Waterbirds around the world

A global overview of the conservation, management and research of the

world's waterbird flyways

Edited by G.C. Boere, C.A. Galbraith and D.A. Stroud

Assisted by L.K. Bridge, I. Colquhoun, D.A. Scott,

D.B.A. Thompson and L.G. Underhill

EDINBURGH, UK: THE STATIONERY OFFICE

Extract only - complete publication at www.jncc.gov.uk/worldwaterbirds

© Scottish Natural Heritage 2006

First published in 2006 by The Stationery Office Limited71 Lothian Road, Edinburgh EH3 9AZ, UK.

Applications for reproduction should be made to Scottish Natural Heritage,Great Glen House, Leachkin Road, Inverness IV3 8NW, UK.

British Library Cataloguing in Publication DataA catalogue record for this book is available from the British Library

ISBN 0 11 497333 4

Recommended citation:Boere, G.C., Galbraith, C.A. & Stroud, D.A. (eds). 2006.

Waterbirds around the world. The Stationery Office, Edinburgh, UK. 960 pp.

Names used for geographical entities do not imply recognition, by the organisers of the Waterbirds around the world conference or othersupporting organisations or governments, of the political status or boundaries of any particular territory. Names of territories used (andany alternatives) are included solely to help users of this publication apply information contained within this volume for waterbird conservation purposes. The views expressed in papers included within this volume do not necessarily represent views of the editorsor the organisations and governments that supported the conference and this publication.

Cover photography: Whooper Swans Cygnus cygnus arriving at Martin Mere, England. Photo: Paul Marshall.

(www.paulmarshallphotography.com)

Copyright of all photographs used in this publication resides with the named photographers.

Extract only - complete publication at www.jncc.gov.uk/worldwaterbirds

ABSTRACTThe monitoring of waterbirds in The Netherlands has a long tradi-tion. It has evolved from various regional, seasonal and species-specific projects into an integrated monitoring scheme which aimsto assess waterbird trends at site and national level, and operateswithin the framework of a governmental network of ecologicalmonitoring. It is also incorporated in international monitoringprogrammes such as the International Waterbird Census (IWC)and the Wadden Sea Trilateral Monitoring and AssessmentProgramme (TMAP). The backbone of the scheme comprisesmonthly counts at all sites supporting large numbers of non-breeding waterbirds. This paper provides an overview of themonitoring scheme, starting with its aims and targets. It describes the census scheme, and discusses the sites that arecovered and the methods used to calculate indices and trends. Italso shows how data are used for policy and management issues,taking the EU Birds Directive as an example. Data from the water-bird monitoring project have been used extensively to designateSpecial Protection Areas in The Netherlands. Population trends inindividual species are used to assess conservation status which isthen translated into the aims of future management plans.Monitoring will gain importance by assessing the numbers andtrends against targets and a system of alerts. As waterbirds arehighly dependent on their food supplies and often respond quicklyto changes in food availability, they are cost-effective indicators ofchanges in the broader ecological quality of their habitats.

INTRODUCTIONThe Netherlands are of outstanding importance for many water-birds. The country’s position along the East Atlantic Flyway, thelarge amount of wetland habitat and the prevailing mild wintersattract internationally important numbers of waterbirds during allstages of their life cycle. The monitoring of waterbirds in TheNetherlands has a long tradition. As early as the 1950s and 1960s,waterbird counts were being carried out regularly. The start of the International Waterbird Census (IWC) in the late1960s further improved the network and stimulated many volun-teer and professional bird-watchers to participate. In addition tothe international midwinter counts, several regional (oftenmonthly) censuses were established in important wetlands,e.g. in the Wadden Sea, along the River Rhine and its tributaries,at Lake ijsselmeer, and in the Delta area in the south-west of thecountry. Counts of geese and swans had already started in the1960s, both in wetlands and in agricultural areas. In response tothe growing need for comprehensive national monitoring data, allthese surveys have recently been combined in a national waterbird

monitoring scheme (Koffijberg et al. 2000). This scheme is part ofthe so-called Network Ecological Monitoring, a governmentalinstrument to collect monitoring data for various groups of speciesincluding, for example, plants, dragonflies and mammals (vanStrien 2005). Through this scheme, data required for internationaltreaties and conventions, such as the European Union (EU) Birdsand Habitats Directives, Ramsar Convention on Wetlands, andAfrican-Eurasian Migratory Waterbird Agreement (AEWA) of theBonn Convention, also become available.

