Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Efficiency of playback for assessing the occurrence of five bird species in Brazilian Atlantic Forest fragments

Efficiency of playback for assessing the occurrence of five bird species in Brazilian Atlantic... Anais da Academia Brasileira de Ciências (2006) 78(4): 629– 644 (Annals of the Brazilian Academy of Sciences) ISSN 0001-3765 www.scielo.br/aabc Efficiency of playback for assessing the occurrence of five bird species in Brazilian Atlantic Forest fragments 1 1 2 DANILO BOSCOLO , JEAN PAUL METZGER and JACQUES M.E. VIELLIARD Departamento de Ecologia, Instituto de Biociências, Universidade de São Paulo Rua do Matão, 321, Travessa 14, 05508-900 São Paulo, SP, Brasil Departmento de Zoologia, Instituto de Biologia, Universidade Estatual de Campinas Caixa Postal 6109, 13083-970 Campinas, SP, Brasil Manuscript received on April 17, 2006; accepted for publication on April 19, 2006; contributed by JACQUES M.E. VIELLIARD* ABSTRACT Playback of bird songs is a useful technique for species detection; however, this method is usually not stan- dardized. We tested playback effi ciency for fi ve Atlantic Forest birds (White-browed Warbler Basileuterus leucoblepharus, Giant Antshrike Batara cinerea, Swallow-tailed Manakin Chiroxiphia caudata, White- shouldered Fire-eye Pyriglena leucoptera and Surucua Trogon Trogon surrucura) for different time of the day, season of the year and species abundance at the Morro Grande Forest Reserve (South-eastern Brazil) and at thirteen forest fragments in a nearby landscape. Vocalizations were broadcasted monthly at sunrise, noon and sunset, during one year. For B. leucoblepharus, C. caudata and T. surrucura, sunrise and noon were more effi cient than sunset. Batara cinerea presented higher effi ciency from July to October. Playback expanded the favourable period for avifaunal surveys in tropical forest, usually restricted to early morning in the breeding season. The playback was effi cient in detecting the presence of all species when the abundance was not too low. But only B. leucoblepharus and T. surrucura showed abundance values signifi cantly related to this effi ciency. The present study provided a precise indication of the best daily and seasonal periods and a confi dence interval to maximize the effi ciency of playback to detect the occurrence of these forest species. Key words: playback, birds, forest fragments, methodology development, Atlantic forest, São Paulo State. INTRODUCTION the detection of the species presence and accurate in the recognition of their absence. Species occurrence data have often been used to One main challenge is to avoid the problem monitor populations in fragmented habitats and to of “ false absence” (Thompson 2002). Cryptic be- understand the effects of fragmentation on popu- haviour or seasonal and daily variations in species lation survival (Hanski 1994, Clergeau and Burel activity can make this task diffi cult, especially if the 1997, Lindenmayer et al. 1999). This kind of data survey is conducted inside dense habitats such as should be easy to collect, enabling surveys of a large neotropical forests (Catchpole and Slater 1995, Sick number of habitat fragments in a short period of 1997), where visual contact is seriously limited. To time. To be effi cient, these methods must be fast in reduce this problem, surveys may be conducted with the assistance of playback calling. *Member Academia Brasileira de Ciências Correspondence to: Jean Paul Metzger Many studies have revealed that the use of play- E-mail: jpm@ib.usp.br An Acad Bras Cienc (2006) 78 (4) 630 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD back increases the number of contacts during bird 1997). The climate type is temperate and rainy, surveys (Johnson et al. 1981, Rosenfi eld et al. 1985, as Köppen’s (1948) Cfa. Mean monthly tempera- ◦ ◦ Johnson and Dinsmore 1986, Kaufmann 1988, Swift ture ranges from 11 Cto27 C and mean annual et al. 1988, Sliwa and Sherry 1992). Some exper- precipitation is about 1400 mm, with the coldest iments indicated that the playback of conspecifi c and driest months between April and August. The songs is effi cient in inducing aggressive responses original vegetation cover was classifi ed as “ dense from various bird species, allowing their detection montane ombrophilous Atlantic Forest” (Veloso et (Milligan 1966, Møller 1992, Catchpole and Slater al. 1991). Nowadays, the region is composed of 1995). Mosher et al. (1990) pointed out that the nearly 360 fragments of second growth forest that use of playback during bird surveys might not only ranges from 1 to 300 ha. The landscape presents increase the contact rate but also reduce the interval 31% of forest fragments in medium to late succes- needed for the fi rst contact. Other studies, how- sional stages and 6% of early successional forests, ever, pointed out that variations in the abundance totalling 37% natural vegetation. The remaining of birds in the forest might have an effect on the area is composed of agricultural fi elds, pastures, results obtained by playback experiments (Glahn exotic forest plantation and human infrastructure 1974, Mosher and Fuller 1996, Schieck 1997). Al- (Metzger 2003). Five km east of the fragmented though playback techniques are widely used to sur- landscape, over the same geomorphologic unit, lies ◦ ◦ vey birds, few studies have been conducted with the the Morro Grande Forest Reserve (23 39’ – 23 48’ S ◦ ◦ aim of standardizing it. According to Morrell et al. and 46 47’ – 46 55’ W), a 9.400 ha continuous late (1991) and Ratcliffe et al. (1998), the standardiza- second growth forest well connected with the ma- tion of the technique can reduce the operation cost ture rainforest of the Paranapiacaba sierra. and increase the possibility of collecting a large data STUDIED SPECIES set in a short time. These factors become even more important when the objective is to sample a large Five bird species were chosen considering three number of fragments. main criteria. Firstly, they were strictly forest spe- The current study was designed to develop an cies that exhibit territorial behaviour, thus being accurate and simple tool to detect the presence and likely to respond to playback (J.M. Goerck, unpub- absence (occurrence pattern) of strictly forest bird lished data, Stotz et al. 1996). Secondly, in or- species inside a tropical fragmented landscape. Our der to consider species that perceive the landscape specifi c objectives were: i) to verify the time of in different ways and present distinct sensitivity to the day and season of the year when some Atlantic the fragmentation of their habitat, the chosen spe- Forest bird species are more effi ciently detected by cies presented also diverse biological characteris- playback; ii) to test if the effi ciency of the tics. Finally, all species were also non-migratory technique is affected by the abundance of the birds. and can be found all year long at the study sites. These species are (Table I): 1 – White-browed Warbler Basileuterus leucoblepharus, a small Paru- MATERIALS AND METHODS lidae weighing only 20 g, found in montane forests. STUDY REGION It feeds on insects, lives in the understory and is The study region was located on the Atlantic Plateau weakly affected by habitat fragmentation (Uezu et of São Paulo (Brazil), between the municipal dis- al. 2005). 