NEST-SITE SELECTION, ECTOPARASITES, AND MITIGATION TECHNIQUES: STUDIES OF BURROWING OWLS AND ARTIFICIAL BURROW SYSTEMS IN SOUTHWESTERN IDAHO

NEST-SITE SELECTION, ECTOPARASITES, AND MITIGATION TECHNIQUES: STUDIES OF BURROWING OWLS AND ARTIFICIAL BURROW SYSTEMS IN SOUTHWESTERN IDAHO by Brian Wade Smith A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Raptor Biology at Boise State University April, 1999

The thesis presented by Brian Wade Smith entitled Nest-site Selection, Ectoparasites, and Mitigation Techniques: Studies of Burrowing Owls and Artificial Burrow Systems in Southwestern Idaho is hereby approved: Advisor Date Committee Member Date Committee Member Date Graduate Dean Date ii

DEDICATION This thesis is dedicated to my loving parents, Richard and Brenda Smith, and to my loyally devoted wife, Rebecca. iii

ACKNOWLEDGMENTS First, I give thanks to all members of my family for their encouragement and support throughout my academic career at Boise State University. I especially thank my wife Rebecca for her patience with me and her assistance on my project. Second, I thank my friends and fellow graduate students for all of their insights and advice around the office, but especially for all the relaxing and recreational events outside the office. I am particularly grateful for my advisor, Dr. Jim Belthoff, who taught me not only the importance of experimental design and statistical power, but also the art of nymph fishing the Big Wood River. Likewise, I thank my committee members, Dr. Al Dufty and Dr. Steve Novak, for their helpful suggestions, critical reviews, and encouragement during this project. Additionally, Dr. Charles Baker provided sound advice, information about parasites, and fulfilling hunting expeditions. Lara Hannon, Ben Nelson, Hilary Smith, and Elizabeth Garcia are greatly appreciated for putting up with my strenuous demands as field technicians. Extreme gratitude is extended to David Anderson, Trent Brown, Ethan Ellsworth, Sean Finn, Brian Herting, Bob Lehman, Steve Lewis, David Lupien, Rebecca Smith, Dave Oleyar, Liz Vandernoot, and Kurt Zwolfer for helping me dig numerous holes in unforgiving soil for my artificial burrow experiments. I also thank Dr. Mark Fuller and Leslie Carpenter for their technical assistance and cooperation on this project. Identification of ectoparasites was provided by R.J. Adams, Dr. Dale H. Clayton (University of Utah), and Dr. Robert Lewis (Iowa State University, retired). In addition, iv

I thank Dr. Marc Bechard, Mike Kochert, Brit Peterson, and Laura Valutis for reporting sightings of burrowing owls. Dan Gossett, Stephanie Gossett, Elna Black, Tom Zariello, and Aimee Pope also provided assistance with this project. Access to owls nesting on private property was granted by the Hayes and Stewart families. Financial and logistical support for this study was provided through challenge cost share grants from the Bureau of Land Management to Dr. Jim Belthoff, by the Department of Biology and Raptor Research Center at Boise State University, and by the Snake River Field Station, Forest and Rangeland Ecosystem Science Center, U.S. Geological Survey, Boise, Idaho. Jim Clark, John Doremus, and John Sullivan facilitated our work in the Boise District and Snake River Birds of Prey National Conservation Area. v

TABLE OF CONTENTS DEDICATION....iii ACKNOWLEDGMENTS... iv LIST OF TABLES....viii LIST OF FIGURES..x GENERAL INTRODUCTION 1 Study Species: Western Burrowing Owls..1 Why Study Nest-site Selection and Ectoparasites?...3 Overview of Chapters One, Two, and Three.5 Literature Cited..7 CHAPTER 1: EFFECTS OF CHAMBER SIZE AND TUNNEL DIAMETER ON NEST-SITE SELECTION IN BURROWING OWLS...12 Abstract..12 Introduction....13 Overcrowded Hypothesis. 16 Study Area and Methods 18 Study Areas..18 Artificial Burrow Placement 19 Site Reuse 22 Locating and Capturing Burrowing Owls 22 Measuring and Marking Owls.23 Owl Monitoring...23 Data Analyses..24 Results 25 Artificial Burrow Experiments 25 Effects of Chamber Size..25 Effects of Tunnel Diameter.30 Reuse of Nest Sites..35 Discussion..42 Chamber Choice Experiment...42 Tunnel Choice Experiment..45 Reuse of Nest Sites..46 Conclusions 48 vi

Literature Cited..49 CHAPTER 2: ECTOPARASITES ON BURROWING OWLS IN SOUTHWESTERN IDAHO: ASSESSMENT OF POTENTIAL EFFECTS ON SITE REUSE AND JUVENILE GROWTH, HEALTH, AND SURVIVAL...54 Abstract..54 Introduction 55 Study Area and Methods 58 Study Areas......58 Locating and Capturing Burrowing Owls 59 Measuring and Marking Owls.60 Ectoparasite Identification...60 Levels of Infestation 61 Juvenile Growth and Body Condition.62 Owl Monitoring...63 Data Analyses..63 Results 64 Ectoparasite Identification...64 Levels of Infestation 66 Site Reuse 66 Juvenile Growth and Body Condition.71 Discussion..73 Ecology of Ectoparasites Collected.73 Levels of Infestation 78 Site Reuse 79 Juvenile Growth and Body Condition.80 Conclusions 82 Literature Cited...83 CHAPTER 3: BURROWING OWLS AND HUMAN DEVELOPMENT: RESULTS OF SHORT-DISTANCE NEST BURROW RELOCATIONS TO AVOID CONSTRUCTION IMPACTS...89 Abstract..89 Introduction 90 Study Area and Methods 92 Study Area...92 Nestling Data...93 Nest Relocation 94 Results 97 Nestling Data...97 Nest Relocation 97 Discussion..99 Literature Cited 102 GENERAL CONCLUSIONS..105 vii

