Biometrics and timing of primary moult of non-breeding Mute Swans Cygnus olor at Lake IJsselmeer, the Netherlands

Transcription

1 Biometics and timing of pimay moult of non-beeding Mute Swans Cygnus olo at Lake IJsselmee, the Nethelands KLAAS VAN DIJK and MENNOBART R. VAN EERDEN This pape examines the influence o f weight and size upon the timing o f wing moult o f488 nonbeeding Mute Swans caught duing the flightless peiod at Lake IJsselmee (the Nethelands) in imaies gew at a ate o f 6.3 mm/day, but the gowth ate slowed down late on. Stat o f wing moult showed a peak in thefist ten days o f July. males began on aveage six days ealie than adult females. Weights ae low, compaed to othe studies. Weight was stongly coelated with size. Thee was a slight tendency fo lage sized swans to stat moult ealie. The coelation between body weight and stat o f wing moult is highly significant, the heaviest swans stated ealiest. The swans lost weight ove time. The low weights and weight-losses might indicate that food is limiting. Seveal studies have been caied out on moulting Mute Swans Cygnus olo in Westen Euope (Andesen-Haild 1981a,b, Mathiasson 1973, 1981a,b). Hitheto little attention has been paid to the timing of pimay moult. This pape epots on the timing of pimay moult of the Dutch population of non-beeding Mute Swans at Lake IJsselmee. The aim of this study is to analyse the possible influence of weight and measuements on the timing of pimay moult. The Mute Swan is a common beeding bid in the lowe pats of the Nethelands with 3,000-4,000 beeding pais (SOVON 1987). Beeding bids stay in thei teitoy with the gowing cygnets duing wing moult. Non-beedes migate to special locations which povide sufficient food and safety fom pedatos. Lake IJsselmee and Lake Gevelingen ae the most impotant sites in the Nethelands. Lake Gevelingen is a salt-wate lake with eelgass Zostea maina as the main food souce. Maximum numbes vay between 1,000 and 2,000 (Meininge et al. 1984, 1985). Lake IJsselmee is a shallow fesh-wate lake of 2,000 km2 suounded by man-made dikes (Fig. 1). A submeged vegetation of geen algae Cladophoa sp. gows along these dikes on bouldes. The algae ae only pesent duing summe and fom the only food souce fo the swans. Mute Swans ae absent in winte, but in summe flocks of moulting non-beedes assemble along the dikes (Fig. 2). The aveage maximum in July is 3,960 individuals (Table 1 ). The Houtibdijk (52 37 N, 5 25 E), length Supplement No. 1 (1991):