2 – Giant Antshrike Batara cinerea, ◦ ◦ tricts of Cotia and Ibiúna (23 35’ – 23 50’ S and the biggest of the fi ve species, weighing approxi- ◦ ◦ 46 45’ – 47 15’ W). Local relief is characterized by mately 140 g. It is an insectivorous Formicariidae hills with more than 15% of inclination and eleva- that lives below the canopy and benefi ts from frag- tions between 850 and 1100 m (Ross and Moroz mentation in the Cotia region (Uezu et al. 2005). An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 631 TABLE I Selected bird species and their characteristics. In: insectivorous; Fr: frugivorous; Bc: below canopy; Ms: middle storey; Us: understorey. Species Feeding Sensitivity to Family Weigh Habitat (common name) habit fragmentation Basileuterus leucoblepharus Parulidae 20 g In Us Low (White-browed Warbler) Batara cinerea Formicariidae 140 g In Bc + Ms Low (Giant Antshrike) Chiroxiphia caudata Pipridae 25 g Fr Ms + Us Medium (Swallow-tailed Manakin) Pyriglena leucoptera Formicariidae 30 g In Ms + Us Medium (White-shouldered Fire-eye) Trogon surrucura Trogonidae 70 g Fr Bc High (Surucua Trogon) 3 – Swallow-tailed Manakin Chiroxiphia caudata, note shorter; C – The vocalization of C. caudata a Pipridae weighing around 25 g known by its char- is characterized by strong calls usually, but not al- acteristic and intricate sexual behaviour (Foster ways, repeated for several times; D – The song of 1981). The species is frugivorous, inhabits the in- P. leucoptera is very penetrating in the forest and termediate strata of the forest and is moderately af- is characterized by an unvarying sequence of the fected by habitat fragmentation (Uezu et al. 2005). same note; and E – Trogon surrucura sings a con- 4 – White-shouldered Fire-eye Pyriglena leucop- stant series of 14 to 17 notes with the last ones tera, another Formicariidae, weighing about 30 g. lower. This species is of medium sensitivity to fragmen- Song recordings were broadcasted using a tation (Uezu et al. 2005) and lives in the middle Sony TCM-5000EV cassette tape recorder with the and low storeys of the forest searching for insects speaker positioned between one and two meters (Sick 1997). 5 – Surucua Trogon Trogon surrucura, above the ground. The volume of the broadcast was from the family Trogonidae, is a large frugivorous set for human ears to hear it at a maximum distance bird that weights approximately 70 g. Because of of 100 m. The intensity of the songs, measured its size and feeding habits, it is very susceptible to with a Bruel & Kjaer decibel meter positioned at fragmentation (Goerck 1997), being present only in 1 m from the speaker was: 79.2 dB for B. leucoble- large forest fragments (Uezu et al. 2005). pharus, 81.6 dB for B. cinerea, 75.0 dB for C. cau- data, 79.0 dB for P. leucoptera and 79.5 dB for T. surrucura. SONG RECORDINGS DAILY AND SEASONAL VARIATION OF Male bird song or call recordings were obtained PLAYBACK EFFICIENCY from the Arquivo Sonoro Neotropical (ASN) of the Universidade Estadual de Campinas (UNICAMP), This experiment was conducted inside the contin- Brazil. The vocalization of each species is distinc- uous forest of the Morro Grande Forest Reserve tive and with unique characteristics (Fig. 1): A – between June 2000 and May 2001, in three distinct The song of B. leucoblepharus consists of a high- selected sites where all fi ve species were known to pitched scale of descending notes; B – Batara cine- be present and could be easily observed or heard rea vocalizes a long and trilled song with the last all year long. In order to reduce the variance in An Acad Bras Cienc (2006) 78 (4) 632 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD Fig. 1 – Sonograms of the vocalizations used for each studied species for the playback surveys. A: B. leucoblepharus; B: B. cinerea; C: C. caudata; D: P. leucoptera; E: T. surrucura. Each sonogram is presented in a distinct scale of time (in seconds) and frequency (in kHz). Extracted from the Arquivo Sonoro Neotropical (ASN), UNICAMP. environmental conditions between sites, their selec- values for a single species may vary from zero to tion was guided according to the forest type and the total number of playback sessions conducted at structure. Playback sessions were conducted every a given month or day period. In order to allow eas- separate month and for different times of the day. ier comparisons, these values were also standardised The sum of detected bird responses to the playback as percentages of positive response. within each tested month or period of the day was To assess detection effi ciency in relation to termed detection effi ciency. Therefore, effi ciency time of the day, three different test periods were An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 633 set: 1) during the two hours after sunrise; 2) at about late responsive birds. On the other hand, it would two hours around noon and 3) during the two hours not be convenient to wait for periods longer than the before sunset. At each period, three playback ses- mean latency time of the species added to its stan- sions were broadcasted for each one of the species dard deviation, as it is unlikely that a bird responds at three different sampling points (nine playback after this and the species could then be documented sessions at nine different points per day). Each play- as absent for the tested point without further loss back session was composed by one minute of play- of time. After the end of the last species broadcast, back, intercalated by thirty seconds of silence, to the observer (DB) waited for fi ve additional minutes account for responsive birds (Johnson et al. 1981), to account for birds with long latency of response. until fi ve minutes were completed (Morrell et al. Playback was not carried out during rainy days due 1991, Sliwa and Sherry 1992). To avoid the sti- to noise interference (Johnson et al. 1981). mulation of the same bird more than once during These procedures were repeated three days the same period, sampling points were set at least a month for twelve consecutive months. For each 200 m away from each other (Sliwa and Sherry species, 27 playback sessions were conducted 1992). The sequence of species broadcasted was monthly, totalling 324 tests apiece, 108 at each pe- set randomly at each point. To prevent habituation riod of the day. Differences in birds detection effi - ciency along the day were verifi ed through G-tests of the birds to the recorded songs no sampling point was tested more than once a month (W.R. Silva, un- in contingency tables, and when signifi cant differ- published data). ences were detected (P< 0.05) an a posteriori χ test At each session, the response by the birds was (Zar 1999) was used. Detection effi ciency through- recorded as negative (no response) or positive. A out the year was tested with circular statistics, com- response was considered positive when changes in paring the twelve months through the Rayleigh’s test the behaviour of the birds were detected after the (Zar 1999), with a signifi cance of P< 0.05. playback begun. These changes were usually rec- PLAYBACK DETECTION EFFICIENCY IN RELATION ognized as increased rate of vocalization and/or al- TO BIRD ABUNDANCE terations in the movement of the birds towards the sound source. If an individual was detected clos- The outcomes of the experiment described above ing in and/or circling around the sound source in were used to determine the best period of the day silence during or after the playback, it was also doc- and season of the year when the relationship be- umented as a responsive bird. Birds detected be- tween playback effi ciency and bird abundance fore the beginning of the playback but that did not should be tested. To conduct that test, thirteen for- show any changes in their behaviour demonstrating est fragments ranging from 13 to 275 ha were stud- specifi c interest for the played song during or af- ied. Fragment selection was conducted in order to ter the broadcast were noted as negative responders reproduce Reserve sites environmental conditions. (no response). To identify how long it takes for each Inside each fragment, a grid of fi ve to nine sam- species to respond to the playback, the time elapsed pling points (depending on fragments area) located between the beginning of the playback and the de- at least 200 m apart was set. The sites used before tection of positive response, termed as the latency in the Morro Grande Forest Reserve were divided time, was also noted. This was made in order to into three different sets of nine points and sampled identify which is the minimum time that someone again. The Reserve sets of nine points were orig- should wait for a response to take place before stat- inally placed at least 2 km apart of each other in ing that the species is absent during that session. order to simulate the spatial confi guration of the A short waiting time may lead to a