LIST OF TABLES Table 1.1a Occupancy (1997) of artificial burrow clusters of three placed around 1995 or 1996 natural nest burrows.....26 1.1b Occupancy (1998) of artificial burrow clusters of three placed around natural burrows in which burrowing owls nested between 1995 1997....28 1.2a Occupancy (1997) of artificial burrow clusters of two placed in suitable burrowing owl habitat and patterns of tunnel-diameter selection for each occupied nest burrow 32 1.2b Occupancy (1998) of artificial burrow clusters of two placed in suitable burrowing owl habitat and patterns of tunnel-diameter selection for each occupied nest burrow 33 1.3 Patterns of site reuse and chamber choice within clusters of three artificial burrow systems that owls used for nesting in 1997 and 1998 38 1.4 Patterns of site reuse and tunnel-diameter choice within clusters of two artificial burrow systems that were used for nesting in 1997 and 1998 40 2.1 Species and demographics of ectoparasites collected from burrowing owls in southwestern Idaho during 1997 and 1998....65 2.2a Patterns of ectoparasite load and site reuse within clusters of two artificial burrow systems that burrowing owls used for nesting in 1997 and 1998 69 2.2b Patterns of ectoparasite load and site reuse within clusters of three artificial burrow systems that burrowing owls used for nesting in 1997 and 1998 70 viii

3.1 Nestling information, relocation measurements, and fate of each relocated nest.98 ix

LIST OF FIGURES Figure 1.1 Prediction matrix of research hypothesis...17 1.2 Configurations of artificial burrow systems...20 1.3 Mean clutch size and number of fledglings for nests in small, medium, and large artificial burrow chambers...31 1.4 Mean clutch size and number of fledglings for nests in 10-cm and 15-cm diameter tunnels...36 2.1 Mean ectoparasite levels in 1997 and 1998 by tunnel diameter and chamber size 67 2.2 Mean number of ectoparasites per brood counted from owls heads and observer s hand for different tunnel diameters and chamber sizes...68 2.2 Mean mass and length of tenth primary per brood as a function of ectoparasite load.72 2.3 Mean hematocrit levels of broods with experimentally removed or unaltered ectoparasite levels..74 2.4 Interaction diagrams for length of tenth primary and mass according to treatment time and level of treatment.75 3.1 Location of occupied artificial nest burrows and direction of their relocation within the study site..95 x

1 GENERAL INTRODUCTION This thesis consists of three chapters describing my investigations of nest-site selection, ectoparasites, and nest relocations of burrowing owls (Athene cunicularia) in southwestern Idaho. The purpose of my field research was to (1) determine how chamber size and tunnel diameter influenced selection of artificial burrow systems, (2) identify ectoparasite species of owls and examine their potential effects on nesting success, site reuse, growth, and survival, and (3) investigate the efficacy of short-distance relocations of occupied burrowing owl nests. Information contained in this thesis should be of particular interest to those investigating nest-site selection processes in cavity-nesting species and to those resource managers involved with the active management and conservation of burrowing owls. Study Species: Western Burrowing Owls Western burrowing owls (Athene cunicularia hypugaea) breed throughout the open and well-drained grasslands, deserts, steppes, prairies, and agricultural lands of western North America, ranging as far north as southern Canada from interior British Columbia east to south-central Manitoba, and as far south as central Mexico (Zarn 1974, Haug et al. 1993). The eastern limit of the breeding range lies roughly along a line from Manitoba to northwest Louisiana (Zarn 1974). Seven other subspecies occur in North and Central America, including the Caribbean Islands (Haug et al. 1993). Most of the

2 birds in North America are migratory or disperse widely, except for populations in southern California, Florida, and the southern plains where owls are partial migrants or sedentary (Coulombe 1971, Haug et al. 1993). Burrowing owls are usually present only during the breeding season (March-October) in Idaho, but their migratory routes and wintering areas remain unknown (King 1996). Burrowing owls typically nest in abandoned mammal burrows where they may lay up to 12 eggs (Haug et al. 1993). In southeastern Idaho, the mean clutch size was 9.9 eggs (N = 30 clutches; Olenick 1990). The mean number of young fledged per nest attempt ranges from 1.6 to 4.9 (Haug et al. 1993). I chose western burrowing owls as study subjects for several reasons. First, they are abundant and usually nest near other pairs (i.e., semi-colonial) in southwestern Idaho, which greatly reduced travel time between nest sites. Next, burrowing owls are experiencing population declines in other regions of North America (Zarn 1974, Collins and Landry 1977, Wedgwood 1978, Martell 1990, Haug et al. 1993, James and Espie 1997). In addition, burrowing owls are tolerant of human and vehicular disturbances at or near their nest sites (Plumpton and Lutz 1993, King 1996, pers. obs.) which was critical given the number of times I needed to visit each occupied nest. Because burrowing owls commonly use artificial burrows (Collins and Landry 1977, Olenick 1990, Botelho 1996), they are an attractive species with which to investigate nest-site selection hypotheses. The use of artificial burrow systems (ABSs) in my study was designed to enhance management techniques for the species while experimentally testing for chamber size and tunnel diameter preferences in burrowing owls. Finally, ectoparasites (e.g., fleas, lice) occur at high densities on some burrowing owl families (J.

3 Belthoff, pers. comm.) and may affect their productivity and behavior. Identifying the role of ectoparasites in burrowing owl survival and reproductive success will enhance management techniques and contribute information about the life-history of this species. In contrast, studying burrowing owls in the field challenged me on a daily basis; their fossorial nature required me to bury the ABSs in the desert soil and my research questions forced me to check occupied burrows on a regular basis. Many logistical problems inherent in studying fossorial organisms were removed with the placement of artificial burrows, but several others seemed to be created (e.g., cattle collapsed several plastic chambers and tunnels). Why Study Nest-site Selection and Ectoparasites? Nest-site selection can influence reproductive success, microclimate, predation, levels of parasitsm, and many other ecological factors (White and Kinney 1974, Ellis 1982, Birchard et al. 1984, Kern and van Riper 1984, Bekoff et al. 1987, Møller l990, van Riper et al. 1993, Zwartjes and Nordell 1998, Hooge et al. 1999). Therefore, if organisms select an inadequate site, reproductive costs may be expressed as a reduction in survival for parents and/or offspring, future fecundity costs for parents and/or offspring, and costs associated with the timing of reproduction (Nur 1988, Møller 1993, Christe et al. 1996a). Therefore, nest-site selection in cavity nesting species is likely influenced by factors that maximize their fitness and likelihood of survival. The use of ABSs allowed me to address specific nest-site selection hypotheses concerning a secondary cavity nesting species. Previous investigators have used ABSs to compensate for a lack of natural burrows (Collins and Landry 1977), or to study

