Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Factors affecting nest site selection and reproductive success of tundra nesting shorebirds Smith, Paul 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_2004-0073.pdf [ 7.25MB ]
Metadata
JSON: 831-1.0091257.json
JSON-LD: 831-1.0091257-ld.json
RDF/XML (Pretty): 831-1.0091257-rdf.xml
RDF/JSON: 831-1.0091257-rdf.json
Turtle: 831-1.0091257-turtle.txt
N-Triples: 831-1.0091257-rdf-ntriples.txt
Original Record: 831-1.0091257-source.json
Full Text
831-1.0091257-fulltext.txt
Citation
831-1.0091257.ris

Full Text

Factors Affecting Nest Site Selection and Reproductive Success of Tundra Nesting Shorebirds By Paul A. Smith B.Sc, Trent University, 2001 A THESIS SUBMITTED IN PARTIAL F U L F I L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF MASTER OF SCIENCE IN THE F A C U L T Y OF G R A D U A T E STUDIES DEPARTMENT OF Z O O L O G Y We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A November 2003 © Paul A. Smith, 2003 Library Authorization In present ing this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shal l make it freely avai lable for reference and study. I further agree that permiss ion 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 f inancial gain shal l not be a l lowed without my written permission. Pau l Smith 05/12/2003 N a m e of Author (please print) Date (dd/mm/yyyy) Title of Thes is : Factors Affecting Nest Site Select ion and Reproduct ive S u c c e s s of Tundra Nest ing Shorebirds Degree: Master of Sc ience Department of Zoo logy The University of British Co lumbia Vancouver , B C C a n a d a Year : 2003 Abstract I studied patterns of nest site choice and reproductive success in tundra nesting shorebirds at East Bay, Nunavut, to determine the factors that affect nest success in habitats with limited structural complexity. From 2000-2002,1 monitored the nests of five species: Black-bellied Plover (Pluvialis squatarola), Red Phalarope (Phalaropus fulicaria), Ruddy Turnstone (Arenaria interpres), White-rumped Sandpiper (Calidris fuscicollis) and Semipalmated Plover (Charadrius semipalmatus). For all species, I evaluated the influence of habitat, food availability, nest distribution and parental behaviour on nest success. I found strong patterns of non-random nest placement and clear evidence of habitat preferences. However, I found little evidence that variation in nest habitat was related to variation in success within or between species. Shorebirds did not prefer to nest in habitats where food was most abundant. Nest success was not consistently higher in preferred nest habitats. Instead, reproductive success may be related to the amount of parental activity near the nest. Red Phalaropes displayed considerable variation in nest site choice; they nested in grass/sedge marshes and also in rocky coastal habitats. I studied their distribution, behaviour and nest success to determine i f their use of coastal nest sites reflects a protective nesting association with an aggressive larid, the Sabine's Gull (Xema sabini). I found and monitored 29 phalarope nests with nearby (< 150 m) Sabine's Gull nests and 26 nests without nearby Sabine's Gulls. Coastal phalarope nests were nearer to Sabine's Gulls than expected by chance. Although I detected no habitat differences between coastal sites with and without Sabine's Gulls, only 3 phalarope nests were found in coastal areas without gulls. Approach experiments and incubation monitoring suggested that phalaropes without nearby gulls behaved more secretively. Phalaropes with nearby gulls had 17-20% higher hatch success in 2000 and 2001. In 2002, lemmings were scarce, predation was high, and phalaropes with nearby Sabine's Gulls did not have increased reproductive success. I found evidence that nest success may be influenced by parental behaviour and nesting associations, but the most significant variation in reproductive success was attributable to cycles of predators and lemmings. Table of Contents A B S T R A C T . . . II T A B L E O F C O N T E N T S I l l LIST O F T A B L E S V LIST O F FIGURES VI A C K N O W L E D G E M E N T S VIII C H A P T E R 1: G E N E R A L INTRODUCTION 1 C H A P T E R 2: E F F E C T S O F NEST SITE, F O O D , AND P A R E N T A L B E H A V I O U R ON SHOREBIRD R E P R O D U C T I V E SUCCESS 6 ABSTRACT 6 INTRODUCTION 6 METHODS 8 Study Area 8 Habitat Classification 9 Shorebirds and Their Predators at East Bay 9 Nest Searching and Monitoring 11 Invertebrate Abundance and Distribution 12 Nest Site Descriptions • 12 Random Site Descriptions 13 Statistical Analyses 13 Artificial Nest Experiments 14 RESULTS 15 Predators and Lemmings 15 Shorebird Nest Sites 15 Factors Affecting Hatch Success 17 DISCUSSION 20 Nest Microhabitat 20 Food Availability and Reproductive Success 21 Adaptive Nest Habitat Preferences? 22 Spatial Distribution of Nests 22 Antipredator Behaviour 23 Concealment and Defence 24 Activity Near the Nest 25 The Role of Predator-Prey Cycles 26 C H A P T E R 3: NEST SITE S E L E C T I O N O F T H E R E D P H A L A R O P E (PHALAROPUS FULICARIA): H A B I T A T AND SOCIAL F A C T O R S 27 ABSTRACT 27 INTRODUCTION 28 METHODS 29 Nest Finding, Monitoring and Habitat Evaluation 29 Predators and Defence 30 Indices of Density and Dispersion 30 Nesting Success 30 Artificial Nest Experiments 31 Flushing Distance Experiments 32 i i i Monitoring Incubation Behaviour 32 RESULTS 34 Principal Components Analysis 34 Nest Abundance and Distribution 34 Predators and Prey 35 Nesting Success 36 Artificial Egg Experiments 37. Flushing Distance Experiments 38 Incubation Behaviour 38 DISCUSSION 39 C H A P T E R 4: G E N E R A L DISCUSSION 45 Nesting Associations 46 Behaviour of Incubating Adults 46 Annual Variation in Predation 47 Implications 48 R E F E R E N C E S 50 T A B L E S 61 FIGURES 75 APPENDIX I - D E T A I L E D DESCRIPTION O F H A B I T A T T Y P E S 91 APPENDIX II - H A B I T A T C H A R A C T E R I S T I C S O F NEST SITES 97 APPENDIX III - NEST D E F E N C E BY SABINE'S G U L L S 98 iv List of Tables Table 1: Features of the habitats of East Bay, Southampton Island, Nunavut . 61 Table 2: The proportion of shorebird nests located in the six habitat categories at East Bay ..62 Table 3: Predicted group membership for shorebird nests, based on a discriminant function 2 2 analysis of habitat variables at scales of lm and 75m 63 Table 4: Habitat variables that discriminate between the nest sites of shorebirds at East Bay 64 Table 5: Indices of aggregation for the distribution of shorebird nests at East Bay 65 Table 6: Habitat variables that discriminate between nests and random sites for shorebirds at East Bay 66 Table 7: Mean distances to the nearest, or five nearest, shorebird neighbours for each species at East Bay 67 Table 8: Component loadings for principal components analyses of Red Phalarope nest sites (1 m2) and nest patches (75m2) 68 Table 9: Observed and expected nearest conspecific neighbour distances and indices of aggregation for nest distributions of Red Phalaropes in inland and coastal habitats 69 Table 10: Mayfield estimates of nest survival over a 19-day incubation period for Red Phalaropes nesting >150m away and <150m away from Sabine's Gulls 70 Table 11: Cox's proportional hazard regression models of nest survival for the Red Phalaropes of East Bay 71 Table 12: Mayfield estimates for daily survival of artificial shorebird nests at East Bay in 2002..... 72 Table 13: Cox's proportional hazard regression models for survival of artificial nests at East Bay in 2002 ..73 Table 14: G L M analyses of incubation behaviour of Red Phalaropes in 2002 74 Table A - l : The mean values for habitat characteristics of shorebird nests at East Bay 97 Table A-2: Behavioural responses of Sabine's Gulls to the intrusion of predators at East Bay , 98 v List of Figures Figure 1: The study area at East Bay, Southampton Island, Nunavut 75 Figure 2: The mean number of sightings per day of arctic foxes (Alopex lagopus), jaegers (Stercorarius parasiticus and S. longicaudus) and lemmings (Dicrostonyx groenlandicus), May-August, 2000-2001, at East Bay 76 Figure 3: Lateral and overhead concealment of shorebird nest sites and random sites at East Bay 77 Figure 4: The distribution of scores from discriminant function analyses of random sites versus nest sites, and failed versus successful nest sites for Black-bellied Plovers .78 Figure 5: The distribution of scores from discriminant function analyses of random sites versus nest sites, and failed versus successful nest sites for Red Phalaropes........ 79 Figure 6: The distribution of scores from discriminant function analyses of random sites versus nest sites, and failed versus successful nest sites for Ruddy Turnstones 80 Figure 8: The distribution of scores from discriminant function analyses of random sites versus nest sites, and failed versus successful nest sites for Semipalmated Plovers 81 Figure 9: Mayfield estimates of the daily mortality rate for shorebird nests at East Bay 83 Figure 10: The daily mortality rate for shorebird nests in each of the six habitat types at East Bay 84 Figure 11: The proportion of artificial nests taken by predators in each of the six habitat types at East Bay 85 Figure 12: The daily mortality rate for shorebird nests at East Bay placed in preferred versus non-preferred habitats 86 Figure 13: The mean volume (ml) of invertebrates per pitfall trap captured in each of the six habitats types at East Bay 87 Figure 14: A schematic diagram of the probes used to log nest temperature and determine incubation behaviour of Red Phalaropes. 88 Figure 15: The distribution of Red Phalarope and Sabine's Gull nests in 2000, 2001 and 2002 89 Figure 16: Daily rate of nest mortality for Red Phalaropes >150 m away and <150 m away from Sabine's Gulls in 2000-2002 90 vi Figure A - l : Mean sedge height for the pond edge habitats of East Bay, Southampton Island, ordered from coastal to inland 92 Figure A-2: Mean percent ground cover of various types for 30 1 m 2 random sites in a patch selected to best represent dry heath 93 Figure A-3: Mean percent ground cover of various types for 30 l m 2 random sites in a patch selected to best represent sedge meadow 93 Figure A-4: Mean percent ground cover of various types for 30 random l m 2 sites in a patch selected to best represent gravel ridge 94 Figure A-5: Mean percent ground cover of various types for 30 lm random sites in a patch selected to best represent moss carpet 95 Figure A-6: Mean percent ground cover of various types for 30 l m 2 random sites in a patch selected to best represent scrub willow ,. 96 Figure A-l: Mean percent ground cover of various types for 30 Inf random sites in a patch selected to best represent intertidal zone habitats 96 vii Acknowledgements First and foremost, I wish to thank Dr. Jamie Smith for his supervision, support and expeditious editing. Special thanks also go to Dr. Grant Gilchrist for supervision, encouragement and assistance from the beginning. I am indebted to Dr. Erica Nol, for sparking my interest in birds, and for her help and guidance along the way. I appreciate the contributions of my committee: Dr. Kathy Martin, Dr. Dolph Schluter and Dr. Judith Myers. I wish to thank my partner Sarah, for her patience and understanding, and my family: Jim, Carolyn, Andrew and Michael, for their support. My father Jim deserves special thanks for constructing the thermistor probes. I am also grateful for the statistical advice of Jane Reid. I also thank Dr. David Larson, Dr. Roger Pickavance and Krista Bowers, of Memorial University of Newfoundland, for their hard work identifying the invertebrate samples from East Bay. Finally, I send thanks to many co-workers, who contributed in many ways: Carolyn Saunders, Anna Hargreaves, Deb Perkins, Eric Davies, Iain Stenhouse, Rachel Bryant, Karen Truman, and Jean-Michel Devink. viii Chapter 1: General Introduction Survival and reproductive success in birds both depend on nest site choice (e.g., Burger 1985, Martin 1992, Wiebe and Martin 1998), and it is widely assumed that nest preferences have evolved to maximize fitness (e.g., Wiens 1989). Studies of nest site selection typically examine nest microhabitat (e.g., Clark and Shutler 1999, Martin 1998), but nest success may also be influenced by patterns of dispersion (Tinbergen et al. 1967), and/or the behaviour of the incubating parents (Cresswell 1997, Ghalambor and Martin 2002). Across a wide range of species, studies have shown that nest microhabitats are selected non-randomly (e.g., Colwell and Oring 1990, Rodrigues 1994, Clark and Shutler 1999). Patterns of nest placement are assumed to reflect the fitness benefits of nesting in preferred habitats (e.g., Cody 1981, Wiens 1989, Martin 1998). Nest success may be a major component of these benefits, as it is a dominant component of avian reproductive success. Nest habitat can affect nest success through the availability of food (Lack 1968, Martin 1987), nest microclimate (Walsberg 1985), and habitat specific predation risk (Martin 1995). Food limitation is widespread in passerines, and variation in food abundance is related to variation in fecundity (Lack 1968, Martin 1987, Murphy 1989). In arctic shorebirds, fecundity is inflexible because clutch size is fixed and there are few opportunities for re-nesting (Pitelka et al. 1974, Arnold 1999). However, temporal and spatial patterns of food abundance may influence reproductive success through effects on adult body condition (Carey 1980). Trade-offs between foraging and parental effort have been noted for sandpipers (Ashkenazie and Safriel 1979, Maxson and Oring 1980) and Lapwings (Walters 1990). Lower body condition is also correlated with increased nest abandonment, reduced nest attentiveness, and increased predation rates in shorebirds (Hegyi and Sasvari 1998). 1 Alternatively, nest sites may be selected because they provide a favourable microclimate (e.g., Horvath 1964, Walsberg 1981, 1985, Gloutney and Clark 1997, Wiebe and Martin 1998). Maintaining eggs at a suitable temperature and re-warming cooled clutches is energetically costly (Vleck 1981, Williams 1996), and these costs can limit reproductive success (Reid et al. 2000), especially in arctic environments. An unfavourable thermal environment can also reduce the viability of eggs (Norton 1972, Webb 1987). Pectoral Sandpipers preferentially line their nests with materials that insulate even when wet (Reid et al. 2002), suggesting that microclimate influences nest design. In the harsh climate faced by tundra nesting shorebirds, microclimate could also be an important component of nest habitat preferences. Predation is the primary cause of nest mortality for many birds (e.g., Ricklefs 1969, Martin 1992), and selection should favour a preference for nest sites that reduce predation risk. Risk of predation varies between nesting guilds and life history may have evolved in response to these differences (Lack 1968, Martin 1993, 1995). Within a guild, risk may be mitigated by adaptation to preferred habitat. For example, in the montane forests of Arizona, birds show a reduced risk of predation when nesting in their preferred microhabitats (e.g., Martin 1998,2001). In forest communities, where habitat is structurally complex, predation may act to limit the number of species in a nesting guild (Martin 1988a), or may promote differentiation between the nest sites of species within a guild (Martin 1988b). In structurally simple landscapes such as tundra however, there are fewer opportunities for specialized or unique nest preferences. Avoiding clutch loss may instead depend on factors such as incubator behaviour (e.g., Cresswell 1997, Ghalambor and Martin 2002) or nest distribution (Tinbergen et al. 1967). Risk of predation can be reduced when incubating adults display either secretive or aggressive behaviours. Secretive or cryptic behaviour near the nest reduces the risk of predation in many species (Wiebe and Martin 1997, Martin et al. 2000). In temperate cavity nesters, the rate of incubation feeding is related to interspecific patterns of nest predation (Martin and 2 Ghalambor 1999), and is reduced when predators are detected near the nest (Ghalambor and Martin 2002). For species without incubation feeding, predation risk could be reduced by limiting the number of incubation recesses, or by foraging far from the nest (Brown 1964, Pitelka etal. 1974) While cryptic behaviour can help birds avoid encounters with predators, many birds exhibit nest defence behaviours to directly reduce their chance of predation (review in Redondo 1989). In ground nesting birds, nests are easily accessible to predators and defensive behaviours may be particularly important. Nest defence in shorebirds varies in intensity among species, from scolding, circling and broken wing displays that distract predators or lead them away from the nest, to aggressive aerial attacks that may injure predators (Gochfeld 1984). Any form of nest defence can be risky (e.g., Jourdain 1936, Myers 1978, Brunton 1986), and shorebirds presumably balance the risk of injury to themselves with the rewards of successful nest defence (Montgomerie and Weatherhead 1988). Interspecific differences in antipredator behaviour can be related to this balance of risk and reward. Aggressive behaviour is most prevalent in large species, such as Black-bellied Plover (Pluvialis squatarola), where both members of the pair remain on the territory. These large species are typically monogamous, and they defend nests jointly, perhaps to reduce the risk to each parent (Larsen 1991). In contrast, smaller or uniparental species are less able to dissuade predators with aggression, and resort to distraction or cryptic behaviours (Larsen et al. 1996). When studying nest site selection, it is important to recognise that nests are placed not only in a habitat type, but also in a location with respect to neighbours. The spatial distribution of nests could affect reproductive success through a variety of mechanisms. If predators concentrate search effort in areas where they have found clutches previously, selection may favour hyperdispersion of nests (e.g., Tinbergen et al. 1967, Croze 1970, Page et al. 1983). Aggregations of nests can result in a lower rate of predation through a predator swamping effect 3 (Robertson 1973, Clark and Robertson 1979). For species with aggressive nest defence, colonial nesting may facilitate coordinated group response of early detection of predators (e.g., Andersson and Wiklund 1978, Larsen et al. 1996). Aggressive species may also affect the nest site selection of timid species. Many birds nest near more aggressive species in apparent protective associations (Durango 1949, Dyrcz et al. 1981, Haemig 2001). Many nesting associations have been reported for tundra breeding birds, with protector species ranging from large shorebirds (e.g., Paulson and Erckmann 1985) to Snowy Owls (Nyctea Scandiaca; e.g., Summers et al. 1994). While selecting a suitable nest site or adopting appropriate antipredator behaviour can improve nest success through the mechanisms described above, there are also other consderations. In tundra systems, almost complete breeding failure occurs in shorebirds and waterfowl in some years (e.g., Summers 1986, Summers and Underhill 1987, Underhill et al. 1993). Populations of lemmings (Lemmus and Dicrostonyx spp.) and voles (Microtus and Clethrionomys spp.) vary greatly in size over time, with fluctuations following periods of three to five years (Krebs 1964, Hanski and Korpimaki 1995). When rodents are abundant, they are the primary prey of arctic foxes (Alopex lagopus) and jaegers (especially Long-tailed Jaeger, Stercorarius longicaudus; Wiley and Lee 1998). Predator populations respond to this abundance both functionally and numerically (Wilson and Bromley 2001, Anglestam et al. 1984). When rodent densities decline, predators turn to the eggs and young of birds as alternative prey (Martin and Baird 1988, Bety et al. 2002). The cyclical variation in predation risk could influence shorebird nest site selection over evolutionary time scales, or on an annual basis i f birds are able to respond to the local abundance of predators. However, as breeding failure in years of high predation is catastrophic, this selective force may instead act on life history. Many factors can influence nest site selection and reproductive success in shorebirds. Given the structural simplicity of the tundra, we might expect habitat to play a lesser role in 4 determining nest fate in tundra habitats than it does in temperate forests. Instead, nest success may be more closely related to the behaviour of incubating adults or the spatial distribution of nests. As nest success is an important component of fitness, an understanding of the relative influence of habitat, behaviour and dispersion on nest success can further our understanding of life history evolution. In the chapters to follow, I examine the factors that affect shorebird reproductive success at East Bay, Southampton Island, Nunavut. Chapter 2 is a study of the relative influence of habitat, food and parental behaviour on the nest success of five species of shorebirds: Black-bellied Plover, Red Phalarope (Phalaropus fulicaria), Ruddy Turnstone (Arenaria interpres), White-rumped Sandpiper (Calidris fuscicollis) and Semipalmated Plover (Charadrius semipalmatus). In chapter 3,1 explore the importance of protective nesting associations by comparing the reproductive success and parental behaviour of Red Phalaropes nesting near and far from the aggressive Sabine's Gull (Xema sabini). In the final chapter, I bring together these inter- and intraspecific patterns, and attempt to show that parental behaviour and spatial distribution of nests are important determinants of reproductive success in tundra habitats. Throughout, I compare success between years with high and low populations of rodents, documenting the overriding influence of cyclic predation on shorebird nest success. 