Agreement on the Conservation of
African-Eurasian Migratory Waterbirds
Doc: AEWA/MOP 4.31
Agenda item: 27 c.
Date: 25 August 2008
4th Session of the Meeting of the Parties
15 – 19 September 2008, Antananarivo, Madagascar
“Flyway Conservation at Work – Review of the Past, Vision for the Future"
DRAFT INTERNATIONAL SINGLE SPECIES ACTION PLAN FOR THE CONSERVATION OF THE BLACK-TAILED GODWIT LIMOSA LIMOSA
This International Single Species Action Plan for the Conservation of the Black-tailed Godwit (Limosa limosa) is based on the EU Black-tailed Godwit Species Management Plan and will supersede it upon approval by the Meeting of the Parties. As such the AEWA SSAP represents and extension of the EU Management Plan to a full flyway scale. This SSAP was initiated in 2006 and was commissioned to Orbicon – the same consultancy which compiled the EU plan. It was compiled by Flemming Pagh Jensen (Orbicon), Arnaud Béchet (Tour du Valat) and Eddy Wimenga (Altenburg & Wymenga). Drafts of the plan went through rigorous consultations with experts and the AEWA Technical Committee followed by official consultation with governmental officials of the range states of the species. The draft plan was endorsed by the Standing Committee at its 5th meeting in June 2008 for submission ot MOP4.
The Action Plan follows the format for Single Species Action Plans approved by the AEWA 2nd Meeting of the Parties in September 2002.
Action requested from the Meeting of the Parties
The Meeting of the Parties is invited to review this SSAP and adopt it for further imlpementation
Agreement on the Conservation of
African-Eurasian Migratory Waterbirds (AEWA)
International Single Species Action Plan for the
Conservation of the Black-tailed Godwit
Limosa l. limosa & L. l. islandica
Prepared with funding from Vogelbescherming Nederland (BirdLife Netherlands)
This AEWA SSAP is based on the EU Management Plan for the Black-tailed Godwit of 2007, which was updated with newly available information and extended to flyway scale
Compiled by Flemming Pagh Jensen1, Arnaud Béchet2 and Eddy Wymenga3
Orbicon, Ringstedvej 20, DK-4000 Roskilde, Denmark
2 Tour du Valat, F-132000 Le Sambuc, France
3 Altenburg & Wymenga (A&W), P.O. Box 32, NL-9269 ZR Veenwouden, The Netherlands
Milestones in the production of the Plan:
Workshop: 1st October 2007 in La Rochelle, France (organised by International Wader Study group)
First draft: December 2007.
Second draft: May 2008
Final version: August 2008
This International Single Species Action Plan requires implementation in the following countries regularly supporting the Western Palearctic population of L. l. limosa and the population of L. l. islandica of the Black-tailed Godwit: Albania, Algeria, Austria, Azerbaijan, Belarus, Belgium, Bulgaria, Burkina Faso, Cameroon, Chad, Croatia, Czech Republic, Denmark, Egypt, Estonia, Ethiopia, Eritrea, Finland, France, Gambia, Germany, Ghana, Greece, Guinea Bissau, Guinea Conakry, Hungary, Iceland, Iran, Republic of Ireland, Israel, Italy, Kazakhstan, Kenya, Latvia, Libya, Lithuania, Mali, Mauritania, Montenegro, Morocco, Netherlands, Niger, Nigeria, Norway, Oman, Poland, Portugal, Romania, Russia, Senegal, Serbia, Sierra Leone, Slovakia, Spain, Sudan, Sweden, Tunisia, Turkey, Ukraine, United Kingdom, United Arab Emirates and Yemen.
This International Single Species Action Plan should be reviewed and updated every 10 years (first revision in 2018).
We would like to thank the following for providing invaluable help and assistance to the preparation of this action plan: José Alves, Nicola J. Crockford, Gerrit Gerritsen, Jennifer A. Gill, Tómas G. Gunnarsson, Guillaume Gelinaud, Jos Hooijmeijer, Hermann Hötker, Rowena H.W. Langston, Jan van der Kamp, Ernst Oosterveld, Theunis Piersma, Pavel Pinchuk, Wolf Teunissen, Sami Timonen, Patrick Triplet, Bertrand Trolliet, Nuno Cidraes Vieira and Leo Zwarts.
Recommended citation: Jensen, F.P., Béchet, A. & Wymenga E. (Compilers) 2008. International Single Species Action Plan for the Conservation of Black-tailed Godwit Limosa limosa. AEWA Technical Series No. XX. Bonn, Germany.