decrease in the fragments. Because the latency time of each spe- quantity of contacts by impeding the detection of cies was already known, the species broadcasting An Acad Bras Cienc (2006) 78 (4) 634 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD sequence was set by this parameter, in order to in- visits to the fragments were needed for the playback crease bird detection probabilities. Consequently, to acquire a similar occurrence pattern as obtained the fi rst broadcasted species was the one with high- by point counts, the three above explained meth- est latency time and the last with the shortest. Each ods were conducted for all possible combinations of study site (fragments or Reserve sites) was sampled one, two, three and four survey days. for four different non-consecutive days at the best season of the year and time of the day. In each RESULTS one of the four days, three sampling points were DETECTION EFFICIENCY IN RELATION TO THE TIME randomly selected for the playback trials, totalising OF DAY AND SEASON OF THE YEAR twelve playback sessions per site and species for each one of the sixteen sites. Each site was sur- The species with the highest percentages of pos- veyed with playback twice at sunrise and twice at itive responses inside the Reserve was B. leuco- noon, between 1 August and 29 September 2001. blepharus (Table II). Mean latency time varied be- The abundance information was estimated tween species from 2 to 8.3 minutes (Table III). The between May 2000 and April 2001 through pas- analysis of variance showed that latency times were sive point counts (Blondel et al. 1970, Vielliard signifi cantly different among species (F = 41.16; and Silva 1990), at the same points used for the P< 0.01). The a posteriori test indicated that the playback (Uezu et al. 2005). This was the best latency time of B. cinerea was signifi cantly longer available abundance estimate for all studied spe- than any other tested species (P< 0.01). The latency cies and could be used as a precise occurrence data presented by C. caudata was different from that of at these sites. All fragments were sampled during T. surrucura (P< 0.01) but not of B. leucoblepharus early morning for one day every two months. At (P = 0.80). No signifi cant differences were found each day, one fragment was selected and fi ve sam- between P. leucoptera and T. surrucura (P = 0.58) ple points were randomly chosen in it. Each point or between P. leucoptera and C. caudata (P = 0.27). was observed for 20 minutes to account for present Three species, B. leucoblepharus, C. caudata birds. By dividing the number of contacts, visual and T. surrucura showed signifi cant differences in and/or auditory, by the number of sampled points detection effi ciency according to the period of the an Index of Point Abundance (IPA) was then calcu- day (Table III). The a posteriori χ test indicated lated for each species in each fragment. that the effi ciency at sunrise and noon was signif- To test if detection effi ciency was related to icantly higher in comparison with the period just bird abundance in the forest, three different methods before sunset (Table III). were used. The playback-determined occurrence According to the Rayleigh’s test, the only spe- of the birds (dependent variable) was related with cies to present a mean season of higher effi ciency their abundance (IPA, independent variable) in a was B. cinerea (z = 6.041; P< 0.01, Fig. 2). 0.05,88 logistic regression (Ratcliffe et al. 1998). A Pearson All other species had uniform distributions of effi - correlation analysis (Zar 1999) was done between ciency throughout the twelve tested months (Fig. 2). the response rates to the playback in each fragment The mean period of higher effi ciency for B. cinerea and the IPA. The response rate was defi ned for each was July and August, but considering the low con- species as the quantity of positive responses in each centration of responses (r = 0.262), one can perceive site divided by twelve, the total number of playback that this is only the central tendency of a larger pe- sessions conducted after four survey days. Finally, riod of fi ve consecutive months ranging from June the percentage of sites where the playback correctly to October. Those fi ve months presented no signif- attested the presence of the birds (defi ned by point icant difference between each other along the given counts results) was calculated. To verify how many period (G = 3.807; P> 0.25). Responses from June An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 635 TABLE II Percentages of detected positive responses to playback according to each tested period of the day and month for the five studied species. B. leucoblepharus B. cinerea C. caudata P. leucoptera T. surrucura Periods Sunrise 77.8 27.8 61.1 39.8 53.7 of the Noon 72.2 25.9 61.1 28.7 45.4 day Sunset 58.3 27.8 41.7 36.1 25.9 January 48.1 14.8 59.3 40.7 25.9 February 70.4 2.,2 55.6 44.4 22.2 March 81.5 3.7 44.4 33.3 55.6 April 70.4 29.6 51.9 22.2 55.6 May 66.7 22.2 59.3 29.6 44.4 June 70.4 48.1 77.8 29.6 59.3 Months July 70.4 40.7 33.3 25.9 51.9 August 59.3 25.9 33.3 33.3 18.5 September 88.9 40.7 48.1 48.1 25.9 October 66.7 48.1 70.4 33.3 55.6 November 70.4 14.8 63.0 40.7 40.7 December 70.4 14.8 59.3 37.0 44.4 TABLE III Mean latency values (in minutes) and playback positive (pos) and negative (neg) responses according to the period of the day for the five studied species (N= 108 for each species and period) inside the Morro Grande Forest Reserve. P: significance level. Playback responses G test a posteriori χ Species Latency N* Sunrise Noon Sunset sr × nn × st (sr + nn) × st sr × nn 2 2 (sr) (nn) (st) G (P) χ (P) χ (P) (2) c(1) c(1) pos neg pos neg pos neg 2.0±1.9 10.09 8.65 0.62 B. leucoblepharus 84 24 78 30 63 45 225 (P<0.01) (P<0.01) (P>0.25) 8.3±8.1 0.12 B. cinerea 30 78 28 80 30 78 88 (P>0.9) 2.6±2.6 10.99 10.21 0.02 C. caudata 66 42 66 42 45 63 177 (P<0.01) (P<0.01) (P>0.99) 3.6±4.1 3.08 P. leucoptera 43 65 31 77 39 69 113 (P>0.1) 4.2±5.1 18.58 15.56 1.18 T. surrucura 58 50 49 59 28 80 135 (P<0.01) (P<0.01) (P>0.9) * The numbers of observations (N) are unequal between species because latency times could only be accounted after positive responses. An Acad Bras Cienc (2006) 78 (4) 636 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD Fig. 2 – Distribution of positive responses (gray dots) to the playback during one year of surveys and Rayleigh’s test results. A: B. cinerea; B: B. leucoblepharus; C: T. surrucura; D: C. caudata; E: P. leucoptera. N: number of playback sessions; r: concentration level, P: signifi cance level. to October were also compared to responses from occurrence only when IPA was less than 0.05. the rest of the year, confi rming that this was the Logistic regression was calculated only for T. period of highest playback effi ciency to detect B. surrucura since it was the only species not present cinerea in the forest (McNemar’s test, χ = 5.1127; at all sampled sites. One single survey day was c1 P< 0.025). enough to incur into a signifi cant regression, indi- cating a strong relation between the two tested vari- DETECTION EFFICIENCY ACCORDING TO THE ables (Table V). ABUNDANCE OF THE BIRDS INSIDE The correlation analyses showed contrasting FOREST FRAGMENTS results. For three species, B. cinerea, C. caudata and P. leucoptera, almost no correlation between Three species were detected in all sampled sites playback response rate and IPA were observed for by both the playback and the point counts (Table any of the 16 possible combinations (Table VI). For IV). The two methods presented similar species oc- currence results, except for two fragments where B. leucoblepharus and T. surrucura, nearly all pos- T. surrucura and P. leucoptera were detected only sible combinations had signifi cant positive correla- tions, showing a strong influence of the abundance by point counts but not by playback (Table IV). In on the detection by playback. both cases, species abundance (IPA) was low, indi- For all species, the percentage of sites where cating that the playback was defi cient in detecting An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 