5 Chapter 2: Effects of Nest Site, Food, and Parental Behaviour on Shorebird Reproductive Success . ABSTRACT I studied the influence of nest habitat, food, nest distribution and parental behaviour on the reproductive success of tundra breeding shorebirds at East Bay, Southampton Island, Nunavut. From 2000-2002,1 monitored the nests of five species: Black-bellied Plover (Pluvialis squatarola), Red Phalarope (Phalaropus fulicaria), Ruddy Turnstone {Arenaria interpres), White-rumped Sandpiper (Calidris fuscicollis) and Semipalmated Plover (Charadrius semipalmatus). For each species, habitat differed between nest sites and random sites. In contrast, habitat differed between successful and failed nest sites only for White-rumped Sandpipers. Although nest success varied between species in all years, artificial nest experiments suggested that interspecific variation in predation rate was not related to habitat type. Shorebirds did not prefer to nest in habitats where food was most abundant. Instead, interspecific patterns of success are consistent with the hypothesis that reproductive success is related to the amount of parental activity near the nest. However, the factor with the greatest influence on nest success was the fluctuating predation pressure, related to the abundance of predators and lemmings. INTRODUCTION Nest sites influence avian ecology from the community level, where they affect species composition (Martin 1988a,b), down to the level of the individual, where they affect survival and reproductive performance (Burger 1985, Martin 1992). Many studies have demonstrated that nest sites are selected actively, such that microhabitat at nest sites differs from random sites (e.g., Colwell and Oring 1990, Clark and Nudds 1991). However, far fewer studies have.clearly 6 demonstrated that such nest microhabitat preferences are adaptive (Martin 1998, Clark and Shutler 1999); i.e., that individuals nesting in preferred microhabitats experience higher nest success in the long run. Reproductive success may be affected when nest habitats differ in food availability (Martin 1987), nest microclimate (Walsberg 1985) or habitat specific predation risk (Martin 1995). However success may also depend on factors other than habitat, such as nest distribution (Tinbergen et al. 1967) or the behaviour of the incubating parents (e.g., Cresswell 1997, Ghalambor and Martin 2002). Variation in food is known to affect fecundity and reproductive success (e.g., Lack 1968, Martin 1987). In shorebirds, clutch size is relatively fixed and opportunities for second broods are rare (Arnold 1999). However, variation in food can influence reproductive success by affecting incubator body condition (Carey 1980) or nest attentiveness (Hegyi and Sasvari 1998). Nest sites may also be selected for a favourable microclimate (e.g., Hprvath 1964, Walsberg 1981, 1985, Gloutney and Clark 1997, Wiebe and Martin 1998). The viability of eggs depends on nest microclimate (Norton 1972, Webb 1987). Incubation is energetically costly (e.g., Carey 1980, Vleck 1981, Williams 1996), and the rate at which clutches of unattended eggs cool can influence both the rate and duration of incubation recesses (Haftorn 1988, Reid et al. 1999). Predation is the primary cause of nest mortality for most birds (e.g., Ricklefs 1969, Martin 1992), and selection should therefore favour nesting strategies that reduce predation risk. In the montane forests of Arizona, nest sites in non-preferred microhabitats suffered increased predation (Martin 1998). Indeed, Martin has argued that habitat specific risk of predation may explain much of the variation in life history of passerines (Martin 1995). However, these effects of habitat on reproductive success take place where nest sites are limiting (Martin 1988a). Tundra breeding birds select nest sites on a simple landscape that lacks the structural complexity of forests. In such a landscape, suitable nest sites may not be limited, and high overlap of nest characteristics among species might be expected. However, even in habitats where nest sites are limited, antipredator or cryptic behaviours may have an overriding affect on nest success (Cresswell 1997, Ghalambor and Martin 2002). Shorebirds exhibit a variety of antipredator behaviours, including distraction displays and aggressive nest defence (Gochfeld 1984). While such behaviours should influence the risk of predation, risk might also be affected by activity near the nest, such as incubation recesses or nest visitation (e.g., Lyon and Montgomerie 1987, Wiebe and Martin 1997, Martin et al. 2000). As parental care and defence strategies are related to mating systems in shorebirds (Larsen 1991, Larsen et al. 1997), the influence of these behavioural factors on nest success can be evaluated with interspecific comparisons. I tested for patterns of nest preference by contrasting nests with random sites in 5 species of shorebirds, and looked for differences between the microhabitat of successful and unsuccessful nests to test for adaptive nest site choice. Relative abundance of food was measured to determine i f shorebirds preferentially nest in habitats with the most abundant prey. I also used an experiment with artificial nests to compare the frequency of predation across habitats. Finally, I used an interspecific comparison to test whether nest success is influenced primarily by habitat or by parental behaviour. METHODS STUDY A R E A Field work was conducted from late May to early August, 2000-2002, in the East Bay Migratory Bird Sanctuary, Southampton Island, Nunavut (63°59'N 081°40'W, Fig. 1). The 8 2 habitats in this 1,200 km sanctuary vary along an elevation gradient. Low lying areas within 1 km of the low tide mark support many brackish ponds (0.1 ha - 5 ha, <1 m deep). These saline areas support few plants and the sand and rock substrates are largely exposed. Inland from the coast, raised gravel beaches one to three metres high occur, and are separated by lower-lying areas with many freshwater ponds similar in size and depth to the brackish coastal ponds. In all areas, ponds dry as the nesting season progresses; large ponds decrease in size and shallow ponds (<30 cm) dry completely. Vegetation in low-lying inland areas is dominated by mosses (Campylium stellatum, Scorpidium scorpioides), sedges (Carex aquatilis, C. subspathacea) and grasses (Arctagrostis latifolid). Drier areas are dominated by dwarf shrubs (Dryas integrifolia, Salix arctica, S. reticulata) (Fontaine and Mallory, in prep.). HABITAT CLASSIFICATION I recognised six habitat types based on appearance, salinity, vegetation and elevation (Table 1). To quantify these classes, I evaluated the vegetation and substrate at 30 random sites within non-randomly selected habitat patches (Appendix I). The study plot was divided into a coastal and an inland portion, with intertidal areas, moss carpets and scrub willow comprising the coastal group, and sedge meadows, gravel ridge and dry heath comprising the inland types. SHOREBIRDS AND THEIR PREDATORS AT EAST B A Y Over 50 bird species have been recorded in the East Bay Migratory Bird Sanctuary (Abraham and Ankney 1986). Five shorebird species commonly nest in the study area: Red Phalarope, White-rumped Sandpiper, Ruddy Turnstone, Black-bellied Plover and Semipalmated Plover. Several other shorebirds occur and likely breed in the area, but are not common in the study plot: the Semipalmated Sandpiper (Calidris pusilla), Purple Sandpiper (Calidris maritima), Dunlin (Calidris alpind), Red Knot (Calidris canutus) American Golden-plover (Pluvialis dominica), Baird's Sandpiper (Calidris bairdii), Pectoral Sandpiper (Calidris melanotos) and Sanderling (Calidris alba). The five shorebird species nesting commonly in the plot vary in reproductive ecology. The Red Phalarope exhibits facultative polyandry. Males are the sole incubators. They show no territoriality, distraction displays or aggression towards predators (Tracy et al. 2002). Red Phalaropes feed on a variety of aquatic and terrestrial invertebrate prey on the breeding grounds, especially the larvae and adults of chironomids and tipulids (Kitchinski and Chernov 1973, in Tracy et al. 2002). White-rumped Sandpipers are polygynous and territorial; only females provide parental care (Parmelee 1992). They feign injury when predators approach and anecdotal information suggests that they select well concealed nest sites (Parmelee et al. 1968). They feed primarily by probing in moss for larvae (especially tipulids), but may also take spiders, beetles and adult tipulids from the substrate (Parmelee 1992). The Ruddy Turnstone is monogamous and territorial. Incubation is shared, but the male's contribution varies geographically and seasonally (Nettleship 2000). Turnstones are highly vigilant and aggressively pursue predators. They feed primarily on dipterans, which they pick from the substrate or find by overturning stones (Macdonald and Parmelee 1962). The Black-bellied Plover is large, territorial and monogamous with both parents incubating and caring for chicks (Paulson 1995). They are highly vigilant and defend their nests aggressively from predators through aerial attack and distraction displays (Drury 1961). They are visual foragers and peck invertebrates off the substrate, especially dipterans, beetles and spiders (Paulson 1995). Semipalmated Plovers are monogamous, territorial and both parents incubate (Nol and Blanken 1999). They are highly vigilant, and both parents take part in distraction displays or 10 scolding (Sullivan Blanken and Nol 1998). They feed primarily on dipteran larvae and adults (Baker 1977). Potential nest predators are abundant at East Bay. Parasitic Jaegers {Stercorarius parasiticus), arctic foxes and Herring Gulls (Larus argentatus) were observed regularly. Peregrine Falcons (Falco peregrinus), Long-tailed Jaegers, Glaucous Gulls (Larus hyperboreus), Sandhill Cranes (Grus canadensis) and Common Raven (Corvus corax) were also observed. The number of these predators observed daily (sightings/day) was used as an index of relative predator abundance between years. As the influence of generalist predators on avian nesting success may depend on the abundance of alternative prey (e.g., Summers 1986, Summers and Underhill 1987, Bety et al. 2002), daily observations of lemmings (Dicrostonyx groenlandicus) were noted. N E S T SEARCHING A N D MONITORING In 2000 and 2001, we searched 7 km of tundra, consisting of 4.5 km in inland habitats and 2.5 km 2 in coastal areas. In 2002, this area was expanded to 12 km 2 : 8.7 km 2 of inland and 3.3 km of coastal habitats. To avoid bias, search effort was allocated evenly across the study area. Search effort was greatest at the onset of the nesting period, from mid to late June. Nests were found through behavioural observation, flushing birds while walking and by two people dragging a 30 m length of 5 mm diameter rope. Nests were marked with a wooden tongue depressor placed 10-20 m from the nest at a random bearing (Reynolds 1985). Eggs were checked at regular intervals and floated to estimate developmental stage (Westerskov 1950, Sandercock 1997). The Mayfield method was used to estimate hatch success and daily mortality rates (Mayfield 1961). Nests hatching at least one chick were considered successful. Most hatch events were observed directly, but small eggshell fragments in the nest lining were also accepted 11 as evidence of hatch (Mabee 1997). Both abandoned and depredated nests were considered to have failed. Mayfield exposure days were terminated at the last active date for nests of unknown fate, and halfway between the last active and first inactive date for nests of known fate found empty (Manolis et al. 2000). Standard errors were calculated following Johnson (1979). Survival rates were compared using the program CONTRAST (Hines and Sauer 1989). INVERTEBRATE A B U N D A N C E A N D DISTRIBUTION To investigate spatial and seasonal patterns of the food resources available to shorebirds, terrestrial invertebrates were sampled throughout the nesting period in 2000-2001. Twenty plastic cups (diameter 11 cm, depth 8 cm) were placed in random locations in systematically chosen patches of each of the six habitat categories. An equal number of black and white traps were used in each habitat. They were placed flush with the substrate and filled to a level of 1 cm with propylene glycol. The distance between traps was at least 4 m. In 2000, all traps were deployed in the western half of the study plot. In 2001, traps were divided evenly between the 2000 sites and new sites in the eastern half of the study plot. Trap contents were filtered through a reusable coffee filter for collection. Traps were emptied three times in 2000: 8 July, 16 July and 24 July and three times in 2001: 7 July, 15 July, 23 July. N E S T SITE DESCRIPTIONS Nest sites of shorebirds were described at two spatial scales. First, I quantified ground 2 2 cover in a 1 m circle centred on the nest (nest site). Second, in a 75 m circle surrounding the nest (nest patch), I recorded the percent cover of each of the six habitat types. I also measured the distance ( ± l m ) to the nearest water and to the nearest dried pond edge (which would have been inundated at nest initiation) with a handheld rangefinder. To measure concealment, 1 used three 12 cm diameter white plastic disks marked with a black grid. Two disks were fastened at 12 right angles and placed atop a third, to provide an identical silhouette from four lateral directions and from overhead. Then, I estimated the proportion of markings obscured to the nearest 5%. Estimates of lateral concealment were made from north, south, east and west, at a distance of 5 m and a height of 40 cm (the approximate height of an arctic fox). Overhead concealment was estimated from eye level, directly above the nest. The height of the rocks or vegetation directly surrounding the nest (i.e. contacting the bottom disk) was measured to the nearest ±1 cm at the north, south, east and west edge. A l l nest habitat and concealment data were collected in late July. Nest coordinates (±3 m) were mapped using Arcview 3.2 (ESRI 1999). Distances between nesting neighbours were calculated with the "Nearest Feature" extension of Arcview (Jenness 2002), and spatial patterning for each species was measured with the Clark and Evans test with Donnelly's modification (Clark and Evans 1954, Donnelly 1978). The significance of deviations from random patterns was determined with Z-tests (Krebs 1989). For each nest, the density of neighbours was estimated as the mean distance to the five nearest shorebird neighbours. RANDOM SITE DESCRIPTIONS In each year, the study area was divided into a 50 x 100 m grid. Eighty grid intersections were selected at random. At each location, I tossed a stick backwards over my head and used the point of the stick as the random site. At these sites, habitat data were collected as for nest sites, 2 2 at 1 m and 75 m scales. Sites falling in ponds were not included in analyses. STATISTICAL ANALYSES A l l analyses were performed with SPSS 10.0.7 (SPSS Inc. 2002). Unless otherwise noted, P-values are two-tailed and means are displayed ± 1 SE. I used discriminant function 13 analysis (DFA) to test for habitat differences between: (1) nest sites of different bird species, (2) nest sites and random sites, and (3) successful and failed nest sites. An initial principal components analysis revealed that multicollinearity in both the 1 m 2 and 75 m 2 nest habitat datasets was weak. Therefore, I used the original percent cover and concealment variables (arcsine transformed) in the DFA's . Variables were entered stepwise using Wilk's Lambda method. Structure coefficients were obtained from discriminant analyses to determine correlations of variables with the discriminant functions. For classification, prior probabilities were set proportional to initial sample size. Chance corrected classifications were assessed for significance (Titus et al. 1984). ARTIFICIAL NEST EXPERIMENTS An artificial nest experiment was conducted in 2002 to assess relative predation pressure in the habitats of East Bay. "Clutches" of two Japanese Quail (Coturnix japonica) eggs were distributed in a stratified random design. As artificial nest locations were selected for another study (see Chapter 3 below), sample sizes vary between habitats. Eggs were laid out from 8-10 July, 2002, and were checked every third day. After three checks, missing or damaged eggs were replaced and the experiment was repeated (eggs laid out 17-19 July) using the same artificial nest sites. Quail eggs resemble small shorebird eggs (e.g., Red Phalarope, White-rumped Sandpiper) in coloration and size (circa 32 mm x 22 mm). The depressions used as nest sites were similar to the simple scrapes used by shorebirds. A coloured nail was hidden under the eggs to facilitate finding the nest i f eggs were taken by predators. Eggs were sterilised in 50°C water for 20 minutes. Nests with at least one egg damaged or removed were considered depredated. 