Azevedo 2010 Whistles emitted

Whistles emitted by Atlantic spotted dolphins(Stenella frontalis) in southeastern Brazil

Alexandre F. Azevedoa�

Departamento de Oceanografia, Laboratório de Mamíferos Aquáticos e Bioindicadores (MAQUA),Universidade do Estado do Rio de Janeiro, Rua São Francisco Xavier 524, Maracanã,Rio de Janeiro 20550-013, Brazil

L. FlachProjeto Boto Cinza, Rua Sta Terezinha, 531 Muriqui, Mangaratiba, Rio de Janeiro 23870-000, Brazil

Tatiana L. Bisi, Luciana G. Andrade, Paulo R. Dorneles, and J. Lailson-BritoDepartamento de Oceanografia, Laboratório de Mamíferos Aquáticos e Bioindicadores (MAQUA),Universidade do Estado do Rio de Janeiro, Rua São Francisco Xavier 524, Maracanã,Rio de Janeiro 20550-013, Brazil

�Received 20 September 2009; revised 22 December 2009; accepted 11 January 2010�

The whistles of Atlantic spotted dolphins have been studied in a few localities of the North AtlanticOcean and those studies revealed that the species emits whistles within a broad frequency range,with a high number of inflection points and presence of harmonics. In the South Atlantic Ocean,there is no information about the sounds produced by Atlantic spotted dolphins. A total of 1092whistles emitted by free-ranging Atlantic spotted dolphins in Southeastern Brazilian coastal waterswere analyzed. Whistles recorded in this study had a broad frequency range from 1.15 to 23.44 kHz.Whistles without harmonics were frequently emitted �N=701; 64.2%� and those signals with zeroup to two inflection points corresponded to 94% of all whistles. Some differences in whistlecharacteristics �inflection points and duration� were found in relation to areas in North AtlanticOcean and whistles were shorter and with a smaller number of inflection points in Brazil. Whistlesproduced by Atlantic spotted dolphins varied between the two behavioral states in which dolphinswere engaged. Whistles were more frequently emitted when dolphins presented behaviors thatincluded fast movement at surface, prey pursuit, aerial behavior, and physical contact. In thesesituations, whistles were on average longer and had higher frequency parameters than those emittedwhen animals were engaged in slowly and moderate traveling. The findings presented herewithreveal that dolphins modified whistle structures within behavioral states.© 2010 Acoustical Society of America. �DOI: 10.1121/1.3308469�

PACS number�s�: 43.80.Ka �WWA� Pages: 2646–2651

I. INTRODUCTION

Dolphins emit two broad categories of acoustic signals:pulsed and tonal sounds �Popper, 1980; Richardson et al.,1998�. Whistles are narrow-band frequency modulatedsounds used primarily for communication between individu-als �Tyack, 1998�. These signals have been the focus of sev-eral bioacoustic studies on delphinids due to the numerouseasily measured characteristics presented by them. In addi-tion, they have been used for characterization and compari-sons among groups and populations �e.g., Rendell et al.,1999; Bazúa-Durán and Au, 2004; Azevedo and Van Sluys,2005; May-Collado and Wartzok, 2009�.

Atlantic spotted dolphins �Stenella frontalis� are foundin warm temperate to tropical waters in Atlantic Ocean �Jef-ferson et al., 2008�. As most delphinids, this species is ahighly vocal dolphin species that produces echolocation clicktrains, burst pulse sounds, and whistles related to behavioralactivities �Perrin, 2002�. Narrow-band signature whistleshave been reported in Atlantic spotted dolphins both in cap-

a�Author to whom correspondence should be addressed. Electronic mail:

2646 J. Acoust. Soc. Am. 127 �4�, April 2010 0001-4966/2010/1

Downloaded 04 Jan 2012 to 200.128.66.36. Redistribution subject to ASA license

tive �Caldwell et al., 1973� and in the wild �Herzing, 1996�.The whistles of Atlantic spotted dolphins have been studiedin a few localities of the North Atlantic Ocean, especially inBahamas and Gulf of Mexico �e.g., Wang et al., 1995; Lam-mers et al., 2003; Baron et al., 2008�. Those studies revealedthat Atlantic spotted dolphins emit whistles with a high num-ber of inflection points, duration between 0.08 and 2.07 s, aswell as with minimum and maximum frequency average val-ues of 7 and 16 kHz, respectively.