1. Biological assessment……………………………………………………...…………………..7
2. Available key knowledge……………………….…………………..…...……………………16
3.1. Factors, which directly affect population level……………………………….20
3.2. Factors, which indirectly affect population level……………………………..22
4. Policies and legislation relevant for management……………………………… ...………..28
4.1. International conservation and legal status of species…….…… …………….28
4.2. National conservation and legal status …….……………….… …….………..28
5. Framework for action…………………………………………………… ………..…….……29
5.1. Priority statement………………….……………………………………… ….29
5.2. Purpose of the plan…………….……………….…………………………. .…31
6. Activities by country/region……………….…………………………….……………… …...33
7. References …………………………………………………………...………………… ……. 43
This International Single Species Action Plan for the Conservation of the Black-tailed Godwit Western Palearctic breeding population Limosa l. limosa and the L. l. islandica of Iceland was commissioned by UNEP/AEWA Secretariat and financially supported by Vogelbescherming Nederland (BirdLife Partner The Netherlands). It has been compiled by a team consisting of Flemming P Jensen of Orbicon (Denmark), Arnaud Béchet of Tour du Valat (France) and Eddy Wymenga of Altenburg & Wymenga (The Netherlands).
The Action Plan follows the format for Single Species Action Plans approved by the AEWA 2nd Meeting of Parties in September 2002.
This AEWA Action Plan builds on the European Commission Management Plan for the Black-tailed Godwit which has been adopted recently by the EU Member States. However, the AEWA plan is upgraded to a flyway level and also includes new data and information which has become available since the completion of the EU plan. This includes new estimates of population size, information on the migration ecology and trend in some countries. The AEWA plan also takes into account the recent (July 2008) temporary five-year suspension of hunting on the Black-tailed Godwit in France.
The Black-tailed Godwit has a widespread but disjunct distribution in the Western Palearctic. Two subspecies occur in this area; islandica which breeds mainly in Iceland and limosa with a main breeding range from The Netherlands to Russia. The populations of both subspecies are migratory and have separated migration systems. In the European part of the migration system subspecies can mix. The species increased during the 20th century throughout Western Palearctic but while the islandica population has continued to increase in numbers and expanded its breeding range, nominate limosa’s has shown range contraction and major declines in most key breeding areas during the last decades. Today the islandica population numbers c.25,000 pairs while the nominate population of the Western Palearctic totals c.110,000 pairs. About 50% of the nominate population breeds in The Netherlands. Due to the continuing decline of nominate Black-tailed Godwits, its status on the IUCN Red List of Threatened Species was changes in 2006 from “Least Concern” to “Near Threatened”.
The nominate form breeds almost exclusively in man-made habitats in particular semi-natural grassland and meadows. In The Netherlands and adjacent Germany and Belgium the majority breeds in intensively managed moist to wet grassland used for dairy farming. In Central and Eastern Europe the godwits mainly breed at flood-plain meadows and wet pastures near lakes and rivers with moderate cattle grazing and haymaking in the late summer. The islandica subspecies breeds in lowland areas, primarily on coastal marshes and dwarf-birch bogs.
Throughout its range nominate godwits face loss and degradation of breeding habitat mainly due to urbanisation and infrastructure development, conversion of grassland into arable land, loss of openness and increasing disturbance. This has lead to a widespread decline of the species, although the development of the eastern populations is generally poorly known. In The Netherlands and Germany, which is the core breeding area of western nominate godwits, the population is currently declining by 5% annually. Intensification of grassland management, landscape changes and increased predation has lead to a very low reproduction in this area and is believed to be that main cause of this decline. In Central and Eastern Europe land use changes, in particular large scale abandonment of farming activities is a serious threat to the godwits when it leads to overgrowth of breeding sites.
During migration and in the winter quarters Black-tailed Godwits have traditionally largely been restricted to estuaries and large inland wetlands. In recent decades, rice fields have become increasingly important during winter in West Africa and in Spain and Portugal during spring migration. This, combined with progressively earlier arrival of godwits to West Africa, due to failed breeding, has created conflicts with farmers, and locally resulted in a loss of 5-6% of adult birds due to hunting.
The goal of this plan is to restore 'Least Concern' status of the Black-tailed Godwit on the global IUCN Global Red List of Threatened Species. The short term objective is therefore to halt the current decline and contraction of distribution while the long-term objective is to restore all Western Palaearctic populations to a favourable conservation status. In addition, the plan aims at maintaining the favourable status of the islandica population.