14 A l l experimental procedures were approved by the Animal Care Committee of the University of British Columbia. RESULTS PREDATORS AND LEMMINGS The frequency of predator and lemming sightings differed substantially between years (Fig. 2). Lemmings were encountered frequently in 2000 and 2001 (0.56/day and 0.70/day, respectively), but none was seen in 43 days of observation in 2002. Fox encounters were rare in 2000 (0.04/day) and more common in 2001 and 2002 (0.19/day in both years). Jaegers were present in all years, but were more than three times as abundant in 2002 compared to 2000/2001. In all years, Parasitic Jaegers were >4 times more abundant than Long-tailed Jaegers. Although about 20 pairs of Herring Gulls bred in the study area in all years, they do not feed heavily on shorebird eggs (Allard, in prep.). I assume that Herring Gull predation on shorebird nests was infrequent and did not vary markedly between years. SHOREBIRD NEST SITES From 2000-2002, 189 shorebird nests were found and monitored: 21 of the Black-bellied Plover, 57 of the Red Phalarope, 63 of the Ruddy Turnstone, 24 of the White-rumped Sandpiper and 24 of the Semipalmated Plover. In addition, one Purple Sandpiper nest and three Semipalmated Sandpiper nests were found, but were not included in analyses. The number of nests found varied between years and sample sizes differ in some analyses because of missing information. No year effect was found in habitat measurements, so data are pooled across years where appropriate. 15 Four of the five common species exhibited clear preferences for one of the six habitat types, while the Red Phalarope used both sedge meadow and scrub willow habitats equally often (Table 2). Nests of the five common species were readily distinguished by DFA, indicating that preferred habitat differed between species. Classification based on the discriminant functions was 72% correct (a 48% improvement on chance, Tables 3-4, ZK = 15.3, P < 0.001). The first function groups nests based on the amount of moss and herbs at the site (1 m ), or the amount of intertidal habitat in the nest patch (75 m2), and accounts for 48% of the variability in nest habitats. The second function uses habitat type (gravel ridge and dry heath versus intertidal) to discriminate between the otherwise similar (sparsely vegetated, low cover) nest sites of Black-bellied Plover, Ruddy Turnstone and Semipalmated Plover. Means of nest site characteristics for each species appear in Appendix II. The concealment of the nests of the Black-bellied Plover, Ruddy Turnstone and Semipalmated Plover was similar to that of random sites (Fig. 3a,b). In contrast, Red Phalaropes and White-rumped Sandpipers had higher lateral and overhead concealment than random sites or than the three other species. The average heights of vegetation or rocks surrounding nests of the five species showed a similar pattern, with taller cover at the nests of White-rumped Sandpipers and Red Phalaropes (Appendix II). Species varied not only in habitat and concealment, but also in patterns of nest dispersion (Table 5). For Black-bellied Plovers, a sufficient number of nests for analysis were available in 2002 only; these nests tended towards a uniform distribution. The nests of Red Phalaropes were significantly clumped in 2000 and 2001, but not in 2002. The nests of both Ruddy Turnstones and Semipalmated Plovers were clumped in coastal habitats. Samples of White-rumped Sandpiper nests were small, and no significant patterns of dispersion were found. The nest sites of all species were readily distinguished from random sites in DFA's (Table 6). The accuracy of all classifications exceeded 80% (>30% improvement on chance). 36 Black-bellied Plovers had more cryptobiotic crust, lichen and exposed rock at the nest site, and less scrub willow habitat in the nest patch than random sites (Fig. 4a). Red Phalaropes selected nest sites with higher concealment and more herbs than random sites (Fig. 5a). Turnstones avoided dry heath and gravel ridges and selected sites with more exposed rock and cryptobiotic crust than random sites (Fig. 6a). Semipalmated Plovers nested in sites with more cryptobiotic crust than random sites and selected intertidal and moss carpet habitats (Fig. 8a). White-rumped Sandpipers selected nests with more willow and lateral concealment than random sites, and preferred to nest in sedge meadow habitats (Fig. 7a). While there were clear patterns of non-random nest placement for all species, there were few differences between successful and failed nest sites in the habitat variables measured (Figs. 4-8,b). Discrimination was weak for Red Phalaropes, Ruddy Turnstones and Semipalmated Plovers. Successful Black-bellied Plover nests tended to be in sites with less lichen, although they preferred nests with more lichen than random sites. Similarly, successful White-rumped Sandpiper nests had more dirt and moss and less sedge meadow habitat than failed sites, though they preferentially selected nest sites in sedge meadow habitats. FACTORS AFFECTING H A T C H SUCCESS Nest success varied markedly between years and between species (Fig. 9): Estimates for daily mortality were generally high in 2002, when lemmings were scarce and predators abundant. Semipalmated Plovers had similar success in 2001 and 2002 (no nests were found in 2000). White-rumped Sandpipers had high mortality in both 2002 and 2000. Daily mortality was lower for Black-bellied Plover, Ruddy Turnstone and Semipalmated Plover (species with biparental care) than for Red Phalarope and White-rumped Sandpiper (species with uniparental care). This difference in estimated daily nest mortality (DMR) between parental care systems is significant (DMRbjparentai = 0.027 ± 0.000, DMRuniparentai = 0.078 ± 0.000, X 2 , = 20.89, P < 0.001). 17 The mean incubation period of the biparental species at East Bay was 24 days, while that of the uniparental species was 21 days. Differences in the DMR's observed between species translate into large differences in hatch success, despite these differences in incubation length. The difference between hatch success of the uniparental species with the best success and the biparental species with the worst success was 69% in 2000, 24% in 2001 and 13% in 2002, but precision for hatch success estimates was low. Interspecific patterns of habitat use were poorly related to hatch success. The strong variation in success between species was unrelated to variation in predation pressure across habitat types. Red Phalarope nests were found in all habitats but gravel ridge (Table 2). White-rumped Sandpipers nested in dry heath and scrub willow habitats, though estimates of nest mortality for these habitats were not higher than for other habitats when the nests of all species were considered (Fig. 10). Only Red Phalarope and White-rumped Sandpiper nested in sedge meadow. This accounts for the higher mortality observed in this habitat type. Rates of loss of artificial nests do not suggest that predation pressure is higher in the habitats preferred by Red Phalaropes and White-rumped Sandpipers (i.e. Sedge Meadow and Scrub Willow, Fig. 11). Sedge meadow habitats had comparatively low rates of artificial nest lost in both the early/mid incubation period and the late/post incubation period. With both periods and all habitats combined, artificial nests had a daily mortality rate lower than that of real nests in the year of the experiment (DMRartjrlcia] nests= 0.040 ± 0.005, D M R r e a l nes.s(2002) = 0.058 ± 0.008, X2] = 3.76, P = 0.05). Density differences around real nests are unlikely to have caused the interspecific differences in nest success. While the mean distance to the 5 nearest nesting neighbours (any shorebird spp.) differs between species (Table 7, F4js4 = 12.9, P < 0.001), densities across the study plot are low (7.4 nests/km ). The mean distance from a nest to the nearest shorebird neighbour was >116 m for all species. Within species, neither the distance to the nearest 18 neighbour nor the mean distance to the 5 nearest neighbours differed between successful and depredated nests (All P's > 0.1). Within species, habitat preference was unrelated to nest success. Though all species had habitat preferences, nest success was not affected by whether individuals nested in preferred versus non-preferred habitats (Fig. 12). While Semipalmated Sandpipers, White-rumped Sandpipers and Red Phalaropes tended towards higher nest success in preferred habitats, Black-bellied Plovers and Ruddy Turnstones showed the opposite trend. As shorebirds may benefit from nesting near food resources, patterns of insect abundance could affect nest distributions. The relative abundance of invertebrates varied between habitats (Fig. 13). The highest volume of invertebrates was captured in dry heath habitats, and the lowest in gravel ridge and intertidal habitats. Trap contents in 2000 were dominated (in volume) by dipterans (Chironomidae, Muscidae, Scatophagidae), carabid beetles (esp. Pterostichus caribou Ball) and spiders (esp. Erigone, Alopecosa and Pardosa spp.). In 2001, many adult fritillary butterflies (Boloria spp.) were captured in dry heath and sedge meadow habitats (30% of volume in dry heath and 27% in sedge meadow). Few Boloria were captured in 2000. In both years, adult tipulids were most common in gravel ridge, sedge meadow and dry heath habitats (ml per trap per period, 2000/2001: gravel ridge = 0.04 / 0.04, sedge meadow = 0.03 / 0.01, dry heath = 0.02 / 0.01). Other dipterans were most abundant in scrub willow, intertidal and moss carpet habitats (ml per trap per period, 2000/2001: scrub willow = 0.52 / 0.33, intertidal = 0.35 / 0.38, moss carpet = 0.29 / 0.59). Beetles and spiders were most abundant in dry heath (ml per trap per period, 2000/2001: Carabidae = 0.32 / 0.37, Araneae = 0.50 / 0.31). Shorebirds did not preferentially use the dry heath habitats where invertebrate prey were most abundant. Also, invertebrates were abundant in moss carpet habitats, though few shorebirds nested there. 19 DISCUSSION As more than 40% of shorebird clutches at East Bay are lost to predators each year, natural selection should strongly favour nesting preferences that minimise predation. While numerous studies have demonstrated non-random nest site selection in birds, few studies have examined fitness consequences of these choices (Martin 1998). Evidence that birds can respond adaptively to spatial or habitat specific variation in nest success is rarer still, particularly for ground nesting birds (Wiebe and Martin 1998, Clark and Shutler 1999). While habitat may influence nest success, other factors such as nest defence or incubator behaviour may have more impact (Cresswell 1997, Ghalambor and Martin 2002), particularly in tundra environments, where the habitat is homogeneous and bird densities are low. NEST MICROHABITAT In this study, habitat features of nest sites differed markedly from random sites for all species. On the basis of habitat alone, nests of all species could be distinguished from random points with >80% accuracy. Four of five species preferred a single habitat type, while the fifth nested equally often in two habitats. Habitat characteristics of nest sites also differed significantly between species. Thus for all species, results suggest that habitat played a significant role in nest site selection. Despite this result, intraspecific variation in nest habitat was not related to variation in nest success. Successful and unsuccessful nests could be discriminated on the basis of habitat only for White-rumped Sandpipers. Variation in nesting habitat among species was also not related to nest success. Though rates of predation were 20-50% higher for White-rumped Sandpiper and Red Phalarope than for Black-bellied Plover, Ruddy Turnstone and Semipalmated Plover, these differences were not 20 related to differences in habitat specific risk of predation. Nor were patterns of nest success between species related to patterns of predation on artificial nests in species' preferred habitats. FOOD AVAILABILITY AND REPRODUCTIVE SUCCESS Nest sites may be chosen for efficient access to food, yet food is often overlooked in studies of nest site selection. In shorebirds, the availability of food could influence nest success by affecting foraging efficiency or the body condition of incubating adults (Carey 1980). The influence of food availability on foraging efficiency also depends on the distance between foraging locations and the nest. This information is limited for the shorebirds of East Bay, though most species seem to feed near the nest. Ruddy Turnstones defend territories, but may also use communal feeding areas (Nettleship 2000). Semipalmated Plovers may feed on the nesting territory, but may also establish separate territories for feeding (Nol and Blanken 1999). Black-bellied Plovers maintain large territories (ca. 50 ha) throughout incubation (Paulson 1995), and may feed near the nest. In my observations, female White-rumped Sandpipers and male Red Phalaropes generally fed within 100 m of the nest. While the abundances of likely shorebird prey varied between habitats, I found no evidence that shorebird nest site choice was constrained by the availability of food. Habitats with the greatest abundance of suitable prey were not preferred. Moss carpet habitats harboured an abundance of invertebrates, but few birds nested there. Adult tipulids, an important shorebird food (Baker 1977), were most abundant in gravel ridge habitats, though only Black-bellied Plover nested there. However, invertebrates collected from pitfall traps probably do not represent the full suite of prey items used by shorebirds. Insect larvae are an important food source for shorebirds, particularly for probing species such as White-rumped Sandpiper, Red Phalarope and Semipalmated Plover. The extent to which the habitat use of winged adult 21 invertebrates reflects the distribution of larvae is unclear. Also, flying insects, like the fritillary butterflies captured in 2001, may be attracted to pitfalls from a distance. ADAPTIVE NEST HABITAT PREFERENCES? Birds' choice of nest microhabitat is widely assumed to be adaptive, yet this assumption has rarely been tested (Clark and Shutler 1999). Adaptive site choice has been demonstrated for passerines breeding in montane forests where nest sites are limited (Martin 1988a); in this case, reproductive success suffers when nests are placed in non-preferred habitats (Martin 1998). Interspecific overlap in habitat preferences increases the risk of predation (Martin 1988a,b), favouring the coexistence of species with different preferred nest microhabitats. My study did not find nest sites to be limited, nor did shorebirds consistently show higher hatch success when nesting in preferred habitats. There were few habitat differences between successful and unsuccessful nest sites in the habitat variables measured. Though nest microhabitats differed between species, those species with more similar nest microhabitats (Black-bellied Plover, Ruddy Turnstone, Semipalmated Plover) did not suffer increased predation. This does not imply that habitat is unimportant in determining nest fate; merely that microhabitat variation in nest sites does not drive variation in predation. In a structurally simple environment such as the tundra, other behavioural factors may play a greater role in determining nest fate. SPATIAL DISTRIBUTION OF NESTS Patterns of nest distribution could affect reproductive success through a variety of mechanisms, such as density dependent predation (e.g., Tinbergen et al. 1967, Lack 1968) or group defence (Larsen et al. 1996). If predators intensify search in areas where they have encountered success, risk of predation may be positively correlated with nest density. In this study, distance to nearest neighbour (a measure of relative density) was not related to nest 22 success within or between species. The density of nests across the study plot was low (7.4 nests/km2) and mean distance to the nearest neighbour was >115 m for all species. Though thresholds for density dependence may be lower for the tundra ecosystem, studies in sub-arctic and temperate areas have not documented density dependent predation at such low densities (Goransson et al. 1975, Sugden and Beyersbergen 1986, Schieck and Hannon 1993, Lariviere and Messier 1998). Density may be negatively correlated with predation for birds that respond aggressively to predators and collaborate in group defence. Alternatively, incubators could benefit from the vigilance and early warning provided by near nesting neighbours (Neuchterlein 1981). While I know of no evidence for group defence in any of the shorebirds nesting at East Bay, Black-bellied Plovers and Ruddy Turnstones exhibit aggressive nest defence (pers. obs., Paulson 1995, Nettleship 2000), and could benefit from group mobbing. Semipalmated Plovers are highly vigilant, engage in distraction displays (Nol and Blanken 1999), and could benefit from early warning. Though these three species had the highest nest success, Ruddy Turnstones and Semipalmated Plovers showed significant clumping in all years and had the shortest nearest neighbour distances, while Black-bellied Plovers showed a random nest distribution and the longest nearest neighbour distances. ANTIPREDATOR BEHAVIOUR The antipredator behaviours of shorebirds are risky (e.g., Brunton 1986), and have evolved because of their success at dissuading or distracting predators (Gochfeld 1984). Because the risk of injury to the defender must be balanced with the reward of successful nest defence (Larsen et al. 1996), interspecific differences in nest success could be related to the level of nest defence. Both aggressive species (Black-bellied Plover and Ruddy Turnstone) had high nest success over all years, though they suffered increased predation in 2002 when predators were abundant and lemmings scarce. However, the highest shorebird nest success was found in Semipalmated Plovers. Semipalmated Plovers engage in vigorous distraction displays (both sexes), but do not attack predators (Nol and Blanken 1999). White-rumped Sandpipers also exhibit distraction behaviours (females only, Parmelee 1992), though they had the lowest nesting success of any shorebird at East Bay. Red Phalaropes show few nest defence behaviours (pers. obs., Tracy et al. 2002). CONCEALMENT AND DEFENCE Cover around the nest may conceal incubators and reduce the dispersion of scent (e.g., Bergerud and Gratson 1988), or may create a favourable microclimate (e.g., Walsberg 1985, Wiebe and Martin 1998). Concealment, however, may also be costly when predators threaten incubating birds, or when the success of antipredator behaviour depends on early detection of predators (Gotmark et al. 1995). In general, the benefits of a concealed nest site might be greater when faced with visual predators such as jaegers than when faced with olfactory predators such as foxes (e.g., Clark and Nudds 1991). The nests of Red Phalaropes and White-rumped Sandpipers had high lateral and overhead concealment, while Black-bellied Plover, Semipalmated Plover and Ruddy Turnstone selected nests with concealment similar to random sites. Red Phalaropes rely on concealment to escape detection from predators (e.g., Ridley 1980, Mayfield 1979). White-rumped Sandpipers exhibit distraction displays, but flush at close range (Parmelee et al. 1968). Both species may select concealed sites to avoid detection by predators. Black-bellied Plovers, Ruddy Turnstones and Semipalmated Plovers all flush at distances >20 m before engaging in antipredator behaviour (Paulson 1995, Nol and Blanken 1999, Nettleship 2000). These species may select exposed sites to facilitate early detection of predators. However, all three species selected nest sites with concealment similar to that of random sites, so concealment may simply be unimportant to site choice. 