Wintering and spring staging Dunlin Calidris alpina in the south Baltic area.

by Christer Persson

Christer Persson, Ljungsätersvägen 43, S236 41
Höllviken, Sweden. E-mail cp.hollviken'at'

Straight text version for printing: text version.

Abstract Dunlin have wintered in the Sound and the south Baltic area since 1971 - 72. Juveniles connected with the wintering population have been recorded from c20 October, but not earlier, adults no sooner than 15 November. They must return from the Waddensea or some area between the Waddensea and the Sound by this time. After New Year the age ratio 2c/3c+ of stationary flocks is appr. 50 : 50. The maximum wintering population of the south Baltic - Belt Sea - Kattegat area was reached by the mid 90's with 10,000 birds. In spite of losses due to frost-bite, starvation and predation, the birds are slow to leave the area during periods of ice; they stick to areas where currents and/or salinity keep the waters open. The rate of nocturnal activity increases with sub-zero temperatures, first as "supplementary" feeding, but protection from frostbite may be of equal importance; with strong wind and ongoing formation of ice the latter prevails.
In spring, the same area may keep some 20,000 staging Dunlin. These birds partly come from the wintering population, but new birds arrive from late February. The body moult starts in early April, often with one single scapular at a time, by mid-May the first birds have complete breeding plumage.


1. Introduction: prehistory
2. Results
2.1. Winter
2.2. Spring conditions, staging
3. Discussion
3.1. The wintering population

1. Introduction: prehistory

Dunlin have been known to occur on the Falsterbo peninsula, S. Sweden in December and January at least since the late 50's (Roos 1962), but no observations were made in the critical month: February. Confirmed wintering was reported from Skälderviken (Fig. 1) in north-western Scania in the winter 1971 - 72, and the county as a whole kept a few hundred wintering Dunlin between 1974 and 1978, with a maximum of 250 in 1974 - 75 (SkOF. Fåglar i Skåne 1980). After a couple of cold winters about 1980 numbers began to rise again, and similar increases were noted in Denmark and Germany in the same period (Meltofte 1993, Müller 1985). In January 1992 a record 78,000 Dunlin were recorded in the Schleswig-Holstein Waddenseea (H.-U. Rösner, quoted in Meltofte et al. 1994). Origin and phenology of the Baltic wintering population remain obscure; its arrival in autumn is masked by a much larger number of transmigrants, its presence in spring by an additional arrival of staging birds.

This section opens up with a census of Dunlin wintering in Kattegat, the Belt Sea, the Sound and the German parts of the Baltic by the early 90's. Next, using phenology and recoveries of Dunlin at some time or other visiting the Sound area, an attempt is made to outline the time schedule and the areas of origin of the wintering population. Finally a few adaptations of wintering Dunlin during cold spells are described.

2. Results

2.1. Winter

2.1.1. Development of wintering in the Baltic, the Belt Sea, the Sound and southern Kattegat. After a couple of cold winters in the Sound area about 1980 the number of wintering Dunlin rose again to new record levels. Three separate centres could be discerned: one at Skälderviken to the north, one at Barsebäck/Landskrona in the central part of the Sound and one at Falsterbo to the south. The development from the winter 1982 / 83 is described in Table I. These figures are a little on the low side but give an adequate picture of the general trend.

Table I. Largest Dunlin flock on the Falsterbo peninsula in December and consecutive wintering in Scania. (Fåglar i Skåne, 1982 - 1994)

WinterFalsterbo pen., DecemberScania: est. wintering


Denmark: Glutz et al. 1975 state: "In Denmark numbers decrease to a few hundred in January". Less than two decades later Meltofte 1993 estimates that 2 - 3,000 Dunlin are present by midwinter in the central and eastern parts of the country, but he adds that numbers depend on the severity of winter weather. Even this figure may underestimate the true extent of "normal" wintering in recent years, simple addition gives a winter population of at least 6.000 individuals in the early 90's (DOF. Fugle på Sjælland 1986-1993, Rapportgruppen ved DOF / Århus amt, K. Biledgaard, R. Christensen, K. Due-Johansen, H.-E. Jørgensen in litt.), and as far as I have been able to find there are no published observations from several potential wintering-grounds between Fyn and Jutland.

