Owl Occurrence and Calling Behavior in a Tropical Rain Forest

8/20/2019 Owl Occurrence and Calling Behavior in a Tropical Rain Forest

1/8

j Raptor Res. 5 2) :107-114¸ 2001 The Raptor Research Foundation, nc.

OWL OCCURRENCE AND CALLING BEHAVIOR IN ATROPICAL RAIN FOREST

PAULA . ENRiQUEZ OCHA AND . Lms RANGEL-SALAZAR ,2Programa egional n Manejo de Vida Silvestre, niversidad acional, Apartado 1350-3000, Heredia, Costa Rica

ABSTRACT.--We stimated owl occurrence number of individuals per km of trail/night) from spontaneouscalls and responses o broadcast ocalizations, uring 12 repetitions f each method over 3 2-km trails romApril-September 995 at La Selva Biological tation, Costa Rica. We found Vermiculated creech-Owls Otusvermiculatus, .75 owls/km) and Crested Owls Lophomix ristata, .42owls/km) o be common, nd MottledOwls (Strix virgata, .32 owls/km) and Black-and-white wls (S. nigrolineata, .10 owls/km) to be uncommon.Spectacled wls Pulsatrixperspidllata, .06 owls/km) and Central American Pygmy-Owls Glauddiumgriseiceps,0.03 owls/km) were rare. When compared o earlier reports, our results suggested hat owl occurrence has

changed n recent years at La Selva Biological Station. Broadcast ocalizations ignificantly ncreased hedetections f only Mottled Owls. Detections f all species ncreased when llumination was ess, lthough hiswas statistically ignificant nly or Crested nd Mottled Owls. Three species ere significantly ssociated ithmoon phase. Owl detections ended o be greater during darker periods when the moon was approximatelyfull (either before moon rise or due to cloud cover). Significantly more Vermiculated Screech-Owls eredetected away rom the summer solstice, hile significantly ewer Crested Owls and Spectacled Owls weredetected away rom the solstice. etections decreased with time after official sunset, lthough his was statis-tically significant only for Spectacled Owls. Our results suggested hat changes n owl calling activity wascorrelated with other species ctivity, he season f the year, and environmental onditions.

KEYWORDS: Vermiculated creech-Owl; tus vermiculatus; Crested wl; Lophostrix cristata; Mottled Owl;Strix virgata; Black-and-white wl;Strix nigrolineata; Central American ygmy-Owl; laucidium griseiceps;Spectacled wl;Pulsatrix perspicillata; ocalizations; ropical orests; osta Rica.

Ocurrencia de Bfihos y Comportamiento del Llamado en un Bosque Tropical Lluvioso

R•sUmEN.--Estimamos a ocurrencia de bfihos (nfimero de individuos por km de transecto por noche)por medio de dos metodos, lamados spont•neos provocaci6n udiriva, con 12 repeticiones or cadametodo en 3 transectos e 2 km cada uno, en La Estaci6n Bio16gica La Selva, Costa Rica de abril aseptiembre e 1995. Otus vermiculatus llamados .75 bfihos/km y respuestas .89 bfihos/km) y Lophostrixcristata 1.42 y 1.39) fueron especies omunes, trix irgata 0.32 y 0.71) y S. nigrolineata 0.10 y 0.19) fueronpoco comunes, y Pulsatrix erspicillata 0.06 y 0.03) y Glaucidium risceiceps 0.03 y 0.04) fueron especiesraras. Cuando realizamos omparaciones on reportes previos de ocurrencia de bfihos en La Selva, nues-tros resultados ugieren ue esta ocurrencia ha estado cambiando n La Selva. E1metodo de provocaci6naudiriva ncrement6 significativamente as detecciones olo para S. virgata. as detecciones ocales e todaslas especies ncrementaron cuando a iluminaci6n disminuy6, aunque ue significativa olo para L. cristatay S. virgata. as vocalizaciones e tres especies stuvieron sociadas ignificativamente on la fase unar. Ladetecci6n de los bfihos ncrement6 durante periodos obscuros n dias cercanos luna 11ena antes quela luna saliera o en dias nubiados). Por otro lado, O. vermiculatus ue significativamente etectado m/rs endias ejanos al solsticio e verano, mientras que L. cristata P. perspicillata ueron detectados m/rs en diascercanos l solsticio e verano. Las detecciones e las especies e bfihos decrecieron on el tiempo despuesdel atardecer. Nuestros esultados ugieren cambios en las poblaciones e bfhos y que la actividad vocalesta correlacionada on la actividad de otras especies, emporada y ciertas condiciones mbiental.

