are several problems with this. First the one-out all-out principle is an artefact of regarding aquatic systems ascollections of individual, independent determinands thathave no mutual interaction. This latter is a very peculiarinterpretation of the meaning of ecological status and onestudy has shown that the one-out all-out principle will tendto downgrade sites unjustifiably (Moss et al., 2003). However,the greater objection is that on this logic, ecology is reducedto a few elements, which is one reason why the Directiveproposed a more complex system to replace previous simplesystems. For example, the status of UK rivers was deter-mined largely on a few chemical symptoms of gross organicpollution. It would seem that UKTAG wishes to annul theDirective completely and return to a traditional water qualityapproach, merely using biological indicators as surrogates forchemical analyses.

6. Conclusion: blocks to progress

The European Commission is concerned about the status ofimplementation of the Directive and examination of the UKsituation confirms that the Commission is right to be so.The Commission (Commission of the European Commu-nities, 2007) reviewed progress based on documents it hadreceived and legal action taken already to counter infringe-ments by member states. It found that the number of waterbodies meeting WFD objectives was low, sometimes only1% and generally fewer than 40% in the longer establishedmember states other than Luxembourg. Significant pres-sures were diffuse pollution and physical degradation and,in southern Europe, over-exploitation of water resources.

The Commission assessed member state performance onfour criteria: conformity of legal transposition; compliancewith Article 3 (which concerns the administrative arrange-ments for the establishment of river basin districts); compli-ance with Article 5 (which concerns the characteristics ofbasins, analysis of problems and economic analysis); andoverall quality of reporting performance. Few states

39S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

transposed the Directive into national legislation by thedeadline and the quality of the transposition was low. Morepromising was establishment of river basin districts anddesignation of competent authorities. Article 5 progress wasgenerally poor with inadequate data. Based on generalreporting quality, the Commission urged focus on completingimplementation of earlier directives (particularly the UrbanWaste Water Treatment Directive and the Nitrates Directive,which are enfolded within the WFD), on creation of economicinstruments and on development of ecological classificationschemes. It felt that the current intercalibration exercise forthese has major deficiencies (see also Cardoso et al., 2005).

Other reports confirm the lack of enthusiasm. The Eur-opean Environmental Bureau (2004) writes of a ‘dark picture ofprogress’ and is particularly concerned at the lack of involve-ment with NGO and other non-government agencies, and anofficial regard of their views as problematic. A workshop of theRoyal Swedish Academy of Agriculture and Forestry (2006)took technical evidence from representatives of Finland,Estonia, Denmark, France, Sweden, and the UK. It found thatimplementation was largely a political process rather thanone dependent on scientific evidence. It concluded that ‘Froma scientific point of view the reference conditions seems (sic) to be likethe Holy Grail: you can search for it forever but you will never find it.Politically, on the other hand, it seems like the reference conditionsand good ecological status are more like rubber bands that can beextended as much or as little as you like’.

My experience of following progress on the implementa-tion of the Directive in the UK is in line with these official andunofficial documents. There appears to be a resistance toecological improvement, a concerted attempt to minimise theeffects of the Directive, whilst staying within a literalinterpretation of its generalities, and an obfuscation of theissues. Organisations, especially large ones, that have pro-ceeded in a particular way often continue in that way, forreasons that their previous investment in it has beensubstantial, even if it is no longer appropriate, and are veryresistant to change. This is known as the sink-cost effect(Scheffer and Westley, 2007; Scheffer et al., 2003).

Smaller organizations, less inherently powerful, prove tobemuchmore adaptable and resilient andmore readily able toadopt new approaches. Loring (2007) gives a graphic examplefor travelling circuses! Approaches to the Water FrameworkDirective in the UK have been plodding. Public consultationhas been either trivial (on the precise boundaries of the RiverBasin Districts, and the establishment of the statutoryenvironment agencies as the competent authorities forexample) or poorly advertised so that for the most part,commercial bodies, who maintain a close watch on legaldevelopments that might affect them, have responded. A richresource of genuine ecological expertise in the universitiesand non-governmental organisations has been avoided infavour of consultancy contracts controlled to deliver apolitically expedient product. The thinking of UKTAG can beexplicitly demonstrated to be confused.

