J Ornithol (2006) 147:591–598 DOI 10.1007/s10336-006-0083-3
O R I G I N A L A RT I C L E
The importance of salt-marsh wetness for seed exploitation by dabbling ducks Anas sp. Ole R. Therkildsen Æ Thomas Bregnballe
Received: 4 July 2005 / Revised: 7 January 2006 / Accepted: 9 April 2006 / Published online: 10 August 2006 Dt. Ornithologen-Gesellschaft e.V. 2006
Abstract The relationship between the inundation of a salt marsh in southeast Denmark not subject to lunar tides and the availability and predation of seeds of the annuals Salicornia spp. and Suae´da maritima by autumn staging dabbling ducks was studied by carrying out exclosure experiments over the course of 2 years. There was a marked difference in the wetness of the salt marsh between the two study years, which resulted in distinct temporal patterns of salt-marsh use by dabbling ducks. In both years, the depletion of seeds of both Salicornia spp. and S. maritima was initiated subsequent to the flooding of the sample transects, which also induced the gradual release of seeds from the plants within the exclosures. Nevertheless, seeds were removed more rapidly in plots visited by dabbling ducks than in the exclosures. The predation of seeds took place as soon as the individual plants had been fully submerged, but before the seeds were released from the plants. The timing of flooding events during early autumn may potentially affect the availability of the salt-marsh seed stock. Therefore, weather conditions may impose critical constraints on the feeding opportunities for dabbling ducks during autumn migration on non-tidal salt marshes. Keywords Anas sp. Æ Dabbling ducks Æ Autumn staging Æ Seed exploitation Æ Flooding
Communicated by F. Bairlein O. R. Therkildsen (&) Æ T. Bregnballe Department of Wildlife Ecology and Biodiversity, National Environmental Research Institute, Kalø, Grena˚vej 14, 8410 Rønde, Denmark e-mail:
[email protected]
Introduction Temperate zone dabbling ducks Anas sp. live in environments where they experience considerable shortterm spatial and temporal variation in food availability. They often exploit temporarily and seasonally flooded wetlands, where they feed on plant material and invertebrates. Dabbling ducks are adapted to exploiting the patchy and unpredictable food resources of such wetlands and, often, they respond opportunistically to weather-related changes in food availability (van Eerden 1997). Dabbling ducks are generally reported to take leaves and seeds during autumn and winter, when this food resource is more abundant (Cramp and Simmons 1977). However, a minimum amount of water is crucial to permit the exploitation of food items like seeds, since dabbling ducks frequently use a highly specialised feeding technique of filtering soft mud or water (Cramp and Simmons 1977). Consequently, the exploitation of small plant seeds in marshes and aquatic invertebrates in saltpans and potholes is precluded during periods of drought. The presence of water in a marsh may affect the numbers of dabbling ducks not only through its effect on food abundance and food availability, but also through its potential impact on predation risk and attractiveness of the marsh as a roosting site. For example, wigeon Anas penelope become more vigilant as they move away from water during feeding bouts and tend to restrict their foraging to areas adjacent to water (von Ka¨nel 1981; Mayhew and Houston 1989). This behaviour is believed to reflect that the predation risk increases with the distance from water (Madsen 1988; Jakobsen and Ugelvik 1994).
123
592
We studied the relationship between the inundation of a salt marsh not subject to lunar tides and the availability of seeds to autumn staging dabbling ducks in a Danish wildlife reserve at the Baltic Sea coast. The availability of submerged vegetation to dabbling ducks in the shallows in this area is heavily influenced by daily changes in water levels (P. Clausen and O.R. Therkildsen, in preparation), which are mainly the result of changes in wind speed and direction. Therefore, the salt marsh constitutes an important alternative feeding area, especially during periods with high water levels, which restrict access to the submerged vegetation on the shallows. To study the influence of water levels on the exploitation of the salt marsh seed stock by dabbling ducks, we carried out exclosure experiments during the autumn of 1997 and 1998. We monitored the ripening and release of seeds of the annuals Salicornia spp. and Suae´da maritima, which dominate the lower parts of the salt marsh and constitute an important food source for dabbling ducks during autumn. The water level on the salt marsh differed markedly between the 2 years, enabling us to assess how salt-marsh wetness affected dabbling duck numbers and their exploitation of the seed resource.