[Traducci6n de autores]

• Present address: Department of Animal Science, Faculty of Agriculture, University of British Columbia, Vancouver,BC V6T 1Z4 Canada and Departamento de Ecologla y Sistematica, 1 Colegio de la Frontera Sur, Aparmdo Postal63-29290, San Crist6bal de las Casas, Chiapas, Mexico.2 Present address: Centre for Applied Conservation Biology, Faculty of Forestry, University of British Columbia, Van-couver, BC V6T 1Z4 Canada.

107


2/8

108 EN•QUEZ ROCHa )•,4I) P,•NGEL-S• VOL. 35, NO. 2

Table 1. Mean owl occurrence or six owl species at La Selva Biological Station, Costa Rica, per km of trail/nightand total occurrences ounted or spontaneous alls and responses o call broadcasts ver 12 repetitions on three 2-km trails from April-September 1995.

METHOD

CALLING RESPONSES

SPECIES OWLS/km TOTAL OWLS/kin TOTAL D • P

Vermiculated Screech-Owl 0.75 54 0.89 65 0.866 0.45Crested Owl 1.42 103 1.39 100 0.57 0.89

Spectacled Owl 0.06 4 0.03 2 1.15 0.14Central American Pygmy-Owl 0.03 3 0.04 3 1.15 0.14Mottled Owl 0.32 23 0.71 47 1.44 0.03Black-and-white Owl 0.10 7 0.19 14 0.86 0.45

• Kolmogorov-Smirnov wo-sample est.

An understanding of the dynamics of raptorpopulations is fundamental in their conservation(Newton 1979, Thiollay 1989). However, imitedpopulation data exist for owls (Mikkola 1992).Most species f owls nhabit tropical forests, but weknow very little about their natural history andecology (Clark et al. 1978, Newton 1979, Mikkola1992). Indeed, further study s essential or theirconservation n tropical areas where forests are de-clining.

Most owl species re active at night; therefore, itlS easier o hear than see hem, and counting theirvocalizations as generally been the most reliableway of detecting them (Springer 1978, Forsman1983, Smith 1987). Two methods based on vocali-zations have commonly been used to survey andestimate occurrence and densities of nocturnal

birds of prey. The first method is based on countsof spontaneously iven calls (spontaneous alling)by owls and the second method uses counts of re-sponses y owls to tape-recorded playbacks Fullerand Mosher 1981, Marion et al. 1981, Mosher etal. 1990). Both methods have been used extensive-ly in surveys f single species n temperate habitats(Smith et al. 1987, Mosher et al. 1990), but rarelyin tropical forests or tropical species of owls (Ger-hardt 1991).

Tropical rain forests surrounding La Selva Bio-logical Station in Costa Rica have come under in-creasing pressure from people in recent years(Braker and Greene 1994, McDade et al. 1994),and the negative mpacts have resulted n declinesin some bird populations (Levey and Stiles 1994).To enable documentation of future declines in owl

populations, we conducted research to develop a

standardized method of detecting the occurrenceof owl species at La Selva Biological Station bycomparing the two methods used to estimate owloccurrence and to identify the environmental fac-tors that might influence owl vocal activity.

STUDY AREA AND METHODS

La Selva Biological Station is located in SarapiquiCounty, Heredia Province, Costa Rica (10ø26'N,83ø59'W). La Selva adjoins Braulio Carrillo National Parkto the south and borders agricultural and pastureland othe north. The station covers 1513 ha at the ecotone be-

tween tropical premontane and very humid tropical or-est life zones (Hartshorn 1983). Elevations ange from35-150 m. Weather conditions are very humid with 4000-4500 mm of annual rainfall, most of which falls n June-July, and November-December. Although there is no pro-nounced dry season, less rain falls from Februarythrough April. March is the driest month of the year,August he warmest (mean temperature = 27.1øC), andJanuary the coldest (mean temperature = 24.7øC)(McDade et al. 1994).