24 ACTIVITY NEAR THE NEST While direct interactions with predators may have direct consequences on nest success, a variety of other behaviours may be modified to reduce predation risk (Gresswell 1997, Ghalambor and Martin 2002). Activity near the nest increases the risk of predation in many species (Wiebe and Martin 1997, Martin et al. 2000). For example, the rate of nest visitation is positively related to the risk of predation in tundra nesting Snow Buntings (Lyon and Montgomerie 1987). In shorebirds, mating system and parental care strategy are important ultimate determinants of incubator behaviour (e.g., Pitelka et al. 1974). Uniparental incubators must leave the nest to feed, whereas a biparental incubator typically feeds while its mate is incubating. As numerous short feeding trips are less stressful for eggs and incubators than long bouts and absences (Reid et al. 1999), uniparental incubators must leave the nest more frequently than a biparental pair must change over. Though it varies with weather and time of day, the mean shift length of Red Phalaropes at East Bay, or of White-rumped Sandpipers on the Melville Peninsula, is less than 1 hour (Chapter 3, Cartar and Montgomerie 1985). In contrast, Semipalmated Plovers at Churchill, Manitoba, have mean shift lengths of approximately 200 minutes (Sullivan Blanken and Nol 1998). The length of incubation bouts is unknown in the Black-bellied Plover, but males and females brooded new chicks for bouts of 12 hours and 7 hours, respectively (Hussell and Page 1976, Paulson 1995). Incubation in Ruddy Turnstones is biparental: in early incubation mean bout duration of females is > 4 hours, but the contribution of males varies (Skipnes 1979 and Bianki 1967, in Nettleship 2000). A marked reduction in the number of trips on and off the nest may account for the higher nest success seen in bi- versus uniparental species. The importance of activity around the nest in determining success is corroborated by the observation that artificial nests had lower daily mortality than real nests. 25 T H E ROLE OF PREDATOR-PREY CYCLES Habitat and behaviour may contribute to nest site selection and reproductive success, but the single greatest influence on nest success across all species was fluctuating annual predation pressure. In tundra systems, predation of the nests of shorebirds and waterfowl is extreme about every fourth year when lemmings are scarce, and rodent predators turn to the eggs of birds (e.g., Summers 1986, Martin and Baird 1988, Bety et al. 2002). Populations of collared lemmings at East Bay probably cycle with a period of three to five years (Krebs 1964, Hanski and Korpimaki 1995). Lemmings were scarce in 2002 compared to 2000-2001. I assume that the dramatically increased predation in 2002 reflects increased predation by arctic foxes and jaegers. Many shorebirds can defend their nests from avian predators, but none are consistently successful at deterring mammalian predators (Larsen et al. 1996). Concealment or reduced activity at nests may also be ineffective at reducing fox predation, as they rely primarily on scent to detect nests (Clark and Nudds 1991). As nest loss in years of low lemming abundance is catastrophic for many species (e.g., Summers and Underhill 1987, Underhill et al. 1993), there may simply be no adaptations in nest site preference or parental behaviour that reduce fox predation in years after a lemming decline. 26 Chapter 3: Nest Site Selection of the Red Phalarope (Phalaropus fulicaria): Habitat and Social Factors A B S T R A C T Red Phalaropes (Phalaropus fulicaria) typically nest in grass/sedge marshes, but nests have also been noted in rocky coastal habitats. I studied the reproductive ecology of Red Phalaropes at East Bay, Southampton Island, Nunavut, to determine i f their use of coastal nest sites reflected a protective nesting association with an aggressive larid, the Sabine's Gull (Xema sabini). From 2000-2002,1 found and monitored 29 phalarope nests with nearby (< 150m) Sabine's Gull nests and 26 nests without nearby Sabine's Gulls. Coastal phalarope nests were nearer to Sabine's Gulls than expected by chance. No habitat differences were detected between coastal sites with and without Sabine's Gulls, but only 3 phalarope nests were found in coastal areas without gulls. Thermistor probes inserted in nests revealed that incubators with nearby gulls behaved less cryptically, taking more frequent and longer incubation recesses. In approach experiments, phalaropes with nearby gulls flushed earlier than those without gulls. In 2000 and 2001, hatch success was 17-20% higher for phalaropes with nearby Sabine's Gulls, but this effect was reversed in 2002. Sabine's Gulls are able to defend their nests from avian predators only, and arctic foxes (Alopex lagopus) were abundant and lemmings (Dicrostonyx groenlandicus) scarce in 2002. An artificial egg experiment in 2002 failed to demonstrate an effect of Sabine's Gulls on nest or egg losses. I suggest that phalaropes select coastal sites near Sabine's Gulls, but that this association is beneficial to phalaropes only in years when arctic foxes prey mainly on lemmings. 27 INTRODUCTION Predation is the primary cause of reproductive failure among birds (Ricklefs 1969, Bohning-Gaese et al. 1993), and selective pressure to mitigate this loss through adaptive nest site choice should be strong. Studies of the adaptive significance of nest site selection typically focus on habitat (e.g., Martin 1992, Martin 1998, Clark and Shutler 1999, chapter 2 above), but success may also depend on the distribution of nests (e.g., McKinney 1965, Tinbergen et al. 1967, Clark and Robertson 1979, Goransson et al. 1975). Spatial patterns of nests are usually considered for single species, but interactions between nesting neighbours are not limited to conspecifics. Many birds nest with more aggressive species in apparent protective associations (e.g., Durango 1949, Dyrcz et al. 1981, Haemig 2001). Such associations are common for Arctic and Subarctic regions, with putative protectors ranging from large waders to raptors (e.g., Popham 1897, Koskimies 1957, Portenko 1972, Paulson and Erckmann 1985, Blomqvist and Elander 1988). A timid bird nesting in association with a bold species can benefit when: (1) the bold species provides early warning which allows the timid species to react with cryptic behaviour ("information parasitism hypothesis", Neuchterlein 1981), and (2) the bold species defends an area around its own nest, indirectly protecting all birds nesting nearby ("defence parasitism hypothesis ", Dyrcz et al. 1981). To identify a protective association several conditions must be met: (1) birds must be able to recognise potential protectors and distinguish them from predators, (2) one species must actively select nest sites near the other and not merely use mutually preferred habitat, and (3) additional breeding success of birds nesting near apparent protectors must exceed any effects of predator swamping. The Red Phalarope is a Holarctic breeder that exhibits high variation in reproductive success (Tracy et al. 2002, this study). Red Phalaropes prefer to nest in grassy marshes (e.g., 28 Kitchinski 1975, Mayfield 1979, Ridley 1980). However, they also nest in rocky and exposed habitats within colonies of Arctic Terns (Hohn 1971, Hilden and Vuolanto 1972). Red Phalaropes at East Bay, Southampton Island, nest in both sedge marshes and rocky coastal areas. Many nests are in coastal areas in or near a Sabine's Gull colony. By assessing the distribution of phalarope and Sabine's Gull nests, recording behaviour of incubating adults, and comparing success for real and artificial nests with and without nearby gulls, I tested whether a protective nesting association exists between phalaropes and Sabine's Gulls. METHODS NEST FINDING, MONITORING AND HABITAT EVALUATION Phalarope nests were found and monitored, and nest habitat evaluated as described in Chapter 2. The conspicuous nests of the Sabine's Gull were found primarily during laying and checked at 3-day intervals. These nests form a loose colony, with most nests <100 m apart (Stenhouse et al. 2001, this study). Laying dates of both species were recorded directly or back-calculated from hatch dates to estimate the timing of breeding between species (incubation for Sabine's Gulls lasts 21-22 days, Stenhouse et al. 2001; for Red Phalaropes, incubation lasts 19 days, this study n = 4 nests). The number of nest habitat variables was reduced through principal components analysis 2 2 (PCA). Separate analyses were conducted for percent cover data at 1 m and 75 m . Variables with low or zero correlations in the correlation matrix were removed prior to analysis. Principal component scores were generated only for readily interpretable components with eigenvalues >1 (Guttman 1954). 29 PREDATORS AND DEFENCE An index of relative abundance (sightings/day) was generated for arctic fox, lemmings and jaegers (see Chapter 2). The mean flushing distance of Sabine's Gulls in response to the approach of predators is 160 m (Appendix III, Stenhouse et al., submitted). Arcview GIS 3.2 software (ESRI 1999) was used to plot nest locations (±3 m accuracy), and all areas within 150 m of any Sabine's Gull nests were defined as "with gulls". Sabine's Gulls at East Bay nest primarily in coastal habitats, so areas without gulls were furthered classified as "coastal without gulls" or "inland" based on habitat characteristics. INDICES OF DENSITY AND DISPERSION Using the GIS map, the minimum distance to the nest of a conspecific was determined for each Red Phalarope nest. These nearest neighbour distances were used to investigate spatial patterning in inland and coastal nesting phalaropes (Clark and Evans 1954). To assess whether phalaropes tend to nest near the gulls, the distances between coastal Red Phalarope nests and Sabine's Gull nests were compared to those between coastal random points and Sabine's Gull nests using Mann-Whitney (/-tests. Random points were generated with the "Generate Randomly Distributed Points" extension for Arcview (Lead 2002). (/-tests were also used to compare the distance to the nearest neighbour of any species for inland and coastal Red Phalarope nests. Both this measure and relative density (nests/km searched) were used to assess nesting densities around phalarope nests in inland and coastal habitats. NESTING SUCCESS Red Phalarope nest success was calculated as in Chapter 2. To examine factors affecting hatch success, I used Cox's proportional hazard regression. I included habitat in the models as 9 9 the first and second principal components of the 1 m and 75 m habitat analyses. I also included 30 the proportion of exposure days where a phalarope nest was within 150 m of an active Sabine's Gull nest (arcsine transformed) as an index of the protection from Sabine's Gulls. The distance between phalarope nests and their nearest neighbours, as well as the number of neighbours within specified distances were tested as covariates. Predictors were assessed individually for significant Wald statistics, and final models were created through a backwards removal likelihood ratio analysis with a remo al criterion of a = 0.05. ARTIFICIAL NEST EXPERIMENTS To assess the protection that Sabine's Gulls provide, an artificial nest experiment was conducted in 2002. Two artificial nests per site were placed at 16 sites in inland habitats ("inland"), 16 sites in coastal habitats >400 m away from any gull nest ("coastal without gulls") and around 16 nests of Sabine's Gulls ("coastal with gulls"). "Clutches" of two Japanese Quail eggs were placed at 50 m and either 100 m (n - 8 sites) or 200 m (n = 8 sites) from randomly selected points or Sabine's Gull nests. An additional six clutches of four eggs each per treatment were also deployed to ensure that using two eggs did not affect the likelihood of predation. The artificial nests were laid out from 8-10 July, 2002, to coincide with mid-incubation in Sabine's Gulls, and were checked every third day. After three checks, missing or damaged eggs were replaced and the experiment was repeated (from 17-19 July) using the same artificial nest sites. By this second round of checks, most Sabine's Gulls had finished nesting and abandoned their territories. Consequently, differences in predation pressure between treatments in this second round should be independent of any protection provided by Sabine's Gulls. Nests with at least one egg damaged or removed were considered depredated. Separate Mayfield estimates of daily survival were calculated for each treatment in each round, and for the 50 m, 100 m and 200 m protected nests. Survival rates were compared with z-statistics and multiple rates were compared using the program CONTRAST (Hines and Sauer 1989). Cox's 31 proportional hazard regression was used to investigate the variables affecting survival time of artificial nests. Because I predicted a priori that the hazard functions for protected, unprotected and inland nests would differ in shape, separate analyses were mn for each treatment. Final models were created through a backwards stepwise likelihood ratio analysis with a removal criterion of a = 0.05. FLUSHING DISTANCE EXPERIMENTS To compare responses of phalaropes to an approaching threat with and without nearby gulls, I approached all nests on foot in 2002. Responses of shorebirds to human intrusion resemble those elicited by other predators (e.g., Armstrong 1956, Gochfeld 1984, Reid and Montgomerie 1985). I approached nest at a steady pace from a random bearing. The distance at which the bird flushed was recorded (±1 m) with a handheld rangefinder. Flush distances for phalaropes with and without nearby gulls were compared with a Mann-Whitney (/-test. I used linear regression to assess the influence of nest concealment and incubation stage on flush distances (e.g., Gotmark et al. 1995, Knight and Temple 1986). To minimise disturbance, approach experiments were combined with regular nest monitoring visits. MONITORING INCUBATION BEHAVIOUR Thermistor probes attached to Hobo Temp-XT data loggers (Onset Instrument Corporation, Pocasset, Massachusetts) were placed in the nests of incubating Red Phalaropes with and without nearby gulls. The probes consisted of a 2 mm, 10 KQ Curve-G thermistor on a 15 m, 24 A W G cable, and a 10 K Q (±1%) reference resistor loop on a 2.5 mm stereo jack (Fig. 14). The tip of the probe (3 mm x 6 mm) was centred in the nest and was level with the top surface of the eggs. It was in contact with the brood patch when the bird was incubating. Readings were taken every 30 seconds, allowing 66 hours of continuous records. Loggers were 32 placed 15 m from the nest in a camouflaged, waterproof housing and the cable between them was buried or concealed. The entire placement procedure lasted less than 10 minutes. Loggers were placed in completed clutches only. Several nests failed before loggers could be deployed. The ten logger systems were distributed opportunistically between nests with and without gulls. Nests were visited every third day to retrieve data. Observations (24 h total) on two nests before and after deployment of the logger system revealed that the probes had no effect on incubator behaviour and that they accurately captured departures of the incubator to within ±30 s. Temperatures neared 40° C when the incubating bird was present, and dropped sharply to ambient temperature when the incubator departed. These temperature traces were used to calculate the proportion of a 24 h period that nests were incubated (% nest attentiveness), the number of recesses/24 h and the mean recess duration per 24 h period. Variables affecting these incubation parameters were examined with general linear models. Nests were included as a random factor and whether a nest was <150 m from a Sabine's Gull nest was included as a fixed factor. Covariates examined included Julian date (and date ), incubation stage (and stage ) and daily average wind-chill (estimated by a weather station in the centre of the study plot). To include habitat and concealment variables (which have no variance within nests), I ran separate analyses with means of the incubation parameters for each nest. Where random or fixed factors were significant individually, they were included in all subsequent tests of cpvariates. Individual predictors were tested with Wald statistics and final models were created through a backwards likelihood ratio analysis (removal criterion of a = 0.05). 33 R E S U L T S PRINCIPAL COMPONENTS ANALYSIS Through principal components analyses, I reduced the number of nest habitat variables to two for the 1 m 2 data and two for the 75 m 2 data. At the 1 m 2 scale, the final matrix included eight variables (Table 8, a), and the first two components captured 26% and 18% of the variability in the original data, respectively. PCI loads highly negatively on moss, sedge and grass (hydrophytic) and positively on lichen and avens (xerophytic). I interpret it as a measure of moisture, with low scores representing moist sites. PC2 loads negatively on rock and dirt, and positively on willow, lichen and organic crust. This component captures the amount of exposed substrate; willow, lichen and dead moss are found in habitats with little exposed substrate. At the 75 m 2 scale, the first two components explained 21% and 19% the variation respectively (Table 8, b). PCI loads positively on dry heath and sedge meadow habitats and negatively on pond margin habitats. I interpret it as a measure of patch size and habitat homogeneity; dry heath and sedge meadow habitats consist of large patches, while pond margins are edge habitats. PC2 loads positively on moss carpet and scrub willow and negatively on the intertidal zone habitat type, and reflects position within the study area. A high PC2 score indicates a central location, where moss carpet and scrub willow are common. NEST ABUNDANCE AND DISTRIBUTION Over three years, Red Phalaropes arrived at the study area on 8-12 June, several days later than most other shorebirds nesting at East Bay. Snow cover at the time of their arrival varied from 95% in 2000 to 8% in 2001. Sabine's Gulls arrived on 3-11 June, before the arrival of phalaropes in each year. The number of phalaropes present in the study area at late-34 courtship/early-incubation (circa 1 week after arrival) differed between years, with 26.6 birds/km2 in 2000, 23.1 birds/km2 in 2001 and only 6.8 birds/km2 in 2002. I found 55 Red Phalarope nests: 25 inland and 30 near the coast (Fig. 15a-c). For all years, the density of phalarope nests was higher in coastal areas (Coastal: 2.9 ± 0.7 nests/km , Inland: 1.6 ± 0.6 nests/km2). Coastal Red Phalarope nests were significantly clumped in all years. In contrast, the distribution of inland phalarope nests was random in all years (Table 9). Of the 30 coastal nests, 27 were within 150 m of the nearest Sabine's Gull nest. These nests are referred to as "with gulls". Two inland nests were also near Sabine's Gulls, for a total of 29 nests with gulls and 26 without. The mean distance of coastal nests to the nearest Sabine's Gull nest for all years was 92 ± 11 m, compared to 787 ± 106 m for inland nests. Coastal phalaropes selected sites significantly nearer to Sabine's Gulls than expected from a random distribution (92 ± 11 m vs. 259 ± 49 m, U= 270, n, = n2= 30, P = 0.007). I detected no habitat 2 2 differences between random coastal sites with.and without gulls (Mests of 1 m and 75 m PC scores, all P's > 0.35). Phalaropes nesting near gulls tended to initiate clutches after the Sabine's Gulls. Of 15 phalarope clutches where lay date of both the phalarope and the nearest Sabine's Gull was known, 10 were initiated after the first Sabine's Gull egg was laid. This measure is conservative as Sabine's Gulls establish territories upon arrival to the breeding grounds (Stenhouse et al. 2001), while Phalaropes do not defend nesting territories (Tracy et al. 2002). PREDATORS AND PREY The frequency of predator and lemming sightings differed substantially between years, (Fig. 2, chapter 2). Lemmings were present in 2000 and 2001, but none was seen in 2002. Predators were present in all years, but sightings did not vary co-vary closely across species. 