Methods Study area The study was carried out in the Ulvshale/Nyord Wildlife Reserve on the coast of Møn in southeast Denmark (5501¢N 1215¢E). It is a coastal wetland consisting of brackish bays with shallow areas surrounding the Ulvshale peninsula and the island of Nyord, with extensive areas of reed beds and salt
Fig. 1 Ulvshale/Nyord Wildlife Reserve on the coast of Møn in southeast Denmark. The arrow indicates the experimental area in which exclosures and transects were placed
123
J Ornithol (2006) 147:591–598
marshes (Fig. 1). Juncus gerardi, Agrostis stolonifera and Festuca rubra dominate the upper parts of the salt marsh, whereas the low-marsh, i.e. depressions and saltpans, are colonised by the annuals Salicornia spp. and S. maritima during summer. In the lower, central part of the salt marsh, the latter two taxa form dense, homogeneous stands of 5–10 cm in height, with Salicornia spp. dominating the lowest parts. Following the establishment of a permanent reserve in 1989, the site has become one of the most important sites in Denmark for staging dabbling ducks (Madsen 1998). During autumn migration, ducks begin to arrive in late August and their numbers peak in October. Some dabbling ducks remain in the area during December–January. Maximum numbers in the autumns of 1985–1999 were 10,000 mallard Anas platyrhynchos, 4,900 teal Anas crecca, 34,500 wigeon, 2,600 shoveler Anas clypeata and 4,250 pintail Anas acuta (Madsen 1998; P. Clausen unpublished data). Although hunting occurred once every third week in autumn 1998 on the southern marsh (205 ha), this had little effect on the numbers of waterbirds on the salt marsh, especially when it was flooded (Bregnballe et al. 2005). In autumn, potential predators such as peregrine falcon Falco peregrinus and white-tailed eagle Haliaeetus albicilla are present, whereas red fox Vulpes vulpes occur throughout the year. The tidal amplitude in southeast Denmark is negligible, e.g. 6 cm at Lolland (Eger 1985). Nevertheless, the wind direction and its force have a substantial effect on water movements between Kattegat and the Baltic Sea and, consequently, on the water levels around islands in southeast Denmark. The average daily water levels in August–December recorded at Rødvig, c. 25 km NNE of Nyord, ranged from –38 cm to +55 cm compared with Danish Ordnance Datum
J Ornithol (2006) 147:591–598
(DNN) in 1997 and from –45 cm to +73 cm to DNN in 1998. The salt marsh is flooded at irregular intervals in autumn and winter, while in summer, precipitation is the main factor determining water levels in saltpans and the wetness of the salt marsh as a whole. Seed sampling To monitor the removal of S. maritima and Salicornia spp. seeds by dabbling ducks on the salt marsh, exclosure experiments were carried out in 1997 and 1998. In the salt marsh, two exclosures, each 3 · 3 m, were erected at the edge of saltpans. These two annuals form homogenous stands of approximately 10 cm in height along saltpan edges during summer. An exclosure consisted of four poles with a bird net surrounded by an electric fence to keep cattle out. Two sample transects were placed at approximately the same elevation at least 50 m from the exclosures at the edges of saltpans, i.e. the areas visited by the ducks. On each sampling date, 20 plants were collected randomly along transects and 15 plants of very similar size were collected in the exclosures (i.e. new plants were collected on each sampling date). A smaller number of plants were collected in the exclosures to avoid emptying them before the end of the experiment. In the case of Salicornia spp., the number of seeds per plant was used to monitor seed release and depletion. In the laboratory, the plants were dried and seeds were counted. For S. maritima, the seed number was expressed as the number of seeds per seed stand, which are placed at leaf bases. Each seed stand contains around 2–4 seeds. Only fully developed seed stands were included to obtain seed numbers (i.e. the upper part of the plants were removed). The mean numbers of seeds remaining on plants within exclosures and along sample plots on each sampling date were compared using the t test procedure in SAS/STAT (1989). van Eerden (1997) found that grazing intensity was highly correlated with food density for wigeon feeding on Salicornia europea. In this study, we measured the rate of seed loss over time in relation to the water level, which was assumed to be independent of plant density.
