The Breeding Success of Tawny Owls (Strix aluco&rpar ...

j RaptorRes.34(4):322-326 ¸ 2000 The Raptor ResearchFoundation, Inc.

THE BREEDINGSUCCESS OF TAWNYOWI,S (STRIXALUCO) IN A MEDITERRANEAN AREA: A LONG-TERM

STUDY IN URBAN ROME

LAMBERTO

RANA]•ZI

Dipartimento di Medicina5])e•imentale e Patologia,Policlinico"Umberto 1" Universitd di Roma "La Sapienza" Viale ReginaElena 324, 1-00161Rome,Italy ALBERTO

MAN GANARO

Via di Donna Olimpia152, 1-00152Rome,Italy LUCA SALVATI

PiazzaE Morosini12, 1-00136 l•me, Italy 6•tsspp.). Vegetation of city parks as well as suburban woodlandswas generally composedof strandsof mixed deciduouswood predominated by oaks (e.g., Quercus ilex). Nests were generally located in natural cavitiesof The breeding biology of the Tawny Owl (Strix aluco) old oaksand pines. The rate of territory occupationwas was studied in northern and central Europe specifically remarkablyhigh in both habitats,so the population denfocusing on the influence of food abundance on clutch sitydid not showsignificantvariationsamongyears(Rans•ze and productivity of young (Southern 1970, Wend- azzi et al. 2000). land 1984). Baudvin (1990) fbund a remarkable positive Proceduresfor mapping territories and locating nestcorrelation belween the reproductive output of Tawny ing sites lbllowed Ranazzi et al. (2000). Although the Owls and the percentageof woodlandrodentsin the diet bulk of the data on breeding successwere obtained fi•om of pairs in central France.Annual fluctuationsin Tawny pairs consecutivelystudied throughout the censuspenOwl breeding successwere directly linked to the abun- od, some pairs, especiallythoseof urban parks,were not dance of woodland rodents,especiallythe yellow-necked continuouslycertsuseddue to the impossibilityto visit their territories in some breeding seasons(e.g., occurmouse (Apodemus flavicollis),the main prey of this owl in rence of summer events or public works in parks and woods and mixed farmlands (Wendland 1984, Baudvin gardensand Tawny Owls nesting sometimeson private 1990,Jedrze. jewskiet al. 1994). Moreover,alternativeprey property). Data lbr 1984-85 seasonswere not considered (e.g., birds and amphibiaus) increased in diet in low due to the small nmnber of records available. The nummouseyears(Plesnikand Dusik 1994). Cyclicfluctuations ber of occupied territories censuscdeach year was 14.3 •n populationsof rodent prey is probablythe main factor _+5.9 (N= 200) fi•oni 1986-99 in urban plots, and 10.1 affecting•I:awnyOwl productivity,but other fhctors,such _+5.8 (N = 111) from 1989-99 in suburbanplots. as weafimr conditions, could be also involved (Kostrzewa Estimatesof the nmnber of young in nestswere made and Kostrzewa1990, Gil-Delgadoet al. 1995, Penteriani by broadcastingcalls of male Tawny Owls on a SANYO portable stereowith 6 W loudspeakerswithin the nesting 1997). area at a distance of' about 50 m l?om known nest sites In Mediterranean areas,very few studieshave fbcused on the study the aunual variations in the breeding sin> (see Ranazzi el al. 2000 lbr details) and listeninglbr recessof any rapmr (Gil-Delgado et al. 1995). The aim of sponses.Generallyall young respondedto callswith their persistent'p•zia' begging calls, so this method was used this studywasto assess the long-term breeding success of' to evalumethe number of successthl pairs and fledgling TawnyOwls in a Mediterranean urban area, checkingfor production (Wendland 1984). Nest site disturbancewas variations,if any,in productivityof young,and comparing reducedby limiting visitsto each territory to only two in them to the breeding t)erfbrmance of other areasin Eu- May-August. This period was chosen to ceususyoung rope. basedon a preliminary assessment in 1984-85 of young vocal activityat eight known nests.All fledglingsbegan METttOI)S to utter the 'ptzia' call in May and theyremainedin fimir The study wascarried out frown1984-99 in five urban parents' territories at least until early August, continucensusplots, inchiding developed areas, small gardens, ouslyuttering their begging calls.These data were conarid city parks (mean densityof TawnyOwl territories --- sistent to those in non-Mediterranean zones (Southern 3 0/kin '•) and l?om 1989-99 in three suburbanplotsof 1970,Wendland 1984). Data gatheredin earlyMay or in Rome, including open land and deciduous woodland Augustwere included only if a control visitwasmade in June or July. When we were uncertain of the exact numpatches (mean densityof Tawny Owl territories = 5.6/ km• Ranazziet al. 2000). Vegetationin small gardens ber of young that begged due to the many callscontem•ncludedpines (Pinuspinea),cypresses (Cupressus semper- poraneouslyuttered, we omitted them l?om analyses. vzrens),cedars(Cedrusspp.), aswell asisolatedoaks(QuerWe agree with Wendland (1984) that this method can KEYWORDS: rl&wnyOwl; Strix aluco; l,•eedingsuccess; annualfluctuations; Mediterranean areas;urbant•me.