Waterbird monitoring in The Netherlands is currently a jointscheme of the Ministry of Agriculture, Nature and Food Quality,the Ministry of Transport, Public Works and Water Management,BirdLife The Netherlands and Statistics Netherlands, and is co-ordinated by the SOVON Dutch Centre for Field Ornithology. At present about 1 500 observers, of whom a large proportionare dedicated volunteers, participate in the scheme each year.This paper provides an overview of the monitoring programme,and gives examples of results and their role in the developmentof environmental policies and management plans.

WHY WATERBIRDS?Statistics on the status and trends in numbers of waterbirds arerequired for the conservation of bird populations and biodiversity,as well as for the conservation of their (wetland) habitats. Thisacknowledges the role of waterbirds as indicator species. Waterbirdnumbers can be assessed rather accurately by counts, and sincethey often respond quickly to changes in their environment, theirstatus can be a powerful indicator of changes in other organisms inthe ecosystem, which are often more difficult to measure or forwhich only recent data are available. At Lake Veluwemeer, forinstance, a close relationship was found between the numbers ofsome waterbird species and the amount of submerged macrophytes(Fig. 1). In the 1970s and 1980s, submerged vegetation was verylimited because of eutrophication. As a consequence, the numbersof waterbirds were low. However, an improvement in water qualityled to an increase in aquatic vegetation, which was followed by anincrease in waterbird numbers. By using this relationship, it waspossible to reconstruct the ups and downs in the abundance ofsubmerged vegetation and other aspects of the ecosystem beforethe 1970s, when only bird numbers were being surveyed. In thisway, qualitative descriptions of the lake could be underpinned with(modelled) data. Several additional relationships exist betweenwaterbird numbers, food supply, water quality parameters, andpolicy and management issues in which waterbirds might act aseffective indicators of underlying processes at lower trophic levels(Noordhuis & Koffijberg 2004).

463

Waterbirds around the world

Long-term waterbird monitoring in The Netherlands: a tool for policyand management

Marc van Roomen1, Kees Koffijberg1, Ruurd Noordhuis2 & Leo Soldaat 3

1 SOVON Dutch Centre for Field Ornithology, Rijksstraatweg 178, 6573 DG, Beek-Ubbergen, The Netherlands. (email: marc.vanroomen@sovon.nl; kees.koffijberg@sovon.nl)2 RIZA, PO Box 17, 8200 AA Lelystad, The Netherlands. (email: R.Noordhuis@riza.rws.minvenw.nl)3 Statistics Netherlands, PO Box 4000, 2270 JM, Voorburg, The Netherlands. (email: LSLT@CBS.nl)

Roomen, M. van, Koffijberg, K., Noordhuis, R. & Soldaat, L. 2006. Long-term waterbird monitoring in The Netherlands: a tool forpolicy and management. Waterbirds around the world. Eds. G.C. Boere, C.A. Galbraith & D.A. Stroud. The Stationery Office,Edinburgh, UK. pp. 463-470.

AIMS OF THE MONITORING PROGRAMMEA clear definition of the aims and targets of a monitoring schemewill help to direct efforts to improve the programme and providethe information that is required. This includes selection of thetarget species which are to be covered and a definition of thegeographical scale for which data are required. Fortunately,waterbirds are relatively easy to count, and it is often the casethat numbers of all species can be assessed when visiting a site.All species of divers, grebes, cormorants, herons and allies,swans, geese, ducks, rails, coots, waders, gulls and terns occur-ring in The Netherlands are included in the counts. Non-nativespecies which have been introduced or escaped are also covered.In addition, a few species of raptors and passerines, which arespecific to wetlands and easy to cover while counting otherwaterbirds, are included. For analyses and trend calculations(see below), priority is given to species which are important in

464

Waterbirds around the world

Fig. 1. Trends in biomass of herbivores and stonewort in Lake Veluwemeer, The Netherlands. A strong correlation is found between waterbird numbers

(expressed in biomass to account for consumption by species of different size) and stonewort biomass (r = 0.96, df = 29).