To achieve the goal and objectives the plan aims to addressing the most urgent issues in a specific, measurable, agreed, realistic and time-bound process. Since many results and proposed conservation actions apply to more than one country, the countries have been grouped into four categories, combining status and threats of the godwit and the political situation of each country: (1) Non-EU Member states, (2) EU Member States with the exception of The Netherlands and neighbouring areas in Germany and Belgium, (3) The Netherlands and neighbouring areas in Germany and Belgium and (4) Countries within the migratory and wintering area of the flyway, consisting of EU Member States, non-EU countries as well as countries on the Middle East and in Africa.
For these four regions the conservation priorities are:
Prevent further breeding habitat loss and degradation, and restore breeding habitats
Reduce chick-mortality and nest destruction where Black-tailed Godwit breeds in intensively managed farmland
Provide adequate support for and protection and management of important Black-tailed Godwit staging and wintering areas
Ensure legal protection of Black-tailed Godwit in all range states
Eliminate illegal hunting
Improve the understanding of the distribution and trend of the eastern breeding populations
Improve the understanding of the migration and wintering areas of the eastern populations
1. Biological assessment
The Black-tailed Godwit is a large wader species which has a widespread but disjunctive distribution in the Palearctic, extending from Iceland across northern Europe to western Siberia. Two subspecies occur in Europe.
The main breeding range of the nominate form L. l. limosa ranges from The Netherlands to Germany, Poland, Belarus, Ukraine and western Russia. Small populations occur in other European countries. Historical data suggest that this population increased during the 20th century in particular in The Netherlands and NW Germany, to reach a maximum population size between c.1940 and 1960. Throughout much of its range, this population has been in decline since then. Today about half of the population breeds in The Netherlands with other significant populations occurring in Russia, Ukraine, Poland and Belarus.
The subspecies L. l. islandica breeds mainly in Iceland, and increased its breeding range and population over the last 100 years, especially between 1960 and 1990. This increase still goes on.
The Black-tailed Godwit belongs to the Scolopacidae family (sandpipers and allies), the subfamily Tringinae (Godwits, Curlews and other sandpipers). In the Western Palearctic two subspecies occur: the nominate race L.l. limosa and L.l. islandica. The breeding populations east of the Yenisei River are separated as subspecies L. l. melanuroides; these are not treated in this action plan.
In the first half of the 20th century, the nominate race has been adapting to man-induced changes of the landscape and has spread into agricultural habitats over much of West and Central Europe (Glutz von Blotzheim et al. 1977, Cramp & Simmons 1983). In recent decades, the North-West European limosa-population has declined in many areas, mainly because of intensification of grassland management and loss of breeding habitat. Between 1970 and 2000 declines occurred in several countries, collectively holding up to 85% of the European population, including The Netherlands, Germany, Poland and Russia (BirdLife International 2004, Hötker et al. 2007).
In The Netherlands, which forms the stronghold of the European population, the population grew between 1920 and 1960 to reach a maximum of 125-135,000 breeding pairs (Mulder 1972). In the 1980s the population was estimated at 85-100,000 pairs (van Dijk 1983, Piersma 1986). The most rapid declines occurred in the 1970s and again since the mid-1990s (Altenburg & Wymenga 2000, Teunissen et al. 2004). The present annual decline is estimated at c. 5% on a national scale (Teunissen & Soldaat 2006), resulting in a current population of c. 55,000 pairs (data SOVON, W. Teunissen pers. com.). This number may even be lower, as in the province of Fryslân, where half of the Dutch population breeds, the annual decline during 2000-2005 was 9% (Oosterveld 2006). Locally there are populations in The Netherlands which increase or are stable (Oosterveld 2006).
In Russia the species was considered relatively common in some regions until the mid-1980s but from then onwards it has been declining in many areas (Sukhanova 2008). In eastern and south-eastern Ukraine numbers were low around 1880-1890, but increased since then to reach high numbers in the 1930-1940s. In the 1980s-1990s a slow decrease started and since 2000 the species declined rapidly (Banik & Vergeles 2003). In the western Ukraine, including the Desna and Dnipro river basins, the same trends occurred, with breeding numbers halved in some areas over 10 years (Gorban pers. com, Voblenko pers. com).
