Nesting and Diet of Long-Eared Owls in Conifer Forests, Oregon

The Condor91:908-912 Q The CooperOrnithological Society 1989

NESTING AND DIET OF LONG-EARED OWLS IN CONIFER FORESTS, OREGON

EVELYN L. BULL AND ANTHONY L. WRIGHTS USDA Forest Service,Pacijic NorthwestResearch Station, La Grande, OR 97850

MARK G. HENJUM OregonDepartment of Fish and Wildlife, La Grande, OR 97850

Abstract. Long-eared Owls (Asiootus) nested in extensive stands of grand fir (Abies grandis) in northeastern Oregon. Nineteen of 20 nests were in dwarf-mistletoe brooms (Arceuthobiumspp.) in Douglas-fir (Pseudotsugamenziesii) trees. Young were raisedat 70% of the nests,and brood size averaged3.0. The northern pocketgopher (Thomomystalpoides) was the primary prey item both in frequency(56%) and biomass (74%). These observations differ from those reported previously for the Long-eared Owl in North America. Key words: Long-earedOwl; nesthabitat; reproduction;diet; Oregon.

INTRODUCTION aspects(3 16-45”) 3 1% are eastern(46-l 36”), 20% The Long-eared Owl (Asio otus) has a circum- are southern (136-225”) and 16% are western polar, holarctic distribution in boreal, temperate, (226-3 15’). Mediterranean, and steppeclimatic zones (Mik- The study area consistsof a mosaic of forests kola 1983). Becauseof its wide distribution and (84% of area) with many shallow-soiled grass- tendency to roost and nest in concentrations, lands (16%). Forest types in the study area (clas- much has been published on its food habits and sified by Burr 1960) are 14% open ponderosa nesting (Hagen 1965, Kallander 1977, Mikkola pine (Pinus ponderosa), 41% ponderosa pine- 1983). However, the majority of information re- Douglas-fir (Pseudotsugamenziesii), and 45% ported has been for owls inhabiting small, scat- grand fir (Abiesgrandis) mixed with Douglas-fir, tered woodlands surrounded by open land ponderosa pine, and western larch (Larix occi- (Wijnandts 1984, Marks 1986). No such infor- dentalis) (45%). Forested stands were uneven- mation is available on this speciesin a predom- aged and were classified into three successional inantly forested habitat in North America. Our stages:(1) stands with trees ~30 cm dbh (di- objectives were to describe the nesting habitat, ameter at breast height); (2) standswith trees 30- reproduction, and diet of Long-eared Owls in 50 cm dbh; and (3) stands with trees >50 cm extensive conifer forestsin northeastern Oregon. dbh. Ninety percent of the area had not been logged in 40 years; the remainder had a partial STUDY AREA removal of the overstory within the last 15 years. The study was conducted on a 5,270-ha area on the Starkey Experimental Forest (Starkey) 35 km METHODS southwestof La Grande in northeastern Oregon. Starkey is characterized by undulating uplands We searched for Long-eared Owls from 15 dissectedby moderate to steeply walled drain- March-25 May in 1987 and from 29 February- ageswith elevations of 1,070-1,524 m. Fifty-one 25 May in 1988, walking routesthrough the study percent of the study area is in slopes (area be- area after sunset. Routes were ~0.5 km apart to tween ridges and draws), 44% in ridges (areas ensurethat the entire area was covered.The study with O-10% slopes), and 5% in draw bottoms. area was searchedonce prior to and once after Thirty-two percent of the slopes are northern 15 April each year. Along a route we stopped every 0.1 km, imitated the call of a Long-eared Owl, and listened 3 min for a response.If an owl was heard, we recorded date, time, and location. * Received27 February 1989. Final acceptance27 July 1989. To locate nests, we returned during the day to 2 Present address:Running Creek Ranch, P.O. Box sites where owls had been heard and searched 1515, Hamilton, MT 59840. for whitewash, pellets, and potential nest plat-