The Atlantic spotted dolphin is the only member of thegenus Stenella that is frequently observed close to shore inBrazil �Moreno et al., 2005�. Group size is relatively smallfor Stenella species, being most of the times smaller than 50individuals �Moreno et al., 2005�. Despite being commonlyobserved, there is no information available about acousticsignals of the species in the area. In this study, we describethe whistle acoustic structure of free-ranging Atlantic spotteddolphins and provide insights on whistle geographical varia-tions by comparing our results with previous studies.

II. METHODS

Acoustic recordings of underwater sounds produced

© 2010 Acoustical Society of America27�4�/2646/6/$25.00

or copyright; see http://asadl.org/journals/doc/ASALIB-home/info/terms.jsp

Bay �23°05�–23°14� S; 44°05�–44°23� W�, southeast-ern Brazil, �Fig. 1�, during one day in April 2007, two daysin October 2007, and one day in December 2008. All surveyswere carried out under similar weather conditions �Beaufortsea states �2�, in a small inboard-powered boat about 8 m inlength. In the four encounters, Atlantic spotted dolphins werein groups of 40–50 animals and calves corresponded to about10%–15%. Besides group size and composition, the fourherds had the same characteristics with individuals movingtogether in apparent association, engaged in same activityand rarely split out in a large area �Table I�. Two distinctbehavioral states were observed and associated with theacoustic recordings: �a� low surface activity: dolphins en-gaged in slowly and moderate traveling and short distancesbetween individuals �less than 10 m�, but no physical contactwas observed; �b� high surface activities: dolphins engagedin socializing and foraging/feeding behaviors with fast moveat surface, prey pursuit, aerial behavior, and physical contactbetween dolphins.

Acoustic recordings were made with the engine turnedoff and were monitored by headphones. Whenever dolphinsounds became weak, we stopped recording and repositionedthe boat. The recording system consisted of a High Tech Inc.hydrophone �model HTI-96-MIN, frequency response: 5 Hzto 30 kHz�1.0 dB, �165 dB re 1 V /�Pa� and a digitalaudio tape recorder Sony TCD-D8 with upper frequencylimit of 24 kHz �sampling rate of 48 kHz�. Whistles were

FIG. 1. Map of Rio de Janeiro coast, southeastern Brazil, where acousticrecordings of underwater sound produced by Atlantic spotted dolphins weremade.

TABLE I. Aggregation size, percentual of calves, time observed, and totaltime of recording, in which Atlantic spotted dolphin whistles were recordedin study area.

MonthAggregation

sizePercentualof calves

Time observed�h�

Time recorded�min�

April 2007 55 10 2.5 56October 2007 40 10 4 58October 2007 50 15 4 90December 2008 50 15 3.5 75

J. Acoust. Soc. Am., Vol. 127, No. 4, April 2010

Downloaded 04 Jan 2012 to 200.128.66.36. Redistribution subject to ASA license

defined as continuous, narrow-band sound emissions with orwithout harmonics �Popper, 1980�. Whistles were analyzedusing the RAVEN 1.1 program �Cornell Laboratory of Orni-thology, New York� with a fast Fourier transform �FFT� sizeof 1024 points, an overlap of 50%, and using a 512 sampleHamming window.

