Feeding Ecology of the Spotted Owl in California

Fall 1980 Lee--SmallestNestling Goshawk 73

Figure1. One-month-oldGoshawk chicks. The adult femaleis in the background.

FEEDING ECOLOGY OF THE SPOTTED OWL IN CALIFORNIA

by Cameron Barrows CaliforniaState University, Long Beach LongBeach, California 90840*

Abstract The SpottedOwl (Strixoccidentalis) from the CoastRange of northernCalifornia and the PeninsularRange of southernCalifornia preys heavily on forest-dwellingmammals, especiallythe dusky-lootedwoodrat (Neotoma fuscipes). Meadows are not regularly huntedby theseowls, despite large numbers of prey there.Indexes of preybiomass di- versityindicate a relativelynarrow prey basefor the SpottedOwl in California.The ScreechOwl (Otusasio) is an uncommonbut regularprey item of the SpottedOwl; mobbingby ScreechOwls in responseto imitatedSpotted Owl callsis described. *Current address:3162 Yellowtail Drive, Los Alamitos, CA 90720

RaptorResearch 14(3):73-78 74 RAPTOR RESEARCH Vol. 14 No. 3

Introduction Forsman(1975) provided a very completeanalysis of SpottedOwl (Strixoccidentalis) preyin Oregon,but comprehensiveregional food data from other locations are nearly absent.Prior to Forsman'sstudy, Marshall (1942) reported prey contents of two Spotted Owl stomachstogether with severalpellets from onelocation in the SierraNevada, Cal- ifornia, and three owl stomachcontents from the northernCascades, Oregon. Other publishedaccounts have describedless than 20 prey itemsfrom a singlelocation, in mostcases from only one stomach (Richardson 1906; Daggett 1913; Dickey 1914; Huey 1932;Miller 1933;Johnson and Russell 1962; Smith 1962; Maser 1965; Kertell 1977). I presentprey data from nine pairs of SpottedOwls in California.These data are then comparedto thosepresented by Forsman(1975) for Oregon.Measures of prey species diversityand biomass diversity are alsoconsidered. This comparison permits the eluci- dationof patternsin SpottedOwl preyselection over a wide geographicarea. StudySite and Methods Prey data from the North CoastRange of northernCalifornia were gatheredfrom four pairsof SpottedOwls from Junethrough August 1977. Two pairswere in the Na- ture conservancy'sNorthern California Coast Range Preserve, Mendocino County. An- other pair wason Point ReyesPeninsula, Marin County,and a fourth pair at SanGero- nimo, Marin County. In the PeninsularRange of southernCalifornia data were collectedfrom five pairsof owlsbetween June 1978 and November1979: one pair eachin the SanJacinto Moun- tains,Riverside County; Santa Ana Mountains,Orange County; Palomar Mountain, San Diego County;and two pairsin the CuyamacaMountains, San Diego County. Vegetationat all locationswas typical of the mixed evergreenforest (Sawyer et al. 1977).A morecomplete description of the habitatcharacteristics at eachsite is givenin Barrowsand Barrows (1978) and Barrows (1980). All preyitems identified were recovered from SpottedOwl pellets,totalling 250 pelletsfrom the North CoastRange and 200 pelletsfrom the PeninsularRange. The accu- racy of usingowl pelletsin a determinationof prey consumptionwas discussed by Err- ington(1930). Mammalian prey itemswere identifiedusing the keysof Ingles(1965) and Glass(1973). Small mammal densities in meadowswere estimatedby observingvisual signsof activity(i.e., runways).Other prey specieswere identifiedby comparisonwith museumspecimens and, in the caseof insects,with living individuals.Prey weightmea- surementsused in the calculationof percentbiomass are from Forsman(1975). Prey biomassdiversity was calculated as the productof the numberof individualsof a particular speciesmultiplied by that species'mean weight.

Results and Discussion The preyspecies and their relative proportion in the SpottedOwls' pellets are shown in table1. Calculationsof percentbiomass are also included as a moredescriptive mea- sureof the value of eachprey speciesin meetingthe owls'energetic requirements. Tablei showstwo majortrends: there is a clearpreference by the owlsfor forest-dwell- ing species;and the dusky-lootedwoodrat (Neotoma fuscipes) accounts for a significant proportionof the totalbiomass consumed by SpottedOwls. The onlynonforest species listed in table i is the Californiameadow mouse (Microtus californicus)whose preferred habitat is meadows. Spotted Owl huntingranges occasion- ally encompassedmeadows which were found to harborlarge numbers of meadowmice. Fall 1980 Barrows--FeedingEcology 75

Table1. SpottedOwl PreyItems from the NorthCoast and PeninsularMountain Rangesof California.