Germany, Poland: According to Glutz et al. 1975 the Baltic proper is evacuated in ice-winters. On the other hand Müller 1985 reports between 1000 and 2000 wintering Dunlins from the early 80's in the Bock area, south of Hiddensee (Rügen). At Bottsand, close to the mouth of the Bay of Kiel, the number of wintering Dunlin has doubled from c. 350 in the season 73 - 74 to a maximum of 730 in 94 - 95, and the area remained occupied even in the ice winters of 78 - 79 and 84 - 85 (but was evacuated in January 96), although there was notable mortality (H. Behmann in litt.). In the German part of the Baltic appr. 1,000 Dunlin were present in January 1993 and 600 in January 1994 (B. Struwe-Juhl in litt.). In W. Poland (Szczecin area) the species is seldom seen in winter (W. Meissner in litt.), and in the Bay of Gdansk area only single specimens (at most 20) have been noted (J. Gromadzka in litt.). I have been unable to obtain information from the Kaliningrad enclave and Lithuania. The extent of wintering in the three countries by 1995 is summarized in Fig. 1; the centre of gravity is situated in southern Kattegat. In a mild winter with little or no ice, the overall area has kept some 10,000 wintering Dunlin, and even this estimate may be conservative.

2.1.2. Establishment of the wintering population. Up till the end of 2000 the late autumn and winter ringings in the Sound area have yielded 16 local recaptures or recoveries of ringed birds (8 ringed as 2c+ / 3c+, 8 as 1c / 2c) between seasons and 3 local recaptures of some length within the same winter. The time-span of these recoveries is shown in Fig. 2. So far no adult bird controlled between 16.11 and 28.2 was ringed outside the same interval, and juveniles ringed in late October were controlled or found dead between 16.11 and 28.2 in later years.

One bird falls just outside the limits set for "wintering" in two consecutive seasons. It was ringed as 1c at Klagshamn 25.10.93 and recaptured at Skanör on 7.11.94, in complete winter plumage including a few distinct "adult buff" coverts (Gromadzka 1986, 1989). (This particular bird was buff, not chestnut). I believe, that it belonged to the population staging at Skanör in late autumn, an adult in a flock of mainly juveniles, and that it was due for departure within a few days. This shows that adult birds of Siberian origin, arriving late in autumn after having completed their moult on or near breeding-grounds, may be faithful to routes as well as winter areas; wind-drift or induction need not be invoked to explain their occurrence (cf. Stanley & Minton 1972). There is another interesting double recovery, that illustrates the loop flown by such birds; the bird in question had old remiges when ringed in July, on its way to moulting-grounds, and must have had fresh remiges (or growing outer primaries) when retrapped the second time in Germany; in one season it moulted on W European moulting-grounds or in the Mediterranean, in another on or near breeding-grounds:

  • 3c+ Skanör, SWEDEN 27.7.97 - V Nizovya Tiligulskogo Limana (46.40 N, 31.09 E), Odessa, UKRAINE 5.5.00 - V Langenwerder (54.02 N, 11.30 E), Wismar, GERMANY 3.10.00.

Still I believe that many birds of Siberian origin, after migrating by way of the Baltic in their first autumn, take a short cut as adults and migrate on a more easterly route to the Mediterranean or the Persian Gulf, most of them do not occur in the Baltic in later years, and this is the reason why there are no controls in subsequent years of the staging population in October - November. These birds disappear completely after their first calendar year, so far there is no sign of them in the summer moult migration of the Baltic.