637 TABLE IV Number of positive responses for each study site obtained after twelve playback sessions per species and abundance values (IPA) in parenthesis (Uezu et al. 2005). Cases where IPA and playback results disagreed are marked with *. Fragments B. leucoblepharus B. cinerea C. caudata P. leucoptera T. surrucura Mioko5 (0.71) 7 (0.73) 9 (1.03) 5 (0.14) 1 (0.11) Takimoto 5 (0.89) 3 (0.53) 9 (1.31) 8 (0.26) 6 (0.23) Pedro 8 (1.43) 6 (0.70) 10 (1.60) 5 (0.34) 3 (0.11) Pedroso 4 (1.69) 6 (0.30) 5 (1.14) 2 (0.37) 1 (0.34) Zezinho 2 (0.31) 5 (0.30) 10 (1.03) 3 (0.26) 4 (0.23) Lila 3 (0.34) 5 (0.13) 8 (1.03) 3 (0.17) 0 (0.00) Dito-André 6 (0.89) 9 (0.63) 9 (0.77) 5 (0.20) 0 (0.00) Beto 6 (0.96) 9 (0.50) 7 (1.18) 5 (0.32) 0 (0.00) Lacerda 4 (0.25) 10 (0.38) 9 (1.46) 0 (0.04)* 0 (0.00) Godoy9 (1.00) 6 (0.50) 8 (1.18) 3 (0.43) 0 (0.00) Nelson 9 (1.69) 8 (0.43) 8 (0.63) 7 (0.54) 0 (0.00) Reizinho 5 (1.43) 6 (0.37) 5 (0.94) 3 (0.29) 0 (0.00) Agostinho 11 (1.40) 3 (0.47) 10 (0.89) 6 (0.23) 0 (0.03)* ReserveA10 (2.09) 4 (0.23) 7 (1.11) 2 (0.34) 6 (0.23) ReserveB6 (1.34) 1 (0.27) 8 (1.40) 1 (0.31) 8 (0.51) ReserveC8 (1.37) 5 (0.23) 10 (1.26) 8 (0.26) 9 (0.17) TABLE V Logistic regression between species occurrences according to playback results and to abundance index (IPA) for T. surrucura. “Combination” indicates which survey days (a, b, c, d) were utilized in each test. P: significance level. B0 is the estimated logistic coefficient. Quantity of survey days Combination Logistic regression B0 P a – 3.40 0.02 b – 2.31 < 0.01 c – 2.25 < 0.01 d – 3.08 < 0.01 a+b – 2.31 < 0.01 a+c – 2.51 < 0.01 a+d – 3.08 < 0.01 b+c – 6.12 < 0.01 b+d – 2.51 < 0.01 c+d – 6.12 < 0.01 a+b+c – 6.12 < 0.01 a+b+d – 2.31 < 0.01 a+c+d – 6.12 < 0.01 b+c+d – 6.12 < 0.01 4 a+b+c+d – 6.12 < 0.01 An Acad Bras Cienc (2006) 78 (4) 638 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD TABLE VI Pearson correlations coefficients (r ) between the efficiency of the playback and the abundance index values (IPA) for all species and all possible combinations of survey days (a, b, c, d). Quantity of B. leucoble- Combinations B. cinerea C. caudata P. leucoptera T. surrucura survey days pharus a 0.39* 0.03 0.05 0.00 0.26* b 0.23 0.10 0.08 0.34* 0.38* c 0.47* 0.05 0.10 0.00 0.36* d 0.05 0.03 0.01 0.00 0.47* a+b 0.50* 0.16 0.11 0.17 0.49* a+c 0.61* 0.05 0.00 0.00 0.36* a+d 0.34* 0.06 0.06 0.00 0.40* b+c 0.42* 0.16 0.00 0.09 0.57* b+d 0.16 0.07 0.09 0.15 0.48* c+d 0.25* 0.07 0.02 0.00 0.52* a+b+c 0.60* 0.14 0.02 0.07 0.54* a+b+d 0.37* 0.12 0.12 0.08 0.50* a+c+d 0.48* 0.07 0.00 0.00 0.45* b+c+d 0.29* 0.12 0.00 0.06 0.57* 4 a+b+c+d 0.47* 0.13 0.03 0.05 0.54* *P< 0.05. the playback correctly attested the bird presence in- derived from their complex courting behaviour, in- creased with the sampling effort (Fig. 3). For two dividuals shall be more prone to react to intruders species, B. leucoblepharus and B. cinerea, maxi- (in this case the broadcasted song) during this pe- mum success (100% of correct presence detection) riod, leading to higher method effi ciency. was obtained with three survey days. Two days were On the other hand, some passive point counts enough to obtain maximum success for C. caudata. studies tend to disagree with this result, consider- The remaining two species did not reach 100% of ing sunrise as the best period for bird surveys, fol- accuracy: the highest success for P. leucoptera was lowed by sunset and at last noon (Järvinen et al. around 95% and for T. surrucura around 90%. 1977, Robbins 1978, Bibby et al. 1993). G.S. Be- tini (unpublished data) detected that T. surrucura, P. leucoptera and Basileuterus culicivorus, a spe- DISCUSSION cies closely related to B. leucoblepharus, were PLAYBACK DETECTION EFFICIENCY:TIME OF THE spontaneously more active during early morning in DAY AND SEASON INFLUENCES contrast to the rest of the day, when their vocal ac- tivity was lower. For three species the best periods of the day to conduct playback census were just after sunrise and These results suggested that for these species at noon. This is in compliance with the daily peak the playback increased the time period when bird of spontaneous activity of some species, such as C. census should be conducted to acquire the high- caudata. This species presents a period of higher est effi ciency from sunrise to the entire morning, spontaneous activity between sunrise and 15:00, up to noon. Otherwise, the playback may induce the main period of sexual activity (Foster 1981). Be- birds to become detectable at any time, maintain- cause of the strong vocal interaction of the males, ing the quantity of contacted individuals equal along An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 639 Fig. 3 – Percentages (±SD) of sites where the playback correctly detected the presence of each species according to the number of survey days. Sites where IPA=0.0 were omitted. the day, in contrast to the census without playback. The only species with signifi cant seasonal vari- This was observed for P. leucoptera at the present ation in playback response was B. cinerea, which study and for Dendroica caerulescens in Jamaica should be surveyed between June and October to (Sliwa and Sherry 1992), which leads to the pattern acquire maximum census effi ciency. This is in ac- of no daily variation in bird detection presented by cordance with the pattern found by A. Uezu (unpub- these species. Of course, in this case the playback lished data), who recorded a more intense peak of had also increased the survey period to further than spontaneous vocal activity of this species between merely sunrise. July and September. The annual period of greater An Acad Bras Cienc (2006) 78 (4) 640 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD playback effi ciency may be linked to the reproduc- those cases, playback will be effi cient to identify tive biology of the species. Many authors use the only close birds, since the playback can only reach reproductive season to proceed with avifaunal sur- nearby individuals. But, even if the song functional veys (Dawson 1981, Mayfi eld 1981, Greenwood et structure is preserved (Aubin et al. 2002), allowing al. 1995, Clergeau and Burel 1997, Ratcliffe et al. birds farther away to recognize their specifi c vocal- 1998, Drapeau et al. 2000), when individuals are ization, the observer may not have the same ability, reinforcing their dominance or establishing new ter- decreasing the quantity of contacts proportionally ritories and consequently are more active. For this to the abundance. reason, they may be more prone to confront intrud- The signifi cant relationship found for Trogon ers, increasing the chances of response to playback. surrucura is probably an effect of the rarity of the However, contrary to the expected, for four of species. According to several researchers (Willis our species there was no seasonal variation in play- 1979, Christiansen and Pitter 1997, Marini 2001, back effi ciency. Even in non-reproductive months, Marsden et al. 2001), trogons are highly affected the birds kept responding with the same intensity by patch size, being present only in large fragments to the playback, producing no signifi cant variation or reserves and in low abundance, as detected in in the quantity of contacts along the tested year. the current study (Uezu et al. 2005). Usually, big This suggests that playback may be used as a sur- canopy frugivorous birds are rare and need large vey method for these bird species all year long and areas to search for food (Goerck 1997, Price et. al. not only during the reproductive season. This might 1999). The size of each individual territory was thus be related to the active stimulation of the birds or presumably large in relation to the distance between may come from a strong bond of the individuals to sample points. As the abundance increases, territo- their territories. ries might overlap or decrease in size (Sherman and Eason 1998) and contact with the birds may become PLAYBACK DETECTION EFFICIENCY ACCORDING TO more frequent, increasing detection rates along with BIRD ABUNDANCE the number of present birds in the forest. On the