, , Raptor Res. 35(4):288-295 J. © 2001 The Raptor Research Foundation, Inc. SYNCHRONOUS AND DELAYED NUMERICAL RESPONSES OF A PREDATORY BIRD COMMUNITY TO A VOLE OUTBREAK ON THE CANADIAN PRAIRIES Ray G. Poulin^ Biology Department, University ofRegina, Regina, SK S4S 0A2 Canada Troy L Wellicome^ Department ofBiological Sciences, University ofAlberta, Edmonton, AB T6G 2E9 Canada L. Danielle Todd Biology Department, University ofRegina, Regina, SK S4S 0A2 Canada — Abstract. In 1997, meadow vole {Microtuspennsylvanicus) populations reached abnormally high levels in the grasslands of Saskatchewan. From 1996-98 on the Regina Plain, we studied the numerical responses of eight predatory birds to the meadow vole outbreak. Populations of Loggerhead Shrikes {iMnius ludov- icianus) and American Kestrels {Falco sparverius) were unaffected by the high-vole year, but six other species exhibited significant numerical responses. Populations of Short-eared Owls {Asioflammeus) and Ferrugi- nous Hawks {Buteo regalis) changed in synchrony with the availability of small mammals. Short-eared Owls were apparently nomadic, as they were common on our study area during the vole high, but were not observed the year before or the year after. In contrast, the Burrowing Owl {Athene cunicularia) population reached a historical low the year that voles were most abundant, but increased substantiallyin the following year. This was the only annual population increase observed for Burrowing Owls in our study area for at least a decade. Sightings ofRed-tailed Hawks {B. jamaicensis) Swainson’s Hawks {B. swainsoni), and North- ern Harriers Circus cyaneus) increased in the year of the vole outbreak and remained at elevated levels in ( the following year Immature buteos were seldom seen before or after the vole peak, but during the vole peak, immatures were common, roosting together in large groups in fields. Key Words: raptors', hawks', owls'. Burrowing Oxvl; Athene cunicularia; Loggerhead Shrike', meadow vole', nu- merical response.', grassland. Respuetas numericas subcronicas y retardadas de una comunidad depredadora de aves a una erupcion de ratones de campo en las praderas canadienses — Resumen. En 1997, las poblaciones del raton de pradera {Microtus pennsylvanicus) alcanzaron niveles anormalmente altos en los pastizales de Saskatchewan. Desdc 1996-98 en la Llanura de Regina, estu- diamos las respuestas numericas de ocho aves depredadoras a la proliferacion de ratones de la pradera. Las poblaciones de Alcaudon Tonto {Lanius ludovicianus) y Cernicalos {Falco sparverius) no fueron afec- tadas por el aho de alta abundancia de ratones, pcro otras seis especies exhibieron respuestas numericas significativas. Las poblaciones del Buho de Orejas (iortas {Asioflamm.eus) y del Gavilan ferruginoso {Buteo regalis) cambiaron sincronicamente con la disponibilidad dc los pequehos mamiferos. Los Buhos de Orejas Corta.s aparentemente fueron nomadas, tanto asi quc fueron comunes en nnestra area de estudio durante la gran abundancia, pero no fueron obsei vados el aho anterior o al aho siguiente. En contraste, la poblacion del Biiho Cavador {Athene cunicularia) alcanzo una baja historica el aho en que los ratones fueron mas abundantes, pero aumentaron sustancialmente al aho siguiente. Este fue el unico incre- mento poblacional anual observado en los Buhos Cavadores en nuestra area de estudio por lo menos en una decada. Los avistamientos de Gavilanes de Gola Rqja {B. jamaicensis) Gavilancs de Swainson {B. swainsoni) y Aguiluchos Nortehos {Circus cyaneus) aumentaron en el aho de explosion de ratones y permanecieron en niveles elevados al aho siguiente. Los buteos inmaduros fueron vistos rara vez antes o despues del pico de abundancia de ratones, pero durante el pico, los inmaduros fueron comunes, perchando juntos en grandes grupos en los campos. [Traduccion de Victor Vanegas Cesar Marquez] y E-mail address; [email protected] ‘ Present address: Canadian Wildlife Service, 4999-98 Ave., Edmonton, AB, T2B 2X3, Canada. 288 , , , December 2001 Numerical Responses of Vole Preuafors 289 Table 1. Summary of search effort employed to survey birds of prey on the Regina Plain, Saskatchewan. Each day that a party recorded raptor sightings from a truck is considered one ‘search day.’ Thus, two trucks covering different routes on the same day were counted as two search days. No. OF Total No. Mean No. Totai. Mean Disiance Search OF Hours OF Hours per Distance per Search Day Ranc.e of Dates Days Searched Search Day Searched (km) km) ( 19 Apr-13 Aug 1996 152 1101 7.2 ± 0.14 41873 275 ± 7.3 21 Apr-19 Aug 1997 158 1028 6.5 ± 0.21 32 397 205 ±7.2 14 Apr-20 Aug 1998 146 1005 6.9 ± 0.15 31 644 216 ± 5.7 Breeding densities of many raptor species vary among species because of differences in size, be- from year to year because of annual fluctuations in havior, and survey techniques (Millsap and Le- their prey (Newton 1976). Such numerical re- Franc 1988), we made no attempt to compare pop- sponses of predators can be either in synchrony ulation indices among species. (S) with their prey, showing no obvious time lags, Four of our eight study species, including the or delayed (D) by one or more years (Galushin endangered Burrowing Owl {Athene cuniculana, 1974). Numerous examples of synchronous and Wellicome and Haug 1995), are designated as a delayed numerical responses to prey have been re- Species at Risk in Canada (Rothfels et al. 1999). corded for raptor species in a variety of habitats; Hopefully, an improved understanding of factors m tundra, Rough-legged Hawk (Buteo lagopus) (S) influencing populations of these species will also (e.g., Virkkola 1992), Gyrfalcon {Falco rusticolus) aid in their conservation. (D) (Niefsen 1999), Short-eared Owl {Asio flam- Methods meus) (S) (Andersson 1981), and Snowy Owl {Nyc- tea scandiaca) (S) (Wiklund and Stigh 1986); in bo- Study Area. This study was conducted in the grassland real, Northern Harrier {Circus cyaneus) (S) ecoregion of Saskatchewan (Harris et al. 1983), in an area roughly bounded by the cities of Regina (50°25'N, (Hamerstrom 1979), Northern Hawk-Owl {Surnia 104°39'W), Moose Jaw (50°23’N, 105°32'W), and Wey- ulula) (S) (Rohner et al. 1995), Boreal Owl (Aego- burn (49°4TN, 103°52'W). The study site encompassed lius funereus) (S or D) (Korpimaki 1992), Great 12 000 km^ of predominantly cultivated land. Over 90% Horned Owl {Bubo virginianus) (D) (Rohner of the original grassland in the area has been converted 1996), and Northern Goshawk {Accipiter gentilis) to cropland (James et al. 1990). European settlement and farming in the region has resulted in roads, usually (S) (Doyle and Smith 2001); in semidesert, Harris’ spaced by 3.2-6.4 km, running east-west and north-south Hawk {Parabuteo unicinctus) (S), Red-backed Hawk in a grid across the study area. {Buteo polyosoma) (S), and Black-chested Eagle {Ger- Raptor Survey. To estimate the relative abundance of anoaetus melanoleucus) (S) (Jaksic et al. 1992); and birds of prey (other than Burrowing Owls) in the study area, we counted Short-eared Owls, Northern Harriers, in the tropics, Barn Owl {Tyto alba) (S) (Wilson et American Kestrels {Falco sparverius) Loggerhead Shrikes al. 1986). {Lanius ludovicianus) Ferruginous Hawks {Buteo regalis), Few studies have attempted to relate numerical Red-tailed Hawks {B. jamaicensis) and Swainson’s Hawks changes of breeding raptors to annual food varia- {B. swainsoni) that we observed while driving in the study area each day. We also counted Prairie Falcons {Falco mex- tion in grassland habitats (Schmutz and Hungle icanus) Merlins {F. columbarius) Great Horned Owls, and 1989, Steenhof et al. 1997). In the present study, Golden, Eagles {Aquila chrysaelos,) but they were too rare , we recorded patterns of inter-annual variation in for analysis. Field vehicles contained tally sheets on which population indices of eight species of predatory investigators recorded observations of predatory birds, birds on the Regina Plain in southern Saskatche- along with the number of km driven and hours worked wan, Canada. We also estimated small mammal each day (Hochachka et al. 2000). Young-of-the-yearwere not included for any species, and migrating individuals availability before, during, and after a meadowvole were excluded by the dates ofour surveys (Table 1). Also, {Microtus pennsylvanicus) peak in the study area. raptor species that migrate through, but do not breed in Our effort and methods for obtaining population our study area (e.g.. Rough-legged Hawk, Snowy Owl, and Gyrfalcon) were excluded from our analysis. Distant indices remained con.sistent within each species , buteos that could not be identihed to species were re- among years. However, given that the proportion corded as ‘unknown buteos.’ We did not record the sex of the population detected undoubtedly varied or breeding status of birds, so non-breeding adults (e.g.. 290 Biology VoL. 35, No. 4 second-year Red-tailed Hawks) were included in the over- 40 all abundance indices. Individuals may sometimes have (A £ been counted more than once per day because the same — deef area was occasionally driven more than once in a day. 30 However, such errors were likely consistent among years, >£« :ltur .-.ifletsfiV.- so our method provided useful indices for comparing 0. among-year population changes within species. •5 20 Our work was conducted during daylight hours, typi- H cally between 0900 and 1800 H. Total search effort was similar among years (Table 1). To account for any varia- cCO 10 tion in search effort, however, estimates of bird abun- 4> dances were expressed as the mean number of individ- 0 uals observed per 100 km traveled per census day. . 1992 1993 1994 1995 1998 Among-year variation in these population indices was as- sessed for each species using one-way analysis ofvariance Figure 1. Mean number of vertebrates cached by Bur- and significant differences were identified with post-hoc rowing Owl pairs in each of 7 yr. Values were calculated Tukey tests. All analyses were conducted with an alpha for each pair by averaging the number of prey counted value of 0.05. in nest stores during checks in both the pre-laying and Burrowing Owl Census. Burrowing Owls were rarer than the other birds ofprey in our study area, so we used laying periods. Bars indicate annual means of all verte- a more intensive method to estimate changes in their brate prey cached and error bars show SE. ‘Other verte- population. Beginning in the second or third week of brate prey’ include sagebrush voles, house mice, shrews, April (1996—98), all sites known to have Burrowing Owls passerines, and tiger salamanders. Prey-cache data were in the previous 5 yr (James et al. 1997, Wellicome et al. collected from 13, 24, 16, 26, 17, 18, and 17 pairs in 1997) were searched for signs of occupancy. We slowly 1992—98, respectively (adapted from Wellicome 2000). drove or walked transects, spaced at ca. 25 m, through suitable nesting areas (i.e., non-cultivated fields), scan- ning each Richardson’s ground squirrel {Spermophilus ri- ative prey abundance in our study area (measured by chardsonii) or badger {Taxidea taxus) burrow for signs of small mammal trapping over a 4-yr period; Wellicome owls, owl pellets, or whitewash. In addition to our search- 2000). To determine cache sizes, we opened all Burrow- es, the Operation Burrowing Owl program in Saskatch- ing Owl nests that were in artificial nest boxes. Artificial ewan had a toll-free telephone number that other biol- nest boxes allowed us to access nest chambers to count ogists and members of the general public were and mark all stored prey items without disturbing the encouraged to use to report Burrowing <^1 sightings physical structure ofthe nest (Wellicome et al. 1997, Wel- (Skeel et al. 2001). We investigated each reported sight- licome 2000). For analyses, we used prey-cache data col- ing within our study area. lected up until 2 wk after the first egg was laid in each We are confident that our census was accurate because nest because this laying period had the highest rate of nesting pastures were small and work on Burrowing Owls prey caching within each season (Wellicome 2000). had been ongoing since 1987 in the area (James et al. Cache size was measured as the mean number of prey 1997, Wellicome et al. 1997). However, as an accuracy items found in each nest, provided the nest had been check, we randomly chose five ofthe 28 townships (each visited at least twice during pre-laying and laying. The 9.6 km X 9.6 km) that contained owls in 1995, and mean of all nests was then calculated to obtain an index searched all grassland fragments and roadside ditches ofrelative abundance ofsmall mammals in the studyarea (regardless ofwhether the sites had any previous records for each year between 1992-98. of owls) within this subset of townships in 1996. We Results searched all suitable habitat within these townships by driving slowly along ditches and walking transects in pas- Almost all cached vertebrate prey were either tures and other grasslands. No new owls were found us- deer mice {Peromyscus maniculatus) or meadow ing this intensive setirch, suggesting that the estimated population size obtained