The wintering population in the western Sahel decreased with 2/3 during the last 20 years, paralleling the development of the western breeding population. Numbers in the Inner Niger Delta (Mali) and the Lake Chad basin have remained more or less stable (Zwarts et al. in press.). Very little information is available regarding the development of the populations wintering further east, such as in the Sudd (Sudan) and in East Africa.
The population of L. l. islandica, which basically consists of the breeding population in Iceland, has been increasing from an estimated 2,000-3,000 individuals in 1900 (Gunnarsson et al. 2005a) to a mid-winter population of 50,000-75,000 birds (Gill et al. 2007).
All populations from the Western Palearctic are migratory. The Icelandic and nominate godwits have clearly separated migration systems.
The majority of adult nominate godwits in West and Central Europe leaves the breeding grounds in late June - July. The populations further east depart later, sometimes as late as September (Dementiev et al.1969). In areas with high densities, godwits roost communally after (and before) the breeding season (Piersma 1983, Gerritsen 1990). Following fattening for about two weeks most adult godwits from The Netherlands fly non-stop to the wintering areas in West Africa (Zwarts et al. in press.). The migration of juveniles lasts from July to September. The majority of juveniles are also believed to migrate non-stop to the winter quarters in West Africa, although a portion of them use areas in the south-west of France and the Iberian peninsula as stop-over sites; from an analysis of the EURING-data it appears that 85% of godwits recovered in France in July are juveniles (Zwarts et al. in press).
The main wintering areas of the north-west European limosa-populations are situated in Senegal (Casamance) and Guinea Bissau and to a lesser extent in the large Sahelian floodplains: the Senegal River Delta and the Inner Niger Delta. The godwits arrive in West Africa from late June to September where they mainly congregate in the rice field zone. Initially the godwits are found mainly on recently ploughed land and just sown seedbeds and parcels (July-September). Later on the godwits use the rice zone more extensively whilst feeding on animal and vegetal matter. During the harvest period (November-December) they basically feed on rice grains (Tréca 1975, 1984, van der Kamp et al. 2008).
When the rice fields in South Senegal (Casamance) and Guinea Bissau are harvested and drying out, the godwits start to migrate north by the end of December. The bulk of the population, including wintering birds from the Senegal Delta, is believed to fly non-stop to the rice fields in Spain and Portugal (Kuijper et al. 2006, Sanchez-Guzman et al. 2007). Wetlands in Morocco were previously important stop-over sites for these godwits. In recent years the Moroccan wetlands have lost much of their significance, although 5,000 – 10,000 birds may still stop-over briefly in January-February (Green 2000, Kuijper et al. 2006).
Colour marking of individual birds has shown that the first limosa return to Spain and Portugal in December and numbers build up in January-February with some also reaching France (Hooijmeijer pers. com.). In March most limosa’s have left Spain and Portugal, and numbers subsequently increase in France and in particular in The Netherlands (Hooijmeijer pers. com.). Godwits arrive in The Netherlands from late February to March with 50% of the population normally present by mid-March (Wymenga 2005a).
The eastern populations (east of Germany) seem to have a more eastern migration route with ringing recoveries from Italy and Turkey and to winter mainly in the Inner Niger Delta, the Lake Chad Basin and possible the Sudd in southern Sudan and further south to Kenya. Some also winter in the Middle East. However, generally very little is known of the movements and winter quarters of these populations. Important staging grounds are found in Azerbaijan, Iran, Greece, Bulgaria (1,000-5,000 at Atanasovsko Lake, Kostadinova & Gramatikov 2007), Kazakhstan, Turkey (the Kizilirmak and the Ceyhan Delta (Doga Dernegi) and Tunisia. On spring migration large flocks have been recorded in March-April in wetlands of the East Mediterranean Basin (Turkey, Greece, Bulgaria, Cyprus), in the Middle East and around the Black Sea (cf. Kube et al. 1998). In Kazakhstan concentrations of 3,000-8,000 birds have been reported in 2005 and 2006 (BirdLife Int. 2007), and up to 2,000 in southern Belarus in the Pripyat floodplain in spring (P. Pinchuk in litt.). In Western Ukraine 600 – 1,100 have been observed on spring migration in the Tuiria, Stokhid floodplains and in the Ukrainian part of the Pripyat (Prypyat) floodplains (Gorban 1999, 2002). In southern and eastern Ukraine fewer godwits are recorded on spring migration. In autumn the numbers passing through Western Ukraine are usually lower than in spring with flock of up to 200 in Volyn and Lviv regions (Gorban 1999, 2002). In the Danube Delta flocks of up to 8,000 were recorded during autumn migration in the 1980s but in recent years numbers have decreased to 1,200 – 2,000 individuals, with the sharpest decline during the last five years (Zhmud pers. com). In the south-central part of Ukraine flocks of up to 500 are regularly observed in August-September in the Kinsburn Regional Landscape Park (Petrovich pers. com.) as wel as in the Karkinitska Bay, Crimea (Tarina pers. com.).
