Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Pre-incubation behaviour of Harlequin ducks (Histrionicus histrionicus) in Labrador : testing the function… Squires, Kelly A. 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-0164.pdf [ 2.44MB ]
Metadata
JSON: 831-1.0075097.json
JSON-LD: 831-1.0075097-ld.json
RDF/XML (Pretty): 831-1.0075097-rdf.xml
RDF/JSON: 831-1.0075097-rdf.json
Turtle: 831-1.0075097-turtle.txt
N-Triples: 831-1.0075097-rdf-ntriples.txt
Original Record: 831-1.0075097-source.json
Full Text
831-1.0075097-fulltext.txt
Citation
831-1.0075097.ris

Full Text

PRE-INCUBATION BEHAVIOUR OF HARLEQUIN DUCKS {Histrionicus histrionicus) IN LABRADOR: TESTING THE FUNCTION OF MALE VIGILANCE AND AGGRESSION by Kelly A . Squi res B.Sc. (Hon), Memorial University of Newfoundland, 1996 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E In T H E F A C U L T Y O F G R A D U A T E S T U D I E S T H E F A C U L T Y O F F O R E S T R Y Department of Forest Sc i ences Centre for Appl ied Conservat ion Resea rch W e accept this thesis as conforming to the required standards T H E U N I V E R S I T Y O F BRIT ISH C O L U M B I A M A R C H 2003 © Kelly A . Squi res In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) A b s t r a c t Male behaviours such as vigi lance and aggress ion are usually a s s u m e d to function in paternity assurance during the pre-incubation period. However, the functions of these behaviours may be to protect females from conspeci f ics and predators during feeding and resting thereby enhancing their ability to form clutches. I studied pre-incubation behaviour in a population of Harlequin Ducks to determine how male vigi lance and aggress ion are related to female foraging, resting and date of clutch initiation. In 2000 and 2001 , 217 hours of time-activity data on 12 paired females and 16 paired males were col lected on two rivers in central Labrador during the pre-incubation period. Ma les face greater risk of losing paternity due to mate loss than to extra-pair copulation ( E P C ) because E P C attempts were rare and did not result in c loacal contact due to female resistance. Most aggress ion was low-level 'head-nodding' by both members of a pair. Though high-level male aggress ion was directed at other males, it occurred during feeding and not resting, despite c lose proximity between resting pairs and unpaired males. High-level aggress ion by males was likely used to defend feeding areas and not to 'mate guard' . Harlequin Ducks were observed in lake outlets throughout the pre-incubation period except during spring snow melt when they were observed feeding c lose to f looded lakeshores. The proportion of time that males were vigilant during feeding bouts was greater when pairs were observed at the lakeshore than in the river outlets (22% versus 5%). Ma les were more vigilant than females and were vigilant when other males were not within sight. The frequency of male vigi lance was unrelated to the female fertile period suggest ing no paternity assurance function. Male vigi lance and female feeding and resting were not positively correlated and vigi lance and aggress ion were unrelated to date of clutch initiation. Ma les may use vigi lance to detect predators to enhance female survival in 'risky' habitat. II Table of Contents Abstract II Acknowledgements iv List of Tab les and Append ices V List of Figures VI Introduction 1 Paternity Assu rance Studies 1 Male Investment in Fema le Nutritional Status and Survival 2 Study Spec ies : Harlequin Duck (Histrionicus histrionicus) 6 Harlequin Duck Study Populat ion: Of Special Concern 7 Object ives 8 Methods 12 Study Si tes 14 Behavioural Observat ions 16 Clutch initiation in Relat ion to T ime of Arrival and S n o w Melt 17 Paternity R isks and Predator Interactions 18 Statistical Ana l yses 18 Resul ts 20 Behaviour in the Pre- lncubat ion Per iod 21 Test ing Predict ions 28 Discuss ion 38 Conc lus ions 42 Literature Ci ted 44 III A c k n o w l e d g e m e n t s Thank you to my family for their generous donations to my 'poor-student fund' and for supporting me through all things. Many thanks to my amaz ing partner Jacquel ine Kennel ly who was there at the start and is here at the end. My thanks to my friends and fellow ' C A C R e r s ' for reminding me to have fun through it al l : Devon Haag , S u s a n Paczek , Kym Wels tead , Richard Fe ldman and Jenn ie Chr is tensen. Thanks to my supervisor Kathy Martin for her patient and steady mentorship and to my committee: Jamie Smith, Peter Marshal l , and Bob Elner. Thank you in particular to Mandy Kellner, Shannon MacLach lan and Zig G a l e for excel lence in field ass is tance and for maintaining the giggles through cold and bugs. A final and huge thank you to Ian Goud ie who generously shared his project with me and without whom none of this would have been possib le. List of Tables and Appendices Table 1. Predict ions of hypotheses to explain the function of male vigi lance and aggress ion in the Harlequin Duck during the pre-incubation period 13 Table 2. Samp le s ize for behavioural observat ions on identified paired Harlequin Ducks , Fig Lake and Nipishish Lake, Labrador, 2000 and 2001 20 Table 3. M e a n (+ S E ) proportions of time spent in behaviours (Figure 3) for observat ions on paired Harlequin Ducks in fast-moving water at Fig Lake and Nipishish Lake and in s low-moving water at Fig Lake, Labrador 25 Table 4. M e a n (+ S E ) proportions of time spent in behaviours (Figure 5) in the pre-egg laying and egg laying period for observat ions on paired Harlequin Ducks , Fig Lake and Nipishish Lake, Labrador 32 Table 5. Resul ts of Spearman Rank correlation between the proportion of time females fed and rested (including preening) during observat ion per iods and the proportion of feeding and resting bouts that males were vigilant in s low- and fast-moving water while other birds were within and were not within sight of the focal pair at Fig Lake, Labrador (Figure 7 and Figure 8) 36 Append ix 1. Ethogram of Harlequin Duck Behaviour 49 Appendix 2. Number and hours of behavioural observat ions on identified paired Harlequin Ducks , Fig Lake and Nipishish Lake, Labrador, 2000 and 2001 50 Appendix 3. M e a n proportions of time spent in behaviours for two males (31, 3R) and six females (3H,3K,3L ,3P,3T,3V) at Fig Lake, Labrador, 2000 and 2001 51 Appendix 4. Clutch initiation in relation to arrival and snow melt for identified, paired female Harlequin Ducks breeding at Fig Lake and Nipishish Lake, Labrador 52 Appendix 5. Interactions of potential predators with Harlequin Ducks at Fig Lake, Labrador, 2000 and 2001 53 V L is t of F i g u r e s Figure 1. Locat ion of Fig Lake and Nipishish Lake study si tes, Labrador 15 Figure 2. Diurnal patterns of behaviours for paired female and male Harlequin Ducks in fast- and s low-moving water at Fig Lake and Nipishish Lake, Labrador 23 Figure 3. Effects of habitat type on the behaviour of paired Harlequin Ducks at Fig Lake, Labrador (*p<0.05, Table 2) 24 Figure 4. M e a n proportion of time spent in low- and high-level aggress ion (Appendix 1) by paired Harlequin Ducks at Fig Lake and Nipishish Lake , Labrador (Table 3).. . .26 Figure 5. Intensity of behaviour in relation to reproductive state of females for paired Harlequin Ducks at Fig Lake and Nipishish Lake, Labrador (p<0.05, Table 4) 31 Figure 6. Number of observat ions of high-level aggress ion (Appendix 1) directed at other individuals by paired male and female Harlequin Ducks at Fig Lake, Labrador over 159 hours of observat ion 33 Figure 7. Relat ionship between the proportion of time that paired female Harlequin Ducks rested (including preening) within 30 minute observat ions and the proportion of the rest bout that the male was vigilant in fast and s low moving water when other birds were within and were not within sight of the focal pair at Fig Lake , Labrador. 34 Figure 8. Relat ionship between the proportion of time that paired female Harlequin Ducks fed within 30 minute observat ions and the proportion of the feeding bout that the male was vigilant in fast and s low moving water when other birds were within and were not within sight of the focal pair at Fig Lake , Labrador 35 Figure 9. Intensity of female feeding and male vigi lance and aggress ion in relation to est imated date of clutch initiation (Appendix 4) for paired Harlequin Ducks at Fig Lake and Nipishish Lake, Labrador 37 VI Introduction Paternity Assurance Studies Socia l monogamy prior to incubation is assumed to be a consequence of male 'mate guarding' for many bird spec ies . This assumpt ion is based on a widely- invoked hypothesis that paired males should maximize their reproductive s u c c e s s by copulating with extra-pair females while ensuring they are not 'cuckolded ' by extra-pair copulation ( E P C ) efforts of other males (Trivers 1972). 'Mate guarding' or male accompaniment of the female before incubation of eggs is assumed to be the most common strategy by which male birds avoid 'cuckoldry' and assure their paternity (Birkhead and Mol ler 1992; Birkhead 1998). Support for this hypothesis is based on observat ions that males of many spec ies follow the movements of their mate more often than do females during the pre-incubation period. Further, in some spec ies , ma les maintain c loser proximity and follow their mate more often during the period that females are most fertile (Birkhead and Mol ler 1992). Studies that find such patterns in male behaviour conclude that males are attempting to deter the E P C attempts of their socia l partners or of other males to assure their own paternity. Ma les may a lso use aggress ion against intruding males (reviewed by Birkhead and Mol ler 1992) and vigi lance (Brodsky 1988) to minimize E P C attempts of other males while accompany ing females. S o m e authors have chal lenged the assumpt ion that mate guarding males can minimize extra-pair paternity ( E P P ) by deterring the E P C attempts of their soc ia l partners or of other males (Stamps 1997). Studies have increasingly recognized the role of females in the evolution of animal (Ahnesjo et al . 1993) and in particular av ian mating sys tems (Cunningham and Birkhead 1997; Gowaty 1994; 1997a; 1997b; Stutchbury and Neudorf 1997). Studies from the ' female perspect ive' have provided widespread ev idence that 1 females exert considerable control over paternity (reviewed in Petr ie and Kempenaers 1998; Stutchbury and Neudorf 1997) through rejection of within and extra-pair copulation attempts (Michl et. al 2002; Lifjeld and Rober tson 1992; Wagner 1991) and through the active solicitation of E P C partners (Martin and Hannon 1988; Double and Cockburn 2000; Kempenaers et. al 1992). Forced extra-pair copulat ion attempts are common in waterfowl (55 spec ies) and other spec ies (reviewed by McK inney and Evarts 1997) and s e e m likely to result in fertilization due to the intromittent c loaca of males. However, the few data from studies of wild populations suggest that most F E P C attempts do not result in paternity. For example, F E P C attempts result in no more than 5 % E P P in g e e s e (Dunn et a l . 