There is manifestly also a lack of appreciation of theconnections between water management and all otheraspects of environmental policy. To quote Carter (2007) onthe planning system in the UK: ‘The contents of land use plans,therefore, tend to reflect political, social and economic priorities. It is

important to acknowledge that many factors affect planningdecisions aside from concerns for the water environment. This islikely to lead to some planning decisions being taken that run counterto achieving the goals of theWFD, or that do not take opportunities tolink more strongly with water issues. For example, planningauthorities may sanction development on floodplains, provide fornew housing in water stressed locations, or not promote the use ofwater efficiency technologies within their planning policies. Ulti-mately, for the requirements of the WFD to be implementedsuccessfully and effectively, political commitment to achieving thegoals of the Directive is crucial’.

Moreover, there is little appreciation of what restoration ofecological status means. There is a well-established back-ground philosophy, exemplified by Higgs (1997) and adoptedby the National Academy of Sciences in the USA:

‘…. restoration is defined as the return of an ecosystem to a closeapproximation of its condition prior to disturbance. In restora-tion, ecological damage to the resource is repaired. Both thestructure and the functions of the ecosystem are recreated.Merely recreating the form without the functions, or the functionsin an artificial configuration bearing little resemblance to anatural resource, does not constitute restoration. The goal is toemulate a natural, functioning, self-regulating system that isintegrated with the ecological landscape in which it occurs’.

It would be naïve to think that restoration to a status onlyslightly different from pristine would be possible in much ofthe area of most of the crowded countries of Europe. Anupland river in such a state would have nearly unbrokenforest around it, top predators like wolves and brown bears,and only the lightest of cultivation or stock keeping in a tinypart of the catchment. This might be possible for the remoterareas of Scotland and Scandinavia but more extensively onlyif a possible runaway greenhouse effect makes most of usextinct (Lovelock, 2006). However, there is a differencebetween admitting these difficulties in the light of arealisation of what high ecological quality really means anda covert attempt to weasel the wording of the Directive.Honest confrontation of our problems is needed not institu-tional denial but no such transparency seems to haveinformed thinking in the UK at least. Moreover, enormoussums are being spent on the WFD throughout Europe, andwill be committed to future monitoring, to make assess-ments of secondary details, in the interests of a falseprecision, when a very simple accurate assessment ofecological quality, based on current ecological understandingcould be rapidly and easily done (Moss, 2007; Moss et al.,2003). It is depressing that a passé chemical hegemony ingovernment agencies is being used to undermine a Directivethat could have sparked an approach to environmentalmanagement that would have helped solve much greaterproblems than those of European waters.

All of which bodes very badly for any attempts to solve theeven greater environmental problems of climate change,world population, poverty and food supply, the end of the oileconomy and the substantial destruction of the Earth'snatural ecosystems. If the arguments developed above con-cerning the meaning and importance on a global scale ofecological quality are accepted, there can be no greater priority

40 S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

than restoration of a substantial proportion of the Earth'snatural capital, its ‘total environment’. This surpasses politicalcompromise. And if ‘The objectives of management of land, waterand living resources are a matter of societal choice’ (UNESCO, 2000),the choice increasingly looks like one for collapse (Diamond,2006) as opposed to survival.

Acknowledgement

Support was provided by the EU Framework 6 Project,EUROLIMPACS (GOCE-CT-2003-505540).

R E F E R E N C E S

Battarbee RW, Curtis CJ, Binney HA, editors. The future of Britain'suplandwaters. London: Ensis Publishing,UniversityCollege; 2005.

Baattrup-Pedersen A, Szoszkiewicz K, Nijboer R, O'Hare M, FerreiraT. Macrophyte communities in unimpacted European streams:variability in assemblage patterns, abundance and diversity.Hydrobiologia 2006;566:179–96.

Baxter CV, Fausch KD, Saunders WC. Tangled webs: reciprocalflows of invertebrate prey link streams and riparian zones.Freshwat Biol 2005;50:201–20.

Bergstrom AK, Jansson M. Atmospheric nitrogen deposition hascaused nitrogen enrichment and eutrophication of lakes in thenorthern hemisphere. Global Change Biol 2006;12:635–43.

Boar RR, Crook CE, Moss B. Regression of Phragmites australisreedswamps and recent changes of water chemistry in theNorfolk Broadland. Aquat Bot 1989;35:41–55.

Calman RJ, Bilbo RE, Schindler DE, Belfield JM. Pacific salmon,nutrients and the dynamics of freshwater ecosystems.Ecosystems 2002;5:399–417.