35 Foxes were more abundant in 2001 and 2002 than in 2000. Jaegers, primarily Parasitic Jaegers, were much more abundant in 2002 than in the two other years. NESTING SUCCESS Nesting success of the Sabine's Gulls varied strongly across years. In 2000, 19 of the 25 (76%) nests hatched at least one chick. In 2001, 16 of 25 (64%) nests hatched young. In 2002, only 1 of 35 nests hatched young. Twenty-nine clutches were taken by predators and five were abandoned. Locations of nests appear in Fig. 15a-c. Mayfield estimates of Phalarope hatch success were low in all years, varying from 31% in 2000 and 35% in 2001 to 5% in 2002. In both 2000 and 2001, nests without gulls had daily mortality rates that were approximately double the rates of nests with gulls. In neither year, however, was this effect significant. In 2002, this trend was reversed, and nests with gulls suffered substantially higher mortality than nests without gulls (Fig. 16). Mayfield estimates for hatch success over a 19-day incubation period showed >17% differences in hatch success between nests with and without gulls in each year (Table 10), but statistical power to identify significant differences was low. The proportion of days that a phalarope nest was <150 m away from an active Sabine's Gull nest did not predict Red Phalarope nest survival. However, there was a significant interaction effect between gull presence and year; high gull presence increased nest survival in 2000 and 2001 and decreased survival in 2002. Habitat (PC2 75 m ) was the only significant main effect for all years combined, with high scores predicting longer survival times and better hatch success (Table 11). Nests with high scores for this component are found in scrub willow and moss carpet habitats, towards the inland boundary of coastal areas (refer to Table 9 for loadings). Similarly, PC2 75 m was a significant predictor of survival for all nests in 2000. In 2001, survival was best predicted by the height of rocks and vegetation surrounding nests; 36 successful nests had less cover than failed nests (Successful: 38 ± 5 mm, Failed: 54 ± 7 mm). For all years, inland nests surrounded by lower vegetation or rocks had better survival (Successful: 44 ± 3 mm, Failed: 63 ± 6 mm). In both cases, the height values for successful nests were intermediate. Random sites (n = 183) at East Bay are surrounded by rocks or vegetation 18 ± 1 mm high. The survival of coastal nests (all years) was predicted successfully by PCI 75 m 2 , a measure of habitat heterogeneity. Sites in fine-grained habitats (high PCI score) showed reduced survival. Nest habitats with high PCI 75 m 2 scores include mossy pond margins and narrow land bridges in intertidal habitats. ARTIFICIAL E G G EXPERIMENTS In both rounds of artificial nest experiments, coastal nests near Sabine's Gulls had the lowest daily survival rate (Table 12). In the first round of the experiment (begun 8 July, 2002), 17 of 36 (47%) coastal nests with gulls had damaged or missing eggs, compared to 8 of 32 (25%) nests for coastal sites without nearby gulls. Predation was lowest at inland sites, with 4 of 32 (13%) nests taken by predators. Predation was generally higher in the second round (Z-tests, all P's < 0.05); 23 of 36 (64%) coastal nests with gulls were preyed on, 19 of 32 (59%) in coastal habitats without gulls and 12 of 32 (38%) inland. The survival rate of artificial nests at distances of 50 m, 100 m and 200 m from Sabine's Gull nests did not differ in either round. No significant differences in survival were observed between 2 egg and 4 egg artificial nests (Table 12), but power here was low. For all treatments and both rounds, 91 nests were depredated. In 20 instances (22%), depredated eggs were punctured by a narrow avian bill (diameter < 5 mm). Ruddy Turnstones are common in coastal habitats of the study area and are known to be opportunistic predators on birds' eggs (e.g., Vuolanto 1968, Parkes et al. 1971, Alberico et al. 1991). However, no real nest at East Bay was ever found with such punctures. When these punctured eggs are not included as 37 depredated, I found no differences in artificial nest survival between protected, coastal unprotected and inland habitats (Table 12). Cox regression identified few factors that significantly influenced artificial nest survival (Table 13). In the first round, less concealed nests in coastal areas without gulls survived better. In the second round, nests in coastal areas with gulls survived better when there was lower surrounding vegetation or rocks and less exposed substrate (higher PC2 1 m 2). For coastal nests with gulls, the number of days that the nearby Sabine's Gull nest was active did not influence survival time in either round of checks. FLUSHING DISTANCE EXPERIMENTS Flush distances were recorded for 5 phalarope nests with and 8 nests without nearby gulls (n = 29 flushes). As flush distance was not related to incubation stage (r2 = 0.01, P = 0.60), means for individuals were used in all analyses. Incubators with nearby gulls flushed at significantly greater distances than birds without gulls (With: 38 ± 8 m (median 48 m), Without: 10 ± 5 m (median 4.6 m), U= 3.0, P = 0.01). However, mean flush distance was also significantly correlated with % overhead nest concealment (r2 = 0.42, P - 0.02), and . concealment differed almost three-fold between treatments (With Gulls: 3.5 ± 1.9%, Without: 9.4 ± 2.2%, U= 8.5, P = 0.08). To control for this difference, I generated residuals from a regression of flush distance on overhead concealment. The residuals for incubators with nearby gulls were larger than for birds without gulls (U= 6.0, P = 0.04), suggesting that flush distances were greater than predicted by the difference in nest concealment. INCUBATION BEHAVIOUR Incubation behaviour was recorded at 9 nests for 42 complete 24 h periods. Four of the nests (15 nest days) were within 150 m of Sabine's Gulls and five nests (27 nest days) were 38 without nearby gulls. For three independent pairs of nests (12 days total paired observations), mean nest attendance was lower for incubators with nearby gulls than for those without (With Gulls: 81.7 ± 4.4%, Without Gulls: 88.3 ± 0.9%). Incubators with gulls achieved this lower nest attendance by taking both more recesses (With Gulls: 29.6 ±3 .9 recesses / 24 h, Without Gulls: 23.0 ± 4.7 recesses / 24 h) and recesses of longer duration (With Gulls: 9.5 ± 2.2 min / recess, Without Gulls: 7.9 ± 1.3 min / recess). Because of small samples, no statistical analyses were conducted on these paired observations. When all observations were analysed with a G L M , nest was identified as a significant random factor for all predicted variables (Table 14, a). With the random factor included, the presence of gulls, date (and date ), incubation stage (and stage ) and weather did not predict any incubation parameter. Means for individuals were used for a second set of analyses that included predictor variables with no variance within a nest. These analyses (Table 14, b) identified a significant effect of gulls, with incubators near gulls exhibiting reduced nest attentiveness. The daily average wind chill best predicted the number of recesses in a 24 h period. The mean recess duration was not significantly related to any of the covariates analysed. These effects must be interpreted with caution, however, as variance in the dataset was reduced artificially by using means rather than individual records in the analyses. D I S C U S S I O N Protective nesting associations have been noted in many birds (e.g., Durango 1949, Dyrcz et al. 1981, Haemig 2001), with a disproportionate number of accounts for tundra breeders (e.g., Popham 1897, Paulson and Erckmann 1985, Blomqvist and Elander 1988). As phalaropes exhibit no nest defence, they could benefit by nesting near an aggressive species. In a protective nesting association, phalaropes must: (1) be able to recognise potential protectors, (2) actively 39 select nest sites near these birds and not merely in similar habitat, and (3) derive benefits that exceed the effects of predator swamping. In all years, phalarope nest densities were highest in coastal areas with gulls, but support for reproductive benefits of this strategy to phalaropes was variable. Birds recognise their predators and generally respond appropriately (e.g., Simmons 1952, McCaffrey 1982). Many species can discriminate between predators that pose different levels of threat (Stenhouse et al., submitted, Walters 1990, Larsen et al. 1996). Direct tests of the ability of birds to identify potential protectors, however, are rare. Perhaps the best examples are for associations between tundra nesting waterfowl (Branta bernicla bernicla, Somateria spectabilis, Anser caerulescens atlanticus) and Snowy Owls. Snowy Owls can defend their nests from all egg predators, but will prey upon waterfowl when lemmings are scarce (Ebbinge and Spaans 2002). Associations between geese and Snowy Owls tend to occur in years of lemming abundance (Underhill et al. 1993, Summer et al. 1994, Lepage et al. 1996, Bety et al. 2001, Ebbinge and Spaans 2002), suggesting that geese can assess the risks and benefits of nesting near these effective protectors that are also potential predators. While I did not test whether phalaropes can differentiate between potential protectors and predators, their nesting preferences and behaviour suggest that they can. Sabine's Gulls are common at East Bay, and they successfully expel most avian predators. They have never been observed to depredate a nest of any species (Stenhouse et al., submitted). I found that phalaropes prefer to nest near Sabine's Gulls, but did not find aggregations near the nests of other aggressive but potentially dangerous species such as Parasitic Jaegers or Herring Gulls (pers. obs.). I observed nesting and feeding phalaropes at East Bay react to alarm calls of Sabine's Gulls with vigilance (n = 16). In contrast, they cowered at the sight of Parasitic Jaegers (n = 5) (PAS). Further, phalaropes on the Taimyr Peninsula also nest near potential protector species such as Black-bellied Plover and Long-tailed Jaeger (Larsen and Grundetjern 1997). It therefore 40 seems likely that phalaropes can discriminate between threatening and harmless species and that they identify Sabine's Gulls as potential protectors. I found strong evidence that Red Phalaropes nesting in coastal habitats selected sites near gulls. The clutches of phalaropes in "with gulls" habitats tended to be laid after Sabine's Gulls initiated their clutches. Their nests were significantly closer to nesting gulls than predicted by random settling, resulting in a significantly clumped distribution in "with gulls" areas. Site choice by phalaropes was not constrained by snow cover at the time of nest initiation, as snow cover was less than 5% in 2001. I could detect no difference between the available coastal habitat and the "with gulls" habitat preferred by phalaropes. While phalaropes preferred coastal areas with gulls, fitness consequences of this preference varied strongly among years. Mean hatch success of phalaropes at East Bay (22%), was similar to that found in other studies (Tracy et al. 2002, Erckmann 1981). Hatch success differed between years, from 31- 35% 2000 and 2001, to only 5% in 2002. Breeding success of the Sabine's Gulls was also highly variable, with >60% success in 2000 and 2001, and only 3% success in 2002. The low hatch success of all birds at East Bay in 2002 coincided with an absence of lemmings and many sightings of foxes and jaegers. Similar correlations between lemming abundance, predation and nest success have been well documented for other arctic birds (e.g., Summers et al. 1998, Bety et al. 2001, 2002, and references therein). Parasitic Jaegers, the most common jaeger species at East Bay, are not dependent on lemmings as prey (Wiley and Lee 1999). However, rodents are the primary prey of arctic foxes across much of the arctic (Stickney 1991, Van Impe 1996). Foxes show numerical and functional responses to changes in lemming abundance (e.g., Wilson and Bromley 2001), and prey on eggs and birds when lemmings are scarce (Larson 1960, Summers and Underhill 1987). At East Bay, foxes were scarce in 2000 but more abundant in 2001, after a high lemming year in 2000. When lemmings were scarce in 2002, fox predation on birds' eggs, an alternative prey, 41 was high. As Sabine's Gulls can expel avian predators but not arctic foxes from their breeding territories (Stenhouse et al., submitted), these fluctuations in predator regime are likely to affect protective associations. Hatch success of Red Phalaropes in areas with gulls was nearly twice that of phalaropes that nested inland in 2000 and 2001, when predation was low. In 2002, hatch success of nests with gulls was 0%, compared to an estimate of 18% for areas without gulls. Artificial egg experiments in 2002 also showed a similar pattern, with equal or higher rates of predation for areas with gulls than those without. If Sabine's Gulls offer protection from avian predators only, Red Phalaropes would not benefit from nesting near them in years of high fox predation. Indeed, the higher densities of nests in coastal areas and the activity of the colonial birds may have enhanced the ability of foxes to locate and depredate nests (Rodgers 1987, Mayer and Ryan 1991, Larsen and Moldsvor 1992, Hogstad 1995). For cryptic species, activity near the nest increases the risk of predation. Reducing the number of incubation recesses decreases detectability by predators in a variety of birds (e.g., Lyon and Montgomerie 1987, Martin et al. 2000, Ghalambor and Martin 2002). Increases in nest attentiveness may also decrease risk of predation by shortening incubation (Martin 1995). Increased nest attentiveness has costs, however, as incubators' body condition may be dependent on the amount of time budgeted for foraging (Carey 1980, Hegyi and Sasvari 1998). Though variability between nests was high in my study, phalaropes in areas without nearby gulls exhibited higher nest attentiveness. These differences may be sufficient to cause differences in body condition of phalaropes with and without gulls. Phalaropes nesting at Barrow, Alaska (71 °N), exhibited 85% nest attendance and lost an average of 14% of their body weight over a 19.5-day incubation period (Schamel and Tracy 1987). Similarly, incubators at the Arctic National Wildlife Refuge (71°N) exhibited 83% attentiveness and lost 14% of their mass (Erckmann 1981). In contrast, incubators at Wales, 42 Alaska (66°N) spent 70% of their time on the nest, and maintained or gained mass over a 22.5-day incubation period (Erckmann 1981). The 7% difference observed in incubation constancy between phalaropes with and without nearby gulls could thus affect body condition of incubators, with potential effects on adult survival, parental care duration or chick survival (Hegyi and Sasvari 1998). Incubators with and without nearby gulls also differed in the distance at which they flushed when nests were approached. The median flush distance of those with nearby gulls was 48 m, while it was 5 m for those without gulls. Phalaropes in sedge marsh habitats, rely on cryptic behaviour to escape detection (e.g., Ridley 1980, Mayfield 1979). There are reports of humans touching incubating males without causing them to flush (Manniche 1910, Mayfield 1979, Tracy et al. 2002). Both eggs and adults are vulnerable to the common predators at East Bay: arctic fox and Parasitic Jaegers (Tracy et al. 2002, Young 1954, Wiley and Lee 1999). Consequently, remaining on the nest as predators approach represents a trade-off between adult survival and clutch loss (e.g., Montgomerie and Weatherhead 1988, Gotmark et al. 1995, Larsen et al. 1996, Wiebe and Martin 1998). If coastal phalaropes rely on the aggressive defence of nearby gulls, the benefit of remaining on the nest as predators approach may be reduced. I found that phalaropes nesting in areas with nearby Sabine's Gulls exhibited different behaviour and experienced reduced rates of predation in 2 of 3 years, when lemmings were abundant. Across the North, the duration of lemming cycles typically varies from 3-5 years (Hanski and Korpimaki 1995). If phalaropes benefit from nesting near Sabine's Gulls in all years but that after a lemming decline, this trait would be adaptive over an average lemming cycle. However, my results do not unequivocally link the differences in phalarope nest success to the presence of Sabine's Gulls. Habitat factors predicted nest survival over all years (but see Chapter 2), while the presence of Sabine's Gulls did not. Nests in scrub willow and moss carpet habitats survived better, as did nests with intermediate concealment. These nest habitat features, 43 however, predict increased survival in the coastal habitats where Sabine's Gulls also nested. Further, phalaropes did not use coastal habitats without gulls, though I detected no habitat differences between coastal areas with and without gulls. The presence of Sabine's Gulls appears to influence Red Phalarope nest distribution, but this association is beneficial only in years when lemmings aren't scarce, and nest predation by arctic foxes is low. 44 Chapter 4: General Discussion When studying nest sites and reproductive success, researchers typically seek patterns of habitat use that have emerged over evolutionary time (e.g., Wiens 1989, Clark and Shutler 1999). These patterns of habitat selection are assumed to reflect optima that minimize the risk of predation and/or increase the chance of future reproduction (Cody 1981, Rosenzweig 1985). However, there is often considerable intraspecific variation in the characteristics of nest sites (e.g., Chapter 3, Marzluff 1988, Wiebe and Martin 1998). By examining this variation and identifying habitat differences between successful and failed nest sites, we can identify the selective processes that have contributed to current patterns. Innumerable studies have identified non-random patterns of nest site use by birds (Clark and Shutler 1999). In chapter 2,1 demonstrated that nesting habitat was selected non-randomly for all five species of shorebirds studied. I also showed that the nest preferences of each species were unique, such that we could discriminate between the nests of each species on the basis of habitat. Using nest sites similar to those of other species may increase the risk of predation by predators that develop