Fig. 2. Trends in the numbers of Gadwall Anas strepera in The

Netherlands and in North-west Europe. The trend in North-west Europe

is taken from Delany et al. (1999). Numbers in The Netherlands have

grown faster than those in North-west Europe (t = -12.4; df = 12,

p < 0.001, based on a non-linear regression analysis).

Fig. 3. Trends in the numbers of Great Crested Grebe Podiceps cristatus

at Lake Veluwemeer, The Netherlands, based on numbers from January

counts (dashed line) and on bird-days calculated from monthly counts in

September-April (solid line).

Fig. 4. Phenology of the number of target species of waterbirds (Annex I of the

EU Birds Directive and indicator species in national freshwater bodies) occurring in

The Netherlands. Species are only included for a given month if their numbers in

that month are equal to at least 80% of the peak count during the 12-month period.

relation to the EU Birds Directive. This aims to safeguard birdspecies and their habitats within the EU, and obliges memberstates to carry our regular monitoring at important sites (SpecialProtection Areas, SPAs).

At present, indices and trends are calculated for 53 species(van Roomen et al. 2004). These are calculated at three differentgeographical scales: individual sites, regions and the country asa whole (national trends). Priorities for trend analyses at sitelevel are SPAs designated under the EU Birds Directive (in mostcases these sites have also been designated as Ramsar sites under

the Ramsar Convention on Wetlands), national freshwater bodies(water bodies managed by the government and overlapping to alarge extent with SPAs), and agricultural sites important forgeese and swans (these often lack any protection status).

In addition to these national aims, the counts also contributeto international monitoring programmes, including theInternational Waterbird Census (IWC) co-ordinated by WetlandsInternational and the Trilateral Monitoring and AssessmentProgramme (TMAP) for the international Wadden Sea. Thelatter is a trilateral co-operation between The Netherlands,

Fig. 5. Monitoring sites in The Netherlands used for the calculation of national trends in populations of waterbirds (except for swans, geese and seaducks).

465

Waterbirds around the world

Germany and Denmark for monitoring and management policiesin the Wadden Sea. International population estimates and trendsare important references for national monitoring. They arenecessary for separating trends at national or site level fromtrends at flyway level (Fig. 2).

OUTLINE OF THE MONITORING PROGRAMMESeveral counts a yearA single count in January is the best strategy for an internationalmonitoring scheme, since the distribution of waterbirds at thistime of year is rather stable, and in many countries a large surveyeffort can only be achieved once a year. However, for monitoringat national level and at the level of individual sites, more countsare required per year, since a single count in January covers onlya tiny part of the annual cycle of waterbirds. For example, speciesoccurring on migration in spring and autumn and those occurringin moulting concentrations in late summer are not covered at allin January, but might be the most important species at many sites.Moreover, as phenology patterns or wintering strategies mightchange in time (e.g. a switch in distribution as a result of globalwarming), national monitoring of waterbirds benefits fromcounts which are carried out at all times of the year. Thus, trendsin January may give a very misleading picture of the overallabundance of a species, as shown in Fig. 3 for the Great CrestedGrebe Podiceps cristatus at Lake Veluwemeer. When only theJanuary counts are considered, it appears that Great CrestedGrebes were very scarce in the 1960s and 1970s, and increasedafter the mid-1980s. However, it is known that the species didoccur in substantial numbers in the 1960s and 1970s, but mainlyin spring and autumn, with only a few birds remaining in January.This only becomes apparent when one takes into account allmonthly counts between September and April. It then becomesclear that the current use of the site is rather similar to that in the1960s and 1970s. The recent increase in January is simply theresult of an increase in the numbers of wintering birds (i.e. aprolonged stay of birds in autumn).