4 burrowing owl breeding biology (Henny and Blus 1981, Olenick 1990, Botelho and Arrowood 1998), nestling growth rates (Landry 1979, Olenick 1990), and the effectiveness of owl relocation or transplant programs (Turner 1985, Harris and Feeney 1989, Martell 1990, Dyer 1991, Haug et al. 1992, Trulio 1995, Delevoryas 1997, Feeney 1997, Trulio 1997). However, there have been no systematic studies to determine which configurations of ABSs burrowing owls prefer. Thus, the limited success of many these studies may be attributable to unfavorable ABS chamber or tunnel dimensions. Ecologists and evolutionary biologists have paid increasing attention to parasitehost interactions, yet the impact of parasites on their host populations is poorly understood (Toft 1991). Recent studies have found that, in some bird species, ectoparasites reduce clutch size, nestling body mass, and the number of young at hatching and fledging (Møller 1990, 1993, Richner et al. 1993, Møller et al. 1994, Christe et al. 1996a, Dufva and Allander 1996, McFadzen and Marzluff 1996), increase feather preening activities (Clayton 1991), increase mortality (Møller 1990, 1993, Richner et al. 1993, McFadzen and Marzluff 1996), increase nestling begging rates (Christe et al. 1996a) and adult provisioning rates (Tripet and Richner 1997), reduce adult sleeping time (Christe et al. 1996b), and reduce nestling hematocrit levels (McFadzen and Marzluff 1996). In contrast, a few studies found no effect of ectoparasitism on reproductive success or survival in adults or nestlings (Roby et al. 1992, Young 1993, Gebhardt- Henrich et al. 1998). In southwestern Idaho, ectoparasites occur in various densities in burrowing owl nests (J. Belthoff, pers. comm., pers. obs.). It is unknown what, if any, effects these ectoparasites have on reproductive success, nestling growth rates, and survival in

5 burrowing owls. Also, very few published accounts concerning the diversity of ectoparasites that infest burrowing owls in Idaho exist. With access to numerous broods and adults (through ABS use), I examined the diversity and estimated abundance of ectoparasites on burrowing owls. Additionally, I performed an experiment to investigate potential effects of ectoparasites on nestling growth, survival, and hematocrit levels. Overview of Chapters One, Two, and Three In Chapter One I present the findings of an artificial burrow experiment that was designed to examine how chamber size and tunnel diameter affect nest-site selection in burrowing owls. I provided two types of artificial burrow clusters: one offered three chamber sizes but a standardized tunnel diameter, and the other offered two tunnel diameters with a standardized chamber size. Upon their return to the study areas, owls presumably selected nest sites based on preferred chamber or tunnel dimensions within the cluster they selected. I monitored reproductive measures (number of eggs, number of young fledged) as a function of chamber or tunnel dimension to examine potential effects of ABS configuration on reproductive success. Burrowing owls preferred the largest chamber size and smallest tunnel diameter available to them in the two different cluster types. Also, significantly more juveniles fledged from the largest chamber than from the second largest chamber in the clusters of three ABSs. This finding is important because most resource managers deploy chambers much smaller than the largest chamber in my experiment. In Chapter One, I discuss hypotheses that may explain the patterns of ABS choice observed in my experiments.

6 Chapter Two focuses on the diversity and abundance of ectoparasites of burrowing owls in southwestern Idaho and their potential effects on site reuse, and nestling growth, survival, and hematocrit levels. I identify the species of ectoparasites I collected from adult and juvenile owls, and describe their relative abundance on a per brood basis. Also, I discuss the results of an experiment I conducted that investigated the effects of ectoparasitism by comparing broods in which I experimentally reduced ectoparasite levels to broods in which natural levels of ectoparasites existed. To reduce ectoparasite levels, I dusted all individuals within a brood and their nests' substrate with an insecticide. I collected three species of flea, one species of lice, and one species of carnid fly from burrowing owls in my study areas. Ectoparasite levels were similar between years of my study and thus, ectoparasite levels had no influence on nest-site reuse between years. Levels of infestation also did not differ between two tunnel diameters or among three chamber sizes used by nesting owls. Ectoparasite level did not affect juvenile owl growth (mass and tenth primary length) or fledging success when analyzing brood averages. In my ectoparasite manipulation experiment, dusted and infested nests did not differ in rates of growth or hematocrit levels. Experimental constraints and the lack of statistical power in my design are discussed in detail. Also, I discuss the ecology and potential effects on burrowing owls of each ectoparasite species I collected. In Chapter Three I present the results of an experiment that examined the efficacy of short-distance relocations of occupied burrowing owl nests. Nests (N = 5) occurred in artificial burrow systems in a field that was being converted from grassland habitat to irrigated agriculture. Nests contained one to five juveniles that ranged from 27 45 d.