from our usual census tech- voles. Other vertebrate prey included sagebrush nique was close to 100% of the actual population (see voles (Lemmiscus curtatus), house mice {Mus mus- Wellicome et al. 1997 for details). culus), shrews {Sorex spp.), passerines, and tiger sal- Small Mammals. Burrowing Owls are generalist hunt- amanders {Ambystoma tigrinum). Excluding 1997, a ers that capture prey species in the same proportions as mean of 70% (range = 40-87%) of the vertebrate are available in the environment (Green et al. 1993, Plumpton and Lutz 1993, Silva et al. 1995, but seeJaksic prey items found in Burrowing Owl nests were et al. 1992) On the Regina Plain study area, vertebrate deer mice, and only 19% (range = 7-32%) ofprey . prey made up between 85-97% oftotal preyvolume mea- items were meadow voles (Fig. 1). In 1997, general sured in food pellets annually during the breeding sea- field observations and snap-trapping data (Welli- son (Wellicome 2000). We obtained an index of small mammal availability by counting the number of small come 2000, Sissons et al. 2001) suggested that the mammals cached inside Burrowing Owl nests, as average small mammal population reached extremely high annual cache size has been shown to reflect annual rel- levels. Such a high abundance of small mammals December 2001 Numerical Responses of Vole Predators 291 100 n 1.7, df = 2, N = 453, P — 0.17) showed significant Burrowing Owls population responses to the 1997 vole increase. 75- The mean number of individuals (±SE) observed S£ per 100 km per search-day from 1996-98, were 0.17 ± 0.03, 0.16 ± 0.03, and 0.18 ± 0.03 for shrikes, and 0.15 ± 0.03, 0.09 ± 0.03, and 0.17 ± 0.04 for kestrels, respectively. There was a highly-significant difference in the number of Short-eared Owls observed amongyears = = V = P < (F 68.4, df 2, 453, 0.001; Fig. 2). In both 1996 and 1998 (years with normal small mam- 1996 1997 1998 4-, mal populations) we observed this species on only , two occasions; whereas, in 1997 (the peak vole year), we recorded a total of 604 observations of this species. There was significant annual variation in the number of Swainson’s Hawks (F — 56.92, df — 2, N = 453, P < 0.001), Red-tailed Hawks (F - 29.1, df = 2, N = 453, P < 0.001), Ferruginous Hawks (F = 13.7, df = 2, V = 453, P < 0.001), and total buteos (including unknown; F = 49.897, df = 2, N — 453, P < 0.000) observed per 100 km per day. 1996 1997 1998 For each of the three species, there were signifi- Year cantly more observations during the year of the Figure 2. Total number of Burrowing Owl pairs on the vole outbreak (Fig. 3). In the year following the study area, and Short-eared Owl population index vole outbreak, when prey numbers returned to a [mean No./lOO km/search day] ± SE) in three separate normal level, both Swainson’s and Red-tailed Hawk ( years. The Burrowing Owl population had a delayed re- populations remained significantly higher than sponse to the high abundance ofsmall mammals in 1997, they were in the year prior to the vole outbreak. but the Short-eared Owl population had a synchronous However, observations of Ferruginous Hawks did response. For Short-eared Owls, results from Tukey tests not remain elevated in the year following the vole are shown with letters above each bar; differing letters peak. For all buteos combined (unidentified, Red- indicate that among-year differences in observation rates tailed, Swainson’s, and Ferruginous Hawks), there were significant. were eight times more observations in 1997 (3413), and four times more observations in 1998 (1667), than there were in 1996 (416). Although was evidently a rare occurrence, as populations had the numbers of adult vs. immature birds were not not been this plentiful since 1969 (Houston 1997). In 1997, 87% of cached prey items were meadow recorded, there was an obvious increase in the fre- quency of immature buteos in 1997. Most obser- voles, making the mean total number ofprey items vations of buteos in 1997 were of 1-yr-old Swain- per nest between three and 16 times higher in son’s and Red-tailed Hawks, but in 1996 and 1998 1997 than in the other years (Fig. 1). almost all observations were of adult, breeding Data have been collected on the population size birds (pers. observ.). of Burrowing Owls in a portion of our study area There was significant annual variation in the since 1987 and there was a decline in every