1999). Further, there is little ev idence that increased mate guarding intensity is correlated with reduced E P P (Johnsen et a l . 1998; Kempanaers et a l . 1995; Mol ler and Ninni 1998; but s e e Chuang -Dobbs et a l . 2001). Thus , most studies show that mate guarding by males is an ineffective paternity assurance strategy (Kempenaers et a l . 1992 and 1995; Johnsen et a l . 1998; Stutchbury and Neudorf 1997). Male Investment in Female Nutritional Status and Survival Males may accompany females prior to incubation for reasons other than to assure paternity by minimizing E P C attempts (Fitch and Shugart 1984; Leffelaar and Robertson 1984; Martin 1984). For example, males may accompany females to invest in female reproduction and survival. Investment benefits males if it confers higher female and ultimately male reproductive performance. Investment may a lso benefit males if it indirectly assures their paternity. Alternatively, females paired to males that invest well may be unwilling to engage in extra-pair copulation or unwilling to form a pair bond with 2 another male. Thus , male investment in female reproduction and survival may be indirect ways in which males assure paternity. O n e way that males may invest in female reproduction is by enhancing female nutritional status. In many raptor and owl spec ies , females do not forage during the pre-incubation period and use food brought to them by their male partner for clutch formation. For example, in kestrels Falco tinnunculus (Meijer et a l . 1989) and osprey Pandion haliaetus (Green and Krebs 1995), males are solely responsib le for female nutritional status prior to incubation. In other spec ies , male investment in female nutritional status may be less vital but still an important component of male accompaniment . Consort ing males may use aggress ion and vigi lance to enhance their socia l partner's ability to acquire and assimi late food resources for clutch formation and incubation. Ma les may use aggress ion to defend resources such as feeding and resting areas. Aggress ive males may allow their socia l partners a c c e s s to preferred feeding and resting areas and to spend more time feeding and resting. Aggress ion by males has been assoc ia ted with increased foraging time of their socia l partners in common eider Somateria mollisima (Ashcroft 1976; Chr is tensen 2000) and barnacle geese (Black 2001). Similarly, male vigi lance for predators may decrease the time females spend vigilant al lowing females to spend more time feeding and resting. Alternatively, males may be vigilant to detect predators to ensure female survival without a subsequent effect of male vigi lance on time spent feeding and resting by females. There is ev idence that male vigi lance is a sexual ly selected trait. In the barnacle goose Branta leucopsis and grey partridge Perdix perdis, females prefer vigilant males (Choudhury and Black 1993; Dahlgren 1990) and vigi lance behaviour and levels of testosterone are positively related (Fusani et a l . 1997), suggest ing that male vigi lance is important to females as well as males. Male vigi lance has been assoc ia ted with increased female foraging time in greater snow geese Chen caerulescens atlantica (Gauthier and Tardiff 1991) and willow Lagopus lagopus and white-tailed Lagopus leucurus ptarmigan (Hannon and Martin 1992; Art iss and Martin 1995). However, in experimental removals of male willow ptarmigan prior to incubation, 'widowed' females did not have lower body weight, survival or reduced reproductive s u c c e s s than paired females (Martin 1984; Hannon and Martin 1992). These results suggest that male accompaniment has either no effect on the weight, survival or reproductive s u c c e s s of female wil low ptarmigan or that females can compensate for its loss. Benefits to paired females such as enhanced female nutritional status and survival are often a s s u m e d to be by-products of male attempts to assure paternity (Artiss and Martin 1995). S o m e studies address ' female-benefits ' hypotheses by predicting that females should follow males as frequently as the reverse (Birkhead and Mol ler 1992). Upon finding male-b iased following, authors reject ' female-benefits ' hypotheses as explanat ions for male at tendance. However, this may not be a valid prediction. If male at tendance functions to enhance female nutritional status and survival and is a sexual ly-selected trait, then males are under select ive pressure to follow their socia l mates. Therefore, this hypothesis a lso predicts male-b iased fol lowing. Indeed, this may explain the function of pursuit flights, which are assumed to be F E P C attempts, but may function in female assessmen t of the quality of male at tendance (Hoi 1997). 4 To my knowledge, only one study has attempted to dist inguish between the ' female nutritional status and survival ' and 'paternity assurance ' hypotheses to explain behavioural components of consortship such as male vigi lance and aggress ion prior to incubation in birds. Male white-tailed ptarmigan are vigilant prior to incubation and female feeding and male vigi lance are positively correlated. Further, males are more likely to become vigilant if females are feeding than if females are engaged in other behaviours. The intensity of male vigi lance does not change with female fertility, suggest ing that male vigi lance evolved to enable females to forage more effectively (Artiss and Martin 1995; Art iss et a l . 1999). Male investment in female nutritional status may be especia l ly important in spec ies in which breeding females face high energy demands over a short t ime period. The energet ic requirements for clutch formation in waterfowl are high compared to spec ies with altricial young (reviewed in Carey 1996). Within migratory waterfowl, transport of stored reserves from winter to breeding areas is facilitated by large body s ize such that smal ler waterfowl spec ies general ly depend more on resources in breeding areas for reproduction than larger waterfowl like geese and swans (A l isauskas and Ankney 1992). Spec ies with high wing loading, character ized by short wing length relative to body s ize and high speed flight, a lso have limited ability to transport stored reserves from winter a reas (Al isauskas and Ankney 1992). Therefore, females of smal l -bodied, migratory waterfowl spec ies with high wing loading may face high energy demands for clutch formation during the short time frame of the pre-incubation period. Male investment may facilitate the process by which females meet these energy demands and this may result in higher reproductive performance. Ma le investment in female 5 survival is likely important in spec ies with long-term pair bonds s ince males benefit by retaining paired status if females survive. Study Species: Harlequin Duck [Histrionicus histrionicus) Harlequin Ducks Histrionicus histrionicus are smal l , migratory s e a ducks (Tribe: Mergini) that form long-term, social ly monogamous pair bonds (Robertson and Goud ie s. 1999; Smith et a l . 2000). During breeding, Harlequin Ducks dive for benthic invertebrates in fast-flowing st reams, filling a feeding niche shared only with the Amer ican Dipper Cinclus mexicanus among North Amer ican birds. Pai r -bonds form in winter (Gowans et a l . 1997) and pairs migrate from coastal wintering areas to inland streams where breeding typically concentrates near lake outlets and inlets or other areas of high benthic invertebrate density (Robertson and Goud ie 1999; Hunt and Ydenberg 2000). Fema les are philopatric to natal s t reams and pairs and unpaired males show philopatry to breeding st reams and winter a reas (Robertson and Goud ie 1999; Rober tson et a l . 2000). Pai red males migrate to moulting a reas at the start of incubation and thus do not assist in incubation or brood rearing (Robertson and Goud ie 1999). Within breeding areas , females may exper ience high competit ion for food resources that are likely patchily distributed (Robertson and Goud ie 1999) and may potentially limit the production of young (Bengston and Ulfstrand 1971). Short wing length relative to body s ize and high speed flight character ize Harlequins Ducks as having high wing loading. Thus , it is likely that transport of stored nutrients from winter a reas is limited in this spec ies and females need to forage in breeding areas to form clutches. These combined characterist ics imply that female Harlequin Ducks exper ience high energy 6 demands during the pre-incubation period. Thus , there is good potential to investigate whether males invest in female nutritional status and survival. It has not been determined whether male Harlequin Ducks 'mate guard' . Thus , it is not known whether males maintain c lose proximity between pairs more often than females and whether males follow their social partners more often or more c losely when females are most fertile. Ma les position themselves between their soc ia l partners and other birds and this has been interpreted as a male 'mate guarding' strategy (Inglis et al . 1989). Alternative functions of male accompaniment prior to incubation have not been tested in the Harlequin Duck. Harlequin Duck Study Population: Of Special Concern Global ly, there are at least two disjunct populations of Harlequin Ducks . In the Paci f ic , Harlequin Ducks winter along the coast of North Amer i ca from Oregon to the Aleut ian Islands and along the coast of eastern Siber ia to northern J a p a n . In the Atlantic, a wintering and breeding population in Iceland may be disjunct from the rest of the Atlantic and thus may represent a third globally disjunct populat ion, but there are few data. Harlequin Ducks winter along the south coasts of Baffin Island and Greenland and the coast of eastern North Amer ica from Maine to southern Newfoundland (Robertson and Goud ie 1999). In eastern North Amer i ca , Harlequin Ducks breed throughout Labrador and Q u e b e c and along the northern peninsula of Newfoundland. Two main wintering areas are in southern Newfoundland and Maine. It was previously thought the eastern distribution represented one population and it was listed as endangered ( C O S E W I C 1990) based 7 on ev idence suggest ing range restrictions and a decl ining population est imated at approximately 1000 adults (Goudie 1991). Recent work has shown that at least 15 of 17 males implanted with satellite transmitters during breeding in northern Labrador and northern Q u e b e c moulted and wintered on the southwest coast of Green land . Migrant males (n=7) fitted with transmitters in eastern Q u e b e c spent the breeding and moulting season in Labrador and wintered in Maine (Brodeur et a l . 