a search image (Martin 1988), but I found no evidence that those species with the most similar nest sites had increased predation. Within species, I failed to identify habitat features that discriminated between successful and unsuccessful nest sites, nor did I find increased success in individuals occupying their preferred nesting habitat. In short, I found clear evidence that nest site selection was non-random, but little evidence that variation in the habitat of nest sites contributed to variation in reproductive success. This finding differs from that of other studies where researchers have identified both patterns of nest site use and selective process, i.e., differences between successful and unsuccessful nest sites that would maintain these preferences (Clark and Shutler 1999). However, there have been few investigations of nest site selection for shorebirds in tundra 45 habitats. My inability to identify habitat differences between successful and unsuccessful nests could simply mean that I failed to measure important habitat variables. Alternatively, the failure could stem from the structural simplicity of the tundra habitats in which I worked. On tundra, all shorebird nests are simple scrapes on the ground. The nest site, therefore, may influence success less than patterns of nest distribution or the behaviour of incubating adults. NESTING ASSOCIATIONS Aggressive nest defence may achieve nest protection, but it entails costs in terms of energy and exposure to risk (Montgomerie and Weatherhead 1988, Larsen et al. 1996). By leaving the aggressive nest defence to others, timid birds may reap the benefits of defence while avoiding the risks. We found that some Red Phalaropes selected nest sites near Sabine's Gulls, and that they had higher reproductive success than phalaropes without nearby gulls in 2 of 3 years. In the third year, however, lemmings were scarce and predation was high. As Sabine's Gulls can defend their nests from avian predators but not mammals, this nesting association did not benefit phalaropes in a year of high fox predation. While few other studies have tested for the presence and effect of nesting associations rigorously, they appear common in arctic areas and may contribute significantly to the reproductive success of tundra nesting birds (e.g., Larsen and Grundetjern 1997, Tremblay et al. 1997). BEHAVIOUR OF INCUBATING ADULTS Even in systems where variation in nest habitat does influence reproductive success, parental behaviour has been shown to have an overriding effect (e.g., Cresswell 1997, Ghalambor and Martin 2002). While I found few effects of habitat on nest success, differences in the rate of predation between species suggest that success might be influenced by the level of activity near the nest. 46 The rate at which birds come and go from the nest influences predation rates in many birds (Lyon and Montgomerie 1987, Wiebe and Martin 1997, Martin et al. 2000). In shorebirds, this rate is influenced by the parental care system. Uniparental incubators leave the nest to feed frequently, while biparental incubators typically incubate for longer bouts and feed while their mate is incubating. Presumably because of increased energetic demands, uniparental incubators leave the nest more frequently than biparental pairs. I found lower nest success for uniparental species than biparental species. While aggressive nest defence is more common in monogamous, biparental species (Larsen 1991), this pattern of nest success was not explained by defence behaviours. The most and least successful species here both exhibited similar anti-predator distraction displays. The importance of activity around the nest in determining success was corroborated by the observation that artificial nests, with no incubators, had lower daily mortality than real nests. The level of activity around a nest may be influenced by the parental care system, but it may also be influenced by individual decisions. Phalaropes nesting outside of the protective umbrella of Sabine's Gulls exhibited more cryptic behaviour than those with gulls nearby. They took fewer and shorter recesses, potentially compromising their body condition. Taking fewer recesses may reduce the detectability of the nest to predators. They also remained on the nest longer in approach experiments than phalaropes with nearby Sabine's Gulls, possibly reflecting a greater reliance on cryptic behaviour to avoid nest predation. ANNUAL VARIATION IN PREDATION Although I identified several factors that contributed to reproductive success, the single most important factor was the increased predation associated with low densities of lemmings. The nest success of Semipalmated Plovers was unaffected, but all other species had nesting success 25-75% lower in 2002, when lemmings were scarce, versus 2000 and 2001. While the 47 relationship between shorebird nesting success and lemming abundance is well documented for the European arctic, this relationship has rarely been reported for shorebirds in the North American arctic (Summers 1986, Blomqvist et al. 2002). IMPLICATIONS The cycle of predators and lemmings appeared to be the single greatest factor influencing reproductive success in tundra nesting shorebirds, but our understanding of this relationship comes primarily from migration counts of juveniles and adults (Blomqvist et al. 2002). Semipalmated Plovers at East Bay had similar reproductive success in a year with lemmings and one without, yet they nested in similar habitats to Ruddy Turnstones and coastal nesting Red Phalaropes. While their ability to evade predators in high predation years remains an enigma, it is clear that a broader investigation of the importance of predator and lemming cycles in the evolution of shorebird breeding strategies is warranted. Though lemmings are widespread across the Arctic, none are present on Coats Island, 100 km south of East Bay (Gaston, pers. comm.). Coats Island, however does have arctic foxes. With similar shorebird communities (Gaston and Ouellet 1997), these two sites offer a unique opportunity to investigate the influence of cyclical predation on shorebird breeding ecology. An understanding of the relative influence of nest habitat, dispersion and incubator behaviour on nest success is crucial to understanding the evolution of life histories (Lessells 1991, Cresswell 1997). If habitat is unimportant in determining nest success, or i f nest sites are not limited and birds are able to make optimal choices, selection may act instead on behavioural traits or spatial patterns of nests. Interspecific comparisons such as those presented here are useful for identifying patterns. However, studies of individuals faced with different levels of predation risk are needed to further 48 our understanding of the factors influencing nest site selection and reproductive success in tundra nesting shorebirds. 49 References Abraham K. , and C. D. Ankney. 1986. Summer birds of East Bay, Southampton Island, Northwest Territories. Canadian Field-Naturalist 100:180-185. Alberico, J. A . R., J. M . Reed, and L. W. Oring. 1991. Nesting near a Common Tern colony increases and decreases Spotted Sandpiper nest predation. Auk 108:904-910. Allard, K . In Prep. Foraging behaviour of Herring Gulls at East Bay, Southampton Island, Nunavut. Ph.D. dissertation, University of New Brunswick, Fredricton, New Brunswick. Andersson, M . , and C. G. Wiklund. 1978. Clumping versus spacing out: experiments on nest predation in Fieldfares (Turdus pilaris). Animal Behaviour 26:1207-1212. Anglestam, P., E. Lindstrom, and P. Widen. 1984. Role of predation in short-term population fluctuations of some birds and mammals in Fennoscandia. Oecologia 62:199-208. Armstrong, E. A . 1956. Distraction displays and the human predator. Ibis, 98:641-655. Arnold, T. W. 1999. What limits clutch size in waders? Journal of Avian Biology 30:216-220. Ashkenazie, S., and U . N . Safriel. 1979. Time-energy budget of the Semipalmated Sandpiper Calidrispusilla at Barrow, Alaska. Ecology 60:783-799. Baker, M . C. 1977. Shorebird food habits in the eastern Canadian arctic. Condor 79:56-62. Bergerud, A. T., and M . W. Gratson. 1988. Survival and breeding strategies of grouse. Pages 578-685 in Adaptive Strategies and Population Ecology of Northern Grouse (A. T. Bergerud and M . W. Gratson, Eds.). University of Minnesota Press, Minneapolis. Bety, J., G. Gauthier, J. Giroux, and E. Korpimaki. 2001. Are goose nesting success and lemming cycles linked? Interplay between nest density and predators. Oikos 93:388-400. Bety J., G. Gauthier, E. Korpimaki, and J. F. Giroux. 2002. Shared predators and indirect trophic interactions: lemming cycles and arctic-nesting geese. Journal of Animal Ecology 71:88-98. Bianki, V . V . 1967. Gulls, shorebirds, and alcids of Kandalaksha Bay. Kandalaksha State Reservation 6 (English translation by Israel Program for Scientific Translations, Jerusalem, Israel, 1977). Blomqvist, S., and M . Elander. 1988. King Eider (Somateria spectabilis) nesting in association with Long-tailed Skua (Stercorarius longicaudus). Arctic 41:138-142. Blomqvist, S., N . Holmgren, S. Akesson, A. Hedenstrom, and J. Pettersson. 2002. Indirect effects of lemming cycles on sandpiper dynamics: 50 years of counts from southern Sweden. Oecologia 133:146-158. 50 Bohning-Gaese, K. , M . L . Taper, and J. H. Brown. 1993. Are declines in North American insectivorous songbirds due to causes on the breeding range? Conservation Biology 7:76-86. Brown, J. L. 1964. The evolution of diversity in avian territorial systems. Wilson Bulletin 76:160-169. Brunton, D. H. 1986. Fatal antipredatory behaviour of a Kildeer. Wilson Bulletin 98:605-607. Burger, J. 1985. Habitat selection in temperate marsh-nesting birds. Pages 253-281 in Habitat Selection in Birds (M. L . Cody, Ed.). Academic Press, Orlando. Cartar, R.V., and R. D. Montgomerie. 1985. The influence of weather on incubation scheduling of the White-rumped Sandpiper (Calidris fuscicolis): a uniparental incubator in a cold environment. Behaviour 95:261-289. Carey, C. 1980. The ecology of avian incubation. Bioscience 30:819-824. Clark, P. J., and F. C. Evans. 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35:445-453. Clark, K. L., and R. J. Robertson. 1979. Spatial and temporal multi-species nesting aggregations in birds as anti-parasite and anti-predator defense. Behavioural Ecology and Sociobiology 5:359-371. Clark, R. G., and T. D. Nudds. 1991. Habitat patch size and duck nesting success: the crucial experiments have not been performed. Wildlife Society Bulletin 19:534-543. Clark, R. G., and D. Shutler. 1999. Avian habitat selection: pattern from process in nest-site use by ducks. Ecology 80:272-287. Cody, M . L. 1981. Habitat selection in birds: the roles of vegetation structure, competitors and productivity. Bioscience 31:107-113. Colwell, M . A. , and L. W. Oring. 1988. Breeding biology of Wilson's Phalarope in southcentral Saskatchewan. Wilson Bulletin 100:567-582. Colwell, M . A. , and L. W. Oring. 1990. Nest site characteristics of prairie shorebirds. Canadian Journal of Zoology 68:297-302. Cresswell, W. 1997. Nest predation: the relative effects of nest characteristics, clutch size and parental behaviour. Animal Behaviour 53:93-103. Croze, H. 1970. Search image in Carrion Crows. Zeitschrifte fur Tierpsychologie 5:1-85. Donnelly, K. 1978. Simulations to determine the variance and edge-effect of total nearest neighbor distance. Pages 91-95 in Simulation Models in Archaeology (I. Hodder, Ed.) Cambridge University Press, London. 51 Drury, W. H. , Jr. 1961. The breeding biology of shorebirds on Bylot Island, Northwest Territories, Canada. Auk 78:176-219. Durango, S. 1949. The nesting association of birds with social insects and with birds of different species. Ibis 91:140-143. Dyrcz, A. , J. Witkowski, and J. Okulewicz. 1981. Nesting of'timid' waders in the vicinity of 'bold' ones as an antipredation adaptation. Ibis 123:542-545. Ebbinge, B. S., and B. Spaans. 2002. How do Brent Geese cope with evil? Complex relationships between predators and prey. Journal fur Omitholgie 143:33-42. Erckmann, W. J., Jr. 1981. The evolution of sex-role reversal and monogamy in shorebirds. Ph.D. dissertation, Univ. of Washington, Seattle. ESRI. 1992. Arcview 3.2. Redlands, California. Fontaine, A . J., and M . L . Mallory, Eds. In Prep. Land cover mapping of East Bay and Harry Gibbons Migratory Bird Sanctuaries, Southampton Island, Nunavut. Canadian Wildlife Service Occasional Paper, Environment Canada, Ottawa. Gaston, A . J., and H. Ouellet. 1997. Birds and mammals of Coats Island, NWT. Arctic 50:101-118. Ghalambor, C. K., and T. E. Martin. 2002. Comparative manipulation of predation risk in incubating birds reveals variability in the plasticity of responses. Behavioural Ecology 13:101-108. Gloutney, M . L., and R. G. Clark. 1997. Nest-site selection by Mallards and Blue-winged Teal in relation to microclimate. Auk 114:381-395. Gochfeld, M . 1984. Antipredator behaviour: aggressive and distraction displays of shorebirds. Pages 289-377 in Behaviour of Marine Animals. Vol . 5, Shorebirds: Breeding Behaviour and Populations (J. Burger and B. Olla, Eds.). Plenum Press, New York. Goranson, G., J. Karlson, S. G. Nilsson, and S. Ulfstrand. 1975. Predation on bird's nests in relation to antipredator aggression and nest density: an experimental study. Oikos 26:117-120. Gotmark, F., D. Blomgvist, O. C. Johansson, and J. Bergkvist. 1995. Nest site selection: a trade-off between concealment and view of the surroundings? Journal of Avian Biology 26:305-312. Guttmann, L. 1954. Some necessary conditions for common factor analysis. Psychometrika 30:179-185. Haemig, P. D. 2001. Symbiotic nesting of birds with formidable animals: a review with application to biodiversity conservation. Biodiversity and Conservation 10:527-540. Haftorn, S. 1988. Incubating female passerines do not let the egg temperature fall below the, 'physiological zero temperature' during their absences from the nest. Ornis Scandinavica 19:97-110. 52 Hanski, I., and E. Korpimaki. 1995. Microtine rodent dynamics in northern Europe: parameterized models for the predator-prey interaction. Ecology 76:840-850. Hegyi, Z., and L. Sasvari. 1998. Parental condition and breeding effort in waders. Journal of Animal Ecology 67:41-53. Hilden, O. and S. Vuolanto. 1972. Breeding biology of the Red-necked Phalarope Phalaropus lobatus in Finland. Ornis Fennica 49:57-85. . Hines, J. E., and J. R. Sauer. 1989. Program CONTRAST - A General Program for the Analysis of Several Survival or Recovery Rate Estimates. US Fish & Wildlife Service, Fish & Wildlife Technical Report 24, Washington, DC. Hogstad, O. 1995. Do avian and mammalian nest predators select for different nest dispersion patterns of Fieldfares (Turdus pilaris)! A 15 year study. Ibis 137: 484-489. Holm, E. O. 1971. Observations on the breeding behaviour of Grey and Red-necked Phalaropes. Ibis 113: 335-348. Horvath, O. 1964. Seasonal differences in Rufous Hummingbird nest height and their relation to nest climate. Ecology 45:235-241. Hussell, D. J. T., and G. W. Page. 1976. Observations on the breeding biology of Black-bellied Plovers on Devon Island, N.W.T. Canada. Wilson Bulletin 88:632-653. Jenness, J. 2002. Nearest features, with distances and bearings, v 3.6d. Arcview extension: http://arcscripts.esri.com. Johnson, D. H. 1979. Estimating nesting success: the Mayfield method and an alternative. Auk 96:651-661. Jourdain, F. C. 1936. The so-called injury feigning in birds. Oologists Rec 16:25-37. Kitchinski, A . A. , and Y . I. Chernov. 1973. Foods of the Grey Phalarope in the tundras of East Siberia. Pages 61-64 in Fauna and Ecology of Waders (symposium). Pt. 1. Moscow University Press, Moscow, USSR. Knight, R. L. and S. A. Temple. 1986. Why does intensity of nest defence increase during the nesting cycle? Auk 103:318-327. Koskimies, J. 1957. Terns and gulls as features of habitat recognition for birds nesting in their colonies. Ornis Fennica 34:1-6. Krebs, C. J. 1964. The lemming cycle at Baker Lake, Northwest Territories, during 1959-1962; Arctic Institute of North America Technical Paper 15:1-104. Krebs, C. J. 1989. Ecological methodology. Harper Collins, New York. Lack, D. 1968. Ecological adaptations for breeding in birds. Methuen, London; 53 Lariviere, S., and F. Messier. 1998. Effect of nearest neighbours on simulated waterfowl nests: can predators recognize high-density nesting patches? Oikos 83:12-20. Larsen, T. 1991. Anti-predator behaviour and mating systems in waders: aggressive nest defence selects for monogamy. Animal Behaviour 41:1057-1062. Larsen, T., and S. Grundetjern. 1997. Optimal choices of neighbour: predator protection among tundra birds. Journal of Avian Biology 28:303-308. Larsen, T. and Moldsvor, J. 1992. Antipredator behaviour and breeding associations of Bar-tailed Godwits and Whimbrels. Auk 109:601-608. Larsen, T., T. A . Sordahl, and I. Byrkjedal. 1996. Factors related to aggressive nest protection behaviour: a comparative study of Holarctic waders. Biological Journal of the Linnean Society 58:409-439. Larson, S. 1960. On the influence of the arctic fox (Alopex lagopus) on the distribution of arctic birds. Oikos 11:276-305. Lead, S. 2002. Generate randomly distributed points. Arcview extension: http ://arcscripts. esri. com Lessels, C M . 1991. The evolution of life histories. Pages 137-168 in Behavioural Ecology: An Evolutionary Approach (J. R. Krebs and N . B. Davies, Eds.) Blackwell Scientific, Oxford. Lyon, B., and R. Montgomerie. 1987. Ecological correlates of incubation feeding: a comparative study of high arctic finches. Ecology 68:713-722. Mabee, T. J., and V . B. Estelle. 2000. Nest fate and vegetation characteristics for Snowy Plover and Killdeer in Colorado, USA. Internation Wader Study Group Bulletin 93:67-72. Macdonald, S. D., and D. F. Parmelee. 1962. Feeding behaviour of the Turnstone in arctic Canada. British Birds 55:241-243. Manniche, A . L. Y . 1910. Terrestrial mammals and birds of northeast Greenland. Meddelelser om Gronland 45, Copenhagen. Manolis, J. C , D. E. Andersen, and F. J. Cuthbert. 2000. Uncertain nest fates in songbird studies and variation in Mayfield estimation. Auk 117:615-626. Martin, A . P., and D. Baird. 1988. Lemming cycles: which Palearctic migrants are affected. Bird Study 35:143-145. Martin, T. E. 1987. Food as a limit on breeding birds: a life-history perspective. Annual Review of Ecology and Systematics 18:453-487. Martin, T. E. 1988a. Habitat and area effects on forest bird assemblages: is nest predation an influence? Ecology 69:74-84. 54 Martin, T. E. 1988b. On the advantage of being different: nest predation and the coexistence of bird species. Proceedings of the National Academy of Sciences (USA) 85:2196-2199. Martin, T. E. 1992. Breeding productivity considerations: what are the appropriate habitat features for measurement? Pages 455-473 in Ecology and Conservation of Neotropical Migrant Landbirds (J.M. Hagan III and D. W. Johnson, Eds.). Smithsonian Institution Press, Washington. Martin, T. E. 1993. Nest predation among vegetation layers and habitat types: revising the dogmas. American Naturalist 141:897-913. Martin, T. E. 1995. Avian life history evolution in relation to nest sites, nest predation, and food. Ecological Monographs 65:101-127. Martin, T. E. 1998. Are microhabitat preferences of coexisitng species under selection and adaptive? Ecology 79:656-670. Martin, T. E. 2001. Abiotic vs. biotic influences on habitat selection of coexisting species: climate change impacts. Ecology 82:175-188. Martin, T. E., and C. K. Ghalambor. 