[9W LONG-EARED OWL NESTING HABITAT 909 forms. Any tree with more than six pellets un- measured with a planimeter on aerial photo- derneath it and within 150 m of a nest was de- graphs. fined as a roost tree. Pellets were identified by The following habitat characteristicsin 40 O.l- size, although Long-eared Owls were observed ha plots selected at random in grand fir forest at approximately half these roosts.We identified types within the study area were recorded: the species of tree, type of roost (mistletoe or successional stage, slope gradient, number of branch), and measured dbh, height of roost, and canopy layers, logging activity (none or partial distance from nest. Platforms with an adult bird removal), and number of trees with dwarf-mis- in an incubation position, with eggs,or with nest- tletoe. lings (live or dead) were defined as nests. There Chi-squareanalyses were usedto compare nest- were three caseswhere nestswere found after the site characteristics with characteristics in ran- nestlings had left the nest but were in trees near- dom plots: (1) in ridges, slopes, and draws; (2) by. The nest was easily identified by the presence in north, east, south, and west aspects;(3) in three of whitewash and pellets underneath it and by forest types; (4) in number of canopy layers; (5) pellets and Long-eared Owl feathers in the plat- in logging activity; and (6) in successionalstage. form. We compared the slope gradient and number of Nest sites were visited once a week to collect trees with dwarf-mistletoe brooms at nest sites pellets and look for dead owls, prey remains, or with the gradient and number of treeswith dwarf fledged young. After the young had left, we mistletoe at random plots (unpaired two-tailed climbed to each nest to collect pellets and prey t-test). Significance was defined when P 5 0.05. remains and to measure (diameter and depth) and describe the nest platform. Skulls and den- RESULTS taries in each pellet were identified with the use REPRODUCTION of keys (Maser and Storm 1970, Glass 1973). Skulls and dentariesof eachspecies were summed, The first Long-eared Owl was heard on 18 March and the number of each specieswas determined in both 1987 and 1988. In 1987 and 1988, re- by the number of skulls or pairs of dentaries, spectively, 25 and 34 owls were heard during whichever was larger. Biomass consumed by the sampling, and 88% and 53% were heard before owls was estimated for each prey speciesaccord- 26 April. Owls were heard calling from an hour ing to Bull et al. (1989). Because of variable before sunset until 7 hr after sunset. Vocal ac- weights of birds and their relative insignificance tivity was greatest within 2 hr of sunset when in the diet, avian specieswere not considered in 63% of the owls were first heard, yet only 26% biomass calculations. of the time spent listening was within this 2-hr The age of northern pocket gophers (Thomo- period. mys talpoides)was determined using a regression Clutch size at five nests ranged from three to formula in Janesand Barss(1985). All dentaries five eggs;we observed females on eggsfrom 15 with a condyloid processlength of 2 20 mm were April to 26 May. Mean brood size of 12 nests classified as adults and those ~20 mm were ju- within a week of when young started branching veniles. We used 90 g for adults and 30 g for (climbing out branches away from the nest) was juveniles in biomass calculations. 3.0 (SD = 1.2, Range = l-4). Young started We recorded the following characteristics of branching from 13 June until 1 July. each nest tree: species,dbh, height, age, and pres- Young were successfullyraised (lone young ence of dwarf-mistletoe brooms (Arceuthobium left the nest) at 14 (70%) of 20 nests. Nine of the spp.). The following habitat characteristicswere nestswere found during incubation, eight during recordedin a 0.1 -ha circular plot centered on the brooding, and three after branching. Fifty-six nest tree: location (ridge, slope, or draw), slope percent of the nestsfound during incubation were aspectand gradient, forest type and successional successful.Remains of depredatednestlings were stage,stem density, number of trees with dwarf- found at two nests, and remains of a depredated mistletoe, distance to opening >0.5 ha, number adult were found at one nest. We found remains of canopy layers, logging activity, and canopy of three fledgedjuveniles that had been killed by closure(four readingsat nest using sphericalden- Accipiterspp. Accipiter pluck sitescontaining re- siometer). Within a 500-m radius of each nest, mains of Long-eared Owls in the vicinity of five the percent of area in forests and openings was Long-eared Owl nest sites suggestedthat other 910 E. L. BULL, A. L. WRIGHT AND M. G. HENJUM

TABLE 1. Diet of Long-earedOwls during nestingin northeasternOregon, 1987-1988, determined from 1,128 pellets (1,123 prey items).