The contour of each whistle was determined by visualanalyses of the frequency modulation by at least two authorsand was then categorized into the following broad classes:ascending �whistles rising in frequency and with no one in-flection point�, descending �whistles falling in frequency andwith no one inflection point�, ascending-descending �initialrising in frequency, one inflection point, and then falling infrequency�, descending-ascending �initial falling in fre-quency, one inflection point, and then rising in frequency�,constant �whistles which the frequency changes 1000 Hz orless during more than 90% of duration�, and multi �morethan one inflection point�.

Nine acoustic parameters from fundamental componentof each whistle were measured: starting frequency �SF�, end-ing frequency �EF�, minimum frequency �MinF�, maximumfrequency �MaxF�, delta frequency �DF� �MaxF−MinF�, du-ration �DUR�, frequency at 1

4 of duration �F 14

�, frequency at 12

of duration �F 12

�, and frequency at 34 of duration �F 3

4�. Pres-

ence of harmonics and number of inflection points �definedas points where the whistle contour changed from ascendingto descending or vice versa� were also determined. The fre-quency variables were measured in kilohertz and the dura-tion in milliseconds. The mean frequency �MeF� was calcu-lated as the average of SF, EF, MinF, and MaxF. Thesewhistle parameters were chosen to be consistent with previ-ous studies of Atlantic spotted dolphins �e.g., Wang et al.,1995; Lammers et al., 2003; Baron et al., 2008� and otherdolphin species �e.g., Bazúa-Durán and Au, 2004; Azevedoand Van Sluys, 2005�. Only whistles for which all parametersof a spectral contour were distinctly measurable were used.

The descriptive statistics for all measured variables in-cludes the minimum values, maximum values, means, andstandard deviations. The paired-sample t test �Zar, 1999� wasapplied to verify if the mean of the end frequency of allwhistles analyzed was significantly different from the startfrequency. The frequency of whistles in each category wascompared between the two behavioral states using Chi-square test �P�0.01� and Mann–Whitney test �P�0.05�was applied to investigate variation in whistle variables be-tween the two behavioral states.

III. RESULTS

A total of 1745 whistles were recorded over 279 min.Five whistles were not analyzed because they were “cut off”by the upper frequency limit of the recording system. Of thattotal �N=1745�, 1092 whistles had adequate signal qualityfor acoustical analyses. Pure tone whistles were the mostfrequent ones �N=701; 64.2%� and whistles with up to 14

Azevedo et al.: Atlantic spotted dolphin whistles in Brazil 2647

or copyright; see http://asadl.org/journals/doc/ASALIB-home/info/terms.jsp

inflection points corresponded to 94% of all whistles.Whistles categorized as ascending were the most commonand corresponded to 47.0% of all whistles. Those with morethan one inflection point represented 16.5% and descending-ascending 15.5%. Ascending-descending �10.3%�, descend-

FIG. 2. �Color online� Examples of whistles emitt

ing �6.5%�, and constant �4.2%� whistles were less frequent.

2648 J. Acoust. Soc. Am., Vol. 127, No. 4, April 2010

Downloaded 04 Jan 2012 to 200.128.66.36. Redistribution subject to ASA license

Examples of whistles emitted by Atlantic spotted dolphinsare presented in Fig. 2. There were significant differences forwhistle contour distribution of the six categories betweenlow and high activities, but ascending whistles were morefrequently emitted in both behavioral states �X2=41.9; df

Atlantic spotted dolphins in southeastern Brazil.

=5; P�0.01; Fig. 3�.

Azevedo et al.: Atlantic spotted dolphin whistles in Brazil

or copyright; see http://asadl.org/journals/doc/ASALIB-home/info/terms.jsp

Whistle duration mean was 360.9�293.2 ms and97.6% of the whistles lasted less than 1 s. The average mini-mum frequency was 8.04�2.51 kHz and maximum fre-quency averaged 13.58�3.64 kHz. The mean of delta fre-quency was 6.25�3.34 kHz. The average mean frequencywas 10.81�2.63 kHz and the whistles had MeFs rangingfrom 2.72 to 21.00 kHz. The end frequency �mean12.76�3.80 kHz� of the Atlantic spotted dolphin whistleswas, on average, higher than the start frequency �mean8.85�3.21 kHz� �paired-sample t-test, t=29.41, df=1091,P�0.01�. Descriptive statistics of all measured whistle pa-rameters are shown in Table II.