NorthCoast Range PeninsularRange (N -- 375) (N = 296) N Percent Percent N Percent Percent Species Occurrence Biomass Occurrence Biomass

M am m als Neotomafuscipes 112 29.9 69.2 98 33.1 79.3 Arborimuslongicaudus 64 17.1 4.0 0 0.0 0.0 Glaucomyssabrinus 56 14.9 14.8 0 0.0 0.0 Peromyscussp. 48 12.8 2.4 91 30.7 6.0 Scapanuslatimanus 4 1.1 0.5 I 0.3 0.2 Microtuscalifornicus 2 0.5 0.1 3 1.0 0.2 Lasiurus cinereus 2 0.5 0.05 0 0.0 0.0 Mustela erminea 1 0.3 0.1 0 0.0 0.0 Sylvilagussp. 1 0.3 0.8 I 0.3 1.0 Eutamiassp. 1 0.3 0.2 I 0.3 0.2 Thomomysbottae 0 0.0 0.0 23 7.8 6.0 unident. mammals 2 0.5 0.1 0 0.0 0.0

Birds Cyanocittastellari 10 3.1 2.7 5 2.1 2.0 Otus as/o 5 1.3 1.6 3 1.0 1.3 Glaucidiumgnoma I 0.3 0.1 0 0.0 0.0 otherbirds a 21 5.6 3.4 18 6.1 3.8

Arthropodsb 45 12.0 0.2 52 17.6 0.3 a. IncludesMelanerpes formicivorus, Colaptes auratus, Turdus rnigratorius, Junco hyernalis, unidentified species. b. Includeslargely insects (Orthoptera, Coleoptera), with a minorrepresentation ofScorpionida andDiplopoda. The low numbersof thisspecies in SpottedOwl pelletsin both regionssuggest that the owlsrarely hunt in suchopen areas. My observationsof SpottedOwls during twilight hoursshowed them to hunt exclusivelyfrom elevatedperches; the absenceof such perchesin openmeadows may account for theirlow utilizationof theseareas. The remainingprey speciescommonly inhabit forested areas. Of thesespecies, the dusky-footedwoodrat, the westernflying squirrel (Glaucomys sabrinus), the red-backed tree mouse(Arborimus longicaudus), and the deermouse (Peromyscus sp.) make up the largestportion of the diet in areasexamined. In the North Coastand PeninsularRanges, California, and in the easternSiskyou Mountainsof southernOregon (Forsman 1975), the dusky-lootedwoodrat comprises over69 percentof the SpottedOwl prey biomass.In otherregions of Oregonreported by Forsman(1975) the dusky-lootedwoodrat constitutes only a minorportion, or is absent,as a prey item in the SpottedOwls' diet. In thoseregions, the flyingsquirrel and thebushy-tailed woodrat (Neotoma cinerea) were the mostcommon prey species. Prey diversity(table 2) givesa furthercharacterization of the SpottedOwl's prey base.Numbers and proportions of preyspecies in eachregion are roughlysimilar, hence the consistentvalues for the speciesdiversity index. However, prey biomassdiversity index,which emphasizes relative mass of individualprey items,is lowerfor California areasand the easternSiskyou Mountains than for any other Oregonsites. Lower prey biomassdiversity values indicate a relativelynarrow prey base,in thiscase a reflection 76 RAPTOR RESEARCH Vol. 14 No. 3 of thelarge numbers of heavydusky-looted woodrats in thediet. Prey biomass diversity maybe a moredescriptive method of depictingthe preybase because it doesnot com- putea woodratand a beetleto be of equalimportance. Table2. RegionalComparison of SpottedOwl PreyDiversity.