Since most countries will have a mixture of resident species,passage migrants and winter visitors, waterbird counts shouldideally cover all months of the year. In The Netherlands, eachmonth of the year appears to be of some importance as regardsthe number of target species occurring in important numbers

(Fig. 4). The main reason for this is the large variation inphenology of the species involved. In freshwater lakes and otherinland sites, the period September-April has the highest abun-dance of waterbirds. Some large lakes are also used in othermonths, e.g. when moulting concentrations occur in latesummer. In estuaries, however, all months of the year are impor-tant. Based on these patterns, it was decided to use monthlycounts between September and April to cover inland sites andmonthly counts throughout the year to cover coastal estuaries(e.g. the Wadden Sea and Delta area) and some large freshwaterlakes (e.g. Lake IJsselmeer). In the Wadden Sea, monthly countsare achieved through a combination of monthly counts at asample of sites and five total counts throughout the year. This area is so large that monthly total counts would be beyondthe capability of the volunteer counters.

Selection of sites for monitoringBecause of the large amount of wetland habitat in TheNetherlands and large part of the year in which importantnumbers of waterbirds are present, it would be a considerabletask to include all water bodies in the monitoring scheme.Therefore, a selection of sites has been identified which coversall important waterbird concentrations (Fig. 5). This selectionincludes all SPAs and national freshwater bodies, and thus allpriorities for site monitoring. For most waterbird species exceptgeese, swans and seaducks, these sites together support a highproportion of the national populations (Fig. 6). The core areasfor geese and swans are in agricultural areas, and additionalcounts are carried out for these species in these areas (Fig. 7).Seaducks occur in large numbers only in the coastal zone of theNorth Sea and open waters of the Wadden Sea. Both of theseareas are not covered by monthly land-based counts, but arecounted once a year (in January) by an aerial survey. Additionalwaterbird counts are made in January at less important sites aspart of the International Waterbird Census.

Missing counts and trend analysisDue to the different history of the various census projects, not allsites have similar and comparable series of data. Some sites havemonthly counts dating back to 1975, while others have countsstarting in the 1980s or lack counts in important months. Even though coverage has increased over the decades (Fig. 8),some counts are still missed because of poor weather conditions(e.g. fog), absence of the observer or other reasons. Thus, whencompiling uniform data sets for trend analyses, considerationhas to be given to the missing counts. Standardized methodshave been developed to estimate numbers at the sites not counted(imputing). In The Netherlands, we have adopted the methodused in the Wetland Bird Survey in the British Isles (for back-ground, see Underhill & Prys-Jones 1994). Species-specificmodels are used to account for site, month and year factors. Thisimputing is performed by using the U-index package (Bell1995). Missing counts are imputed on the basis of the smallestcount units, and count units are stratified according to theirseasonal and annual patterns (Soldaat et al. 2004).

Trend analyses are based on annual estimates of occurrenceper species for a certain geographical area (either site, regionalor national level). This annual estimate of occurrence is definedas the seasonal sum of the counted and estimated (imputed)numbers throughout the year (or in September-April, depending

466

Waterbirds around the world

Fig. 6. The proportions of the national January totals of 28 species of

waterbirds that occur within the monitoring sites shown in Fig. 5.

on the site; see above). Trends are calculated by using a linearregression of the log-transformed seasonal sums, and areassessed for different periods (e.g. since the start of monitoringor over the last 10 years). The output is classified according tocertain definitions, including definitions of stable and fluctu-ating trends. In the near future, trend calculations will beextended by incorporating flexible trends (Visser 2004) whichare better able to deal with strong fluctuations.

USE OF WATERBIRD DATA IN RELATION TO THE EUBIRDS DIRECTIVEThe EU Birds Directive has become increasingly important fornature conservation policy and management in The Netherlands(and other EU countries). Together with the EU HabitatsDirective, a network of sites, the so-called Natura 2000 sites, hasbeen set up within the EU for the conservation of biodiversity.Special Protection Areas (SPAs) are a part of this Natura 2000

467

Waterbirds around the world

Fig. 7. Staging sites of swans and geese in The Netherlands used for the calculation of national trends.

network. As a first step, national waterbird counts (and breedingbird surveys) were used to identify SPAs on the basis of the listof target species (species listed in Annex I of the Birds Directive)and 1% threshold values of the Ramsar Convention (van Roomen et al. 2000). To date, 79 sites have been designatedas SPAs in The Netherlands. These include all important wetlandsites but do not cover many agricultural sites important for populations of geese and swans.