2002). Thus , it is now known that Harlequin Ducks breeding in eastern North Amer i ca compr ise two winter populations, one in eastern North Amer ica and one in Green land (Thomas and Robert 2001). Movement between the wintering areas does occur as s o m e wintering males implanted with satellite transmitters in Maine moulted in Green land (http: / / lavoieverte.qc.ec.gc.ca/faune/sauvagine/html/ap_captures_2001_e.html). It has not been determined whether these winter populations a lso represent separate breeding populat ions. Though Threatened status was recommended , the eastern wintering population was downlisted two categories to one of Special Concern in May 2001 based on ev idence of an increasing population est imated at 1800 adults (Thomas and Robert 2001). Objectives In this thesis, I descr ibe the behaviour of Harlequin Ducks in the pre-incubation period. S e c o n d , I attempt to distinguish between the ' female nutritional status', ' female survival ' and 'paternity assurance by minimizing E P C attempts' hypotheses to explain the function of male vigi lance and aggress ion. I a s s u m e that aggress ion and vigi lance are energetical ly costly to males and thus are maximized when benefits to males are greatest. Prior to testing predictions, I descr ibe the nature and ex is tence of pre-condit ions relevant to these hypotheses: 8 1. S e x dif ferences in behaviour. Greater proportion of t ime spent in vigi lance and aggress ion by males would indicate opportunity for females to benefit from male investment in these behaviours. Similarly, more time spent feeding and in energy-conserv ing behaviours like resting and preening by females may indicate the value of these behaviours to females in egg production. 2. De layed clutch initiation relative to arrival and availability of bare ground. I predicted that clutch initiation would occur soon after arrival and ground was bare of snow if females used primarily stored reserves transported from winter areas for clutch formation. Initiation soon after ground is avai lable would indicate that females do not rely heavily on nutrients acquired at breeding areas for clutch formation and initiation. Rel iance on stored reserves from winter a reas for clutch initiation would indicate minimal potential for male investment in female nutritional status to enhance the clutch initiation and formation ability of females during the pre-incubation period. De layed clutch initiation relative to timing of arrival and availability of bare ground may indicate that females need to feed at breeding areas to initiate egg laying. Thus , de layed clutch initiation would indicate the potential for male investment to enhance female nutritional status and ability to form and initiate clutches. 3. Select ive pressure for male 'mate guarding' through the potential for mate loss by males and paternity loss through E P C s . Fema le receptivity to E P C attempts would indicate that males risk paternity loss from E P C s . Thus , males may use aggress ion and vigi lance as 'mate guarding' strategies to minimize E P C attempts. Lack of female receptivity and a low f requency of E P C attempts suggest that male aggress ion and vigi lance are not used to minimize E P C s and may instead be used to enhance female nutritional status and survival. Incidences of female-control led mate switching indicate that males risk losing their socia l partners. Thus , males may use aggress ion to 'mate guard ' by deterring courtship by other males. Alternatively, if male vigi lance and aggress ion function to enhance female nutritional status and survival, then females may use levels of male vigi lance and aggress ion to a s s e s s male quality. Ma les may thus assure paternity indirectly by being aggress ive and vigilant to ensure that females are unwilling to engage in E P C or pair with other males. 4. The reaction of Harlequin Ducks to predators and the f requency of interactions to determine situations in which male vigi lance is most important for female survival. Harlequin Ducks may be most suscept ib le to either aerial or terrestrial predators. If Harlequin Ducks show little reaction to potential terrestrial predators and the f requency of interactions with terrestrial predators is low relative to interactions with aerial predators, then male vigi lance may more important further rather than near the shore. Under the "paternity assurance ' hypothesis, I predicted that male vigi lance and aggress ion would be highest during the egg-laying period (Table 1). Fema les of most avian spec ies store sperm for about 10 days; however, the viability of sperm in the female reproductive tract decl ines with time. Fertil ization of an ovum occurs about 24 hours prior to being laid and thus copulation occurr ing a few days before and during the egg-laying stage is more likely to result in fertilization than previous copulat ions. In birds, the last male to inseminate a female general ly has high probability of paternity. Thus , E P C s during this time can fertilize eggs and thus male vigi lance and aggress ion should be maximized during egg-laying if used to deter E P C attempts (Birkhead and Mol ler 1992). If used to enhance female nutritional status, levels of male vigi lance and 10 aggress ion would be highest when the rate of nutrient acquisit ion is highest or when female nutritional requirements for clutch formation are highest. The existence of a lipid threshold required to initiate egg-laying (Esler et a l . 2001) implies higher rate of feeding prior to egg- laying. However, nutritional requirements in other waterfowl spec ies are highest during the first few days of egg-laying (Al isauskas and Ankney 1992). Thus, levels of male vigi lance and aggression may be highest during the pre-laying stage or during the laying stage. Observat ions of highest vigi lance and aggress ion prior to egg-laying would indicate that both behaviours are not used by males to assure paternity by minimization of E P C attempts. Higher male vigi lance during egg-laying would provide support for both hypotheses. If males use vigi lance to ensure female survival, levels of male vigi lance should be constant throughout the season . Second ly , I predicted that males would be most vigilant in the absence of conspeci f ics if v igi lance functions to enhance female nutritional status and survival . Ma les would be vigilant only in the presence of other males if v igi lance is related to paternity assurance. Third, I predicted that males would be aggress ive to males and females if being used to enhance female nutritional status through the defence of feeding and resting areas. Ma les would be aggress ive only to other males if used to deter E P C attempts and assure paternity. Fourth, I predicted that male vigi lance would be positively related to female feeding and body maintenance behaviours like resting and preening if v igi lance functions to enhance female nutritional status. Thus, the proportion of feeding and resting bouts that males spend vigilant should be positively correlated in relation to the proportion of the 11 observat ion period that females spend feeding and resting. V ig i lance should be constant in relation to female feeding and resting if functioning to detect predators to enhance female survival. Finally, the ' female nutritional status' hypothesis predicts that females paired to vigilant and aggress ive males spend more time feeding and ultimately have higher reproductive performance. I used date of clutch initiation as an index of reproductive performance and predicted a negative correlation between date of clutch initiation and the proportion of time that females spent feeding and males spent vigilant and aggress ive. M e t h o d s In 1999, the Institute for Environmental Monitoring and R e s e a r c h ( IEMR) at G o o s e Bay, Labrador and the Atlantic Co-operat ive Wildlife Resea rch Network ( A C W E R N ) with support from the Innu Nation initiated a research project at Fig Lake in central Labrador to investigate the effects of military jet low-level flying on breeding Harlequin Ducks . Fig Lake was designated a treatment site and was overflown by low-flying N A T O military jets periodically throughout this study. Flights were infrequent during the pre-incubation period. Nipishish Lake is located outside of the Low-Leve l Training A r e a of the Department of National Defence, C a n a d a and in 2000 w a s establ ished as a control site for the overall project. 12 Table 1. Predictions of hypotheses to explain the function of male vigilance and aggression towards conspeci f ics in the Harlequin Duck during the pre-incubation period. Female Nutritional Status Female Survival Paternity Assu rance Male Vig i lance Higher f requency in pre- lay ing or laying s tage Cons tan t throughout s e a s o n H igher f requency in laying s tage H ighest f requency in a b s e n c e of conspec i f i cs Pos i t i ve correlat ion with fema le feed ing and rest ing H ighest f requency in a b s e n c e of conspec i f i cs Cons tan t f requency On l y w h e n other ma les in sight N o relat ion Nega t i ve correlat ion with t ime of c lutch initiation N o relat ion N o relat ion Male Aggress ion Higher f requency in pre- lay ing or laying s tage H igher f requency in laying s tage Di rec ted to m a l e s and f ema les D i rec ted only at other ma les Negat i ve corre lat ion with t ime of N o relat ion c lutch initiation 13 Study Sites The study was conducted from May to July 2000 and 2001 at Fig Lake (53°03'N, 63°09'W) and May to July 2001 at Nipishish Lake (54°02'N, 60°15'W) in central Labrador, C a n a d a (Figure 1). Fig Lake is located in the Lake Melvi l le Ecoregion of the Boreal Shie ld E c o z o n e about 80 kilometers southeast of Churchi l l Fal ls (Ecological Stratification Work ing Group, 1996). The area surrounding Fig Lake is character ized by large string bogs and open and c losed Black Spruce-dominated stands (Picea mariana). B a l s a m Fir {Abies balsamea) is found in low density on wetter soi l . The understory of upland, open stands is dominated by Car ibou Lichen (Cladonia spp) while feathermosses, Er icaceous shrubs such as Labrador T e a (Ledum groenlandicum Oedef), and severa l Vaccinium spec ies dominate the understory of c losed canopy stands. Wil low (salix spp.), Sweetga le (Myrica gale), A lder (Alnus spp), White Birch (Betula papyrifera), and Eastern Larch (Larix laricina) are found in riparian areas. Nipishish Lake is roughly 50 km north of Sheshatsu i and Happy Va l l ey -Goose Bay and is part of the Mecat ina River Ecoregion of the Ta iga Shie ld E c o z o n e (Ecological Stratification Work ing Group, 1996). The area around Nipishish Lake w a s burned about 20 years ago and is sparsely vegetated by patches of low shrubs and Black Spruce refugia in wetter a reas . Both lakes are surrounded by low-lying moraines and eskers and narrow to a shal low low-banked river no more than 50 meters at its widest. Downriver of the lake outlets, each river widens along its course into a number of areas of s low water. Both lake outlets and downriver of the outlets contain mid-stream boulders and is lands of various s izes the smal ler of which become submerged at high water levels during spring snow melt. 14 N \ f • A Quebec • • ' p i >.J V Scale (km) 0 100 200 J & \ ; % ^ Fig Lake 3 Figure 1. Location of Fig Lake and Nipishish Lake study sites, Labrador. 