year number of Northern Harriers observed per 100 except 1998 (James et al. 1997, Wellicome et al. km = = P < per day (F 51.3, df 2453, 0.001; Fig. 1997). The only recorded increase in the number 3). Harriers were much more abundant in 1997 of Burrowing Owls was between 1997-98 (Fig. 2) than in 1996, but did not decrease in 1998 to 1996 and the fewest Burrowing Owls in the past decade levels. occurred in 1997, the year of the meadow vole out- break. Discussion Neither Loggerhead Shrikes (F = 0.08, df = 2, Two of the eight avian predators in this study N — 453, P — 0.93) nor American Kestrels (F = showed no significant numerical response to the 292 Biology VoL. 35, No. 4 Red-tailed likely for shrikes, as they construct their own stick Hawk nests in shrubs or small trees (Yosef 1996). The remaining six avian predators in this grass- land study showed significant numerical responses to the vole high in 1997. In general, local increases in bird populations in response to elevated prey numbers can result from increased reproductive output in situ and/or immigration from peripheral populations (Solomon 1949). The numerical re- sponse of the Short-eared Owl to prey can un- 1997 1998 doubtedly be attributed to immigration because Short-eared Owls were rare on the study area in 1996, but suddenly became very common in 1997 Northern with the increase in voles. Such synchronous re- Harrier sponses are characteristic of species with nomadic lifestyles (Galushin 1974). In concordance with our results in the grasslands, the Short-eared Owl has been described as nomadic also in boreal (Korpi- maki and Norrdahl 1991) and tundra habitats (An- dersson 1981). The species’ specialized diet, simple nest-site requirements, and large clutch size seem 1996 1997 1998 to make it particularly well suited to a lifestyle of nomadism (Holt and Leasure 1993). Year Although the Burrowing Owl shares some of Figure 3. Population indices ([mean No./100 km/ these general characteristics with the Short-eared search day] ± SE) for Swainson’s Hawks, Red-tailed Owl, it exhibited an opposite response to the prey Hawks, Ferruginous Hawks, and Northern Harriers. All high. Burrowing Owls in our study were at their four species exhibited a synchronous response to the lowest during the vole peak but increased in the high abundance of small mammals in 1997 and all spe- subsequent year (Fig. 2) The 1-yr delay in the pop- cies except tbe Ferruginous Hawk were higher in 1998 . ulation’s response to the vole outbreak suggests than in 1996. Results from Tukey tests are shown with that these owls are not nomadic, as they do not letters above each bar; differing letters indicate that among-year differences in observation rates were signifi- search actively for nesting sites based on the cur- cant. Note that patterns of change in population indices rent availability of prey in an area, at least not at a among years can be compared, but that indices should large geographic scale. Given that the species is not not be compared among species, as detectability likely nomadic then, other mechanisms must explain its differed among species. observed numerical response to the prey high. Al- though clutch size was no higher for Burrowing Owls during the vole high in 1997, both nestling survival and fledging success were substantially el- small mammal high in 1997; populations of Log- evated in that year compared to other years (Wel- gerhead Shrikes and American Kestrels remained licomc 2000). In addition, post-fledging survival stable over all three survey years. This lack of re- was significantly higher in 1997 than it was in years sponse might have been an artifact of the relatively following (Todd 2001) or preceding the vole high small populations of these two species in our study (Glayton 1997). Furthermore, the percent offledg- area. Alternatively, it is possible that these two spe- lings from 1997 that returned to breed in the pop- cies relied heavily on prey items other than small ulation in 1998 was twice that ofreturns from other mammals. For example, they may have fed pre- years (R. Poulin, T. Wellicome, and L. Todd un- dominantly on insects. Another possibility is that, publ. data). These factors, alone or in combina- rather than food supply, availability of nesting sites tion, seem to have contributed to the delayed nu- limited their populations. This explanation might merical response