As for the rest, the birds left behind when the staging birds disappear by mid-November, one particular point should be emphasized again: juvenile Dunlins which have spent their first winter in S Scania have wintered again in the Sound area as adults. The same site tenacity applies for Bottsand in the south Baltic (H. Behmann in litt.) as well. One possibility is for adults to arrive from east in late autumn after having completed their moult on breeding grounds, and, as a matter of fact, a handful of adults do occur among juveniles till 25 October in the Sound area and at Langenwerder (Brenning 1987). There comes a period of between two weeks and a month in each autumn, however, when adults are practically absent, and the amounts present by midwinter can be found nowhere throughout the autumn. All potential sites have been sifted, night and day; they are occupied almost exclusively by immature birds. Then, suddenly, adult birds occur again, like a bolt from the blue. They are ostentatious, keep close together, are easy to spot and recognize from the first minute of their arrival, and the remaining juveniles join them. An interesting point is, that their arrival coincides with the final departure of staging birds at Skanör. (In 1996 the first colour-ringed birds (two) were spotted on the isle of Måkläppen/Falsterbo on 16 November, in 1997 the first (one) near Barsebäck on 14 November, both observations by Peter Olsson).

On the other hand, the full Scanian pattern is not automatically reproduced at Bottsand. Here, birds belonging to the wintering flock have been recorded at least between 20 August and 23 May, i.e. for nine months, and birds ringed during winter months have been retrapped at the mouth of Vistula/Poland on 30 July and at Ottenby/Sweden on 13 August (the same bird again recaptured at Bottsand on 23 August) and 15 August. But important: Henning Behmann still assumes, that the intermediate visitors continue to moulting-grounds in the Waddensea. These results are based on 77(!) Dunlin ringed between 15 November and 15 March in the early 90's, in addition there are numerous sightings in winter of birds colour-ringed in autumn. Bottsand has a very strategic position, based on one tenth of the field-work I was doing by the same time in the Sound area, Behmann received five times as much information! It is estimated that 80-90 % of the wintering flock has consisted of 1y birds up till the season 1994-95. In December 1995 a much higher proportion of all Dunlin present were adults, however, and many birds ringed in the winter 1994-95 were present. (H. Behmann in litt.).

2.1.3. Proximate origin of the wintering flock. There are five recoveries of winter birds, showing that they may take part in the summer moult migration in the Baltic:

  • 2c+ Skanör 25.2.93 - V Säppi (61° 29' N, 21° 21' E), Turku-Pori, FINLAND 13.7.93. 835 km NNE
  • 1c Bjärred (55° 44' N, 13° 00' E), SWEDEN 24.11.91 - V Kosakowo (54° 38' N, 18° 31' E), mouth of Rewa, POLAND 15.7.93. 371 km ESE
  • V 2c+ Barsebäck 29.12.95 - • 3c+ Ottenby (56°12' N, 16° 24' E), SWEDEN 18.8.92. 223 km W
  • V 2c+ Barsebäck 30.12.95 - • 2c+ Swibna (54° 22' N, 18° 56' E), mouth of Vistula, POLAND 19.7.95. 414 km WNW.
  • V 2c+ Barsebäck 29.12.96 - • 2c Ottenby, SWEDEN 17.8.92. 223 km W.

The early dates in Finland and Poland suggest that the birds were females migrating to moulting-grounds and finishing their moult in late August or early September - but where did they spend the rest of the summer and autumn? The Pscore of moulting birds in mid-July seldom exceeds 20 (Gromadzka 1986, Holmgren et al. 1993b; also see Ginn & Melville 1983) and many birds migrate with old remiges, which means that there is 1 - 2 months to account for before the winter plumage is complete. Adult summer migrants don't stay for long at south Baltic stopover sites (Dierschke 1996 has the correct evaluation here; stopover times of 1 - 4 days reported by Brenning 1987, Gromadzka 1986 and Holmgren et al. 1993a probably were caused by minor fitness losses in walk-in traps, cf Persson 2006), and there is no permanent moulting flock in SW Scania or E Denmark in summer, nor does any German site serve as a moulting area (H. Behmann in litt.). If the birds involved are faithful to their wintering grounds, this in turn brings forth the possibility of moult in the Waddensea (or Kattegat?) and consecutive dispersion, involving return to the east for wintering. The development of adult ratios from the first half of September to February is shown in Fig. 3. In 122 birds caught at Barsebäck between Christmas and New Year 1995 and 96 the age ratio was exactly 50 : 50.