593
refer to this index of water level as the water level index (WLI). Dabbling duck numbers Counts of dabbling ducks on the salt marsh were made from 1 September until the end of December. The counts were made 2–3 times a week from a 7-m high observation tower using a 30· or a 20–60· zoom telescope.
Results Wetness of the salt marsh Following the dry, warm summer of 1997 (Rosenørn and Lindhardt 2000), most saltpans had dried out by late September, leaving the salt marsh almost dry. The saltpan WLI increased gradually until early October, when the salt marsh, including the experimental area, was flooded, inundating most saltpans. The WLI remained around index 10 throughout the study period, except for early November, when the WLI was above index 13 for a short period (Fig. 2). Following the cool, but otherwise normal, summer of 1998 (Rosenørn and Lindhardt 2000), water was present in most of the large saltpans, resulting in a WLI in September of around 7 (higher than at the same time in 1997). In 1998, the experimental area was partially flooded already in early September and, towards the end of the month, all plants were submerged. The WLI remained around index 7 until mid-October, after which, it gradually increased until late October, when even the saltpans on the highest parts of the salt marsh were flooded. The WLI remained above index 10 throughout November (Fig. 2). In both study years, the degree of inundation of the salt marsh increased most markedly when the water level in the surrounding shallows reached beyond a certain level, i.e. high surrounding water levels rather than precipitation caused the most significant increases in inundation (T. Bregnballe, unpublished data). Dabbling duck numbers
Water level The water level on the salt marsh was monitored during September–November 1997 and 1998. The water level in 13 of the largest saltpans was estimated at 1–7day intervals. The water level in each saltpan was given a score from 0 to 14, where 0 represented a completely dry saltpan and 14 represented a flooded saltpan. We
In 1997, dabbling ducks were almost totally absent from the salt marsh until early October. Their numbers increased rapidly following the flooding of the lower parts of the salt marsh between 2 and 3 October. Unfortunately, bad weather made it impossible to count the study area before 8 October. However, from observations in the field, it was clear that substantial
123
594
J Ornithol (2006) 147:591–598
Fig. 2 Water level index (WLI) in 13 saltpans on the salt marsh during autumn 1997 and 1998. See text for method
numbers of dabbling ducks moved into the area immediately after the flooding, as indicated in the figures. From late October until early November, only small numbers of dabbling ducks were observed, but following the flooding of the higher parts of the salt marsh, the numbers increased markedly in early November (Fig. 3a, b). In 1998, dabbling ducks were present on the salt marsh from early September, whereas few were observed during October. From late October, their numbers increased gradually towards mid-November, when ice covered the feeding grounds (Fig. 4a, b). Overall, more dabbling ducks were observed in the study area in 1998 than in 1997.
Salicornia spp. In both years, seeds of Salicornia spp. ripened in late August. In 1997, the seeds were gradually released as the plants dried by late September. Subsequent to the flooding in early October, significantly more seeds remained on the plants in the exclosures (Fig. 3b). In 1998, on 1st and 10th September, there was no significant difference between the number of seeds remaining on the plants in the exclosures and along the transects, whereas on all other sampling dates, significantly more seeds remained on the plants in the exclosures than along the transects (Fig. 4b).
Release and removal of seeds
Discussion
Suae´da maritima
Several factors affect the use of wetlands by waterbirds. The water level, height and density of vegetation, food abundance, predation risk, access to alternative feeding sites and human disturbance are amongst the potential compounding factors which determine the extent to which waterbirds exploit wetlands. In particular, the water level has been recognised to have a strong influence on the distribution of waterbirds and the water depth has repeatedly been shown to play a key role in determining where dabbling ducks and other waterbirds forage (Clausen 1998, 2000; Colwell and Taft 2000; Isola et al. 2000; O.R. Therkildsen and P. Clausen, in preparation). The close relationship between waterbirds and the presence of water has been ascribed to the requirement of nearby water for drinking and bathing while feeding (Owen 1972, 1973; Rijnsdorp 1986), whereas Mayhew and Houston (1989) provided evidence that wigeon
In both study years, seeds of S. maritima ripened in late August. In 1997, the seeds remained on the plants until the flooding in early October, after which, they were rapidly released. On 12th and 17th October, significantly more seeds remained on the plants in the exclosures than along the sample transects, whereas on all other sample dates, there was no significant difference between exclosures and transects (Fig. 3a). In 1998, seeds were gradually released from the plants following the partial flooding in mid-September. On 1st September, there was no significant difference between the number of seeds remaining on the plants in the exclosures and along transects, whereas on all other sampling dates, significantly more seeds remained on the plants in the exclosures than along the transects (Fig. 4a).