Comparisons of acoustic whistle parameters recordedduring low and high activities found significant differencesin most of whistle variables. Averages of duration, starting

FIG. 3. Distribution of the six categories of whistles emitted by Atlanticspotted dolphins during low and high activities.

TABLE II. Descriptive statistics �mean�standard deviation; minimum-maxRio de Janeiro coast, southeastern Brazil. The frequency variables were mvariation in whistle variables among the two behavioral states.

Parameters All �N=1092� Low activity �N

Duration 0.36�0.29 0.34�0.30.05–3.62 0.05–3.6

Starting frequency 8.85�3.21 8.00�3.1.15–21.88 1.15–21.1

Ending frequency 12.76�3.80 12.50�3.1.56–22.35 1.56–22.3

Minimum frequency 8.04�2.51 7.94�2.41.15–20.09 1.15–20.0

Maximum frequency 13.58�3.64 13.07�3.3.00–23.44 3.23–23.4

Delta frequency 5.53�3.52 5.13�3.50.11–16.47 0.11–16.4

Mean frequency 10.81�2.63 10.50�2.2.72–21.00 3.01–21.0

Frequency at 14 of duration 9.63�3.00 9.12�2.8

2.71–22.19 2.71–22.1

Frequency at 12 of duration 10.77�3.04 10.36�2.

3.00–23.44 3.11–23.4

Frequency at 34 of duration 11.55�3.02 11.26�3.

2.88–22.74 3.17–22.7Inflections 0.74�1.30 0.47�0.8

0–14 0–8

J. Acoust. Soc. Am., Vol. 127, No. 4, April 2010

Downloaded 04 Jan 2012 to 200.128.66.36. Redistribution subject to ASA license

frequency, maximum frequency, delta frequency, mean fre-quency, frequency at 1

4 of duration, frequency at 12 of dura-

tion, frequency at 34 of duration, and number of inflections

points of the whistles emitted during high activities werehigher than those produced in low activities �Table II; Mann–Whitney test; P�0.05�. Additionally, whistles were pro-duced more frequently during high surface activities �9.2/min� than during low surface activities �3.9/min�.

IV. DISCUSSION

Atlantic spotted dolphin emitted a varied repertoire ofwhistles simple in structure. Whistles with up to two inflec-tions were more abundant. This is not in accordance withprevious studies of free-ranging Atlantic spotted dolphins inNorth Atlantic Ocean. Wang et al. �1995� and Baron et al.�2008� reported that the species produced whistles with ahigher number of inflection points �Table III�. Wang et al.�1995� found a mean number of inflection points of 3.43 inBahamas, and Baron et al. �2008� reported the value of 2.22as the average in Gulf of Mexico, while we found 0.72 insoutheastern Brazilian coast. These differences may be re-lated to longer whistles reported by those authors, since theyfound whistles with mean values of duration 127% and 80%higher than in Rio de Janeiro coast. Differences betweenthose characteristic whistles of Atlantic spotted dolphinsfrom South Atlantic Ocean and other sites may be caused byadaptation to environmental conditions, social relationship,behavioral variation, as well as anatomical features related tosound production, as it has been suggested for other dolphinspecies �Steiner, 1981; Rendell et al., 1999; May-Colladoet al., 2007a�. Another whistle character that differs from

� for 11 acoustic parameters values of Atlantic spotted dolphin whistles ind in kHz and the duration in s. Mann–Whitney test �P�0.05� results for