2 imay moult in Mute Swans Figue 2. Flock of moulting Mute Swans along the Houtibdijk, 28 June hoto Klaas van Dijh km, holds the lagest numbes with up to 3,000 individuals. Numeous ecoveies eveal that the MuteSwansofLakeUsselmeeaealmostexclusively of Dutch oigin (Renssen 1981, Van Dijk 1991). Exchange with bids fom the lage moulting sites in Denmak does not appaently occu (Andesen- Haild 1981a). Methods In 1983 small boats with out-boad engines wee used tocaptue flightless swans along the Houtibdijk on 21 June (12), 5/6 July (155), 18/19 July (89), 25 July (76), 8/9 August (107) and 26 August (49). Thee was at least a delay of one hou between catching and pocessing the swans ashoe. Afte being sexed (cloacal examination) and inged, leg colou was used to detemine colou moph (gey o leucistic/ olish, Andcsen-Haild 1978). Swans wee classified as fist summe (bom in 1982) by the pesence of small gey/bown (gey moph) o light beige (leucistic moph) feathes on the lowe back, supplemented by infomation about bill-colou and knob-size. The othe swans weeagedasadults (bom befoe 1982), although some could not be divided popely and wee put in the categoy unaged (bom in o befoe 1982). Length (to the neaest mm) of the flattened plo (outemost pimay) fom the tip to its oigin fom the skin was used to descibe pimay moult. The stat of pimay moult is defined as the day when the new pimaies appea. Wehave taken adelay of thee days fo the few flightless swans which still had thei old pimaies and a delay of 1.5 day fo swans lacking all pimaies. Fo all othe individuals gowth ates of plo (deived fom ecaptues) wee used to wok back to the stat-date of pimay moult. Five stuctual measuements (all to the neaest mm) wee taken. The width of the foot-web (between outemost toe-webs, fully stetched), the wing length (maximum chod with a stopped ule) with subsequently the length of plo subtacted to obtain a moult-independent wing-size, the length of the foe-am (adius and ulna, measued with slide callipes) and the length of the tip of bill to the neaest end of the nostil and the neaest end of the eye. These last two measuements wee taken fom pojected colouslides, taken of each swan following a standad pocedue. Weight was measued with esola sping balances to the neaest 0.1 kg. All measuements wee caied out by the fist autho. The analysis is based on488 bids (263males, 220 females and 5 of unknown sex) of which 29 wee caught twice. The swans wee divided into 157 fist summe (IS), 320 adult and 11 unaged bids. The leucistic moph was ecoded in 41.4 % of the bids ( n = 486), divided into 33.3 % of males ( n = 261) and 505% of females (n=218). Statistical analyses wee caied out with SSS. Significance was detemined at the 0.05 level. Results Timing o f wing moult Gowth ates of plo ae pesented in Figue 3. These data show a constant ate fo quite a long peiod, but suggest a pobable slowing down in the latte stages. To claify that gowth ate deceases as moult pogesses, we ha ve divided the ecaught swans into thee diffeent goups. The fist goupof 15 bids with pi 0 < 200 mm at ecaptue had an aveage gowth ate of 6.3 mm/day (SD = 0.28). Thee wee no significant diffeences (t-tests) between ages (sexes combined) and between sexes (ages combined). The second goupofsix swans caughtwithpl0< 100 mm and ecaught with plo > 200 mm had an aveage gowth of 5.7 mm/day (SD = 0.59). The thid goup consisted of seven swans caught withpl0>100 mm (X = 165) and ecaught withpl 0 > 200mm (X=244). These bids had a gowth ate of 5.0 mm/day (SD = 0.63), which is significantly lowe than the gowth ate of the fist goup (t-test, <0.005). Fo easons of simplicity we state that pi 0 gows with a constant ate of 6.3 mm/day fom the stat uptoalengthof200 Table 1. Numbes of M ute Swans at Lake IJsselmee, based on aeial counts. *Afsluitdijk not counted June 3,484 3,432 3,613 4,103 3,071 1,489* 2,177 2,415 3,040 4,030 July 3,510 4,542 4,284 4,748 4,134 3,600 3,709 2,870 3,560 4,686 August 3,458 4,052 4,232 4,129 3,949 3,302 2,867 2,521 3,056 3,220

3 2 9 8 Klaas van Dijk and Mennobat R. van Eeden LENGTH OF IO (MM) WHEN RECAUGHT Figue 3. Gowth ates of plo of 28 flightless Mute Swans caught twice in Aveage gowth ate is indicated fo plo < 200 mm. Open dots efe to bids caught with plo < 100 mm and ecaught with plo > 200 mm. mm and aftewads with a ate of 5.0 mm. We could not coect fa a likely futhe decease in gowth ate as plo appoaches its final length ( mm). This is a mino poblem as only five swans wee caught with pi 0 longe than250mm (maximum 268 mm). The timing of the onset of wing moult is vaiable (Fig. 4). The fist swan must have stated at the end of May and some pobably still had to stat by the end of August All goups exhibit a peak in the onset of pimay moult in the fist ten days of July. males began on aveage on 3 July (SD = 15 days) and adult females began on aveage at 9 July (SD = 18). males began on 11 July (SD = 15) and fist summe females on 14 July (SD = 18). males stated significantly ealie than adult females (t-test,/><0.001)andadultmales stated significantly ealie than fist summe males ( O.OOOl). Othe diffeences wee not significanl Stuctual size Figue 4. Timing of stat of pimay moult of flightless Mute Swans caught a t I>ake Ûsselmee in A = adult male ( n ~ 180), B = fist summe male ( n = 80), C = adult female ( n = 137), D = fist summe female ( n = 11). The stuctual measuements ae shown in Table 2. Males wee significantly lage than females fo all five measuements (t-tests, <0.005). Thee wee no significant diffeences between fist summe and adult males, except fo the tip of bill to nostil (0.025 < <0.05). females wee significantly lage sized than fist summe females, except fo the tip of bill to nostil. These diffeences wee geatest in footweb ( <0.005) and less ponounced in wing,

4 imay moult in Mute Swans WEIGHT (KG) JULY AUGUST Figue 5. Weights of moulting M ute Swans in The data epesent actual weights at pocessing. Bas indicate 95% C.I., sample sizes (boken down by sex and age) ae indicated unde each gaph.