other hand, there was a lack of corre- Playback was a very effi cient technique to detect lation between the response rate and IPA for three the presence of these species in all sampled sites species. In the case of C. caudata, the lack of cor- where the point counts detected the occurrence of relation may be an effect of the sexual behaviour of the birds. It incurred in correct presence/absence the species. Male manakins are constantly engaged detection in at least 98% of the studied cases. How- in vocal duels for dominance in the hierarchy of the ever, there was not always a direct relationship be- group (Foster 1981) and present a natural aggres- tween the abundance of the birds and the detection sive behaviour that makes the individuals extremely effi ciency. responsive, increasing the detection rate after play- For B. leucoblepharus and T. surrucura there back. So, even in low abundance sites, the quantity was a signifi cant positive relationship between the of contacts shall be high. response rate and IPA, indicating that playback For B. cinerea, the lack of correlation was not capacity to identify sites where the species is pre- due to the behaviour of the individuals but to the sent increased with the abundance. According to properties of its vocalization. This species song Schieck (1997) vocalizations with frequencies is very loud and the responsive individual was fre- higher than 2 kHz may suffer strong attenuation quently far away from the playback source, indicat- in the forest. Aubin et al. (2002, 2004) stated ing that the song could travel long distances in the that the song of B. leucoblepharus is particularly forest and still be recognizable for the receiver. As high-pitched and susceptible to a strong attenua- a result, individuals might still be able to identify tion when propagating through dense vegetation. In An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 641 and respond to the broadcasted song even if they tion of correct detection did not reach 100% after are further than 100 m from the sound source. Be- four visits due to fragments with very low abun- cause of that, even in places with low abundance, dance. There was a threshold of approximately 0.05 it is possible to locate distant individuals, creating of IPA, below which the bird detection through the non-correlation pattern. playback is improbable for a sampling effort of four In the case of P. leucoptera no results pointed days. An increase of the sampling effort could solve out to any clear cause to the lack of correlation be- this problem. tween detection effi ciency and abundance. The de- tected pattern of response may be related not only CONCLUSION to bird abundance. For instance, if for any reason The results of the present study pointed out to a the access to food resources decline at a certain site, large variation in Atlantic Forest bird species re- individual birds may tend to become more aggres- sponse to playback and to the importance of method- sive in order to defend their territories (Armstrong ological studies in order to standardize census meth- 1991, Carrascal et al. 1998) and therefore more re- ods. Future researchers should try to understand sponsive to playback. In places of higher resource each species behaviour before applying the method. availability, even if the species is more abundant, If known, the latency time and the variation of play- there is no need to defend each territory so strongly back responses according to the time of the day, and the response rate may then decrease. Therefore, season of the year, abundance and sampling effort the abundance of a species did not appear to be the could help to reduce the problem of “ false absence” sole or main factor to influence the occurrence de- detections and maximize the effi ciency of the play- tection effi ciency of these birds. According to each back to collect occurrence data, assuring good re- species, other factors may modify this relationship. sults for quick access surveys. When compared to passive methods, the play- EFFECTS OF SAMPLING EFFORT back expanded the favourable time to conduct avi- According to Bibby et al. (2000) the chances of faunal surveys for these Atlantic Forest birds from detecting new individuals should increase along the fi rst hours of the day to the entire morning, up with the sample effort, independently of the sam- to noon. Also the season of the year is not an im- pling method. That was the case for all species, pediment to census these birds with playback. Four since the proportion of correct presence detection of the fi ve studied species might be surveyed with through playback increased continuously with the the same effi ciency of detection all year long. Bar- number of survey days. The main question here is ring when the abundance is too low, the playback how large should the sampling effort be to obtain an technique can detect with only three survey days accurate incidence data set, without wasting more per site the occurrence pattern of these birds as ac- time than necessary at a single site, in order to be curately as the passive point counts. These patterns able to quickly sample a large number of sites? Con- may also be applied to other bird species not studied sidering a confi dence threshold of 90%, one visit in the present work. was enough to accurately detect C. caudata, while two visits were necessary for B. leucoblepharus and ACKNOWLEDGMENTS B. cinerea, and three for P. leucoptera.For T. sur- rucura, four visits were necessary to obtain a con- This research was part of the thematic project fi dence value close to 90%. For a confi dence of “ Biodiversity conservation in fragmented land- 100%, two visits were necessary for C. caudata and scapes at the Atlantic Plateau of São Paulo” and was three for B. leucoblepharus and B. cinerea. In the fi nancially supported by the Programa de Pesquisas case of P. leucoptera and T. surrucura, the propor- em Caracterização, Conservação e Uso Sustentável An Acad Bras Cienc (2006) 78 (4) 642 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD da Biodiversidade da Fundação de Amparo à Pes- REFERENCES quisa do Estado de São Paulo (BIOTA-FAPESP) ARMSTRONG DP. 1991. Aggressiveness of breeding (n. 99/05123-4; master fellowship for DB territorial honeyeaters corresponds to seasonal n. 00/01358-6). We would like to thank Ilse Storch, changes in nectar availability. Behav Ecol Sociobiol Wesley R. Silva and José Carlos Motta Jr who re- 29: 103– 111. viewed previous versions of this manuscript and AUBIN T, MATHEVON N, VIELLIARD JME AND SILVA José Roberto Nali along with all the staff from Com- ML. 2002. How an inadequate acoustic signal suc- panhia de Saneamento Básico do Estado de São ceeds in the extreme environment of a tropical for- Paulo (SABESP) and the Morro Grande Forest est? The song of the White-browed Warbler. Beijing, Reserve for their hospitality and support. China: Abstracts XXIII Int Ornith Congr. AUBIN T, MATHEVON N, SILVA ML, VIELLIARD JME RESUMO AND SEBE F. 2004. How a simple and stereotyped A técnica de play-back é muito útil para a detecção de acoustic signal transmits individual information: the aves, mas este método geralmente não é padronizado. Sua song of the White-browed Warbler Basileuterus leu- efi ciência em atestar a ocorrência de cinco espécies de coblepharus. An Acad Bras Cienc 76: 335– 344. aves da Mata Atlântica (Pula-pula-assobiador Basileute- BIBBY CJ, BURGESS ND AND HILL DA. 1993. Bird rus leucoblepharus, Batará Batara cinerea, Tangará Chi- Census Techniques. San Diego: Academic Press. roxiphia caudata, Olho-de-fogo Pyriglena leucoptera e BIBBY CJ, BURGESS ND, HILL DA AND MUSTOE nd Surucuá-de-barriga-vermelha Trogon surrucura) foi ana- SH. 2000. Bird Census Techniques. 