UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Timing of reproduction by red-tailed hawks, northern goshawks and great horned owls in the Kluane Boreal… Doyle, Frank I. 2001

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

Item Metadata

Download

Media
831-ubc_2001-0021.pdf [ 5.36MB ]
Metadata
JSON: 831-1.0099513.json
JSON-LD: 831-1.0099513-ld.json
RDF/XML (Pretty): 831-1.0099513-rdf.xml
RDF/JSON: 831-1.0099513-rdf.json
Turtle: 831-1.0099513-turtle.txt
N-Triples: 831-1.0099513-rdf-ntriples.txt
Original Record: 831-1.0099513-source.json
Full Text
831-1.0099513-fulltext.txt
Citation
831-1.0099513.ris

Full Text

TIMING OF REPRODUCTION B Y RED-TAILED H A W K S , N O R T H E R N G O S H A W K S A N D G R E A T HORNED OWLS IN T H E K L U A N E B O R E A L FOREST OF SOUTHWESTERN Y U K O N . by F R A N K I. D O Y L E A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF T H E REQUIREMENTS FOR T H E D E G R E E OF M A S T E R OF SCIENCE in T H E F A C U L T Y OF G R A D U A T E STUDIES Department of Zoology We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH C O L U M B I A November 2000 © Frank I. Doyle, 2000  In presenting  this thesis in partial fulfilment of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may department or by  his or her  representatives.  be granted by the head of  It is understood that copying or  publication of this thesis for financial gain shall not be allowed without my permission.  zLOOL&ty^f  Department of  The University of British Columbia Vancouver, Canada Date  DE-6  (2/88)  ^  J  H  /  O  O  my  written  ABSTRACT  In this thesis I examine the timing of breeding in 3 raptorial birds, red-tailed hawks (Buteo jamaicensis), goshawks (Accipiter gentilis) and great horned owls (Bubo virginianus). Specifically, I test Lack's 1954 theory that birds typically begin to breed such that the young bird's greatest demand for food will coincide later with the greatest abundance of available prey. Lack's theory predicts that birds which successfully match the timing of breeding to the peak in prey fledge more young than pairs which do not. This study was part of the larger Kluane Boreal Forest Ecosystem Project (Krebs et al. In press). Detailed information on weather, prey density and timing of the peaks in prey availability was gathered annually. I examined the timing of breeding over 8 years (1989 1996) in an environment with both harsh winter weather and cyclic prey populations. The 3 species differ in both life history and morphology and yet share the same prey base for much of the year. These shared prey enabled me to explore if and how the three species adjusted their breeding time to exploit the peaks in prey availability. The migrant red-tailed hawk bred every year and adjusted breeding to match the peak in prey availability. It bred early when its main prey, the Arctic ground squirrel, bred early thus ensuring that prey consumption needs of young red-tailed hawks corresponded with the synchronous emergence of young ground squirrels. In contrast, the resident great horned owl only bred when its main prey was abundant. Great horned owls have a long breeding period, and therefore bred before a peak in prey was available to match the peak in prey consumption demands of its young. The other resident, the goshawk, has a shorter breeding period than the owl, and bred such that a broad peak in available prey matched the peak consumption needs of its young. However, it is unclear if the goshawk adjusted breeding to match the predicted peak in prey, or if it had an average breeding time that corresponded to the average breeding time of its preferred prey. Slight annual variation in the timing of breeding in the goshawk, like that of the other resident, the owl, may have been in response to fluctuations in winter prey density. These fluctuations in prey availability could through its influence on the raptors body condition, reduced the bird's ability to breed at the average time to which they have become adapted. Alternatively, goshawks may have adjusted their timing of breeding in response to a predicted later broad peak in prey, initiated by the  increased availability of those prey during mating. Within a year, seasonal declines in the number of young that fledged were absent in the two resident species, but present for the migrant red-tailed hawk. Reduced fledging success in red-tailed hawks was attributed to attacks on the young by blackflies. Attacks from blackflies were so intense in 2 of the years that all young hawks from 5 of the 12 intensively monitored nests died. These nests were all late and the young were <3 weeks of age at the time of blackfly emergence.  Attacks by blackflies on young red-tailed hawks,  largely explained why the birds failed to attain maximum reproductive success every year, even though they matched peak needs by their young to the peak in prey availability. The resident goshawks and great horned owls only bred when they could do so successfully, while all red-tailed hawks bred every year. I propose that migration by the redtailed hawk avoided the loss of body condition caused by low prey abundance in winter. However, the red-tailed hawk had a lower overall success rate, with only 25% of pairs fledging young in the least successful year. Over all years, when birds did breed, red-tailed hawks typically fledged 1-2 young and both goshawks and horned owls 2 -3 young.  T A B L E OF CONTENTS  ABSTRACT  ii  TABLES  viii  FIGURES  ix  APPENDIX  xi  ACKNOWLEDGEMENTS INTRODUCTION  xii 1  Environmental factors affecting body condition and timing of breeding  2  Life history and timing of breeding  4  Genotypic effects on timing of breeding  4  THIS S T U D Y  5  METHODS  6  STUDY A R E A  6  L O C A T I N G PAIRS A N D NESTS  7  Red-tailed hawk  7  Northern goshawks  8  Great horned owl  9  M O N I T O R I N G NESTS, NESTLINGS A N D DIET  9  D E T E R M I N A T I O N P R E Y SPECIES A N D A G E I N PELLETS A N D R E M A I N S  11  A N N U A L C H A N G E S I N P R E Y DENSITIES A N D TIMING OF T H E A N N U A L P E A K S I N P R E Y  11  Methods used to estimate prey densities  12  T I M I N G OF T H E A N N U A L P E A K S I N P R E Y A B U N D A N C E  13  Mammals  13  Avian prey  13  G R A P H I C A L A N A L Y S I S OF PEAKS I N P R E Y A B U N D A N C E , DIET A N D P R E Y CONSUMPTION .. 14 WEATHER  15  BLACKFLY AND ABUNDANCE  16  DEFINITIONS OF V A R I A B L E S  16  STATISTICAL A N A L Y S I S  17  RESULTS.  Prey and Weather  18  PREY AVAILABILITY  18  Mammals  18  Birds  18  V A R I A T I O N I N A N N U A L P R E Y DENSITY  19  Mammals  19  Birds  20  Prey weights  20  WEATHER  20  Correlation between precipitation and temperature  21  P R E Y DENSITY, TIMING OF REPRODUCTION A N D W E A T H E R  21  Mammals  22  Birds  22  Correlations between population density and timing of breeding in prey  22  BLACKFLY ABUNDANCE  23  TIMING OF B R E E D I N G B Y T H E THREE RAPTORS  23  (A) RED-TAILED H A W K  23  Annual variation in timing of breeding and reproductive performance  23  Links between timing of breeding and reproductive performance  23  Prey availability and timing of the peak in prey consumption by the young  24  Diet and reproductive performance  25  Weather and the annual changes in reproductive performance  26  Weather, blackflies and reproductive performance  26  Summary: Red-tailed hawk  27  (B) N O R T H E R N G O S H A W K  27  Annual variation in timing of breeding and reproductive performance  27  Timing of breeding and reproductive performance  28  Prey availability and timing of the peak in prey consumption by the young  28  Diet and reproductive performance  29  Weather, blackflies and reproductive performance  30  Summary: Northern goshawk  30  (C) G R E A T H O R N E D O W L RESULTS  31  Annual variation in timing of breeding and reproductive performance  31  Timing of breeding and reproductive performance  31  Prey and timing in peak food demands of the young  31  Diet and reproductive performance  32  Weather, blackflies and reproductive performance  33  Summary: Great horned owl  33  DISCUSSION  34  EFFECTS OF WEATHER ON BREEDING  35  Weather as a cue in the timing of breeding  35  Effects of weather on the reproductive performance of raptors and their prey  35  Summary: Weather  36  T I M I N G OF B R E E D I N G : R E D - T A I L E D H A W K  37  Responses to prey density  37  Responses to juvenile prey  37  Constraints on breeding time  38  T I M I N G OF B R E E D I N G : N O R T H E R N G O S H A W K  40  Responses to prey density  40  Responses to juvenile prey  41  Constraints on breeding time  41  T I M I N G OF B R E E D I N G : G R E A T H O R N E D O W L  43  Responses to prey density  43  Responses to juvenile prey  43  Constraints on breeding time  44  Summary: Timing of breeding in raptors  46  G E N E T I C CONSTRAINTS O N TIMING OF B R E E D I N G  46  S Y N C H R O N I Z I N G P R E Y A V A I L A B I L I T Y WITH TIMING OF B R E E D I N G  46  I N T E R A C T I N G E F F E C T S OF FOOD A N D A G E O N V A R I A B I L I T Y O F B R E E D I N G D A T E S  48  A S S U M P T I O N S A N D BIASES I N THIS THESIS  49  L A C K ' S T H E O R Y A N D ITS A L T E R N A T I V E S  51  C O M P A R I S O N WITH OTHER STUDIES OF T H E TIMING O F B R E E D I N G I N BIRDS  53  FUTURE RESEARCH  54  L I T E R A T U R E CITED  56  TABLES Table 1. Comparisons between the life history characteristics and morphology of the raptors  71  Table 2. Characteristics of timing of litters and broods in prey  72  Table 3. Mean parturition dates for snowshoe hares  73  Table 4. Red-tailed hawk laying and fledging dates  74  Table 5. Red-tailed hawk reproductive performance (1990 - 1996)  75  Table 6. Red-tailed hawk. Seasonal variation in fledging success  76  Table 7. Red-tailed hawk. Diet  77  Table 8. Red-tailed hawk. Blackflies recorded at nests  78  Table 9. Red-tailed hawk. Causes of breeding failures  79  Table 10. Northern goshawk laying and fledging dates  80  Table 11. Northern goshawk reproductive performance from 1989 - 1996  81  Table 12. Northern goshawk. Seasonal variation in fledging success  82  Table 13. Northern goshawk diet  83  Table 14. Great horned owl mean laying and fledging dates  84  Table 15. Great horned owl reproductive performance (1989 - 1996)  85  Table 16. Great horned owl. Seasonal variation in fledging success  86  Table 17. Great horned owl diet  87  Table 18. Correlations between a raptors main prey and reproduction  88  Table 19. Correlations between the annual peak in the raptors main prey and reproduction  89  viii  FIGURES  Figure 1. Relationship between prey vulnerability and its inceasing biomass  15  Figure 2. Map of the study area at Kluane Lake, southwestern Yukon  90  Figure 3. Dates when young of the common prey taken by the three raptors become available  91  Figure 4. Annual index of prey species abundance  94  Figure 5. Correlation between March/April temperature and ground squirrel emergence dates  95  Figure 6. Annual mean fledging date of all three raptors  96  Figure 7. Red-tailed hawk. Annual fledging date and the proportion of monitored pairs fledged young  97  Figure 8. Red-tailed hawk. Numbr fledged per pair vesrus timing of breeding  97  Figure 9. Red-tailed hawk. Relationship between breeding and prey consumption  98  Figure 10. Red-tailed hawk. Diet  100  Figure 11. Red-tailed hawk. Fledge date versus ground squirrel emergence date  101  Figure 12. Red-tailed hawk. Proportion fledging young and ground squirrel emergence date  101  Figure 13. Red-tailed hawk. Fledged per pair and ground squirrel emergence date  102  Figure 14. Red tailed hawk. Ground squirrel in the diet versus squirrel density  103  Figure 15. Red-tailed hawk. Fledging date versus temperature in March and April  104  Figure 16. Red-tailed hawk. Number fledged from successful nest and temperature in June Figure 17. Red-tailed hawk. Nest failures  105 106  Figure 18. Northern goshawk. Number fledging per pair versus timing of breeding .... 107 Figure 19. Northern goshawk. Relationship between breeding dates and prey consumption Figure 20. Northern goshawk. Diet  108 110  Figure 21. Northern goshawk. Relationship between the annual fledging date and hare density  Ill  Figure 22. Northern goshawk. Relationship between pairs located annually and hare  IX  density  Ill  Figure 23. Northern goshawk. Timing of the 1 hare litter and the proportion of pairs st  fledging young  112  Figure 24. Northern goshawk. Hare population density vesus percentage in the diet.... 113 Figure 25. Great horned owl. Fledge date vesus proportion of pairs fledging young.... 114 Figure 26. Great horned owl. Fledge date and number of young fledged  115  Figure 27. Great horned owl. Relationship between breeding dates and prey consumption  116  Figure 28. Great horned owl. Diet  118  Figure 29. Great horned owl. Hare population density versus percentage juvenile hare in the diet Figure 30. Great horned owl. Fledge date versus hare density in spring  119 119  Figure 31. Great horned owl. Proportion of snowshoe hares in the diet veusus fledge date Figure 32. Great horned owl. Number fledged per pair and snowshoe hare density  120 121  Figure 33. Great horned owl. Number fledged per pair versus and proportion of juvenile hares in the diet  121  APPENDIX  Appendix 1. Young Arctic ground squirrel emergence dates  122  Appendix 2. Young red squirrel emergence dates  123  Appendix 3. Spruce grouse hatch dates  124  Appendix 4. Snowshoe hare population density  125  Appendix 5. Arctic ground squirrel population density  126  Appendix 6. Red Squirrel population density  127  Appendix 7. Vole (Microtus  spp. and Clethrionomys  rutilus)  population density  128  Appendix 8. Spruce grouse population density index  129  Appendix 9. Passerine population density  130  Appendix 10. Mean weights of prey:  131  Appendix 11. Mean daily maximum temperature  132  Appendix 12. Mean maximum temperature and precipitation  133  Appendix 13. Number of days month when daily temperatures went above freezing.... 134 Appendix 14. Mean daily precipitation  135  Appendix 15. Number of precipitation days  136  Appendix 16. Relationship between precipitation in February and red squirrel density in fall  137  Appendix 17. Relationship between rain days in July and the density index of spruce grouse in winter  138  Appendix 18. Red-tailed hawk. Annual ranking of prey species contribution to the diets  139  Appendix 19. Northern goshawk. Annual ranking of prey species contribution to the diets  140  Appendix 20. Great horned owl. Annual ranking of prey species contribution to the diets  141  xi  ACKNOWLEDGEMENTS  This thesis, possibly more than any other to come from the work on the boreal forest ecosystem project at Kluane (funded by N.S.E.R.C.), is indebted to the information and help provided by all those involved. I am very grateful to all those who contributed. In particular, as advisors and colleagues the encouragement and guidance provided by my supervisor, Jamie Smith and the other project leaders, Rudy Boonstra, Stan Boutin and Charles Krebs was instrumental in making this thesis possible. I also owe a special thank you to Villis Nams, to whom I am indebted to for his inspiration, enthusiasm and guidance. In the field, a special thanks for teamwork and professionalism to the task in hand, goes out equally to all those who put up with the long hours, blackflies, bears and my English idiosyncrasies. As their annual turn came these dedicate individuals were rewarded by the doubtful pleasure of dangling from the tree tops, while being attacked by enraged adult raptors protecting their young. To the team, Brendan Delehanty, Tamie Hucal, Kirsten Madsen and Shawna Pelech, thank you. To Christoph Rohner a special thanks to his eye for detail, and methodological scrutiny within the framework of hectic summer days and nights. Without his contribution on the ecology of great horned owls at Kluane during the peak and crash of the hare cycle this thesis would not have been possible. Thanks also to Troy Wellicome for his assistance with red-tailed hawk data collection at the peak of the hare cycle. On the thesis text and content I am grateful for long hours and pages of critical comments from my supervisor Jamie Smith, and from the other members of my committee Charles Krebs and B i l l Neill. I am also indebted to my colleague Shawna Pelech, who was always there to critically read and comment on the next draft of this thesis. For the direction and focus of the research, and needed encouragement, I would also like to thank Andrea Byrom and Lance Barrett-Lennard. M y parents, Kathleen and Kevin, have a special role to play in this thesis, as their enthusiasm and love of nature brought me to this place in time. To Robert Kenward, my first teacher and mentor, thank you. In my work at Kluane, the everyday friendship and support, provided by the community at the Arctic Institute and Silver City, was pivotal in making this research and thesis possible. In particular, I would like to thank Andy Williams for the TOOLS to get the job done and Jan Williams for the INGREDIENTS to keep me going. Finally, a special thanks to my partner in life, Cathy, who through her enthusiasm and critical insights, supported and guided this research both in the field and during the production of this manuscript.  DEDICATED TO MY FAMILY  CATHY. GARETH. GLYN AND EVAN  INTRODUCTION  The underlying mechanisms that set the timing of a bird's breeding season are the circadian rhythm and circannual cycles, which together determine the overall pattern of the bird's seasonal physiology (Gill 1995). The circannual cycles are modified by predictive cues, particularly photoperiod, which, through the production of hormones, triggers the initiation of gonadal development (Jacobs and Wingfield 2000). Within this framework, the fine-scale tuning of life history stages, including the timing of breeding in birds, has been the focus of much research and debate over the past 40 years (Ricklefs 2000). This research is founded largely on theories postulated by Lack (1954, 1968). Lack's central idea (1954) was that the timing of breeding in birds has evolved so that in their natural habitat, each species produces, on average, the greatest possible number of surviving young. Lack concluded that birds begin to breed so that a young bird's greatest demand for food will coincide later with the greatest abundance of available prey. There are both phenotypic and genotypic components to the way in which a young bird's greatest demand for food is matched with a peak in prey abundance (Lack 1954, Drent and Daan 1980, Roff 1986). Individuals that are successful doing this should presumably produce more surviving young and thus be favored by natural selection. Work by Lack (1950) on coal tits (Parus ater) and long term studies on great tits (Parus major) (Perrins 1991, Nager and van Noordwijk 1995), tree swallows (Tachycineta bicolor) (Hussell and Quinney 1985), sparrowhawks (Accipiter nisus) (Newton and Marquiss 1982, 1984) and coots (Fulica atra) (Brinkhof 1995) support this hypothesis. These studies have shown that when peak food demand and peak food supplies coincide, more chicks are fledged per nesting attempt than when they do not. If the match between the average time of breeding and the average peak in food supplies were simply a consequence of symmetrical normalizing selection, there should be approximately as many early breeders as there are late breeders. There are, in fact, more late breeders than there are early breeders. These late breeders not only fledge fewer young per breeding attempt, but also lay smaller clutches (Perrins 1970, Drent and Daan 1980, Dann et al. 1986, Steeger and Ydenberg 1993). Late breeders however can compensate in part by  1  having smaller broods with heavier fledged weights and earlier fledging, which both increase survival of the young (Drent and Daan 1980, Perrrins and McCleery 1989, Dijkstra et al. 1990, Steeger and Ydenberg 1993). The primary hypothesis to explain the existence of late breeders in a population is that a lack of food early in the season prevents the formation of eggs (Perrins 1970). Thus, only the birds in the best condition may be able to lay at the optimal time. This theory has been tested extensively and in one recent review (Newton 1998), addition of food to 23 species prior to laying advanced laying in 96% of cases and increased chick production in 80% of cases. In raptors, a study on sparrowhawks has linked the timing of laying to improved body condition brought about by the appearance of fledgling birds in the diet (Newton and Marquiss 1982). Further, sparrowhawks breeding in poor quality territories were in poorer body condition than those pairs from good quality territories, they also bred later and produced fewer young (Newton and Marquiss 1984). In another study, of a raptor guild in the central Canadian Arctic (Poole and Bromley 1987), the appearance of a species main prey was linked to the timing of breeding, however, no information was available on body condition or breeding success with timing of breeding. In contrast to the abundance of late breeders, there is a scarcity of early breeders. In insectivorous species, this may arise due to the scarcity of food prior to the rapid increase in available prey in spring. This early scarcity preventing females attaining breeding condition too early in the season (Perrins 1970). Early breeders may be able to compensate for initiating laying too soon by extending the length of their incubation period (van Noordwijk et al. 1995). Similarly, growth of the young can be retarded by poor food supply (Ricklefs 1968). Through these influences, the timing of a chick's peak in prey requirements may still match a peak in prey, even in those birds that initiated laying early. In raptors, no evidence is available to suggest that some bird's bred too early, however, a similar rapid increase in prey is frequently observed, and it is possible that similar adjustment mechanisms are at work.  Environmental factors affecting body condition and timing of breeding  Environmental factors can work both directly and indirectly to affect body condition and the timing of breeding. In two studies, on great tits (Parus major) (Perrins and McLeery  2  1989, Nager and van Noordwijk 1995), temperature increased the abundance of food, which through improved body condition caused the initiation of laying, and in turn synchronized the needs of the young with the later peak in food abundance.  In raptors, a similar correlation  between weather and the timing of reproduction has been observed (Cave 1968, Newton 1977, Olsen and Olsen 1989, Penteriani 1997). These raptor studies did not look in detail at how weather affected the timing of the peak(s) in prey abundance. However the link between weather and timing of breeding in insectivorous birds (prey of many raptors), suggests that fluctuations in weather conditions would also indirectly affect body condition in raptors as their prey (insectivorous birds) increased in abundance. The seasonal impact of cold winter temperatures and associated lows in prey abundance, affects body condition (Drent and Daan 1980, Daan et al. 1989, Hornfeldt and Eklund 1989). Many birds build up fat stores in fall as a hedge against unpredictable feeding conditions (review in Gill 1995), yet birds still die in winter due to winter weather and/or food shortages (Arcese et al. 1992, Newton 1998). Reduced prey abundance may also be unpredictable, with cold weather in spring inhibiting the emergence of some prey species from hibernation (Michener 1979), and unfavorable winds delaying the arrival of migrant prey (review in Kerlinger 1989). Raptors may also be unable to obtain enough food when wet weather reduces hunting success (Newton 1979, Mearns and Newton 1988, Olsen 1989, Kostrzewa and Kostrzewa 1990), which may thereby deplete the available energy for breeding (Kennedy 1970). Finally, in migrant birds reduced body condition can be caused by unfavorable weather on the flight back to the breeding area (review in Gill 1995). For raptor species breeding at higher latitudes a further constraint on food supply and hence body condition is the cyclic nature of many prey populations (Hornfeldt 1978, Korpimaki and Lagerstrom 1988, Boutin et al. 1995). In these environments, both resident and migratory species of raptor experience extended periods of low prey densities (Adamcik et al. 1979, Rohner 1994, Doyle and Smith 1994). A bird's age can also affect the timing of breeding independently of the direct effects of weather and food abundance. Birds in their first breeding season may not have acquired enough hunting/foraging experience to attain good body condition to allow them to breed early in the year (Perrins 1965, Smith et al. 1980).  