15 Behavioural Observations Harlequin Ducks were captured in mist nets p laced less than one metre above the water 's surface at the narrowing of the Fig Lake and Nipishish Lake outlets. W h e n water levels were high, mist nets were a lso placed in smal l emphemera l coves c lose to the shore of Fig Lake. Individuals were marked with one metal and one coloured tarsus band with a unique alphanumeric code. Rad io transmitters (Holohil Sys tems , C a n a d a ; 5-8 g, - 2 % body mass) were mounted dorsally on seven adult females, two in both years, during the pre-incubation period using the anchor suture method (Pietz et al . 1995). Reproduct ive data such as nest location, date of clutch initiation and nest fate were col lected for radio-marked females. Approximately 15 and 20 pairs bred at Fig Lake and Nipishish Lake respectively. Behaviour data were col lected throughout the day (0400-2100) on identified birds. Observat ion periods were 30 minutes duration using a focal , instantaneous sampl ing technique with a 15 second interval (Martin and Bateson 1986). Observat ions were made from blinds using a 20-60x spotting scope , binoculars, or unaided. Bands were read when individuals hauled out on midstream rocks to preen and rest. To avoid bias, pairs or individuals were chosen for observat ion such that s o m e observat ions began on swimming and diving birds for which identification was not initially possib le. In these cases , birds were monitored until identified, even if this entai led observing birds after the observat ion period was completed. Somet imes identification w a s not made and these data were not included in ana lyses . S o m e females could be identified by distinctive feather patterns. Sequent ia l observat ions on the s a m e pair were separated by at least 30 minutes. Observat ions less than 10 minutes and data on paired birds 16 with less than 10 observat ions were not included in the ana lyses . Observat ions in the hour after disturbance by humans or jet overflights were not included in the ana lyses. The proportion of t ime spent in each behaviour was est imated by dividing the number of intervals scored for each behaviour by the total number of intervals within the 30 minute observat ion period. The behaviours recorded were feeding, preening, resting, s leeping, aggress ion, courtship and copulat ion, vigi lance, alert, and locomotion (Appendix 1). I divided aggress ion into low (head nodding) and high (approach, rush, attack) levels and into aggress ion occurring during feeding and while hauled out. For observat ions col lected at the Fig Lake site, the presence of other birds within sight of the focal pair was noted during observat ion sess ions in order to test my predict ions regarding the effect of conspeci f ics on the intensity of male vigi lance. Targets of aggress ion by the focal pair were recorded to determine whether males use aggress ion for guarding their socia l partners to assure paternity or to enhance female nutritional status by defending feeding or resting areas . Clutch initiation in Relation to Time of Arrival and Snow Melt To estimate the number of days between arrival and clutch initiation I assumed that the arrival date of a female was the date of first observat ion. I est imated the date of clutch initiation to be 11 days prior to the start of incubation (Robertson and Goud ie 1999). I est imated the start of incubation to be one day prior to the first date that males were consistently observed without the female. Nests of radio-tagged females were monitored remotely to determine nest fate and hatch date. Clutch initiation was est imated as 41 days prior to est imated hatch date. In s o m e instances hatch date was est imated by ageing young using plumage characterist ics (Gol lop and Marshal l 1954, adapted by Cass i re r and Groves 1994). 17 Paternity Risks and Predator Interactions I recorded instances of mate switching and attempted to determine the sex responsible to determine whether males risk losing their soc ia l partners to other males. If so , males may use aggress ion to deter courtship by other males . Alternatively, males may use vigi lance and aggress ion to invest in female reproduction and survival as an indirect way to assure paternity by ensuring that females are unwilling to pair with other males or engage in E P C s . To determine whether males should guard mates to ensure paternity by deterring E P C attempts, I recorded observat ions of attempted extra-pair copulation at the Fig Lake study site. I a lso recorded the identity and gender of initiators and targeted individuals, descr ibed the response of the targeted individual and categor ized it as receptive or non-receptive. I attempted to identify and descr ibe the behaviour of predators at Fig Lake and descr ibed the reaction of birds during and outside of observat ion periods. Statistical Analyses Proport ion data were used in all statistical ana lyses of t ime spent feeding, resting, preening, and in aggress ion and vigi lance. The mean proportions of t ime spent in these behaviours were calculated for each individual. Ave rages and dif ferences between paired averages were normally distributed (Shapiro-Wilk p>0.05) so I used t-tests (two-tai led, a = 0.05). I pooled data for the two males for which I had two years of behaviour data because the mean proportion of time spent in behaviours differed by less than one standard error of the mean except for preen and rest for male 31 (Appendix 3). I pooled data for the 6 females for which I had two years of behaviour data because there were no significant dif ferences between years for any of the behaviour categor ies (feed: 18 p=0.12; rest: p=0.16; preen: p=0.18; aggress ion: p=0.49; paired t-tests; Appendix 3). I recorded observat ions on Harlequin Ducks in fast- and s low-moving water at the Fig Lake site and tested for differences in the mean proportion of t ime spent in behaviours between these habitats with paired t-tests. Prior to performing statistical tests related to pre-condit ions or predictions, I est imated time of day effects on behaviour by determining whether there were large differences or trends in the average proportion of time spent in behaviours over one hour intervals. I used independent sample t-tests to test for gender dif ferences in the mean proportion of time spent in behaviours. I used paired t-tests to determine whether males spent more time vigilant and aggress ive in the pre-laying or laying periods. In order to determine when the rate of food acquisit ion by females was highest, I tested for dif ferences between the pre-laying and laying periods in the proportion of time that females spent feeding. To account for the effect of time of season on the proportion of t ime spent in behaviours between reproductive states of the female, I used a genera l ized linear model (GLM) to perform covar iance analys is using observat ion periods as the sample unit. These ana lyses were performed on non-transformed proportions s ince these data did not become normally distributed with arcsine square root transformation. I a lso included habitat type (fast- and s low-moving water) as an independent categorical variable to determine whether reproductive state had a similar effect on the proportion of time spent in behaviours in both habitat types. 19 To test my prediction that male vigi lance is positively related to female feeding and body maintenance behaviour (resting and preening), I used non-parametr ic Spea rman rank correlation. For this analys is , I dist inguished observat ion periods according to whether other birds were within sight of the focal pair to test the effect of conspeci f ic presence on levels of male vigi lance. Finally, I calculated Spearman rank correlations between the average proportions of time that females fed and that males were vigilant and aggress ive with female clutch initiation date to test my prediction of a negative correlation. Results I col lected 454 observat ions (217 hours) on 16 paired males and 12 paired females in 2000 and 2001 (Table 1 and Appendix 2) during the pre-incubation period, early May to late June . Table 2. Sample size for behavioural observations on identified paired Harlequin Ducks, Fig Lake and Nipishish Lake, Labrador, 2000 and 2001. Study Site Number of Number of Habitat Number of Hours of Fema les Ma les (Relative Observat ions Observat ion Water Speed) Fig Lake 7 11 Fast 256 122.9 S low 88 41.2 Nipishish 5 5 Fast 110 53.2 Lake 12 16 454 217 20 Behaviour in the Pre-incubation Period Harlequin Ducks were sighted repeatedly at particular local ized sites but the location of sightings varied during the season . Prior to snow melt, Harlequin Ducks were only observed at lake outlets because the rest of the watercourse was f rozen. W h e n water levels rose with snow melt, Harlequin Ducks frequented smal l ephemera l coves along the f looded lakeshore of Fig and Nipishish Lakes . At peak water levels, most midstream boulders and smal l is lands were submerged and Harlequin Ducks were infrequently sighted in the lake outlets. W h e n water levels receded, Harlequin Ducks were rarely observed along lakeshores and for the remainder of the pre-incubation period, were more frequently sighted in the lake outlets than downriver of the lake outlets. A lmost all observat ions recorded in the lake outlets and most recorded downriver of the lake outlets were in fast-moving water near the middle of the watercourse. Observat ions recorded at the lakeshore were in s low-moving water. Feed ing general ly occurred within one to two metres of the water 's edge. Harlequin Ducks were observed flying downriver from the lake outlets at dusk and upriver into the lake outlets at dawn. There were few patterns in the mean proportion of time Harlequin Ducks spent in different behaviours over one hour intervals, except that birds rested more and fed less at dusk (Figure 2). Pa i rs and unpaired males often fed and rested in c lose proximity to each other (about 1 to 5 metres). Diving was often synchronous within pairs with females usually submerging about one second before males. Similarly, groups of birds consist ing of pairs and unpaired males often submerged within a few seconds of each other when in c