Prey item Weight % Biomass % Frequency

Mammals Northern pocket gopher 30,9@ 74.4 55.7 Vole (Mcrotusspp.) 26 15.2 20.5 Deer mouse (Peromyscusmanicuhztus) 16 11.8 Bushy-tailed woodrat (Neotomacinerea) 156 :.; 0.7 Red-backed vole (Clethrionomysgappert] Shrew (Sorex spp.) 195 0.50:9 ::: Other mammalsb 0.5 0.3 Birds Insects ‘Weight (g) usedfor biomasscalculations: 30 g usedfor juvenile and 90 g usedfor adult pocketgophers. bIncluded prey comprising~0.5% of diet: heathervole (Phenacomys intermedius) (27 g). yellow-pinechipmunk (Eutamiur amoenus) (56 g), and northernflying squirrel(Glaucomys sabrinus) (115 g).

adults had been killed. In addition, a Long-eared (SD = 24.65) of age. Nest height averaged 9.8 m Owl leg was found in a Great Horned Owl (B&o (SD = 3.24). Nineteen of the nestswere in dwarf- virginianus) nest. mistletoe brooms, and one was a stick nest- probably built by an accipiter. The mean di- DIET mensions of nest platforms were 49 (SD = 13.5) We collected 1,128 pellets from Long-eared Owl x 35 cm (SD = 9.98) with a mean depth of 23 nests and roosts located within 150 m of occu- cm (SD = 8.93). All of the nests were supported pied nests. The northern pocket gopher was the by branches, but the actual nest contents were primary prey item both in frequency and bio- needles at 14 nests, twigs at four nests, and grass mass (Table 1). Of the pocket gopher dentaries at one nest. The canopy closure over nests av- 72% were from juveniles and 28% were from eraged 62% (SD = 22.03). adults. Adults accountedfor 40% of the biomass, There was a significant difference in nest use and juveniles for 34%. Microtus spp. made up among forest types, as all 20 nests occurred in 20.5%ofthediet(15.4%M. montanus, 1.7%M. the grand fir type (x2 = 24.4, df = 1, P < 0.01). longicaudus,and 3.4% unknown). Owls selectednest sites in unlogged stands (x2 = Forty-four bird skulls were found in the pellets, 3.96, df = 1, P = 0.05) with a high number of and feather remains of 20 birds were found in dwarf-mistletoe brooms (t = 2.7, df = 58, P = or near nests. These feathers included remains 0.01). Plots at nest sites contained an average of ofthree Mountain Bluebirds (Sialia currucoides), 9.1 trees (SD = 8.11) with dwarf-mistletoe two Western Bluebirds (S. mexicana), two brooms, while random plots contained an av- Northern Flickers (Colaptes auratus), two Wil- erage of 3.7 (SD = 5.29). liamson’s Sapsuckers (Sphyrapicus thyroideus), By comparing nest sites with random plots, one Stellar’s Jay (Cyanocitta stelleri), one Clark’s there were no significant differencesamong slope Nutcracker (Nucifraga columbiana), one Brown- aspects (x2 = 3.19, df = 3, P = 0.38); ridges, headed Cowbird (Molothrus ater), one American slopes, or draws (x2 = 1.17, df = 2, P = 0.57); Robin (Turdus migratorius), one European Star- slope gradient (t = 0.01, df = 58, P < 0.50); ling (Sturnus vulgaris), one Dark-eyed Junco number of canopy layers (x2 = 0.62, df = 1, P = (Junco hyemalis), one Chipping Sparrow (Spi- 0.45); or successionalstages (x2 = 3.42, df = 2, zella passerina),and one warbler (Vermivora sp.). P = 0.20). One insect was identified as a ponderous borer Nest sites contained an average of 85.5 stems (Ergates spiculatus). < 10 cm dbh (SD = 55.9), 47.7 stems lo-50 cm dbh (SD = 10.8), and 1.6 stems >50 cm dbh NESTING HABITAT (SD = 1.79) in 0. l-ha plots. Canopy closure We located 20 nests in 1987-l 988. All nests were around the nest tree averaged 80% (SD = 15.7). in Douglas-fir trees which averaged 31 cm dbh Nineteen of the 20 nests were not within 30 m (SD = 9.97), 21 m tall (SD = 5.91), and 87 years of an opening (> 1 ha in size); average distance LONG-EARED OWL NESTING HABITAT 911 to an opening was 105 m (SD = 58.8). Within Peromyscus (Marti 1974), and heteromyids 500 m of nests the amount of area in forest and (Marks 1984) in the breeding diet. opening averaged 79% (SD = 6.48) and 2 1% (SD Even though Long-eared Owls in our study = 6.09) respectively. area preyed heavily on a different species,ju- At least six pellets were collected from each of venile pocket gophers,the prey-size distribution 37 roost treesin the vicinity of