8� High activity �N=704� Mann–Whitney test �P�0.05�

0.37�0.26 Yes0.06–1.879.04�3.31 Yes1.55–21.88

12.91�3.75 No2.76–22.128.10�2.54 No1.55–18.90

13.86�3.54 Yes3.00–22.985.76�3.50 Yes0.29–16.44

10.98�2.58 Yes2.72–20.209.91�3.05 Yes3.00–21.96

11.00�3.05 Yes3.00–22.82

11.70�2.97 Yes2.88–22.460.90�1.45 Yes

0–14

imumeasure

=38

42018857977447710299841049

Azevedo et al.: Atlantic spotted dolphin whistles in Brazil 2649

or copyright; see http://asadl.org/journals/doc/ASALIB-home/info/terms.jsp

those observed in previous studies was the presence of har-monics. Our findings revealed that Atlantic spotted dolphinspreferentially produced whistles without harmonics, whereasLammers et al. �2003� reported that the majority of thewhistles produced by the species had one or more harmonicsin Bahamas. However, these differences should be inter-preted with caution because Lammers et al. �2003� used abroadband recording system that capture more of thewhistles frequency range of this species.

Atlantic spotted dolphins have been shown to emitwhistles within a broad frequency range �Wang et al., 1995;Lammers et al., 2003; Baron et al., 2008�. Whistles recordedin our study also had wide frequency range �1.15–23.44kHz�, revealing that the species in Atlantic South Oceanemits whistles in frequencies close to 1 kHz and higher than20 kHz. Several odontocete species, including Atlantic spot-ted dolphin, emit whistles with fundamental frequency ex-tending into the ultrasonic range �Lammers et al., 2003;May-Collado et al., 2007b�. But, in fact, the usage ofwhistles with fundamental frequencies above 20 kHz seemsto be not usual by Atlantic spotted dolphins. Our results re-vealed only 4.7% of whistles with fundamental frequencyhigher than 20 kHz and only five whistles with fundamentalfrequency higher than 24 kHz. Additionally, Lammers et al.�2003� reported 27.4 kHz as the highest maximum funda-mental frequency measured for Atlantic spotted dolphin, inBahamas, but only 3.6% of all whistles had fundamentalfrequencies above 20 kHz.

Based on the relationship between starting and endingfrequencies, Bazúa-Durán and Au �2002� suggested that So-talia spp., Delphinus delphis, Stenella attenuata, S. clymene,S. longirostris, and Lagenorhynchus albirostris share someof the characteristics of their whistles. Considering that hy-pothesis, the Atlantic spotted dolphin can be included in thatgroup since our findings showed a clear predominance of

TABLE III. Values of whistle parameters for Atlantic spotted dolphins repo

Parameters This studya �N=1092� Wang et al. �1995�b �

Duration 0.36�0.29 0.82�0.400.05–3.62 0.08–2.07

Starting frequency 8.85�3.21 10.95�3.211.15–21.88 5.47–19.13

Ending frequency 12.76�3.80 11.56�3.661.56–22.35 6.25–19.79

Minimum frequency 8.04�2.51 7.91�1.181.15–20.09 5.00–11.09

Maximum frequency 13.58�3.64 16.04�1.893.00–23.44 9.92–19.79

Mean frequency 10.81�2.632.72–21.00

Delta Frequency 5.53�3.520.11–16.47

Inflections 0.74�1.30 3.43�2.030–14 0–11

aFrequency response of recording system: 24 kHz.bFrequency response of recording system: 25 kHz.cFrequency response of recording system: 130 kHz.dFrequency response of recording system: 24 kHz.

ascending whistles and the ending frequency was, on aver-

2650 J. Acoust. Soc. Am., Vol. 127, No. 4, April 2010

Downloaded 04 Jan 2012 to 200.128.66.36. Redistribution subject to ASA license

age, higher than starting frequency. However, our analysis issimple and the similarity of the whistles between Atlanticspotted dolphin and other delphinids deserves further explo-ration.