Prey Species Prey Biomass Location N Diversitya Diversitya Oregon(7areas] • 2,453 2.43-1.18 2.20-1.50

Oregon,eastern SiskyouMountains b 176 1.81 1.02

Northern California CoastRange 375 1.97 1.15

Southern California PeninsularRange 296 1.62 0.87 a. Shannon-Wienerdiversity index, H'=- pi logpi. b. Calculatedfrom Forsman1975; (tables 1-8). In regionswhere dusky-lootedwoodrats commonly occur, such as thosereported fromCalifornia and the easternSiskyou Mountains, these mammals appear to be prefer- entiallytaken by SpottedOwls. Forsman (1975) suggested that SpottedOwls prey se- lectivelyon larger,arboreal mammals over more abundant, smaller, terrestrial species. It is unclear,however, whether it is largersize or semiarborealhabits or a combination of thesefactors that stimulateSpotted Owls to selectsuch a high proportionof dusky- looted woodrats. Future researchmight examinethe possiblerelationship between woodrat popu- lationsand the reproductivebiology and/or habitatselection of SpottedOwls in Cali- fornia.Other studieshave revealeda correlationbetween prey numbersand the repro- ductivesuccess of owls(Craighead and Craighead1956; Southern 1970; Adamcik et al. 1978),and of otherraptors (Hamerstrom 1979). Spotted Owls do not breedevery year (Forsman1975; Barrows unpubl. data). The dominanceof woodratsin SpottedOwl diets from the areasreported in thisstudy and the knowledgethat woodratpopulations can fluctuategreatly from year to year (Linsdaleand Tevis 1951)suggest a possiblecause andeffect relationship in SpottedOwl breedingpatterns. Of the nonmammalianprey (table1), the occurrenceof the ScreechOwl (Otusasio) in the SpottedOwls' diet warrantsparticular mention. So-called cannibalism by Spotted Owlswas reported by Richardson(1906) and Daggett (1913) after finding remains of the PygmyOwl (Glaucidiumgnoma) in SpottedOwl stomachs.Forsman (1975) reported the Screech,the Pygmy,and the Saw-WhetOwl (Aegoliusacadicas) as SpottedOwl prey items.While callingfor SpottedOwls during summer nights, using vocal imitations of theircalls, I wasfrequently "mobbed" by ScreechOwls. Immediately following my imi- tationof a SpottedOwl call,Screech Owls in the nearvicinity frequently began calling in a raucous,chattering manner. The ScreechOwls were usually in familygroups of one or two adultsaccompanied by severalowlets. Both adults and immatures used the chat- teringcalls. Adult ScreechOwls moved sporadically through the treessurrounding me while calling;the owletsremained more or lessstationary. The adult and immature Fall 1980 Barrows--FeedingEcology 77

ScreechOwls were easily located with a flashlightwhile engaged in thismobbing behavior.A highproportion of ScreechOwl remainsin the SpottedOwl pelletswere re- coveredduring July, the sameperiod when the mobbingbehavior of ScreechOwl fam- ily groupswas most pronounced. The chattering call by ScreechOwls in responseto a SpottedOwls' call was also heard, but at a lowerintensity, throughout the late summer andfall months.The effect of the ScreechOwls' mobbing behavior on SpottedOwls is unknown,but it certainlymade my detectionof thesesmall owls easier. Conclusions Generalconclusions which can be drawnfrom thisanalysis of SpottedOwl feeding ecologyare: 1. The owls'diet consistslargely of forest-dwellingmammalian species, with prey itemsaveraging more than 100 g. 2. Meadowsor largeopen expanses were not regularlyused by SpottedOwls for hunting,despite potentially large numbers of preythere. 3. Whenavailable, the dusky-looted woodrat comprises a high proportion of Spotted Owls'diet in Californiaand may be preferentiallyselected by the owls. I haveno prey datafrom the SierraNevada or dataon temporalvariation in prey selection.The needfor moredata pertainingto the relationshipbetween prey and SpottedOwl breedingpatterns is discussedabove. These gaps need to be filledbefore ourknowledge of SpottedOwl feedingecology is complete. Acknowledgments I wishto conveymy appreciationfor the useof the Nature Conservancy'sNorthern CaliforniaCoast Range Preserve and CuyamacaRancho State Park asbases for my re- gionalstudy. For their helpfulcomments on the manuscript,I thankKatherine Barrows and Charles Collins.