1999. Males feeding females during incubation. I. Required by microclimate or constrained by nest predation? American Naturalist 153:131-139. Martin, T. E., J. Scott, and C. Menge. 2000. Nest predation increases with parental activity: separating nest site and parental activity effect. Procedings of the Royal Society of London B 267:2287-2293. Marzluff, J. M . 1988. Do Pinyon Jays alter nest placement based on prior experience? Animal Behaviour 36:1-10. Maxson, S. J., and L. W. Oring. 1980. Breeding season time and energy budgets of the polyandrous Spotted Sandpiper. Behaviour 74:200-263. Mayer, P. M . and M . R. Ryan. 1991. Survival rates of artificial Piping Plover nests in American Avocet colonies. Condor 93:753-755. Mayfield, H. F. 1961. Nesting success calculated from exposure. Wilson Bulletin 73:255-261. Mayfield, H. F. 1979. Red Phalaropes breeding on Bathurst Island. Living Bird 17:7-39. McCaffery, B. J. 1982. Geographic variation in predator-mobbing behaviour - a hypothesis. Wader Study Group Bulletin 35:30. McKinney, F. 1965. Spacing and chasing in breeding ducks. Wildfowl 16:92-106. Montgomerie, R. D., and P. J. Weatherhead. 1988. Risks and rewards of nest defence by parent birds. Quarterly Review of Biology 63: 167-187. 55 Murphy, M . T. 1989. Life history variability in North American breeding tyrant flycathers: phylogeny, size, or ecology? Oikos 54:3-14. Myers, J. P. 1978. One deleterious effect of mobbing in the Southern Lapwing (Vanellus chilensis). Auk 95:419. Nettleship, D. N . 2000. Ruddy Turnstone (Arenaria interpres). No. 537 in The Birds of North America (A. Poole and F. Gi l l , Eds.). The Birds of North America Inc., Philadelphia. Nol, E., and M . S. Blanken. 1999. Semipalmated Plover (Charadrius semipalmatus). No. 444 in The Birds of North America (A. Poole and F. Gil l , Eds.). The Birds of North America Inc., Philadelphia. Norton, D. W. 1972. Incubation schedules of four species of calidridine sandpipers at Barrow, Alaska. Condor 74:164-176. Nuechterlein, G. L . 1981. 'Information parasitism' in mixed colonies of Western Grebes and Forster's Terns. Animal Behaviour 29:985-989. Page, G. W., L. E. Stenzel, D. W. Winkler, and C. W. Swarth. 1983. Spacing out at Mono Lake: breeding success, nest density, and predation in the Snowy Plover. Auk 100:13-24. Parkes, K. C , A . Poole, and H. Lapham. 1971. The Ruddy Turnstone as an egg predator. Wilson Bulletin 83:306-308. Parmelee, D. F. 1992. White-rumped sandpiper, Calidris fuscicollis. No. 29 in The Birds of North America (A. Poole and F. Gills, Eds.). The Academy of Natural Sciences, Pennsylvania; The American Ornithologists Union, Washington. Parmelee, D. F., W. D. Grenier, and D. W. Graul. 1968. Summer schedule and breeding biology of the White-rumped Sandpiper in the central Canadian Arctic. Wilson Bulletin 80:5-29. Paulson, D. R. 1995. Black-bellied plover, Pluvialis squatorola. No. 186 in The Birds of North America (A. Poole and F. Gills, Eds.). The Academy of Natural Sciences, Pennsylvania; The American Ornithologists Union, Washington. Paulson, D. R. and W. J. Erckmann. 1985. Buff-breasted sandpipers nesting in association with Black-bellied Plovers. Condor 87:429-430. Pitelka, F. A. , R. T. Holmes, and S. F. MacLean, Jr. 1974. Ecology and evolution of social organization in arctic sandpipers. American Zoologist 14:185-204. Popham, H. L. 1897. Notes on birds observed on the Yenisei river, Siberia, in 1895. Ibis 3:89-108. Portenko, L. A . 1972. Birds of the Chukchi Peninsula and Wrangel Island, Vol . 1. Amerind Publishing, New Delhi. 56 Redondo, T. 1989. Avian nest defence: theoretical models and evidence. Behaviour 111:161-195. Reid, J. M . , P. Monaghan, and G. D. Ruxton. 1999. The effect of clutch cooling rate on starling, Sturnus vulgaris, incubation strategy. Animal Behaviour 58:1161-1167. Reid, J. M . , P. Monaghan, and G. D. Ruxton. 2000. Resource allocation between reproductive phases: the importance of thermal conditions in determining the cost of incubation. Proceedings of the Royal Society of London 267:37-41. Reid, J. M . , W. Cresswell, S. Holt, R. J. Mellanby, D. P. Whitfield, and G. D. Ruxton. 2002. Nest scrape design and clutch heat loss in Pectoral Sandpipers. Functional Ecology 16:305-312. Reid, M . L., and R. D. Montgomerie. 1985. Seasonal patterns of nest defence by Baird's Sandpipers. Canadian Journal of Zoology 63:2207-2211. Reynolds, J. D. 1985. Sandhill Crane use of nest markers as cues for predation. Wilson Bulletin 97:106-108. Ricklefs, R. E. 1969. An analysis of nesting mortality in birds. Smithsonian Contributions in Zoology 9:1-48. Ridley, M . W. 1980. The breeding behavior and feeding ecology of Grey Phalaropes Phalaropus fulicarius in Svalbard. Ibis 122:210-226. Robertson, R. J. 1973. Optimal niche space of the Red-Winged Blackbird: spatial and temporal patterns of nesting activity and success. Ecology 54:1085-1093. . .. . Rodgers, J. A. , Jr. 1987. On the antipredator advantages of coloniality: a word of caution. Wilson Bulletin 99:269-271. Rodrigues, R. 1994. Microsite variables influencing nest-site selection by tundra birds. Ecological Applications 4:110-116. Rosenzweig, M . L. 1985. Some theoretical aspects of habitat selection. Pages 517-540 in Habitat Selection in Birds (M, L. Cody, Ed.). Academic Press, Toronto. Sandercock, B. K. 1997. Factors affecting demography of Western Sandpipers, Calidris mauri, and Semipalmated Sandpipers, Calidris pusilla, at Nome, Alaska. Ph.D. dissertation, Simon Fraser University, Vancouver. Schamel, D., and D. M . Tracy. 1987. Latitudinal trends in breeding Red Phalaropes. Journal of Field Ornithology 58:126-134. Schieck, J. O., and S. J. Hannon. 1993. Clutch predation, cover, and the overdispersion of nests of the Willow Ptarmigan. Ecology 74:743-750. Simmons, K. E. L. 1952. The nature of the predator-reactions of breeding birds. Behaviour 4:161-171. 57 Skipnes, K. 1979. The breeding behaviour of a pair of Turnstones, Arenaria interpres. Var Fuglefauna 2:86-94. (Norwegian with English summary) SPSS Inc. 2000. SPSS for windows 10.0.7. Chicago, Illinois, USA. Stenhouse, I. J. 2003. The reproductive ecology of the Sabine's Gull (Xema sabini) in the Eastern Canadian Arctic. Ph.D. dissertation, Memorial University, St. John's. Stenhouse, I. J., H. G. Gilchrist, and W. A . Montevecchi. 2001. Reproductive biology of Sabine's Gull in the Eastern Canadian Arctic. Condor 103:98-107. Stenhouse, I. J., H. G. Gilchrist, and W. A . Montevecchi. Submitted. Predation, anti-predator behaviour, and the influence of alternative prey on Sabine's Gulls (Xema sabini). Behavioral Ecology & Sociobiology. Stickney, A . 1991. Seasonal patterns of prey availability and the foraging behavior of arctic foxes (Alopex lagopus) in a waterfowl nesting area. Canadian Journal of Zoology 69:2853-2859. Sugden, L. G., and G. W. Beyersbergen. 1986. Effect of density and concealment on American Crow predation of simulated duck nests. Journal of Wildlife Management 50:9-14. Sullivan Blanken, M . , and E. Nol. 1998. Factors affecting parental behavior in Semipalmated Plovers. Auk 115:166-174. Summers R.W. 1986. Breeding production of Dark-bellied Brent Geese (Brantab. bernicla) in relation to lemming cycles. Bird Study 33:105-108. Summers R. W., Underhill L . G. 1987. Factors related to breeding production of Brent Geese (Branta b. bernicla) and waders (Charadrii). Bird Study 37:161-171. Summers, R. W., L. G. Underhill, E. E. Syroechkovski Jr., H . G. Lappo, R. P. Piys-Jones, and V . Karpov. 1994. The breeding biology of Dark-bellied Brent Geese Branta b. bernicla and King eiders Somateria spectabilis on the northeastern Taimyr Peninsula, especially in relation to Snowy Owl Nyctea scandiaca nests. Wildfowl 45:110-118. Tinbergen, N . , M . Impekoven, and D. Franck. 1967. An experiment on spacing-out as a defence against predation. Behaviour 28: 307-321. Titus, K., J. A . Mosher, and B. K . Williams. 1984. Chance-corrected classification for use in discriminant analysis: ecological applications/American Midland Naturalist 111:1-7. Tracy, D. M . , D. Schamel, and J. Dale. 2002. Red Phalarope (Phalaropus fulicarius). No. 698 in The Birds of North America (A. Poole and F. Gil l , Eds.). The Birds of North America Inc., Philadelphia. Tremblay, J. P., G. Gauthier, D. Lepage, and A . Desrochers. 1997. Factors affecting nesting success in Greater Snow Geese: effects of habitat and association with Snowy Owls. Wilson Bulletin 109:449-461. 58 Underbill, L .G. , R. P. Prys-Jones, E. E. Syroechkovski Jr., N . M . Groen, V . Karpov, H. G. Lappo, M . W. J. van Roomen, A . Rybkin, H. Schekkerman, H. Spiekman, and R.W. Summers. 1993. Breeding of waders (Charadrii) and Brent Geese Branta bernicla bernicla at Pronchishcheva lake, northeastern Taimyr, Russia, in a peak and a decreasing lemming year. Ibis 135:277-292. Van Impe, J. 1996. Long-term reproductive performance in White-fronted Geese (Anser a. albifrons) and Tundra Bean Geese (A.fabalis rossicus) wintering in Zeeland, Netherlands. Bird Study 40:280-289. Vleck, C. M . 1981. Energetic cost of incubation in the Zebra Finch. Condor 89:874-898. Vuolanto, S. 1968. On the breeding biology of the turnstone at Norrskar, Gulf of Bothnia. Ornis Fennica 45:19-24. Walsberg, G. E. 1981. Nest-site selection and the radiative environment of the Warbling Vireo. Condor 83:86-88. Walsberg, G. E. 1985. Physiological consequences of microhabitat selection. Pages 389-413 in Habitat Selection in Birds (M. L. Cody, Ed.). Academic Press, Orlando. Walters, J. R. 1990. Anti-predatory behavior of Lapwings: field evidence of discriminative abilities. Wilson Bulletin 102:49-70. Westerskov, K. 1950. Methods for determining the age of game bird eggs. Journal of Wildlife Management, 14:56-67. Webb, D. R. 1987. Thermal tolerance of avian embryos: a review. Condor 89:874-898. Wiebe, K. L., and K. Martin. 1997. Effects of predation, body condition and temperature on incubation ryfhyms of White-tailed Ptarmigan. Wildlife Biology 3:143-151. Wiebe, K. L. , and K. Martin. 1998. Costs and benefits of nest cover for ptarmigan: changes within and between years. Animal Behaviour 56:1137-1144. Wiens, J. A . 1989. The ecology of bird communities. Volume 1. Foundations and Patterns. Cambridge University Press, Cambridge. Wiley, R. H. , and D. S. Lee. 1998. Long-tailed Jaeger (Stercorarius longicaudus). No. 365 in The Birds of North America (A. Poole and F. Gi l l , Eds.). The Birds of North America Inc., Philadelphia. Wiley, R. H., and D. S. Lee. 1999. Parasitic Jaeger (Stercorarius parasiticus). No. 445 in The Birds of North America (A. Poole and F. Gil l , Eds.). The Birds of North America Inc., Philadelphia Williams, J. B. 1996. Energetics of avian incubation. Pages 375-415 in Avian Energetics and Nutritional Ecology (C. Carey, Ed.) Chapman and Hall, London. 59 Wilson, D. J., and R. G. Bromley. 2001. Functional and numerical responses of predators to cyclic lemming abundance: effects on loss of goose nests. Canadian Journal of Zoology 79:525-532. Young, C. G. 1954. Phalaropes, storm petrels and Barn Owl at sea. Ibis 96:311. 60 Tables Table 1: Features of the habitats of East Bay, Southampton Island, Nunavut (see Appendix I). Habitat Type Distinguishing Features Intertidal Zone Intertidal or within splash range of fall storms Dead/dormant moss (organic crust) occurs Bare substrate dominant, living moss and graminoids sparse and patchy Moss Carpet Pond edges in coastal areas Living moss covers substrate Sparse to moderate abundance of grasses and sedges Numerous herbs, but patchy and sparse Scrub Willow Drier areas in central and northern portions of plot (0.5-1 km inland) Salix spp. abundant Herbs, grasses, sedges and lichens common Substrate of bare soil and small rocks Dry Heath Drier areas > lkm inland Ericaceous shrubs dominant; a dense cover of mountain avens (Dryas integrifolia) is a distinguishing feature Willows and lichens abundant Herbs moderate in richness and abundance Substrate variable: soil, rock and gravel. Relief varies from flat to extremely hummocked Sedge Meadow Moist areas and pond edges inland Moss covers substrate, few rocks present Sedges/grasses tall (>50mm) and dense Herbs abundant and diverse Relief varies from flat to highly hummocked Gravel Ridge Bare gravel dominant Flora sparse and depauperate Visibly raised from surrounding areas Colonised sparsely by mountain avens at low edges 61 Table 2: The proportion of shorebird nests located in the six habitat categories at East Bay, Nunavut. Species n Gravel Ridge Dry Heath Sedge Meadow Scrub Willow Moss Carpet Intertidal Black-bellied Plover 21 12 62 2 24 0 0 Red Phalarope 57 0 14 29 30 12 15 Ruddy Turnstone 63 0 2 2 59 17 21 White-rumped Sandpiper 24 0 17 69 15 0 0 Semipalmated Plover 24 0 0 0 6 2 92 62 Table 3: Predicted species (i.e. group membership) for shorebird nests, based on a discriminant function analysis of habitat variables at scales of l m 2 and 75m2. Four letter species codes are displayed: Black-bellied Plover (BBPL), Red Phalarope (REPH), Ruddy Turnstone (RUTU), White-rumped Sandpiper (WRSA), Semipalmated Plover (SEPL). Predicted Species Actual Species B B P L REPH R U T U WRSA SEPL n B B P L 0.71 0.00 0.29 0.00 0.00 21 REPH 0.07 0.56 0.21 0.12 0.04 • 57 RUTU 0.02 0.03 0.83 0.02 0.11 63 WRSA 0.04 0.13 0.13 0.71 0.00 24 SEPL 0.00 0.00 0.17 0.00 0.83 24 63 Table 4: Habitat variables that discriminate between the nest sites of 5 shorebirds at East Bay, Nunavut. Standardised canonical discriminant function coefficients are displayed. Only variables contributing significantly to the discrimination are presented. Function Variable 1 2 3 4 l m 2 Moss 0.54 -0.18 0.22 -0.13 Willow 0.30 -0.16 0.30 -0.19 Herbs 0.44 0.00 -0.34 0.16 75m2 Gravel Ridge 0.08 0.50 0.19 0.22 Dry Heath 0.26 0.78 0.46 0.40 Sedge Meadow 0.31 0.08 0.71 -0.06 Intertidal -0.40 -0.40 0.84 0.44 Lateral Cone. 0.22 -0.25 -0.49 0.57 Overhead Cone. 0.36 -0.13 -0.05 0.30 % of Variance 47.7 29.9 13.4 9.0 Sig. <0.0001 O.0001 <0.0001 <0.0001 1-4 2-4 3-4 4 64 Table 5: Indices of aggregation for the distribution of shorebird nests at East Bay, Nunavut. Significant index values <1 indicate clumping, while significant index values >1 indicate overdispersion. Samples of fewer than seven nests were not analyzed (Donelly 1978). Year Species n Nearest Neighbour ( m ± S E ) Index of Aggregation 2000 Black-bellied Plover 2 Red Phalarope 20 243 ± 43 0.75* Ruddy Turnstone 8 333 ± 7 5 0.60* White-rumped Sandpiper 7 2001 Black-bellied Plover • 2 Red Phalarope 16 211 ± 4 4 0.57** Ruddy Turnstone 19 181 ± 2 4 0.54*** Semipalmated Plover 14 159 ± 71 0 4 0 * * * White-rumped Sandpiper 7 2002 Black-bellied Plover 17 494 ± 65 1.05 Red Phalarope 19 479±120 1.08 Ruddy Turnstone 38 142 ± 11 0 4 7 * * * Semipalmated Plover 10 372 ± 8 8 0.58* White-rumped Sandpiper 10 478 ± 79 0.75 * P < 0.05, ** P < 0.01, *** P < 0.001 65 Table 6: Habitat variables that discriminated between nests and random sites for shorebirds at East Bay, Nunavut. Separate analyses were run for each species. A l l habitat variables measured are displayed but standardised canonical discriminant function coefficients are shown only for variables contributing significantly to discrimination. Four letter species codes are displayed: Black-bellied Plover (BBPL), Red Phalarope (REPH), Ruddy Turnstone (RUTU), White-rumped Sandpiper (WRSA), Semipalmated Plover (SEPL). Species B B P L REPH R U T U SEPL WRSA Rock Dirt Crust Lichen Moss Willow Avens Sedge/Grass Herb 0.61 0.97 0.75 0.80 0.69 0.56 -0.58 -0.43 0.40 0.32 -0.39 0.39 0.65 75m2 Sedge Meadow Dry Heath Gravel Ridge Scrub Willow Moss Carpet Intertidal Lat. Cone. Over. Cone. -0.29 -0.66 -0.57 -0.64 0.41 -0.38 1.12 -0.51 0.68 0.58 0.67 0.44 % Correct Classification 9.7.6 82.1 % Improvement on Chance 47.6 32.1 86.5 89.5 83.3 36.5 39.5 33.3 A l l Chance Corrected P's <0.0001 66 Table 7: Mean distances to the nearest, or five nearest, shorebird neighbours for each species at East Bay, Nunavut. Both metrics represent distances to nests of any shorebirds, not only conspecifics. 1 Nearest 5 Nearest Species Mean (m) ± SE Mean (m) ± SE Black-bellied Plover 126 ± 1 8 238 ± 2 2 Red Phalarope 256 ± 54 447 ± 57 Ruddy Turnstone 116 ± 9 207 ± 11 Semipalmated Plover 117 ± 14 231 ± 16 White-rumped Sandpiper 153 ± 2 1 342 ± 3 1 67 T a b l e 8: Component loadings for principal components analyses of Red Phalarope nest site (A) and nest patch (B) characteristics. At the site level, original variables are estimated percent covers in a 1 m 2 circle centred on the nest. Crust refers to an organic crust of lichen and dead moss (cryptobiotic crust). At the patch level, original variables are estimated percent covers of habitat types in a 75 m 2 circle around the nest. A) Component 1 2 Rock 0.51 -0.68 Crust 0.43 0.41 Lichen 0.53 0.36 Moss -0.82 0.24 Dirt -0.22 -0.44 Avens 0.53 0.29 Willow 0.01 0.58 Sedge/Grass -0.61 0.13 B) Component 1 2 Gravel Ridge -0.26 -0.06 Dry Heath -0.56 0.05 Scrub Willow 0.11 0.72 Sedge Meadow -0.51 -0.25 Moss Carpet 0.55 0.44 Intertidal 0.49 -0.72 Water 0.55 -0.16 68 Table 9: Observed (Obs.) and expected (Exp.) nearest conspecific neighbour distances and indices of aggregation (R) for nest distributions of Red Phalaropes in inland and coastal habitats, 2000-2002. Y e a r Nest Locat ion n Obs. ± S E ( m ) E x p . ± S E (m) R (Obs./Exp.) z P 2000 Inland 13 287 ± 56 337 ± 5 5 0.85 -0.92 0.36 Coast 7 163 ± 5 6 365 ± 82 0.45 -2.46 0.01 2001 Inland 5 353 ± 50 599±159 0.59 -1.55 0.12 Coast 11 142 ± 4 9 364 ± 50 0.39 -4.46 0.00 2002 Inland 7 738 ± 96 675±151 1.09 0.42 0.67 Coast 12 159 ± 15 329 ± 59 0.48 -2.87 0.00 69 Table 10 : Mayfield estimates of nest survival (±95% confidence intervals) over a 19-day incubation period for Red Phalaropes nesting >150m away (No Gulls) and <150m away (With Gulls) from Sabine's Gulls in 2000-2002. Nests Year No Gulls With Gulls 2000 Period Survival 0.23 0.43 95% CI 0.09 - 0.58 0.19-0.97 n 12 8 2001 Period Survival 0.25 0.42 95% CI 0.06 - 0.93 0.19-0.88 n 6 10 2002 Period Survival 0.18 0.00 95% CI 0.04 - 0.68 0.00 - 0.08 n 8 11 70 T a b l e 11: Cox's proportional hazard regression models of nest survival for the Red Phalaropes of East Bay. Final models were created through backwards removal using likelihood ratios (criteria: a = 0.05). Only covariates with significant Wald statistics for > 1 group of nests are displayed. Variables assessed, but not found significant for any groups of nests include: with/without gulls, % overhead concealment, % lateral concealment, PC 1-1 m 2 , PC2-1 m 2 , nearest Sabine's Gull nest, nearest Red Phalarope nest, and number of Sabine's Gull nests within 150 m. Protected Days / Avg. Height of Full Model Variables Included Survival Days * Veg etation PCI -75m2 PC2-75m in Final Model Year (mm) Nests Wald P Wald P Wald P Wald P X2 df P A l l Years, all nests 54 3.13 0.21 0.36 0.55 1.54 0.22 4.09 0.04* 4.19 1,54 0.04* P C 2 - 75m 2 2000 20 0.16 0.69 1.95 0.16 5.24 0.02* 5.80 1,20 0.02* P C 2 - 75m 2 2001 15 5.28 0^02* 0.07 0.80 1.17 0.28 6.23 1, 15 0.01* Average Height of Surr. Vegetation 2002 19 0.35 0.55 2.96 0.09 0.38 0.54 Inland 25 1.36 0.24 5.34 0.02* 0.11 0.74 0.99 0.32 5.68 1,25 0.02* Average Height of Surr. Vegetation Coastal 29 7.08 0.03* 2.51. 