We estimated he occurrence of owl species y record-ing spontaneous alling and responses o playback f owlcalls. Thirty permanent survey stations were spread at200-m intervals over 3 2-km trails (10 stations per trail)Each station was surveyed on 24 different nights fromApril-September 1995 twice per month (N = 12 times)counting number of owls detected rom spontaneous o-calizations without previous conspecific broadcasting(hereafter calls ), and twice per month (N = 12 times)counting owls ollowing broadcasts f tape-recorded owlvocalizations (hereafter responses ). Travel time be-tween survey tations anged from 5-8 min. Surveys tart-ed during local twilight and continued br approximately5.5 hr. Local twilight occurred fi•om about 1830-1900 Hduring the period of the study.

To determine calls, we walked our survey rails andlistened or owl calls for 10 min at each survey station,recording all individual owls hat we heard or saw duringthis period (Anderson et al. 1985, Kocheft 1986, Fuller


3/8

JUNE 2001 TROPICAL OWL OCCURRENCE 109

2 3

0.15 0.15

,o.,la o., n n.05 0,0501' .... O,

iiiA M J J A $ A M J ,J A

Months

Figure 1. Average occurrence per km of trail/night, bymonth from April-September 1995, for (a) VermiculatedScreech-Owl, b) Crested Owl, (c) Spectacled Owl, (d)Central American Pygmy-Owl, (e) Mottled Owl, and (f)Black-and-white Owl based on spontaneous alls (white

bars) and responses o broadcasts black bars) at La SelvaBiological Station, Costa Rica. Error bar represents onestandard error. Table 1 shows the total count of each

species over 19 repetitions by each method over 3 9-kmtrails.

and Mosher 1987). To avoid counting he same ndivid-ual, we estimated ts location by compass riangulation(Bell 1964). While one of us triangulated an owl, theother person always emained at the survey station to re-cord other individuals (same or different species) vocal-izing within the 10-min period. We located ndividuals ona map of La Selva (scale 113 000).

Three days after counting calls, we used the same railand survey stations o conduct a response count. Prior toour study, we recorded locally the characteristic ocali-zations of Vermiculated Screech-Owl (Otus vermiculatus),Crested Owl (Lophostrix ristata), Spectacled Owl (Pulsa-trix perspicillata), Mottled Owl (Strix virgata) and Black-and-white Owl (S. nigrolineata). We used a Uher 4000RLtape recorder and unidirectional Electro-Voice micro-phone to obtain vocalizations. We used a 3-min editedsequence of each species or playback and broadcast twith a Sony CFM-1605 cassette ecorder. We divided eachtransect nto two equal segments, hen randomly selected

the playback sequence of the five species nd played hecall of only one species at each station (Kocherr 1986).Any one species' call was broadcast at only two stationsalong any one trail during any one night. A samplingperiod per station consisted f 3 min of broadcasting allfollowed by 7 rain of silence and listening. During a sin-

gle sampling period, we recorded responses f any owlto playback. We determined also he location of respond-ing individuals by triangulation. Although the CentralAmerican Pygmy-Owl Glaucidium risceiceps) as not in-

cluded in our original protocol, this species alled spon-taneously nd responded o playback of interspecific callsduring our study. Therefore, we included it in our resultsand analyses.

We recorded the following temporal and environmen-tal conditions at each survey station: hour (number ofhours after the official local sunset), days rom solstice(the absolute number of days before or after the summersolstice, fter Cooper 1981), moon phase (0 during pe-riod around new moon and 4 during period around fullmoon, with gradations n between, after Palmer 1987),illumination (from 0 for totally dark to 3 for as bright asnight can be, after Palmer 1987), cloud cover (from 0for no cloud to 3 for totally overcast, fter Palmer 1987),and precipitation (0 for no rain to 3 for heavy rain). Wealso considered he mean temperature (maximum plusminimum divided by two) for the survey date at the SelvaBiological Station.

Our sampling unit was any single trail surveyed anysingle night, and we expressed ur owl occurrence ndexas the average number of individuals sighted or heard ina night per km of trail surveyed (Brower et al. 1990)The total occurrence value was the total number of in-

dividuals of any species etected with one of the two sur-vey methods over the 6 mo of study (i.e., the total overthe 12 repetitions by each method over 3 2-km trails)We reduced pseudo-replication y combining results orany one month.