7 Nest chambers and tunnels were moved (range: 72.5 258 m) from original nest areas to selected locations in a buffer strip that surrounded the field. Juveniles were transported to the new sites, but adults were expected to move the short distances on their own. Overall, two families (40%) accepted their relocation sites, two families (40%) exhibited "site tenacity" towards their original nest areas, and one family (20%) dispersed from the immediate vicinity. I describe the characteristics of each successful and unsuccessful relocation. I also introduce hypotheses to potentially explain why certain relocations were or were not successful. I conclude this chapter with a discussion of the efficacy of active and passive relocation efforts, problems I encountered with my experiment, and suggestions to increase effectiveness of mitigation techniques similar to those I employed. Literature Cited Bekoff, M., A.C. Scott, and D.A. Conner. 1987. Nonrandom nest-site selection in evening grosbeaks. Condor 89:819-829. Birchard, G.F., D.L. Kilgore, and D.F. Boggs. 1984. Respiratory gas concentrations and temperatures within the burrows of three species of burrow-nesting birds. Wilson Bull. 96:451-456. Botelho, E.S. 1996. Behavioral ecology and parental care of breeding western burrowing owls (Athene cunicularia hypugaea) in southern New Mexico, U.S.A. Ph.D. Dissertation, New Mexico State Univ., Las Cruces. Botelho, E.S., and P.C. Arrowood. 1998. The effect of burrow site use on the reproductive success of a partially migratory population of western burrowing owls (Speotyto cunicularia hypugaea). J. Raptor Res. 32:233-240. Christe, P., H. Richner, and A. Oppliger. 1996a. Begging, food provisioning, and nestling competition in great tit broods infested with ectoparasites. Behav. Ecol. 7:127-131.

Christe, P., H. Richner, and A. Oppliger. 1996b. Of great tits and fleas: sleep baby sleep. Anim. Behav. 52:1087-1092. Clayton, D.H. 1991. Coevolution of avian grooming and ectoparasite avoidance. Pp. 258-289. In: J.E. Loye and M. Zuk, eds. Bird-parasite interactions: ecology, evolution and behavior. Oxford Univ. Press, New York. Collins, C.T., and R.E. Landry. 1977. Artificial nest burrows for burrowing owls. N. Amer. Bird Bander 2:151-154. Coulombe, H.N. 1971. Behavior and population ecology of the burrowing owl, Athene cunicularia, in the Imperial Valley of California. Condor 73:162-176. Delevoryas, P. 1997. Relocation of burrowing owls during courtship period. In: J.L. Lincer and K. Steenhof, eds. Proceedings of the first international burrowing owl symposium. J. Raptor Res. Report 9:138-144. Dufva, R., and K. Allander. 1996. Variable effects of the hen flea Ceratophyllus gallinae on the breeding success of the great tit Parus major in relation to weather conditions. Ibis 138:772-777. Dyer, O. 1991. Reintroduction of burrowing owls (Athene cunicularia) to the South Okanagan Valley, British Columbia (1983-1988). Provincial Mus. of Alberta Nat. Hist. Occasional Paper No. 15. Ellis, J.H. 1982. The thermal nest environment and parental behavior of a burrowing bird, the bank swallow. Condor 84:441-443. Feeney, L.R. 1997. Burrowing owl site tenacity associated with relocation efforts. In: J.L. Lincer and K. Steenhof, eds. Proceedings of the first international burrowing owl symposium. J. Raptor Res. Report 9:132-137. Gebhardt-Henrich, S.G., P. Heeb, H. Richner, and F. Tripet. 1998. Does loss of mass during breeding correlate with reproductive success? A study on blue tits Parus caeruleus. Ibis 140:210-213. Harris, R.D., and L. Feeney. 1990. Restoration of habitat for burrowing owls (Athene cunicularia). Pp. 251-259. In: H.G. Hughes and T.M. Bonnicksen, eds. Restoration 89: the new management challenge. Proceedings of the society for ecological restoration first annual meeting. The Univ. of Wisconsin Arboretum, Madison. 8

Haug, E.A., D. Hjertaas, S. Brechtel, K. De Smet, O. Dyer, G. Holroyd, P. James, and J. Schmutz. 1992. National recovery plan for the burrowing owl. A report prepared for the Committee for the Recovery of Nationally Endangered Wildlife. Canadian Wildlife Federation, Ottawa. Haug, E.A., B.A. Millsap, and M.S. Martell. 1993. Burrowing owl (Speotyto cunicularia). In: A. Poole and F. Gill, eds. The Birds of North America, No. 61. The Academy of Natural Sciences, Philadelphia; The American Ornithologists' Union, Washington, D.C. Henny, C.J., and L.J. Blus. 1981. Artificial burrows provide new insight into burrowing owl nesting biology. Raptor Res. 15:82-85. Hooge, P.N., M.T. Stanback, and W.D. Koenig. 1999. Nest-site selection in the acorn woodpecker. Auk 116:45-54. James, P.C., and R.H.M. Espie. 1997. Current status of the burrowing owl in North America: an agency survey. In: J.L. Lincer and K. Steenhof, eds. Proceedings of the first international burrowing owl symposium. J. Raptor Res. Report 9:3-5. Kern, M., and C. van Riper III. 1984. Altitudinal variations in nests of the Hawaiian honeycreeper Hemignathus virens. Condor 86:443-453. King, R.A. 1996. Post-fledging dispersal and behavioral ecology of burrowing owls in southwestern Idaho. Unpubl. M.S. Thesis, Boise State Univ., Idaho. Landry, R.E. 1979. Growth and development of the burrowing owl. Unpubl. M.S. Thesis, California State Univ., Long Beach. Martell, M.S. 1990. Reintroduction of burrowing owls into Minnesota: a feasibility study. Unpubl. M.S. Thesis, Univ. of Minnesota, Minneapolis. McFadzen, M.E., and J.M. Marzluff. 1996. Mortality of prairie falcons during the fledging-dependence period. Condor 98:791-800. Møller, A.P. 1990. Effects of parasitism by the haematophagous mite on reproduction in the barn swallow. Ecology 71:2345-2357. Møller, A.P. 1993. Ectoparasites increase the cost of reproduction in their hosts. J. Anim. Ecol. 62:309-322. Møller, A.P., F. de Lope, J. Moreno, G. Gonzalez, and J.J. Perez. 1994. Ectoparasites and host energetics: house martin bugs and house martin nestlings. Oecologia 98:263-268. 9