2.1.4. Winter ecology. In mild winters the wintering Dunlin in the Sound area seem to lead a very unproblematic life: there is little predation to be seen and no tides necessitating daily movements. Under such conditions the birds seem to sleep themselves through the days; they are excellent energy conservers.

This uneventful state is interrupted when there is frost; the flocks then immediately occupy the best feeding-grounds of their overall wintering area and start to exploit their resources, day and night. The increased activity continues as long as the cold weather prevails. If the weather ameliorates, the old habit of sleeping through the days is quickly resumed, on the other hand: if an ice-sheet covers coastal waters, a real struggle for survival begins. Suddenly all Dunlin seem to disappear - but even with the whole of the littoral zone covered by ice and temperatures approaching -15° C in the night-time they obviously do not abandon the area. Instead they hold on - with losses due to predation and frost-bite - to shores or islets in the Belts and the Sound (Saltholm), where currents and salinity keep the waters open. Here they are difficult to spot and easily overlooked. Even in the Bay of Kiel (H. Behmann in litt.) and the Bay of Wismar (V. Dierschke in litt.), small flocks have stood their ground in ice-winters. Sewage outlets and, in recent years, nuclear plants have offered additional resorts. The number of Dunlin observed decreases, many birds are predated or freeze to death, but the quick return of Dunlin flocks to wind flats with rising temperatures demonstrates, that the survivors were never far away.

The hard core of the Scanian winter-flock is the group of adults arriving in November-December; these experienced birds are capable of holding out successfully under very severe weather conditions. Even after the ice has settled, water-levels will continue to oscillate with wind and air-pressure; sometimes, with low levels, the birds may find protection and insulation under the ice at the edges of ice-sheets. As long as the water is open but freezing, most birds will stick to belly-deep water (protecting the legs from frost) overnight. If the water-repellent ability of the feather plumage is impaired (and it is, a parallel to the decay of the feathering of Coots Fulica atra when temperatures sink below -10° C), this adaptation may be counterproductive; at dawn single birds find themselves weighed down by ice-lumps the size of a walnut attached to tail-feathers. Such birds are quickly predated, particularly by large Laridae.

The sudden death of Nereis diversicolor when water temperatures fall below c. +0.5_ C in waters of average Baltic salinity (V. Dierschke in litt.) gives a welcome respite; at Barsebäck tens of thousands of dead Nereis kept 200 Dunlin alive between 27 and 31.12.95. This resource will be accessible for approximately one week, and if the birds move north with the ice, they will again meet with dying Nereis in areas of higher salinity. With sub-zero temperatures and the emergence of this very timely source of prey, the Dunlin flocks will feed around the clock. Two birds weighing 42 and 43.5 g when ringed 27.12.95 died within a few days, and there was heavy predation from female Sparrowhawks Accipiter nisus - the Dunlin so stiffened or numbed by the cold that they did not bother to alight when attacked. The survivors, with mean weights 47 g on 27 and 28 December and almost 49 g on 29 - 31 December left in fairly good condition (flying by the wind!) on January 1st. (Ulf Lundwall, Peter Olsson). It should be added that the effects of salinity are notable even within the Sound; the first area to be evacuated when coastal waters get completely covered by ice is the Falsterbo peninsula, while small Dunlin flocks are known to have held on to Skälderviken at the northern "mouth" even in sustained ice winters - here there will always be some open water. The ultimate retreats Torekov at Skälderviken/Scania and Fyns hoved north of Kerteminde/Fyn both are known as regular wintering sites for Purple Sandpiper Calidris maritima as well. Four recoveries from the winter 95 - 96 illustrate the range of Dunlin trying to avoid the most severe conditions; they stick to the overall Baltic-Sound-Belt area:

  • 1c Foteviken (55° 27' N, 12° 58' E) 27.10.91 - x Bröndbystrand (55° 37' N, 12° 25' E), Sealand, DENMARK 14.1.96. 39 km WNW.
  • 1c Foteviken 17.10.94 - x Snekkersten (56° 00' N, 12° 36' E), Sealand, DENMARK 3.1.96. 65 km NNW.
  • 1c Barsebäck 6.10.95 - x Laboe (54° 24' N, 10° 13' E), Schleswig-Holstein, GERMANY 2.1.96. 232 km SW.
  • 2c+ Barsebäck 30.12.95 - x Kerteminde (55° 27' N. 10° 40' E), Great Belt, Fyn, DENMARK 7.1.96. 144 km WSW.