3  There is also evidence that body condition can be affected by parasitism. By lowering adult condition, parasites may delay the onset of breeding, and may affect the condition of the young by increasing the time needed for growth (review in Newton 1998). In cliff swallows (Petrochelidon pyrrhonota), ectoparatism lengthened the incubation period and caused a decline in seasonal reproductive success (Brown and Brown 1999). Blood parasites in Tengmalm's owls (Aegolius funereus) cause anemia and loss of body condition in females, delaying breeding and reducing clutch size (Korpimaki, et al. 1993). In fledgling great horned owls (Rohner and Hunter 1996) and nestling red-tailed hawks (Smith et al. 1998), blood parasite infections carried by ornithophilic blackflies (Simulium spp.) influence body condition and can cause death.  Life history and timing of breeding  Life history characteristics which can influence timing of breeding include the period of time needed for migration, that which is needed to raise young to independence, breeding location (birds breeding at high latitude or high elevation may have to wait for snow melt at nest sites), and the timing of breeding in the previous year. Migrants have to cope with the energy demands of moult and the time needed to complete migration in spring and fall (Gill 1995, Hemborg 1999). Timing of breeding in migrants may therefore depend on the time needed for migration, rather than on a peak in prey abundance.  Large species such as  raptors may also be restricted from breeding at the optimum time because their young take a long time to reach full body size, and need a long period of post fledging care to develop hunting skills (Newton 1979). Breeding in birds, particularly at high latitudes can also depend on the time of snowmelt from nest sites, e.g. in tundra-nesting geese (review in Newton 1977). The reproductive success of a pair in one year may influence the timing of breeding by the same pair the following year. Pairs that missed the peak in-prey in one year and fail to fledge young, adjust their timing of breeding in the following year to match the time of the peak prey in the previous year (Nager and van Noordwijk 1995).  Genotypic  effects on timing of breeding  4  In contrast to fine scale phenotypic adaptations in the timing of breeding, heritable genotypic effects on the timing of reproduction work on a broader time scale. Studies on tits (Parus sppj by van Balen (1973) and Perrins (1990), suggested a genetic component to the timing of breeding, whereby differences in the timing of prey peaks in different habitats, caused gene selection for laying at different dates in different habitats. Gene flow of birds between different habitats may result in part of a population not matching timing of breeding to the peak in prey. A recent study on blue tits (Parus caeruleus) in the South of France has confirmed this prediction. Blue tits breeding in a dominant deciduous habitat matched peak food requirements of the young with prey availability. In contrast, birds breeding in coniferous woodland did not (Blondel et al. 1993). On the island of Corsica, however, segregation of oak and coniferous in different valleys allowed local adaptation to evolve in a situation where gene flow was restricted by a geographic barrier (Lambrechts and Dias 1993).  THIS STUDY  In this study, I took the opportunity presented by a long-term ecosystem project (1986 -1996), set in the boreal forest of the southwestern Yukon (Boutin et al. 1995, Krebs et al. 1995, Krebs et al. In press), to look at the timing of breeding in three large cohabiting raptors. This environment exhibits extremes in seasonal weather and dramatic seasonal and annual changes in prey densities. Many of the prey are cyclic and many others are absent in the winter. The study species, the red-tailed hawk (Buteo jamaicensis), northern goshawk (Accipiter gentilis) and great horned owl (Bubo virginianus), all bred on overlapping territories and shared the same prey base in summer. In contrast, they differed in morphology and life histories. I test Lack's (1954) theory that birds annually breed at the optimum time to maximize prey delivery to their young. I explore the genetic, physiological and ecological mechanisms involved in matches and departures from predictions of Lack's theory. I focus on the breeding performance and timing of breeding, seasonal and annual peaks in prey abundance, weather patterns, and the influence of a parasitic blackfly on the young raptors. Previous work on these 3 species (Mclnvaille and Keith 1974, Rohner 1994)  5  examined environmental influences on the timing of breeding but did not relate the annual peak in prey abundance to the consumption needs of young. The red-tailed hawk (Table 1) is an annual migrant that spends its winter in the southern United States. It arrives in mid-late April and typically lays its eggs in the first week of May. Birds migrate south in early-mid September. This bird is diurnal and preys mainly on Arctic ground squirrels (Spermophilus parryii) and red squirrels (Tamiasciurus hudsonicus). During the peak in snowshoe hare (Lepus americanus) numbers, however, juvenile hares were the main prey (this study). Long term breeding densities are stable (review in Preston and Beane 1993). Productivity varies with prey density and with rainfall during the early nestling period (Adamcik et al. 1979). No direct links between timing of breeding and prey densities or weather conditions have been observed. The northern goshawk (Table 1) is similar in size to the red-tailed hawk and this raptor is resident at Kluane in most years but may migrate locally during periods of low prey abundance (see page 50). It is diurnal and takes a wide range of prey, including snowshoe hare and grouse at Kluane (Doyle and Smith 1994). Laying occurs in late April-early May and young leave their nest area in late August. Breeding densities and productivity in this species are linked to prey densities (McGowan 1974, Doyle and Smith 1994) and to weather conditions in spring (Kostrzewa and Kostrzewa 1990, Penteriani 1997). The great horned owl is the largest of the 3 species (Table 1), and is a nocturnal resident. Its main prey is the snowshoe hare (Rohner and Krebs 1996). Laying occurs in mid March and young leave their natal territory in September-October (Rohner 1996). Breeding densities, productivity and timing of breeding in this owl have been correlated with prey densities (Mclnvaille and Keith 1974, Rohner 1994).  METHODS  STUDY AREA  M y research took place adjacent to Kluane National Park (60°57'N, 138°12'W), in a 400-km area of the Shakwak Trench, a broad glacial valley bounded by alpine areas to the 2  northwest and southeast (Figure 2). Along the northwest side of the valley a major highway  6  (Alaska Highway) stretches for 30 km, and the rest of the valley is bisected by about 50 km of abandoned secondary roads and trails. Most of the study area is accessible only on foot. The valley bottom averages 900 m above sea level and is mostly covered by closed spruce forest (50%), interspersed with shrub thickets (33%), grassy meadows (7%), old burns, eskers, a few small lakes, marshes and ponds. Few large tracts of any one vegetative type are present, and all 3 raptor species in this study are found throughout the area. The dominant tree species is white spruce (Picea glauca), with some aspen (Populus tremuloides) and scattered balsam poplar (Populus balsamifera) (Boutin et al. 1995). The dominant shrubs are gray willow (Salix glauca), bog birch (Betula glandulosa) and soapberry (Sherpherdia canadensis). Trees at Kluane are mostly small (canopy height 8-13m). On the borders of the valley the vegetation changes first to open sub-alpine forest of spruce and dense willows and finally to open tundra at about 1400 m above sea level. The large Kluane Lake (ca. 300-km ) borders the study area to the northwest. The 2  climate is cold continental, with a growing season from mid-May through mid-August and with snow cover from October through to May. A rain shadow cast by the mountains to the southeast results in only moderate precipitation (<3mm a day for each month. Accumulated snow depth averages about 55 cm by late winter (Krebs et al. 1986).  L O C A T I N G PAIRS AND NESTS  Work took place at two levels. First, I used a general approach to collecting information on all local raptors. A designated "Intensive Raptor Study Area" was established (Figure 2) within the context of the spatial boundaries of the larger Ecosystem Project. This 100-km area was situated centrally and was heavily used by the team working on the larger 2  project. Team members were trained to identify raptors and to keep daily records of raptor sightings. These observations were reported to me and used to compile maps of the location of breeding/territorial birds and their nests (Boutin et al. 1995). Second, other researchers and I used species-specific approaches to locate birds and their nests from March to late June each year as follows.  7  Red-tailed hawk  From mid April - early June red-tails gave conspicuous aerial displays above nest areas. Observation of these displays enabled us to locate pairs and nests distributed throughout both the intensive area and the larger study area. The distinctive plumage patterns of many birds (Doyle 1996) aided in assigning birds to known nests. Transects were conducted on foot throughout the intensive study area after hatching in late June and July. At this time, birds and their nests were readily located as the birds took flight from the nests and gave alarm calls as the nests were approached. Red-tailed hawks nests were regularly spaced throughout the study area within a year (Doyle 1996) and this distribution assisted in the location of pairs. Some nests were also located by tracking radioed prey taken by hawks.  Northern goshawk  Nests of this secretive resident are difficult to locate. I therefore used six methods to search for birds and their nests (Doyle and Smith 1994). (1) Goshawks exhibit strong nest site fidelity. Prior to laying and incubation, all known nests and nest areas were systematically searched for goshawk sign (plucking posts, whitewash, moulted feathers) or active nests. (2) Prior to the onset of incubation in April falling lid box traps baited with pigeons (Kenward et al. 1983) were placed throughout the study area (1988 - 1992). Trapped adult birds were fitted with tail-mount radio transmitters (Kenward 1978), age was determined by the distinctive plumage patterns of adult (age >3 years) and immature birds. Daily monitoring of these birds and pinpointing centers of activity, helped in the location of nests in the following months. (3) Systematic searches of the entire Intensive Study Area were conducted in April-June each year from 1989-1996 when I looked for goshawk sign. (4) During the searches I used playback alarm calls of the goshawk to elicit a response from territorial birds close to their nests (Kimmel and Yahner 1990, Kennedy and Stahlecker 1993). Transects were placed 200 meters apart and a playback call was broadcast every 200 meters along each transect. (5) Transects were also conducted opportunistically outside of the "intensive area" but within the larger study area. These were typically along trails and  8  forest roads and involved both observation and the use of playback. (6) Radio transmitters attached to prey by other studies monitoring prey populations (Krebs et al. In Press) also revealed the location of a few pairs and their nests.  Great horned owl  Great horned owls were the first predatory birds to breed each year. Breeding birds were found during incubation in March and April by locating calling birds when they called just after dusk or just before dawn. At this time males and females typically perform a hooting duet near the nest (Rohner and Doyle 1992). Triangulation of this location using compass bearings from several vantage points was followed by ground searches to locate roosting birds and their nests. Nests were typically located on "witches' brooms", fungusinduced clumps of dense foliage in White Spruce (Picea glauca). Systematic searching of these clumps for feathers or whitewash (guano) aided in the pinpointing of the nest. In an intensive study of horned owls from 1989 - 1991 (Rohner 1994) some pairs and nests was located using radio transmitters fitted to female owls (Kenward 1987). Age and breeding status of many birds was known from trapping adult birds at nests; fledgling juveniles were also tagged and radio marked before release. Radio transmitters attached to prey (Boutin et al. 1995) also helped to locate nests outside the intensive study area.  MONITORING NESTS, NESTLINGS AND DIET  To minimize nest desertion due to human intrusion (Fyfe and Olendorff 1976), nest visits were conducted when birds were no longer incubating eggs (the period during which birds are most likely to desert). Similarly, disturbance was minimized during the late nestling phase by not climbing nest trees, a period when I knew that premature fledging of young could take place. Nests were monitored until all young fledged. Red-tailed hawk nests were monitored weekly by observation from the ground. Once the chicks were two weeks of age, and visible from the ground, all nests that could be climbed were checked every 3 - 4 days. During these visits pellets and prey remains were collected from the nest and beneath it, and the young were weighed and measured. In 1989 -  9  1990, attempts were made to monitor nestling red-tailed hawks and their diet with the use of platforms (as described below for great horned owls), but these attempts were abandoned as many adults would not feed the tethered young. At nests that were climbed, the length of the 4th primary feather was used to estimate age of chicks (Petersen and Thompson 1977). In this hawk, blackflies (Simulium annulum group) were often visible feeding on and flying around the young particularly around the neck. The number of these flies was estimated (< 5 , 5 - 10, 10 - 20, 20 - 50, 50 - 70, > 70 per nest) during visits to the nests. Goshawk nests and nestlings were monitored every 3 - 4 days from the ground with the use of binoculars, or when necessary, by the use of a X20 telescope and a blind, until the chicks were close to fledging. Near to the expected time of fledging, nests were monitored more frequently and the approximate age of the chicks was estimated by backdating from the fledging day (Ehrlich, et al. 1988). Information on the nestling diet during breeding was obtained from regular collection of pellets and prey remains from beneath the nest. Pellets and remains were also collected from plucking sites on fallen trees, stumps and knolls within 200 meters of the nest. Trees with great horned owl nests were climbed soon after hatching to count and age chicks (Rohner 1994), using the length of the fourth primary feather (Petersen and Thompson 1977). From 1989 - 1996, pellets were collected during weekly visits to the nests and from 1989 - 1994 chicks were tethered to 1.5 X 1.5m wooden slatted, platforms for 4-6 weeks, starting at approximately 30 days of age. These platforms were placed 3 - 4 m off the ground and beneath the canopy. Slatting on the platform allowed prey remains and pellets to fall through into a collection bag (Petersen and Keir 1976, Rohner 1994). In 1995 and 1996, survival of chicks was estimated and pellets and prey were collected, weekly. Pellets were also collected throughout the study from beneath roost perches of adults at nest sites. These were located by direct observation, from whitewash, or by using radio transmitters fitted to roosting birds. As with goshawks, nests and nestlings of this species were also monitored every 3 - 4 days from the ground with the use of binoculars, or when necessary, by the use of a X20 telescope and a blind, until the chicks were close to fledging. Near to the expected time of fledging, nests were monitored more frequently and the approximate age of the chicks was estimated by backdating from the fledging day (Ehrlich, et al. 1988).  10  DETERMINING PREY SPECIES AND A G E IN PELLETS AND REMAINS  Pellets were oven-dried and teased apart using tweezers. Most prey species were identified using a reference collection obtained on site. Some items were identified using reference material from the Vertebrate Museum at the University of British Columbia. Other items were identified using reference guides for fur, feathers and teeth (Banfield 1974, Moore et al. 191'4, Godfrey 1986, Nagorsen 1993). After identification, all diagnostic items were counted and a minimum number of prey individuals per pellet was established (Marti 1987). The average age and biomass of the individual juvenile hares present in pellets and remains were calculated using the hind foot length. Juvenile weight was estimated using the regression equation, weight (g) = -302.2 + 10.2 x R H F (right hind foot measure in mm) (r = 0.92, n = 1051, M . O'Donoghue unpub. data), and juvenile age was estimated by using an estimated growth rate of 16.3 g per day (O'Donoghue and Krebs 1992). The ages of Arctic ground squirrels in prey remains were estimated from skeletal measurements combined with appearance and reproductive status as compared to reference specimens collected in the study area. The ages of Arctic ground squirrels in pellets, however, could not be determined. I therefore assumed that the annual ratio of adult to juveniles in pellets equaled that found in prey remains. Weights of the different prey species were obtained from individuals trapped in the study area at that time (Table 2). Mass estimates of passerine species are from Dunning (1984). The number of prey individuals in prey remains was obtained from a minimum count of diagnostic parts that represented a prey individual (e.g. squirrel tails, bird breastbones, etc.), found at nest sites through the nestling/fledging period. This method was used because observations from blinds showed that cached prey and parts of prey could take many days to consume. Large prey, in particular, would be overestimated if all items were counted as new individuals on each visit to a nest.  A N N U A L CHANGES IN PREY DENSITIES AND TIMING OF T H E ANNUAL PEAKS IN PREY  II  As part of the larger long-term (1986 - 1996) ecosystem project, most common smaller mammal species were studied intensively and indices of abundance were obtained for all terrestrial avian species (More detailed information was obtained for spruce grouse (Dendragrapus canadensis, see Boutin et al. 1995)). In this research, control grids set up throughout the study area (Figure 2) were used to estimate mammalian and avian prey numbers in early spring and late summer. These surveys yielded estimates of winter survival and summer recruitment (see below). In particular, studies of the three main mammalian prey (snowshoe hares, ground squirrels and red squirrels) and one common avian prey (spruce grouse) enabled me to calculate when young of these species became available to raptors.  Methods used to estimate prey densities  Snowshoe hares were trapped on two 32.5 ha control grids (Controls #1 and #2, Figure 2) (86 traps per grid), during two trapping sessions each covering 4 - 6 nights in March and again in October. Numbers were estimated by using Program Capture (Otis et al. 1978, Boulanger 1993). Arctic ground squirrels were trapped on two 7.5 ha control grids (Ground squirrel control 1 and 2) during 2 consecutive days in May and again in late July. Population size was estimated using the model selected by Program C A P T U R E (see Boutin et al. 1995 for more details). Red squirrel population densities were estimated from two 7.5 ha control grids (Controls #1 and #3), that were trapped repeatedly in May and August (Boutin et al. 1993). Densities of voles (Clethrionomys rutilus, Microtus spp.) and deer mice (Peromyscus maniculatus) were obtained from four 2.8 ha control grids (Control #1 and #2, Figure 2), by live trapping for 2 days in May and August (Boutin et al. 1995). Spruce grouse population change was estimated from the number of birds observed per hour in the study area by observers on foot from September to April (Boutin et al. 1995). Information on population trends of willow ptarmigan {Lagopus lagopus) came from Chilkat Pass, 150km to the southwest (Hannon 1983). Territorial males were counted on a 57 ha control plot from 1980 through 1992 (Boutin et al. 1995). Passerine population density was estimated in early June, using 5-minute point counts at 11 regularly placed stations on two 34 ha control  12  grids (Controls #1 and #2, Figure 2), from 1988-1992 and 1995-1996 (Folkard and Smith 1995, Smith 2001).  TIMING OF T H E ANNUAL PEAKS IN PREY ABUNDANCE  Mammals Annual peaks in snowshoe hare abundance occur with the birth of the young, which are eaten by raptors from birth onwards (O'Donoghue 1994). Mean parturition dates for hares (2-3 litters/year) were measured in 1989 - 1992 and 1994 - 1996 (Sovell 1993, O'Donoghue and Boutin 1995, Stefan 1998). B y trapping pregnant females and placing them in large individual cages. The hares were then monitored closely until the young were born, when the mother and leverets were released at the original trap sites within their home ranges (see O'Donoghue 1991 and Stefan 1998). Arctic ground squirrel young become available as food for raptors when they emerge from natal burrows in mid-June (Hubbs 1994, Byrom 1997). The annual timing of emergence was measured in 1991 - 1996. Emergence dates were obtained by intensive observations of the squirrels at burrows and by trapping juveniles to estimate litter sizes (Hubbs 1994, Byrom 1997, Karels unpub. data). Similarly, the young red squirrels become available when they emerge from nests. Boutin (unpub. data), provided emergence dates from all years. As with the other prey, a peak in small mammal abundance occurred when young emerged from nests. Mice and voles have multiple litters through the summer, with the first litter being the most synchronous (Gilbert and Krebs 1991, Mihok 1988). Mean annual dates for first litters were obtained from Gilbert and Krebs (1991) and from Mihok (1988). This information was used to estimate when these prey groups became available to raptors.  Avian prey  Young spruce grouse become available as food to raptors on hatching. Hatching dates were available from 1990 - 1994 (K. Martin pers. comm.). Young passerines also  13  become available to raptors after they fledge (Newton and Marquiss 1982). Work at Kluane 1988 - 89 by Folkard (1990) and by J.N.M. Smith (pers. comm.) provided approximate fledging times of the more common species.  GRAPHICAL ANALYSIS OF PEAKS IN PREY ABUNDANCE, DIET, AND PREY CONSUMPTION  To examine relationships between peak prey abundance and prey consumption, I used graphical displays for each species (Figure 9, 19 and 27). These graphs display horizontal blocks, the depth of which represents the percentage biomass contribution of each prey to the diet. The percentage biomass contribution of each prey includes both the adult and young proportion contributions to the diet. It was not always feasible to estimate when young of a prey appeared in the diet (see page 11). Furthermore, while nestlings/fledglings were being fed, the density of adult prey was relatively constant compared to the sudden availability of juvenile prey. The Julian date at which the peak numbers in prey occurred represents the mean date that the young of that prey become available to the raptors (e.g. birth of hare litter(s), emergence of young ground squirrels, etc,). I used the mean date rather than the latest date, a point at which most prey were assumed to be available, as the best representation of the peak in prey. I did so because geographic variation in timing of litters, (unpubl. data) within the study area, made it impractical to determine the exact date of the peak in availability for all raptor pairs. The mean date also takes into account the inevitable decrease in young prey caused by predation and deaths from other causes. The decline in numbers of juvenile prey as they grow is probably roughly balanced by increased biomass of juvenile prey as they grow (Figure 1).  