Nur, N. 1988. The cost of reproduction in birds: an examination of the evidence. Ardea 76:155-168. Olenick, B.E. 1990. Breeding biology of burrowing owls using artificial nest burrows in southeastern Idaho. Unpubl. M.S. Thesis. Idaho State Univ., Pocatello. Plumpton, D.L., and R.S. Lutz. 1993. Nesting habitat use by burrowing owls in Colorado. J. Raptor Res. 27:175-179. Richner, H., A. Oppliger, and P. Christe. 1993. Effect of an ectoparasite on reproduction in great tits. J. Anim. Ecol. 62:703-710. Roby, D.D., K.L. Brink, and K. Wittmann. 1992. Effects of bird blowfly parasitism on eastern bluebird and tree swallow nestlings. Wilson Bull. 104:630-643. Toft, C.A. 1991. Current theory of host-parasite interactions. Pp. 3-15. In: J.E. Loye and M. Zuk, eds. Bird-parasite interactions: ecology, evolution and behavior. Oxford Univ. Press, New York. Tripet, F., and H. Richner. 1997. Host responses to ectoparasites: food compensation by parent blue tits. Oikos 78:557-561. Trulio, L.A. 1995. Passive relocation: a method to preserve burrowing owls on disturbed sites. J. Field Ornithol. 66:99-106. Trulio, L.A. 1997. Strategies for protecting western burrowing owls (Speotyto cunicularia hypugaea) from human activities. Pp. 461-465. In: J.R. Duncan, D.H. Johnson and T.H. Nicholls, eds. Biology and conservation of owls of the northern hemisphere: proceedings of the second international owl symposium. U.S. Dept. of Agric. Gen. Tech. Rept. NC-190. Turner, J. 1985. The burrowing owl transplant program in the South Okanagan Valley of British Columbia. Unpubl. Rept. for JT Biotech Associates, Inc., Princeton, British Columbia. 63 pp. van Riper III, C., M.D. Kern, and M.K. Sogge. 1993. Changing nest placement of Hawaiian common amakihi during the breeding cycle. Wilson Bull. 105:436-447. Wedgwood, J.A. 1978. The status of the burrowing owl in Canada. A report prepared for the Committee on the Status of Endangered Wildlife in Canada. Canadian Wildlife Service, Ottawa. White, F.N., and J.L. Kinney. 1974. Avian incubation. Science 186:107-115. 10

Young, B.E. 1993. Effects of the parasitic botfly Philornis carinatus on nestling house wrens, Troglodytes aedon, in Costa Rica. Oecologia 93:256-262. Zarn, M. 1974. Burrowing owl, Athene cunicularia hypugaea. Report No. 11, Habitat management series for unique or endangered species. Bureau of Land Mgmt., Denver, Colorado. 25 pp. Zwartjes, P.W., and S.E. Nordell. 1998. Patterns of cavity-entrance orientation by gilded flickers (Colaptes chrysoides) in cardón cactus. Auk 115:119-126. 11

12 CHAPTER 1: EFFECTS OF CHAMBER SIZE AND TUNNEL DIAMETER ON NEST-SITE SELECTION IN BURROWING OWLS Abstract Using field experiments, I examined choice for artificial burrow configuration by nesting burrowing owls (Athene cunicularia) in southwestern Idaho during 1997 1998. To assess potential preference for chamber size, I placed clusters of three artificial burrows around natural nest sites used in 1995, 1996, and 1997. Each cluster contained three different chamber sizes (small, medium, and large), each with a standard tunnel diameter. To assess potential choice for tunnel diameter, I placed clusters of two artificial burrows in suitable burrowing owl habitat. Each of these clusters offered two tunnel diameters (10 and 15 cm), each with a standardized chamber. Overall, 77 clusters of three artificial burrows were available for use by nesting pairs of owls during the two years of this study. In 1997, nesting pairs of burrowing owls occupied 18 clusters; 16 pairs used large, one used medium, and one used small chambers. In 1998, twenty-seven clusters were occupied by nesting pairs, with 16 pairs using large, six using medium, and five using small chambers. The distribution of use in each year, as well as for both years combined, differs significantly from uniform and indicates that burrowing owls prefer to nest in artificial burrow systems with large nest chambers. Despite such preference,

13 there was no effect of chamber size on clutch size, but the number of fledglings per nest was significantly greater in large chambers when compared with medium chambers. In 1997 and 1998, seventy-two clusters of two were available. Overall, owls occupied 44 of these; 30 pairs used burrows with 10-cm diameters and 14 pairs used burrows with 15-cm diameter tunnels. This distribution of use also differs significantly from uniform and indicates that burrowing owls prefer to nest in burrows with small-diameter tunnels. Thus, my results indicate strongly that burrowing owls prefer to nest in artificial burrows with (1) large chambers and (2) small-diameter tunnels. Burrowing owls may select artificial burrows with these configurations to reduce negative effects of overcrowding, to gain the most stable microclimate for developing juveniles, or to deter large grounddwelling mammalian predators. Introduction Relatively few bird species nest underground, possibly because of the increased risk of nest failure. Underground nests are susceptible to flooding or to ground-dwelling predators (Coulombe 1971, Haug et al. 1993). On the other hand, the burrow environment is a climatically stable nesting environment that protects birds from temperature extremes (White et al. 1978, Ellis 1982). Despite such benefits, the availability of underground nest sites may be limited depending on whether a species digs its own burrow (bank swallow, Riparia riparia, Ellis 1982; Florida burrowing owl, Athene cunicularia floridana, Haug et al. 1993, Millsap and Bear 1997) or relies on other fossorial organisms or geological cavities for nest burrows (western burrowing owl, Athene cunicularia hypugaea, Haug et al. 1993). For those species that rely on pre-