exhibited by Burrowing Owls to be plausible for kestrels, as they are obligate sec- the vole high. Interestingly, the only study other ondary-cavity nesters (Bent 1938), but seems less than ours to examine the reaction of a Burrowing December 2001 Numerical Responses oe Vole Predators 293 Owl population to a prey high, showed that owls were most plentiful on the landscape; whereas, in a Chilean semi-desert also exhibited a delayed adults would be forced to be generalists, eating numerical response (Jaksic et al. 1997). whatever prey was available in their breeding ter- Unlike Short-eared Owls, Burrowing Owls con- ritories each year. Further research is needed to sume a wide variety of prey items, and their diet test this hypothesis because, although studies sug- often changes depending on the availability ofprey gest that adult breeders in these species are gen- in the environment (Green et al. 1993, Plumpton eralist predators faithful to their breeding sites, and Lutz 1993, Silva et al. 1995,Jaksic et al. 1992). little is known about the ranging behavior of im- Thus, Burrowing Owls are not overly reliant on any mature hawks (Preston and Beane 1993, England one type of prey, and can switch to take advantage et al. 1997). of peaks in several prey species (Fig. 1). The above scenario does not explain why Red- Ferruginous Hawks showed a synchronous re- tailed and Swainson’s Hawk numbers were higher sponse to the vole outbreak, reaching their highest in 1998 than they were in 1996. This delayed par- A relative population size in 1997. nomadic ten- tial response to the vole high could have been dency has been suggested for breeding popula- caused by an increase in adult survivorship, by an tions of this species (Schmutz and Hungle 1989, increase in breeding fidelity, or by a number of Bechard and Schmutz 1995). However, unlike immature birds from 1997 returning to breed in Short-eared Owls, Ferruginous Hawks did not ap- the study area in 1998. pear to react strongly to the voles and remained Northern Harriers showed a synchronous in- an uncommon species through the course of our crease with the meadow vole peak in our study. study (Fig. 3). We did not examine reproduction, This is in agreement with the results ofa long-term but if Ferruginous Hawks fledged more young in study in Wisconsin, which found that harriers fluc- 1997 than in other years, we might not expect to tuated in synchrony with meadow voles during a see an increase in the breeding population until 16-yr period (Hamerstrom 1979). However, similar 1999 when those fledglings reached breeding age to the populations of Swainson’s and Red-tailed (Bechard and Schmutz 1995). Alternatively, per- Hawks, the population of Northern Harriers re- haps these hawks specialized on Richardson’s mained at higher levels in 1998 than in 1996, sug- ground squirrels in our area, as has been noted in gesting that perhaps they experienced high pro- other studies (e.g., Schmutz and Hungle 1989), in ductivity in 1997 and/or high survivorship and site which case Ferruginous Hawks would be expected fidelity over the winter of 1997-98. Northern Har- to show little reaction to vole populations. riers commonly feed on meadow voles, but they Populations of Red-tailed and Swainson’s Hawks also supplement their diet with several other prey showed elements of both synchronous and delayed species (MacWhirter and Bildstein 1996). Unlike responses. We noted, though, that most of the Short-eared Owls, the more generalized diet of hawks in the high-food year were non-breeding, harriers likely allows them to remain on the Regina immature birds. It was common in the 1997 breed- Plain as a resident population in years of more ing season to see dozens ofimmature buteos roost- moderate vole abundance. ing communally in fields. In 1996 and 1998, the only similar densities of hawks occurred when Acknowledgments adults congregated to feed near tractors that were We wish to acknowledge the efforts of R. Sissons, J tilling fields, and those observations were compar- Skilnick, D. Grier, and D. Junor for their thoroughness atively rare. This raises the intriguing possibility in helping watch the skies. We are grateful to C.S. Hous- that different age-classes ofthese species might use ton, N.L. Brown, and D.K. Rosenberg for critical reviews of this manuscript, and thank M. Brigham, D. Schock, different strategies for distributing