123
J Ornithol (2006) 147:591–598
595
Fig. 3 a The cumulative number of dabbling ducks on the salt marsh at Nyord during autumn 1997 and the average number of seeds per seed stand on individual plants of Suae´da maritima in exclosures and along sample transects (±SE). The grey bar indicates the time at which plants in the experimental area were submerged. The dotted line indicates that substantial numbers of dabbling ducks moved into the study area at a time when bad weather made it impossible to count the area. b The cumulative number of dabbling ducks on the salt marsh at Nyord during autumn 1997 and the average number of seeds remaining on individual plants of Salicornia spp. in exclosures and along sample transects (±SE). The grey bar indicates the time at which plants in the experimental area were submerged. The dotted line indicates that substantial numbers of dabbling ducks moved into the study area at a time when bad weather made it impossible to count the area
feeding close to water lowered predation risk. However, our study shows that the presence of water is also a prerequisite for the exploitation of small seeds by dabbling ducks. The marked difference in the timing of the inundation of the salt marsh between the two study years enabled us to describe the presence of dabbling ducks in relation to the water level on the salt marsh and their exploitation of the salt-marsh seed stock during autumn.
From observations in the field, it was clear that teal and wigeon were predominantly feeding on the lower parts of the salt marsh (i.e. areas dominated by S. maritima and Salicornia spp.). In contrast, mallard were generally grazing on the upper parts, whereas pintail were feeding in the larger saltpans by upending, possibly on invertebrates (Therkildsen, unpublished data). Therefore, it seems plausible to assume that teal and wigeon were responsible for the majority of the removal of seeds. Likewise, van Eerden (1997) found
123
596
J Ornithol (2006) 147:591–598
Fig. 4 a The cumulative number of dabbling ducks on the salt marsh at Nyord during autumn 1998 and the average number of seeds per seed stand on individual plants of S. maritima along sample transects and in exclosures (±SE). The light grey bar indicates the partial flooding of plants in the experimental area, whereas the dark grey bar indicates the time at which plants were submerged. b The cumulative number of dabbling ducks on the salt marsh at Nyord during autumn 1998 and the average number of seeds remaining on individual plants of Salicornia spp. along sample transects and in exclosures (±SE). The grey bar indicates the time at which plants were submerged
that both teal and wigeon exploited seeds of S. europaea in Lauwerszee, NL. However, wigeon consumed more seeds than teal, which was a result of unfavourable water levels for teal to exploit this food source. Our study showed a close relationship between the timing and extent of flooding of the salt marsh and the increase in dabbling duck usage of the marsh. Hence, a marked difference in the wetness of the salt marsh between the two study years resulted in distinct temporal patterns of salt-marsh use by dabbling ducks. In particular, the sudden flooding of the otherwise dry salt
123
marsh in early October 1997 resulted in an abrupt increase in numbers. In both years, the depletion of seeds of both Salicornia spp. and S. maritima was initiated subsequent to the flooding of the sample transects, which also induced the gradual release of seeds from the plants within the exclosures. However, seeds were removed more rapidly in areas visited by dabbling ducks than in the exclosures, e.g. predation of seeds took place as soon as the individual plants had been fully submerged, but before the seeds were released from the plants.