One interesting recovery from Bottsand belongs in this context; again the bird was present at a remarkably early date at Bottsand (cf the note ending section 2). Vresen (Fig. 1) is a typical retreat for Dunlin under severe winter conditions: an islet south of the Great Belt bridge:

  • 1c Bottsand 27.9.98 - V Vresen (55° 13' N, 10° 52' E), Great Belt, Fyn, DENMARK 8.1, 27.12.2000.

Even earlier presence of juveniles wintering in the Baltic is indicated by another recovery from Vresen, listed in the Swedish annual report for 2002:

  • 1c Falsterbo 5.9.01 - V Vresen, DENMARK 18.12.2001. 124 km W, 3 m, 13 d.

Turning to Germany, two factors are adverse to wintering in the Hiddensee-Bock area: low salinity (early freezing date, poor benthic fauna in winter) and varying water levels (high levels with northerly winds in late autumn). Large flocks may still lag behind in November and 1 - 2,000 Dunlin wintered in the 80's (Müller 1985), but in recent years many birds obviously moved west to neighbouring Schleswig-Holstein when temperatures fell below zero; 175 inds. were seen in January 1993 and 125 in January 1994 (B. Struwe-Juhl in litt.). It should be added that the nature reserve Bock has been visited only once a month in recent years. In general winter Dunlin do not move around much once they have settled in N. Germany; at Bottsand their daily movement does not exceed a few km (H. Behmann in litt.). Three other German sites: Peenemünder Haken, Langenwerder/Poel and Schlei (entrance to Schleswig) will hold flocks in late autumn and into winter, but they are abandoned if there is ice. During one ice-winter flight surveys revealed a winter flock on an islet in the Bay of Wismar, however; this may be where the Langenwerder birds go when there are sub-zero temperatures (V. Dierschke in litt.). A fourth site, the nature reserve "Grüner Brink" on Fehmarn may communicate with Hyllekrog on Lolland; over years there are a lot of winter observations, but neither locality is quoted as a permanent wintering site.

One area remains, the most important of all: the south end of Kattegat and the Belt Sea. Until recently Dunlins were not ringed, or even studied here; the only source of information was recoveries of ringed birds shot by hunters (and there are a number of December and January recoveries from Fyn and Jutland in the sixties, the pre-history of the wintering may be traced here). Keeping count of Dunlin flocks in such an area must be extremely difficult; there are hundreds of Danish islets - and wind-flats surrounding them! - seldom visited by ornithologists in winter. In addition the birds move between adjacent sites, obviously more than in the Baltic proper. So, the important sites on northern Sealand (Saltbæk vig, Sejerø bugt, Isefjord, Roskilde fjord), eastern Jutland (Horsens fjord, Mariager fjord), the isles of Samsø and Endelave, and northern Fyn marked in Fig. 1 should not be equalled with the same number of independent winter flocks, but nor do they reflect the movements of one or two flocks, roaming the whole area. There are dozens of flocks, foraging in restricted areas and shifting from one site to another (especially from the mouth to the interior of the fjords) mainly when they are forced to, by high water or ice. Such movements are easily overlooked, particularly in the waters between Sealand and Lolland-Falster (as they probably are between Fehmarn and Lolland and between the Falsterbo peninsula and Sealand/Møn). I have been unable to locate any material from the area between Fyn and Jutland, but Dunlins are known to occur here in winter, too.