themselves geo- and C. Aldridge for helpful comments on earlier drafts graphically with respect to prey. That is, adults may We thank K. Scalise for enabling this research through choose to be faithful to nesting sites (which may her fund-raising and coordinating efforts. Financial and be limited) returning to the same territories each in-kind support for this project was provided by World , year regardless of prey; whereas, immature buteos Wildlife Fund, Saskatchewan Environment and Resource may opt for a nomadic lifestyle, searching at a large Management, Foothills Pipe Lines, TransGas, Trans- Canada Pipeline, InterProvincial Pipe Lines Inc. Envi- geograpliic scale and settling in areas with high ronmental Youth Corps Canada, Wildlife Preservation prey availability. In this scenario, immature hawks Trust Canada, Nature Saskatchewan, and Operation Bur- could specialize on hunting voles wherever they rowing Owl. 294 Biology VoL. 35, No. 4 A Literature Cited 1997. long-term study of vertebrate predator re- sponses to an El Nino (ENSO) disturbance in western Andersson, M. 1981. Reproductive tactics of the long- South America. Oikos 78:341—354. tailed skua Stercorarius longicaudus. Oikos 37:287-294. James, P.C., TJ. Ethier, and M.K. Toutloff. 1997. Pa- Bechard, M.J. and J.K. Schmutz. 1995. Ferruginous rameters of a declining Burrowing Owl population in Hawk {Buteo regalis). In A. Poole and F. Gill [Eds.], Saskatchewan. Pages 3-5 mJ.L. Lincer and K. Steen- The birds of North America, No. 172. The Academy hof [Eds.], The Burrowing Owl, its biology and man- of Natural Sciences, Philadelphia, PA and American agement including the proceedings of the first inter- DC Ornithologists’ Union, Washington, U.S.A. national Burrowing Owl symposium. Raptor Res J. Bent, A.C. 1938. Life histories of North American birds Report 9. ofprey. Part 2. Dover Publ. Inc., New York, NYU.S.A. , G.A. Fox, AND TJ. Ethier. 1990, Is the opera- ClAYTON, K.M. 1997. Post-fledging ecology ofBurrowing tional use of strychnine to control ground squirrels Owls in Alberta and Saskatchewan: dispersal, survival, detrimental to Burrowing Owls? RaptorRes. 24:120- J. habitat use, and diet. M.S. thesis, Univ. Saskatchewan, 123. Saskatoon, SK Canada. Korpimaki, E. 1992. Population dynamics of Eennoscan- Doyle, F.I. andJ.N.M. Smith. 2001. Raptors and scaven- dian owls in relation to wintering conditions and be- gers. Pages 377-404 in C.J. Krebs, S. Boutin, and R. tween-year fluctuations of food. Pages 1-10 in C.A. Boonstra [Eds.], Ecosystem dynamics of the boreal Galbraith, I.R. Taylor, and S. Percival [Eds.], The ecol- forest: the Kluane project. Oxford Univ. Press, New ogy and conservation of European owls. UK Nature NY York, U.S.A. Conservation No. Joint Nature Conservation Com- 5, England, A.S., M.J. Bechard, and C.S. Houston. 1997. mittee, Peterborough U.K. Swainson’s Hawk {B. swainsoni). In A. Poole and F. AND K. Norrdahl. 1991. Numerical and function- Gill [Eds.], The birds ofNorth America, No. 265. The al responses of kestrels. Short-eared Owls, and Long- Academy of Natural Sciences, Philadelphia, PA and eared Owls to vole densities. Ecology 72:814-826. American Ornithologists’ Union, Washington, DG MacWhirter, R.B. and K.G. Bildstein. 1996. Northern U.S.A. Harrier (Circus cyaneus). InA. Poole and F. Gill [Eds.], Gai.ushin, V.M. 1974. Synchronous fluctuations in pop- The birds of North America, No. 210. The Academy ulations of some raptors and their prey. Ibis 116:127- of Natural Sciences, Philadelphia, PA and American 134. Ornithologists’ Union, Washington, DC U.S.A. Green, A.G., R.E. Fitzner, R.G. Anthony, and L.E. Roc^ Millsap, B.A. and M.N. LeFranc,Jr. 1988. Road-transect ERS. 1993. Comparative diets of Burrowing Owls in counts for raptors: How reliable are they? RaptorRes. J. Oregon and Washington. Northwest Sci. 67:88-93. 22:8-16. Hamerstrom, F. 1979. Effect of prey on predator: voles Newton, I. 1976. Population limitation in diurnal rap- and harriers. 96:370-374. tors. Can. Field-Nat. 90:274—300. Harris, W.C., A. Kabzems, A.L. Kosowan, G.A. Padbur, Nielsen, O.K. 1999. Gyrfalcon predation on ptarmigan, and J.S. Rowe. 1983. Ecological regions of Saskatch- numerical and functional responses./. Anim. Ecol. 68. ewan. Sask. Parks and Renewable Resources Tech. 1034-1050. Bull. 10, Saskatoon, SK Canada. Pl.UMPTON, D.L. AND R.S. LuTZ. 1993. Prey selection and Hochachka, W.M., K. Martin, F. Doyi.e, and C.J. Krebs. food habits of Burrowing Owls in Colorado. Great Ba- 2000. Monitoring vertebrate populations using obser- sin Nat. 53:299-304. vational data. Can. Zool. 78:521-529. Preston, C.R. and R.D. Beane. 