We considered a species o be rare if the overall meanestimated occurrence was 0.85owls per night/km of trail. We then calculated monthlyaverages or each species o evaluate changes n owl de-tectability throughout the study period. Fluctuations nmonthly averages n spontaneous calls and responseswere tested using Kolmogorov-Smirnov goodness-of-fit.For significant fluctuations, we determined significantcategories using 95% Bonferroni confidence intervals(Neu et al. 1974). Occurrence distributions esulting

from the two sampling methods were compared with Kol-mogorov-Smirnov wo-sample ests. To identify environ-mental variables that may have correlated with calls andresponses, we used Spearman rank correlation analysis(Sokal and Rohlf 1981).

RESULTS

Vermiculated Screech-Owls and Crested Owls

were common (Table 1). Mottled Owls and Black-and-white Owls were uncommon, while SpectacledOwls and Central American Pygmy-Owls ere rare.Vermiculated Screech-, Crested, and Mottled Owls

were recorded in all study periods (Fig. 1). Signif-icant monthly variation in detections occurred forthe Vermiculated Screech-Owl, Spectacled Owl,and Central American Pygmy-Owl Table 2). Ver-miculated Screech-Owls were detected less often


4/8

110 ENRiQUEZ ROCHA AND RANGEL-SALAZAR VOL. 35, No. 2

Table 2. Kolmogorov-Smirnov D) goodness of fit values or fluctuations n mean monthly counts of six owls speciesusing two survey methods rom April-September 1995 at La Selva Biological Station, Costa Rica.

MONTHLY MEAN

SPECIES METHOD ( + SD) D

Vermiculated Screech-Owl

Crested Owl

Spectacled Owl

Central American Pygmy-Owl

Mottled Owl

Black-and-white Owl

Spontaneous alls 9.0 _+ 5.1 0.317'Responses 10.7 __ 4.8 0.217Spontaneous alls 17.2 + 5.3 0.279Responses 16.7 _ 5.2 0.182Spontaneous alls 0.7 _+ 0.8 0.293Responses 0.3 _+ 0.5 0.407**Spontaneous alls 0.3 ___ .5 0.407**Responses 0.5 - 0.8 0.390**Spontaneous alls 3.8 _ 2.4 0.189Responses 7.8 +_ 2.7 0.200

Spontaneous alls 1.2 -+ 1.2 0.224Responses 2.3 +-- 1.9 0.251

* P --< 0.05.

** P < 0.01.

during June than in most other months. Specta-cled Owls were detected only from April-June,while detections of the Central American Pygmy-Owl were scattered (Fig. 1, Table 2).

The broadcast survey method provided signifi-

cantly more detections of Mottled Owls (Table 1).The two survey methods provided similar resultsfor the Vermiculated Screech-Owl and the Crested

Owl. Detections of other species were too few tomeaningfully compare or the two methods.

Vermiculated Screech-Owls were positively asso-ciated with the number of days before or after thesummer solstice Tables 3, 4, Fig. 1). Generally oth-er species were less detected away from the sum-mer solstice Fig. 1). For the spontaneous all tech-nique the decrease away from the solstice was

significant or Crested Owls and Spectacled Owls(Table 3), while with the response to broadcasttechnique the decrease away rom solstice was sig-nificant only for Mottled Owls (Table 4).

For every species, he number of detections wasnegatively associated with the amount of illumina-tion, though this relationship was statistically ig-nificant for only Mottled Owls (Table 3) and Crest-ed Owls (Table 4). Detections were positivelyassociated with moon phase, significantly so forVermiculated Screech-Owls, Crested Owls and Mot-

tied Owls (Table 3). The detection rate was highduring dark periods when the moon phase wasnearly full, either before the moon rose over thehorizon or when clouds obscured he moonlight.

Detections were generally negatively associated

with the number of hours since sunset, but this

relationship was significant only for SpectacledOwls with the spontaneous call technique (Table3).

Our results ndicated that some of the owl spe-cies have increased in numbers at La Selva over the

past 35 years (Table 5). In our study, VermiculatedScreech-Owls and Crested Owls were common, incontrast with previous ecords at La Selva.