2.2. Spring conditions, staging

2.2.1. The numbers staging between Kattegat and the Baltic. Possible origin. According to Meltofte et al. 1994 and Rösner 1997 up to 1.12 million Dunlin have been counted in the overall Waddensea area in May, i.e. the vast majority of all birds wintering in W. Europe congregate here before returning to their arctic breeding-grounds. The highest densities were recorded in the Schleswig-Holstein and Jutland sectors of the Waddensea, the birds pushing to get as far to the north (and east) as possible. Still, the eastern limit is by no means reached with the Danish Waddensea; the whole wintering area - see Fig. 1 - from northern Kattegat to German Vorpommern (Rügen) receives an additional extra of staging Dunlin in early spring. The annual bird report of the easternmost of the Danish isles, "Fugle på Sjælland 1990" (DOF. 1991) remarks: "Similar to last winter (1989), Dunlin seemingly arrived from SW from mid-February, due to the warm weather." This applies for Scania in mild winters as well; the winter flock abandons its traditional grounds on Måkläppen and the spits of Skanör and moves some 10 km east, to the more eutrophic Foteviken area. At the same time new birds are added: 500 - 1,000 birds in late March, 1,000 - 2,000 in April. Colour-ringed birds from winter are present: 30.3.95 one in 1,000 (Paul Eric Jönsson), 7.3.97 one in 200 (Peter Olsson) and 19.4.98 five in at least 1,000 (Henning Behmann, Christer Persson). One recovery indicates the wintering area of the fresh arrivals, the bird in question had shortened the flight distance to its breeding-grounds by some 1,100 kms. This recovery calls to mind the suggestion by Goede, Nieboer & Zegers 1990, that April staging birds from the Dutch Waddensea area move to new staging areas in the Baltic in May:

  • 2c+ Rumney Great Wharf (51° 30' N, 3° 06' E), Glamorgan, WALES 5.11.94 - V Lilla Hammarsnäs 13.4.95. 1147 km ENE.

Maximum spring flock size from the same years as in Table VI are given in Table II below:

Table II. Largest Dunlin flock in the Falsterbo-Foteviken area in spring. (Fåglar i Skåne, 1982 - 1993)

YearFlock sizeDate

826104 April
83130011 May
84150030 April
8590021 April
87240029 May
888005 April
90151018 April
91170027 March
93130021 March

The Saltholm-Amager area (Copenhagen), Isefjord/Roskilde Fjord, Saltbæk Vig (both N. Sealand), Northern Fyn, the Samsö/Endelave area etc. also harbour flocks of the same order of magnitude. In Germany, Hiddensee (Bessin) and the Gellen-Bock area may keep from a few thousand to 10,000 birds (more than 5,000 in 1991 Kube et al. 1994; at most 1,500 in 1993 Helbig et al. 1994); Langenwerder, Fehmarn, Bottsand and Schlei a few hundred to 1,000 from late March. It seems as if spring flocks of Dunlin are less frequently reported and/or published by annual report editors, so there is more uncertainty surrounding the numbers in spring. If 10,000 Dunlin have wintered in the Kattegat - South Baltic area in the best years, the number of staging birds in the same area has been at least twice that sum: 20,000, and possibly more. This may appear a droplet in the sea - less than 2 % of the maximum numbers staging in the Waddensea (Rösner 1997) - but from a Scandinavian point of view it could be an important contingent: a major part of the Swedish and Finnish breeding populations. (The Swedish breeding population is estimated at 51,000 breeding pairs (SOF. 1990), in my view a gross over-estimate; today those birds can be found nowhere in Swedish Lapland. The decline may have been caused by reinforced predation on waders during two decades without "lemming years").

2.2.2. Biometry and morphology of staging spring birds. Biometry may give one clue to the provenience of Dunlin staging in spring in the Sound area. 52 April birds had bill-lengths 26.0 - 36.6 mm, mean 31.3 ± 0.4 mm, s. d. 2.7 mm, wing-lengths 109.5 - 127, mean 117.6 ± 0.6 mm, s. d. 4.1 mm. Under all circumstances these values indicate a surplus of males. With mean values 31.4 (males) and 35.0 mm (females) in spring birds from the Waddensea (Rösner 1997, also see "Bill length distributions..."), a male : female ratio of e.g. 60 : 40 would have resulted in an overall bill length mean value of appr. 32.8 mm, so there is reason to believe that the staging birds belong to a short-billed population of western and / or northern origin.