2 ed., San lisada de acordo com o horário do dia, estação do ano e Diego: Academic Press. abundância das espécies na Reserva Florestal do Morro BLONDEL J, FERRY C AND FROCHOT B. 1970. La Grande (São Paulo, Brasil) e em treze fragmentos flo- méthode des indices ponctuels d’ abondance (I.P.A.) ou des relevés d’ avifaune par “ stations d’ écoute”. restais de uma paisagem adjacente. Gravações das vo- Alauda 38: 55– 71. calizações foram emitidas mensalmente ao amanhecer, CARRASCAL LM, SENAR JC, MOSETICH I, URIBE meio do dia e entardecer, durante um ano. Para B. leu- F AND DOMENECH J. 1998. Interactions among coblepharus, C. caudata e T. surrucura, o amanhecer e o environmental stress, body condition, nutritional sta- meio do dia apresentaram-se mais efi cientes que o entar- tus, and dominance in Great Tits. Auk 115: 727– 738. decer. Batara cinerea apresentou maior efi ciência de ju- CATCHPOLE CK AND SLATER PJB. 1995. Bird Song: lho a outubro. O uso do play-back aumentou o período Biological themes and variations. Cambridge: Cam- de maior efi ciência dos levantamentos de aves em matas bridge University Press. tropicais, comumente restrito ao período da manhã du- CHRISTIANSEN MM AND PITTER E. 1997. Species rante a estação reprodutiva. A técnica utilizada foi efi - loss in a forest bird community near Lagoa Santa ciente na detecção da ocorrência de todas as espécies, in Southeastern Brazil. Biol Conserv 80: 23– 32. salvo quando sua abundância se apresentou muito baixa. CLERGEAU P AND BUREL F. 1997. The role of spatio- Porém, apenas B. leucoblepharus e T. surrucura apre- temporal patch connectivity at the landscape level: sentaram abundância signifi cativa e positivamente rela- an example in a bird distribution. Landscape and cionada com essa efi ciência. O presente estudo propor- Urban Planning 38: 37– 43. ciona uma indicação precisa sobre o melhor horário do dia DAWSON DG. 1981. Counting birds for a relative mea- e época do ano, assim como um intervalo de confi ança, sure (index) of density. In: RALPH CJ AND SCOTT com o intuito de aumentar a efi ciência do play-back em JM (Eds), Estimating the numbers of terrestrial detectar a ocorrência dessas espécies. birds. Studies in Avian Biology 6: 12– 16. Palavras-chave: play-back, aves, fragmentos florestais, DRAPEAU P, LEDUC A, GIROUX JF, SAVARD JPL, desenvolvimento metodológico, Mata Atlântica, Estado BERGERON Y AND VICKERY WL. 2000. Land- de São Paulo. scape-scale disturbances and changes in bird com- An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 643 munities of boreal mixed-wood forests. Ecol Monogr MAYFIELD HF. 1981. Problems in estimating population 70: 423– 444. size through counts of singing males. In: RALPH CJ AND SCOTT JM (Eds), Estimating the numbers of FOSTER MS. 1981. Cooperative behaviour and social terrestrial birds. Studies in Avian Biology 6: 220– organization of the Swallow-tailed Manakin (Chiro- xiphia caudata). Behav Ecol Sociobiol 9: 167– 177. METZGER JP (Coord). 2003. Biodiversity conserva- GLAHN JF. 1974. Study of breeding rails with rec- tion in fragmented landscapes at the Atlantic Plateau orded calls in north-central Colorado. Wilson Bull of São Paulo. São Paulo: Scientifi c report to the 86: 206– 214. BIOTA-FAPESP program. GOERCK JM. 1997. Patterns of rarity in the birds of the MILLIGAN M. 1966. Vocal responses of White-crowned Atlantic Forest of Brazil. Conserv Biol 11: 112– 118. Sparrows to recorded songs of their own and another GREENWOOD JJD, BAILLIE SR, GREGORY RD, species. Anim Behav 14: 356– 361. PEACH WJ AND FULLER RJ. 1995. Some new MØLLER AP. 1992. Interspecifi c response to playback approaches to conservation monitoring of British of bird song. Ethology 90: 315– 320. breeding birds. Ibis 137: 16– 28. MORRELL TE, YAHNER RH AND HARKNESS WL. HANSKI I. 1994. Patch-occupancy dynamics in frag- 1991. Factors affecting detection of Great Horned mented landscapes. Trends Ecol Evol 9: 131– 135. Owls by using broadcast vocalizations. Wildlife Soc JÄRVINEN O, VÄISÄNEN RA AND HAILA Y. 1977. Bull 19: 481– 488. Bird census results in different years, stages of the MOSHER JA AND FULLER MR. 1996. Surveying wood- breeding season and times of the day. Ornis Fennica land hawks with broadcasts of Great Horned Owl 54: 108– 118. vocalizations. Wildlife Soc Bull 24: 531– 536. JOHNSON RR AND DINSMORE JJ. 1986. The use of MOSHER JA, FULLER MR AND KOPENY M. 1990. tape-recorded calls to count Virginia Rails and Soras. Surveying woodland raptors by broadcast of conspe- Wilson Bull 98: 303– 306. cifi c vocalizations. J Field Ornith 61: 453– 461. JOHNSON RR, BROWN BT, HAIGHT LT AND SIMPSON PRICE OF, WOINARSKI JZ AND ROBINSON D. 1999. JM. 1981. Playback recording as a special avian Very large area requirements for frugivorous birds censusing technique. In: RALPH CJ AND SCOTT in monsoon rainforests of the Northern Territory, JM (Eds), Estimating the numbers of terrestrial birds. Australia. Biol Conserv 91: 169– 180. Studies in Avian Biology 6: 68– 75. RATCLIFFE N, VUGHAN D, WHYTE C AND SHEP- KAUFMANN GW. 1988. The usefulness of taped Spot- HERD M. 1998. Development of playback census less Crake calls as a census technique. Wilson Bull methods for Storm Petrels Hydrobates pelagicus. 100: 682– 686. Bird Study 45: 302– 312. KÖPPEN W. 1948. Climatologia. Mexico City: Ed ROBBINS CS. 1978. Census techniques for forest birds. Fondo Cultura Economica. In: DEGRAED RM (Ed), Management of southern LINDENMAYER DB, M CCARTHY MA AND POPE ML. forest for nongame birds. Atlanta: USDA General 1999. Arboreal marsupial incidence in eucalypt Technical Report SE-GTR 14: 142– 163. patches in south-eastern Australia: a test of Hans- ROSENFIELD RN, BIELEFELDT J, ANDERSON RK ki’s incidence function metapopulation for patch AND SMITH WA. 1985. Taped calls as an aid in occupancy. Oikos 84: 99– 109. locating Cooper’s Hawk nests. Wildlife Soc Bull 13: MARINI MA. 2001. Effects of forest fragmentation on 62– 63. birds of the cerrado region, Brazil. Bird Conserv ROSS JLS AND MOROZ IC. 1997. Mapa Geomorfo- Intern 11: 13– 25. lógico do Estado de São Paulo, escala 1:500.000. Volume 1. São Paulo: Universidade de São Paulo. MARSDEN SJ, W HIFFIN M AND GALETTI M. 2001. Bird diversity and abundance in forest fragments and SCHIECK J. 1997. Biased detection of bird vocaliza- Eucalyptus plantations around an Atlantic forest re- tions affects comparisons of bird abundance among serve, Brazil. Biodiv Conserv 10: 737– 751. forested habitats. Condor 99: 179– 190. An Acad Bras Cienc (2006) 78 (4) 644 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD SHERMAN PT AND EASON PK. 1998. Size determi- UEZU A, METZGER JP AND VIELLIARD JME. 2005. nants in territories with inflexible boundaries: ma- Effects of structural and functional connectivity and nipulation experiments on White-winged Trumpe- patch size on the abundance of seven Atlantic Forest ters’ territories. Ecology 79: 1147– 1159. bird species. Biol Conserv 123: 507– 519. SICK H. 1997. Ornitologia Brasileira. Rio de Janeiro: VELOSO HP, RANGEL FILHO ALR AND LIMA JCA. Editora Nova Fronteira. 1991. Classifi cação da vegetação brasileira adap- tada a um sistema universal. Rio de Janeiro: IBGE. SLIWA A AND SHERRY TW. 1992. Surveying win- tering warbler populations in Jamaica: point counts VIELLIARD J AND SILVA WR. 1990. Nova metodolo- with and without broadcast vocalizations. Condor gia de levantamento quantitativo da avifauna e pri- 94: 924– 936. meiros resultados do interior do Estado de São Pau- lo, Brasil. In: IV ENCONTRO NACIONAL DE ANI- STOTZ DF, FITZPATRICK JM, PARKER III TA AND LHADORES DE AVES, Recife. Universidade Federal MOSKOVITS DK. 1996. Neotropical birds. Ecology Rural de Pernambuco, PE, Brasil, p. 117– 151. and conservation. Chicago: University of Chicago Press. EO. 1979. The composition of avian communi- WILLIS ties in three remanescent woodlots in southern Brazil. SWIFT BL, ORMAN SR AND OZARD JW. 1988. Re- Papéis Avuls Zool 33: 1– 25. sponse of Least Bitterns to tape-recorded calls. Wil- son Bull 100: 496– 499. ZAR JH. 1999. Biostatistical analysis. Upper Saddle River, New Jersey: Prentice Hall International Edi- THOMPSON WL. 2002. Towards reliable bird surveys: tions. accounting for individuals present but not detected. Auk 119: 18– 25. An Acad Bras Cienc (2006) 78 (4) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Anais da Academia Brasileira de Ciências Unpaywall

Efficiency of playback for assessing the occurrence of five bird species in Brazilian Atlantic Forest fragments

Anais da Academia Brasileira de CiênciasDec 1, 2006

Loading next page...