DISCUSSION

The calling activity of some owls appears o beaffected by moon cycles and light levels (Smith etal. 1987). At La Selva, moon phase and days romsummer solstice ffected vocal activity of some butnot all of the owl species we encountered. Illumi-

nation had a negative effect with the calling activityof all owl species occurring more during darknights and during dark periods of otherwise brightnights. This might seem to be contradictory withthe positive association f vocalizations with moonphase but most of these responses ccurred duringdark moments around the full moon before the

moon appeared in the horizon and increased l-lumination, or when clouds obscured the moon.By comparison, Northern Saw-whet Owls (Aegoliusacadicus), Boreal Owls (A. funereus), and Western

Screech-Owls Otus kennicottii) ave been reportedto respond more during bright nights and fullmoon periods (Palmer 1987). Northern SpottedOwls (Strix occidentalis) esponded more duringquarter ebb (last quarter) and new moon (Ganey


5/8

JUNE 2001 TROPICAL OWL OCCURRENCE 111

1990), whereas the response of Eastern Screech-Owls (Otus asio) to broadcasts has varied littleacross nvironmental conditions Carpenter 1987,

Smith 1987). The responsiveness f owls o broad-casts of conspecific alls apparently varies by spe-cies and geography

Vermiculated Screech-Owls, Central AmericanPygmy-Owls, nd Spectacled Owls exhibited signif-icant variability of vocal activity mong months. Wewere unable to correlate calling periods withbreeding seasons, because ittle breeding infor-mation is available or most tropical owl species.Nonetheless, the differences that we observed in

vocal activity of these species may have been asso-

ciated with their breeding phenology. ndeed, un-derstanding he year-long calling behavior of theseowls s important for understanding heir breedingphenology and to document population trendsthrough repeated monitoring (Mathisen andMathisen 1968).

For most species n our study, survey results didnot differ between the two techniques. MottledOwls were more frequently detected using callplayback techniques than by using spontaneouscalling. Gerhardt (1991) reported similar results n

Guatemala. In North America, Eastern Screech-Owls and Elf Owls (Micrathene whitneyi) have alsobeen recorded more frequently using playbacktechniques (Johnson et al. 1981). Broadcast vocal-ization has been an effective and useful method for

counting temperate owls (McGarigal and Fraser1985, Bosakowski t al. 1987, Zubergoitia nd Cam-pos 1998). In contrast, Great Horned Owls (Bubovirginianus) were detected more often from spon-taneous calls than from responds to broadcasts(Springer 1978). Intra- and interspecific ompeti-

tion could play a major role in response patternswhile using the playback method (Enr/quez andRangel-Salazar 997).

Slud (1960) mentioned only one record for theVermiculated Screech-Owl and recorded the Crest-

ed Owl as rare at La Selva Biological Station. Sub-sequently, anzen (1991) reported Crested Owls asoccasional and Vermiculated Screech-Owls as un-

common. Spectacled Owls and Central AmericanPygmy-Owls ere previously eported as commonby Slud (1960) and later as uncommon (Janzen

1991, Levey and Stiles 1994). During our study, werecorded so few individuals of both species hat weclassified them as rare. Mottled Owls were abun-

dant or common, but we found them to be uncom-mon. Barn Owls (Tyt0 alba) occur at La Selva but


6/8

112 ENRiQUEZ ROCH• •2ND R•NGEt.-S^•Z^R VOL. 35, No. 2

Table 4. Correlation coefficients of responses y four owl species o broadcasts f calls with respect o temporal andenvironmental variables rom April-September 1995 at La Selva Biological Station, Costa Rica.

SPECIES

BLACK-AND-WHITE

VARIABLES N) V. SCREECH-OWL CRESTED OWL MOTTLED OWL OWL

Days rom solstice 18) 0.325* 0.209 -0.402'* -0.049Hour (5) -0.112 -0.133 -0.923 0.137Illumination (4) -0.854 -0.984* -0.841 -0.822Cloud cover (4) -0.023 0.241 -0.315 -0.513Moon phase (4) 0.283 0.413 0.353 0.389Temperature 8) 0.198 0.487 0.345 0.169Precipitation 4) - 0.136 - 0.253 - 0.294 - 0.302* P < 0.05.

** P < 0.01.

were not recorded near our survey rails; herefore,we did not estimate any occurrence or this species.We did not record Andean Pygmy-Owls Glauci-dium jardinii) or Striped Owl (Asio clamator),though both species have occasionally been re-corded at La Selva during Christmas Bird Counts(GBG 1985-94). Previous owl records at La Selvawere part of attempts o ascertain occurrence of allavifauna without specifically argeting owls, so ourstudy should have detected owls at a higher ratethan previous studies.

The Crested Owl has been reported to be asso-ciated with mature forest and mature second-

growth forests for nesting and roosting (Marcot1995). However, it apparently tolerates imited de-

forestation (Hume 1991, K6nig et al. 1999). Crest-ed Owls have apparently ncreased n abundanceat La Selva, whether because of breeding successor immigration of individuals displaced by cuttingof surrounding forests.