lose proximity. Fol lowing feeding bouts in the lake outlets or downriver of the outlets, 21 birds typically hauled out of the water onto boulders or smal l , sparsely-vegetated islands near the middle of the watercourse and c lose to feeding areas . W h e n observed feeding along the lakeshore, birds hauled out c lose to feeding a reas but usually further away from shore than feeding areas. By averaging the means for the proportion of time hauled out for each bird, I calculated the mean proportion of t ime spent out of the water as 51 .0% + 2 for females and 48 .7% + 2 for males (n = 217 hours). At the Fig Lake site, males were vigilant more often while females fed in s low-moving water c lose to the shore than in fast-moving water at the middle of the watercourse (21% + 5 versus 5 % + 3, t = -4.46, p <0.002, paired t-test). Fema les spent more time feeding (40% + 5 versus 3 3 % + 2, t = -5.35, p < 0.002, paired t-test) and both males and females spent less time resting in s low water, near-shore a reas than in fast-moving water (Figure 3, Table 3). Sex Differences in Behaviour Males spent more time vigilant than females during feeding and resting bouts except for during rest bouts when pairs were hauled out near the lakeshore. The greatest difference in the f requency of vigi lance behaviour between males and females occurred when pairs fed in the s low-moving water of the lakeshore (Figure 3, Tab le 3). Pai red males and females both spent less than 3 % of the time engaged in aggress ion (Figure 4, Table 3). Most aggress ion was low-level 'head nodding' and usual ly both members of a pair head nodded at the s a m e time toward other pairs and unpaired males. Almost all high level aggress ion occurred during feeding (Figure 6). 22 Fast Moving Water slow Moving Water 4 6 8 10 12 14 16 18 20 22 4 6 8 10 12 14 16 18 20 22 Time of Day Figure 2. Diurnal patterns of behaviours for paired female and male Harlequin Ducks in fast- and slow-moving water at Fig Lake and Nipishish Lake, Labrador. 0.3 0.2 -0.1 -H 0.0 J o C 0.5 g '•c Q. 0.4 ] O Q_ CO CD Paired Males * • Fast-moving water O Slow-moving water 0 0 o o Paired Females o o o cn c '•£ a> <i> 5 £ o » _ 3 CO X 0 (0 _ P o E o o o E w 'E o t: o u _ < 0) o c ro 'at > o c ro > Figure 3. Effects of habitat type on the behaviour of paired Harlequin Ducks at Fig Lake, Labrador (*p<0.05, Table 2). 92 CD 3 Q3_ Q . CD J O CO *—' CD CD ^ =s g. Q . CD 5' c/>J?> o = 5 ¥ < ro E?" 1 ? CD =» 5T " c? CD - i ^ i £ ^ ro 3 W CD - Z3 _1 O II $ CD 3 CD CD CO O " CD _ T CD < o' c: —i c r CD CD CD _5 CD_ CD CO 0) Q . C ? 3 CD w 0) CO t-5* "D CD — i ' CD Q . CD CO •-CO Q . \f - i CD _J O CD CO c r CD _r CD < o' c —» cr CD £ CD CD o f CO 03 _5 Q . C0_ O CD CD CO C D ' o" CD _5 I—•* Q . if —i CD _3 O CD CO 5' cr g CL CD 2. CD 3 CD_ CD CO ro > CD O o 4^  O O O b o o CO o o ro o b o ro O o •o c_ pj 6' 3 O b o o to O O O 3 o o' 3 o b CD o b o b o o ro o oo o b o b o ro o oo o b o b o b o o oo TJ 3 CD 3 O b CO o o -»• ro CO o oo o b o o b '->• o o b b ro o ro A O b o o b oo 73 Fe > CD Fe co C/) CD CO Q_ ~ i CD CO « o' O O O CO co b CO co ro o o o b b b ro ro o ro o o o b ->J o ro o o o b b b cn cn o o p o b b cn ro CO o o o co CO b CO o CO o o o b b b ro ro o co o o o ro b cn ro o o o b b b o o o o b CO 00 o o o o ho ro fo CD CD ro o O o bo b bo —^  CO o Jig' E. oT c 0 o b co o b o b o CO o b o b o oo A O b o ro CD to « 8 o b o b o ro o b cn o CO o o b o to o b cn o b o ro A o b o p b < o c CD CD 1 + I co I m CD CO CD CD CD 1 + I col m co o CD CD CD 3 CD CD CD 03 23 1 + |C0| m 03 CO CD 03 13 1 + I col m co o 03 CD Q3_ CD II ICDI CD 0.002 5 0.000 Males Low Aggression High Aggression Males Females During Feeding Hauled Out Figure 4. Mean proportion of time spent in low- and high-level aggression (Appendix 1) by paired Harlequin Ducks at Fig Lake and Nipishish Lake, Labrador (Table 3). Clutch Initiation in Relation to Time of Arrival and Snow Melt I est imated date of clutch initiation for 15 nests of 9 females (Appendix 4). The median time between arrival and clutch initiation was 16.1 days (range 5 - 33; Appendix 3). I may have underest imated time between arrival and initiation for three females (3L, 3V, 3K) in 2000 as they may have arrived earlier but were not observed . However, I am confident of the accuracy of my estimation of the arrival date of female 3 P in 2001 because I was present on the study site before any Harlequin Ducks were observed. I found the nest of this female and am sure of my est imate of clutch initiation date and arrival to initiation interval of 7 days. Thus , at least one female w a s able to initiate egg-laying as soon as one week after arriving. The median time between clutch initiation and the first day of bare ground in riparian areas was 12 days (range 2-35). About half of the nests were initiated 0 to 18 days before all the snow had melted and the rest 2 to 11 days after all snow had melted (Appendix 4). 26 Paternity Risks I observed one instance of mate switching at the start of the breeding period at Fig Lake in 2000. A n unbanded male was observed following a pair (band ID 3 P female and 3B male) over two consecut ive days. These banded birds were paired in 1999. Both males were observed in s leep posture while swimming behind the female with the male 3B between the female and unbanded male. After the third day, the 3 P female had switched socia l partners and remained paired with the unbanded male for the remainder of the s e a s o n . Male 3B remained unpaired for the remainder of the breeding season and returned unpaired in 2001. Unpaired males fol lowed pairs closely, somet imes feeding and resting within 1 to 3 meters. In some c a s e s , up to three other unpaired males pursued a pair. I observed three E P C attempts by paired males and one by an unpaired male over 159 hours of observat ion at the Fig Lake site over both years. Pai red females responded aggressively to the male attempting to mount in each c a s e and none of the attempts resulted in successfu l mounting. At Nipishish Lake, a female was chased by two unpaired males while her mate was captured. O n e of the males persisted longer than the other and fol lowed her c losely for about 20 minutes. The female attacked the male on four separate occas ions during this time. On four other occas ions paired males left their partners to feed nearby (5-10 metres). Unpaired males approached the female during three of these occas ions and in each instance, the partner male returned to within 1 m of the female. 27 Predator Interactions Harlequin Ducks tilted their heads to the side in response to the presence of large birds overhead. Feed ing Harlequin Ducks did not react further to Osprey diving for fish within c lose proximity ( -10 metres) or to overflights of Herring Gul ls . Most predator interactions involved Bald Eag les at the outlet of Fig Lake. Harlequin Ducks reacted to Bald Eag les by f locking closely together in or out of the water. Harlequin Ducks also reacted to potential ground predators - one was unidentified and the other may have been a wolverine (Appendix 5) - by hauling out of the water. S o m e birds flew away in reaction to terrestrial predators. Harlequin Ducks reacted to a River Otter by flocking and swimming in a tight circle. Only two interactions involved directed movements by predators - a Northern Hawk Owl and Bald Eag le . Testing Predictions I predicted that the intensity of male vigi lance would be higher during egg-laying when females are most fertile if males use vigi lance to assure paternity. Under the ' female nutritional status' hypothesis, I predicted higher male vigi lance when female energy requirements for clutch formation or rate of nutrient acquisit ion were highest, which may have been in the pre-laying or laying stage. I predicted no change in the level of male vigi lance if used to enhance female survival. Ave raged over all observat ions in both habitats, males spent significantly more time vigilant during feeding bouts and females spent significantly more time feeding during the laying period than prior to egg-laying (Figure 5, Table 4). I performed covar iance analys is using G L M with date as a covariate and habitat type (fast- and s low-moving water) and reproductive state (pre-laying and laying) as 28 categorical var iables on observat ions recorded at the Fig Lake site (n = 361). Date had no significant effect on the proportion of time that females spent feeding (p = 0.95). Habitat type (p = 0.01) but not reproductive state (p = 0.16) had a significant effect on the proportion of time that females fed with higher feeding in s low-moving water in both the pre-laying and laying periods. The interaction term (habitat type x reproductive state) was not significant (p = 0.34) so females fed more in s low water irrespective of reproductive stage. T ime of season had a significant positive effect on the proportion of t ime that males spent vigilant during feeding (p < 0.001). The interaction term w a s a lso significant (p < 0.001). W h e n in fast-moving water, males spent on average 4 . 8 % and 4 . 5 % of the time vigilant during feeding in the pre-egg laying and laying per iods respectively. W h e n in s low-moving water, males spent on average 1 1 % and 2 6 % of the time vigilant during feeding in the pre-egg laying and laying periods respectively. Thus , males were only more vigilant when females were most fertile in s low water. I predicted that males would be aggress ive to other males and females if males use aggress ion to defend habitat resources but only to other males if aggress ion is used to assure paternity or maintain the pair bond. Ma les were observed aggressively approaching and rushing at pairs, unpaired males and males of other pairs (Figure 6). Ma les attacked other males more than females and all at tacks except one were on paired males. O n e paired male attacked a paired female but it w a s unclear whether this was an attempt to force copulat ion. The paired male forcefully pecked at the head of the paired female and may have been attempting to mount but w a s unsuccessfu l . Pai red females were never observed attacking other females or males of other pairs but 29 rushed at unpaired males. Pai red males were only observed rushing unpaired males while following their mate. I predicted that males would be vigilant only when other males were within sight if v igi lance functioned to assure paternity. I a lso predicted that male vigi lance would be positively related to female feeding and body maintenance behaviours if functioning to enhance female nutritional status. Pai red males were vigilant when no other birds were in sight of the focal pair. The presence of other birds did not affect the intensity of male vigi lance (Figure 7 and Figure 8). The magnitude of the correlat ions between the proportions of time that females fed and rested and the proportions of feeding and resting bouts that males were vigilant were low (Table 5). Three general patterns in the data can be observed. First, the higher proportion of time that females spent resting, the lower the proportion of the rest bout that males spent vigilant (Figure 7). Otherwise, there was little relationship. S e c o n d , the frequency of male vigi lance in fast-moving water var ied between about 0 and 2 0 % of feeding bouts when females fed for more than 2 0 % of the observat ion period. Third, males were vigilant for 0 % to 100% of feeding bouts when females fed in s low-moving nearshore water. I predicted that date of clutch initiation and intensity of male vigi lance and aggress ion would be negatively correlated if these behaviours function to enhance female foraging, clutch-forming ability and ultimately reproductive s u c c e s s . Under this prediction, I assumed that early-initiating females spend more time feeding and I tested this assumpt ion. There was no correlation between clutch initiation date and the mean proportion of time that females fed and that males were vigilant and aggress ive (Figure 9). 