14  • Time afterbirth/emergence  Figure 1. Qualitative relationship between prey vulnerability and its increase in biomass with time.  WEATHER  Weather data were obtained from a meteorological station 40-km northwest of the study area, at a similar elevation and adjacent to the same lake. Temperature and precipitation were recorded daily and means for individual months and pairs of months (February - March, March - April, etc.) were calculated. I also used the numbers of rainy days in a month and paired months, and the temperature and rainfall, to test if variation in weather influenced prey and raptor reproduction. In the Richardson's ground squirrel (Spermophilus richardsonii), a close relative of the Arctic ground squirrel, there is a strong correlation between spring temperature and the timing of breeding (Michener 1979). I therefore correlated the timing of breeding in ground squirrels and other prey species, with late winter and early spring temperatures (February May), on a monthly and bi-monthly basis. In this analysis, spring is defined as beginning in  15  the month(s) when mean maximum daily temperatures exceeded freezing on >50% of days. Temperatures above freezing cause snow melt, and trigger the swelling and opening of dormant buds in plants (Chabot and Mooney 1985).  BLACKFLY AND ABUNDANCE  A sample of the ornithophilic blackflies seen feeding on nestling hawks was collected and identified by B. Hunter, University of Guelph. Timing of blackfly emergence was not recorded at Kluane, but two studies on the flies in the boreal forest, one in Saskatchewan (Mason and Kusters 1990) and the other in Ontario (Bennett 1960) were used to predict the likely pattern in emergence and peak abundance of these parasites.  DEFINITIONS OF VARIABLES  Densities are the number of pairs in the 100-km "intensive raptor area". A pair 2  refers to a territory or nest at which adult bird(s) were present during repeated visits to the site. Chicks fledged per pair is the mean number of young fledged per monitored pair. Fledging date is the time that the last young fledged from the nest. Successful pairs are those that fledged young. Proportion of pairsfledgingyoung is the proportion of pairs located and consistently monitored through the breeding season, that fledged young. The number of young fledged per pair/per quartile is calculated by dividing the annual range of breeding dates into 4 equal parts, starting with the earliest breeding date that year, and calculating the number of young fledged per pair in each quarter of time (e.g. Lower quartile = earliest period). Peak in prev availability is the mean date of the numerical peak in prey abundance. Peak in prev consumption is the mean date that fledglings in a nest or the mean annual date of fledglings from all nests peaked in their food (biomass) consumption requirements. For prey species, density always refers to the population abundance on control grids or another systematic index of abundance in spring. In some cases, density was also measured at other times of the year. These times are mentioned under the appropriate results.  16  STATISTICAL ANALYSIS  I emphasize graphical display of the relationships between variables, as usually only one value per variable is available for 5-8 years and therefore statistical power is low. Where sample sizes allow, standard statistical procedures are used. In most cases, P values are presented for their descriptive value rather than as tests of statistical hypothesis (Arcese et al. 1992) and stress is placed on the magnitude of "r" values as supportive evidence to the strength or weakness of biologically observed correlations. A l l statistical results integral to this analysis were therefore not only supported by graphical representation, but also by supportive information (e.g., Statistical relevance of prey species abundance to the reproductive performance of a raptor is also supported by other data showing the relevance of that prey in the diet.). I used 'Statistica' (StatSoft Inc. 1995) for all statistical analysis. Proportional and percentage data were arcsine square root transformed before analysis (Zar 1984). When regression was used, linear regression was employed for most analyses (keeping in mind that the small sample size and skew may have influenced the strength of any correlation, and therefore "r" is calculated only as a support to any biologically observed links). The high degree of skew in most data and the small samples made multiple regression impractical or inaccurate. Independent and dependent variables were chosen based on biological expectations. Diet width was measured using Levins' measure (1968), B = 1 ^ ^ , where p is the y  proportion of prey taxon j in the diet.  This gave an index of values from 1 to n, where 1 is  the narrowest diet width possible. In this way, I could explore the importance of individual prey species to individual raptor species over the study, by correlating diet width and prey population densities. To test the statistical significance of the annual variation in fledging dates I used oneway A N O V A s . Kruskal -Wallis tests were used to examine differences in the number of chicks fledging from pairs that bred early versus late. Alpha was set at 0.05 unless otherwise stated.  17  RESULTS.  P R E Y AND W E A T H E R  P R E Y AVAILABILITY  Mammals  Young snowshoe hares are available as food for raptors from the day of birth. Parturition dates were monitored closely from 1989 to 1992 and from 1994 to 1996 (Table 3) (Sovell 1993, O'Donoghue and Boutin 1995, Stefan 1998). Hares produced an average of three litters per female each summer (Table 2). Birth of the first litter took place in the last third of May in all but one-year (Figure 3a). Adult females gave birth synchronously within a year (S.D. = 2 - 6 days). This first litter was followed by a second litter 35 days later in late June and a third litter 35 days later in late July - early August (Table 3). Arctic ground squirrel young became available when they emerged from burrows. Ground squirrels have only one litter per summer (Table 2) which emerged synchronously (S.D. 3 - 7 days) in the third week of June in 5 out of 6 years (Appendix 1, Figure 3b). Red squirrels usually have one litter (Table 2, S. Boutin, pers. comm.), with a wide range of individual litter dates within a year (> 40 days), and great variation in the mean litter date between years (26 May - 30 July, Appendix 2, Figure 3c). No detailed information on the timing of breeding in mice and voles was available during this study, however, two studies in the boreal forest show that the first litter appears in mid-June, with young born every 20 days until September in some years (Gilbert and Krebs 1991, Mihok 1988).  Birds  Spruce grouse have one brood (Table 2) and their precocial young become available to raptors after hatch in mid-June (Appendix 3). Between years, mean dates ranged from the 12 - 22 June (Figure 3d). Our sample sizes are small within years ( 1 - 5 broods), but hatching appears to be synchronous among broods ( 0 - 3 days, K.Martin unpub.data). Typical fledging dates for passerines frequently found in the raptors' are: gray jays  18  (Perisoreus canadensis) in late April - early May; American robins (Turdus migratorius), boreal chickadees (Parus hudsonicus) and dark-eyed juncos (Junco hyemalis) 10th June, chipping sparrow (Spizella passerina) and yellow-rumped warbler (Dendroica coronata) 20th June, thrushes other warblers, and flycatchers early July (Folkard 1994 and J.N.M. Smith pers. comm.). No information is available on the inter-annual variation in fledging dates.  VARIATION IN ANNUAL PREY DENSITY  Mammals  Snowshoe hare population densities in the study area followed a cyclic pattern, with a slow increase followed by a rapid decline (Figure 4a, Krebs et al. 1995). Hare densities on two 32.5-hectare plots (Appendix 4) reached a peak in the fall of 1989 with 2.35 (range N/A) hares per hectare. This was followed by a peak in spring densities of 1.51 (range 1.43 1.60) hares per hectare in 1990. Densities then declined to a springtime low of 0.07 (0.02 0.11) hares per hectare in 1994. Arctic ground squirrel densities on two 7.5 ha plots were monitored from 1990 -1996 (Appendix 5) and showed a similar pattern of increase and decline to hares (Figure 4a, Boutin et al. 1995), but with a 1 year lag and lower amplitude. Spring and fall populations peaked in 1991, with 2.16 (range 1.72 - 2.59) and 2.92 (1.52 - 4.32) squirrels per hectare. Spring densities fell to 0.91 (0.70 - 1.13) and 0.84 (0.80 - 0.86) individuals ha  1  in 1993 and  1994 (Boutin et al. 1995, C.J. Krebs unpubl.data). Red squirrel densities (Appendix 6) varied erratically but varied less than those of hares or ground squirrels. Between 1988 and 1996, densities estimated on two 7.5 hectare plots, ranged from 22 - 41 individuals per 10 hectares in spring and 19 - 45 individuals in the fall (Figure 4b). Peak densities occurred from fall 1993 (42 per 10 ha.) through spring 1994 (41). Lowest numbers were in 1988 (22 per 10 ha in spring, 22 in fall) and in 1990 (23, spring, 19 fall) (Boutin et al. 1995, S. Boutin unpubl.data). Vole populations were less predictable than those of hares and ground squirrels (Figure 4b, Boutin et al. 1995, R. Boonstra unpubl.data).  On the two 2.8 control plots, high '  19  Microtus spp. densities (Appendix 7) occurred in 1992 - 1994 and high red-backed vole (Clethrionomys rutilus) densities from 1991 to 1994. Peak Microtus densities were in 1988 and 1993, with 6.0 individuals per ha" in the spring of both years. The low of 0.04 per ha 1  -1  was recorded in spring 1996. Red-backed voles peaked in 1992 with 5.0 individuals per hectare in spring and declined to zero in 1995 and 1996.  Birds An index of spruce grouse sightings (Appendix 8) revealed a similar trend to hares and ground squirrels (Figure 4a). Peak spring densities occurred in 1989, a year before the peak in spring hare densities. In 1989, 37 birds were seen per 100 field hours in summer (May - August) and 11 birds per 100 field hours in the winter (Sept - April). Numbers declined in 1991 and 1992 to 5 and 6 birds respectively per 100 field hours in summer and 2 and 3 birds per 100 hours respectively in winter. Passerine population densities were not estimated every year (Appendix 9), but a marked increase in the population density on the 36-ha control plots took place between the two periods of monitoring (1988 - 1992 and 1994 - 1996) (Figure 3b). Peak detections occurred in 1996 with 113 birds per 36 hectares and low numbers in 1989, with 42 detections per 36 hectares (Folkard and Smith 1995, J.N.M. Smith unpubl.data.)  Prey weights  Prey weights (Appendix 10) were obtained from animals trapped in the study area, during the period that prey were taken as prey by the raptors. Adult hares were the heaviest commonly taken prey at 1406.6 grams (+S.D. 149.3, n = 50), while the lightest voles/mice weighed 30 grams.  WEATHER  The mean maximum temperature and mean precipitation per month from 1989 1996, (Appendices 11, 14), were all within the standard deviation of the long term means for  20  1973 - 1996 (Appendix 12 a-b). Although weather varied from year to year no one year deviated markedly from the mean. Variation in the timing of spring each year was measured by mean maximum monthly temperatures and by the onset of consistently above freezing temperature each year. In February and March mean temperatures were consistently below freezing while those in April and M a y were consistently above freezing (Appendix 11).  In March, an average of 15  days (+S.D. 4.32) reached above freezing (Appendix 13). B y April, 28 days (+S.D. 3.16) exceeded freezing. The shift from winter to spring therefore takes place through March and April.  In contrast, 1992 was an exceptional year, with cold maximum temperatures in late  spring (Appendix 13).  T w o other years, 1989 and 1995, were also cold in early spring. In  March 1989, only 7 days were above freezing and in March 1995, only eight days were above freezing (8 days less than the mean) (Appendix 13). Mean daily precipitation (Appendix 14) from February to M a y averaged less than 1 mm, then increased in June and July. The wettest June was in 1990, with 2.4 mm of rain or snow per day and the wettest July was in 1992 with 3.6 mm per day. In contrast 1995 and 1996 each had only 0.3 mm precipitation per day in June. The driest July was in 1989, when precipitation was 0.3 mm per day. When months were paired (Appendix 14), May/June and June/July in 1991 were the wettest periods recorded, and the same periods in 1994 were the driest.  July 1993 saw the largest number of rain days, with rain on 20 days, and June 1995  the driest with rain on only one day. M a y of 1992, was also particularly wet with 13 rain days.  Correlation between precipitation and temperature  Long term (1973 - 1996) mean maximum monthly temperature and precipitation, (Appendix 12 a-b) were negatively correlated (r = -0.87, P<0.03), through the raptor breeding period (February-July).  PREY DENSITY, TIMING OF REPRODUCTION AND WEATHER  21  Mammals  Densities of prey were significantly correlated to the timing of breeding in that prey only in the red squirrels (r = 0.71, P<0.033, n = 9). No correlation or trend was seen between densities of ground squirrels and snowshoe hares and the timing of breeding in those species. Ground squirrel young emerged later the cooler and wetter the weather was in March and April (r = -0.89 P<0.019, n = 6), and the low the mean temperature was in March and s  April (Figure 5). In fall, ground squirrel numbers were negatively correlated to the number of rain days from the previous February - April. In red squirrels the only significant correlation was between the fall population density in a year, and precipitation the preceding February (Appendix 16). Juvenile survival was lower, litters were smaller or fewer females bred in years of high precipitation. The mean timing of breeding varied by more than 2 months, but was unconnected to spring weather. No significant correlations were seen between weather and Microtus and Clethrionomys population trends.  Birds In birds detailed information on the annual variation in the timing of breeding is only know for spruce grouse, and the timing of breeding in this species was weekly correlated to it's density (r = 0.59, P<0.21, n = 6). In passerines, the annual density indices were negatively correlated to both the number of rainy days (r = -0.89, P<0.019, n = 6) and to the total rainfall in June (r = 0.81 s  s  P<0.05, n = 6). Spruce grouse densities in winter were negatively correlated to both rainfall and the number of days precipitation the previous July (Appendix 17).  Correlations between population density and timing of breeding in prey  I looked for correlations between prey species to identify species with shared patterns in the timing of breeding, and breeding performance. Only one significant correlation was seen, with spring population density of snowshoe hares positively correlated with spring population density of ground squirrels (Figure 4a). To identify if any causal links are  22  present between these two prey and the later reproductive performance of a raptor therefore has to be supported by evidence showing the presence and relevance of that prey in the diet.  B L A C K F L Y ABUNDANCE  Timing of blackfly emergence was not recorded at Kluane, but the two studies on blackflies (Simulium spp.) in the boreal forest showed the same pattern, with flies predictably appearing in late May and early June and peaking in early July.  TIMING OF BREEDING BY T H E THREE RAPTORS  (A)  RED-TAILED HAWK  Annual variation in timing of breeding and reproductive performance  The mean fledging date of red-tailed hawks varied little in over seven years, with fledging always occurring in the 3 week of July (Table 4). The range of individual fledging rd  dates varied annually from 3 - 2 5 days (±8.3 S.D., n = 7). Mean fledging date was earliest in 1994 and latest in 1992 and 1995 (Figure 6a). The density of pairs varied from 9 - 1 2 per 100 km (Table 5) over the study. A l l females attempted to breed each year and the proportion that fledged young ranged from 0.25 in 1992 to 1.0 in 1994. The number of chicks fledged per pair (Table 5) ranged from 0.6 (±0.9 S.D. n = 13) in 1992 and 1995 to 1.8 (±0.6 S.D. n = 12) in 1994. Of successful pairs, the number of young fledged per pair ranged from 1.4 (±0.5 S.D. n = 16) in 1995 to 2.2 (±0.6 S.D. n = 13) in 1996. In neither case were these differences in breeding performance statistically significant (1-way A N O V A ) .  Links between timing of breeding and reproductive performance  The proportion of pairs that fledged young was higher in years with an earlier annual mean fledging date (Figure 7). The number of young fledged per pair also increased the  23  earlier the hawks bred (pooling data over all years, r = -0.91, P<0.005, n = 76, Figure 8). There was also a positive correlation between the number of chicks fledged per pair and the proportion of pairs that fledged young (r = 0.88, P<0.008, n = 7). Within a year, fledging success of pairs also varied with the timing of breeding (Kruskal-Wallis test X = 9.92, d.f. = 2, P<0.007, n = 21, Table 6). Over the seven study 2  years (Table 6), the mean number of chicks fledged by early breeders (1.70, +0.86 S.D., n = 21, lower quartile) was higher than for late breeders (1.24, +0.77 S.D., upper quartile).  Prey availability and timing of the peak in prey consumption by the young  Fitch et al. (1946) showed that the peak prey consumption by red-tailed hawks occurs at the point of inflection in the growth curve (the period just prior to the slow down in the chick growth rate) of the nestlings (Figure 9). This point of inflection at Kluane was calculated using the length of the 4 primary feather and the fledging dates of young. The th  peak prey consumption at Kluane varied annually between 15 - 25 of June (Julian days 165 175). Five main prey types were identified in pellets at nests (Table 7, Figure 10). Redtailed hawks specialized on medium sized mammals, with arctic ground squirrel contributing the largest proportion of the biomass in 5 of 7 years (Appendix 18). Red squirrel contributed the second largest total biomass overall and was the dominant prey in 1994. Snowshoe hares made up the third largest total contribution to the diet and were taken more than ground squirrels in 1990 when hare populations were at their peak (Figure 4a). In all years, birds, voles and mice contributed very little to the biomass consumed by each nest (<10.1 % collectively and <5 % each annually, Table 7). Juvenile hares were consumed more frequently than adults overall (Table 7) and they contributed a greater proportion of the hare biomass in 4 of the 7 years, particularly when hares were abundant. They made up 42 % of the biomass consumed at the peak in spring hare densities, in 1990. At the low in hare numbers in 1994 and 1995, juvenile hares made up only 1.7% of the biomass. The first peak in juvenile hare numbers occurred with the birth of the first hare litter, which was always 10-30 days prior to the peak in consumption by the young (Figure 9). Birth of the second hare litter ranged from slightly before the peak prey  24  consumption in 1995, at the peak 1994, and after the peak in the remaining 4 years (Figure 9). Juvenile hares taken by red-tails averaged 326 grams (n = 34), when hares were estimated to be 20 days old (range 3 to 32 days). These hares were typically from the first litter, and at 20 days of age they were therefore usually taken < 1 week (5 out 6 years that diets were monitored) prior to the peak in prey consumption by the young. I could not determine the age of red squirrels in the diet from pellet analysis, but juveniles were seen in prey remains at nests. However, the peak in red squirrel emergence occurred after the peak in prey consumption in five of the six years (Figure 9). In the year (1994) that it occurred before the peak in prey consumption, red squirrels were the dominant prey (51.5% of the biomass consumed, Table 7). Ground squirrel juveniles were taken more frequently than adults in 5 of the 7 years (Table 7), when they also contributed a greater proportion of the ground squirrel biomass in the diet. The percentage contribution of juvenile ground squirrels to the biomass consumed ranged from 13.0% in 1990 to 44.0% in 1995. Emergence of juvenile ground squirrels occurred at the peak in prey consumption in 5 of the 6 years (Figure 9). In the one-year that it was after the peak (1992), ground squirrels still made up a large percentage of the prey biomass (Figure 10), and breeding success was very low (see below). On occasions juvenile small mammals and birds were seen in prey remains at nests, but as I saw these prey groups did not contribute a large percentage of the biomass to the diet (Figure 10). It is unlikely that their presence influenced the timing of breeding in this hawk.  Diet and reproductive performance  The mean timing of breeding was unrelated to annual variation in abundance in any of the main prey species, and no correlations were found between any breeding parameter and the annual abundance of prey. In contrast, the annual timing of fledging was correlated to the timing of the annual peak in both ground squirrel (Figure 11) and red squirrel (r = 0.82, P<0.023, n = 7) populations. No annual relationships were seen between timing of fledging and peaks in other prey. More red-tailed hawks fledged young (Figure 12), and more young fledged per pair (r = 0.74, P<0.07, n = 5) when juvenile ground squirrels emerged early in a year. The number  25  of red-tailed hawk pairs fledging young (r = 0.71, P<0.05, n = 8), and the number of young fledged per pair (Figure 13) were also correlated to the timing of red squirrel emergence. The proportion of each prey (excluding ground squirrels) in the diet of red-tailed hawks was positively correlated (P<0.05, n = 6) with the annual abundance of that prey in the environment that year. Arctic ground squirrels, however, contributed a high percentage of the biomass in the hawks' diet, even after the ground squirrel population declined sharply in 1992 and 1993 (Figure 14).  Weather and the annual changes in reproductive performance  Breeding was late in years with low March and April temperatures (Figure 15). Increasing temperatures in June were negatively correlated with the number of chicks fledging per pair (Figure 16). There is no correlation between March and April temperatures and temperatures seen later in June. No correlation was seen between rainfall or rain days and the number of young fledged per pair in June.  Weather, blackflies and reproductive performance  From 1992 to 1995 the number of blackflies was recorded every 2 - 3 days, at 5-7 nests annually. In these four years, whenever >70 flies were recorded on or around the young during a visit, all chicks died by the next nest visit (n = 12, Table 8). A l l 5 nest failures noted in those years were associated with severe blackfly attacks. In contrast none of 13 nest failed in 1993 and 1994 when there were fewer blackflies. Over all four years, the annual proportion of nest failures was significantly correlated with the mean number of flies seen per nest visit (P<0.05, n = 4). Nest failures following high blackfly infestation occurred when the chicks were <20 days old. These young called incessantly and did not exhibit weight loss prior to death (on five occasions dead chicks were found lying intact in or under the nest with full gullets, in the presence of one or both adults. Over the whole study area (Table 9), nest failure rates followed a similar pattern to those associated with blackflies, with a significant difference in the proportion of nests  26  failing from unknown cause (not predation or starvation) in 1992/1995 (n = 28), compared to 1993/1994 (n = 27), ( X = 199.06, d.f. = 1, P<0.000). It was also observed that during the 2  period that blackflies were monitored (1992 to 1995), that it was the late breeders which had the higher failure rate (r = 1.00, P<0.00, n = 4). This correlation was also significant over s  the entire study period (Figure 17) when the nests that failed from starvation or predation were excluded from the analysis.  