The Spectacled Owl is associated with dense or-est with old trees, and may also occur in forest edg-es and plantations (K6nig et al. 1999). It was re-corded as common by Slud (1960) and rare in ourstudy. Although it tolerates more deforestationthan the Crested Owl (Hume 1991), our resultssuggest hat it has apparently decreased t La Selva.

The Mottled Owl has been reported to be themost abundant owl in Neotropical rainforests andhas a wide distribution range (Gerhardt et al.

Table 5. Owl species bundance A = Abundant, C = Common, U = Uncommon, R = Rare, O = Occasional, 2= only one record, -- = no report) over three decades t La Selva Biological Station, Costa Rica.

STUDIES 1

S T J L AND S CBC E and RSPECIES (1960) (1990) (1991) (1994) (1985-94) (1995)

Barn Owl -- U U U U 09Vermiculated

Screech-OM 02 U U U UG GCrested OM R O O UG UG G

Spectacled Owl G U U U U RCentral American

Pygmy-OM C U U U U RAndean Pygmy-Owl .... O --Motfled OM A G G UC UG UBlack-and-white Owl -- R R R R U

Striped OWl .... O --1S = Slud (1960), T = Taylor (unpublished ata), J = Janzen (1991), L and S = Levey and Stiles 1994), CBC = Christmas •rdCounts, La Selva (1985-94), E and R = (1995), this study.


7/8

JUNE 2001 TROmC•L OWL OCCURRENCE 113

1994). Mottled Owls are considered tolerant tosome deforestation, and can be found near forestedges, semi-open areas, and secondary mature for-

ests (Stiles and Skutch 1989). Nonetheless, MottledOwl populations may have decreased at La Selvaduring the past 30 years. The Black-and-white Owlwas also uncommon during our study. Unlike theMottled Owl, it is considered o be rare throughoutits range and has been classified as uncommon inCosta Rica (Stiles and Skutch 1989). Black-and-white Owls are associated with dense and mature

forests Marcor 1995), and La Selva might act as arefuge for it.

Our results suggest hat owl occurrence has been

changing at La Selva. Levey and Stiles (1994) pre-dicted changes n populations of birds, and habitatloss has been identified as the principal factor af-fecting populations of tropical raptors (Thiollay1994). In the Neotropics, protected areas such asreserves and national parks often function as ref-uges for many species of wildlife that depend onforested habitats (Terborgh et al. 1990). Specta-cled Owls and Mottled Owls have been thought totolerate forest fragmentation (Hume 1991), butthey appear to have decreased n occurrence at La

Selva. Further studies of the ecology of tropicalowls would be useful n identifying the life historycomponents most affected by human disturbance.

ACKNOWLEDGMENTS

We are grateful to R.P. Gerhardt, T. Will, and D.Johnson for their useful comments and improving ourEnglish. L. Sierra, M. McCoy, J. Ram, and P. Marshallprovided invaluable advice. S.H. Anderson and ananonymous reviewer gave constructive comments on anearly draft of this manuscript, as did C. Crocker-Bedfordon early and late drafts. We also thank B. Young (Orga-

nization fbr Tropical Studies), G. Rabb (Chicago Zoolog-real Society), the Costa Rican Wildlife Foundation, andthe U.S. Fish and Wildlife Service, for financial supportfor this study. The Florida Museum of Natural History,Owl Research Institute, and Idea Wild donated fieldequipment. For their field assistance e thank R. Barran-res and F. Morales.

LITERATURE CITED

ANDERSON, .E., o.J. RONGSTAD, NDW.R. MYrTON. 1985.Line transect analysis of raptor occurrence alongroads. Wildl. Soc. Bull. 13:533-539.

BELL,R.E. 1964. A sound triangulation method for count-ing Barred Owls. Wilson Bull. 76:292-294.

BOSAKOWSKI,., R. SPEI,SER, ND. BENZINGER.987. Dis-tribution, density, and habitat relationships of theBarred Owl in North New Jersey. Pages 135-143 inR.W. Nero, RJ. Clark, RJ. Knapton, and R.H. Hamre

[EDS.], Biology and conservation of northern forestowls. Winnipeg, Manitoba, Canada.