30 0.5 -i 0.4 0.3 fc! 0.2 0.1 H 0.0 • P r e - l a y i n g O L a y i n g # O Male V ig i lance Male V ig i lance Ma le F e m a l e Dur ing Feed ing Wh i l e Hau led Out Agg ress ion Feed ing Figure 5. Intensity of behaviour in relation to reproductive state of females for paired Harlequin Ducks at Fig Lake and Nipishish Lake, Labrador (p<0.05, Table 4). 31 Table 4. Mean (+ SE) proportions of time spent in behaviours (Figure 5) in the pre-egg laying and egg laying period for observations on paired Harlequin Ducks, Fig Lake and Nipishish Lake, Labrador. Behaviour Pre- laying Laying Ma les (n = 12) Mean + S E Mean + S E t P Vig i lance (Feeding) Vig i lance (Haul Out) Aggress ion 0.05 0.01 0.12 0 .02 0.02 0 .004 0.10 0 .02 0.12 0.01 0.02 0 .005 -3.80 0.22 0.97 0.003 0.83 0.36 Fema les (n = 8) Feed 0.32 0 .03 0.37 0 .03 -1.32 0.23 1 Pa i red sample t-tests; significant dif ferences in bold 32 CO £Z o CD CO j Q o CD - Q E 25 -20 -15 -10 -5 -0 25 i 20 15 10 5 0 25 i 20 15 10 5 0 Approach Pair Rush Pair Attack Unpaired Male • Female • Male Unpaired Male Unpaired Male Paired Male Paired Male Paired Male Paired Female Figure 6. Number of observations of high-level aggression (Appendix 1) directed at other individuals by paired male and female Harlequin Ducks at Fig Lake, Labrador over 159 hours of observation. Fast Moving Water c 03 1.0 0.8 0.6 CD > 0.4 -\ Q) CD g 0.2 CO CD CO 0.0 o • • o • • o • • • • • • •• • o o • • o • • ° . • V • • • • • • • • •• • ° o • Go « • ° « o • • o • • • • • o o o * oam o oo • o o o • • c r y 0.0 0.2 0.4 0.6 0.8 1.0 13 O CO D) C 1.0 Slow Moving Water CO 0.8 o c .9 0.6 o Q. O 0.4 -I CL 0.2 0.0 o, • Other Birds in Sight oother Birds Not in Sight o o o o o o o o ~ • o o a o CD o o o o o o • • o o o # o 8 ° o o o ocg 0.4 0.6 1 0.8 1.0 0.0 0.2 Proportion of Time Females Rested Figure 7. Relationship between the proportion of time that paired female Harlequin Ducks rested (including preening) within 30 minute observations and the proportion of the rest bout that the male was vigilant in fast and slow moving water when other birds were within and were not within sight of the focal pair at Fig Lake, Labrador. c 03 > CD i _ CD (7) 0 03 O c o o CD-CD 1.0 n 0.8 H 0.6 H 0.4 0.2 0.0 Fast Moving Water • Other Birds in Sight • o • OOther Birds Not in Sight CP o o o o o o - ° • • • ••• . o • cup o* m o o o • ° o * • • n ° cnxpcm) « x > o » o — c » c « > « r c > c r Q ^3 • • • w_ 3 o 00 D) 1.0 T J CD CD 0.0 0.2 0.4 Slow Moving Water CP • OCD« CO O O 0.6 o o • o °o o p 0.8 1.0 0.8 0.6 0.4 0.2 \ 0.0 - lo 9 Q o o o Q Q O o o» o o o o CP o o o • • o o <9 o o CD o o • o o o o 0.0 0.2 0.4 0.6 0.8 1.0 Proportion of Time Females Fed Figure 8. Relationship between the proportion of time that paired female Harlequin Ducks fed within 30 minute observations and the proportion of the feeding bout that the male was vigilant in fast and slow moving water when other birds were within and were not within sight of the focal pair at Fig Lake, Labrador. I 0) £ CD" CD < Q . ( Q O _ ** o CD o ho co o o o X CD cn o co o b cn o CD ro CO z co' o — _ c ? ^ c o 1 1 ! CD Q3 9- 3 =3 2 CQ CD ro co p I. o o ro o co cn o o CD O cn o cn o ro co CO co' X co g CD CD < Q.CD O W C 3 o CD cn co co o o ro ro oo A O © o CO co O CO CO ro ro CO D c CD _ CD < C ? * ~ CD Q3 9- 3 Z ! O ( Q CD ro p co o b o co CO CO o CD O 00 co o o ? oo ^ 3 -• O w a. s ® (/) J 2 o 00 CD 0) < 5' c CD 0) 3 1 + (0 m CD D) 3 1 + CO m n a) </> • 5 o < 5" CQ % CD CO o o < 5' CQ CD CD E g o Q_ O C CD 0 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.5 0.4 0.3 0.2 0.1 0.0 0.35 0.30 0.25 0.20 0.15 0.10 0.05 Female Feeding Male Vigilance Male Aggression rs = 0.05 p = 0.86 rs = 0.14 p = 0.62 rs = 0.23 p = 0.43 135 140 145 150 155 160 Estimated Julian Date of Clutch Initiation Figure 9. Intensity of female feeding and male vigilance and aggression in relation to estimated date of clutch initiation (Appendix 4) for paired Harlequin Ducks at Fig Lake and Nipishish Lake, Labrador. Discussion Habitat Use and Behaviour in the Pre-incubation Period Harlequin Ducks are typically assoc ia ted with fast-moving water during breeding. However, s low-moving water may be important habitat when water levels are high. I observed Harlequin Ducks in s low-moving water near f looded lakeshores during spring snow melt. Birds were infrequently observed in fast-moving water at the lake outlets and downriver of the lake outlets during the period of snow melt perhaps because most midstream refugia were submerged or because the current was too strong to allow efficient foraging. U s e of s low-moving water during the breeding season has been observed in Idaho during periods of high water (Kuchel 1977), irrespective of water levels In Iceland and northern Labrador (Inglis et a l . 1989; Rodway 1998) and has been inferred from diet (Robert and Cloutier 2001). It is a lso possib le that s low-moving water near lakeshores may be food-rich habitat for Harlequin Ducks during the pre-incubation period. Birds may have been responding to a peak in abundance of aquat ic invertebrates such as ephemeropteran larvae. I observed these larvae at lakeshores at high water levels during snow melt. These larvae were pelagic, abundant and perhaps required less energy to consume than benthic spec ies found in fast-moving water such as black fly larvae (Simuli idae). Harlequin Ducks may be vulnerable to predators at c lose proximity to shore. Harlequin Ducks have been observed to swim into fast currents to move quickly away from predators - a predator avo idance tactic that is not an option in s low-moving water (Robertson and Goud ie 1999). Midstream refugia may be an important habitat element for Harlequin Ducks because they provide safe resting a reas away from shores. Pai red 38 female and male Harlequin Ducks spent 5 1 % and 4 9 % of the daylight hours hauled out on midstream boulders or sparsely-vegetated islands c lose to feeding areas during the pre-incubation period. These values are comparable to other est imates of 4 7 % in Iceland (Inglis et a l . 1989) and 58 -59% in Alberta during the pre-incubation period (Hunt 1998). Simi lar observat ions exist of the use of midstream refugia near foraging areas for resting and preening in other areas (Bengtson 1972; Inglis et a l . 1989, Dzinbal and Jarv is 1984; Hunt and Ydenberg 2000). During the pre-incubation period, I observed Harlequin Ducks feeding in smal l , local ized sites. T h e s e were likely areas of high invertebrate density, such as the lake outlets and the inlets and outlets of smal l 's teadies' - a reas downriver but c lose to the lake outlets where the river widened into ca lm water (Richardson and M a c k a y 1991). Pai rs and unpaired males frequently fed c lose together and hauled out within one metre of one another near feeding sites. This pattern of habitat use suggests that resources are patchily distributed within watersheds and concentrated at lake outlets, suggest ing that they are a lso patchily distributed across landscapes, a conclus ion a lso suggested in a recent study in northern Labrador (Heath 2001). Female Nutritional Status and Survival versus Paternity Assurance Paired male Harlequin Ducks may lose their mates to unpaired males during the pre-incubation period. One unpaired male formed a pair bond with an already-paired female after following the female and her mate over a two day period. Though unpaired males fol lowed pairs, I observed only one E P C attempt by an unpaired male, suggest ing that unpaired males may be attempting to form pair bonds and not to engage in E P C . Extra-pair copulation attempts were not observed in a breeding population in Iceland (Inglis et a l . 1989). During copulat ion, female waterfowl rotate their pelvis to ach ieve c loacal contact and though the c loaca of male waterfowl is intromittent, contact and successfu l fertilization likely requires female cooperat ion. In the four E P C attempts that I observed, females responded aggress ive ly in each E P C attempt and it is doubtful that any in fertilization because none resulted in c loacal contact. G iven the low frequency of E P C attempts, the aggress ive response of females and that mate-switching occurred, I suggest that paired male Harlequin Ducks do not 'mate guard ' to minimize E P C attempts but instead assure paternity indirectly by ensuring that females do not pair with another male. Investment in female nutritional status and survival may be a male strategy to ensure female fidelity to the pair bond and thus male paternity. Aggress ion was infrequent and consisted mostly of 'head nodding' often by both members of a pair s imultaneously when intruders approached. Pai rs head nodded at unpaired males and at pairs while hauled out and during feeding bouts suggest ing that it is used to defend resting and feeding areas. High-level aggress ion occurred primarily during feeding and not resting bouts, despite observat ions of unpaired males and pairs hauled out within one metre on the s a m e rock (Inglis et a l . 1989; Hunt 1998). Ma les directed high-level aggress ion at other males, which supports the 'paternity assurance ' hypothesis. However, males may have also directed high-level aggress ion at pairs and because high-level aggress ion occurred during feeding, I suggest that males use high-level aggress ion to defend feeding areas. Fema les directed high-level aggress ion only to unpaired males. Pai red males fol lowed females when they rushed unpaired males. Fema les may have been communicat ing resistance to courtship by unpaired males or this may have been courtship behaviour by paired females to unpaired males. Ma les 40 rush females during winter courtship prior to the formation of pair bonds (Gowans et al . 