nests.These roosts was consistentwith that reported by Marti (1976) averaged 39 m (SD = 18.2) from the nest. Sev- and Marks (1984) where most ofthe prey weighed enty percent of these roosts were in Douglas-fir, ~60 g. The juvenile pocket gopherswere prob- 24% in grand fir, and 5% in ponderosa pine. The ably taken above ground, as Howard and Childs roost trees averaged 31 cm dbh (SD = 12.88). (1959) reported juveniles traveling along the Fifty-seven percent of the roost sites were in ground in March, April, and May. Scheffer(1954) dwarf-mistletoe brooms. The remainder were on noted a spring and early summer dispersal of branches. Roost sites averaged 4 m (SD = 1.60) subadult pocket gophersabove ground. off the ground. The high incidence of pocket gophers in the diet and the extensive amount of forest (86%) in our study area suggeststhat Long-eared Owls DISCUSSION were foraging in forest stands. The 14% of the We discovered several characteristics of Long- area in openings consistedof shallow-soiled scab eared Owl nesting habitat not previously re- flats-areas infrequently inhabited by pocket go- ported in North America. First, Long-eared Owls phers due to the shallow soil. Davis et al. (1938) nested in extensive conifer forests and were de- seldom found pocket gophersin soils < 10 cm in pendent on dwarf-mistletoe brooms in Douglas- depth. Consequently, we think the birds foraged fir as nest sites.In other studies,Long-eared Owls for pocket gopherslargely in open forestedstands typically used stick nests built by corvids (Craig or along edges.Only Armstrong (1958) reported and Trost 1979, Wijnandts 1984, Marks 1986). much foraging in woodlands; all other authors Mikkola (1983) reported three of 256 nests in (Randle and Austin 1952, Marti 1974, Mikkola witches’ brooms in Britain. We think that the 1983, Marks 1984) observed foraging in open owls nested in the smallest, most inconspicuous country. platform that would accommodate them, unlike Our observations revealed that theseowls may Mikkola’s (1983) observation that they used large be more abundant in conifer forests than pre- old nests. viously thought and may point to a more op- Second, all nest sites were in dense, unlogged portunistic nature of this species. stands of grand fir with no apparent regard for slope aspector gradient. We think that the birds ACKNOWLEDGMENTS chosethe grand fir forest type becauseof the cov- We thank H. D. Cooper,R. D. Dixon, and J. E. Hoh- er afforded by the nearly closed canopy. In con- mann for their assistancewith fieldwork. G. Alcom trast, Mikkola (1983) reported that most Long- and C. Wood assistedwith identification of bird feath- eared Owls nested in small patchesof woodland ers. C. D. Marti, J. S. Marks, J. W. Thomas, and T. among open meadows and fields. R. Madden reviewed the manuscript.Funding was pro- We think the dense canopy cover reduced pre- vided by the USDA Forest Service’s Pacific Northwest ResearchStation and Oregon Department of Fish and dation. We found depredated adults or young at Wildlife Nongame Fund. eight of 2Q nest sites. In spite of this predation we found higher nest success(70%) than Marks LITERATURE CITED reported in Idaho (34-51%; 1986). Only Craig and Trost (1979) reported a pre- ARMSTRONG,W. H. 1958. Nesting and food habits of the long-eared owl in Michigan. Museum, dominance of northern pocket gophers(43% by Michigan State Univ. Biol. Series 1:61-96. biomass) in the diet of Long-eared Owls. Janes BULL. E. L.. M. G. HENJUM.AND R. S. ROHWEDER. and Barss(1985) noted that pocket gopherstaken I989. Diet and optimal foragingofgreat grayowls. by Long-eared Owls were usually small with a J. Wildl. Manage. 53:47-50. mean weight of 41 g. Other studies reported a BURR,J. A. 1960. Soil survey, StarkeyExperimental Forest and Range, Union and Umatilla Counties, predominance of Microtus (Armstrong 1958, Oregon. USDA Soil Conserv. Serv. and For. Serv. Glue and Hammond 1974, Hagen 1965, Mik- C RAIG,T. H., AND C. H. TROST. 1979. The biology kola 1983) Apodemus(Wooller and Triggs 1968), and nestingdensity of breedingAmerican Kestrels 912 E. L. BULL, A. L. WRIGHT AND M. G. HENJUM