 
/lp/unpaywall/efficiency-of-playback-for-assessing-the-occurrence-of-five-bird-OA9zXV0D6K

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0001-3765
DOI
10.1590/s0001-37652006000400003
Publisher site
See Article on Publisher Site

Abstract

Anais da Academia Brasileira de Ciências (2006) 78(4): 629– 644 (Annals of the Brazilian Academy of Sciences) ISSN 0001-3765 www.scielo.br/aabc Efficiency of playback for assessing the occurrence of five bird species in Brazilian Atlantic Forest fragments 1 1 2 DANILO BOSCOLO , JEAN PAUL METZGER and JACQUES M.E. VIELLIARD Departamento de Ecologia, Instituto de Biociências, Universidade de São Paulo Rua do Matão, 321, Travessa 14, 05508-900 São Paulo, SP, Brasil Departmento de Zoologia, Instituto de Biologia, Universidade Estatual de Campinas Caixa Postal 6109, 13083-970 Campinas, SP, Brasil Manuscript received on April 17, 2006; accepted for publication on April 19, 2006; contributed by JACQUES M.E. VIELLIARD* ABSTRACT Playback of bird songs is a useful technique for species detection; however, this method is usually not stan- dardized. We tested playback effi ciency for fi ve Atlantic Forest birds (White-browed Warbler Basileuterus leucoblepharus, Giant Antshrike Batara cinerea, Swallow-tailed Manakin Chiroxiphia caudata, White- shouldered Fire-eye Pyriglena leucoptera and Surucua Trogon Trogon surrucura) for different time of the day, season of the year and species abundance at the Morro Grande Forest Reserve (South-eastern Brazil) and at thirteen forest fragments in a nearby landscape. Vocalizations were broadcasted monthly at sunrise, noon and sunset, during one year. For B. leucoblepharus, C. caudata and T. surrucura, sunrise and noon were more effi cient than sunset. Batara cinerea presented higher effi ciency from July to October. Playback expanded the favourable period for avifaunal surveys in tropical forest, usually restricted to early morning in the breeding season. The playback was effi cient in detecting the presence of all species when the abundance was not too low. But only B. leucoblepharus and T. surrucura showed abundance values signifi cantly related to this effi ciency. The present study provided a precise indication of the best daily and seasonal periods and a confi dence interval to maximize the effi ciency of playback to detect the occurrence of these forest species. Key words: playback, birds, forest fragments, methodology development, Atlantic forest, São Paulo State. INTRODUCTION the detection of the species presence and accurate in the recognition of their absence. Species occurrence data have often been used to One main challenge is to avoid the problem monitor populations in fragmented habitats and to of “ false absence” (Thompson 2002). Cryptic be- understand the effects of fragmentation on popu- haviour or seasonal and daily variations in species lation survival (Hanski 1994, Clergeau and Burel activity can make this task diffi cult, especially if the 1997, Lindenmayer et al. 1999). This kind of data survey is conducted inside dense habitats such as should be easy to collect, enabling surveys of a large neotropical forests (Catchpole and Slater 1995, Sick number of habitat fragments in a short period of 1997), where visual contact is seriously limited. To time. To be effi cient, these methods must be fast in reduce this problem, surveys may be conducted with the assistance of playback calling. *Member Academia Brasileira de Ciências Correspondence to: Jean Paul Metzger Many studies have revealed that the use of play- E-mail: jpm@ib.usp.br An Acad Bras Cienc (2006) 78 (4) 630 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD back increases the number of contacts during bird 1997). The climate type is temperate and rainy, surveys (Johnson et al. 1981, Rosenfi eld et al. 1985, as Köppen’s (1948) Cfa. Mean monthly tempera- ◦ ◦ Johnson and Dinsmore 1986, Kaufmann 1988, Swift ture ranges from 11 Cto27 C and mean annual et al. 1988, Sliwa and Sherry 1992). Some exper- precipitation is about 1400 mm, with the coldest iments indicated that the playback of conspecifi c and driest months between April and August. The songs is effi cient in inducing aggressive responses original vegetation cover was classifi ed as “ dense from various bird species, allowing their detection montane ombrophilous Atlantic Forest” (Veloso et (Milligan 1966, Møller 1992, Catchpole and Slater al. 1991). Nowadays, the region is composed of 1995). Mosher et al. (1990) pointed out that the nearly 360 fragments of second growth forest that use of playback during bird surveys might not only ranges from 1 to 300 ha. The landscape presents increase the contact rate but also reduce the interval 31% of forest fragments in medium to late succes- needed for the fi rst contact. Other studies, how- sional stages and 6% of early successional forests, ever, pointed out that variations in the abundance totalling 37% natural vegetation. The remaining of birds in the forest might have an effect on the area is composed of agricultural fi elds, pastures, results obtained by playback experiments (Glahn exotic forest plantation and human infrastructure 1974, Mosher and Fuller 1996, Schieck 1997). Al- (Metzger 2003). Five km east of the fragmented though playback techniques are widely used to sur- landscape, over the same geomorphologic unit, lies ◦ ◦ vey birds, few studies have been conducted with the the Morro Grande Forest Reserve (23 39’ – 23 48’ S ◦ ◦ aim of standardizing it. According to Morrell et al. and 46 47’ – 46 55’ W), a 9.400 ha continuous late (1991) and Ratcliffe et al. (1998), the standardiza- second growth forest well connected with the ma- tion of the technique can reduce the operation cost ture rainforest of the Paranapiacaba sierra. and increase the possibility of collecting a large data STUDIED SPECIES set in a short time. These factors become even more important when the objective is to sample a large Five bird species were chosen considering three number of fragments. main criteria. Firstly, they were strictly forest spe- The current study was designed to develop an cies that exhibit territorial behaviour, thus being accurate and simple tool to detect the presence and likely to respond to playback (J.M. Goerck, unpub- absence (occurrence pattern) of strictly forest bird lished data, Stotz et al. 1996). Secondly, in or- species inside a tropical fragmented landscape. Our der to consider species that perceive the landscape specifi c objectives were: i) to verify the time of in different ways and present distinct sensitivity to the day and season of the year when some Atlantic the fragmentation of their habitat, the chosen spe- Forest bird species are more effi ciently detected by cies presented also diverse biological characteris- playback; ii) to test if the effi ciency of the tics. Finally, all species were also non-migratory technique is affected by the abundance of the birds. and can be found all year long at the study sites. These species are (Table I): 1 – White-browed Warbler Basileuterus leucoblepharus, a small Paru- MATERIALS AND METHODS lidae weighing only 20 g, found in montane forests. STUDY REGION It feeds on insects, lives in the understory and is The study region was located on the Atlantic Plateau weakly affected by habitat fragmentation (Uezu et of São Paulo (Brazil), between the municipal dis- al. 2005). 2 – Giant Antshrike Batara cinerea, ◦ ◦ tricts of Cotia and Ibiúna (23 35’ – 23 50’ S and the biggest of the fi ve species, weighing approxi- ◦ ◦ 46 45’ – 47 15’ W). Local relief is characterized by mately 140 g. It is an insectivorous Formicariidae hills with more than 15% of inclination and eleva- that lives below the canopy and benefi ts from frag- tions between 850 and 1100 m (Ross and Moroz mentation in the Cotia region (Uezu et al. 2005). An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 631 TABLE I Selected bird species and their characteristics. In: insectivorous; Fr: frugivorous; Bc: below canopy; Ms: middle storey; Us: understorey. Species Feeding Sensitivity to Family Weigh Habitat (common name) habit fragmentation Basileuterus leucoblepharus Parulidae 20 g In Us Low (White-browed Warbler) Batara cinerea Formicariidae 140 g In Bc + Ms Low (Giant Antshrike) Chiroxiphia caudata Pipridae 25 g Fr Ms + Us Medium (Swallow-tailed Manakin) Pyriglena leucoptera Formicariidae 30 g In Ms + Us Medium (White-shouldered Fire-eye) Trogon surrucura Trogonidae 70 g Fr Bc High (Surucua Trogon) 3 – Swallow-tailed Manakin Chiroxiphia caudata, note shorter; C – The vocalization of C. caudata a Pipridae weighing around 25 g known by its char- is characterized by strong calls usually, but not al- acteristic and intricate sexual behaviour (Foster ways, repeated for several times; D – The song of 1981). The species is frugivorous, inhabits the in- P. leucoptera is very penetrating in the forest and termediate strata of the forest and is moderately af- is characterized by an unvarying sequence of the fected by habitat fragmentation (Uezu et al. 2005). same note; and E – Trogon surrucura sings a con- 4 – White-shouldered Fire-eye Pyriglena leucop- stant series of 14 to 17 notes with the last ones tera, another Formicariidae, weighing about 30 g. lower. This species is of medium sensitivity to fragmen- Song recordings were broadcasted using a tation (Uezu et al. 2005) and lives in the middle Sony TCM-5000EV cassette tape recorder with the and low storeys of the forest searching for insects speaker positioned between one and two meters (Sick 1997). 