Camargo JA, Alonso A. Ecological and toxicological effects ofinorganic nitrogen pollution in aquatic ecosystems: a globalassessment. Env Internat 2006;32:831–40.

Cardoso AC, Solimini AG, Premazzi G, Birk S, Hale P, Rafael T, et al.Report on harmonisation of freshwater biological methods.EUR 2005;21769 EN.

Carstensen J. Statistical principles for ecological status classificationof Water Framework Directive monitoring data. Mar Poll Bull2007;55:3–15.

Carter JG. Spatial planning, water and the Water FrameworkDirective: insights from theory and practice. Geog J2007;173:330–42.

Chapman PM. Bioassay testing for Australia as part of water-quality assessment programs. Austr J Ecol 1995;20:7–19.

Commission of the European Communities. Towards sustainablewater management in the European Union—first stage in theimplementation of the Water Framework Directive. 2000/60/ECCOM 2007;128:1–13.

Cowx I. Introduction of fish species into European fresh waters:economic successes or ecological disasters? Bull Franc PechePisc 1997;344/5:57–77.

Daly HE. Ecological economics and the ecology of economics.Cheltenham, UK: E Elgar Publcations; 1999.

Darwin C. On the origin of species bymeans of natural selection, orthe preservation of favoured races in the struggle for life. 1sted. London: John Murray; 1859.

Dawkins R. The blind watchmaker. New York: Norton; 1986.Diamond J. Collapse: how societies choose to fail or succeed.

London: Penguin Books; 2006.Dworak T, Gonzalez C, Laaser C, Interwies E. The need for new

monitoring tools to implement the WFD. Env Sci Pol2007;8:30130–6.

Elster HJ. Das limnologische Seetypensystem, Ruckblick undAusblich. Verh. int. theor. angew. Limnol 1958;13:101–20.

European Community. Directive 2000/60/EC of October 23 2000 ofthe European Parliament and of the Council establishing aframework for community action in the field of water policy.Off J Eur Comm 2000;L327:1–72.

European Environmental Bureau. EU Water Policy. Making theWater framework Directive work. The quality of nationaltransposition and implementation. A snapshot. Results of aquestionnaire by the European Environmental Bureau. 2004.

Everard M. Selection of taxa as indicators of river and freshwaterwetland quality in theUK. Aquat Cons: Mar Freshwat Ecos 2007.doi:10.1002/aqc.896 .Published online in Wiley InterScience(www.interscience.wiley.com).

Fitkau EJ. Role of caimans in the nutrient regime of mouth-lakes inAmazon effluents (a hypothesis). Biotropica 1970;2:138–42.

Furch K. Water chemistry of the Amazon Basin: the distribution ofchemical elements among freshwaters. In: Sioli H, editor. TheAmazon. Dordrecht: Junk; 1984. p. 167–200.

Garcia-Berthou E, Alcaraz C, Pou-Rovira Q, Zamora L, Coenders G,Feo C. Introduction pathways and establishment rates ofinvasive aquatic species in Europe. Can J Fish Aquat Sci2005;62:453–63.

Godwin H. The history of the British flora. A factual basis forphytogeography. Cambridge, UK: Cambridge University Press;1956.

Hayden BP. Ecosystem feedbacks on climate at the landscapescale. Phil Trans Roy Soc Lond 1998;353:5–18.

Hassall C, Thompson DJ, French GC, Harvey I. Historical changes inthe phenology of British Odonata are related to climate. GlobalCh Biol 2007;13(5):933–41.

Hatton-Ellis T. The hitchhiker's guide to the Water FrameworkDirective. Aquatic Conserv: Mar Freshw Ecosyst 2008;18:111–6.

Heiskanen, A.S., van de Bund, W., Cardoso, A.C., Noges, P.,Towards good ecological status of surface waters in Europe—interpretation and harmonization of the concept. Wat SciTechnol 2004;49:169–177.

HeringD, Johnson RK, Buffagni A. Linking organismgroups—majorresults and conclusions from the STAR Project. Hydrobiologia2006;566:109–13.

Hickling R, Roy DB, Hill JK, Thomas CD. A northward shift of rangemargins in British Odonata. Global Ch Biol 2005;11(3):502–6.

Higgs ES. What is good ecological restoration. Cons Biol 1997;11(2):338–48.