Variations in whistle parameters of dolphins and whis-tling rate related to behavioral states have been usually re-ported for delphinids �e. g. Herzing, 1996; Van Parijs et al.,2000; Bazúa-Durán and Au, 2004� and they seem to be re-lated to whistle functions, which include to transmit emo-tional states and to coordinate dolphin activities �Tyack,1998�. Whistles produced by Atlantic spotted dolphins variedamong the two behavioral states in which dolphins were en-gaged. When dolphins presented behaviors with fast move-ment at surface, prey pursuit, aerial behavior, and physicalcontact, the whistles were more frequently emitted and were,on average, longer and had higher frequency parameters thanwhen animals were engaged in slowly and moderate travel-ing. Additionally, dolphins emitted whistles more complex inshape when engaged in high surface activities.

V. CONCLUSIONS

This is the first description of the whistles produced byAtlantic spotted dolphins in South Atlantic Ocean. Dolphinsemitted whistles within a wide frequency range and somedifferences in signal characteristics were found in compari-son to areas of North Atlantic Ocean. Additionally, dolphinsmodified whistle structures within behavioral states, maybeas a result of whistle communication functions.

ACKNOWLEDGMENTS

This study was funded by Conselho Nacional para oDesenvolvimento Científico e Tecnológico �CNPq-Brazil;Edital Universal No. 476255/2007-4�. We particularly thankAdriana Nascimento Gomes, Sylvia Chada, and Sílvia

in some studies cited in this paper.

� Lammers et al. �2003�c �N=220� Baron et al. �2008�d �N=27�

0.44�0.30 0.65�0.04

9.08�0.25

13.13�0.35

7.1�1.5 7.51�0.17

14.5�2.5 15.84�0.32

10.9�2.0 ND

7.4�2.0 8.33�0.31

2.22�0.14

rted

N=80

Peixoto of the Estação Ecológica Tamoios �ESEC Tamoios-

Azevedo et al.: Atlantic spotted dolphin whistles in Brazil

or copyright; see http://asadl.org/journals/doc/ASALIB-home/info/terms.jsp

ICMBio� and Faculdade de Oceanografia �UERJ� which pro-vided logistical support. We also thank Mr. José Carlos tohelp us during all data collection. Dr. Azevedo has a researchgrant from CNPq �Grant No. 304826/2008-1�. J.L.-B. is aresearcher of “Prociência” Program-FAPERJ/UERJ. T.L.B.has a scholarship from CAPES. Being a team, the interns ofthe Laboratório de Mamíferos Aquáticos e Bioindicadores�MAQUA-UERJ� are acknowledged.

Azevedo, A. F., and Van Sluys, M. �2005�. “Whistles of tucuxi dolphins�Sotalia fluviatilis� in Brazil: Comparisons among populations,” J. Acoust.Soc. Am. 117, 1456–1464.

Baron, S. C., Marinez, A., Garrison, L. P., and Keith, E. O. �2008�. “Differ-ences in acoustic signals from delphinids in the western North Atlantic andnorthern Gulf of Mexico,” Marine Mammal Sci. 24, 42–56.

Bazúa-Durán, M. C., and Au, W. W. L. �2002�. “Whistles of Hawaiianspinner dolphins,” J. Acoust. Soc. Am. 112, 3064–3072.

Bazúa-Durán, M. C., and Au, W. W. L. �2004�. “Geographic variations inthe whistles of spinner dolphins �Stenella longirostris� of the main Hawai-ian islands,” J. Acoust. Soc. Am. 116, 3757–3769.

Caldwell, D. K., Caldwell, M. C., and Miller, J. F. �1973�. “Statistical evi-dence for individual signature whistles in the spotted dolphin, Stenellaplagiodon,” Cetology 16, 1–21.

Herzing, D. L. �1996�. “Vocalizations and associated underwater behavior offree-ranging Atlantic spotted dolphins, Stenella frontalis and bottlenosedolphins, Tursiops truncatus,” Aquat. Mamm. 22, 61–79.