0.11 4.88 0.03* 3.09 0.08 12.56 3, 29 0.01* PCI - 75m 2, Prot./Surv. Days*Year 71 Table 12: Mayfield estimates for daily survival of artificial shorebird nests at East Bay in 2002. Nest were placed 50 m, 100 m or 200 m away from Sabine's Gulls' nests (Coastal With Gulls), random coastal sites >400 m away from any Sabine's Gull nest (Coastal No Gulls) or random inland sites (Inland). Except for a small sample of 4 egg nests, all nests contained 2 eggs of the Japanese Quail (Coturnix japonica). Eggs were exposed between July 8 t h and 19 th for Round 1, and between July 17 th and 28 t h in Round 2. Round 1 Round 2 Treatment Daily Survival ± S E Difference Between Groups Daily Survival ± S E Difference Between Groups Coastal With Gulls 36 0.937 ±0.015 0.884 ± 0.023 Coastal No Gulls 32 0.972 ±0.010 X 2 2 =9.69, 0.905 ±0.021 X22 =8.18, Inland 32 0.987 ± 0.007 /> = 0.01 0.952 ±0.013 P = 0.02 Coastal With Gulls Only 50m 18 0.940 ±0.021 0.885 ±0.031 100m 9 0.983 ±0.017 X2 2= 3.26, 0.884 ± 0.049 x22=o.oo, 200m 9 0.943 ± 0.028 P = 0.20 0.882 ± 0.045 P= 1.00 Inland 4 Egg Nests 6 0.871 ± 0.060 Z = 1.32, 2 Egg Nests 32 0.952 ± 0.013 P = 0.19 Coastal No Gulls 4 Egg Nests 6 0.889 ±0.052 Z = 0.29, 2 Egg Nests 32 0.905 ±0.021 P=0.77 * Coastal With Gulls 36 0.968 ±0.010 0.927 ±0.017 * Coastal No Gulls 32 0.979 ±0.008 X22=2A4, 0.921 ±0.018 X 2 2 =3.32, * Inland 32 0.987 ± 0.007 P=0.30 0.957 ±0.013 P = 0.20 * With punctured eggs (possibly by Ruddy Turnstone, Arenaria interpres) not included as depredated. 72 Table 13: Cox's proportional hazard regression models for survival of artificial nests at East Bay in 2002. Final models were created through backwards removal using likelihood ratios (criteria: a = 0.05). Only covariates with significant Wald statistics for > 1 group of nests are displayed. Variables assessed but not found significant for any groups of nests include: days "with gulls", % lateral concealment, PC1-1 m2 and PC 1-75 m 2 . % Overhead Cone. Avg. Height of Veg. (mm) PC2-lm~ PC2-75m- Full Model Variables Included in Final Model Wald Wald Wald Wald X2 df Round 1 Inland 32 0.92 0.34 0.18 0.68 Coastal No Gulls 32 5.87 0.02* 0.36 0.55 Coastal With Gulls 36 1.54 0.22 0.82 0.36 0.91 0.34 2.28 0.13 0.32 0.57 0.13 0.72 2.91 0.09 0.14 0.71 7.59 1,32 0.006 % Overhead Concealment Round 2 Inland 32 0.23 0.63 0.16 0.69 0.46 0.50 1.62 0.20 Coastal No Gulls 32 1.07 0.30 0.00 0.95 1.50 0.22 0.81 0.37 Coastal With Gulls 36 1.69 0.19 4.87 0.03* 4.72 0.03* 4.67 0.03* 8.37 2,36 0.015 Avg. Height of Veg , PC2-lm 2 73 Table 14: G L M analyses of incubation behaviour of Red Phalaropes at East Bay in 2002. Final models were created through backwards removal likelihood ratio analysis (criteria: a = 0.05). Separate analyses were conducted for A) all complete 24 h records (n = 42 records from 9 incubators) and B) means for incubators (n = 9). Where "Nest" (random factor) or "Gulls" (with/without: fixed factor) were significant, they were included in all subsequent tests of individual predictors. Only covariates with significant F statistics for > 1 dependent variable are displayed. Variables assessed but not found significant in any analyses include: date (and date2), incubation stage (and stage2), habitat principal components, % lateral and overhead concealment and height of the materials surrounding the nest cup. Nest Gulls? Daily Avg. Predicted (random f.) (fixed f.) Windchill Full Model F P F P F P F df P Variables A) Al l Complete Days % Attentiveness 14.0 0.00* 0.4 0.51 1.7 0.20 14.0 9, 42 0.00* Nest No. Recesses/24hr 3.5 0.01* 0.0 0.92 0.1 0.77 3.5 9, 42 0.01* Nest Mean Recess Dur. (Min.) 2.9 0.01* 0.1 0.76 0.9 0.36 2.9 9, 42 0.01* Nest B) Averages for Incubators % Attentiveness 9.9 0.02* 2.2 0.19 9.9 1,9 0.02* Gulls? No. Recesses/24hr 3.7 0.10 7.3 0.03* 7.3 1,9 0.03* Windchill Mean Recess Dur. (Min.) 2.6 0.15 1.8 0.22 - -74 Figures 0.8 Arctic Fox 2000 2001 2002 Year Figure 2: The number of sightings per day of arctic foxes (Aiopex lagopus), jaegers {Stercorarius parasiticus and S. longicaudus) and lemmings (Dicrostonyx groenlandicus), May-August, 2000-2001, at East Bay, Nunavut, Canada. Note that jaegers were the most abundant predator in all years and are represented with a different scale. 76 30 c I 20 ro <u o c o O ro ro 10 i 57 24 J L 179 T 21 _T_ 63 24 T o 30 20 c E ro <v u c o o T3 ro £ 10 > X. Random BBPL RUTU SEPL REPH WRSA Site Species Figure 3: Lateral (A) and overhead (B) concealment of shorebird nest sites and random sites at East Bay. Means are displayed ±SE. Four letter species codes are displayed: Black-bellied Plover (BBPL), Red Phalarope (REPH), Ruddy Turnstone (RUTU), White-rumped Sandpiper (WRSA), Semipalmated Plover (SEPL). 77 Black-bellied Plover c o | 0) </) n O n— o c o r o Q . O 0.5 0.4 -0.3 • 0.2 • 0.1 -0.0 Random Nests 1 • 1 More crust, lichen, rock at site. Less scrub willow habitat. -4 -2 0 More lichen at site. More sedge meadow habitat. Discriminant Function Score Figure 4: The distribution of scores from discriminant function analyses of random sites versus nest sites (A), and failed versus successful nest sites (B) for Black-bellied Plovers. Habitat variables listed below the abscissa describe the basis of the discrimination. Variables with loadings >|0.40| are included, to a maximum of the four most important. Significant discriminations are denoted with an asterisk. 78 (rt c o to > u (rt o o c o r o a o i_ Q. 1.0 Red Phalarope 0.8 -I 0.6 0.4 0.2 0.0 • • I Random CZ] Nests -—mr-a- n Iil More lateral and overhead concealment, more herbs at site. >• -2 0 More scrub willow habitat. Discriminant Function Score Figure 5: The distribution of scores from discriminant function analyses of random sites versus nest sites (A), and failed versus successful nest sites (B) for Red Phalaropes. Habitat variables listed below the abscissa describe the basis of the discrimination. Variables with loadings >|0.40| are included, to a maximum of the four most important. Significant discriminations are denoted with an asterisk. 79 (rt C O '-5 l_ (rt o o c o o a o i_ a. 0.5 Ruddy Turnstone 0.4 0.3 H 0.2 0.1 -0.0 Random I I Nests n i I L More rock and crust at site. Less dry heath and gravel ridge habitat. -4 -2 0 More lateral concealment. Discriminant Function Score Figure 6: The distribution of scores from discriminant function analyses of random sites versus nest sites (A), and failed versus successful nest sites (B) for Ruddy Turnstones. Habitat variables listed below the abscissa describe the basis of the discrimination. Variables with loadings >|0.40| are included, to a maximum of the four most important. Significant discriminations are denoted with an asterisk. 80 Semipalmated Plover c o TO > 0) V) n O *-o c o tS o Q. o a. 2 1.0 0.4 -J 0.2 0.0 More crust at site. More intertidal zone and moss carpet habitats. B :_ i , ttm Failed mi3 Successful NI I -2 0 More moss carpet habitat. Discriminant Function Score Figure 7: The distribution of scores from discriminant function analyses of random sites versus nest sites (A), and failed versus successful nest sites (b) for Semipalmated Plovers. Habitat variables listed below the abscissa describe the basis of the discrimination. Variables with loadings >|0.40| are included, to a maximum of the four most important. Significant discriminations are denoted with an asterisk. 8 ! White-rumped Sandpiper c o > 0) (/> O * -o c o r o a o More willow and lateral concealment at site. More sedge meadow habitat -2 0 2 More dirt and moss at site. Less sedge meadow habitat. Discriminant Function Score Figure 8: The distribution of scores from discriminant function analyses of random sites versus nest sites (A), and failed versus successful nest sites (B) for White-rumped Sandpipers. Habitat variables listed below the abscissa describe the basis of the discrimination. Variables with loadings >|0.40| are included, to a maximum of the four most important. Significant discriminations are denoted with an asterisk. 82 0.20 , I 1 — « r p i I—i BBPL REPH RUTU SEPL WRSA Species Figure 9 : Mayfield estimates of the daily mortality rate (±SE) for shorebird nests at East Bay 2000-2002. The number of nests monitored in each year is displayed above the bars. 83 0.10 0.08 -ra a: £T 0.06 ra Habitat Type Figure 10: The daily mortality rate (±SE) for shorebird nests in each of the six habitat types at East Bay. Nests of all species found in a given habitat were combined to generate indices of predation pressure across habitats. 84 Round 1 [ = ] Round 2 Habitat Type Figure 1 1 : The proportion of artificial nests taken by predators in each of the six habitat types at East Bay. Nests contained two Japanese Quail (Coturnix japonica) eggs, and were deployed from 8-10 July until 17-19 July, 2002 (Round 1), and at the same sites from 17-19 July until 26-28 July, 2002 (Round 2). The numbers of nests placed in each habitat are displayed above the bars. 85 0.15 BBPL REPH RUTU SEPL WRSA Species F i g u r e 12: The daily mortality rate (±SE) for shorebird nests at East Bay placed in preferred versus non-preferred habitats. Preferred habitat is the habitat type where nests were most commonly placed (see Table 2). 86 Figure 13: The mean volume (ml) of invertebrates per pitfall trap captured in each of the six habitats types at East Bay. Traps were emptied three times in 2000: 8 July, 16 July and 24 July and three times in 2001: 7 July, 15 July, 23 July. The mean volume per trap per collection period is presented separately for 2000 and 2001. 87 2.5mm 10Kfl± 1% 7 | \ i o K n Curve G(9) Figure 14: A schematic diagram of the probes used to log nest temperature and determine incubation behaviour of Red Phalaropes. 88 East Bay Red Phalarope Nests ^ ^ ^ ^ ^ A K Sabine's Gull Nests N Areas "with g ulls" A j 1 Coastal Habitats /A QO Inland Habitats Figure 15: The distribution of Red Phalarope and Sabine's Gull nests in 2000 (A), 2001 (B) and 2002 (C). Areas within 150 m of Sabine's Gull nests are marked as "with gulls". The northern boundary of the coastal habitats is the approximate high tide mark of East Bay. Intensive search was limited to the central 7 km 2 of the 12 km 2 plot shown in 2000 and 2001. The raised beach where camp was located is marked. 89 No Gulls 2000 With Gulls No Gulls With Gulls 2001 No With Gulls Gulls 2002 Figure 16: Daily rate of nest mortality (±SE) for Red Phalaropes >150 m away (No Gulls) and <150 m away (With Gulls) from Sabine's Gulls in 2000-2002. An asterisk denotes a difference with P< 0.05. 90 Appendix I - Detailed Description of Habitat Types Study area habitats were subjectively classified into 6 visually distinct types: sedge meadow, dry heath, gravel ridge, moss carpet, scrub willow, and intertidal areas. Patches considered representative of each habitat type were selected, and within each, 30 random sites were generated with a random distance and compass bearing. At each site, ground cover proportions for flora and substrate were recorded, the dominant plant species noted and the substrate characteristics quantified. For habitats with a distinct hummocked relief, the maximum height of 10 randomly selected hummocks was recorded. For intertidal zone, moss carpet and sedge meadow, maximum sedge height was measured (± 1mm) at 35 randomly selected points. The sets of 30 random sites from each habitat category were used to generate means and confidence intervals for graphical representation. Due to extreme deviations from normality, not eliminated through arcsine transformation, statistical comparisons of percent covers were made using Mann-Whitney (/-tests. Sedge and hummock heights were compared between habitats using independent sample Mests. Habitat types are not distributed evenly throughout the study area, but rather in bands parallel to shore. Intertidal zone habitat is found exclusively near the shore, where it is the dominant type. Above this, moss carpet borders the ponds, with scrub willow found farther from the margins. Several hundred metres south, away from the shore, moss carpet still borders the ponds, but sedges are noticeably and significantly taller (the later is termed moss carpet A , the former, moss carpet B). Farther away from the shore, sedge meadows replace moss carpets in the moist areas and the scrub willow is replaced by dry heath. This transition of pond edge habitat type, from intertidal zone, to moss carpet A , to moss carpet B, to sedge meadow is accompanied by a significant change in the height of vegetation (ANOVA, F 3 / 1 1 1 = 78.01, P < 91 0.001, Fig. A - l ) , suggestive of an underlying gradient in plant productivity. Complete species lists for all East Bay habitats appear in Fontaine and Mallory (in prep.). 80 Intertidal Zone Moss Carpet 6 Moss Carpet A Sedge Meadow Shore-* Habitat Type > Inland Figure A - l : Mean sedge height for the pond edge habitats of East Bay, Southampton Island, ordered from coastal to inland. For each habitat, n = 30 measurements. Error bars represent 95% CPs. D R Y H E A T H The dry heath habitat is dominated by mountain avens (Dryas integrifolia) and exposed substrate (Fig. A-2), with some patches of saxifrage (Saxifraga oppositifolia, S. hirculus) and other herbs (e.g., Papaver radicatum, Draba spp.). Much of the moss present in this habitat is blackened as a cryptobiotic or cryptogamic crust. Fruticose lichens are common. The relief varies substantially between flat and extremely hummocked. Hummocked dry heath is the most highly hummocked of all habitat types, with an average hummock height of 150 ± 32 mm, significantly higher than that of hummocked sedge meadows (t = 5.56, P < 0.001). 92 Q) > O O +-> c <D O L _ <D CL 100 90 80 70 60 50 40 30 20 10 0 Dry Heath Moss Sedge Rock /Grave l Soil Avens Wil low Herbs Cover Type Figure A-2: Mean percent ground cover of various types for 30 1 m random sites in a patch selected to best represent dry heath. Error bars represent 95% CI's. SEDGE MEADOW With little exposed substrate and an abundance of tall grarhminoids (Carex aquatilis, C. subspathacaea, C. marina, C. glacialis, C. fuliginosa, Arctagrostis latifolia), this is the most lushly vegetated habitat of the study plot (Fig. A-3). Moss covers the peat substrate entirely (e.g., Campylium stellatum, Scorpidium scorpioides). Sedges and grasses are significantly higher than in any other habitat (Fig. A - l ) , and herbs are common (especially Draba spp.). As with dry heath, relief varies from flat to highly hummocked. (D > O O 100 90 80 70 60 50 40 c 2 30 0 20 10 0 Sedge Meadow _______ _____ Moss Wi l low Herbs Sedge Rock /Grave l Soil Avens Cover Type 2 Figure A-3: Mean percent ground cover of various types for 30 lm random sites in a patch selected to best represent sedge meadow. Error bars represent 95% CI's. 93 G R A V E L R I D G E This habitat is composed almost entirely of exposed shale fragments (Fig. A-4) . Ericaceous shrubs (esp. Dryas integrifolia) and some herbs (e.g., Papaver radicatum, Pedicularis lanatd) are present in low abundance. 100 Moss Sedge Rock/Gravel Soil Avens Willow Herbs Cover Type Figure A-4: Mean percent ground cover of various types for 30 random l m sites in a patch selected to best represent gravel ridge. Error bars represent 95% CI's. Moss C A R P E T While both moss carpet and sedge meadow have nearly complete moss ground cover, moss carpet is distinguished by sparser gramminoids (Latest, Z = 5.81, P < 0.0001). Several species of herbs are present, but rarely comprise any significant proportion of the ground cover (Fig. A-5). Moss carpet ' A ' and ' B ' differ only in sedge height (See Fig. A - l ) . 94 CD > o O (D O CL 100 90 80 70 60 50 40 30 20 10 0 Moss Carpet Moss Sedge Rock/Grave l Soil Avens Wil low Herbs Cover Type Figure A-5: Mean percent ground cover of various types for 30 lm random sites in a patch selected to best represent moss carpet. Error bars represent 95% CPs. SCRUB WILLOW An abundant and frequently used habitat, scrub willow is highly variable. The distinguishing feature of this habitat is the high proportion of willow, significantly higher than in any other habitat (e.g., Scrub Willow v. Dry Heath, (/-test, Z = -4.00, P < 0.0001, Fig. A-6). Mountain avens is present in some scrub willow patches, but the average percent ground cover is less than in dry heath (cZ-test, Z = 6.62, P <0.0001). Though not a significant element of the ground cover, herbs in this habitat are diverse. 95 > O O 0) o CD CL 100 90 80 70 60 50 40 30 20 10 0 Scrub Willow Moss Sedge Rock/Grave l Soil Avens Cover Type Wil low Herbs Figure A-6: Mean percent ground cover of various types for 30 l m 2 random sites in a patch selected to best represent scrub willow. Error bars represent 95% CI's. INTERTIDAL ZONE This barren habitat is comprised largely of exposed substrate: predominantly large rocks and silt (Fig. A-7). Farther from the shore, in the southern regions of the intertidal zone, dwarfed gramminoids and bryophytes are present, though rarely as a significant element of the ground cover. 0) > O O 0 o CL 100 90 80 70 60 50 40 30 20 10 0 Intertidal Zone Willow Herbs Moss Sedge Rock/Grave l Soil Avens Cover Type Figure A-7: Mean percent ground cover of various types for 30 lm random sites in a patch selected to best represent intertidal zone. Error bars represent 95% CI's. 96 Appendix II - Habitat Characteristics of Nest Sites Table A - l : The mean values for habitat characteristics (±SE) of shorebird nests at East Bay, Nunavut, from 2000-2002. Four letter species codes are displayed: Black-bellied Plover (BBPL), Red Phalarope (REPH), Ruddy Turnstone (RUTU), White-rumped Sandpiper (WRSA), Semipalmated Plover (SEPL). Species B B P L REPH R U T U WRSA SEPL n = 21 N = 57 n = 63 n = 24 N = 24 % Cover Water 0 ± 0 1 ± 1 0 ± 0 0 ± 0 0 ± 0 at l m 2 Rock 29 ± 5 26 ± 3 37 ± 2 3 ± 1 50 ± 5 BLS 23 ± 4 18 ± 2 29 ± 2 14 ± 2 32 ± 5 Lichen 9 ± 2 2 ± 0 4 ± 1 3 ± 1 2 ± 1 Moss 3 ± 2 24 ± 3 14 ± 1 45 ± 4 6 ± 2 Dirt 9 ± 2 8 ± 2 8 ± 1 9 3 10 ± 3 Aven 25 ± 4 6 ± 2 2 ± 1 9 ± 3 0 ± 0 Willow 3 ± 1 7 ± 1 6 ± 1 13 ± 2 0 ± 0 Sedge /Grass 1 ± 0 9 ± 2 2 ± 0 15 ± 2 5 ± 2 Misc. Herbs 0 ± 0 5 ± 2 2 ± 0 0 ± 0 1 ± 1 % Cover Gravel Ridge 10± 7 0 ± 0 0 ± 0 0 ± 0 0 ± 0 at 75m2 Dry Heath 60 ± 10 11 ± 4 2 ± 1 13 ± 6 0 ± o Scrub Willow 19± 8 23 ± 5 45 ± 5 14 ± 5 6 ± 20 Sedge Meadow 5 ± 3 25 ± 5 3 ± 2 70 ± 7 0 ± 0 Moss Carpet 3 ± 2 14 ± 3 25 ± 3 3 ± 2 8 ± 17 Intertidal 1 ± 1 9 ± 3 18 ± 4 0 ± 0 81 ± 6 Rock 0 ± 0 3 ± 2 1 ± 1 0 ± 0 0 ± 0 Water 0 ± 0 5 ± 2 2 ± 1 0 ± 0 0 ± 0 Dried Pond 0 ± 0 11 ± 3 3 ± 1 1 ± 1 4 ± 7 Nearest Water 39 ± 5 22 ± 3 12 ± 1 47 ± 6 19 ± 12 Nearest Dried Pond 20 ± 2 9 ± 2 7 ± 1 20 ± 4 7 ± 11 Overhead Concealment 0 ± 0 10 ± 1 1 ± 0 10 ± 2 0 ± 1 Avg. Lateral Cone. 6 ± 1 23 ± 2 8 ± 1 14 ± 2 6 ± 6 Avg. Height Vegetation 16 ± 2 45 ± 3 18 ± 1 54 ± 4 12 ± 8 97 Appendix III - Nest Defence by Sabine's Guls Table A-2: Behavioural responses of Sabine's Gulls to the intrusion of predators at East Bay in 2000 and 2001. The number of encounters (n), the predators' distance from the nest at which the attack was initiated and the proportion of attacks involving group defence were recorded through behavioural observation. The intensity of response (Response Score) varied as follows: 0 = no response, 1 = flies out, 2 = chases, 3 = chases and calls, 4 = swoops/dives, 5 = strikes predator. Data adapted from Stenhouse, unpublished. Year Predator n Distance from Mean Group nest (m) „ ^ Defence (%) Score 2000 HerringGull 27 126.4±64.0 2.3 ±1.1 55 Parasitic Jaeger 2 162.5 ±17.7 4.5 ± 0.7 100 2001 HerringGull 53 166.4 ±73.6 3.1 ±0 .9 55 Parasitic Jaeger 18 189.2 ±101.9 3.2 ±1 .3 33 Glaucous Gull 2 275.0 ±106.1 4.0 ± 0.0 100 Peregrine Falcon 1 350 100 Arctic fox 250.0 ±70.7 3.5 ±0 .7 100 Human 19 145.5 ±92.8 1.9 ± 1.4 38 98 

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.831.1-0091257/manifest

Comment

Related Items