existing burrows, nest-site selection depends on burrow availability in suitable habitat 14 (Dyer 1991, Botelho and Arrowood 1998), and selection influences the success and productivity of the nesting attempt (Coulombe 1971). Western burrowing owls have declined significantly throughout much of their range as a result of agricultural conversion, destruction of colonial rodents and other burrowing mammals, and urbanization (Zarn 1974, Dyer 1991, Haug and Didiuk 1993, Trulio 1995, James and Espie 1997). For this reason, they are listed as federally endangered in Canada and have special status (endangered, threatened, or species of special concern) in 11 central or western U.S. states (James and Espie 1997). To compensate for the lack of natural burrows or to reintroduce burrowing owls to former portions of their range, wildlife managers have deployed artificial burrow systems (ABSs). Collins and Landry (1977) developed the first documented ABS study in California. They used ABSs to eliminate the possibility of nest burrow collapse and to entice owls to nest in areas not previously used (because of a lack of natural burrows) and that were protected from disturbance. Since then, ABSs have been used to study burrowing owl breeding biology (Henny and Blus 1981, Olenick 1990, Botelho and Arrowood 1998), nestling growth rates (Landry 1979, Olenick 1990), and the effectiveness of owl relocation or transplant programs (Turner 1985, Harris and Feeney 1990, Martell 1990, Dyer 1991, Haug et al. 1992, Trulio 1995, Delevoryas 1997, Feeney 1997, Trulio 1997; also see Chapter Three). Burrows typically offer a micro-environment that is significantly different from ambient conditions (Ellis 1982, Birchard et al. 1984) in terms of temperature and gas concentrations (e.g., CO 2 and O 2 ). Because western burrowing owls rely on burrows dug

15 by fossorial mammals, the dimensions of the tunnel and nest chamber vary depending on the original excavator, which can affect the microclimate of the underground system (Ellis 1982, Birchard et al. 1984). Western burrowing owls, which are considered secondary cavity nesters because they usually do not dig their own burrows, commonly modify tunnels and probably chambers of natural burrows (Thomsen 1971, Zarn 1974). These modifications may enhance characteristics within burrow systems, but it remains unknown what attributes burrowing owls seek within their burrows to meet their preferences for a nest burrow. Understanding their requirements for nest and roost burrows may be an essential component in burrowing owl management by providing insight into ecological or physiological constraints placed on underground nesting species. While information on the above-ground features of burrowing owl nest sites is available in the literature (Rich 1986, Plumpton and Lutz 1993), there is virtually no information on below-ground features of burrows important to nesting owls (Haug et al. 1993). Furthermore, no previous ABS studies have compared the effects of chamber or tunnel configurations on reproductive success in burrowing owls and their use of ABSs. Collins and Landry (1977) speculate that actual ABS chamber dimensions are not critical as long as one right-angle turn occurs in the tunnel. Their design, a 30-cm x 30-cm x 20- cm wooden chamber with a 10-cm tunnel, has been used extensively (Landry 1979, Henny and Blus 1981, Harris and Feeney 1990, Martell 1990, Trulio 1995). Some investigators have used 19-L (5 gal) plastic buckets for the nest chamber (Dyer 1991, Botelho 1996, D.A. Beig unpubl. data). Only Dyer (1991) has used both wooden and plastic bucket ABSs in the same study, but only because the wooden chambers

deteriorated quickly. Therefore, despite the claim by Collins and Landry (1977), it 16 remains unknown if chamber dimension affects the selection of ABSs or the reproductive success of western burrowing owls using ABSs. The objective of my study was to examine the effect of three chamber types (small, medium, and large) and two tunnel diameters (small and large) on nest-site selection by burrowing owls in a field experiment. Also, I wanted to compare the reproductive success (i.e., clutch size and number of young fledged per nesting attempt) among owls nesting in the different ABS types. Given that burrowing owls produce such large clutches of eggs (range up to 12, Haug et al. 1993), it seems logical that available space in both chambers and tunnels might affect nest-site choice. Thus, I designed my experiments to test what I termed the Overcrowded Hypothesis for nest-site selection. Overcrowded Hypothesis Because burrowing owls commonly have clutch sizes reaching 12 eggs, they may prefer large chamber sizes and tunnel diameters to accommodate many nestlings (Fig. 1.1). During the nestling stage, young owls commonly emerge from the burrow in the mornings and evenings but avoid mid-day heat by staying within the burrow system (Coulombe 1971). Moreover, nestlings often return to the burrow once they acquire several prey items, making the entire burrow system an important location for digestion, rest, and protection from predators (Botelho 1996). Although fledging can occur at 30-45 d post-hatching (Haug et al. 1993), burrowing owl broods remain near and continue to inhabit natal and nearby satellite burrows until they disperse (Zarn 1974, King 1996). This hypothesis predicts that the long period of dependence on the natal burrow biases an adult pair s selection process towards a burrow system with a large chamber size and

A. 17 EFFECTS OF ABS CHAMBER SIZE Overcrowded Hypothesis Small and Medium Chambers Rarely selected for nesting Large Chamber Often selected for nesting B. EFFECTS OF ABS TUNNEL SIZE Overcrowded Hypothesis Small-diameter tunnel Rarely selected for nesting Large-diameter tunnel Often selected for nesting Figure 1.1. Prediction matrix of research hypothesis. Predictions of the Overcrowded Hypothesis for effects of (A) chamber size and (B) tunnel diameter on nest-site selection in artificial burrow systems.

tunnel diameter to accommodate activities of even the largest broods. Therefore, the 18 Overcrowded Hypothesis predicts that large ABS chambers and large-diameter tunnels will be selected more often than the two smaller ABS chambers or the smaller-diameter tunnel. Study Area and Methods Study Areas I examined the Overcrowded Hypothesis and nest-site selection by burrowing owls in two study areas in southwestern Idaho. The first was located in Ada County, approximately 3.2 km south of Kuna (43 25' N, 116 25' W) and 23 km north of the Snake River Canyon. Vegetation in this area is characterized by big sagebrush (Artemisia tridentata) shrubland and disturbed grasslands dominated by cheatgrass (Bromus tectorum) and tumble mustard (Sisymbrium altissimum). Surrounding areas contain irrigated agricultural fields (primarily alfalfa, mint, and sugar beets), scattered residential homes, and several large dairy farms. The topography is flat to rolling with elevations ranging from 841 m to 896 m. Rock outcrops and a few isolated buttes (e.g., Kuna Butte, elev. 986 m) exist in the region. Temperatures range from -20 to 45 C, and annual precipitation averages less than 20 cm (NOAA 1985). In this area there is a relatively high density of burrows excavated by American badgers (Taxidea taxus). Burrowing owls commonly use badger burrows for nesting and shelter throughout the breeding season, especially if the burrows are near agricultural fields (Gleason 1978, Leptich 1994, pers. obs.).