322

DECEMBER 2000

SHORT COMMUNICATIONS

323

1

Breeding

Success Fledglings perI

successful

pair

Figure 1. Annual variationsin the rate of breeding successand mean number of fledglingsper successfulpair in

tlrbanRome.Numberof breeding attempts studiedaregivenin parentheses. result in small errors, but it allowscheckingof yotmg in many natural cavitynestsin a relativelyshort time. The breeding successwasassessed using the following indexes: (1) breedingsuccess (at leastone youngfledged); (2) number of fledglingsper successful pair; and (3) number of fledglingsper breeding pair. All data are presentedas mean +SD. Meteorological data were acquired from a weather stationwithin the studyarea. The rate of breed]ng success was comparedby Xe contingencytablesincluding numbers of successful and failed pairs.Parametric testswereusedwhen data showeda normal frequency distribution. Comparisonsbetween different habitatsor studyareaswere generallyperformed by Student'st-tests and one-wayANOVAs usingyearlymeansand SDs. For somebibliographicdata sets,only meanswere available, so we could not test for differencesamong years.A minimum probabilitylevel of P < 0.05 wasacceptedand all testswere two-tailed.Statisticalanalyseswere performed by STATISTICA 4.5 and PRIMER 1.0 PC packages. RESULTS

Out of a total of 311 breedingattemptsstudiedfrom 1986-99, in urban plots 119 (59.5%, N = 200) failed, 37 (18.5%) produced 1 fledgling,24 (12.0%) produced2 fledglings, 16 (8.0%) produced 3 fledglings, and 4 (2.0%) produced 4 fledglings. In suburban plots, 57

breeding attempts (51.3%, N = 111) failed, 26 (23.4%) produced 1 fledgling, 20 (18.0%) produced 2 fledglings, and 8 (7.2%) produced 3 fledglings.No significantdifferencesin any breeding parameterswere found among yearsin both urban (breedingsuccess: Xe= 9.9, df = 13, P = 0.703; mean number of fledglings per successful pair: b•,67 = 1.4, P = 0.177; and mean number of fledglings per breeding pair: F•,•86 = 1.5, P = 0.126; Fig. 1) and suburbanplots (breedingsuccess: Xe = 9.3, df = 10, P = 0.501; mean number of fledglings per successful pair: F•0,43= 1.4, P = 0.177;and mean number of fledglings per breeding pair: F•0,•00= 0.8, P = 0.636; Fig. 2) Differencesin breedingparametersbetweenthe two hal> itats studiedwere also not significant (breeding success: Xe = 1.6, df = 1, P = 0.204; mean number of fledglings per successful pair: te• = 1.4, P = 0.187; and mean number of fledglingsper breedingpair: res= 0.1, P = 0.937). The mean number of fledglings per breeding pair recorded in suburban plots of Rome was comparable to those

observed

in similar

habitats

of Berlin

Grunewald

(t30-- 0.7, P = 0.464) and Oxford Wytham Wood (tee= 0.5, P = 0.597), but lower than those observed in mixed woodlands of Cote d'Or (t•0 = 5.1, P < 0.001) and farmlands of Hradec Kralowe (t]3 = 6.6, P < 0.001), where

Breeding Success

......

Fledglings per successful

pair

Figure 2. Annual variationsin the rate of breeding successand mean number of fledglingsper successfulpair in suburban Rome. The number of breeding attempts studied are given in parentheses.

324

SHORT COMMUNICATIONS

all Tawny Owl pairs bred in nest boxes.In northern European areas (see Table 1 for statisticsand references), significant differences among yearsin the rate of breeding successwere observed over a long-term period in