Summary: Red-tailed hawk  The density of pairs throughout the study period varied little and all females attempted to breed every year. Pairs breeding early in a season fledged more young than pairs breeding later, and when the mean breeding date for the population was late, fewer young were fledged per pair. The main prey, the Arctic ground squirrel continued to make up a large percentage of the prey biomass even when its densities declined. Timing of the peak in prey requirements by the young matched the timing of the peak in availability of ground squirrels (particularly juveniles). Red squirrels contributed the second largest percentage biomass to the diet and timing of breeding by red-tails was correlated to this biomass, but less clearly than the correlation to Arctic ground squirrel. In years when birds bred late, and late within a year, young red-tailed hawks were subject to mortality associated with attacks by blackflies.  (B)  NORTHERN GOSHAWK  Annual variation in timing of breeding and reproductive performance  There was only a 6 day spread in mean goshawk fledging dates (Table 10, Figure 6b) over the six years of data. Fledging dates ranged over 1-25 days within a year (+S.D. 0.58 13.7). The density of pairs (Table 11) ranged from 1-5 pairs per 100 km over the 8 years of 2  the study. Of the 31 breeding attempts located throughout the study, only one failed to reach the nestling stage, and typically most breeding attempts produced young (Table 11). Five of the six failures at the nestling stage are due to predation. The number of chicks fledged per  27  pair per year, ranged from 0.0 from the one nest in 1992, to 3.8 chicks per pair from four nests in 1996. Of the pairs that fledged young, the number of young fledged per pair ranged from 2.0 (±1.4 S.D.) in 1989 to 3.8 (± 0.5 S.D.) in 1990 and 1996. Sample sizes were too small to test for differences in success among years  Timing of breeding and reproductive performance  The timing of breeding was unrelated to other reproductive variables. The range of fledging dates was large within years (Figure 6b), but the number of young fledged from nests showed little variation within a year (Table 12), and no seasonal trend in the number of young fledged within a year was apparent (Figure 18).  Prey availability and timing of the peak in prey consumption by the young  Information on the timing of peak food consumption by young was obtained from Schnell (1958), who found a peak in prey around the point of inflection in the growth curve of the nestlings (Figure 19). Assuming that Kluane chicks had similar growth and consumption rates as those studied by Schnell, peak prey consumption varied annually between the 10 - 25 of June (Julian days 161 - 176), when the chicks were approximately 30 th  days old. Annual variation in the diet was determined by identifying prey remains and pellets found in and around nests (Table 13). Five main prey types were identified, with three, hares, grouse and ground squirrels, each contributing >10% of the biomass in most years. Snowshoe hare contributed the largest proportion of the biomass in five of eight years and grouse contributed the largest proportion in the remaining three years. I was able to separate adult snowshoe hares from juveniles in the goshawk diet. In 6 of 8 years, adult hares contributed more biomass to the diet than did juveniles, but juveniles were taken every year and in 1993 and 1994 (the low in the hare cycle) they made up a larger proportion of the diet than did adults (Table 13). Juvenile hares were available throughout the nestling period, but were typically taken when they averaged 698 grams (S.D. ±56.1, n = 25) when they were around 43 days (range 3 to 69) of age. At this age, juvenile hares were  28  therefore most frequently taken 1-2 weeks after the estimated peak in prey consumption requirements of the young goshawks (Figure 19). Juvenile hares from the second litter were first available (1-2 weeks) after the peak in prey consumption in 4 of 6 years, and prior to fledging in 5 years, but as we saw juvenile hares were typically not taken as prey by goshawks until they were 4-6 weeks of age. Grouse replaced snowshoe hares as the main prey when the hare population was at its cyclic low (1993 - 1995, Figure 20). Grouse abundance peaked 10-15 days prior to the peak in prey consumption by the young (Figure 19), but their small size (<50g) throughout this time, suggests that they would not have a significant impact on the overall goshawk diet at this time. Ground squirrels were taken consistently in all years of the study, but typically contributed <25% of the biomass consumed (Table 13). Peaks in ground squirrel abundance typically occurred at or just after the peak in prey consumption requirements of the young (Figure 19). Red squirrels were also taken frequently (Table 13 and 19) but typically contributed <10% of the prey biomass. The annual peak in red squirrel abundance occurred both prior to and after the peak consumption requirements (Figure 19).  Diet and reproductive performance  Goshawks bred significantly earlier in years when hare (Figure 21) and spruce grouse densities were high (r = -0.94, P<0.019, n = 5). When hare densities were high (1989 1991) a wide range of breeding dates was seen (17 - 25 days). When prey populations (hare and grouse) started to increase from 1994 onwards, breeding by goshawks was more synchronous with only a 1 - 7 day annual range (Table 10, Figure 6b). More goshawk pairs were located in years with high hare numbers in spring (Figure 22), and the proportion of pairs successfully fledging young increased with hare and grouse densities. However, I found no correlation between the number of chicks fledged per successful nest and prey abundance. No correlation was found between the annual abundance of any other prey species and the reproductive variables in goshawks. Snowshoe hares contributed the largest proportion of biomass to the goshawk diet overall. Within a year, the first hare litter was available as food for the young hawks  29  throughout their development and the second litter was born soon after the peak in needs of the young (Figure 19). The proportion of pairs that fledged young each year was higher when the first hare litter was born early (Figure 23). The density of hares was positively correlated with the proportional biomass of hares in the diet (Figure 24). This was also true for another frequently taken prey, spruce grouse, but in this species the correlation was weak (r = 0.58, P<0.17,n = 7).  Weather, blackflies and reproductive performance  The only correlation observed between weather and reproduction was between high rainfall in May and an increase in the number of chicks fledged per pair (r = 0.91 P<0.012 n = 6). No blackflies were seen at goshawk nests during prolonged observations from blinds. Although flies may have been missed while observing nests from a distance, there was no evidence of blackfly deaths as seen in red-tailed hawks.  Summary: Northern goshawk  The density of pairs varied during the study and was positively correlated with hare density in spring. In most years, hares, grouse and ground squirrels each contributed >10% of the prey biomass. Hares were the main prey during the period of high hare densities and grouse were the main prey at low hare densities. Timing of breeding in goshawks was correlated to the density of hares and earlier breeding in hares was correlated to an increase in the proportion of breeding pairs that successfully fledged young. However, no correlation was seen between the timing of breeding and the number of pairs fledging young or the number of young fledged per pair or per successful pair. The peak prey requirements of young were met every year by a broad peak in abundance of the three main prey species. Juvenile hares were available throughout the nestling phase, but were typically taken 1 -2 weeks after the estimated peak in the young goshawks prey requirements. Increased rainfall in May was correlated to an increase in the number of fledged young. No links were seen between reproductive performance and blackflies.  30  (C)  GREAT HORNED OWL  Annual variation in timing of breeding and reproductive performance  There was a 30-day range in mean fledging dates of great horned owls (Table 14, Figure 6c). However, in five of the six years that owls bred, mean fledging date varied by <11 days. In the remaining year (1994) the sole pair that bred, fledged younglO days later (June 29 ), than any other nesting attempts monitored (n = 59). Annually, individual th  fledging dates ranged from 4 - 40 days (+S.D. 3.73 - 9.94 days), with young in 1989-1991 fledging on average a week earlier than those breeding in 1994 -1996. Densities of pairs varied 2.5 fold through the study (10 - 24 pairs 100km ) (Table 15). 2  The number of pairs that did not breed or failed early in the laying/young nestling period, ranged from 100% in 1992 to 14% in 1989 and 1990. The number of young fledged per pahranged from 0 in 1992 and 1993 to 2.1 (±2.1 S.D.) and 2.2 (±1.9 S.D.) in 1989 and 1990. Of the pairs that fledged young, the number of young fledged per pair ranged from 2.6 (±0.8 S.D.) in 1990 to 1.8 (±0.8 S.D.) in 1991.  Timing of breeding and reproductive performance  In years with early breeding, more pairs fledged young (Figure 25). However, no significant variation was seen in the number of young fledged per quartile in a year (KruskalWallis test, Table 16, Figure 26).  Prey and timing in peak food demands of the young  No direct information was available on peak prey consumption by nestling great horned owls. Other raptors, however, show peaks in prey consumption around the point of inflection in chick weight gain (Fitch et al. 1946, Schnell 1958, Cave 1968, Newton 1986, and Watson 1997), and I assumed that the same applies horned owl young. Weight gain of  31  nestling great horned owl chicks came from studies by Reid (1925) and Springer (1979), and work by Newton (1986) showed that chick weight gain was correlated to prey delivery rates. Therefore in great horned owl chicks, peak prey consumption occurs at about 20 days of age, which at Kluane occurs approximately around 30th of April (Julian day 120) (Figure 27). The main prey of great horned owls in five of the seven years was snowshoe hare, and in the remaining years mice and voles were the main prey (Figure 28). Other prey combined (birds, ground squirrel and red squirrels) contributed <15% of the prey biomass in 5 of 7 years. A l l prey peaked in abundance after the peak in prey consumption requirements of the young (Figure 27), except inl994 when one owl pair bred very late. The first prey peak occurred when the first hare litter was born, 20 days after the peak in prey requirements (Figure 27). Great horned owls took hares of all ages including newborns (snowshoe hares give birth above ground). In particular, they focused on young <20 days of age, taking them in preference to older young (Rohner and Krebs 1996). In years when hares were abundant (1989-1991), juvenile hares contributed a substantial proportion (>47 %) of the prey biomass. At this age (<20 days), young from the first hare litter were still available as prey to the owls prior to fledging every year (n = 5). Young from the second hare litters were only available prior to fledging in 1994, and typically only became available late in the fledgling period (Figure 27).  Diet and reproductive performance  Annual variation in diet was estimated from pellets and prey remains (Table 17). Five main prey were identified. Snowshoe hares contributed the largest proportion of biomass in 5 of 7 years (1989-1995) (Figure 28). Over all years, hare density in spring was correlated to biomass of hares fed to the owl chicks, with significantly more juvenile hares being taken the higher the density of hares (Figure 29). In the two years that hares did not contribute the largest proportion of the diet, mice and voles contributed up to 44% of the biomass eaten. Annual ranking of the prey by biomass over the study (Appendix 20) shows that hares followed by ground squirrels and mice/voles were the three main prey species. Juvenile hares were taken 4.5 times more frequently than were adults (Table 17). Adult  32  hares, however, still contributed most of the hare biomass in the diet in 5 of 7 years, because owls took many young hares soon after birth, when they weighed < lOOg. Owl density was positively correlated with hare densities the previous year (Rohner 1994).. Laying in the horned owl took place earlier in years with high spring hare densities (Figure 30), and when hare densities were low and consequently contributed less to the diet, birds either bred late (1994) or did not breed at all (Figure 31). In years of high hare abundance (1989 -1991) a wide range of breeding dates was seen (17 - 39 days). At low prey densities in (1995-1996) pairs bred more synchronously, with a 4 - 6 day range (Figure 6c). More young were fledged per pair (Figure 32) and more active territories fledged young (r = 0.85, P<0.02, n = 8) when snowshoe hare densities were high. Increasing relative biomass of hares, particularly juveniles in the diet of the owls, was correlated with the number of young fledged per territory (Figure 33). High hare densities in the previous fall (as an indication of hare densities that winter) may also have contributed to breeding success of those pairs that bred, with hare densities correlated to the number of young fledged per breeding pair (r = 0.80, P<0.018, n = 8). The date of breeding, the number of chicks fledged per pair, and the proportion of nests that fledged young (r = 0.88, P<0.008, n = 8) were also correlated to the population density of ground squirrels. This result was expected as hare and ground squirrel densities were correlated.  Weather, blackflies and reproductive performance  The reproductive performance of the great horned owl was not correlated to the annual changes in weather. Blackflies, however, were seen at nests in 1989 - 1991 when 15% (n = 86) of fledged chicks died from blood infections carried by blackflies (Rohner and Hunter 1996).  Summary: Great horned owl  33  Density of pairs varied 2.5 fold and was correlated to the density of hares, but with a 1-year time lag. The mean timing of breeding and the proportion of pairs fledging young were both positively correlated with hare densities. Timing of breeding within a year was not correlated with the number of young fledged per pair. Prey populations peaked well after the peak in prey consumption requirements of the young, but the main prey, juvenile hares, were available in the late nestling and fledgling periods, a period of continued growth in the owl.  DISCUSSION.  Adaptive adjustments in life history patterns are all made within a framework of constraints imposed by the environment and the genetic and phenotypic make up of a species (Stearns 1976, Roff 1986). I identified several constraints that influence the timing of breeding in these raptors: weather, prey availability, cyclic prey, body size, length of breeding season, winter survival and attacks on the young by blackflies. Within the influence of these constraints, I now test if Lack's (1954) hypothesis, 'that breeding is timed so that the young's greatest demand for food coincides with the greatest abundance of available prey'. First, I explore the annual influence of weather on the timing of breeding and the breeding success of these raptors. For each raptor I then look at the timing and success of breeding in relation to prey density and the timing of the annual peaks in prey. I conclude this look at individual species, by examining how prey and weather combine with the biological constraints affecting the different raptors to determine the timing of breeding. I next explore mechanisms that could allow birds to breed so that the brood's prey demand is synchronized to the peak in prey availability. I then consider the influence of a bird's age on the timing of breeding, and explore other assumptions made in the overall analysis. I conclude with a general discussion on the applicability of Lack's hypothesis to these raptors, and compare my results with those on other similar birds.  34  E F F E C T S OF WEATHER ON BREEDING  Weather as a cue in the timing of breeding  In this study, weather may have influenced the timing of breeding in the red-tailed hawk, through its influence on the timing of emergence of its main prey, the Arctic ground squirrel. The higher the mean maximum temperature in March and April, the earlier both ground squirrels (Figure 5) and red-tailed hawks bred (Figures 11 and 15). While the effects of weather on migration could speed up or slow down the arrival of red-tailed hawks, most of the birds annually arrived at Kluane in late April. The timing of breeding by resident great horned owls and goshawks at Kluane were not related to either weather or variation in breeding by their main prey, the snowshoe hare. A 10 year study (1984 - 1993) of goshawks in the Mediterranean (Penteriani 1997), however, found that cold, wet weather was associated with later breeding. A possible direct effect of weather on the timing of breeding is that cold temperatures could prevent snowmelt on nests (reviewed in Newton 1977). However, snow on nest platforms certainly did not prevent owls from laying in winter conditions every year at Kluane, nor was it likely to have influenced timing of breeding in the other two species, both of which could readily build new nests if older nests were not available. There was no hint that nest sites were limiting, and individual pairs of all three species used many nests over the study. Only on two occasions did different species use the same nest in more than one year.  Effects of weather on the reproductive performance of raptors and their prey  Poor weather affects breeding success (production of young) in many bird species including raptors (Newton 1998). In this study, only two patterns were observed between weather and breeding success. In goshawks, more young fledged per successful pair in years when May was wet. A similar pattern has been observed in Australian raptors, where increased prey production in wet conditions increases breeding success in several species, including the brown goshawk (Accipiter fasciatus didimus, review in Olsen 1995). However  35  in the Kluane study, no positive correlation between increased rainfall and prey populations was observed. The other observed pattern was in red-tailed hawks, where fewer young fledged from successful red-tailed hawk nests in years of high mean maximum daily temperatures in June (Figure 16). High temperatures in summer may cause reduced breeding success in golden eagles due to heat stress (Beecham and Kochert 1975). Like golden eagles, red-tails have exposed nests and chicks may be vulnerable to heat stress. However, at Kluane overall productivity, of pairs in most years was likely influenced by factors (blackflies and prey abundance discussed later) other than weather, with no correlation between the number of pairs fledging young or fledging success of all pairs and with temperature.  Goshawk and  great horned owls breed earlier than red-tail's, and showed no increase in mortality during hot years. In other studies, low temperature reduced breeding success in both goshawks and kestrels (Falco tinnunculus, Kostrzewa and Kostrzewa 1990) and golden eagle (Aquila chrysaetos, Beecham and Kochert 1975). This was not observed at Kluane. It is possible that experienced individuals may have chosen sites more protected from the affects of weather, and thereby attained a higher productivity than inexperienced birds, as seen in a 12 year study of peregrine falcons (Falco peregrinus) (Olsen 1995). However, the strong correlation seen between breeding success in both goshawks and great horned owls and with the availability of their cyclic prey, suggests that weather did not play a significant role in the productivity of pairs. In red-tails, individual experience with weather may have influenced the nest site selection and subsequent success of some pairs, but the age of most birds in this study was unknown. If weather did have an impact on the success of the more experienced pairs of red-tails, its influence on the overall productivity of the population is probably small, as all nest failures at intensively monitored nests (Table 8) were associated with the abundance of blackflies (see P. 38). Blackfly numbers were not correlated to any weather patterns seen at the time of the nest failures.  Summary: Weather  In conclusion, the timing of breeding by red-tailed hawks may have been affected by the timing of emergence of Arctic ground squirrels, which in turn may have been affected by  36  temperature in March and April. Wet weather in May was associated with the number of young fledging from successful goshawk nests, possibly by influencing prey abundance. In red-tailed hawks, hot weather in June may influence the number of young fledging from successful nests through heat stress, but it had no significant impact on the number of pairs fledging young or the number of young fledged per pair.  TIMING OF BREEDING: RED-TAILED HAWK  Responses to prey density  Red-tailed hawks, migrate south for the winter, and thus avoid winter food shortages at Kluane. The stable density of red-tailed hawks seen throughout the study (Table 5) and the fact that all pairs attempted to breed each year suggests that body condition in spring was always sufficient to initiate breeding. Red-tailed hawks started to breed immediately on arrival, suggesting females did not depend on local food supplies. Breeding by red-tailed hawks in Alberta was also independent of prey densities (Adamcik et al. 1979).  