Literature Cited Adamcik,R. S., A. W. Todd,and L. B. Keith. 1978.Demographic and dietaryresponses of Great HornedOwls duringa SnowshoeHare fluctuation.Canadian Field-Nat. 92(2):156-166. Barrows,C. 1980.Summer roost selection by SpottedOwls: an adaptationto heat stress. M.S. thesis,California State University, Long Beach.51 pp. Barrows,C., and K. Barrows.1978. Roostcharacteristics and behavioralthermoregula- tion in the SpottedOwl. WesternBirds 9(1):1-8. Craighead,J. j., and F. C. Craighead.1956. Hawks, Owls, and wildlife. Stackpole,Har- rison,PA. 443 pp. Daggett,F. S. 1913. Anotherinstance of cannibalismin the SpottedOwl. Condor 15:40-41. Dickey,D. R. 1914.The nestingof the SpottedOwl. Condor16:193-202. Errington, P. L. 1930. The pellet analysismethod of raptor food study. Condor 32:292-296. Glass,B. P. 1973.A key to the skullsof northAmerican mammals. Department of Zoolo- gy, OklahomaState University. 59 pp. Hamerstrom, F. N. 1979. Effect of prey on predators: voles and harriers. Auk 96(2):370-374. Huey,L. M. 1932.Note on the foodof an ArizonaSpotted Owl. Condor34:100-101. 78 RAPTOR RESEARCH Vol. 14 No. 3

Ingles,L. G. 1965.Mammals of the Pacificstates. Stanford University Press, Stanford, CA 5O6pp. Johnson,N. K., and W. C. Russell.1962. Distributional data on certainowls in the western Great Basin. Condor 64:513-514. Kertell, K. 1977. The Spotted Owl at Zion National Park, Utah. Western Birds 8(4):147-150. Linsdale,j. M., and L. P. Tevis, Jr. 1951. The dusky-footedwoodrat. Univ. of Calif. Press,Berkeley, CA. 664 pp. Marshall,J. T., Jr. 1942.Food and habitat of the SpottedOwl. Condor42:66-67. Maser,C. 1965.Spotted Owl preyson duskytree mice.Murrelet 46(3):46. Miller, A. 1933. The red tree mousepreyed upon by the SpottedOwl. J. Mammal. 14:162. Richardson,C. H. 1906.Cannibalism in owls.Condor 8(2):57. Sawyer,J., D. Thornburgh,and J. Griffen. 1977. Mixed evergreenforest. Pages 359-415 in M. Barbourand J. Major, eds.,Terrestrial vegetation of California.Wiley and Sons, New York. Smith,C. 1963.First breedingrecord of the SpottedOwl in BritishColumbia. Condor 65:440. Southern,J. N. 1970.The naturalcontrol of a populationof Tawny Owls(Strix aluco). J. Zool. London 162:197-285.

ANNOUNCEMENTS REQUEST FOR OBSERVATIONS

Pleasereport any fall/winter sightingsof adultHarlan's race Red-tailed Hawk (Buteo iamaicensisharlani). Include plumage description or photographsalong with sighting locations. Historical information is also welcome. Sendinformation to: David Mindell, c/o Bureauof Land Management,4700 E. 72nd, Anchorage,AK 99507.

CONFERENCE FOR INTERNATIONAL ASSOCIATION FOR FALCONRY AND CONSERVATION OF BIRDS OF PREY The InternationalAssociation for Falconryand Conservationof Birdsof Prey is or- ganizinga conference,'Understanding the Goshawk',to be heldin Oxford,Great Brit- ain, from 29th Septemberto 1st October1981. This followsthe 1977 conferenceon 'Birdof PreyManagement Techniques', which was attended by ornithologists,vets and falconers from 20 countries. Goshawkpopulations have recently increased in someparts of Europe,following re- strictionson pesticideusage, and there is renewedcontroversy between hunting and protectioninterests about goshawk control. This conference will present recent research findingson goshawk population dynamics, predation and management, thus providing a basisfor formaland informaldiscussion of how someof the conflictsmight be resolved. Theprogram will be of 1/•-hourpapers by mainspeakers, with some shorter contribu- tionsand no late-night sessions! Proceedings will be published. Thoseinterested in givingpapers or otherwiseparticipating please contact: Dr. Rob- ert Kenward,Institute of TerrestrialEcology, Monks Wood, AbbotsRipton, HUNT- INGDON, Great Britain.