5 3 0 0 Klaas van Dijk and Mennobat R. van Eeden Table 2. Stuctual measuements (mm) of flightless Mute Swans caught in Values ae mean ± SE, numbe of individuals ( n ) and ange. Wing Foe-am Foot web Nostil to bill tip Eye to bill tip male 264.8±0.06 (176) ±0.06 (178) ±0.06 (179) (179) (179) female 250.3±0.08 (133) ±0.09 (136) (137) (137) (137) male 264.1±0.10 (79) ±0.08 (80) ±0.09 (79) (79) (79) female (76) ±0.12 (77) ±0.09 (76) (77) (77) foe-am and tip of bill to nostil (0.025 < < 0.05). Weight changes Mean weights and anges fo each age and sex class ae shown in Table 3. The heaviest male (12.8 kg) weighed moe than twice as much as the lightest female (5.4 kg). Diffeences between sexes (boken down by age) and between ages (boken down by sex) wee vey significant (ttests, <0.0005). The aveage weights deceased in the couse of time (Fig. 5). This downwad tend is noticeable fom the beginning of July Table 3. Aveage body weights (kg) of flightless Mute Swans caught in Data epesent actual weights at pocessing. N Mean 1 SE Range Males Females Males Females until the beginning of August. The aveage weights emained constant duing August. The data fom ecaptued swans confim this patten of weight losses (Fig. 6). These ecaptued bids lost on aveage 25 g (SD = 27.0) pe day. Thee wee no significant diffeences between ages (sexes combined) and between sexes (ages combined) in the median weight deceases of the ecaptued swans (Mann-Whitney U-tests). Relationship between timing o f moult, weight and size JULY AUGUST Figue 6. Changes in body weight of 19 flightless Mute Swans caught twice in Numbes indicate diffeent sex and age classes: (1) adult male, (2) adult female (3) fist summe male and (4) fist summe female. We ha ve used conveted weights fo the analysis of the influence of weight on the othe paametes. We assume that each swan loses 25 g/day at a constant ate duing wing moult. The conveted w eights wee detem ined by inceasing the actual weights by 25 g fo each day between catching and the calculated day of stat of pimay moult. The elationship between weightandsizeisobvious (Table 4). Stuctually lage sw ans ae heavie. Both wing measuements give the best coelations. Models of thee-dim ensional com binations of measuements of wing, foot-web and bill do not impove the coelations with weight.