J Ornithol (2006) 147:591–598
Presumably, the submergence in water disrupts the osmotic balance of salt-marsh plants (Mitsch and Gosselink 2000), leading to a gradual breakdown of cell walls. This allows dabbling ducks to strip partially degraded plants of seeds rather than filtering released seeds from the water surface. This was supported by observations in the field immediately after the plants became covered by water. Hence, the submergence of plants induced both the release of seeds and facilitated the physical removal of seeds directly from the plants by dabbling ducks. Some seeds were released before the flooding in both years, except for Salicornia spp. in 1997. Seeds released from the plants were concentrated on the water surface, where they were available to feeding dabbling ducks (personal observation). Euliss and Harris (1987) found that both pintail and green-winged teal Anas carolinensis would feed heavily on food particles that were concentrated in this manner. This may be part of the reason why dabbling ducks were still present on the salt marsh, even though no seeds remained on the plants, suggesting that, in both years, dabbling ducks were actually filtering seeds from the water until late in the season. In both years, seeds of Salicornia spp. had been depleted by mid-October. In 1997, seeds of S. maritima were depleted in early November, whereas in 1998, seeds had been depleted towards mid-October. In 1997 and 1998, the flooding of the highest parts of the salt marsh in early November and late October, respectively, was followed by a marked increase in dabbling duck numbers. Presumably, this was a result of the flooding making seeds from grasses on the higher parts of the salt marsh available to dabbling ducks. Therefore, the availability of other seeds than Salicornia spp. and S. maritima may have contributed to the continued presence of dabbling ducks on the salt marsh later in the season. Evidently, in both study years, dabbling ducks responded to an increase in the extent of inundation of the salt marsh. Hence, the depletion of the seed stock was caused by the presence of dabbling ducks on the salt marsh. Since no exploitation of the seed stock took place before the flooding, we conclude that the wetness of the salt marsh had a direct effect on the availability of seeds. However, other factors may also have affected the seed predation. The timing of flooding events during early autumn may, potentially, have a major effect on the feeding opportunities for dabbling ducks during autumn migration. In contrast to managed wetlands with manipulated water levels and rather predictable flooding of food resources in intertidal areas, weather
597
conditions during autumn may impose critical constraints on feeding opportunities for dabbling ducks on salt marshes situated in non-tidal areas. In conclusion, unfavourable weather conditions during early autumn may hinder the increase of dabbling duck numbers on salt marshes because an otherwise attractive food source (seeds) is unavailable until flooding has occurred. In particular, this can be critical to teal, because their small and shorter neck restrict access to submerged macrophytes, so that virtually no such food is available to this species at high and intermediate water levels in the surrounding shallows (P. Clausen and O.R. Therkildsen, in preparation). Therefore, to assess the importance of a salt marsh to dabbling ducks during autumn staging, one has to take into account the water level and, consequently, the food availability to each individual species, both in the surrounding shallows and on the salt marsh in question.
Zusammenfassung Die Bedeutung der Feuchtigkeit von Salzwiesen fu¨r die Samennutzung durch Schwimmenten (Anas sp.) ¨ ber zwei Jahre wurde durch Auschlussexperimente in U einer Salzwiese im su¨do¨stlichen Da¨nemark, die nicht dem Gezeitenwechsel unterliegt, untersucht, welchen ¨ berflutung und der Zusammenhang es zwischen U Verfu¨gbarkeit und Nutzung von Sa¨mereien der einja¨hrigen Pflanzen Salicornia spp. und Suae¨da maritima durch rastende Schwimmenten wa¨hrend des Herbstzuges gibt. Zwischen den beiden Jahren gab es einen erheblichen Unterschied in der Feuchtigkeit der Salzwiese, was zu unterschiedlichen zeitlichen Mustern der Nutzung der Salzwiese durch Schwimmenten fu¨hrte. In beiden Jahren begann die Freisetzung der Samen sowohl bei Salicornia spp. als auch ¨ berflubei Suae¨da maritima unmittelbar nach der U tung der Probenfla¨che; gleichzeitig begann auch die graduelle Freisetzung der Samen der Pflanzen in den abgesperrten Bereichen. Auf Fla¨chen mit Enten wurden die Samen rascher entfernt als in den Ausschlussfla¨chen. Das Samenfressen durch die Enten begann unmittelbar, nachdem die einzelne Pflanze vollsta¨ndig untergetaucht war, aber noch bevor die Samen von der Pflanze selbst freigesetzt wurden. Das zeitliche Auftreten der Flutereignisse wa¨hrend des fru¨hen Herbstes ko¨nnte mo¨glicherweise die Verfu¨gbarkeit des Samenangebots von Salzwiesen beeinflussen. Daher ko¨nnten die Wetterbedingungen zu
123
598
kritischen Einschra¨nkungen im Nahrungsangebot fu¨r Schwimmenten auf dem Herbstzug auf tidenunabha¨ngigen Salzwiesen fu¨hren. Acknowledgements Per A. Kjær, Ebbe Bøgebjerg and Jens Peder Hounisen are thanked for their assistance in the field. Fugleværnsfonden and landowners on Nyord are thanked for permission to work on their land. We thank Tony Fox for his valuable comments on an earlier draft of this paper. The experiments on which this paper is based comply with the current laws of Denmark.