Illies J. Limnofauna Europea. 2nd ed. Jena: G. Fischer; 1978.Intergovernmental Panel on Climate Change (IPCC). Climate

change 2007: impacts, adaptation and vulnerability. Summaryfor policymakers. IPCC Secretariat, Geneva; 2007.

James C, Fisher J, Moss B. Nitrogen driven lakes: the Shropshireand Cheshire Meres? Arch. Hydrobiol. 2003;164:249–66.

James C, Fisher J, Russell V, Collings S, Moss B. Nitrate availabilityand hydrophyte species richness in shallow lakes. FreshwatBiol 2005;50:1049–63.

Jeschke JM, Strayer DL. Invasion success of vertebrates in Europeand North America. Proc Nat Acad Sci 2005;102:798–7202.

Kelly M, Juggins S, Guthrie R, Pritchard S, Jamieson J, Rippey B, etal. Assessment of ecological status in U.K. rivers using diatoms.Freshwat Biol 2008;53:403–22.

Kidd S, Shaw D. Integrated water resource management andinstitutional integration: realising the potential of spatialplanning in England. Geog J 2007;173:312–29.

Leopold A. Thinking like a mountain. A sand county almanac andsketches here and there. New York: Oxford University Press;1949.

Leopold LB, Wolman MG, Miller JP. Fluvial processes ingeomorphology. New York, USA: Dover; 1995.

Likens GE, Bormann FH, Pierce RS, Eaton JS, Johnson NM.Biogeochemistry of a forested ecosystem. New York: Springer-Verlag; 1971.

41S C I E N C E O F T H E T O T A L E N V I R O N M E N T 4 0 0 ( 2 0 0 8 ) 3 2 – 4 1

LoringPA.Themost resilient showonearth: the circusasamodel forviewing identity, change, and chaos. Ecol. Soc 2007;12(1):9[online] URL: http://www.ecologyandsociety.org/vol12/iss1/art9/.

Lovelock JE. The ages of Gaia: a biography of our living earth. NewYork and London: W. W. Norton; 1988.

Lovelock JE. The living earth. Nature 2003;426:769–80.Lovelock J. The revenge of Gaia. London: Penguin Books; 2006.Lovelock J, Margulis L. Atmospheric homeostasis: the Gaia

hypothesis. Tellus 1974;26:2–10.Maberly SC, King L, Dent MM, Jones RI, Gibson CE. Nutrient

limitation of phytoplankton and periphyton growth in uplandlakes. Freshwat Biol 2002;47:2136–52.

Margulis L, Lovelock JE. Biological modulation of the earth'satmosphere. Icarus 1974;21:1–19.

Millennial Ecosystem Assessment Board. Millennial EcosystemAssessment Synthesis Report. United Nations EnvironmentProgramme, New York, USA; 2005.

Miyasaka SC, Habte M. Plant mechanisms and mycorrhizalsymbioses to increase phosphorus uptake efficiency. CommSoil Sci Plant Anal 2001;32:1101–47.

Moss B, Johnes P, Phillips G. August Thienemann and LochLomond — an approach to the design of a system formonitoring the state of north-temperate standing waters.Hydrobiologia 1994;290:1–12.

Moss B. Words and water. Br Ecol Soc Bull 2001:8–11 March.Moss B. Shallow lakes, the water framework directive and life.

What should it all be about? Hydrobiologia 2007;584:381–94.MossB, StephenD,AlvarezC, Becares E,VanDeBundW,Collings SE,

et al. The determination of ecological status in shallow lakes— atested system (ECOFRAME) for implementation of the EuropeanWater Framework Directive. Aquat Cons Mar Freshwat Ecos2003;13:507–49.

Noble RAA, Cowx IG. Development of fish-based methods forassessment of ecological status in English and Welsh rivers.Fish Man Ecol 2007;14:495–508.

Noges P, Van de Bund W, Cardoso AC, Heiskanen AS. Impact ofclimatic variability on parameters used in typology andecological quality assessment of surface waters—implicationson the Water Framework Directive. Hydrobiologia2007;584:373–9.

Parmesan C. Ecological and evolutionary responses to recentclimate change. Ann Rev Ecol Evol Syst 2006;37:637–69.

Quevauviller P. The progress state of European implementation ofthe WFD. Houille Blanche-Revue Internationale De L Eau2006;4:55–7.

Rao M, Terborgh J, Nunez P. Increased herbivory in forest isolates:implications for plant community structure and composition.Cons Biol 2001;15:624–33.