6 imay moult in Mute Swans Table 4. Coelations between body weight and stuctual sizes of Mute Swans caught in esented ae eason s coelation coefficients (). Significance is indicatedby: NS >0.05, * < <0.05, ** 0.005< <0.025, * * * < Wing Foe- a Foot web Nostil/bill Eye/bill Male 0.23 *** 0.32 k-k-k 0.21 *** 0.15 ** 0.17 ** Female 0.60 kkk k-k-k 0.48 *** 0.38 k-k-k 0.43 *** Male 0.48 *** 0.41 k k k 0.38 *** 0.12 NS 0.05 NS Female 0.49 *** 0.49 kkk 0.33 k-k-k 0.27 ** 0.35 *** Table 5. Coelations between size and stat of pimay moult of Mute Swans caught in Fo explanation see Table 4. Wing Foe-am Foot web Nostil/bill Eye/bill Male 0.09 NS NS 0.12 NS NS ** Female NS NS NS NS * Male ** ** NS NS NS Female NS kk NS kk *** Table 6. Coelations between body weight and stat of pimay moult of Mute Swans caught in Fo explanation see Table 4. Male kkk Female kkk Male kkk Female kkk Thee is only a slight tendency that lage sized swans stat ealie with pimay moult than smalle sized individuals (Table 5). The coelations between body weight and the stat of wing moult ae much bette (T able 6). Heavie swans begin to moult ealie than lighte ones, so a bid s condition influences the timing of pimay moult. Results of multiple egession analyses (Table 7) confim that weight is the most impotant vaiable in all age and sex classes to explain diffeences in the timing of the stat of wing moult. The contibution of size measuements is negligible. Discussion It should be emembeed that thee ae seveal souces of potential bias. Fo instance, the catching method esults in a sample of swans which ae unable to fly. Bids not yet in moult, o aleady able to fly again could not be caught. Consideing the accuacy of weights, swans often defaecate one to thee times between captue and pocessing. Fesh doppings weigh aound 0.1 kg (Bacon & Coleman 1986). Assumptions about the time-lag between shedding the old feathes and appeaance of the new, about constant weight loss and about the independence of gowth ate and weight loss upon each othe and upon size and the stat of moult may also cause bias. Moe detailed eseach is necessay to make clea if these assumptions ae eliable. Finally, the data ae based on a study of just one yea. As many watefowl species show consideable annual fluctuations in the timing of beeding, migation and moult, one should be caeful in compaing these data with othe findings. The gowth ate of plo in the constant stage (6.3 mm/day) is compaable with the scace data fom liteatue. Mathiasson ( 1973) ecoded a ate of 6.5 mm ( n = 7) fo p5 and Andesen- Haild (1978) mentions aate of 6-7 mm. Baue & Glutz von Blotzheim (1968) quote a ate of 7 mm. Thee have been elatively few studies which have measued feathe gowth ates and most elate to captive bids (Ginn & Melville 1983). Owen & Ogilvie (1979) showed that wild Banacle Geese Banta leucopsis have equal gowth ates of pimaies iespective of age o sex. They suppose a pobable slowing down of

7 3 0 2 Klaas van Dijk and Mennobal R. van Eeden Table 7. Relative impotance of weight and size fo pediction of stat of pimay moult of Mute Swans caught in esented ae stepwize multiple egession analyses with stat of pimay moult as dependent vaiable and weight and size measuements as independent vaiables with = 0.05 as the limit fo inclusion and deletion, 2 is the squaed multiple coelation coefficient. Step Vaiable 2 male 1 weight 0.21 < foot-web 0.26 < foe-am eye/bill nostil/bill female 1 weight 0.14 < wing male 1 weight 0.23 <0.001 female 1 weight 0.17 <0.001 gowth in the latte stages. Owen & King (1979) clealy demonstate with seveal ecaptues of wild MalladAnas platyhynchos that the gowth ate of the longest pimay is equal fo males and females and significantly declines as moult pogesses. Ou findings agee with these studies. The swans of Lake IJsselmee ange in stat of pimay moult fom the end of May to the end of August with a peak in the fist ten days of July (Fig. 4). Non-beedes along the Swedish west coast stated between mid-june and the beginning of Septembe with a peak aound the end of July and the beginning of August (Mathiasson 1973). The Swedish swans showed a clea tendency fo males to stat moult somewhat ealie than females, coesponding with ou findings that adult males begin six days ealie than adult fem ales. U npublished data (Zwanenwekgoep Avifauna Goningen) confim that paied bids with no offsping which stay in thei teitoy duing wing moult show a simila diffeence between the sexes. males have a longe flightless peiod due to a simila gowth ate of the pimaies compaed to that of the females, combined with a longe wing. The adaptation that males stat wing moult ealie is useful fo both membes of the pai in ode to egain flight ability at the same time. Visual obsevations suggest that a high popotion of the moultes at Lake IJsselmee ae paied. This agees with the age distibution of inged swans which shows that 41 % ae olde than thee yeas (Van Dijk 1990). The weights of the Dutch non-beeding swans (Table 3) ae low in compaison to othe data. Andcsen-Haild (1981b) mentions aveages of adult males fom diffeent moulting aeas in Denmak vaying fom 10.3 to 11.6 kg. Diffeences wee mainly caused by diffeences in food quality. Mathiasson (1981b) found in Sweden that fist summe males weighed on aveage 10.0 kg and fist summe females 7.8 kg. Reynolds (1972) pesents data fo fist summe swans fom England, whee males weighed on aveage 11.0 kg and females 8.2 kg. The low Dutch weights might indicate that the availability and/o quality of the food at Lake IJsselmee is not sufficient to suppot all the swans adequately. Although theaveage weights of the Dutch swans ae low, they still decease in the couse of time (Figs. 5, 6). The loss of weight is in ageement with data fom Denmak, whee the swans lost up to 20% of thei body mass duing moult (Andesen-Haild 1981b). An English analysis, howeve, could not find a distinct decease in weight duing moult (Bacon & Coleman 1986). The weight losses of the Dutch swans may be anothe indication that food is athe scace at Lake IJsselmee. Body mass is the most impotant paamete in expl aining the vaiation in stat of wing moult (Tables 6,7). Ou findings that moult stats fist in the heaviest individuals agees with Danish investigations (Andesen-Haild 1981b). Futhe studies at Lake IJsselmee ae necessay to test the hypothesis that food availability might influence timing of wing moult though body condition.