and Long-earedOwls on the Big Lost River, south- MARKS.J. S. 1986. Nest-site characteristicsand re- eastern Idaho. Wilson Bull. 9 1:50-6 1. productive successof Long-eared Owls in south- DAVIS, W. B., R. R. -sax, AND J. M. ARENDALE, western Idaho. Wilson Bull. 98:547-560. JR. 1938. Distribution of pocket gophers (Geo- MARTI, C. D. 1974. Feeding ecology of four sym- mys breviceps)in relation to soils. J. Mammal. 19: panic owls. Condor 76~45-61. 412-418. MARTI, C. D. 1976. A review of prey selectionby the GLASS,B. P. 1973. Key to the skullsof North Amer- Long-eared Owl. Condor 78:331-336. ican mammals. Oklahoma State Univ., Stillwater. MASER,C., AND R. M. STORM. 1970. A key to Mi- GLUE, D. E., AND G. F. HAMMOND. 1974. Feeding crotinae of the Pacific Northwest (Oregon. Wash- ecologyof the Long-eared Owl in Britain and Ire- ington, Idaho). Oregon State Univ. Bookstores, land. Br. Birds 67:361-369. Corvallis. HAGEN,Y. 1965. The food, population fluctuations MIKKOLA, H. 1983. Owls of Europe. Buteo Books, and ecologyof the long-eared owl (Asio otus(L.)) Vermillion, SD. in Norway. Norwegian State Game Res. Inst. Pa- RANDLE, W., AND R. AUST~. 1952. Ecologicalnotes pers Ser. 2, No. 23. on long-earedand saw-whet owls in southwestern HOWARD,W. E., AND H. E. CHILDS,JR. 1959. Ecology Ohio. Ecology 33:422-426. of pocket gophers with emphasis on Thomomys SCHEFFER,T. H. 1954. Concerningthe pocket gopher bottaeMewa. J. Aaric. Science 29~277-358. in mole range. Wash. Agric. Exp. Stn. Circ. 242: JANES, S. W., AND J. M. BARSS. 1985. Predation by l-4. three owl specieson northern pocket gophers of WLJNANDTS,H. 1984. Ecological energetics of the different body mass. Oecologia 67:76-g 1. Long-eared Owl (Asio otus). Ardea 72: l-92. KALLANDER,H. 1977. Food of the long-eared owl WOOLLER,R. D., A& G. S. Tmczs. 1968. Food of Asio otusin Sweden. Omis Fenn. 54:79-84. the Long-earedOwl in Inverness-shire.Bird Study MARKS, J. S. 1984. Feeding ecologyof breedinglong- 15:164-165. eared owls in southwesternIdaho. Can. J. Zool. 62:1528-1533.