Designation of important sites is only a first step towardsbetter protection of the species involved. The conservationstatus of the species needs to be assessed regularly, both atnational and site level, and monitoring plays an important rolein evaluating trends in abundance. A methodology for assessingthe conservation status of species against the formulated targetsis currently being developed. Overviews of the trends in water-bird species listed in the EU Birds Directive are already avail-able at site level (Fig. 9). However, judgements on thefavourable or unfavourable conservation status of a species (andthus the direction of the observed trends) are not always easy tomake. EU environmental policies not only cover biodiversity,but also deal, for example, with improved environmental condi-tions in water bodies. The Water Framework Directive (WFD),for instance, aims to tackle the problem of eutrophication. Adecline in the nutrients in water bodies, however, might changethe food web and thus have a (negative) effect on the abundanceof some waterbirds. An example of this paradox is the relation-ship between the numbers of Tufted Ducks Aythya fuligula andstocks of the Zebra Mussel Dreissena polymorpha at LateMarkermeer. As the levels of eutrophication were reduced,mussel stocks declined and the water became more turbid. Thenumbers of Tufted Ducks responded negatively and declined(Fig. 10). From the viewpoint of the Water FrameworkDirective, the lower eutrophication levels can be regarded as asuccess, but according to the EU Birds Directive, site manage-ment should undertake action to halt the decline in TuftedDucks. Therefore, definitions of favourable conservation statusshould not simply refer to bird numbers, but should also takeinto account the ecological potential of a site.

DISCUSSION The monitoring programmeThe monitoring of waterbirds in The Netherlands is onlypossible at the present time because of the high interest and co-operation of many volunteer bird-watchers. The participationand commitment of many site managers and institutions are alsoessential. We are very fortunate that as early as the 1970smonthly censuses had started in several important wetlandsystems. These censuses provided a firm basis for the currentmonitoring scheme. Although we now have a smooth-runningand well-used scheme, a number of compromises had to bemade while re-designing the programme.

Firstly, because of the limitations to observer effort, we workwith a selection of monitoring sites, e.g. SPAs, national fresh-water bodies and internationally important goose and swanstaging areas (Figs. 5 & 7). One could argue that these sites onlyrepresent optimal sites and do not take into account dynamics insite use. As is known from studies of individual species (e.g. ofDark-bellied Brent Goose Branta bernicla bernicla, Ebbinge1992), optimal sites are often the first to become saturated whena population increases, and less optimal sites are abandoned assoon as a population experiences a downward trend. The impactof such processes is thought to be very small in our monitoringscheme. For many species, 50% or more of the national popula-tion is included within the selection of sites. For some goosespecies, nearly the entire national (or even international) popula-tion is covered by the current selection of monitoring sites. Thus, the dynamics in site use are well taken into account, as alarge number of sub-optimal sites are already part of the networkof monitoring sites. The January census, which covers manymore sites, might eventually indicate sites that are becomingincreasingly important, at least in winter.

Secondly, underestimation of bird abundance might occur,since the monthly counts will not always cover all the migrationpeaks of a species, especially when these occur during a periodof only a few days or always between the mid-monthly counts.Therefore, monitoring in the Wadden Sea within the frameworkof the Trilateral Monitoring and Assessment Programme wasinitially set up with twice-monthly counts in sample census areas(Rösner 1994, Günther 2003). However, in the Dutch WaddenSea, no representative sample of sites could be established.Hence, the errors in estimates of bird abundance caused by thenon-representative sample of sites were larger than in the lessaccurate numbers derived from monthly counts in the entire area(van Roomen et al. 2002). Monthly counts represent reliable esti-mates of the abundance of birds, especially for the purposes oflong-term trend analysis. Only in the case of very small sites, orto derive numbers for qualification purposes (e.g. when desig-nating SPAs), are additional counts needed to arrive at a morereliable estimate of numbers or to assess peak counts.