Raven JA, Andrews M, Quigg A. The evolution of oligotrophy:implications for the breeding of crop plants for low inputagricultural systems. Ann Appl Biol 2005;146:261–80.

Ripple WJ, Beschta RI. Wolves, elk, willows, and tropic cascades inthe upper Gallantin Range of South-western Montana, USA.Forest Ecol Manage 2004a;200:161–81.

Ripple WJ, Beschta RI. Wolves and the ecology of fear: canpredation risk structure ecosystems? BioScience2004b;54:755–66.

Regan CT. The freshwater fishes of the British Isles. London:Methuen; 1911.

Royal Swedish Academy of Agriculture and Forestry. WaterFramework Directive implementation in a European

perspective. Report of a workshop. Kungl. Skogs- ochLantbruksakademiens; 2006. p. 8.

Sandin L, Johnson RK. Ecoregions and benthic macroinvertebrateassemblages of Swedish streams. J N Amer Benthol Soc2000;19:462–74.

Sandin L, Verdonschot PFM. Stream and river typologies — majorresults and conclusions from the STAR Project. Hydrobiologia2006;566:33–7.

Scheffer M, Westley FR. The evolutionary basis of rigidity: locks incells, minds, and society. Ecol Soc 2007;12:36 [online] URL:http://www.ecologyandsociety.org/vol12/iss2/art36/.

Scheffer M, Westley F, Brock W. Slow response of societies to newproblems: causes and costs. Ecosystems 2003;6:493–502.

Smith GR, Temple KG, Dingfelder HA, Vaala DA. Effects of nitrateon the interaction of the tadpoles of two ranids (Rana calamitansand Rana catesbeiana). Aquat Ecol 2006;40:125–30.

Smith RA, Alexander RB, Schwarz GE. Natural backgroundconcentrations of nutrients in streams and rivers of theconterminous United States. Env Sci Technol 2003;37:3039–47.

Steyaert P, Ollivier G. How ecological assumptions frame technicaland social change. Ecol Soc 2007;12:1–18.

Surridge B, Harris B. Science-driven integrated river basin man-agement: a mirage? Interdisc Sci Rev 2007;32:298–312.

Terborgh J. The big things that run the world— a sequel toWilson,E.O. Cons Biol 1988;2:402–3.

Terborgh J, Feeley K, Silman M, Nunez P, Balukjian B. Vegetationdynamics of predator-free land-bridge islands. J. Ecol.2006;94:253–63.

UKTAG. Environmental Standards and Conditions (Phase 1) 2006.Updated Report November 2007. Bristol, UK: EnvironmentAgency; 2007a.

UKTAG. Phase 2 report of June 2007 (UK standards and conditions(Phase 2) SR1-2007. Bristol, UK: Environment Agency; 2007b.

UKTAG. Recommendations on surface water classificationschemes for the purposes of the Water Framework Directive.December 2007. Bristol, UK: Environment Agency; 2007c.

UKTAG. Recommendations on surface water classificationschemes for the purposes of the Water Framework Directive.December 2007. Bristol, UK: Environment Agency; 2007d(updated).

United Nations Educational, Social and Cultural Organisation(UNESCO). Solving the puzzle. The ecosystem approach andbiosphere reserves. Paris: UNESCO; 2000).

Varley ME. British freshwater fishes: factors affecting theirdistribution. London: Fishing News Books; 1967.

Vaughan IP, DiamondM, Gurnell AM, Hall KA, Jenkins A, Milner NJ,et al. Integrating ecology with hydromorphology: a priority forriver science and management. Aquat Cons Mar FreshwatEcosyst 2007. doi:10.1002/aqc.895 . Published online in WileyInterScience (www.interscience.wiley.com).

Verdonschot PFM. Evaluation of the use of Water FrameworkDirective typology descriptors, reference sites and spatial scalein macroinvertebrate stream typology. Hydrobiologia2006;566:39–58.

Wade AJ, Neal C, Butterfield D, Futter MN. Assessing nitrogendynamics in European ecosystems, integrating measurementand modelling: conclusions. Hydrol & Earth Syst Sci2004;8:846–57.

Wasson JG, Chandesris A, Pella H, Blanc L. Hydro-ecoregions: afunctional approach of river typology for the European WaterFramework Directive Ingénieries-EAT, vol. 40. ; 2004. p. 1–12.