19 My second study area was in Elmore County, approximately 8 km north-northeast of Grand View (43 00' N, 116 00' W) and adjacent to State Highway 67. This area is a mosaic of irrigated agriculture and disturbed grasslands. Elevations range from 853 m to 922 m. The area contains very few homes, several paved and dirt roads, and an electrical substation. The Snake River is approximately 7 km south-southwest of this study area. Temperatures here range from -29 and 43 C, and precipitation averages 26 cm per year (NOAA 1985). Both study areas were once typical shrubsteppe communities dominated by large expanses of big sagebrush (Hironaka et al. 1983). Range fires and other disturbances have converted much of the shrublands to exotic annual communities dominated by cheatgrass and tumble mustard. In general, the Kuna Butte study area contains more native shrubland than the seemingly more disturbed Grand View study area (Belthoff and King 1997). Artificial Burrow Placement Before burrowing owls arrived from wintering areas, I deployed clusters of two and three artificial burrow systems in or around the two study areas. The clusters of three artificial burrows tested for chamber size preferences and encircled natural burrows that owls used for nesting in 1995, 1996, or 1997 (Belthoff and King 1997, Belthoff and Smith 1998; Fig. 1.2). Within clusters of three, each artificial burrow consisted of a 15- cm diameter tunnel made of flexible, perforated plastic pipe and a plastic nest chamber. Each cluster had chambers of three sizes: a 30-cm x 30-cm x 20-cm (17-L; 4.5-gal) plastic container, a 19-L (5-gal) bucket with a 30-cm diameter, and a 50-cm x 35-cm x 40-cm (68-L; 18-gal) plastic tub. These chamber dimensions fall within the range of

A. 20 Small 5 m Medium Natural Burrow Large 5 m 5 m B. Small-diameter tunnel Large-diameter tunnel 3 m Figure 1.2. (A) Configuration of chambers around natural burrows for the chamber choice experiment and (B) configuration of chambers for the tunnel diameter choice experiment (see text for explanations of both).

dimensions of natural burrows that owls have used for nesting (Haug et al. 1993). All 21 entrances within a cluster were equidistant (5 m) from and were oriented in the same direction as the historical nest burrow entrance, and tunnel entrances were 120 degrees apart from each other. I randomly assigned chamber size to locations in each cluster. All ABS tunnels were 2 m long with a 90-degree turn between the entrance and the ABS chamber. Each tunnel sloped downward (20-30 degrees) toward the chamber within the range typical of nest burrows within both study areas (Belthoff and King 1997). The tunnel inserted into the chamber on a level plane, and the top of each ABS chamber was at least 30 cm underground. To increase the probability of ABS use during my study, I (1) supplied each chamber with approximately 3 cm of fresh soil (obtained while digging holes for ABS placement) for nesting substrate, (2) placed a wooden perch in the center of the cluster (as in King 1996), (3) created dirt mounds at ABS entrances to mimic those of natural burrows, (4) scattered debris (i.e., manure, prey remains, pellets) from previous nests onto the entrance-mounds, and (5) used rocks to block entrances to the historical nest burrow and any other suitable burrow within a 10-m radius of the historical nest burrow to prevent their use. The rocks were removed after juveniles fledged so the previously blocked burrows could be used as refuge burrows (satellites) if desired. I placed clusters of two artificial burrows in suitable habitat that lacked nesting pairs of burrowing owls (rather than around historical nests as in the chamber choice experiment). Clusters of two were designed to test for preference of tunnel diameter (Fig. 1.2). One artificial burrow had a 15-cm diameter tunnel, the other had a 10-cm diameter tunnel, and chamber size was standardized (19-L plastic buckets for both chambers). I placed the two burrows adjacent to one another with 3 m between each entrance. I

22 oriented tunnel entrances in a south-southeast direction and a north-norteast direction in the Kuna Butte and Grand View study areas, respectively, which is typical for orientation of natural nest sites in these areas (Belthoff and King 1997). I placed a wooden perch between the tunnel entrances in all clusters of two and used the same tunnel lengths and slopes, chamber depths, and general methods as those for clusters of three. Site Reuse If owls used a cluster of three for nesting in 1997, I rearranged the position of chambers within the cluster before owls returned in 1998. In doing so, I stipulated that the chamber used for nesting in 1997 could not be placed in the same location as the previous year. Doing this allowed me to make certain that owls chose ABSs based on chamber size rather than a specific site within a cluster. In addition to the rearrangement, I cleaned all chambers and tunnels of any material that remained from 1997 and supplied them with fresh soil as nesting substrate. If owls used a cluster of two for nesting in 1997, I switched the location of the ABSs within the cluster. This also allowed me to determine if owls chose an ABS based on tunnel diameter or a specific site within a cluster. Again, I supplied each tunnel and chamber with fresh soil after removing all materials that remained from 1997. Locating and Capturing Burrowing Owls To locate nesting owls, I revisted historical nest sites (surrounded by ABSs in 1997) and searched suitable habitat in both study areas on foot and from automobiles. I performed most surveys during daylight hours. After locating owls, I regularly monitored their nesting activities. In 1998 I placed new clusters of three ABSs around nests found at natural burrows in 1997.