Thirdly, we use imputing techniques to arrive at estimatesfor sites not covered by the counts. In recent years, this hasinvolved only a few sites, but when using the older data, theamount of imputed data increases and this might result in erro-neous trend estimates. Moreover, the calculation of standarderrors is theoretically questionable because we are not dealingwith a sample of sites but with more or less complete counts.The results from a recent experiment with “artificial” gaps in acomplete data set have shown that trend estimates might befairly reliable even with a high proportion of imputed data

468

Waterbirds around the world

Fig. 8. The availability of count data for freshwater monitoring sites in

The Netherlands since 1975/76. Coverage is expressed as a percentage

of all possible main count unit/month combinations. Column C indi-

cates complete coverage of all sites.

(Soldaat et al. 2004). The results of this study are now beingused to fine-tune the imputing technique.

Data for policy and managementThe need for monitoring data has increased substantially in thelast decade. While the first waterbird counts focussed on estimatesof total population size and identification of key sites, the moni-toring of individual key sites has become increasingly importantand is an obligation in the implementation of the EU Birds andHabitats Directives. As a consequence, periodical assessment ofthe conservation status of species is required, and managementplans have to be written to formulate conservation objectives.

Developing alert limits will be an essential tool to judge theobserved trends and to point at detrimental or negative impacts atthe designated sites (de Nobel et al. 2002, Austin et al. 2003).

Furthermore, in the case of negative trends it will be impor-tant to separate between local (site) trends, national trends andinternational (flyway) trends. Comparing site trends with inter-national trends will help policy makers and site managers tomake the right decisions regarding the underlying causes. If thetrends at a particular site seem to be caused by local factors, therelationship between the waterbirds and their food supply needsto be investigated. The numbers of waterbirds are often regu-lated by their food supply or their possibilities to exploit this

Fig. 9. The proportion of decreasing, increasing, stable and fluctuating trends in non-breeding populations of waterbird species listed in the EU Birds

Directive, for each of the relevant Special Protection Areas in The Netherlands (SPAs that only support terrestrial species are not taken into account).

The size of the circles represents the number of species per site.

469

Waterbirds around the world

supply (carrying capacity theory). As waterbirds can be readilydivided into guilds according to their food preferences (herbi-vores, fish-eating birds, etc.), and as long time-series of countdata exist, re-constructions of the development in numbers(monitoring) will be of great help in unravelling the causes ofany unfavourable development. The same relationships willhelp to formulate effective measures to restore a favourableconservation status.

ACKNOWLEDGEMENTSFirst of all, we would like to thank all observers participating inthe waterbird monitoring project in The Netherlands. The water-bird monitoring scheme is financially supported by the Ministryof Agriculture, Nature and Food Quality, the Ministry ofTransport, Public Works and Water Management, and BirdLifeThe Netherlands (Vogelbescherming Nederland). JeroenNienhuis and Erik van Winden (SOVON) transformed the firstdrafts of the figures into more presentable versions.

REFERENCESAustin, G.E., Armitage, M.J.S., Atkinson, P.W., Burton,

N.H.K., Leech, D.I., Marshall, A.R., Mellan, H.J.,Musgrove, A.J., Pollit, M. & Rehfisch, M.M. 2003.WeBS Alerts 1999/2000: Changes in numbers ofwintering birds in the United Kingdom, its constituentcountries, Special Protection Areas (SPAs) and Sites ofSpecial Scientific Interest (SSSIs). BTO Research ReportNo. 306. British Trust for Ornithology, Thetford, UK.

Bell, M.C. 1995. UINDEX4. A computer programme for esti-mating population index numbers by the Underhillmethod. The Wildfowl & Wetlands Trust, Slimbridge,UK.