Jefferson, T. A., Webber, M. A., and Pitman, R. L. �2008�. Marine Mammalsof the World, A Comprehensive Guide to their Identification �Elsevier,Amsterdam�.

Lammers, M. O., Au, W. W. L., and Herzing, H. L. �2003�. “The broadbandsocial acoustic signaling behavior of spinner and spotted dolphins,” J.Acoust. Soc. Am. 114, 1629–1639.

May-Collado, L. J., Agnarsson, I., and Wartzok, D. �2007a�. “Phylogeneticreview of tonal sound production in whales in relation to sociality,” BMCEvol. Biol. 7, 136.

J. Acoust. Soc. Am., Vol. 127, No. 4, April 2010

Downloaded 04 Jan 2012 to 200.128.66.36. Redistribution subject to ASA license

May-Collado, L. J., Agnarsson, I., and Wartzok, D. �2007b�. “Reexaminingthe relationship between body size and tonal signals frequency in whales:A comparative approach using a novel phylogeny,” Marine Mammal Sci.23, 524–552.

May-Collado, L. J., and Wartzok, D. �2009�. “A characterization of Guyanadolphin �Sotalia guianensis� whistles from Costa Rica: The importance ofbroadband recording systems,” J. Acoust. Soc. Am. 125, 1202–1213.

Moreno, I. B., Zerbini, A. N., Danilewicz, D., Santos, M. C. O., Simões-Lopes, P. C., Lailson-Brito, J., Jr., and Azevedo, A. F. �2005�. “Distribu-tion and habitat characteristics of dolphins of the genus Stenella �Cetacea:Delphinidae� in the southwest Atlantic Ocean,” Mar. Ecol. Prog. Ser. 300,229–240.

Perrin, W. F. �2002�. “Stenella frontalis,” Mamm. Species 702, 1–6.Popper, A. N. �1980�. “Sound emission and detection by delphinids,” in

Cetacean Behaviour: Mechanisms and Functions, edited by L. M. Her-man, �Wiley, New York�, pp. 1–52.

Rendell, L. E., Matthews, J. N., Gill, A., Gordon, J. C. D., and MacDonald,D. W. �1999�. “Quantitative analysis of tonal calls from five odontocetespecies, examining interspecific and intraspecific variation,” J. Zool. 249,403–410.

Richardson, W. J., Greene, C. R. J., Malme, C. I., and Thomsom, D. H.�1998�. Marine Mammals and Noise �Academic, New York�.

Steiner, W. W. �1981�. “Species-specific differences in pure tonal whistlevocalizations of five western North Atlantic dolphin species,” Behav. Ecol.Sociobiol. 9, 241–246.

Tyack, P. L. �1998�. “Acoustic communication under sea,” in Animal Acous-tics Communication: Sound Analysis and Research Methods, edited by S.L. Hopp, M. J. Owen, and C. S. Evans �Springer, New York�, pp. 163–219.

Van Parijs, S. M., Parra, G. J., and Corkeron, P. J. �2000�. “Sounds producedby Australian Irrawaddy dolphins, Orcaella brevirostris,” J. Acoust. Soc.Am. 108, 1938–1940.

Wang, D., Wursig, B., and Evans, W. E. �1995�. “Comparisons of whistlesamong seven odontocete species,” in Sensory Systems of Aquatic Mam-mals, edited by R. A. Kastelien, J. A. Thomas, and P. E. Nachtigal, �DeSpil, Woerden, The Netherlands�, pp. 299–323.

Zar, J. H., Biostatistical Analysis �Prentice-Hall, Englewood Cliffs, NJ��1999�.

Azevedo et al.: Atlantic spotted dolphin whistles in Brazil 2651

or copyright; see http://asadl.org/journals/doc/ASALIB-home/info/terms.jsp