1997). Most nests were initiated about two weeks after arrival at the study sites and about 12 days after bare ground was avai lable in riparian areas during the period of snow-melt. However, two females initiated nests within four days of bare ground had becoming avai lable in riparian areas. O n e of these females did so in both years of the study -within seven days of arrival in one year - and successfu l ly produced broods in each year. Thus , at least some females can initiate clutches soon after arrival and ground becomes avai lable, implying the use of stored reserves. However , the delayed clutch initiation for most females suggested they needed to feed before initiating laying. My results are supported by observat ions made in Alberta that female body weight is low on arrival and increases during the pre-incubation period (Hunt 1998; Hunt and Ydenberg 2000). The use of exogenous resources by female Harlequin Ducks during the pre-incubation period represents a pre-condition for male investment in female nutritional status. Male investment through vigi lance and aggress ion may al low females to forage more and initiate clutches early resulting in higher reproductive performance. I observed that paired male Harlequin Ducks spent more time vigilant during feeding and while hauled out than paired females when observed in fast- and s low-moving water, which supports the ' female nutritional status' hypothesis. Ma les were vigilant when other birds were not in sight suggest ing vigi lance was not used to assure paternity. Further, though males were more vigilant during the female 's fertile period in s low-moving water, the frequency of male vigi lance in fast-moving water did not increase with female fertility. Thus , the 41 f requency of male vigi lance was unrelated to the fertile state of females suggest ing that vigi lance does not function to assure paternity by minimization of E P C attempts. Ma les were vigilant about five t imes more frequently when females fed in s low-moving nearshore water than in fast-moving water. However, the proportion of time females spent feeding in fast-moving water and in sel f -maintenance in both habitat types was not positively correlated with the proportion of time that males spent vigilant. Further, neither the mean proportion of time that females fed or that males were vigilant were related to date of clutch initiation. Other factors such as female age or body condition on arrival may influence date of initiation. I observed interactions between Harlequin Ducks and terrestrial predators suggest ing that areas of s low-moving water c lose to shore represent 'risky' habitat. Therefore, it s e e m s likely that males were vigilant for predators to ensure female survival and were most vigilant in the 'risky' s low-moving water c lose to the lakeshore. C o n c l u s i o n s Aggress ion during the pre-incubation period was infrequent and likely was used to defend food and resting areas by females and males. Most females initiated clutches 12 days after ground was avai lable for nesting suggest ing that food on the breeding grounds was used to initiate and possibly form clutches. However , the proportion of t ime that females fed was not related to clutch initiation date suggest ing that female age or body condit ion on arrival determine egg-laying date. Extra-pair copulat ion attempts were observed only rarely and probably did not result in fertilization s ince none of the E P C s observed resulted in successfu l mounting. Ma les were vigilant while females fed, rested, and preened, particularly when females fed in s low-moving water c lose to shore. Males were vigilant when other males were not within sight. Ma le vigi lance was not positively correlated with female feeding or resting. I conc lude that male vigi lance is used to detect predators to enhance female survival particularly in potentially 'risky' nearshore areas . Literature Cited Ahnes jo I. 1993. The role of females in influencing mating patterns. Behavioral Ecology 4:187-189. A l i sauskas R, C D Ankney . 1992. The cost of egg laying and its relationship to nutrient reserves in waterfowl, p. 30-61 In B . D . J . Batt, A . D . Afton, M . G . Anderson , C D . Ankney , D.H. Johnson , J .A . Kad lec , and G .L . Krapu (Eds) Eco logy and management of breeding waterfowl. University of Minnesota P r e s s , Minneapol is, Minnesota. Art iss T, K Martin. 1995. Male vigi lance in White-tai led Ptarmigan, Lagopus leucurus: Mate guarding or predator detect ion? Animal Behaviour 49:1249-1258. Art iss T, W M Hochachka , K Martin. 1999. Fema le foraging and male vigi lance in White-tailed Ptarmigan Lagopus leucurus: opportunism or behavioural coordinat ion? Behavioral Eco logy and Sociobio logy 46:429-434. Ashcroft R E . 1976. A function of the pair bond in the C o m m o n Eider. Wildfowl 27 :101-105. Bengtson S - A , S Ulfstrand. 1971. Food resources and breeding f requency of the Harlequin Duck Histrionicus histrionicus in Iceland. O ikos 22:235-239. Birkhead T R , A P Moller. 1992. Spe rm competit ion in birds: evolutionary causes and consequences . Academ ic P ress , London. Birkhead T R . 1998. Spe rm competit ion in Birds: mechan isms and function, p. 579-622 In T R Birkhead and A P Mol ler (Eds) Spe rm competit ion and sexual select ion. A c a d e m i c P r e s s , London. Black J M . 2001. F i tness consequences of long-term pair bonds in Barnac le G e e s e : monogamy in the extreme. Behavioral Eco logy 12:640-645. Brodeur S , J - P L Sava rd , M Robert, P Laporte, P Lamothe, R D Ti tman, S Marchand , S Gil l i land, G Fitzgerald. 2002. Harlequin Duck Histrionicus histrionicus population structure in eastern Nearct ic. Journal of Av ian Biology 33:127-137. Brodsky L M . 1988. Mating tactics of male Rock Ptarmigan Lagopus mutus: a conditional mating strategy. An imal Behaviour 36:335-342. Carey C . 1996. Fema le reproductive energet ics, p. 324-374 In C . Ca rey (Ed) Av ian energet ics and nutritional ecology. C h a p m a n and Hal l , N e w York. Cass i re r E F , C R Groves . 1994. Eco logy of Harlequin Ducks in northern Idaho. Idaho Department of F ish and G a m e . Bo ise , Idaho. Choudhury S , J M Black. 1993. Mate-select ion behaviour and sampl ing strategies in 44 geese . An imal Behaviour 46:747-757. Chr is tensen TK . 2000. Fema le pre-nesting foraging and male vigi lance in C o m m o n Eider Somateria mollissima. Bird Study 47:311-319. Chuang -Dobbs H C , M S Webster , R T Holmes. 2001. The effect iveness of mate guarding by male Black-throated Blue Warblers. Behaviora l Eco logy 12: 541-546. C O S E W I C (Committee on the Status of Endangered Wildlife in Canada) . 1990. List of spec ies with designated status. Ottawa, Ontario. Cunn ingham E, T Bi rkhead. 1997. Female roles in perspect ive. Trends in Eco logy and Evolution 12:337-338. Dahlgren J . Fema les choose vigilant males: an experiment with the monogamous Grey Partr idge. An imal Behaviour 39:646-651. Double M, A Cockburn . 2000. Pre-dawn infidelity: Fema les control extra-pair mating in Superb Fairy-wrens. Proceed ings of the Roya l Soc iety Biological Sc i ences Ser ies B 267:165-170. Dunn P O , A D Afton, M L Gloutney, R T A l i sauskas . 1999. Forced copulation results in few extrapair fertilizations in R o s s ' s and Lesse r S n o w geese . An imal Behaviour 57 :1071-1081. Ecologica l Stratification Working Group. 1996. A national ecological framework for C a n a d a . Agriculture and Agr i -Food C a n a d a , Resea rch Branch , Centre for Land and Biological Resou rces Resea rch and Environment C a n a d a , State of the Environment Directorate, Ecozone Ana lys is Branch , Ottawa/Hul l , C a n a d a . Es ler D, J B Grand , A D Afton. 2001. Intraspecific variation in nutrient reserve use during clutch formation by Lesse r S c a u p . Condor 103:81—820. Fitch M A , G W Shugart . 1984. Requi rements for mixed reproductive strategies in avian spec ies . Amer ican Naturalist 124:116-126. Fusan i L, L Bean i , C Lupo, F Dess i -Fulgher i . 1997. Sexual ly se lected vigi lance behaviour of the Grey Partr idge is affected by p lasma androgen levels. Animal Behaviour 54:1013-1018. Gauthier G , J Tardif. Fema le feeding and male vigi lance during nesting in Greater S n o w geese . Condor 93: 701-711. Gol lop J B , W H Marshal l . 1954. A guide for aging duck broods in the field. Miss iss ippi F lyway Counc i l Technica l Sect ion. 14pp. Northern Prair ie Wildlife Resea rch Center : http : / /www .npwrc.usqs.qov/resource/tools/aqeduck/aqeduck.htm (Version 14NOV97) . Goud ie R l . 1991. Status report on the harlequin duck (eastern population) Histrionicus 45 histrionicus. Canad ian Wildlife Serv ice (Atlantic Region) report prepared for the Commit tee on the Status of Endangered Wildlife in C a n a d a , Ot tawa. G o w a n s B, G J Rober tson, F C o o k e . 1997. Behaviour and chronology of pair formation by Harlequin Ducks Histrionicus histrionicus. Wildfowl 48:135-146. Gowaty P A . 1994. Architects of sperm competit ion. Trends in Eco logy and Evolution 9:160-163. Gowaty P A . 1997(a). Pr inciples of females ' perspect ives in avian behavioral ecology. Journal of Av ian Biology 28:95-102. Gowaty P A . 1997(b). Sexua l dialect ics, sexual select ion, and variation in reproductive behavior, p. 351-384 In P A Gowaty (Ed) Femin ism and evolutionary biology. C h a p m a n and Hall , New York. G reen D J , E A Krebs. 1995. Courtship feeding in Ospreys Pandion haliaetus: A criterion for mate assessmen t? Ibis 137:35-43. Hannon S J , K Martin. 1992. Monogamy in Wil low Ptarmigan - is male vigi lance important for reproductive s u c c e s s and survival of f ema les? Animal Behaviour 43:747-757. Heath J P . 2001 . Factors influencing breeding distributions of Harlequin Ducks Histrionicus histrionicus in northern Labrador: a mult i-scale approach. Master of Sc ience thesis, Memoria l University of Newfoundland, Newfoundland, C a n a d a . Hoi H. 1997. A s s e s s m e n t of the quality of copulation partners in the monogamous bearded tit. An imal Behaviour 53: 277-286. Hunt W A . 1998. The ecology of Harlequin Ducks Histrionicus histrionicus breeding in Jaspe r National Park, C a n a d a . Master of Sc ience thesis, S imon Fraser University, Burnaby, British Co lumbia , C a n a d a . Hunt W A , R Ydenberg . 2000. Harlequins Histrionicus histrionicus in a rocky mountain watershed 1: background and general breeding ecology. Wildfowl 51:155-168. Inglis IR, J Lazarus , R Torrence. 1989. The pre-nesting behaviour and time budget of the Harlequin Duck Histrionicus histrionicus. Wildfowl 40:55-73. Johnsen A , J T Lifjeld, P A Rohde , C R Pr immer, H El legren. 1998. Sexua l conflict over fertil izations: Fema le Bluethroats escape male paternity guards. Behavioral Eco logy & Sociobio logy 43:401-408. Kempenaers B, G R Verheyen , B M V a n Der, T Burke, C V a n Broeckhoven, A A Dhondt. 