The majority of godwits wintering in the Inner Niger Delta in Mali probably originate from these eastern populations (Zwarts et al. in press.). Black-tailed Godwits first occupy the fringes of the delta but subsequently concentrate in the central delta when the flood recedes. In late January –February they feed mainly on small bivalves for pre-migratory fattening. Departure is mainly in the second decade of March (Wymenga et al. 2002, Zwarts et al. in press.).
Birds belonging to islandica initially migrate to England and France in September-October but by mid-winter most have left the UK, migrating to France and in particular Spain and Portugal (Gill et al. 2007, Triplet et al. 2007) with a few reaching Morocco (J. Alves pers. com.). Numbers of wintering islandica in Spain and Portugal have increased significantly since the 1960s, but fluctuate much between years (J. Alves pers. com.). During spring migration in March/April many islandica from Portugal and France, before continuing to Iceland, first move to The Netherlands (Gerritsen & Tijsen 2003) or eastern England where they forage primarily on grasslands (Gill et al. 2007).
In Portugal, both islandica and limosa are present in winter. Although it is not possible to differentiate the proportion of the two subspecies, it is likely that the majority (50-70%) in January is limosa. Counts from the Tejo Estuary between early December and late February show a steady increase in numbers until the first week of February, when numbers can peak at 80,000 birds, suggesting the return of birds from “wintering” grounds further south.
In general, long term data on survival and productivity are lacking, in particular for the eastern limosa-populations. The annual adult survival of islandica has been estimated at 87-94% (Gill et al. 2001a).
In a Dutch study from 1984-1987 the annual adult survival was 81.4%, with no significant survival difference between sexes (Groen & Hemerik 2002). A recent study of Roodbergen et al. (in press) suggests adult survival rates of western limosa and islandica of 0.97-0.98. Another Dutch study mentioned adult survival rates of 0.94 between 1994 and 2007 (Kentie et al. 2007). Recent colourring studies suggest annual survival rates of c. 81 – 96% (Both et al. 2006, J. Schröder in prep.), though national estimates from ring-recoveries suggest annual survival rates of c. 80% (van Noordwijk & Thomson 2008). Using EURING-data, completed with ringing data from the Dutch Centre for Avian Migration and Demography, Zwarts et al. (in press.) showed that the adult survival has increased over the past decennia. In The Netherlands chick survival decreased from 17-42% in the 1980s to 0-24% in 2003-2005 (Schekkerman et al. in press).
In Germany, 0.91 fledging per pair was recorded (Bairlein & Bergner 1995) while a Dutch study from 1984-1987 showed a productivity of 0.58-1.18 fledged chicks per pair – lowest in cold and wet springs – with decreasing net productivity in the course of the study (Groen & Hemerik 2002). In The Netherlands reproductive success has declined dramatically from c. 0.7 chicks per pair per year in the 1980s to 0.1–0.4 chicks per pair per year in 2003-2005 (Schekkerman & Beintema 2007, Schekkerman et al. in press). This is probably far below the threshold for a sustainable population.
Highly gregarious. Flock size varies with the highest concentrations occurring at roosts in early spring when tens of thousands are found together (Snow & Perrins 1998).
In dispersed colonies and sub-colonies. Age of first breeding normally two years or older. However, limosa may breed in the first year.
According to Snow & Perrins (1998) laying in the West and Central Europe is from early to mid-April (mean laying date of first egg in The Netherlands is around 15 April) while further north in Iceland laying begins in late May. The single brood is found on the ground in short or fairly short vegetation. If lost early in the season some limosa produce second clutch. Clutch size is normally 4 (Beintema 1991).
The incubation is 22-24 days.
Mainly invertebrates such as insects, annelids, earthworms and molluscs, small crustacean and arachnids (Snow & Perrins 1998). Populations wintering in West Africa mainly feed on plant material, in particular rice grains.
Chicks of the islandica and limosa population feed mostly on invertebrates gleaned from vegetation (Gunnarsson et al. 2006; Schekkerman & Beintema 2007).