References Bregnballe T, Hounisen JP, Bøgebjerg E (2005) Regulering af jagt pa˚ vandfugle i kystzonen: Forsøg med dato-regulering pa˚ Nyord. Reasearch Notes from the National Environmental Research Institute, Denmark, no 205, 68 pp Clausen P (1998) Choosing between feeding on Zostera and saltmarsh: factors affecting habitat use by brent geese in spring. Norsk Polarinstitutt Skrifter 200:269–286 Clausen P (2000) Modelling water level influence on habitat choice and food availability for Zostera feeding brent geese in non-tidal areas. Wildl Biol 6:75–87 Colwell MA, Taft OW (2000) Waterbird communities in managed wetlands varying in water depth. Waterbirds 23(1):45– 55 Cramp S, Simmons KEL (1977) Handbook of the birds of Europe, the Middle East and North Africa: the birds of the western Palearctic. Vol 1. Ostriches to ducks. Oxford University Press, Oxford, 722 pp Eger L (1985) Vandstande i Østersøen: Varsling af stormflod. PhD dissertation, Institute of Mathematics and Operations Research, Technical University of Denmark, 74 pp Euliss NH, Harris SW (1987) Feeding ecology of northern pintails and green-winged Teal wintering in California. J Wildl Manage 51(4):724–732
123
J Ornithol (2006) 147:591–598 Isola CR, Colwell MA, Safran RJ, Taft OW (2000) Interspecific differences in habitat use by shorebirds and waterfowl foraging in managed wetlands of California’s San Joaquin Valley. Waterbirds 23(2):196–203 Jakobsen OW, Ugelvik M (1994) Grazing and vigilant behaviour of breeding Eurasian Wigeon Anas penelope in relation to distance from water. Wildfowl 45:119–123 Madsen J (1988) Autumn feeding ecology of herbivorous wildfowl in the Danish Wadden Sea, and impact of food supplies and shooting on movements. Danish Rev Game Biol 13:1– 32 Madsen J (1998) Experimental refuges for migratory waterfowl in Danish wetlands. II. Tests of hunting disturbance effects. J Appl Ecol 35(3):398–417 Mayhew P, Houston D (1989) Feeding site selection by Wigeon Anas penelope in relation to water. Ibis 131(1):1–8 Mitsch WJ, Gosselink JG (2000) Wetlands. Wiley, New York, 722 pp Owen M (1972) Some factors affecting food intake and selection in white-fronted Geese. J Anim Ecol 41(1):79–92 Owen M (1973) The winter feeding ecology of Wigeon at Bridgewater Bay, Somerset. Ibis 115:227–243 Rijnsdorp AD (1986) Winter ecology and food of Wigeon in inland pastures in the Netherlands. Ardea 74(2):121–128 Rosenørn S, Lindhardt K (2000) Dansk vejr i 100 a˚r. Egmont Lademann, København SAS/STAT (1989) User’s guide, version 6, 4th edn, vol 1–2. SAS Institute Inc., Cary, NC van Eerden MR (1997) Waterfowl movements in relation to food stocks. In: van Eerden MR (ed) Patchwork: patch use, habitat exploitation and carrying capacity for water birds in Dutch freshwater wetlands. PhD thesis, University of Groningen, The Netherlands, pp 151–164 von Ka¨nel A (1981) Winter feeding ecology of wigeon Anas penelope at the Ouse Washes, England. Ibis 123(3):438–449