Delany, S., Reyes, C., Hubert, E., Pihl, S., Rees, E., Haanstra,L. & van Strien, A. 1999. Results from the InternationalWaterbird Census in the Western Palearctic andSouthwest Asia, 1995 and 1996. Wetlands InternationalPublication No. 54, Wageningen, The Netherlands.

de Nobel, P., van Turnhout, C., van der Winden, J. &Foppen, R. 2002. An alert system for bird population

changes on a national level and for EU Bird Directivemonitoring: a Dutch approach. SOVON ResearchReport 2002/04. SOVON Dutch Centre for FieldOrnithology, Beek-Ubbergen, The Netherlands.

Ebbinge, B.S. 1992. Regulation of numbers of Dark-belliedBrent Geese on spring staging areas. Ardea 80: 203-228.

Günther, K. 2003. Rastvogel Monitoring im Schleswig-Holsteinischen Wattenmeer 2001-2002. WWF, Husum,Germany.

Koffijberg, K., van Roomen, M., Berrevoets, C. &Noordhuis, R. 2000. Tellen van watervogels inNederland: verdere ontwikkelingen en integratie vanaf2000. SOVON-Onderzoeksrapport 2000/05. SOVONVogelonderzoek Nederland, Beek-Ubbergen, TheNetherlands.

Noordhuis, R. & Koffijberg, K. 2004. Watervogels als indica-toren: presentatie van trends in relatie tot beleidsdoel-stellingen. RIZA-rapport 2004.003/SOVON-onderzoeksrapport 2004/01. RIZA, Lelystad, TheNetherlands.

Rösner, H.-U. 1994. Population indices for migratory birds inthe Schleswig-Holstein Wadden Sea from 1987 to 1993.Ophelia, Supplement 6: 171-186.

Soldaat, L., van Winden, E., van Turnhout, C., Berrevoets,C., van Roomen, M. & van Strien, A. 2004. Deberekening van indexen en trends bij het watervo-gelmeetnet. SOVON-onderzoeksrapport 2004/02.Centraal bureau voor de Statistiek, Voorburg/Heerlen,The Netherlands.

Underhill, L.G. & Prys-Jones, R.P. 1994. Index numbers forwaterbird populations. (I) Review and methodology.Journal of Applied Ecology 31: 463-480.

van Roomen, M., Boele, A., van der Weide, M.J.M., vanWinden, E.A.J. & Zoetebier, D. 2000. Belangrijkevogelgebieden in Nederland, 1993-97. Actueel overzichtvan Europese vogelwaarden in aangewezen en aan tewijzen speciale beschermingszones en andere belan-grijke gebieden. SOVON-informatierapport 2000/01.SOVON Vogelonderzoek Nederland, Beek-Ubbergen,The Netherlands.

van Roomen, M., van Turnhout, C., Nienhuis, J., Willems, F.& van Winden, E. 2002. Monitoring van watervogelsals niet-broedvogel in de Nederlandse Waddenzee: eval-uatie huidige opzet en voorstellen voor de toekomst.SOVON-onderzoeksrapport 2002/01. SOVONVogelonderzoek Nederland, Beek-Ubbergen, TheNetherlands.

van Roomen, M., van Winden, E., Koffijberg, K., Boele, A.,Hustings, F., Kleefstra, R., Schoppers, J., vanTurnhout, C., SOVON Ganzen- en zwanenwerkgroep& Soldaat, L. 2004. Watervogels in Nederland in2002/2003. SOVON-monitoringrapport 2004/02, RIZA-rapport BM04/09, SOVON Vogelonderzoek Nederland,Beek-Ubbergen, The Netherlands.

van Strien, A. 2005. Landelijke natuurmeetnetten van het NEMin 2004. Centraal Bureau voor de Statistiek,Voorburg/Heerlen, The Netherlands.

Visser, H. 2004. Estimation and detection of flexible trends.Atmospheric Environment 38: 4135-4145.

470

Waterbirds around the world

Fig. 10. Trends in the numbers of Tufted Duck Aythya fuligula and levels

of eutrophication (expressed as mg/l of phosphorus) at Lake

Markermeer, The Netherlands.