1993. Red-tailed Hawk J. Hoi.t, D.W. and S.M. Leasure. 1993. Short-eared Owl (Buteojamaicensis) InA. Poole and F. Gill [Eds.], The . {Asio flammeus). In A. Poole and F. Gill [Eds.], The birds ofNorth America, No. 52. The Academy ofNat- birds ofNorth America, No. 62. The Academy ofNat- ural Sciences, Philadelphia, PA and American Orni- ural Sciences, Philadelphia, PA and American Orni- thologists’ Union, Washington, DC U.S.A. thologists’ Union, Washington, DC U.S.A. Rohner, C., J.N.M. Smifh, Stroman, M. Joyce, F.I J. Houston, C.S. 1997. Banding of Asio owls in south-cen- Doyi.e, and R. Boonstra. 1995. Northern Hawk Owls tral Saskatchewan. Pages 237—242 in J.R. Duncan, in the ncarctic boreal forest: prey selection and pop- D.H. Johnson, and T.H. Nicholls [Eds.], Biology and ulation consequences of multiple prey cycles. Condor conservation ofowls ofthe northern hemisphere: 2nd 97:208-220. international symposium. USDA Gen. Tech. Rep. NC- 1996. The numerical response of Great Horned . MN 190, St. Paul, U.S.A. Owls to the snowshoe hare cycle: consequences of Jaksic, F.M.,J.E.Jimenez, S.A. Castro, and P. Feinsinger. non-territorial ‘floaters’ on demography./. Anim. Ecol 1992. Numerical and functional response ofpredators 65:359-370. to a long-term decline in mammalian prey at a semi- Rothfels, M., L. Twolan, and S. Nadeau. 1999. RENEW arid Neotropical site. Oecologia 89:90-101. Report No. 9: 1998-1999. Minister Public Works and , S.I. Sii.va, P.L. Meserve., and J.R. Gutierrez. Ck)v. Serv., Ottawa, ON Canada. [http://www.cws-scfec December 2001 Numerical Responses of Vole Predators 295 gc.ca/es/renew/RENEW98_99/RenewJPDF/renew9e. Wellicome, T.I. 2000. Effects offood on reproduction in pdf] Burrowing Owls {Athene cunicularia) during three stag- SCHMUTZ,J.K. AND DJ. HUNGLE. 1989. Populations ofFer- es of the breeding season. Ph.D. dissertation, Univ. ruginous and Swainson’s Hawks increase in synchrony Alberta, Edmonton, AB Canada. with ground squirrels. Can. f. Zool. 67:2596-2601. AND E.A. Haug. 1995. Second update ofstatus re- Suva, S.I., 1. Lazo, F. Silva-Aranguiz, F.M. Jaksic, P.L. port on the Burrowing Owl Speotyto cuniculariain Can- Meserve, and J.R. Gutierrez. 1995. Numerical and ada. Committee on the Status ofEndangered Wildlife functional response of Burrowing Owls to long-term in Canada, Ottawa, ON Canada. mammal fluctuations in Chile. J. Raptor Res. 29:250- G.L. Holroyd, K. Scause and E.R. Wiltse, 1997. , 255. The effects of predator exclusion and food supple- Sissons, R.A., K.L. Scalise, and T.I. Wellicome. 2001. mentation on Burrowing Owl {Speotyto cunicularia) Nocturnal foraging and habitat use by male Burrow- m population change in Saskatchewan. Pages 487-497 ing Owls in a heavily-cultivated region of southern Saskatchewan./. Raptor Res. 35:304—309. J.R. Duncan, D.H.Johnson, and T.H. Nicholls [Eds J, Biology and conservation of owls of the northern Skeel, M.A., Keith, and C.S. Palaschuk. 2001. A pop- J. hemisphere: 2nd international symposium. USDA ulation decline recorded by Operation Burrowing MN Owl in Saskatchewan./. RaptorRes. 35:371-377. Gen. Tech. Rep. NC-190, St. Paul, U.S.A. Solomon, M.F. 1949. The natural control ofanimal pop- Wiklund, C.G. AND J. Stigh. 1986. Breeding density of ulations./. Anim. Ecol. 18:1-35. Snowy Owls Nyctea scandiaca in relation to food, nest Steenhoe, K., M.N. Kochert, and T.L. McDonald. 1997. sites, and weather. Ornis Scand. 17:268—274. Interactive effects of prey and weather on Golden Fa- Wilson, R.T., M.P. Wilson, and J.W. Durkin. 1986. gle reproduction./. Anim. Ecol. 66:350-362. Breeding biology of the Barn Owl Tyto alba in central Todd, L.D. 2001. Survival and dispersal ofjuvenile Bur- Mali. Ibis 128:81-90. rowing Owls {Athene cunicularia) during the post- Yosef, R. 1996. Loggerhead Shrike {Lanius ludovicianus) fledging, pre-migratory period. M.S. thesis, Univ. Re- In A. Poole and F. Gill [Eds.], The birds of North gina, Regina, SK Canada. America, No. 231. The Academy of Natural Sciences, ViRKKOLA, R. 1992. Fluctuations of vole-eating birds of Philadelphia, PA and American Ornithologists’ prey in northern Finland. OrnisFenn. 69:97-100. Union, Washington, DC U.S.A.