BROKER, .E. ^ND H.W. GRF, •NF,. 994. Population biolo-gy, life histories, abundance, demography, and pred-ator-prey interactions. Pages 244-255 in L.A. McDade,K.S. Bawa, H.A. Hespenheide, and G.S. Hartshorn[EDS.], La Selva, ecology and natural history of a Neo-tropical rain forest. Univ. Chicago Press, Chicago, ILU.S.A.

BROWER, ., J. ZAR, AND C. VON ENDE. 1990. Field andlaboratory methods for general ecology. William GBrown, Dubuque, IA U.S.A.

CARPENTER, .T. 1987. Effects of environmental variableson responses of Eastern Screech-Owls o playback.Pages 77-280 in R.W. Nero, RJ. Clark, RJ. Knapton,and R.H. Hamre [EDs.], Biology and conservation ofnorthern forest owls. Winnipeg, Manitoba, Canada

GLARK, .J., D.G. SMITH, ANDL.H. KELSO. 978. Work•ngbibliography of owls of the world. Tech. $er. 1. Natl.Wildl. Fed., Washington, DG U.S.A.

COOPER, J. 1981. Relative abundance of Georgia capn-mulgids based on call-counts. Wilson ull. 93:363-371

ENRiQUEZ, .L. ANDJ.L. RANGEL-SALAZAR.997. Intra-andinterspecific calling in a tropical owl community. Pag-es 525-532 in J.R. Duncan, D.H. Johnson, and T.HNicholls [EDS.], Biology and conservation of owls ofthe Northern Hemisphere. Gem Tech. Rep. NC-190.USDA Forest Service, Denver, CO U.S.A.

FORSMAN, .D. 1983. Methods and materials for locanngand studying Spotted Owls. Gem Tech. Rep. PNW-162. USDA Forest Service, Denver, CO U.S.A.

FULLER,M.R. ^ND J.A. MOSHER. 981. Methods of detect-ing and counting raptors: a review. Pages 235-246 •nJ. Ralph and M. Scott [EDS.], Estimating the numbersof terrestrial birds. Stud. Arian Biol. No. 6.

-- •ND --.. 1987. Raptor survey echniques. Pag-es 37-65 in B.A.G. Pendleton, B.A. Millsap, K WCline, and D.M. Bird [EDs.], Raptor managementtechniques manual. Natl. Wildl. Fed., Washington, DCU.S.A.

G•NF•Y, .L. 1990. Calling behavior of Spotted Owls •nNorthern Arizona. Condor 92:485-490.

GERHARDT, .P. 1991. Response of the Mottled Owl (C•c-caba virgata) to broadcast of conspecific call. J. FzeldOrnithol. 62:239-244.

--, N. BONILL^, D.M. GERHARDT, ND CJ. FLATTEN.1994. Breeding biology and home range of two C•c-caba owls. Wilson Bull. 106:629-639.

HARTSHORN, .S. 1983. Plants: ntroduction. Pages 118-183 in D.H. Janzen [ED.], Costa Rica natural historyUniv. Chicago Press, Chicago IL U.S.A.

HUME, R. 1991. OMs of the world. Running Press, Phd-adelphia, PA U.S.A.

J^NZF. , D.H. 1991. Historia natural de Costa Rica. Uni-versidad de Costa Rica, San Jos6, Costa Rica.

JOHNSON, R., B. BROWN, L.T. Iq•IOHT, ̂ ND J.M. SIMPSON.1981. Playback recording as a special avian censusing


8/8

114 ENPdQUF. ROCHA ANI) RaNGF. -SAL•ZAP, VOL. 35, NO. 2

technique. Pages 68-75 in J. Ralph and M. Scott[EDS.], Estimating the numbers of terrestrial birds.Stud. Arian Biol. No. 6.

KOCHERT, .N. 1986. Raptors. Pages 313-349 inA. Coop-errider, R.J. Boyd, and H.R. Stuart [EDS.], nventoryand monitoring of wildlife habitat. USDI, Denver, COU.S.A.

KONIG, C., F. WEm•4, AND J-H. BEeKING. 1999. Owls: aguide to owls of the world. Yale Univ. Press, New Ha-ven, CT U.S.A.

LLVEY,DJ. AND F.G. STILES. 994. Birds: ecology, behaviorand taxonomic affinities. Pages 217-228 in L.A.McDade, K.S. Bawa, H.A. Hespenheide, and G.S.Hartshorn [EDS.], La Selva ecology and natural his-tory of Neotropical rain forest. Univ. Chicago Press,Chicago, L U.S.A.