Responses to juvenile prey  The timing of peak food needs in young red-tailed hawks was positively (Figure 11) and closely correlated with the peak abundance of their main prey, young ground squirrels (Figure 9). Over the 7 years that diet information was collected, juveniles contributed over half of the ground squirrel biomass in the diet in 5 of 7 years (Table 7). In the two years that juvenile ground squirrels contributed less to the diet than adults, juveniles of other prey species were either very abundant (juvenile hares at the peak in hare numbers in 1990) or were available exceptionally early (red squirrels 1994, Figure 9). As expected if juveniles were a critical prey item, the timing of breeding was positively correlated with the peak in abundance of juvenile Arctic ground squirrel (Table 4, Figure 11). When the peak in squirrel young occurred early in a year, more young fledged per pair (Table 19). It would also be expected that more young should fledge per pair when birds breed such that the peak prey  37  consumption requirements match a peak in juvenile prey, but this was not seen (see discussion below for constraints). Constraints on breeding time The red-tailed hawk differs from both the owl and goshawk by being a migrant. Its breeding period like the goshawks is shorter than that of the great horned owl (Table 1). However, it starts to lay about a week later than the goshawk and the young depend on the adults for 10 - 20 days longer (Preston and Beane 1993). This migrant hawk arrives in late April as winter ends and when the majority of individuals of its main prey, the ground squirrel, are emerging from hibernation (Byrom 1997, T. Karels pers.com.). Its young are independent before the onset of winter, which may allow it some room to adapt its time of breeding to match the peak in prey availability. However, this flexibility in timing of laying is possibly tempered by the fact that young and adults both need time to migrate south. As a consequence of these time restrictions, it is probable that birds lay soon after their arrival, and therefore prior to a time when the local availability of prey has had time to influence the females body condition, and subsequently her ability to lay and the number of eggs that are laid. In addition, although they generally breed so that peaks in prey availability and consumption by the young coincide, later breeding was associated with mortality attributed to attacks on the young by blackflies. Across years, late breeding date was correlated with increasing nest failure (Figure 7). Although blackfly emergence was not recorded at Kluane, two studies in the boreal forest (Bennett 1960, Mason and Kusters 1990), showed that blackflies appeared predictably in late May and early June, peaking in numbers in early July. A l l deaths in this study took place in chicks which were 1-3 weeks of age in early July, a period of maximum growth rate in the young hawks (Fitch et al. 1946, Figure 9), and a period when young birds are vulnerable to any additional stress (Gill 1995). Therefore, I suggest that the blackflies caused the death of young, by either blood loss or through parasite infection at this critical stage in growth. Blackflies were also associated with mortality of nestling red-tailed hawks in Wyoming (Smith et al. 1998),.where death was attributed to blood loss, trauma and a parasitic (Leucocytozoon) infection. Further evidence of the influence of blackflies, comes from an earlier study at Kluane, where Leucocytozoon  38  infections caused death of fledged horned owls in 1991 (Hunter and Rohner 1996). The flies that attacked young owls were from the same group as seen attacking red-tailed hawks in this study. The breeding strategy of red-tailed hawks differed markedly from that of great horned owls and possibly goshawks (Squires and Reynolds 1997) in that all adult females attempted to breed, and this may in part explain why several cases of bigamy were recorded (Doyle 1996). Breeding every year appears to work well if breeding is initiated early in a year, however, if breeding starts late, then the investment into a brood may be wasted, with death of young caused by blackflies. Late breeding may also have survival consequences for the adults, as they have to accumulate the energy to allow moult and for migration (Wijnandts 1984, Newton 1998). The close link between red-tailed hawks and ground squirrels was expected from earlier work (Fitch et al. 1946, Luttich et al. 1970). Red-tails hunt by soaring over open habitat and from elevated perches (Preston and Beane 1993), and ground squirrels are associated with open habitat at Kluane (Boutin et al. 1995). Consequently, it was not surprising that ground squirrels were the main prey by biomass (Figure 10), and were selected in some years (Figure 14). However, red squirrels and occasionally hares also contributed a large percentage of the biomass eaten in many years. I suggest that the close link observed between red-tails and ground squirrels is driven by red-tails reproductive strategy of breeding every year, a strategy that is unusual in raptors breeding at higher latitudes. Typically, raptors like the goshawk and great horned owl in this study, exhibit large fluctuations in the numbers of birds breeding annually, which is in turn associated with fluctuation in prey density (Newton 1979). If the annually breeding redtailed hawk, bred such that prey requirements of its young matched a broad peak in peak abundance, and yet those prey were highly cyclic (hares, grouse, voles), or had a wide annual range of peak abundance dates (red squirrel. Fig. 3c), then in many years breeding attempts could fail. Conversely, the annual timing in the peak of the red-tailed hawk main prey, the ground squirrels is relatively constant (<7 days in 5 of 6 years, Fig. 3b), and compared to other cyclic prey, its population does not decline to as low a population density (Fig. 4). Breeding to match this dependable peak should therefore provide the red-tails with enough prey to feed their young in most years.  39  In summary, the migrant red-tailed hawk avoids winter food shortages and attempts to breed every year. Breeding is synchronized so that a peak in its prey matches a peak in prey requirements, however, late breeding can result in attacks on the young by blackflies, causing death, and consequently a reduction in the number of young fledging.  TIMING OF BREEDING: NORTHERN GOSHAWK  Responses to prey density  In contrast to the red-tailed hawk, the decline in breeding population and probably the density of northern goshawks (Table 11) suggest that like the great horned owl, it is affected by winter food abundance. The population density of goshawks was correlated with just one prey species - the snowshoe hare, its most common prey (Table 18) and in the two years of low hare densities after the hare crash, 1992 and 1993, only one goshawk pair was detected. Timing of breeding was also correlated to snowshoe hare density, suggesting that the abundance of this prey determined the annual timing of breeding. Three other prey contributed to the winter diet, spruce grouse, willow ptarmigan and red-squirrel. Grouse and ptarmigan, like the snowshoe hare are winter residents and they contributed the largest biomass to the diet during the hare low (1993 - 1995, Figure 20), and the second largest proportion of biomass overall (Appendix 19). Spruce grouse were eaten more often than ptarmigan in all years (F.D. pers.obs.). Goshawks in Europe also depend on tetraonids in winter (Linden and Wikman 1983, Widen 1987). At Kluane, the density of spruce grouse was not correlated to the number of breeding goshawks. In contrast, the proportion of pairs fledging young increased significantly with an increase in grouse density. These results suggest that, although goshawk production responds to spruce grouse abundance, spruce grouse alone cannot support goshawks, or influence their overall reproductive strategy. Linden and Wikman (1983), also found that brood sizes of goshawks remained relatively constant during a period of declining hazel grouse (Bonasa bonasia) populations, but that the number of breeding goshawks declined. For the other common tetraonid at Kluane the willow ptarmigan, no local information on annual populations is available, although it is known that females in the region move from  40  shrub tundra to the forest in winter and are therefore available as prey (Gruys 1991). Ptarmigan and hare populations at the Chilkat Pass (200 km south from my study site, 1979 1992, Hannon 1983) and elsewhere (Hannon et al. 1998) fluctuate in parallel with hares at Kluane. However, the influence of ptarmigan on breeding in goshawks is probably small, as female goshawks are fed by the male at nest sites at least 1-2 months prior to lay, a period of time which presumably affects her body condition. Yet, at many nests ptarmigan remains were absent (3 of 8 in 1990), suggesting that they do not play a dominant role in the timing of breeding (Doyle and Smith 1994) in the population as a whole. No correlation was seen in breeding density or the timing of breeding with the abundance of the other common winter prey, the red squirrel. This was somewhat surprising as this prey is common (<90% of pellets) in the diet of goshawks in British Columbia (Doyle and Mahon 1998, Mahon and Doyle 1999). However, the goshawks in the British Columbia studies are smaller than birds in the Kluane study (F.D. unpubl. data), and may therefore be more adept at catching this small arboreal agile prey.  Responses to juvenile prey  In goshawks, the peak of prey consumption by young was matched to a broad peak in abundance of several prey (Figure 19), with a peak in juvenile hares and grouse available before the peak in prey requirements in all years. Ground squirrel abundance peaked around the peak in requirements in all years, and red squirrels peaked annually from 30 days prior to and 30 days after the peak in prey requirements. However, no significant link between the timing in abundance in any one prey and the timing of breeding in the goshawk was observed, and this broad peak in available prey, occurred at the same time annually. The only significant correlation between the timing of the peak in prey, and reproductive output, was between goshawks and snowshoe hares, with more young fledging per, in years when the 1 hare litter was born early (Figure 23). st  Constraints on breeding time  41  The goshawk was resident in most (or possibly all) years, and differed from the other resident the great horned owl by having a broader annual diet (Table 13). Like the red-tailed hawk, it has a shorter breeding period than the owl (Table 1), and times its breeding to match the peak prey consumption by the young with the availability of a broad peak in juvenile prey (Figure 19). Hatching occurs just after the first large influx of young naive prey, with the appearance of fledging gray jays (Perisoreus canadensis, Krebs et al. in press). In the fall, young goshawks become independent in late August (Doyle and Mahon 1998), before migration and hibernation markedly decrease the availability of prey. Like the horned owl, goshawks appear to breed only when prey are plentiful, and breeding attempts are typically successful (Table 11). As with the owl, lifetime reproductive success may be improved by not breeding at all if its body condition is poor in spring. In this case, goshawks are resident on territories at least 1-2 months prior to laying and therefore their body condition is in a large part dependent on winter prey densities. In addition, breeding adults not only have to raise young successfully but also to maintain their own body condition in winter and spring (Wijnandts 1984, Newton 1998). Breeding only when a threshold body condition is attained (a requirement for a minimum threshold in body condition as indicated by body weight prior to'breeding, is well documented in raptors) (review in Newton 1979), increases the chance that young will survive to independence, and that adults will survive to breed in later years. Such a strategy is well suited to cyclic prey species. Nevertheless, goshawks face several constraints. In winters without abundant hares many birds died (all age groups, F.D. unpubl. data, Doyle and Smith 1994), some through predation by the larger great horned owls (O'Donoghue et al. 1995, Doyle and Smith 2000). Further, cold weather early in the breeding season could cause the abandonment of nests, as the energy demands of incubation may be outweighed by the need to maintain body condition (Mertens 1980, Wijnandts 1984, Riedstra 1998). Late breeding by goshawks did not expose small young to mortality from blackflies. Further, no losses of young due to blackfly attacks have been observed in other studies  ;)  (Squires and Reynolds 1997). In summary, goshawks, like the great horned owl, are resident in winter and only bred when they could do so successfully. Because of their broad peak in prey, it is unclear if there  42  would be any benefit to fine scale adjustments in timing. Mortality of yearlings and adults was high in the first winter with low hare abundance.  TIMING OF BREEDING: G R E A T HORNED OWL  Responses to prey density In great horned owls, the crash of their main prey the snowshoe hare caused a dramatic decline in both the density and breeding population (Table 15). This decline in owl numbers was not correlated closely with hare densities due to a 1-2 year time lag (Rohner 1994). In contrast, the decline in hare densities immediately lowered the proportion of pairs fledging young (Table 18). Only one pair fledged young over 3 years at the low in hare cycle (1992-1994). A threshold in body condition or in prey density (snowshoe hares) was therefore reached beneath which they did not breed or even engage in territorial or courtship hooting (Doyle 1997). In Alberta, Adamcik et al. (1978) also found that great horned owls did not attempt to breed when cyclic hare densities were at their lowest, and a similar relationship has been seen between breeding and food supply in many other raptors (Southern 1970, Village 1981, Korpimaki 1985, Daan et al. 1986). However, despite the dependence on hares, no link between hare density and the timing of breeding was observed at Kluane. The great horned owl bred 5 - 6 weeks before the goshawk and red-tailed hawk (Table 1), which may also may explain why the owl is so dependent on snowshoe hares in the boreal forest. Unlike the goshawk, the owl showed no reproductive response to increasing densities of grouse. Birds contributed only 2.5% of the prey biomass in the horned owl diet in 1995, while they contributed >50% of the prey biomass taken by goshawks (Table 13). The hare is the only common large nocturnal prey available during the timing of laying in great horned owls (Banfield 1974). Smaller prey (mice and voles) are active, but the owl did not breed when this prey group was at its peak from 1992 - 1994. This lack of response may be due to the half-meter snow depths typical at Kluane when owls lay (Boutin et al. 1995). Deep snow affords mice and voles protection from avian predation (Taitt et al. 1981, Sonerud 1986, Korpimaki 1992).  43  Responses to juvenile prey Timing of breeding in the great horned owl was not correlated closely with the peak in the abundance of its main prey, the snowshoe hare (Figure 27). However, when young (< 20 day) juvenile hares were available, they were taken in large numbers and in preference to older hares (Rohner and Krebs 1996). It is also likely that the peak food consumption of young owls does not occur until young leave the nest (Rohner and Hunter 1996). Fledgling great horned owls continue growing for another 20-30 days (Reed 1925), while red-tailed hawks and goshawks fledge at adult weight (Fitch et al. 1946, Schnell 1958). Juvenile hares (first litter) became available to owls in the late nestling and early fledgling periods. During the period of peak hare numbers (1989 - 1991) (Figure 27), they were commonly taken and numerically far exceeded any other prey in the owl diet (Table 17). However, if there were a strong relationship between the availability of juvenile hares and the peak consumption requirements of the young, I would have expected that birds that bred too early or late in a year would fledge fewer young (Table 16). This was not seen.  Constraints on breeding time The great horned needs nearly 50% more time to raise its young than the goshawk and red-tailed hawk (Table 1). Much of this additional time (40-70 days) is used to feed the young after fledging but prior to independence. This and the short boreal summer (Chabot and Mooney 1985) may force these owls to breed early in a year, thus enabling their young to become independent before the onset of winter. This may explain why great horned owls lay (Table 14) in mid-March, when mean maximum daily temperatures are below freezing (Appendix 11), and when the subsequent daily energy expenditures for incubation are greater than if the birds bred later. Great horned owls rely heavily on a cyclically abundant prey, the snowshoe hare. Hares contribute most biomass to the diets (Figure 28), and few other prey are taken at the peak in the hare cycle (1989-1991, Table 17). During the low in hare numbers (1992-1993, Appendix 4), no birds bred (Table 15). When owls began to lay in March, few other prey  44  other than hares were available. Although mice and voles were taken in large numbers by owls in summer (Table 17) particularly in 1994-1995 when few hares were taken. However, despite increasing densities of voles in 1991 and 1992 breeding by the owls declined, presumably because voles were inaccessible under the snow in winter. When owls did breed with increasing hare densities they were remarkably successful, and nearly all chicks survived to fledge (Table 15). It is possible that birds in poor condition do not attempt to breed. Birds need to attain a certain body condition to survive the winter (Wijnandts 1984, Newton 1998), and there may be a second and higher threshold of body condition that allows breeding (Stearns 1993). Poor body condition in the owls was likely at the low in the hare cycle in the winter 1992, when many owls died from starvation and from predation by other great horned owls (Rohner 1996, Doyle and Smith 2000, Rohner and Doyle 2000). A further constraint on late breeding in great horned owls may be attacks on young by blackflies. Blackflies caused death of fledged juveniles in years of reduced food supply at Kluane (Hunter and Rohner 1996). Owls may breed early to reduce this mortality. However, great horned owls also breed early across their large range (Houston et al. 1998), suggesting that the extended time needed for breeding determines the early onset of breeding, unless of course blackflies are found throughout their range? Comparative studies of owl populations in areas with and without blackflies are needed to resolve this issue. With its long breeding period and the possible effects on reproductive success through attacks on the young by blackflies, why did the owls not breed significantly earlier in years of high hare abundance?  It is possible that the ability of the birds to breed any earlier  could be effected by cold temperatures, with the average mean maximum daily temperatures in February > 8° C colder, than in March (Appendix 11). Or it may be that the appearance of juvenile hares in the fledgling period was necessary to maximize reproductive success, by increasing the survival of young. This later suggestion is supported by a brood manipulation experiment by Rohner and Smith (1996), which concluded that when young were 35 -78 days old, prey was not superabundant even at the cyclic peak in its prey (snowshoe hare). In summary, the horned owl breeds only in years when food is abundant, and when it does breed it is invariably successful. It does not adjust its timing of breeding to match shifting peaks in prey availability.  45  Summary: Timing of breeding in the raptors  These three raptors have different breeding strategies. Great horned owls only breed if prey is abundant in winter. Timing of breeding in the owl is such that the peak demands of its young is prior to a peak in prey, however when they do breed they are almost invariably successful. Goshawks, like owls are constrained in the ability to breed by the abundance of winter prey, but when they do breed, their shorter breeding period allows them to breed such that a broad peak in abundance of young prey is available in all years (Figure 19). The redtailed hawk avoids local winter food shortages by migrating, and is therefore in good enough condition to attempt to breed every year. The length of its breeding period allows it to fine tune breeding so that the young's food requirements coincide with a peak in food availability (Figure 7). However, breeding late in a year to match the peak in prey can in particularly late years, result in the mortality of young due to blackflies.  G E N E T I C CONSTRAINTS O N T I M I N G OF B R E E D I N G  This study did not reveal any suggestion of heritable effects on the timing of breeding of the type seen in blue tits in the Mediterranean (Blondel et al. 1993). However, this is not surprising as the study area is centered within the homogenous boreal forest habitat, in contrast to the different habitat types, in which the blue tit study was conducted. However, a genetic influence may affect the latest annual breeding dates at which the resident owl and semi-resident goshawk will breed. Breeding too late in a season may see the death of both young and adults with the early onset of winter at this latitude. Similarly, there may also be a genetic selection in red-tails, whereby breeding late in the season is selected against, as bird need time after breeding (both adults and their young) to attain a threshold in body condition to allow them time to migrate.  S Y N C H R O N I Z I N G P R E Y A V A I L A B I L I T Y W I T H T I M I N G OF B R E E D I N G  46  Apart from the possible cues of temperature as seen in the red-tailed hawk, how can raptors time laying such that a later peak in prey consumption by their young is matched to a peak (s) in prey availability? A possible mechanism is the predictable cue to a future peak in prey brought about by mating in that prey, particularly if the act of mating increases the vulnerability of that prey, which will bring about a proximate boost in body condition of the raptor. Species known to become more vulnerable to predation during courtship include the male black grouse (Tetrao tetrix) (Angelstam 1984) which are vulnerable to during lekking, and similarly, male frogs that are preyed upon by carnivorous bats while calling (Tuttle and Ryan 1981). In sparrowhawks, (Accipiter nisus), Newton and Marquiss (1982) hypothesized that breeding condition was affected by the appearance of young fledged passerines (this hawk's main prey), and this in turn synchronized peak nestling prey demands with prey availability. In a guild of raptors in the central Canadian Arctic, the timing of laying was closely associated with the observed appearance of their main prey, ptarmigan and ground squirrels (Poole and Bromley 1987). At Kluane, predation rates on male ground squirrels are high during mating (R. Boonstra. unpub.data), and mating in ground squirrels takes place as soon as the females emerge from their hibernacula in late April and early May (Michener 1979, Hubbs and Boonstra 1997, Byrom 1997). Similarly the first red-tailed hawks observed at Kluane in late April, display and build nests immediately upon their arrival (F.D. pers. obs.), with their young reaching peak food consumption requirements soon after the emergence of the young ground squirrels. It is possible therefore, that the red-tailed hawk is using the increased availability of its main prey the ground squirrel during mating as a cue to the initiation of laying, such that a later peak in prey (juvenile ground squirrels) matches the peak in the young hawks prey requirements. Goshawks may be using a similar mechanism. However, by being resident and with the ability to take a wider range of large prey than the red-tailed hawk, it may be using the peak in availability of several prey as a cue to timing of breeding, through their influence on the females body condition. A l l four of its main prey, snowshoe hare, spruce grouse, willow ptarmigan and Arctic ground squirrel, conduct courtship activity from mid-April onwards (Hannon et al. 1998, Boutin et al. 1995), and as we saw both grouse spp. and Arctic ground  47  squirrels are more vulnerable during mating. Hares may also be more vulnerable during mating, however the evidence is anecdotal. Hares appearing oblivious to risk during mating chases at Kluane (F.D. unpubl.obs.), with chases frequent during the day, an uncharacteristic activity period for this normally crepuscular animal. These observations suggest that hares may be more vulnerable to predation at this time. The incubation period of the goshawk (36 - 38 days) (Ehrlich et al. 1988) closely matches the gestation period of hares (35 days) (Stefan 1998). Mating activity from mid-April onwards in the hares and the other three main prey, therefore leads to a later peak in prey abundance at a time when the prey consumption demands of the young goshawks are also at their peak (Figure 19). In conclusion prey species can become more vulnerable to predation during courtship and this could increase body condition in raptors, and stimulate breeding. If this happens synchronization between prey availability and peak food requirements of the young red-tails and possibly goshawks is ensured because the time needed for gestation/emergence of young prey, and that for incubation and early development in these hawks are the same.  