VOL. 34, No. 4

rope. Both harsh weather conditions and high levelsof competition with other predators may increase fluctuations in population densityaswell asin reproductiveoutput (Kostrzewa and Kostrzewa 1990, SelSs 1998). In both Berlin (X2 = 1382.9, df = 20, P < 0.001) and OxRome, nighttime temperatures throughout the nesting ford (X9 = 57.2, df = 12, P < 0.001), but no differences seasonare generallyhigher than 10øC.From March-May were observed in Cote d'Or (X9 = 11.7, df = 9, P = 1984-98, the averagetemperaturewas 10.7 _+3.1øC(N 0.233) and Hradec Kralowe (X9 = 7.7, df = 3, P = 0.068), = 15 years),while during the post-fledgingperiod, it was probably due to the limited number of study years.As often higher than 20øC (average temperature in June for the productivityof owl pairs, we could test for differ- 1984-98 = 17.9 _+1.2øC,N = 15 years).Also rainfall was encesamong yearsonly in two areas,Hradec Kraloweand generallylow in both spring and early summerwith rainKielder. In Kielder, the mean number of fledglingsper fhll in March-June 1984-98 averaging189.9 +_60.8 mm breeding pair showed a significant difference among (N = 15 years). This provided good weather conditions years (t148 = 4.4, P < 0.001). In Hradec Kralowe, the for rearing young and reducing energy requirements difference in productivityamong yearswasnot significant when nestlingswere growing (Gil-Delgado et al. 1995). (ts0 = 0.9, P = 0.350), again probablydue to the small As alreadyobservedfor some raptors in cities (Sodhi et sample considered (N = 183 breeding attemptsobserved al. 1992, Tella et al. 1996), competitionlevelswith other during four consecutiveyears). predators were substantiallyreduced in the study area, where only the Eurasian Kestrel, the Little Owl (Athene DISCUSSION noctua),and the red fox (Vulpesvulpes)reachedrelatively The breeding successof Tawny Owls in Rome showed high densitiesin some Tawny Owl habitats. Therefore, weak annual fluctuationswhen compared with those of both mild weather conditionsand low level of trophlc some northern populations, although the overall pro- competition with other predators may have further acduction of young did not vary significantlyamong differ- counted for the long-term stability of the Tawny Owl ent populations nesting in natural cavities.In northern breeding successin Rome. Europe, significant fluctuations in breeding parameters are specificallylinked to the abundance of woodland ro- RESl)MEN.-•Lareproducci6n de Strix alucofue estudiada dents, which make up the main part of the Tawny Owl desde 1984-99 yen 1989-99 en la Roma urbana y subdiet. In low rodent-years,breeding success is significantly urbana respectivamente.El exito de anidaci6n y el nulower than in high rodent years (Southern 1970, Wend- mero medio de pichonespor parejasexitosasy por paland 1984, Petty 1989), and alternativeprey,suchasbirds rejas en reproduccion no tuvo variacionessignificativas and amphibians,increasein the diet (Baudvin 1990, Pies- en un periodD de 15 y 11 aftos en habitatsurbanosy mk and Dusik 1994). By contrast,as rodent fluctuations suburbanosrespectivamente.Comparado con las poblaare weaklyobservedin Mediterranean areas (Rizzo et al. cionesdel norte de Europa, los factoresprincipalesque 1993, Gil-Delgado et al. 1995) and their abundance gen- afectan la estabilidadde la poblaci6n reproductivade erally decreasesalong an urban gradient (Galeotti 1994), Strixalucoen Roma pueden estar relacionadoscon la deTawny Owl reproduction in Rome should be lessaffected pendenciade roedoresen bosques,los cualestienen una by this factor. In fact, in Mediterranean urban habitats as disponibilidad limitada en las fireas urbanas mediterr•well as in most coastaland arid hilly areas, small mam- neas. Tambien es importante su alta dependencia de mals are a minor component of the diet which is com- fuentes alternas de alimento, tales como avesy geckos. posed mainly of birds and insects,especiallybeetles, as Las condicionesclim•ticas durante la primaveray el verwell as geckos,bats, frogs, and snails (Manganaro et al. and son favorablespara la cria de juveniles,permitiendo 1999, Ranazzi et al. 2000). The availabilityof these prey una considerablereducci6n de energia requerida en la throughout the breeding season(Capula et al. 1993, Riz- estaci6nreproductiva. Existe tambien una limitada comzo et al. 1993, Gil-Delgado et al. 1995, Manganaro et al. petencia por cornidacon otros depredadores. 1999) allows Tawny Owls to avoid concentrating their [Traducci6n de C6sarM•rquez] predation on few mammal species,probably providing young with a comparableamount of prey each year.The ACKNOWLEDGMENTS use of alternativeprey suchasinsectsand stablebreeding V. Penteriani, S. Redpath, and P. Sunde revised the successhave been observed in Mediterranean populamanuscriptproviding useful commentsand criticism.We tions of the Eurasian Kestrel (Falco tinnunculus), a rodentare grateful to R. Ranazzifor his invaluablehelp in fieldeating raptor that, in northern Europe, showssignificant work. Thanks are also due to F. Mangianti (Ufficio Cenvariationsin its reproductive output due to cyclicfluctu- trale di Ecologia Agraria, Roma-CollegioRomano) for ations of its main prey (Rizzo et al. 1993, Gil-Delgadoet providing meteorological data. We appreciate the imal 1995, Piattella et al. 1999). On the other hand, other provemenu made in English usagebyJ.N. Westthrough factors affecting Tawny Owl reproductive output in the Associationof Field Ornithologists'program of edinorthern Europe are probably reduced in southern Eu- torial assistance.