1992. Extra-pair paternity results from female preference for high-quality males in the Blue Tit. Nature 357:494-496. Kempenaers B, G R Verheyen , A A Dhondt. 1995. Mate guarding and copulation 46 behaviour in monogamous and polygynous blue tits: Do males follow a best-of-a-bad-job strategy? Behavioral Eco logy & Sociobio logy 36:33-42. Kuchel C R . 1977. S o m e aspects of the behaviour and ecology of Harlequin Ducks breeding in Glac ier National Park, Montana. Master of Sc i ence thesis, University of Montana, Missou la , U S A . Leffelaar D, R J Rober tson. 1984. Do male Tree Swal lows guard their mates? Behavioral Eco logy & Sociobio logy 16:73-80. Lifjeld JT , R J Rober tson. 1992. Fema le control of extra-pair fertilization in Tree Swal lows. Behavioral Ecology & Sociobio logy 31:89-96. Martin K. 1984. Reproduct ive defence priorities of male Wi l low Ptarmigan Lagopus Lagopus: enhancing mate survivial or extending paternity opt ions? Behavioral Eco logy & Sociobio logy 16:57-64. Martin, K and S Hannon. 1988. Early pair and extra-pair copulat ions in Wi l low Ptarmigan Lagopus Lagopus. Condor 90:245-246. Martin P, P Bateson. 1986. Measur ing Behaviour. Cambr idge University P ress , Cambr idge. McK inney F, S Evarts. 1997. Sexua l coercion in waterfowl and other birds, p. 163-195 In P G Parker and NT Burley (Eds) Av ian reproductive tactics: female and male perspect ives. Ornithological Monographs No. 49. Meijer T, D M a s m a n , S Daan . 1989. Energet ics of reproduction in female Kestrels. Auk 106:549-559. Michl G , J Torok, S C Griffith, B C She ldon . 2002. Exper imental analys is of sperm competit ion mechan isms in a wild bird population. Proceed ings of the National A c a d e m y of Sc iences of the United States of Amer i ca 99:5466-5470. Mol ler A P , P Ninni. 1998. Spe rm competit ion and sexual select ion: A meta-analysis of paternity studies of birds. Behavioral Eco logy & Sociob io logy 43:345-358. Petrie M, B Kempenaers . 1998. Extra-pair paternity in birds: Explaining variation between spec ies and populations. Trends in Eco logy & Evolut ion 13:52-58. Pietz P J , D A Brandt, G L Krapu, D A Buhl . 1995. Modif ied transmitter attachment method for adult ducks. Journal of Field Ornithology. 66:408-417. R ichardson J S , R J Mackay . 1991. Lake outlets and the distribution of filter feeders: an assessmen t of hypotheses. O ikos 62:370-380. Robert M, L Cloutier. 2001. S u m m e r food habits of Harlequin Ducks in eastern North Amer i ca . Wi lson Bulletin 113:78-84. 47 Robertson G J , R l Goud ie . 1999. Harlequin Duck Histrionicus histrionicus. The Birds of North Amer i ca In: A . Poo le and F. Gil l (Eds)(No.466):The Birds of North Amer ica , Phi ladelphia. Rober tson G J , F C o o k e , R l Goud ie , W S Boyd . 2000. Spac ing patterns, mating systems, and winter philopatry in Harlequin Ducks . Auk 117:299-307. Rodway M S . 1998. Activity patterns, diet and feeding eff iciency of Harlequin Ducks breeding in northern Labrador. Canad ian Journal of Zoo logy 76:902-909. Smith C M , F C o o k e , G J Rober tson, R l Goud ie , W S Boyd . 2000. Long-term pair bonds in Harlequin Ducks . Condor 102:201-205. S tamps J . 1997. The role of females in extrapair copulat ions in social ly monogamous territorial animals p. 294-319 In P A Gowaty (Ed) Femin ism and evolutionary biology. C h a p m a n and Hal l , New York. Stutchbury B, DL Neudorf. 1997. Fema le control, breeding synchrony, and the evolution of extra-pair mating sys tems, p. 103-121 In P G Parker and NT Burley (Eds) Av ian reproductive tactics: female and male perspect ives. Ornithological Monographs No. 49. Thomas P W , M Robert. 2001. Updated C O S E W I C status report, eastern North Amer ican Harlequin Duck Histrionicus histrionicus - Commit tee of the Status of Endangered Wildlife in C a n a d a , Ottawa/Hul l , C a n a d a . Trivers R L . 1996. Parental investment and sexual select ion, p. 136-79. In B Campbel l (Ed) Sexua l Select ion and the Descent of M a n , Ch icago , Ald ine. Wagner R H . 1991. Ev idence that female razorbil ls control extra-pair copulat ions. Behaviour 118:157-169. 48 6P > CD a. C o £ c: CD * Q. r§ a 0) CT EL § = : O rt- CD o a> s'f CD co C L CD o CD CD O O c c CO CO CD CD p. p. CO < a? CO' 3 o 3 CO CD O CD 03 3 3 3 =!. CO 3 1 = « o QJ O 3. ° - g QJ 5. II OJ CD Q. ° EL g - CO 3 ' CO o > o CQ c CQ rtship —I CD rtship Loi ssic Win nig CD" < < CD CD o o o 3 - o o HI C T3 « 3. c CD w ji£ •5' o" 3 =T =f CD CD CD O O <= > ^ CO = QJ 0) " = CD 0) 0) 3 CD 3 o c CD C L 5 15 53-S CD ^ 5 o co co co o a Q3 O 3 CD ^3 CD "- - 3 T 3 O CD CX o CD Q} ED pj_ | § "O CD CU CD O 3 CO 3 O CD 3 Q_ CQ_ ' co' CD C L CP o CD 3 is CD X CD OJ C L 2 O CL > ^ > 73 73 3 o ti c c CD < " O CO CO O CD 3 3 " 3 " > 03 > > t §. s -o o o 3 o o' 3 < C/3 TI oT § «< 3 m > = CQ " O <Q co 3 " < ' CD CD c l ^ DD c c o' o' co co 73 73 c c 3 3 0) 03 3 ^ < o 0) —I C L CO 3 3 5-CD O i-f CD C L 0) c C L CD o -> 03 CO o o 2! 3 3 CD CO EJ" C T 3 =;• CD C L CO c o' 3 " CD 03 C L 15' CT C C CO C L CD s 5 O O CD | 3 CD =fi 2- 9: 3 ° S s if O CQ —h S = 03 O *i CD 3 C L O c CD o 3 " CD C L 3 CD O 7? 3 CO CO =tl CD •a -o 3 ' CO X 03 c_ CD C L O c < CQ oT O CD D c 3 3 ' co T l CD CD Q. 3 ' CO CO CD S Q3 ™ C L 3 o < CD 3 CD 3 #-»-W (/> 03 3 CD - 8 B 03 § Q. Q. 3 g . 1 5' 5' 1 co co o CD ~" °- o o « co 3 S o 3 0) S 3 < o CQ co 2. oT o i-3 3 - -o -CD CO C L § C 3 u, 3 - CD =r CQ 3 a? to CD " C L 3 « S-S CT CD 3 01 CO c 3 C L a CD -, CD ° - O ° 3 " 3 " ^ 0) CD T 3 C_ Q . QJ &•§ i . O ° " 0) < -1 o § 9L § CD ® 3 0 i -CO CD O-CT >< CO C L CD CO C L CD CD Q> C L 73 CD Cfl CD CD 3 CD CD CL W 73 CD 2. n o g CD "O < 0 CO CD 3" < 5' c o w CT CD CQ o 3 03 I = CD rt. 03 C C L o — 7 T 3 CD — 1° CD CO 3 CO o 3 S" CO o 01 3 3_ 3 ' co 3 o < CD 3 CD 3 3 c C L CD CO CO T3_ 03 CO 3 " CT 0J 3 CO I co co DO -n CD C — M Q) P , Q . OJ _. O 3 — . OJ 3 0) CT ? | a ® CD C L 3 CD C L ^ Q -o -6-S -o -h CD CD C L CD ^. Q. O CO | 3 " CD CD CD Q. 3 ' < CD CO CD CO 3 CQ 03 1—* CD co" CO c 3< o CD co c 3 C L CD CD CD O CD CO O O 3 c •o CD 3 C L Appendix 2. Number and hours of behavioural observations on identified paired Harlequin Ducks, Fig Lake and Nipishish Lake, Labrador, 2000 and 2001. Site Band Identification Y e a r Habitat Number of Hours of (Relative Observat ions Observat ion Fema le Male Water Speed) (Year Total) F ig Lake Nip ish ish Lake OS 0T 2000 Fast 47 22.5 S l o w 4 2 3 P Unhanded 2000 Fast 10 4.8 3 P 5 P 2001 S low 12 5.6 3L 5 N 2000 Fas t 27 13 S l o w 3 1.4 3L 5 A 2001 Fast 8 3.7 S l o w 5 2.1 3T 3 R 2000 Fas t 35 16 S low 4 2 3T 3 R 2001 Fas t 16 8 S l o w 3 1.5 3 H 31 2000 Fast 39 19 S l o w 7 3.5 3 H 31 2001 Fas t 15 7.3 S l o w 13 5.6 3 V OK 2000 Fast 22 11 S low 14 7 3 V Unbanded 2001 Fast 15 7.2 S low 4 2 3 K 5 K 2000 Fas t 14 7 S l o w 9 4.5 3 K 5 U 2001 Fas t 8 3.4 S l o w 10 4 2 R Unbanded 2001 Fast 10 4.8 6 V 6 B 2001 Fas t 30 14.8 j anded 6 G 2001 Fas t 32 15.4 6 H 6 J 2001 Fas t 11 4.9 6 P 6 M 2001 Fast 27 13.3 50 Appendix 3. Mean proportions of time spent in behaviours for two males (31, 3R) and six females (3H,3K,3L,3P,3T,3V) at Fig Lake, Labrador, 2000 and 2001. 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 Vigilance During Feeding Vigilance While Hauled Out m n ^ i L n ! _ 1 31 3R 31 3 R Feed 1.0 0.8 0.6 0.4 • I I in in ::l In in 3I 3 R Preen n 0.0 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 Rest 31 3 R Agonism Li 3I 3R 31 3R Feed 3H 3K 3L 3 P 3T 3V Preen 1.0 0.8 0.6 0.4 0.2 -o.o Rest I IS I I \l\ In1 In1 Ifi jfl 0.06 0.05 0.04 0.03 0.02 0.01 -0.00 3H 3K 3L 3 P 3T Agonism 3V 1 3H 3K 3L 3 P 3T 3V 3H 3K 3L 3 P 3T 3 V 51 Appendix 4. Clutch initiation in relation to arrival and snow melt for identified, paired female Harlequin Ducks breeding at Fig Lake and Nipishish Lake, Labrador. Year Sex Band ID Arrival Date of Clutch Initiation Arrival 1 (Days) Snow Melt 2 (Days) Snow Melt 3 (Days) Date of Hatch 2 0 0 0 F OS 6 M a y 2 J u n e ? >27 >16 >-4 Fa i led M 0 T 2 0 0 0 F 3 P 6 M a y 21 M a y 15 4 -18 1 Ju ly M u n b a n d e d 2001 F 3 P R 9 M a y 16 M a y 7 4 -8 25 J u n e * M 5 P 2 0 0 0 F 3 L R 15 M a y 28 M a y 13 11 11 11 Ju ly* M 5 N 2001 F 3 L 20 M a y 30 M a y * 10 18 6 Nes t predated M 5 A 9 M a y 2 0 0 0 F 3 T 6 M a y 6 J u n e 31 20 -2 Fa i led M 3 R 2001 F 3 T 9 M a y 27 M a y 18 15 3 Fa i led M 3 R 2 0 0 0 F 3 H 17 M a y 6 J u n e 20 20 -2 Fa i led M 31 14 M a y 2001 F 3 H 10 M a y 24 M a y 14 12 0 Fa i led M 31 2 0 0 0 F 3 V 19 M a y 21 J u n e * * 33 3 5 13 Fa i led M OK 2001 F 3 V 10 M a y 21 M a y 11 9 -3 Fa i led M u n b a n d e d 2 0 0 0 F 3 K 22 M a y 27 M a y 5 10 -12 Fa i led M 5 K 2001 F 3 K R 20 M a y 27 M a y 7 15 3 Nes t predated M 5 U 2001 F 2 R 17 M a y ? Fa i led M u n b a n d e d 2001 F 6 V R 13 M a y 28 M a y 15 2 2 Fa i led M 6 B 2001 F u n b a n d e d 17 M a y ? ? M 6 G 2001 F 6 H R 17 M a y 1 J u n e 15 6 2 Fa i led M 6 J 2001 F 6 P R ? ? Fa i led M 6 M * B a s e d on nest observat ion * *Poss ib le re-nest R F e m a l e s with radio transmitters Number of days between clutch initiation and 1 arrival 2first day of bare ground in riparian areas 3first day of bare ground in study area. 52 Appendix 5. Interactions of potential predators with Harlequin Ducks at Fig Lake, Labrador, 2000 and 2001. Potential Predator Behaviour of Predator Date Locat ion React ion Ba ld E a g l e Nor thern H a w k O w l W o l v e r i n e ? R i ve r Otter Unident i f ied on G r o u n d C i r c l ed a b o v e , M a y 11 2 0 0 0 did not at tack H o v e r e d about 5 met res a b o v e , did not at tack C i r c l ed a b o v e , did not attack C i r c l ed above , d id not attack C i r c l ed a b o v e , did not attack D i rec ted flight d o w n w a r d , no contact N o d i rec ted m o v e m e n t Outlet M a y 19 2 0 0 0 J u n e 7 2 0 0 0 J u n e 8 2 0 0 0 M a y 23 2 0 0 0 N o d i rec ted m o v e m e n t Not o b s e r v e d Outlet Out let Out let Out let M a y 13 2001 Outlet M a y 16 2001 M a y 20 2 0 0 0 Out let L a k e s h o r e C o v e Al l b irds f locked in water , s o m e birds hau led out in ' head up' alert posture to B a l d eag le c i rc l ing a b o v e ; T w o unpa i red m a l e s padd led in a tight c i rc le in ' head up ' alert posture Al l b irds f locked in water , s w a m in tight c i rc le in ' head up' alert posture A l l b i rds f l ocked and hau led out of water in alert posture A l l b irds f l ocked in water, s o m e birds hau led out in ' h e a d up' alert posture T w o pai rs d o v e , su r faced downs t ream under ove rhang ing vegetat ion at sho re Al l b irds f lock at midd le of wa te rcou rse , all hau l out in ' head up' alert posture, s o m e bi rds fly upr iver A l l b irds f lock in water , sw im in tight c i rc le in ' head up ' alert posture , s o m e birds fly upr iver Al l b i rds f lock at midd le of wa te rcourse , all haul out in ' h e a d up' alert posture, s o m e birds fly upr iver 53 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075097/manifest

Comment

Related Items