I N T E R A C T I N G E F F E C T S O F F O O D A N D A G E O N V A R I A B I L I T Y OF B R E E D I N G D A T E S  The two raptors that depended most on winter prey, the great horned owl and northern goshawk both bred slightly earlier in years of high prey abundance than in years of moderate prey abundance (Fig. 21 and 31), and at very low prey abundance, little breeding took place at all. Both species showed a greater range of breeding dates in years of high prey abundance (1989 - 1991) than in years of moderate prey abundance (1994 - 1996, Fig. 6bc ) . Why did the range of breeding dates increase with increased prey abundance? In particular, did a change in the range of breeding dates affect the factors seen to be influencing the timing of breeding? In the European kestrel (Falco tinnunculus), birds in good condition breed early while birds in poor condition breed later (Cave 1968). In poor food years, only individuals in good body condition were able to lay and did so later than in years when food was abundant. In many birds, older individuals breed earlier (Snow 1958, Newton 1976, Perrins and McLeery 1989, Hochachka 1990), and individuals with greater hunting experience often have a higher body condition (Lack 1966). At Kluane, most pairs of horned owls and  48  goshawks bred during the period of high prey abundance, and few pairs bred during periods of low prey abundance (Tables 11 and 15). As a consequence of this breeding pattern, young birds ( 1 - 3 years old) entered the breeding population frequently near the peak in prey densities, but probably did not after the period of low prey densities (Doyle and Smith 1994, Rohner 1995). One likely consequence of the relationship between body condition, age and prey abundance is that during the peak in prey abundance (1989 - 1991), the experienced adults would have bred early, while young adults would have bred late, thereby creating a wide range in breeding dates. In contrast, the number of young fledged from early versus late breeding pairs (Table 12 and 16), in both goshawks and horned owls did not decline as expected if many late breeders were younger birds (Newton 1976, Smith 1988, Saether 1990, Nager and van Noordwijk 1995). However, a decrease in productivity by young birds is not seen in all species (Porter and Wiemeyer 1972), but whether or not it is just young birds in poor condition (Newton and Marquiss 1984), a greater range of breeding dates with increasing densities of prey still exists in this study. A full knowledge of age structure may therefore be needed to understand the interaction of food and timing of breeding. It is also possible that the greater range of breeding dates with increasing prey populations is a result of larger sample sizes, which may consequently lead to a wider spread in random numbers. This hypothesis appears unlikely in goshawks with sample sizes in years of moderate prey abundance very similar to those in years of high prey abundance (Figure 6b), however it could influence the pattern in breeding dates observed in great horned owls (Figure 6c).  A S S U M P T I O N S A N D BIASES I N THIS THESIS  I assumed that all northern goshawks remained resident in winter at high prey densities, as birds were frequently seen in all months (Doyle and Smith 1994). During periods of low prey density however, parts of the population may have been transient or migratory. In addition I could not census this species accurately unless pairs were breeding. Goshawks are winter residents at low prey densities in sub-alpine boreal forest in central British Columbia (F.D. unpubl. data), but avian populations further north exhibit a higher degree of migratory behavior (Kerlinger 1989).  Evidence for the existence of migrant  49  goshawks comes from migration stations, where large numbers of birds fly south in periods of dramatic hare declines (Mueller and Berger 1967, Hofslund 1973). However, there is no evidence of a large northerly migration in spring, and it is not know if birds seen flying south are migrants or transients. Indeed large numbers of goshawks reported as migrants by a raptor banding station 500 km north (Mclntyre and Ambrose 1999) of this study site were not necessarily migrating (birds were not always flying in the direction of other migrating raptors, Mclntyre pers. comm.). Goshawks elsewhere move locally (Squires and Ruggiero 1995) perhaps to areas of predictably higher winter prey abundance (Marcstrom and Kenward 1981). Whatever the winter status of Kluane birds, goshawks are usually found around their nest areas in late March and early April (Squires and Reynolds 1997) a period when winter prey conditions are still in effect in the Yukon. I therefore assumed here that most birds were resident in the boreal forest throughout this study. A second assumption here is that the interval from laying to fledging is fixed for each species. I estimated lay date by backdating from fledge date or by using 4 primary feather th  measurements, using incubation and nestling periods from the literature. Both the length of the incubation (Aldrich and Raveling 1983) and the time it takes the young to develop (Bryant 1978, Martin 1987, Lindholm et al. 1994) can increase when prey are scarce. In years of poor food supply, fledging dates could therefore appear late even though laying dates were the same. The red-tailed hawk migrates to southern United States in winter. I have assumed that its winter prey is not limiting, and it is in good body condition upon its arrival at Kluane. M y only evidence for this is that all birds attempted to breed soon after their arrival which suggest all birds were in breeding condition (Cave 1968). M y study focused on the timing of the prey consumption peak and its relationship to the peak in prey availability. Little is known about post fledging survival and the lifetime reproductive success of raptors (Korpimaki 1992). At this study site, only the work on horned owls by Rohner and Hunter (1996) examined survival of the young, and then only in 3 years (1989-1991). In this study, Rohner found that first winter survival of fledged young varied from 75% during high prey densities to 20% during declining prey densities. If this variation is applied to early versus late breeders, conclusions about the timing of breeding in relation to peak prey availability and reproductive performance may be false. However, in  50  the species that most closely followed a peak in prey, the red-tailed hawk, as many pairs bred early versus late in relation to the peak in prey. It therefore appears that any effect that the timing of breeding in relation to the peak in prey, had on the subsequent survival of the fledged young, was not applicable in this case. Survival of fledgling raptors varies across species, with higher survival of early-fledged young in the sparrowhawk (Newton 1986), and higher survival of heavier young in tawny owls (Strix aluco, Overskaug et al. 1999). Conversely under cyclic prey conditions, no relationship was seen between timing of breeding in a year and survival of fledged young in the Tengmalm's owl (Korpimaki and Lagerstrom 1988). Finally, a potential bias in the data, particularly for goshawks and red-tailed hawks is that I located a sub-set of breeders with relatively high breeding success, while pairs that failed early were missed. This was particularly likely early in the study, and therefore information on goshawks and great horned owls from 1988 and from red-tailed hawks in 1989 was omitted from this analysis. After this period, I consider this bias less likely because most pairs re-used traditional nest sites or nest areas. In red-tailed hawks, the stable population density supports this assumption. In goshawks and great horned owls, none of the reproductive variables that I looked at suggest that many birds were missed, with our "resident" raptors either breeding or not depending on prey abundance.  For all species I  therefore considered it unlikely that a strong bias due to variation in detection influenced the overall patterns observed.  L A C K ' S T H E O R Y A N D ITS A L T E R N A T I V E S  Of the three species studied, the two species with shorter breeding periods, the redtailed hawk and northern goshawk both corroborated Lack's 1954 theory that birds breed when a peak in prey matches the young's peak prey consumption needs. In the third species, the great horned owl, the short summer and long period of dependence on the adults probably led to the bird's inability to match the peak in prey requirements of the young to a peak in prey availability. From these observations, the next step in the analysis was to see if species and individuals that matched the consumption needs of the young with the peak in prey  51  availability reproduced better than those that did not. It appears though that either matching these peaks is not important in this system or that all individuals that breed are matching the peak so successfully that no pairs breed too early or late in relation to the peak in prey. Lack's 1954 theory, of course, is not the only possible explanation of the timing of breeding in birds; it is possible that other environmental or physiological processes are at work to initiate the timing of breeding. The circannual rhythm causes the initiation of breeding at a set time each year in many passerines (Gill 1995). However, a circannual rhythm is not the main force driving the timing of breeding in these raptors, as the goshawk and great horned owl did not breed every year and red-tailed hawks varied their timing of breeding annually, in synchrony with its main prey. It is possible however, that the resident great horned owl was inhibited from breeding to early during the peak in hare density, by a circannual mechanism, such as photoperiod. Similarly, it is also possible that great horned owls exhibit a genotypic response to the timing of laying at a time when harsh winter conditions may still prevail. Another mechanism that could allow a fine adjustment to the timing of breeding is variation in the length of the incubation and /or the nestling period. Some birds adjust the period of incubation (Aldrich and Raveling 1983, Nilsson and Smith 1988), and thus can improve the match of peak consumption by young to a peak in prey. In the Yukon, this possibility seems unlikely, as any reduction in the time spent incubating eggs increases the risk of predation, and /or the death of the embryo from extremes in environmental conditions. Unlike other birds (Martin 1987), raptors have few eggs and little opportunity to lay again, therefore the loss of even a single egg is a large loss in reproductive investment. It is more likely that development of the young is delayed by a lower food delivery rate (Ricklefs 1968, Martin 1987, Lindholm et al. 1994). This mechanism could work in 'early breeders' in both red-tailed hawks and goshawks, with relatively fine scale adjustment needed (<7 days) to match a peak in prey. If, however, this was the case, I would expect a loss of young (Ricklefs 1968), but this was not observed. Alternatively, Bryant (1978) observed that early breeders fledged smaller young, without the loss of any chicks. In the Kluane study, the detailed morphological measurements needed to look at these differences were not available. A final consideration is that birds might breed in spring under favorable weather conditions, and then simply eat what is available? Under such conditions, the timing of  52  breeding by red-tailed hawks or goshawks and their prey might match simply because predators and prey are developing at a similar rate, so that a peak in young prey occurs at a time when consumption demand of the young raptors is at its peak. This hypothesis seems unlikely in general, as other raptors at Kluane that eat smaller prey do not fit this pattern. Both northern harriers (Circus cyaneus) and boreal owls (Aegolius funereus) have a far longer incubation and nestling periods than the development times of their main prey (voles and small birds, Doyle and Smith 2000). It is also unlikely that if this was just an example of parallel development times of young predators and prey, that the timing in prey availability and prey requirement needs, of both the young red-tailed hawks and goshawks would be significantly correlated. It therefore seems likely that when red-tailed hawks and goshawks bred in synchrony with prey, it was a direct response to a predicted peak in this prey.  C O M P A R I S O N W I T H O T H E R STUDIES OF T H E T I M I N G O F B R E E D I N G IN BIRDS  In raptors, the only comparable studies are those on the sparrowhawk by Newton and Marquiss (1982, 1984) and on the kestrel by Cave (1968). These studies both lacked detailed information on prey populations, but still provide a useful basis for discussion. Like the red-tailed hawk, both of these species breed in the summer months under the backdrop of increasing prey populations, and a peak in available food occurred during the period of peak consumption of the young, thereby supporting Lack's hypothesis. Although the exact mechanism behind the match in prey demands of the young and prey availability are unknown, breeding date were correlated to spring temperatures, and a seasonal decline in breeding success was also seen, both features that I observed in the red-tailed hawk. At all stages breeding performance in kestrels and sparrowhawks was correlated to female body condition, with earlier breeding correlated to increasing body condition. In kestrels, body condition was in turn linked to the development of the oocytes in the ovary (Cave 1968), and in both species, breeding performance was also linked to the age of the breeders, with younger birds breeding later and fledging fewer young. In insectivorous species (Hochachka 1990, Perrins 1991, Brinkhof 1995, Nager and van Noordwijk 1995), fledging success is typically correlated to breeding date, with early breeders fledging more young. This is the same broad pattern as seen in the red-tailed  53  hawks in this study, but the reasons here were quite different. Late breeding in insectivorous species was linked to body condition and age, which either directly prevented the birds raising the maximum possible young, or birds modified their brood size as predicted by the brood reduction hypothesis (Drent and Daan 1980, Hbgstedt 1980, Dijkstra et al. 1990, Tinbergen and Dann 1990). In contrast, decreased fledging success late in the year in redtailed hawks was associated with parasitism of the nestlings by blackflies. Of the pairs that did fledge young, the number of young fledged was not correlated (r = -0.29, n = 7) with the timing of breeding in a year, suggesting that clutch size did not decline over the breeding season as would have been predicted by the brood reduction hypothesis. Goshawks and great horned owls both showed a similar pattern, with no reduction in the number of young fledged per pair.  FUTURE RESEARCH  Information on the diet of all three species comes from the detailed analysis of pellets (great horned owls and red-tailed hawks) or from prey remains (goshawk). While this information may be a reasonable representation of the prey brought to the nest over a season, it clearly does not adequately reflect fine scale selection. With sufficient detail from individual nests we might find that a specific prey (perhaps only available for a brief time) is essential to the survival of the nestlings. This study focused on reproductive success as defined by the ability of a breeding attempt to fledge young in a year and at certain times in a year. However, lifetime reproductive success is the most important criteria for individuals (Krebs 1994), and this needs to be looked at to see if birds that match prey consumption requirements with the peak in prey availability, are more successful than those breeding at other times. Across the range of these 3 raptor species (all are found throughout much of N . America) many different prey types and prey densities are encountered. How do they time their breeding in different areas in response to different prey densities and to the timing of the peak(s) in available prey? Are the mechanisms seen in this study not applicable in populations breeding further south where weather is less severe and non-cyclic prey populations predominate?  54  Similarly, across their range, these raptors must occur in areas with very different blackfly densities. By identifying such areas and monitoring the effects of blackflies on breeding success in early versus late breeding birds we can estimate whether this selection pressure regularly influences the timing of breeding in raptors.  55  LITERATURE CITED  Adamcik, R.S., A.W. Todd and L . B . Keith, 1978. Demographic and dietary responses of great horned owls during a snowshoe hare fluctuation. Canadian Field-Naturalist 92:156-166. Adamcik, R.S., A.W. Todd and L . B . Keith. 1979. Demographic and dietary responses of redtailed hawks during a snowshoe hare fluctuation. Canadian Field-Naturalist 93:16-27. Aldrich, T.W. and R. G. Raveling. 1983. Effects of experience and body weight on incubation behavior of Canada geese. Auk 100:670-679 Anglestram, P. 1984. Sexual and seasonal differences in mortality of the black grouse in boreal Sweden. Ornis Scand. 15:123-124. Arcese, P., J.N.M. Smith, W . M . Hochachka, C M . Rogers and D. Ludwig. 1992. Stability, regulation, and the determination of abundance in an insular song sparrow population. Ecology 73(3):805-822. Aumann, T. 1990. The breeding behaviour of a brown goshawk pair. Australian Bird Watcher 13:156-163 Banfield, A.W.F. 1974. The mammals of Canada. University of Toronto Press. Beecham, J.J. and M . N . Kochert. 1975. Breeding biology of the golden eagle in southwestern Idaho. Wilson Bulletin 87:506-513. Bell, F. D . 1995. Basic Biostatistics. First Edition. Indiana State University. Bennet, G. F. 1960. On some ornithophilic blood sucking diptera in Algonquin Park, Ontario, Canada. Canadian Journal of Zoology 38:377-389. Blondel, J.,P.C. Dias, M.Maistre, and P. Perret. 1993. Habitat heterogeneity and life-history variation of mediterranean blue tits (Parus caeruleus). Auk 110(3):511-520. Boulanger, J. 1993. Evaluation of capture-recapture estimators using a cyclic snowshoe hare  56  population. M.Sc. Thesis. University of British Columbia, Vancouver. Boutin, S. 1990. Food supplementation experiments with terrestrial vertebrates: patterns, problems and the future. Canadian Journal of Zoology 68:203-220. Boutin, S. and K.W. Larsen. 1993. Does food availability affect growth and survival of males and females differently in a promiscuous small mammal, Tamiasciurus hudsonics ? Journal of Animal Ecology 62:364-370. Boutin, S., C. J Krebs, R. Boonstra, M.R.T. Dale, S. J. Hannon, K . Martin, A.R.E. Sinclair, J.N.M.; Smith, R. Turkington, M . Blower, A. Byrom, F. I. Doyle, C Doyle, D. Hik, L . Hofer, A. Hubbs, T. Karels, D.L. Murray, V . Nams. M . O'Donoghue, C. Rohner, and S. Schweiger. 1995. Population changes of the vertebrate community during a snowshoe hare cycle in Canada's boreal forest. Oikos 74:69-80. Brinkhof, M . W. G. 1995. Timing of reproduction - an experimental study in coots. Ph.D.Thesis Netherlands. Brown, C R . and M . B . Brown. 1999. Fitness components associated with laying date in the cliff swallow. Condor 101:230-245. Bryant, D . M . 1978. Environmental influences on growth and survival of nestling house martins Delichon urbica. Ibis 120 (3): 271-283 Byrom, A . E . 1997. Population ecology of Arctic ground squirrels in the boreal forest during the decline and low phase of a snowshoe hare cycle. Ph.D. Thesis. University of British Columbia, Vancouver. Canadian Wildlife Service. In press. Birds of the Yukon. Cave, A. J. 1968. The breeding of the kestrel (Falco tinnunculus), in the reclaimed area Oostelijk Flevoland. Netherlands Journal of Zoology 18(3):313-407. Chabot, B. F. and H.A. Mooney. 1985. Physiological ecology of North American plant communities. New York: Chapman and Hall.  57  Craighead, J.J. and F.C. Craighead. 1956. Hawks, owls and wildlife. Dover Press. New York. Daan, S., C. Dijkstra, R. Drent and T. Meijer. 1986. Food supply and the annual timing of avian reproduction. In Proceedings Acta X I X Congress Internationalis Ornithologici. Eds. H . Oullet. 392-407. Daan, S., C. Djikstra and J. M Tinbergen. 1990. Family planning in the Kestrel (Falco tinnunculus): the ultimate control of covariation of laying date and clutch size. Behaviour 114:83-115. Doyle, F. I. 1995. Bigamy in red-tailed hawks in southwestern Yukon. Journal of Raptor Research 30(l):38-40. Doyle, F.I. 1997. In. Biology and conservation of owls of the Northern Hemisphere. Eds. Duncan, J.R.,D.H. Johnson and T.H. Nicholls :609. Doyle, F.I. and T. Mahon. 1998. Inventory of the northern goshawk in the Kispiox Forest District. Ministry of Environment, Lands and Parks. Smithers. Doyle, F.I. and J.N.M. Smith, 2000. Avian Predators. Goshawks and other raptors. Chapter 16. In: Ecosystem Dynamics of the Boreal Forest: The Kluane Project (eds. Krebs, C.J., S. Boutin and R. Boonstra) Oxford University Press, New York. (In press).  Doyle, F. I. and J.M.N. Smith. 1994. Population responses of northern goshawks to the 10year cycle in numbers of snowshoe hares. Studies in Avian Biology 16:122-129. Drent, R. H . and S. Daan. 1980. The prudent parent: energetic adjustments in avian breeding. Ardea 68:225-252. Dunning, J.B. 1984. Body weights of 686 species of North American birds. Western Bird Banding Association, Monograph. 1:1-38. Ehrlich, P. R., D.S. Dobkin and D. Wheye. 1988. The birder's handbook. New York: Simon and Schuster.  58  Fitch, H . S., F. Swenson and D.F. Tillotson 1946. Behavior and food habits of the red-tailed hawk. Condor 48:205-240. Folkard, N . F. G. 1990. A n experimental study of the plant-arthropod-bird food chain in the southwestern Yukon. M.Sc. Thesis University of British Columbia. Vancouver. Folkard, N . F. G. and J.N.M. Smith. 1995. Evidence for bottom-up effects in the boreal forest: do passerine birds respond to large-scale experimental fertilization? Canadian Journal of Zoology 73:2231-2237. Fyfe, R. W. and R. R. Olendorff 1976. Minimizing the dangers of nesting studies to raptors and other sensitive species. Canadian Wildlife Service. 23:1-17. Gilbert, B. S. and C.J. Krebs. 1991. Population dynamics of Clethrionomys and Peromyscus in southwestern Yukon 1973-1989. Holarctic Ecology. 14:250-259. Gill, F. B . 1995. Ornithology. Second edition.New York: W.H. Freeman and Company. Godfrey, W.E. 1986. The birds of Canada. Revised edition. National Museums of Canada. Green, R. 1976. Breeding behaviour of ospreys in Scotland. Ibis 118:475-490. Gruys, R. 1991. Winter movements and mortality of ptarmigan. M.Sc. Dept. of Zoology. University of Alberta. Hofslund, P.B. 1973. An invasion of goshawks. Journal of Raptor Research 7 (3-4):107-108. Hogstedt, G. 1980. Evolution of clutch size in birds: adaptive variation in relation to territory quality. Science 210:1148-1150. Hannon, S.J. 1983. Spacing and breeding density of willow ptarmigan in response to an experimental alteration of sex ratio. Journal of Animal Ecology 60:353-367. Hannon, S.J., P.K. Eason and K. Martin. 1998. Willow Ptarmigan. In The birds of North America. No. 369. (A. Poole and F. Gill, Eds.). Philadelphia: The Academy of Natural Sciences: Washington, D . C : The American Ornithologists Union.  59  Hemborg, C. 1999. Annual variation in the timing of breeding and moulting in male and female pied flycatchers (Ficedula hypoleucd). Ibis 141:226-232. Hochacka, W. 1990. Seasonal decline in reproductive performance of song sparrows. Ecology 7(4): 1279-1288. Hornfeldt, B . 1978. Synchronous population fluctuations in voles, small game, owls and tularemia in northern Sweden. Oecologia 32: 141-152. Hornfeldt, B . and U . Eklund. 1989. The effect of food on laying date and clutch-size in Tengmalm's owl Aegoliusfunerus. Ibis 132:395-406. Houston, C.S., Smith, D.G. and C. Rohner. 1998. Great horned owl. In Birds of North America. No. 378. (A. Poole and F. Gill, Eds.). Philadelphia: The Academy of Natural Sciences: Washington, D.C.: The American Ornithologists Union. Hubbs, A . H . 1994. The effects of food and predation on population regulation of the Arctic ground squirrel during the peak and decline of a snowshoe hare cycle. M.Sc. Thesis. University of Toronto. Hunter, D.B. and C. Rohner. 1996. Blood-sucking blackflies (Simuliidae) and blood parasites (Leucocytozoon sp.) as contributing factors of mortality in fledgling great horned owls. Journal of Wildlife Diseases. Hussell, D . J. T. and T.E. Quinney. 1985. Food abundance and clutch size of tree swallows Tachycineta bicolor. Ibis 129:243-258. Jacobs, J.D. and J.C. Wingfield. 2000. Endocrine control of life-cycle stages: A constraint on response to the environment? Condor 102:35-51. Johnson, S.J. 1986. Development of hunting and self-sufficiency in juvenile red-tailed hawks. Journal of Raptor Research 20(l):29-34 Kennedy, P. L . and D.W. Stahlecker. 1993. Responsiveness of nesting northern goshawks to taped broadcasts of 3 conspecific calls. Journal of Wildlife Management 57(2):249-  60  257. Kennedy, R.J. 1970. Direct effects of rain on birds: a review. British Birds 63:401-414 Kenward, R. E. 1978. Radio transmitters mounted on hawks. Ornis Scandinavia 9:220-223. Kenward, R. E., M . Karlbom and V . Marcstrom, 1983. The price of success in goshawk trapping. Journal of Raptor Research 17(3):84-91. Kerlinger, P. 1989. Flight strategies of Migrating Hawks. The University of Chicago Press, Chicago. Kirnmel, J. T. and R.H. Yahner. 1990. Response of northern goshawks to taped conspecific and great horned owl calls. Journal of Raptor Research. 24(4): 107-112. Korpimaki, E . 