After breeding Icelandic godwits move to moulting sites in the UK, in particular the Wash, Humber and Dee estuaries.
Continental godwits start moulting at least 2-3 primaries on breeding grounds, migrate south in suspended moult, and continue to moult in their African whereabouts (Timmerman 1985).
Originally mires, wet moor land, blanket bogs, flooded grasslands, river valley fens and marshy margins of lakes, damp grassy steppes and probably estuarine habitats. Some birds still breed in such habitats, especially in Iceland and eastern part of Western Palearctic.
The majority of the European population now breeds in open, secondary habitat: meadows, semi-natural grasslands and intensively managed grassland. In Central and Eastern Europe flood-plain meadows and wet pastures near lakes and rivers are key breeding habitats. Wet or moist grassland on clay, clay-on peat or peat soil is a feature of the lowlands that supports the majority of the breeding numbers in the countries surrounding the North Sea (cf. Wymenga et al. 2006). In The Netherlands and North West Germany the population reached its maximum in the 1960s and 1970s in open, moist to wet, rather extensively used grasslands1.
Mown grasslands are selected over grazed pasture. In intensively managed grasslands in The Netherlands nest site selection is positively influenced by increasing ground water level (Verhulst et al. 2007), but significant breeding densities are also possible at lower ground water levels (80-100 cm below surface level, Oosterveld (2006), depending on soil structure, spatial configuration of feeding and breeding habitats, and on grassland management. In agricultural grasslands, areas mown annually hold higher densities than areas with grazing only (e.g. in Denmark, Thorup 1998, in The Netherlands, Buker & Groen 1989, and in Sweden, Larsson 1976). In Hungary, Black-tailed Godwits prefer either extensive or intensive pastures depending on the biogeographical region considered (Baldi et al. 2005). In Dutch grasslands chicks strongly prefer tall but not dense grass (>15-20 cm), either not yet mown or re-grown after first cut (Schekkerman & Beintema 2007).
In Iceland islandica breeds in lowland areas, primarily on coastal marshes and dwarf-birch bogs (Gunnarsson et al. 2006). The expansion from south-west Iceland (around 1900) to the major basins in the north and west (1920s-1940s) and then the east and north-east of Iceland (1970s-1980s) was characterised by an increase in the proportion of dwarf-birch bog sites occupied (Gunnarsson et al. 2005a). The lowland areas of Iceland have seen widespread drainage of wetlands and increases in numbers of hayfields since the 1960s, and godwits are now frequently recorded feeding on hayfields during the summer season (Gunnarsson et al. 2005a).
The nominate race winter predominantly in open freshwater and brackish habitats south of the Sahara while islandica winters in estuarine habitats along the Atlantic coast from Britain south to Morocco (Beintema & Melter 1997). Most of these birds winter on "soft coasts", mainly estuaries and areas of inter-tidal mud, but substantial numbers of islandica winter on floodlands in Ireland (e.g. Delany 1996). These birds also feed in adjacent grassland as the tide limits the time they can feed on tidal mudflats and where prey are subject to strong seasonal depletion (Gill et al. 2007). Mudflats are a key staging habitat in Iceland when the birds arrive in spring, especially in cold years (Gunnarsson et al. 2005b). Some islandica also use the Iberian rice fields in December-February.
Post-harvest moist and flooded rice fields are important habitats for limosa in West Africa and in Portugal/Spain (Roux 1973, Tréca 1984, Altenburg et al. 1985, Bos et al. 2006). Godwits also winter in natural freshwater habitats, like the floodplains of the Senegal and Niger Rivers (Wymenga et al. 2002, Kuijper et al. 2006, see before). On the Iberian Peninsula large concentrations use the rice fields adjacent to the Tagus Estuary in Portugal, and in the Sado estuary (Kuijper et al. 2006, Sanchez-Guzman et al. 2007); the birds feed on rice grains left after the previous harvest. Despite an increasing preference for rice fields, intertidal feeding (especially on Scrobicularia plana and Nereis diversicolor) and use of salt-pans remain important at the Tagus site (Moreira 1994), but the latter feeding behaviour may well refer to islandica. In extremely wet winters, a significant proportion of the birds in Portugal feed on pasture land and stubbles, which are partially flooded (R. Rufino pers. com.). In Tunisia Black-tailed Godwits are found wintering in the salines of Thyna (south of Sfax), a very salt environment (Kuijper et al. 2006, Azafzaf & Feltrup-Azafzaf 2007).