8 Refeences imay moult in Mute Swans Andesen-Haild, Knoppsvanen. Skav, Copenhagen. (In Danish). Andesen-Haild,. 1981a. Migation of Cygnus olo inged in Denmak in winte and duing moult. In: G.V.T. Matthews & M. Smat (Eds.) oc. 2nd Int. Swan Symp., Sappoo 1980, IWRB, Slimbidge. p Andesen-Haild,. 1981b. Weight changes in Cygnus olo. In: G.V.T. Matthews & M. Smat (Eds.) oc. 2nd Int. Swan Symp., Sappoo 1980, IWRB, Slimbidge. p Bacon,.J. & Coleman, A.E An analysis of weight changes in the Mute Swan Cygnus olo. Bid Study 33: Baue, K.M. & Glutz von Blotzheim, U.N Handbuch de Vögel Mitteleuopas, band 2. Akademische Velagsgesellschaft, Fankfut am Main. Van Dijk, K Hekomst en leeftijdssamenstelling van uiende Knobbelzwanen Cygnus olo op het IJsselmee. Limosa 64. (In Dutch with English summay). Ginn, H.B. & Melville, D.S Moult in Bids. BTO, Ting. Mathiasson, S A moulting population of non-beeding Mute Swans with special efeence to flight-feathe moult, feeding ecology and habitat selection. Wildfowl 24: Mathiasson, S a. The moulting gound s elation to beeding and winteing aeas as evealed by neck-banded Cygnus olo. In: G.V.T. Matthews & M. Smat (Eds.) oc. 2nd Int. Swan. Symp. Sappoo 1980, IWRB, Slimbidge. p Mathiasson, S. 1981b. Weight and gowth ates of mophological chaactes of Cygnus olo. In: G.V.T. Matthews & M. Smat (Eds.) oc. 2nd Int. Swan Symp., Sappoo 1980, IWRB, Slimbidge. p Meininge,.L., Baptist, H.J.M. & Slob, G.J Vogeltellingen in het Deltagebied in 1975/ /80. Rijkswatestaat, Middelbug. (In Dutch). Meininge,.L., Baptist, H.J.M. & Slob, G.J Vogeltellingen in het Deltagebied in 1980/ /84. Rijkswatestaat, Middelbug. (In Dutch). Owen, M. & King, R The duation of the flightless peiod in fee-living Mallad. Bid Study 26: Owen, M. & Ogilvie, M.A Wing molt and weights of Banacle Geese in Spitsbegen. Condo 81: Renssen, Th. A Vespeiding van een locale populatie Knobbelzwanen Cygnus olo. Limosa 54: (In Dutch with English summay). Reynolds, C.M Mute Swan weights in elation to beeding. Wildfowl 23: SOVON Atlas van de Nedelandse Vogels. SOVON, Anhem. (In Dutch). K laas van D ijk1, Univesity of Goningen, Zoological Laboatoy,.O. Box 14, 9750 AA Haen, The Nethelands. M ennobat R. van E eden2, Rijkswatestaat, Diectoate Flevoland,.O. Box 600, 8200 A Lelystad, The Nethelands. 'esent addess: Veaeestaat 48, 9718 SN, Goningen, The Nethelands. 2Request fo epints.