1985. Rapid tracking of microtine populations by their avian predators: possible evidence for stabilizing predation. Oikos 45(2):281-284. Korpimaki, E. 1992. Fluctuating food abundance determines the lifetime reproductive success of male Tengmalm's owls. Journal of Animal Ecology 61:103-111. Korpimaki, E. and M . Lagerstrom. 1988. Survival and natal dispersal of fledgings of Tengmalm's owl in relation to fluctuating food conditions and hatching date. Journal of Animal Ecology 57:433-441. Korpimaki, E.,H. Hakkarainen and G.F. Bennett. 1993. Blood parasites and reproductive success of Tengmalm's owls: detrimental effects on females but not on males? Functional Ecology 7:420-426. Kostrzewa, A. and R. Kostrzewa. 1990. The relationship of spring and summer weather with density and breeding performance of the buzzard Buteo buteo, goshawk Accipiter gentilis and kestrel Falco tinnunculus. Ibis 132:550-559. Krebs, C. J. 1994. Ecology. Harper Collins College Publishers. Krebs, C.J., B.S. Gilbert, S. Boutin, A.R.E. Sinclair and J.N.M. Smith. 1986. Population  61  biology of snowshoe hares. I. Demography of food-supplemented populations in the southern Yukon, 1976 - 84. Journal of Animal Ecology 55:963-982. Krebs, C. J., B.S. Gilbert, S. Boutin and R. Boonstra. 1987. Estimation of snowshoe hare population density from turd transects. Canadian Journal of Zoology 65:565-567. Krebs, C. J., S. Boutin, R. Boonstra, A.R.E. Sinclair, J.N.M. Smith, M.R.T. Dale, K. Martin and R. Turkington. 1995. Impact of food and predation on the snowshoe hare cycle. Science 269:1112-1115. Krebs, C. J., S. Boutin and R. Boonstra. (In Press). Ecosystem Dynamics of the Boreal Forest: The Kluane Project. Oxford University Press, New York. Lack, D. 1954. The natural regulation of animal numbers. Oxford: Clarenden Press. Lack, D. 1966. Population studies of birds. Oxford: Clarenden Press. Lambrechts, M . M . and P. C. Dias. 1993. Differences in the onset of laying between island and mainland Mediterranean blue tits (Parus caeruleus): phenotypic plasticity or genetic differences? Ibis 135:451-455.  Levins, R. 1968. Evolution in Changing Environments. Princeton University Press. New Jersey. Linden, H . and M . Wikman. 1983. Goshawk predation on tetraonids: Availability of prey and diet of the predator in the breeding season. Journal of Animal Ecology 52:953-968. Lindholm, A., G.Gauthier and A. Desrochers. 1994. Effects of hatch date and food supply on gosling growth in Arctic-nesting greater snow geese. Condor 96:898-908. Luttich, S.N.,D.H. Rusch, E.C. Meslow and L . B . Keith. 1970. Ecology of red-tailed hawk predation in Alberta. Ecology 51:190-203. Marcstrom, V . and R.E. Kenward. 1981. Movements of wintering goshawks in Sweden. Swedish Wildlife Research 12:1-36.  62  Marti, C D . 1987. Raptor food habits studies. Pages 67-80. In B.A. Giron Pendleton, B.A. Milsap, K.W. Cline and D. M . Bird (Eds), Raptor management techniques manual. Nat. Wildl.Fed. Washington, D.C.  Martin, T.E. 1987. Food as a limit on breeding birds: a life-history perspective. Annual Review of Ecological Systematics 18:453-487. Mclnville, W.B. and L . B . Keith. 1974. Predator-prey relations and breeding biology of the great horned owl and red-tailed hawk in central Alberta. Canadian Field Naturalist. 88(l):l-20. Mclntyre, C L . and R.E. Ambrose. 1999. Raptor migration in autumn through the Upper Tanana River Valley, Alaska. Western Birds 30:33-38. Mason, P.G. and P . M . Kusters. 1990. Seasonal activity of female blackflies (Diptera: Simulidae) in pastures in northeastern Saskatchewan. Canadian Entomologist 1990. 9/10:825-835. Mearns, R. and I. Newton. 1988. Factors affecting breeding success of peregrines in south Scotland. Journal of Animal Ecology 57:903-916. Mertens, J.A.L. 1980. The energy requirements for incubation in great tits and other bird species. Ardea 68:185-192. Michener, G.R. 1979. The circannual cycle of Richardson's ground squirrels in southern Alberta. Journal of Mammology 60:760-768. Mihok, S. 1988. The role of mortality in models of small rodent population dynamics. Oikos 52:215-218. Moore, T.D.,L.E. Spence and C E . Dugnolle. 1974. Identification of the dorsal guard hairs of some mammals of Wyoming. Wyoming Fish and Game Dept., Cheyenne, Wyo. Mueller, H . C and D.D. Berger. 1967. Some observations and comments on the periodic invasion of goshawks. Auk 84:183-191.  63  Nager, R.G. and A.J. van Noordwijk. 1995. Approximate and ultimate aspects of phenotypic plasticity in timing of great tit breeding in a heterogeneous environment. The American Naturalist 146(3): 454-474. Nagorsen, D.W. 1993. An identification manual to the small mammals of British Columbia. Special Report - Ministry of Environment, B C . Newton, I. 1976. Population limitation in diurnal raptors. The Canadian Field-Naturalist 90(3): 274-300. Newton, I. 1977. Timing and success of breeding in tundra-nesting geese. Pp: 113-126. In Evolutionary ecology' (eds. B . Stonehouse & C M . Perrins).London. Newton, I. 1980. The role of food in limiting bird numbers. Ardea 68:11-30. Newton, I. 1986. The sparrowhawk. T & A. D. Poyser. Newton, I. 1998. Population limitation in birds. Academic Press. Newton, I. and M . Marquiss. 1982. Food, predation and breeding season in sparrowhawks (Accipiternisus). Journal of Zoology. London. 197:221-240. Newton, I. and M . Marquiss. 1984. Seasonal trend in the breeding performance of sparrowhawks. Journal of Animal Ecology 53:809-829. Nilsson, J.A. and Smith, H.G. 1988. Incubation feeding as a male tactic for early hatching. Animal Behaviour 36:641-647. Norrdahl, K . and E. Korpimaki. 1998. Does mobility or sex of voles affect risk of predation by mammalian predators. Ecology 79(l):226-232. ODonoghue, M . 1991. Reproduction, juvenile survival and movements of juvenile snowshoe hares at a cyclic population peak. M.Sc. Thesis. University of British Columbia, Vancouver. O'Donoghue, M . 1997. Responses of coyotes and lynx to the snowshoe hare cycle. Ph.D.  64  Thesis. University of British Columbia, Vancouver. O'Donoghue, M . and S. Boutin. 1995. Does reproductive synchrony affect juvenile snowshoe hares. Canadian Journal of Zoology 70:1787-1791. O'Donoghue, M . and C.J. Krebs. 1992. Effects of supplemental food on snowshoe hare reproduction and juvenile growth at a cyclic population peak. Journal of Animal Ecology 61:1-135. Olsen, P.D. 1995. Australian birds of prey. The Johns Hopkins University Press. Baltimore. Maryland. Olsen, P. D . and J. Olsen. 1989. Breeding of the peregrine falcon Falco peregrinus: I. Weather, nest spacing and territory occupancy. Emu 88(4): 195-201. Otis, D.L., K.P. Burnham, G.C. White and D.R. Anderson. 1978. Statistical inference from capture data on closed animal populations. Wildlife Monographs 62: 1-135. Overskaug, K . J . P . Boldstad, P. Sunde and I.I. Oien. 1999. Fledgling behavior and survival in northern tawny owls. Condor 101:169-174. Penteriani, V . 1997. Long-term study of a goshawk breeding population on a mediterranean mountain (Abruzzi appenines, Central Italy): density, breeding performance and diet. Journal of Raptor Research 31(4):308-312. Perrins, C M . 1965. Population fluctuations and clutch size in the great tit (Parsus major). Journal of Animal Ecology 34:601-647 Perrins, C. M . 1970. The timing of birds' breeding seasons. Ibis 112:242-255. Perrins, C. M . 1991. Tits and their caterpillar food supply. Ibis 133:49-54. Perrins, C. M . and R.H. McCleery. 1989. Laying dates and clutch size in the great tit. Wilson Bulletin 101(2):236-253. Petersen, L.R. and J.R. Weir. 1976. Tether platforms - and improved technique for raptor  65  food habits study. Journal of Raptor Research 10: 21-28. Petersen, L. R. and D.R. Thompson. 1977. Aging nestling raptors by 4th-primary measurements. Journal of Wildlife Management 41(3):587-590. Porter, R.D. and S.N. Wiemeyer. 1972. Reproductive patterns in captive American kestrels. Condor 74:56-73. Poole, K . G . and R.G. Bromley. 1987. Interrelationship within a raptor guild in the central Canadian Arctic. Canadian Journal of Zoology 66:2275-2282. Preston, C R . and R.D. Beane. 1993. Red-tailed hawk. In Birds of North America. No. 52. (A. Poole and F. Gill, Eds.). Philadelphia: The Academy of Natural Sciences: Washington, D . C : The American Ornithologists Union. Reed, B.P. 1925. Growth, development, and reactions of young great horned owls. Auk 42:14-31. Reynolds, R.T., S.M. Joy and D.G. Leslie. 1994. Nest productivity, fidelity, and spacing of northern goshawks in northern Arizona. Studies in Avian Biology 16:106-113 Riedstra, B, C. Dijkstra and S. Dann. 1988. Daily energy expenditure of male and female marsh harrier nestlings. Auk 115(3):635-64. Ricklefs, R.E. 1968. Patterns of growth in birds. Ibis 110:419-451. Ricklefs, R.E. 2000. Lack, Skutch, and Moreau: The early development of life-history thinking. Condor 102:3-8. Roff, D.A. 1986. The evolution of wing dimorphism in insects. Evolution 40:1009-1020. Rohner, C. 1994. The numerical response of great horned owls to the snowshoe hare cycle in the boreal forest. Ph.D. Thesis. Univerisity of British Columbia. Rohner, C. 1995. Great horned owls and snowshoe hares: what causes the time lag in the numerical response of predators to cyclic prey? Oikos 74:61-68.  66  Rohner, C. 1996. The numerical response of great horned owls to the snowshoe hare cycle: consequences of non-territorial 'floaters' on demography. Journal of Animal Ecology 65:359-370. Rohner, C. and F.I. Doyle. 1992. Methods of locating great horned owl nests in the boreal forest. Journal of Raptor Research 26(l):33-35. Rohner, C. and F.I. Doyle. 2000. Great horned owl. Chapter 15. In: Ecosystem Dynamics of the Boreal Forest: The Kluane Project (eds. Krebs, C.J., S. Boutin and R. Boonstra) Oxford University Press, New York. (In press). Rohner, C. and D.B. Hunter. 1994. Survival of great horned owls (Bubo virginianus) during the first year of life: blackflies, blood parasites and the paradox of high post fledging mortality. Canadian Journal of Zoology. Rohner, C. and D.B. Hunter. 1996. First-year survival of great horned owls during a peak and decline of the snowshoe hare cycle. Canadian Journal of Zoology 74:1092-1097. Rohner, C. and C.J. Krebs. 1996. Owl predation on snowshoe hares: consequences of antipredator behaviour. Oecologia 108:303-310. Rohner, C. and J.N.M. Smith. 1996. Brood size manipulations in great horned owls Bubo virginianus: are predators food limited at the peak of prey cycles? Ibis 138:236-242. Rohner, C , J.N.M. Smith, J. Stroman, M . Joyce, F. I. Doyle and R. Boonstra. 1995. Northern hawk-owls in the Nearctic boreal forest: Prey selection and population consequences of multiple prey cycles. Condor 97:208-220. Saither, B.E. 1990. Age-specific variation in reproductive performance of birds. Current Ornithology 7:251-283. Schnell, J. H . 1958. Nesting behavior and food habits of goshawks in the Sierra Nevada of California. Condor 60:377-403. Simmons, R. 1986. Food provisioning, nestling growth and experimental manipulation of  67  brood size in the African redbreasted sparrowhawk Accipiter rufiventris. Ornis Scandinavica 17:31-40. Smith, K . G. 1988. Clutch-size dependent asynchronous hatching and brood reduction in Junco hyemalis. Auk 105:200-203. Smith, J.N.M., R.D. Montgomerie, M.J. Taitt and Y. Yom-Tov. 1980. A winter feeding experiment on an island song sparrow population. Oecologia 47:164-170. Smith, R.N.,S.T. Cain, S.H. Anderson, J.R. Dunk and E.S. Williams. 1998. Blackfly-induced mortality of nestling red-tailed hawks. Auk 115(2):368-375. Smith, J.N.M. and Folkard, N . F. G. In Press. "Other herbivores and small predators: arthropods, birds and mammals". Chapter 12. In "Ecosystem Dynamics of the Boreal Forest. The Kluane Project". Eds. Krebs, C. J., S. Boutin and R. Boonstra. In Press. Oxford University Press, New York. Snow, D.W. 1958. A study of blackbirds. London, Allen & Unwin. Sonerud, G.A. 1986. Effects of snow cover on seasonal changes in diet, habitat, and regional distribution of raptors that prey on small mammals in boreal zones of Fennoscandia. Holarctic. Ecology 9: 33-47. Southern, H . N . 1970. The natural control of a population of tawny owls. Journal of Zoology 162:197-285. Sovell, J.R. 1993. Attempt to determine the influence of parasitism on a snowshoe hare population during the peak and initial decline phase of a hare cycle. M.Sc. Thesis. University of Alberta. Springer, M . A . 1979. Growth analysis for aging: Female great horned owl. Ohio Journal of Science 79(l):37-39. Squires, J.R. and R.T. Reynolds. 1997. Northern goshawk. Birds of North America. No. 298. (A. Poole and F. Gill, Eds.). Philadelphia: The Academy of Natural Sciences:  68  Washington, D.C.: The American Ornithologists Union. Squires, J.R. and L.F. Ruggerio. 1995. Winter movements of adult northern goshawks that nested in southcentral Wyoming. Journal of Raptor Research 29(l):5-9 StatSoft, Inc. 1995. Statistica for Windows. Version 5.1. Tulsa, O K . Stearns, S.C. 1976. Life-history tactics: A review of the ideas. Quarterly Review of Biology 51:3-47. Stefan, C.I. 1998. Reproduction and pre-weaning juvenile survival in a cyclic population of snowshoe hares. M.Sc. Thesis. Steeger, C. and R.C. Ydenberg. 1993. Clutch size and initiation date of ospreys: natural patterns and the effect of a natural delay. Canadian Journal of Zoology 71:21412146. Steenhof, K., M . N . Kochert, L . B . Carpenter and R.N. Lehman. 1999. Long-term prarie falcon population changes in relation to prey abundance, weather, land uses, and habitat conditions. Condor 101:24-41. Taitt, M . J., J.H.W. Gipps, C.J. Krebs and Z. Dundjerski. 1981. The effect of extra food and cover  on declining populations of Microtus townsendii. Canadian Journal of  Zoology 59:1593-1597. Tinbergen, J.M. and S. Daan. 1990. Family planning in the great tit (Parsus major): optimal clutch size as integration of parent and offspring fitness. Behaviour 114:161-190. Tuttle, M . D . and M.J. Ryan. 1981. Bat predation and the evolution of frog vocalization in the Neotropics. Science 214:677-678. Village, A . 1981. The diet and breeding of long-eared owls in relation to vole numbers. Bird Study 28:215-224. van Balen, J.H. 1973. A comparative study of the breeding ecology of the great tit (Parsus major) in different habitats. Ardea 61:1-93.  69  van Noordwijk, A.J., McCleery and C M . Perrins. 1995. Selection for the timing of great tit breeding in relation to caterpillar growth and temperature. Journal of Ecology 64:451458. Watson, J.W. 1997. The golden eagle. Published by T & A . D . Poyser, London. Wieden, P. 1987. Goshawk predation during winter, spring and summer in a boreal forest area of central Sweden. Holarctic Ecology 10:1-7 Wijnandts, H . 1984. Ecological energetics of the long-eared owl (Asio otus). Ardea 72:1-92. Zar, J. H . 1984. Biostatistical analysis. New Jersey, USA: Prentice- Hall Inc.  70  Table 1. Comparisons between the life history characteristics and morphology of the great horned owl, redtailed hawk and northern goshawk (Doyle and Smith 1994, Rohner 1994, Rohner unpub.data., Doyle unpub. data). Mass: goshawks and great horned owlsfromindividuals trapped on the study site, red-tailed hawks from Craighead and Craighead (1956). Status: R = Resident, M = Migrant, * Breeding period = laying - dispersal of young, Estimated lay date: Estimated by backdatingfromknown aged young, pooled means of all years.  Species  Status  Mass Range (g.)  Great horned owl  R  1140- 1700  Breeding*  Estimated  Hunting  Main  Period  Lay date  Period  Prey  # Days  Median  180-210  23 March  Nocturnal  Snowshoe hare Mice/voles  Red-tailed hawk  M  1028 - 1224  130 - 140  6 May  Diurnal  Ground squirrel Red squirrel Snowshoe hare  Northern goshawk  R  775- 1335  110- 125  29 April  Diurnal  Snowshoe hare Grouse Ground squirrel  71  Table 2. Characteristics of timing of litters and broods in the species (groups) commonly found in the diets of raptors. (* Timing of litters and broods for the different species, occur at the same time annually, +/- week) (unpub.data)  Number of Species  Snowshoe hare  litters/broods  Litters/broods  (Range)  synchronous  3 (2-4)  Yes  Yes  Yes  Yes  No  No  Yes  Yes  Ground squirrel  Red squirrel  Mice/voles  *Timing of  1 (1-2)  litters/broods predictable annually  >1  Passerine spp.  Spruce grouse  72  Table 3. Mean parturition dates for snowshoe hares within the study area (1989 - 1992, 1994 -1996). (O'Donoghue 1991, Stefan 1998, Sovell unpub. data)  First Litter  Second Litter  Mean  Mean  Parturition (N) Date (+S.D.)  Third Litter  Mean  Parturition Range  (N) Date (+S.D.)  Parturition Range  (N) Date (+S.D)  1989  (8) 25/05 (1.86) 23/05 to 29/05  (12) 2/07 (3.36) 26/06 to 7/07  1990  (14) 25/05 (2.40) 20/05 to 29/05  (21) 27/06 (2.40) 21/06 to 29/06  (19) 31/07 (1.96) 28/07 to 8/08  1991  (7) 29/05 (2.92) 27/05 to 03/06  (11) 6/07 (2.23) 2/07 to 9/07  No Litter Recorded  1992  (5) 30/05(5.86) 21/05 to  (3) 3/07 (7.55) 26/06 to 11/07  No Litter Recorded  (9) 18/06 (3.06) 24/06 to 24/06  (10) 25/07 (3.8) 20/07 to 2/08  5/06  (7)  Ran£  8/08 (3.36) 2/08 to 13/08  1993 1994  (11) 13/05 (2.65)  9/05 to 17/06  1995  (6) 23/05 (5.82) 16/05 to  1996  (12) 22/05 (2.35) 19/05 to 28/06  2/06  (9) 20/06 (3.47) 13/06 to 27/06 (14) 29/06 (3.46) 22/06 to 4/07  (8) 29/07 (2.7) 28/07 to 2/08 (13) 2/08 (2.2) 30/07 to 5/08  Table 4. Red-tailed hawk laying and fledging dates (Julian days) from 1990 to 1996. Laying date estimated by aging nestlings and assuming a 30 day incubation period.  Estimated Year  (# Nests)  Mean laying date (May 1 = 121)  Mean  fledging  +S.D.  Range offledgingdates  date (July 1 = 182)  122  195  5.07  192- 206  1991  121  195  4.10  192- 201  1992  125  200  4.56  191 - 206  1990  12  1993  10  123  195  5.97  193 - 218  1994  14  119  192  5.01  185 - 202  1995  13  125  200  6.10  191-  212  1996  11  122  196  4.37  189-  192  74  Table 5. Red-tailed hawk reproductive performance (1990 - 1996).  Year  1990  1991  1992  1993  1994  1995  1996  Sample size (Active Territories)  14  12  13  11  12  16  13  Pairs per 100 km  10  11  12  9  10  10  10  2  Mean number of chicks  1.1(1.1)  1.2(1.1)  0.6(0.9)  1.6(0.8)  1.8(0.6)  0.6(0.8)  1.7(1.1)  fledged per pair (+S.D.)  Proportion of pairs that  0.6  0.6  0.25  0.9  1.0  1.8(0.8)  1.9(0.7)  1.8(0.5)  1.7(0.7)  1.6(0.6)  0.47  0.78  fledged young  Number of youngfledgingper  1.4(0.5) 2.2(0.6)  successful pair/nest (+S.D.)  75  Table 6. Seasonal variation in the number of red-tailed hawk chicks fledged per pair per quartile. Number of nests in parentheses.  Year  Lower Quartile  Middle Quartiles  Upper Quartile  1990  2.00(3)  1.33(6)  0.66(3)  1991  1.50(2)  1.75 (4)  1.67(3)  1992  1.00(2)  0.70(5)  0.00(2)  1993  2.00(3)  1.25(4)  1.67.(3)  1994  1.50(4)  2.00(5)  1.50(4)  1995  2.00(3)  0.50(6)  1.00(3)  1996  1.67(3)  2.40(5)  1.67(3)  Mean  1.70(20)  1.40(35)  1.24(21)  +S.D.  0.86  1.01  0.77  Table 7. Red-tailed hawk diet for 1989-1995 at Kluane Lake, using minimum counts of prey species identified in pellets collected during the breeding season. Biomass of each category of prey is expressed as a percentage of the total biomass in the diet (A) = Adult, (J) = Juvenile.  Year  Age  1989  1990  1991  1992  1993  1994  1995  Class Percentage of total prey biomass (n)  Snowshoe hare  A  0.0  J  7.4(3) 42.0(45)  Arctic ground squirrel A J  Red squirrel  4.0(1)  8.4(2)  8.4(2)  0.0  2.4(1)  0.0  5.8(6)  4.8(5)  1.7 (2)  1.7(3)  9.3 (4)  37.9(10) 17.2(12)  27.0(18) 39.0(26)  21.1 (16)  15.5 (18)  3.6(1)  13.0 (7) 21.1 (30)  29.4 (40) 18.4(25)  29.8(46)  24.5 (58)  44.0(24)  39.1 (22) 13.5(20)  25.4(36) 24.7 (35)  37.3 (60)  51.5(127)  35.5(21)  Mice/voles  1.7 (9)  1.1 (15)  1.8(24)  2.2(29)  5.6(86)  2.1 (49)  2.3 (13)  Avian spp.  1.1 (3)  1.3 (9)  2.1 (14)  2.5 (17)  4.5 (34)  2.2(26)  5.4 (15)  Number of prey Total Biomass (kg) Number of Nests Diet Width  54  132  143  139  244  284  78  13.3  35.1  33.5  33.5  38.1  58.3  14.0  4  10  8  8  10  9  7  3.3  3.8  3.7  3.9  3.8  3.2  4.1  * Juvenile hare weight = 326.5 grams (n = 34) using measurements of right hind feet obtained from juvenile hare remains in nests and right hind foot regression equation (see Methods).  77  Table 8. Blackflies at red-tailed hawk nests. Nest failures, the % of nest failures attributed to high blackfly  numbers Q>70 flies), and age of chicks at the peak in fly numbers.  % of visits # Nests # Visits  with blackflies  flies  per  visit (range)  % nest  % failure  Mean chick  failures with >70 flies age at peak in flies  1992  5  25  68  11.8(0-100)  40  100  1993  6  34  15  1.5(0-20)  0  1994  7  63  32  2.7(0-50)  1995  7  25  44  14.8(0-100)  Mean fledging day (Julian)  19 (S.D. 11.5)  199  0  25 (S.D. 1.4)  194  0  0  27 (S.D. 5.4)  192  43  100  18 (S.D. 11.7)  200  78  Table 9. Annual number of red-tailed hawk breeding failures throughout the study period and the reasons for failed breeding attempts. In the case of 'Unknown' breeding failures, no sign of the young being depredated or dying from starvation was evident; chicks were either absent or intact and dead in and under the nest. Nest failure due to starvation occurred at bigamous nests (Doyle 1996).  Nest Failure  Nest Failure due  Failures  # Pairs  Number of  due to predation  to starvation in  due to "unknown"  Monitored  nest failures  or starvation  bigamous nests  causes  1990  14  5  0  0  5  1991  14  4  0  1  3  1992  13  10  3  2  5  1993  11  1  0  0  1  1994  16  0  0  0  0  1995  18  10  0  0  10  1996  13  2  0  0  0  Table 10. Northern goshawk laying and fledging dates (Julian days) from 1989- 1991 and 1994 -1996. Laying date estimated by aging nestlings and assuming a 36 day incubation period.  Year  (#)  Estimated mean  Laying date (May 1 = 121)  Mean fledging  S.D.  Range of fledging date  date (July 1 = 182)  1989  3  120  191  13.7  182 - 207  1990  8  115  186  ;.oo  173 - 197  1991  5  118  189  6.22  179 - 196  1992  1 breeding attempt failed  1993  1 pair fledged young. Date not recorded.  1994  3  123  194  3.51  191 - 197  1995  3  126  197  2.30  195 - 199  1996  4  120  191  0.58  189 - 190  80  Table 11. Northern goshawk reproductive performance from 1989 - 1996.  Year  1989  1990  1991  Sample size (Active Territories)  3  9  7  1  1  Pairs per 100 km  2  5  3  3  1  0.0  1.0  2  Mean number of chicks  1.3(1.5)  3.3(1.3)  1.3(1.3)  0.89  0.57  1992  1993  1994  1995  1996  3  3  4  2  2  1  2.3(0.6) 3.0(0.0)  3.8(0.5)  fledged per pair (±S.D.)  Proportion of pairs that  0.67  0  1.00  —  —  1.00  1.00  1.00  fledged young  Number of young fledged per  2.0(1.4) 3.8(0.5)  2.3(0.5)  2.3(0.6) 3.0(0.0)  3.8(0.5)  successful pair/nest (+S.D.)  Proportion of chicks fledging  0.57  1.00  0.70  0  ?  1.00  1.00  1.00  Table 12. Seasonal variation in the number of northern goshawk chicks fledged per pair per quartile. Number of nests in parentheses.  Year  Lower Quartile  Middle Quartiles  Upper Quartile  1989  3.00(1)  0.00(1)  1.00(1)  1990  4.00 (2)  3.50 (4)  4.00 (2)  1991  2.00(1)  1.66(3)  2.00(1)  1992  - — (0)  - — (0)  0.00 (0)  1993  - — (0)  (0)  (0)  1994  2.00(1)  3.00(1)  2.00(1)  1995  - — (0)  3.00 (2)  3.00 (1)  1996  4.00 (1)  3.50(2)  4.00(1)  Mean  3.17(6)  2.69(13)  2.87(7)  +S.D.  0.98  1.32  1.21  Table 13. Northern goshawk diet (1989-1996) at Kluane Lake, using minimum counts of prey species identified from remains collected during the breeding season. Biomass of each prey category is expressed as a percentage of the total biomass in the diet. (A) = Adult, (J) = Juvenile.  Year  Age 1989  1990  1991  1992  1993  1994  1995  1996  Class Percentage of total prey biomass (n)  Snowshoe hare  A  39.7(13)  J * 13.6(9)  31.6(32)  45.6(16)  44.6(1)  0.0(0)  10.3(2)  25.5(4)  52.0(6)  27.9(57)  19.7 (14)  0.0(0)  25.5 (2)  17.8 (7)  4.2 (2)  8.6(3)  Arctic ground squirrel  17.9(16)  27.4(76)  17.1 (17)  16.3 (1)  18.8(2)  20.6(11)  12.4(8)  12.7 (6)  Red squirrel  15.1(31)  1.6(10)  7.7(17)  0.0(0)  4.1(1)  11.4(14)  5.4(8)  6.4(9)  Grouse  12.8(12)  10.7(31)  9.0(9)  31.2(2)  36.1 (4)  36.0(20)  50.6(34)  18.3 (9)  0.9(8)  0.8(23)  0.8(8)  7.9(5)  15.5(7)  3.8(21)  1.8(12)  2.1(10)  Other Avian spp.  Number of prey Total Biomass (kg) Number of Nests  Diet Width  89  229  81  9  16  75  68  43  46.2  142.6  49.4  3.2  5.4  27.4  33.2  24.4  3  8  4  4.3  3.4  4.0  1  1  3  3  4  2.6  3.9  4.1  3.8  5.1  * Juvenile hare weight = 697.8 grams (n = 25) using measurements of right hind feet obtained from juvenile hare remains in nests and right hind foot regression equation (see Methods).  