Birds in complete winter plumage were still noted 12.4 - 14.4.95, included among them the above bird from Wales; it had migrated in full winter plumage from The Irish Sea to the Sound in order to moult and stage in Scania. Scapulars are grown one single feather at a time and often among the first feathers to be shed, the first fresh ones (on one side) seen 9.4.97, 12.4.95, this one-sided pattern even applying to newly arrived schinzii males (two 9.4.97). One alpina male 29.4.95 had appr. 70 % of belly and back in fresh plumage, another bird, probably a female on the same date no more than 5 - 10 %, two 2c birds (both probably females) on 30.4.01 had belly-patches appr. 10 and 20 % black, a female on 10.5.97 50 % of belly and back fresh. The first bird with practically fresh breeding plumage dates from 13.5.97. The effective individual moulting period seems to be approximately one month when birds start in early April, but there are birds which have barely begun in late April. The whole process is likely to speed up in May with increasing food supply in the main feeding-grounds: pasture and wind-flats.


3.1. Wintering

3.1.1. Needs of supplementary feeding versus need of protection from frostbite in midwinter. In mild weather - say 4 - 5° C by midnight and 8 - 10° C by noon - Baltic wintering Dunlins are excellent energy conservers; foraging seems to claim only a small fraction of their time, much of the day the birds sleep, breast to the wind, limping on one leg with heads laid backwards and bills hidden in scapulars. In the night-time they can not be found at wrack beds or exposed wind flats under such conditions, and nocturnal foraging must be negligible. This pattern changes when temperatures approach zero, particularly if there is an additional wind chill. With sub-zero temperatures, wintering Dunlin will forage throughout the night at rewarding sites. This behaviour calls the 'supplementary' hypothesis of McNeil et al. 1992 to mind; dark hours must be devoted to cover energy losses. Dierschke 1996, adding four German references, remarks that the tactile foraging of Dunlin enables the bird to collect the necessary food at will, in the daytime, or at night. There is no build-up of body reserves (see Fig. 9 of "Phenology and biometry...") to cover overnight losses in winter months, as reported from the Waddensea in December by Smit & Wolff 1981; instead mean weights are steadily decreasing from a peak by mid-November. Midwinter scarcity of food resources in the Sound area may be involved here, but birds under real strain (Barsebäck, New Year) still managed to maintain weights only 3 - 5 grams below local mean weights for January - February.

Still, there is one reservation to be made. With night temperatures well below zero, the wrack beds will become rock-hard and ice-covered and the water full of small ice-particles. Under such conditions there is no point in foraging, instead the legs should be protected from frost at any cost. Goss-Custard 1969 stated that British-wintering Redshanks Tringa totanus fed at night only in winter. This does not apply to wind flats in the Sound area; here there are nocturnal feeders already in September and the tendency is accentuated as the season progresses. With ice, wind and sub-zero temperatures wintering Tringa totanus robusta will abandon all attempts at foraging, however, instead entering belly-deep water and staying there, rocked to and fro and even swimming for a second or two when icy rollers pass. This movement combined with constant preening may help to keep ice from attaching to the feathers, and as a matter of fact there is no evidence that the increased nocturnal activity noted in Green Sandpipers during cold spells in midwinter (Smith et al. 1999) should be attributed to foraging activities; the birds might as well have been very actively moving and preening amidst streaming water. (So: the 'supplementary' feeding may be a phantom in this particular case). The same act is not performed as easily by the smaller Dunlin, but it is at least attempted when there is little wind; with wind they will instead crouch in small depressions along the shoreline. There comes a limit when 'supplementary' food intake is no longer first option, instead it is a matter of avoiding frostbite. In general weights are low, casualties occur, and live birds with ice on bill, back and tail-feathers are found after such nights.

3.1.2. Wintering far to the north: Why take heavy winter losses in some years? The late return of adults to the Sound is puzzling - and the fact, that they return to an "inferior" (compared to the moulting area in the Waddensea) area. And why should juveniles share their hardship at all? It goes without saying, that all resources available must be divided, but where lies the advantage that persuades Dunlin to take heavy winter losses in two or three years out of ten?