DECEMBER 2000

S•ORT

COMMUNICATIONS

326 LITERATURE

SHORTCOMMUNIGATIONS

VOL. 34, NO. 4

CITED

PLESNIK, J. ANDM. DUSIK.1994. Reproductiveoutput of the TawnyOwl Strixalucoin relation to smallmammal dynamicsin intensivelycultivated farmland. Pages chouette hulotte Strix alucoen Bourgogne (France). 531-535 in B.-U. Meyburg and R.D. Ghancellor Uccelli d 7talia 15:30-38. [EI)s.], Raptor conservationtoday. World Working CAPUIA,M., L. LUISELLI, ANDL. RUGIERO.1993. ComparGroup on Birds of Prey and Owls and Pica Press,Berative ecologyin sympatricPodarcismuralisand Podarcis lin, Germany. sicula(Reptilia: Lacertidae) from the historicalcentre RANAZZI, k., A. MANGANARO,R. RANAZZI,AND k. SALVATI. of Rome: what about competition and niche segre2000. Density,territory size,breedingsuccess and diet gation in an urban habitat?Boll. Zool.60:287-291. of a TawnyOwl (Strix aluco)population in a MediterGALEOTTI,P. 1994. Patternsof territory sizeand defence ranean urban arean (Rome, Italy). Alauda 68:133level in rural and urban TawnyOwl (Strix aluco)pop143. ulations.J. Zool.,London234:641-658. Rizzo, M.C., L. MIGLIORE, AND B. MASSA.1993. Insects, GIL-DELGADO, J.A., J. VERDEJO, ANI) E. BARB),.1995. Nessmall mammalsand breeding performanceof farmfling diet and fledgling production of Eurasian Kesland populations of the Common Kestrel (Falcotintrels (Falcotinnunculus)in EasternSpain.J. RaptorRes. nunculus).Pages11-18 in M.K. Nicholls and R. Clarke

BAUDVIN, H. 1990. Bilan de 10 annees d'etude

sur la

29:240-244.

JEDRZEJEWSKI, W., B.JEDRZEJEWSKA, K. ZUB,AJ. RUPRECHT, ANDC. BYSTROWSKI. 1994. Resourceuseby TawnyOwls Strix aluco in relation

to rodent

fluctuations

[EI)s.], Biologyand conservationof smallfalcons.Proceedings of the 1991 Hawk and Owl Trust Conference, The Hawk and Owl Trust, London, U.K.

in Bialow-

SEL•S,V. 1998.Doesfood competitionfrom red fox (Vulieza National Park, Poland.J. Avian Biol. 25:308-312. pesvulpes)influencethe breedingdensityof Goshawk KOSTRZEW^, A. ANI)R. KOSTRZEWA. 1990. The relationship (Accipitergentilis)?Evidence from a natural experiof spring and summer weather with density and ment. J. Zool.,London246:325-335. breeding performance of the Buzzard Buteo buteo, SODHI,N.S., P.C. JAMES,I.G. WARKENTIN, ANDL.W. OLIGoshawkAccipiter gentilisand KestrelFalcotinnunculus. PHANT.1992. Breeding ecologyof urban Merlins (FalIbis 132:550-559. cocolumbarius). Can.J. Zool.70:1477-1483. MANGANAROA., L. SALVATI,S. FATTORINI, AND L. RANAZZI. SOUTHERN, H.N. 1970.The naturalcontrolof a popula1999. Predation on geckosby urban TawnyOwls (Strix tion of Tawny Owls ( Strix aluco).J. Zool.,London162. aluco). Avocetta23:73-75.

197-285.

PENTERIANI, V. 1997. Long-term study of a Goshawk TELLA,J.L., E HIRALDO,J.A. DON•g, ZAR-SANCHO, ANDJJ breeding population on a Mediterranean mountain NEGRO.1996. Costsand benefits of urban nestingin (Abruzzi Apennines,central Italy): density,breeding the LesserKestrel.Pages53-60 in D. Bird, D. Varland, performance and diet. J. RaptorRes.31:308-312. and JJ. Negro [EI)s.], Raptors in human landscapes. PETTY,SJ. 1989. Productivityand densityof TawnyOwls Raptor Research Foundation, Academic Press, London, U.K. Strixalucoin relation to the structureof a spruceforest in Britain. Ann. Zool. Fenn. 26:227-233. WENDLAND, V. 1984. The influence of prey fluctuations PIATTELLA, E., L. SALVATI,A. MANGANARO,AND S. FATTOon the breeding successof the TawnyOwl Strixaluco RINI. 1999. Spatial and temporal variationsin the diet of the Common Kestrel (Falco tinnunculus) in urban Rome, Italy.J. RaptorRes.33:172-175.

Ibis 126:284-295.

Received31 December 1999; accepted22July 2000