83  Table 14. Great horned owl mean laying and fledging dates (Julian days) (1989 - 1996) (Rohner 1994, Doyle unpub. data). (Days from laying to fledging from Hoffmeister and Setzer 1947, Craighead and Craighead 1956)  Year (# Nests) Laying date (April 1 = 91)  Mean fledging date (June 1 = 152)  S.D. Range of fledging dates  1989  12  72  150  9.94  128- 168  1990  18  74  152  3.73  140- 158  1991  20  77  155  6.35  143- 167  1992  No breeding  1993  No breeding  1994  1  102  180  1995  4  83  161  7.14  154- 170  1996  5  81  159  1.52  157- 161  84  Table 15. Great horned owl reproductive performance (1989 - 1996) (Rohner 1994, unpub.data)  Year  1989  1990  1991  1992  1993  1994  1995  Sample size (Active Territories)  14  21  27  25  17  13  12  9  Pairs per 100 km  15  19  22  24  14  10  11  10  2.1 (2.1) 2.2(1.9) 1.3(1.0)  0  0  0.2(0.0) 0.6(1.1)  0.86  0  0  0.08  2.4(0.8) 2.6(0.6) 1.8(0.8)  0  0  2.0(0.0)  0.97  0  0  2  Mean number of chicks fledged  1996  1.1 (1.2)  per pair (+S.D.)  Proportion of pairs that  0.86  0.74  0.33  0.67  fledged young  Number of young fledged per  2.3(0.5)  2.0(0.6)  successful pair/nest (+S.D.)  Proportion of chicks fledged  0.98  0.90  1.00  0.77  0.87  85  Table 16. Seasonal variation in the number of great horned owl chicks fledged per pair. Number of nests in parentheses (Rohner and Doyle unpub.data).  Year  Lower Quartile  Middle Quartiles  Upper Quartile  1989  2.67 (3)  2.33 (6)  2.33 (3)  1990  2.60(5)  2.63 (8)  2.80(5)  1991  1.80(5)  2.00(11)  2.00(5)  1992  No birds breeding  1993  No birds breeding  1994  Only one pair breed  1995  3.00(1)  1.00(2)  2.00(1)  1996  3.00(1)  1.67(3)  2.00(1)  Mean  2.40(15)  2.13 (30)  2.33(15)  +S.D.  0.63  0.86  0.82  Table 17. Great horned owl diet 1989-1995 at Kluane Lake, using minimum counts of prey species identified in pellets collected during the breeding season. Biomass of each prey category is expressed as a percentage of the total biomass in the diet. (Rohner 1994, Doyle unpub.data)  Year  1989  1990  1991  1992  1993  1994  1995  Percentage mass (n)  Adult hare  53.8(47)  42.8 (42)  50.5 (33)  Juvenile hare  29.3(163)  47.3 (295) 33.7(140)  0  (0)  22.7 (1)  40.0(12)  77.6(1)  18.8(4)  10.7 (3)  8.0(15)  12.2(1)  Ground squirrel  6.3 (22)  5.6(22)  12.2(32)  22.5(3)  11.4 (2)  9.3 (11)  0.0  Red squirrel  8.5(37)  2.2(11)  1.8(6)  6.0(1)  9.1 (2)  33.2(49)  0.0  Mice/voles  1.1 (44)  1.2(54)  1.4(42)  33.4(52)  43.6(88)  Avian spp.  1.0(4)  0.9(4)  0.3(1)  19.3(3)  2.5(5)  Number of prey  317  428  254  Total Mass (kg)  122.4  137.3  91.44  Number of Nests  10  10  10  6  6  4  Diet Width  2.1  1.6  2.0  2.0  1.6  3.2  7.5 (102)  6.8(4)  1.5(21)  3.3(2)  63  101  210  12  4.67  6.16  41.46  1.8 1  87  Table 18. Correlations between the annual abundance of a raptors main prey, andfivereproductive parameters. (* = P<0.05)  Number of Species  Main  Timing of Density of  prey  breeding  pairs  Proportion  young fledged  Number of young  of pairs that  per successful  fledged per pair  fledged young  pair  Great horned owl  Hare  0.76  0.39  0.89*  0.85*  Northern  Hare  0.93*  0.89*  0.36  -0.38  0.39  -0.15  0.31  -0.13  -0.43  0.53  0.54  goshawk Red-tailed hawk  Ground squirrel  88  Table 19. Correlations between the annual peak in abundance of a raptors main prey, and five reproductive parameters. (* = P<0.05)  Number of  Species  Density of  Proportion  young fledged  Number of young  of pairs that  per successful  fledged per pair  fledged young  pah-  Main  Timing of  prey  Breeding  owl  Hare  -0.44  -0.74  0.19  0.22  -0.47  Northern  Hare  -0.51  -0.10  0.50  0.84*  0.09  0.86*  0.86*  0.80  0.89*  0.01  pairs  Great horned  goshawk Red-tailed  Ground  hawk  squirrel  89  Figure 2. Map of the study area at Kluane Lake, southwestern Yukon. The map shows control plots, which were used to monitor, prey populations and the boundary of the 'Intensive Raptor Area' within which population trends and reproductive parameters of the raptors were monitored.  90  Figure 3a - d. Dates when young of the common prey taken by the three raptors became available. Sampl sizes, means (squares), standard deviations (squares) and ranges (whiskers) are given. Julian Date 1 June st  152.  3. a) 162  Snowshoe Hare  CD  cn 05  156  co oo  CO  in 0  -ST CO  150  >N  ca  To 5  T3  JQ  CO  c5  n = 7  144  T3  c  CO  n = 11  n = 5  S ^  138  CD O)  132  n = 12  n = 14 n = 6  c  (0 CC  126  1989  1990  1991  1992  1993  1994  1995  1996  1993  1994  1995  1996  Years  3. b) CD  co co  205 200  O)  c o  >• Io •- S-  I 8 I IO) cu E  CD CD OJ  c  CO  CC  1989  1990  1991  1992  Years  92  3. c)  270  co 05 CD  00 00  230  cn  co c  Red Squirrel  . .  3 to >, >* 03 190  O c  2.  w aj 03 "O  •D c (D CO o c CD O) CD E cu —  n = 374  150  n = 92  110  tr  70  n = 374  n = 186  o  CD O) c 03  n = 177  n = 113  1988  1989  n = 365  1990  1991  1992  n = 373  "= 29 1  1993  1994  1995  1996  1993  1994  1995  1996  Years  3. d)  176  160  1988  1989  1990  1991  1992 Years  93  a)  Figure 4. a) Snowshoe hares, Arctic ground squirrel and spruce grousefrom1988-1996. b) Passerine spp., Microtus spp., red backed vole and red squirrelfrom1988-1996. Snowshoe hare, Arctic ground squirrel, red squirrel and small mamma] indices are based on spring trapping estimates. Spruce grouse population index: birds seen per 100 hrs on foot in the field. Passerine spp: Point counts of birds May/June (Folkard and Smith 1995 and unpub. data).  94  Figure 5. Mean maximum temperature in March and April combined with Arctic ground squirrel emergence date.  95  a) Red-tailed  1992  Hawk  1993 Years  205 200  to  3  165 1991  1992  1996  1993  Years  Great  1992  Horned  Owl  1993  Years  Figure 6. Annual mean fledging date (Squares), Standard Deviation (square) and range of dates (whiskers), a) Red-tailed hawk, b) Northern goshawk, c) Great horned owl.  96  Figure 7. Red-tailed hawk. Relationship between mean annual fledging date (Julian days) and the proportion of monitored pairs that fledged young.  4  8  12  16  20  Relative Fledging Date (1990 - 1996) (n = 76)  Figure 8. Numbers of red-tailed hawk young fledging per pair and the timing of fledging within the year for each attempt (0 = earliest nest of year) (days sincefirstnest fledged young) that year.  97  Figure 9. Relationship between hatch and fledge dates (Julian days), prey consumption needs of the young, and mean emergence of young prey for the red-tailed hawk. Solid line indicates the mean weight of the young as a percentage of the adult weight. The dashed line (—) represents the peak daily prey consumption requirements of the young (Fitch et al. 1946). The height of the bands represents the percentage biomass contribution of each prey species to the diet over the breeding season. Key: Gr.spp = Grouse species, Gd.sq. = Ground squirrel, Rd.sq. = Red squirrel  98  99  Figure 10. Estimated percentage contributions of each prey species to the total biomass of prey identified from pellets of red-tailed hawk collected annually at nest sites.  100  Figure 11. Mean annual fledging date (Julian days) of red-tailed hawk and average emergence date of young Arctic ground squirrels.  Figure 12. Proportions of monitored pairs of red-tailed hawks that fledged young and average emergence date (Julian days) of young Arctic ground squirrels.  101  Figure 13. Number of young red-tailed hawk fledged per pair and mean annual emergence date (Julian days) of young red squirrels.  102  Figure 14. Arctic ground squirrel biomass in the red tailed hawk diet, and the population density of arctic ground squirrels (# per hectare) in spring.  103  1995 1992  1996  y = 200.12 - 1.546x r = -0.79  1990  1991 1993  1994  191 1  2  3  4  Temperature (mean) in March and April (Celsius)  Figure 15. Mean temperature in March and April and timing of fledging by red-tailed hawks.  104  2.6  17.2  17.8  18.4  19.0  19.6  20.2  T e m p e r a t u r e ( m e a n d a i l y m a x i m u m ) in J u n e ( C e l s i u s )  Figure 16. Relationship between the number of young red-tailed hawks fledged from successful nest and the mean daily maximum temperature (Celsius) in June.  105  Figure 17. Relationship between mean annual fledging date (Julian days) and percentage of nest failures in redtailed hawks that could not be attributed to a known cause (starvation, predation, etc,).  106  4.5  •  -2  • •  4  10  16  22  28  Relative Fledging Date (1989 - 1996) (n = 26)  Figure 18. Numbers of young northern goshawks fledging per pair and the timing of fledging within the year for each attempt (0 = earliest nest of year) that year.  ,107  Figure 19. Relationship between hatch and fledge dates (Julian days), prey consumption needs of the young, and mean timing emergence of young prey for northern goshawks. Solid line indicates the mean weight of the young as a percentage of the adult weight. The dashed line (—) represents the peak daily prey consumption requirements of the young (Schnell. 1958). The height of the bands represents the percentage biomass contribution of each prey species to the diet over the breeding period. Key: Gr.spp = Grouse species, Gd.sq. = Ground squirrel, Rd.sq. = Red squirrel  108  Hatch 152  132  132  172  152  172 1994 (Julian days)  1991 (Julian days)  Fled  9 l Rd.sq. J e  100  100  3D  SO  eo  eo  dO  iO  |20  20  Gd.sq. Harn ^nj lit j  •  iliiiililllll •HHHHMI  i  Hatch 132  152  - Hare 1st lit. 172 1995 (Julian days)  192  212  132  152  172 1996 (Julian days)  109  110  Snowshoe hare density in spring (# per hectare)  Figure 21. Northern goshawk. Relationship between the mean annual fledging date (Julian days) and snowshoe hare population density (# per hectare) in spring.  Annual snowshoe hare density (# per ha.) in spring  Figure 22. Number of northern goshawk pairs located annually and snowshoe hare population density (# per hectare) in spring.  Ill  Figure 24. Snowshoe hare population density in spring (# per hectare) and the percentage biomass contributed by hares to the estimated total prey biomass in the diets of goshawks.  113  Figure 25. Mean annual fledging date (Julian days) and the proportion of great horned owls pairs that fledged young (Rohner 1994, and this study).  3.5  3.0  • •  ••••••••<  2.5  •• •  2.0  • •  1.5  1.0  0.5  0.0  -0.5  10  20  30  40  Relative Fledging Date (1989 - 1996) (n = 60)  Figure 26. Numbers of great horned owl young fledging per pair and the timing of fledging within the year for each attempt (0 = earliest nest of year) (days sincefirstnest fledged young) that year. (1989 - 1996) (Rohner 1994, and this study).  115  Figure 27. Relationship between hatch and fledge dates (Julian days), prey consumption needs of the young, and mean emergence of young prey for great horned owls. The solid line indicates the percentage mean weight of the young in proportion to the adult weightfromhatching until premature fledging (Springer 1979). The dashed line (—) indicates weight gain of chicks that were monitored from late in the nestling phase until late into the post fledging period (Reed 1925). The height of the bands represents the percentage biomass contribution of each prey species to the diet over the breeding period. Key: Gr.spp = Grouse species, Gd.sq. = Ground squirrel, Rd.sq. = Red squirrel  116  : Rd.sq.  Fledge  \  Hare 2nd lit.  Atrian and Mundae spp. Gd.sq. _1  \ ' Hare 1st lit.  112  132  152  172  192  1990 (Julian days)  1989 (Julian days)  Muridae spp. •  Fledge  /Wain and Muridae spp  100  100  Hare 2nd lit. ao  so  60  60  20  20  : Rd.sq.-j— Gd.sq.  Hare 1st lit  Hare 1st lit.  Hatch 112  132  152 1991 (Julian days)  172  192  112  132  152  172  1994 (Julian days)  1995 (Julian days)  117  Figure 28. Annual estimated biomass contribution of the different prey species to the great horned owl diet (from the number of prey individuals identified in pellets) (Rohner 1994, and this study).  0.55  A n n u a l (mean) s n o w s h o e hare density (# p e r ha.) in spring  Figure 29. Snowshoe hare density in spring (# per hectare) and annual percentage biomass contribution of juvenile snowshoe hares to the great horned owl diet (from individuals identified in pellets) (Rohner 1994, and this study).  • 1994  y = 1 7 0 . 3 3 4 - 16.291X r = -0.76  • 1995 1996  ^  ^  1991  -  ^  ^ - v ^  1990  •  1989 -0.2  0.0  0.2  0.4  0.6  0.8  1.0  1.2  1.4  A n n u a l s n o w s h o e h a r e d e n s i t y (# p e r h a . ) in s p r i n g  Figure 30. Mean fledging date of great horned owls (Julian days) and snowshoe hare population density in spring (# per hectare) (Rohner 1994, and this study).  119  185  • T3  J  1994 (only 1 pair bred) 175  o> 170 co •o g 165 O) TD CD 5=  (No birds bred in 1992 and 1993, years when % biomass of the diet <0.34)  1995  •  160  1991  c  CO  •  I 155  * 1990  "co = 150 c  1989 • .1  < 145  0.3  0.5  0.7  0.9  1.1  1.  Snowshoe hare % (arcsine) biomass contribution to the diet  Figure 31. Proportion biomass contribution of snowshoe hares to the diet of great horned owls and the annual fledging date (Julian days).  Figure 32. Mean number of young great horned owls fledged per pair and snowshoe hare population density in spring (# per hectare) (Rohner 1994, and this study).  Figure 33. Mean number of young great horned owlsfledgedper monitored pair and the proportion biomass contribution of juvenile snowshoe hares in the diet (from individuals identified in pellets) (Rohner 1994, and this study).  121  Appendix 1. Emergence of young Arctic ground squirrel dates in the study area (1991 -1996) (Hubbs 1994, Byrom 1997, Karels unpub. data).  Year  Range  Average  1991  21/06 to 24/06  22/06  1992  04/07 to 16/07  10/07  1993  14/06 to 21/06  17/06  1994  13/06 to 19/06  16/06  1995  20/06 to 27/06  23/06  1996  17/06 to 23/06  20/06  122  Appendix 2. Emergence dates of young red squirrel ( l litter) within the study area sl  (1988 - 1996) (S.Boutin unpub.data).  Range  Year  (N)  Mean  +S.D.  1988  113  21/06/88  13.39  22/05 to 22/07  1989  186  1/06/89  10.61  1/05 to 25/06  1990  92  4/07/90  13.35  9/06 to 24/07  1991  177  27/06/91  13.23  27/05 to 24/07  1992  374  22/07/92  21.27  15/06 to 13/09  1993  374  3/07/93  36.99  17/05 to 11/09  1994  365  26/05/94  13.31  21/04 to 5/07  1995  373  30/07/95  22.98  6/04 to 19/09  1996  129  31/05/96  18.10  4/05 to 13/07  123  Appendix 3. Spruce grouse hatch dates obtained from radioed adult female birds within the study area. (K.Martin unpub.data)  Year  Mean hatch Date  (N)  1988  +S.D  Range  4  18/06/88  0.00  1  15/06/90  2  22/06/92  0.70  22/06 - 23/06  15/06/93  1.15  15/06- 17/06  12/06/94  0.00  12/06/95  1.52  10/06 - 13/06  16/06/96  3.65  12/06 - 21/06  1989  1990  1991  1992  1993  1994  2  1995  1996  5  Appendix 4. Snowshoe hare population density within the study area (1988 - 1996). The average densityfromtwo 32.5 ha plots is presented as animals per ha (Boutin et al. 1995 and unpub. data).  Spring  Fall  Average  Average  Density  Range  Year  Density  Ran ge  1988  0.52  0.14- 0.90  2.28  N/A  1989  0.93  0.27- 1.59  2.35  N/A  1990  1.51  1.43- 1.60  1.62  N/A  1991  1.02  0.89- 1.15  1.08  N/A  1992  0.29  0.25- 0.34  0.13  N/A  1993  0.09  0.07- 0.12  0.20  N/A  1994  0.07  0.02- 0.11  0.53  N/A  1995  0.13  0.08- 0.18  1.07  N/A  1996  0.49  0.39- 0.59  1.81  N/A  125  Appendix 5. Arctic ground squirrel population density within the study area (1990 - 1996). The average density from three 7.5 ha plots is presented as animals per ha (Boutin et al. 1995, T.Karels unpub.data).  Spring  Fall  Average  Average  Range  Year  Density  Range  Density  1990  1.75  1.48-2.02  1.24  1.11 -1.36  1991  2.16  1.72-2.59  2.92  1.52-4.32  1992  1.04  0.71 - 1.36  1.01  0.91 - 1.11  1993  0.91  0.70- 1.13  1.48  1.25-1.70  1994  0.84  0.80-0.86  2.25  2.00 - 2.50  1995  1.16  0.63 - 1.70  2.40  1.63 - 3.20  1996  1.55  1.50-1.60  2.44  2.34 - 2.50  Appendix 6. Red Squirrel population density within the study area (1988 -1996). The average densityfromtwo 10 ha plots is presented as the total number of. individuals caught per ha (Boutin et al. 1995, unpub. data.)  Spring  Fall  Average  Average Year  Density  Range  Density  Range  1988  22  19-25  22  18-25  1989  32  22-42  38  23-53  1990  23  19-27  19  15-23  1991  26  12-42  29  20-39  1992  28  23-32  34  26-43  1993  27  22-31  42  39-45  1994  41  34-47  45  38-52  1995  25  21-29  25  25-26  1996  34  26-41  Appendix 7. Vole (Microtus spp. and Clethrionomys rutilus) population densities within the study area (1988 - 1996). The average density from two 2.8 ha plots is presented as the minimum number of animals alive per ha (Boutin et al. 1995, R.Boonstra unpub.data).  Microtus spp. density  Clethrionomys rutilus density  Spring  Fall  Year  Spring  Fall  1988  6.0  12.0  1.2  1.6  1989  1.8  6.5  1.2  3.2  1990  4.0  3.5  1.2  1.5  1991  1.0  2.0  2.5  5.0  1992  3.5  6.0  5.0  10.0  1993  6.0  11.0  1.2  5.0  1994  3.5  3.0  2.5  2.5  1995  1.8  6.5  0.0  3.0  1996  0.04  0.03  0.0  0.6  Appendix 8. Index of spruce grouse population density within the study area (1988 - 1995), obtained from sightings of birds by researchers working on foot in the field. Sightings were standardized to observations per 100 hours in the field (Boutin et al. 1995). Summer = May - August, Winter = September - April.  Winter Observations  Year  Summer Observations  1988  14  10  1989  37  11  1990  10  8  1991  5  2  1992  6  3  1993  13  3  1994  45  5  1995  63  4  1996  —  —  129  Appendix 9. Passerine population density (2 X 36 hectare plots) within the study area (1988 - 1996). The average density from two 36 ha plots is presented as the pooled number of detections, of the seven most abundant songbirds, per ha (Folkard and Smith 1995 and J.N.M. Smith unpub. data).  Year  Density  1988  56  1989  42  1990  43  1991  45  1992  55  1993  —  1994  —  1995  68  1996  113  Appendix 10. Mean weights of prey: Prey weights were obtained from a random sample trapped on control plots in the study area, during the period when they were fed to the nestling raptors, unless otherwise stated.  Mean Species  (N)  Weight (g  +S.D.  Snowshoe hare (Adult)  50  1406.6  149.3 Obtained from actual raptor prey  Snowshoe hare (Juvenile) *  Arctic ground squirrel (Adult)  50  503.1  10.9  Arctic ground squirrel (Juvenile)  50  246.5  10.1  Red squirrel **  106  223.6  8.2  Ptarmigan (Adult)  790  467.5  Spruce grouse (Adult)  17  566.3  Bird ***  Comments  50.0  Adult and juvenile  (Mossop 1988) 42.5  Est. Using weights in Dunning 1984. and pers. comm. Jamie Smith  Cricetidae spp. (Adult)  30.0  Estimated weight. (Rohner et al. 1995)  *  See individual raptor species accounts for weight of juvenile hares fed to their young.  **  Random sample including both adults and juveniles.  *** Bird species used in sample are those documented as common species in the study area (Folkard and Smith 1995).  131  Appendix 11. Mean daily maximum temperature (Celsius). Recorded at Burwash Landing airport, 40km northwest of the study area (unpub.data). Cool months in bold, warm months underlined.  a) Individual months  yEAR  FEBRUARY  SD  MARCH  SD  APRIL  SD  MAY  SD  JUNE  SD  JULY  SD  1989  -9.9  7.8  -6.8  8.3  8.1  4.6  14.6  4  17.6  2.7  23.1  3.2  1990  -18.8  10.6  -0.4  5.1  8.4  3  14.6  4.3  17.9  3.1  20.8  3.7  1991  -7.4  8.6  -1.7  4.7  6.6  3.7  14.2  3.3  18.4  6.1  17.4  2.9  1992  -11.3  10.3  -0.5  7.7  3.5  5.2  9.6  4.3  18.3  4.4  18.7  3.8  1993  -6.4  10.9  -0.8  3.7  8.1  3.2  15.5  4.8  18.6  2.6  20.3  3.8  1994  -16.4  LZ  -0.7  7.9  83  3.4  12.6  3.2  18.6  3.4  20.8  3.5  1995  -7.7  7.4  -5.9  9  8.1  3.7  15.5  4.3  19.9  4.7  19.2  3.8  5.9  5.5  12.3  4.4  17.4  4  19.1  3.6  -7.8  1996  b)  8.4  -2.6  5  Pairs of months (Mean daily maximum temperature of the two combined months).  YEAR  FEB/MAR  S.D.  MAR/APR  S.D.  APR/MAY  S.D.  MAY/JUNE  S.D.  JUNE/JULY  S.D.  1989  -8.3  8.2  0.52  10.04  11.12  5.21  15.81  3.87  20.42  4.04  1990  -8.8  12.7  4.34  5.78  11.55  4.86  16.25  4.07  19.39  3.68  2.37  5.88  10.43  5.14  16.23  5.27  17.89  4.72  1991  -4.4  7.4  1992  -5.7  10.5  1.45  6.84  6.59  5.61  13.85  6.13  18.48  4.09  1993  -3.4  8.4  3.59  5.63  11.84  5.53  16.98  4.16  19.43  3.36  1994  -8.2  10.9  3.87  7.69  10.67  3.85  15.57  4.44  19.71  3.55  1995  -6.7  8.3  1  9.9  11.84  5.5  17.6  5  19.5  4.3  1996  -5.1  7.3  1.56  6.8  9.14  5.9  14.8  4.9  18.3  3.9  132  a)  §  -20  February  March  April  May  Months b)  February  April  May Months  Appendix 12 a-b. Mean daily maximum temperature and mean daily precipitation (1973 - 1996) from Burwash Landing airport, 40km northwest of the study area (unpubl. data). Mean (square) and standard deviation (whiskers).  Appendix 13. Number of days month when daily temperatures went above freezing. (1989 -1996) (Burwash Landing weather station unpub.data). Cool months in bold.  Year  February  March  April  1989  1  7  30  1990  3  19  30  1991  6  15  29  1992  6  20  21  1993  7  12  30  1994  0  18  30  1995  5  8  30  1996  7  13  27  Appendix 14. Mean daily precipitation at Burwash Landing airport, 40km northwest of the study area (unpub.data). Wet months in bold.  a) Individual months  FEBRUARY  S.D.  MARCH  S.D.  1989  0.1  0.4  0.6  1.5  1990  0.5  0.9  0.3  1.1  JUNE  S.D.  JULY  S.D.  0.9  1.8  3.6  0.3  0.4  2.1  2.6  2.1  0.9  2.4  0  0.3  1  2.4  6.6  2.7  4.2  S.D.  MAY  S.D.  0  0  0.3  0.6  1.9  0.9  APRIL  0.2  0.5  0.3  1.1  0  1992  0.1  0.3  0.1  0.2  0.6  1.5  0.9  1.5  0.7  1.6  3.6  7  1993  0.2  0.6  0  0  0.1  0.3  0.4  1.3  0.9  2.2  2.1  3  1994  0.1  0.4  0.3  0.8  0.3  1.1  0.3  1  1.4  4  0.9  1.9  1995  0.3  0.6  0.4  0.8  0  0  0.6  1.7  0.3  1.4  3.2  7.4  1996  0.3  0.6  0.2  0.5  0.4  1.9  0.9  2.6  0.3  0.8  3.3  6.4  JUNE/JULY  S.D.  1991  b) Mean daily precipitation of pairs of months.  1989  MAY/JUNE  S.D.  0.66  1.03  2.67  1  2.68  2.1  1.73  2.93  1.76  3.05  FEB/MAR  S.D.  MAR/APRIL  S.D.  APR/MAY  S.D.  0.38  1.22  0.3  1.16  0.17  1990  0.41  1.08  0.5  1.58  0.79  1991  0.29  1.08  0.17  0.92  0.14  0.74  1.31  4.77  2.57  5.46  1.16  0.77  1.6  0.81  1.55  2.2  5.26  1992  0.09  0.29  0.34  1993  0.11  0.48  0.04  0.22  0.24  1.03  0.55  1.32  1.57  2.53  1994  0.23  0.71  0.33  0.99  0.33  1.09  0.56  1.38  0.97  1.88  1995  0.35  0.8  0.18  0.6  0.32  1.3  0.25  1.4  2.1  5.9  1996  0.23  0.6  0.27  1.4  0.63  2.3  0.37  0.9  2.1  5  Appendix 15. Number of precipitation days at Burwash Landing airport, 40km northwest of the study (unpub.data). Peak rain days in bold.  a) Individual months  JUNE  JUL\  FEBRUARY  MARCH  APRIL  MAY  1989  4  7  0  6  13  6  1990  2  5  5  8  17  7  1991  7  5  0  3  10  19  1992  4  1  8  13  8  14  1993  3  3  2  6  10  20  1994  7  7  4  6  11  8  1995  8  8  0  7  1  15  1996  10  8  2  9  7  15  b) Paired months (Number of precipitation days in the two months combined).  1989  JUNE/JULY  FEB/MAR  MAR/APRIL  APR/MAY  MAY/JUNE  11  7  6  19  19  1990  7  10  13  25  24  1991  12  5  3  13  29  1992  5  9  21  21  22  1993  6  5  8  16  30  1994  14  11  10  17  19  1995  16  8  7  8  16  1996  18  10  11  16  22  Appendix 17. Relationship between rain days in July and the density index of spruce grouse in winter (observation per 100 field hours).  Appendix 18. Annual ranking of prey species contribution to the diets (biomass of prey in pellets 1 = lowest) in red-tailed hawks (1989 - 1995).  1989  1990  1991  1992  1993  1994  1995  TOTAL  Snowshoe hare  3  5  3  3  1  3  3  21  Arctic ground squirrel  5  4  5  5  5  4  5  33  Red squirrel  4  3  2  4  4  5  4  26  Mice/voles  2  1  1  1  3  1  1  10  Avain spp.  1  2  2  2  2  2  2  13  Year  139  Appendix 19. Annual ranking of prey species contribution to the diets (biomass of prey in pellets, 1 = lowest) of northern goshawks (1989 - 1996).  Year  1989  1990  1991  1992  1993  Snowshoe hare  Arctic ground squirrel  4  4  4  Red squirrel  Grouse  Other Avian spp.  1  1  1  3  3  1  1  1994  1995  1996  4  5  TOTAL  37  27  2  2  15  5  4  31  10  140  Appendix 20. Annual ranking of prey species contribution to the diets (biomass of prey in pellets, 1 = lowest) of great horned owls (1989 - 1995).  1989  1990  1991  1992  1993  1994  1995  TOTAL  Snowshoe hare  5  5  5  2  4  5  5  31  Arctic ground squirrel  3  4  4  4  3  3  Red squirrel  4  3  3  1  2  4  1  18  Mice/voles  2  2  2  5  5  2  4  22  Avian spp.  1  1  1  3  1  1  3  11  Year  1  22  141  

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.831.1-0099513/manifest

Comment

Related Items