The first possible explanation considers the physical preconditions: there is less salt stress (e.g. Klaassen & Ens 1990) from Baltic or Sound water, allowing higher or swifter food intake than in e.g. the Dutch Waddensea, since all prey is of marine origin (Corophium, small bivalves etc) and carries the salinity of the surrounding sea-water. In addition there are no tidal cycles; the prey is available as long as there is no wind-induced highwater. The combination of less salt intake, less exposure to wind chill on leeward beaches and less flight expenditure (prompted by tide cycles) together with strict energy-conserving behaviour may well result in an energy budget at least as favourable as Waddensea conditions will allow. There may be less physical stress in the Baltic, less motion, less plumage wear; all such small gains outweigh the increased mortality of ice-winters.

The second possible explanation departs from the fact, that the wintering in this case is also a sort of preliminary staging - the birds occupying a staging area as early as possible. And this early arrival in turn must be connected with the fact, that the staging area has a limited carrying capacity, it cannot carry hundreds of thousands of moulting birds in spring (cf. Meltofte et al 1994, Rösner 1997). Still, if 10,000 Dunlin have wintered in the Baltic area in the best years, the maximum number of staging birds is at least twice that sum. When Rösner 1990, 1997 asserts, that there are no important spring congregations between the Waddensea and the breeding area, he shuts his eye to this small, but from a Fenno-Scandian point of view significant contingent. The March / April birds have on average shorter wings and shorter bills than any other contingent migrating by way of the Sound area, and - in spite of a possible male surplus - there is reason to believe, that the average longitude of their breeding area lies even closer than that of the original wintering population. These birds, with an intermediate staging area far to the north or/and east, bring to mind the April staging birds from the Waddensea with relatively low fat levels, described by Goede et al. 1990, in my opinion the growth of the staging flock of the Foteviken area in late April and early May must be attributed to rather late additions from the Waddensea area. Also note the concentration of staging birds in the German and Danish Waddensea emphasized by Meltofte et al 1994.

In W Russian or northern breeding areas (Kola Peninsula, the White Sea, Novaja Zemlja, Svalbard), the incidence of "Atlantic" (wet and chilly) and "Siberian" (warm) springs is best anticipated from the Baltic vantage point, and quick departure is facilitated if birds have reached 56° N (Halland, Blekinge) or 57° N (Öland, Gotland, Latvia, Estonia) in early May. There is much to be gained from early arrival in arctic areas: in Swedish Lappland Dunlin were already hatching by midsummer 2000 in an area, where most birds hade barely laid eggs by the same time throughout the nineties, due to late melting of thick snow-cover. How did they know, or anticipate? By all likelihood they had been staging close enough to anticipate an early spring. Here is another aspect, where an alleged "trial and error" progress of juveniles must give way to necessity: the striving of different populations for optimal staging areas and early departure. With an incisive wording it could be said: There is no option whatsoever for juveniles wintering in the Sound area, the choice has already been made for them: it is the staging area of their parents. And the staging juveniles in October / November are in a way heading for their optimal staging area already at that time; again no options, the schedule has been set by "experience", carried by combined heredity and tradition. There is a close interaction between the location of wintering and staging areas, this is why Black Sea staging birds fly south-east as early as possible, and experience/heredity must be introduced at the earliest possible stage (the November "turning point") in order to economize with all food resources between wintering and breeding-grounds.

  • To "Phenology and biometry of Dunlin Calidris alpina migrating by way of the Sound area, S. Sweden"
  • To "Migrating Dunlin Calidris alpina in the Baltic area: the moult issue"
  • To "Risk-prone or risk-averse? Dunlin Calidris alpina migrating with and without moult-gaps in the Baltic area"
  • To "Wintering and spring staging Dunlin Calidris alpina in the south Baltic area"
  • To "Migratory progress of juvenile and adult Dunlin Calidris alpina from two perspectives: the Baltic and the Waddensea"
  • To "Bill-length distributions in Dunlin Calidris alpina"
  • To the bill length account
  • About "adult buff" coverts
  • To the Meissner scale
  • To Dunlin references A - J
  • To Dunlin references K - Z
  • Back to start page

    First published 3.2.03, last changed 29.5.03.