MASCOT, B.G. 1995. Owls of old forests of the world.

USDA Forest Service, Gem Tech. Rep. PNW-GTR-343,Portland, OR U.S.A.

MAP, ON, W.R., T.E. O'MEA•,A, AND D.S. MAEHR. 1981. Use

of play back recording n sampling elusive or secretivebirds. Pages 81-85 in J. Ralph and M. Scott [EDS.],Estimating the numbers of terrestrial birds. Stud. Ari-an Biology o. 6.

MATHISEN, .E. AND A. MATHISEN. 1968. Species abun-dance of diurnal raptors in the Panhandle of Nebras-ka. Wilson Bull. 80: 479-486.

MCGARIGAL,K. AND J.D. FRASER. 985. Barred Owl re-sponses o recorded vocalizations. Condor 7:552-553.

MCDADE, L.A., K.S. BAWA, H.A. HESPENHEIDE, AND G.S.HARTSHORN EDS.]. 1994. La Selva ecology and natu-ral history of Neotropical rain forest. Univ. ChicagoPress, Chicago, IL U.S.A.

MIKI•OLA,H. 1992. Wood owls. Pages 108-140 in J.A.Burton [ED.], Owls of the world. Peter Lowe, Euro-book, Italy.

MOSHER, .A., M.R. FULLER, AND M. KOPENY. 990. Sur-veying woodland raptors by broadcast of conspecificvocalizations. . Field Ornithol. 1:453-461.

NEWTON,. 1979. Population ecology of raptors. PitmanPress, Hertfordshire, U.K.

NEU,C.W., R.R. BYEP, , AND .M. PEEK. 974. A techniquefbr analysis of utilizations-availability ata. J. Wildl.Manage. 38:541-545.

PALMER, .A. 1987. Annual, seasonal, nd nightly varia-tion in calling activity of Boreal and Northern Saw-whet Owls. Pages 162-168 in R.W. Nero, R.J. Clark,R.J. Knapton, and R.H. Hamre [EDS.], Biology andconservation of northern forest owls. Winnipeg, Man-itoba, Canada.

SLUD, P. 1960. The birds of finca La Selva , Costa Rica:a tropical wet forest locality. Bull. Am. Mus. Nat. Hist121.

SUITH,D. 1987. Owl census echniques. Pages 304-307in R.W. Nero, R.J. Clark, R.J. Knapton, and R.H. Ham-re [EDS.], Biology and conservation f northern forestowls. Winnipeg, Manitoba, Canada.

, A. DEVINE, AND D. W^SH. 1987. Censusingscreech owls n southern Connecticut. Pages 55-267in R.W. Nero, R.J. Clark, R.J. Knapton, and R.H. Ham-re [EDS.], Biology and conservation of northern forestowls. Winnipeg, Manitoba, Canada.

SoI•m, R.R. AND FJ. ROHLF. 1981. Biometry. W.H. Free-man, San Francisco, CA U.S.A.

SPRINGER, .A. 1978. Foot surveys ersus owl calling sur-veys: a comparative study of two Great Horned Owlcensusing echniques. nland Bird Banding News 0:83-93.

STILES,G. ANDA. SKUTCH. 989. A guide to the birds ofCosta Rica. Cornell Univ. Press, Ithaca, NY U.S.A.

TEPmORGH, ., S.K. ROBINSON, T.A. PAmrmR, II, C.A.MUNN, ANDN. PIERPONT. 990. Structure and orga-nization of an Amazonian forest bird community. EcolMonogr. 0:213-238.

THIOLLAY, .-M. 1989. Censusing of diurnal raptors in aprimary rain forest: comparative methods and speciesdetectability. . Raptor Res. 23:72-84.

1994. A world review of tropical forest raptorscurrent trends, research objectives and conservationstrategy. Pages 231-239 in B.-U. Meyburg and R.DChancellor [EDS.], Raptor conservation today. WorldWorking Group on Birds of Prey and Owls, Berlin,Germany.

ZUBERGOITIA,. AND L.F. CAMPOS. 998. Censusing owlsin large areas: a comparison between methods. Ardeo-la 4547-53.

Received 1 April 1999; accepted 17 February 2001

Owl Occurrence and Calling Behavior in a Tropical Rain Forest

Documents