23 To capture owls I used Havahart traps and noose rods as described in Belthoff et al. (1995) and King (1996). I also used one-way basket traps to capture adults as they departed artificial burrows. These traps consisted of a 0.5-m section of flexible plastic pipe (10-cm diameter), a small piece of transparent Plexiglas, and an enclosure made of chicken wire with 2.5-cm diameter openings. The Plexiglas was fastened to one end of the pipe by a hinge from above which allowed the door to swing upwards when an owl attempted to exit the pipe. This end of the pipe was inserted into the wire basket. The open end of the pipe was inserted into artificial burrow tunnels when I needed to determine the status of a nest in an ABS. Digging down to the artificial chamber caused any adults in the nest chamber to enter the basket and become trapped when the hinged door closed behind it. Measuring and Marking Owls Upon capture, I recorded each owl's mass (to the nearest gram), wing length, tarsus length, tail length, and length of exposed culmen (all to the nearest 0.5 mm). I classified adult owls as female if they had a well-developed brood patch. I classified the remainder of adults as males based on their lack of a brood patch, relatively lighter plumage, and behavior near nests (Thomsen 1971, Zarn 1974). I could not discern sex of young owls based on appearance or morphological measurements. I fitted all owls with a United States Geological Survey aluminum leg band (size 4) and three plastic, colored leg bands (National Band and Tag Co., Newport, KY) for future identification. Owl Monitoring I conducted follow-up visits at nests in artificial burrows to determine if they were successful. For owls nesting in ABSs, I determined minimum number of eggs produced,

nestling survival, and the number of young fledged. Successful nests in ABSs had at 24 least one young owl survive to fledging age (> 28 d). Data Analyses I performed chi-squared goodness of fit tests to examine choice of nest chamber size or tunnel diameter by nesting owls. I used a similar test to examine site reoccupancy within an ABS cluster. Using analysis of variance (ANOVA), I also examined effects of chamber size and tunnel diameter on number of eggs and fledglings. When the ANOVA indicated significant differences existed, I performed pair-wise mean comparisons using the least significant difference (LSD) procedure. Because some clusters in both experiments suffered damage which rendered them unavailable to owls choosing sites within a cluster, I eliminated them from analyses of choice (chi-squared analyses) but included them in analyses of productivity (i.e., mean number of eggs and fledglings) by chamber or tunnel dimension. Additionally, I included one cluster of three in the analysis of choice but eliminated it from analyses of productivity because I accidentally disturbed the nest when the eggs were hatching, and only three of nine apparently viable eggs hatched. In 1998, six owl pairs attempted a second nest after their first attempt failed. I used first nesting attempts in all analyses and used the second attempts (1) in choice analyses only if the pair re-nested at a new cluster and (2) in all analyses of productivity. Because two dependent variables (eggs, young) were collected from the same experimental units and analyzed, I used Bonferroni adjusted P-values (Zar 1996). Results were considered significant when P < 0.025 for productivity analyses. Otherwise, results were considered significant when P < 0.05. All means are given as x ± SE.

25 RESULTS Artificial Burrow Experiments Effects of Chamber Size I assessed preference for nest chamber size using clusters of three artificial burrows placed around historical nesting sites. In 1997, I placed 34 clusters of three (N = 17 in each study area), of which nesting owls used 21 (62%; Table 1.1a). I removed from the experiment three clusters in which owls nested; cattle trampled two clusters prior to nest initiation and one cluster had been placed on an unacceptably steep slope according to the cluster deployment protocol described earlier. Removing these three clusters from the analysis does not change the conclusion of the chamber choice experiment. Of the remaining 31 clusters, nesting owls occupied 18 (58.1%). Of these 18 pairs of owls, 16 used the large chamber, one pair used the medium chamber, and one pair used the small chamber. This distribution of chamber use differs significantly from uniform (χ 2 = 25.0, 2 df, P < 0.001). In 1998, I deployed 15 additional clusters of three artificial burrows (N = 5 in Kuna Butte and N = 10 in Grand View) but had to remove three of the original 34 clusters because of land development (Table 1.1b). This resulted in 46 clusters of three available in 1998. Of these, burrowing owls nested in 28 (61%; Table 1.1b). I removed from the experiment two clusters in which owls nested because cattle damaged them. Nesting owls occupied 26 (59.1%) of the remaining 44 clusters. These 26 pairs nested in large (N = 15), medium (N = 6), and small (N = 5) chambers. This distribution of chamber selection also differs significantly from uniform (χ 2 = 7.0, 2 df, P < 0.03).

Table 1.1a. Occupancy (1997) of artificial burrow clusters of three placed around 1995 or 1996 natural nest burrows. Patterns of chamber size selection, clutch size, and number of fledglings are indicated for each cluster. An asterisk (*) indicates the minimum number of eggs as some nests were not checked until hatching began. 26 Nest Name Occupied? Chamber # Eggs Laid Size Used (# Fledged a ) Kuna Butte Study Area Sewage Pond #1 Yes Large 11* (8) Sewage Pond #2 Yes Medium 9* (3) b Kuna Butte #1b Yes Large 8 (3) Kuna Butte #5 No - - Kuna Butte #7 No - - Effluent Field North #1 Yes Large 9 (8) Effluent Field South #1 Yes Large 9 (8) Kuna Cave #2 No - - Kuna Cave #3 No - - Swan Falls #3 No - - Swan Falls #4 Yes Large 11 (10) Prison #1 Yes Large 7 (7) J. Hayes #1 No - - J. Hayes #2 No - - J. Hayes #3 No - - B. Stewart #1 Yes Medium c 9 (8) B. Stewart #2 Yes Small d 8 (4) Grand View Study Area Trailer #1 Yes Large 10 (5) Trailer #2 Yes Large 10* (5) Trailer #4 Yes Large 12* (8) Well #1 No - - Well #2 Yes Large 11 (5) Baha #1 Yes Small 9* (4) Substation East #1 Yes Large 8* (8) Trailer View #1 Yes Medium 8* (7) Grand View #2 No - - Grand View #3 No - - Grand View #19 Yes Large 10 (3) Substation South #1 Yes Large 10 (7) Substation South #2 Yes Large 11* (6) Substation South #4 No - -

Table 1.1a. Continued. 27 Nest Name Occupied? Chamber # Eggs Laid Size Used (# Fledged) Substation South #5 Yes Large 12* (6) Substation South #6 Yes Large 9 (4) Substation South #7 No - - a juvenile burrowing owls considered fledged at 28 d old. b six eggs did not hatch. c cattle trampled large chamber, causing it to collapse prior to nest-site selection. d cattle trampled large chamber and tunnel to medium chamber, causing them to collapse prior to nest-site selection.