5 – Surucua Trogon Trogon surrucura, above the ground. The volume of the broadcast was from the family Trogonidae, is a large frugivorous set for human ears to hear it at a maximum distance bird that weights approximately 70 g. Because of of 100 m. The intensity of the songs, measured its size and feeding habits, it is very susceptible to with a Bruel & Kjaer decibel meter positioned at fragmentation (Goerck 1997), being present only in 1 m from the speaker was: 79.2 dB for B. leucoble- large forest fragments (Uezu et al. 2005). pharus, 81.6 dB for B. cinerea, 75.0 dB for C. cau- data, 79.0 dB for P. leucoptera and 79.5 dB for T. surrucura. SONG RECORDINGS DAILY AND SEASONAL VARIATION OF Male bird song or call recordings were obtained PLAYBACK EFFICIENCY from the Arquivo Sonoro Neotropical (ASN) of the Universidade Estadual de Campinas (UNICAMP), This experiment was conducted inside the contin- Brazil. The vocalization of each species is distinc- uous forest of the Morro Grande Forest Reserve tive and with unique characteristics (Fig. 1): A – between June 2000 and May 2001, in three distinct The song of B. leucoblepharus consists of a high- selected sites where all fi ve species were known to pitched scale of descending notes; B – Batara cine- be present and could be easily observed or heard rea vocalizes a long and trilled song with the last all year long. In order to reduce the variance in An Acad Bras Cienc (2006) 78 (4) 632 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD Fig. 1 – Sonograms of the vocalizations used for each studied species for the playback surveys. A: B. leucoblepharus; B: B. cinerea; C: C. caudata; D: P. leucoptera; E: T. surrucura. Each sonogram is presented in a distinct scale of time (in seconds) and frequency (in kHz). Extracted from the Arquivo Sonoro Neotropical (ASN), UNICAMP. environmental conditions between sites, their selec- values for a single species may vary from zero to tion was guided according to the forest type and the total number of playback sessions conducted at structure. Playback sessions were conducted every a given month or day period. In order to allow eas- separate month and for different times of the day. ier comparisons, these values were also standardised The sum of detected bird responses to the playback as percentages of positive response. within each tested month or period of the day was To assess detection effi ciency in relation to termed detection effi ciency. Therefore, effi ciency time of the day, three different test periods were An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 633 set: 1) during the two hours after sunrise; 2) at about late responsive birds. On the other hand, it would two hours around noon and 3) during the two hours not be convenient to wait for periods longer than the before sunset. At each period, three playback ses- mean latency time of the species added to its stan- sions were broadcasted for each one of the species dard deviation, as it is unlikely that a bird responds at three different sampling points (nine playback after this and the species could then be documented sessions at nine different points per day). Each play- as absent for the tested point without further loss back session was composed by one minute of play- of time. After the end of the last species broadcast, back, intercalated by thirty seconds of silence, to the observer (DB) waited for fi ve additional minutes account for responsive birds (Johnson et al. 1981), to account for birds with long latency of response. until fi ve minutes were completed (Morrell et al. Playback was not carried out during rainy days due 1991, Sliwa and Sherry 1992). To avoid the sti- to noise interference (Johnson et al. 1981). mulation of the same bird more than once during These procedures were repeated three days the same period, sampling points were set at least a month for twelve consecutive months. For each 200 m away from each other (Sliwa and Sherry species, 27 playback sessions were conducted 1992). The sequence of species broadcasted was monthly, totalling 324 tests apiece, 108 at each pe- set randomly at each point. To prevent habituation riod of the day. Differences in birds detection effi - ciency along the day were verifi ed through G-tests of the birds to the recorded songs no sampling point was tested more than once a month (W.R. Silva, un- in contingency tables, and when signifi cant differ- published data). ences were detected (P< 0.05) an a posteriori χ test At each session, the response by the birds was (Zar 1999) was used. Detection effi ciency through- recorded as negative (no response) or positive. A out the year was tested with circular statistics, com- response was considered positive when changes in paring the twelve months through the Rayleigh’s test the behaviour of the birds were detected after the (Zar 1999), with a signifi cance of P< 0.05. playback begun. These changes were usually rec- PLAYBACK DETECTION EFFICIENCY IN RELATION ognized as increased rate of vocalization and/or al- TO BIRD ABUNDANCE terations in the movement of the birds towards the sound source. If an individual was detected clos- The outcomes of the experiment described above ing in and/or circling around the sound source in were used to determine the best period of the day silence during or after the playback, it was also doc- and season of the year when the relationship be- umented as a responsive bird. Birds detected be- tween playback effi ciency and bird abundance fore the beginning of the playback but that did not should be tested. To conduct that test, thirteen for- show any changes in their behaviour demonstrating est fragments ranging from 13 to 275 ha were stud- specifi c interest for the played song during or af- ied. Fragment selection was conducted in order to ter the broadcast were noted as negative responders reproduce Reserve sites environmental conditions. (no response). To identify how long it takes for each Inside each fragment, a grid of fi ve to nine sam- species to respond to the playback, the time elapsed pling points (depending on fragments area) located between the beginning of the playback and the de- at least 200 m apart was set. The sites used before tection of positive response, termed as the latency in the Morro Grande Forest Reserve were divided time, was also noted. This was made in order to into three different sets of nine points and sampled identify which is the minimum time that someone again. The Reserve sets of nine points were orig- should wait for a response to take place before stat- inally placed at least 2 km apart of each other in ing that the species is absent during that session. order to simulate the spatial confi guration of the A short waiting time may lead to a decrease in the fragments. Because the latency time of each spe- quantity of contacts by impeding the detection of cies was already known, the species broadcasting An Acad Bras Cienc (2006) 78 (4) 634 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD sequence was set by this parameter, in order to in- visits to the fragments were needed for the playback crease bird detection probabilities. Consequently, to acquire a similar occurrence pattern as obtained the fi rst broadcasted species was the one with high- by point counts, the three above explained meth- est latency time and the last with the shortest. Each ods were conducted for all possible combinations of study site (fragments or Reserve sites) was sampled one, two, three and four survey days. for four different non-consecutive days at the best season of the year and time of the day. In each RESULTS one of the four days, three sampling points were DETECTION EFFICIENCY IN RELATION TO THE TIME randomly selected for the playback trials, totalising OF DAY AND SEASON OF THE YEAR twelve playback sessions per site and species for each one of the sixteen sites. Each site was sur- The species with the highest percentages of pos- veyed with playback twice at sunrise and twice at itive responses inside the Reserve was B. leuco- noon, between 1 August and 29 September 2001. blepharus (Table II). Mean latency time varied be- The abundance information was estimated tween species from 2 to 8.3 minutes (Table III). The between May 2000 and April 2001 through pas- analysis of variance showed that latency times were sive point counts (Blondel et al. 1970, Vielliard signifi cantly different among species (F = 41.16; and Silva 1990), at the same points used for the P< 0.01). The a posteriori test indicated that the playback (Uezu et al. 2005). This was the best latency time of B. cinerea was signifi cantly longer available abundance estimate for all studied spe- than any other tested species (P< 0.01). The latency cies and could be used as a precise occurrence data presented by C. caudata was different from that of at these sites. All fragments were sampled during T. surrucura (P< 0.01) but not of B. leucoblepharus early morning for one day every two months. At (P = 0.80). No signifi cant differences were found each day, one fragment was selected and fi ve sam- between P. leucoptera and T. surrucura (P = 0.58) ple points were randomly chosen in it. Each point or between P. leucoptera and C. caudata (P = 0.27). was observed for 20 minutes to account for present Three species, B. leucoblepharus, C. caudata birds. By dividing the number of contacts, visual and T. surrucura showed signifi cant differences in and/or auditory, by the number of sampled points detection effi ciency according to the period of the an Index of Point Abundance (IPA) was then calcu- day (Table III). The a posteriori χ test indicated lated for each species in each fragment. that the effi ciency at sunrise and noon was signif- To test if detection effi ciency was related to icantly higher in comparison with the period just bird abundance in the forest, three different methods before sunset (Table III). were used. The playback-determined occurrence According to the Rayleigh’s test, the only spe- of the birds (dependent variable) was related with cies to present a mean season of higher effi ciency their abundance (IPA, independent variable) in a was B. cinerea (z = 6.041; P< 0.01, Fig. 2). 0.05,88 logistic regression (Ratcliffe et al. 1998). A Pearson All other species had uniform distributions of effi - correlation analysis (Zar 1999) was done between ciency throughout the twelve tested months (Fig. 2). the response rates to the playback in each fragment The mean period of higher effi ciency for B. cinerea and the IPA. The response rate was defi ned for each was July and August, but considering the low con- species as the quantity of positive responses in each centration of responses (r = 0.262), one can perceive site divided by twelve, the total number of playback that this is only the central tendency of a larger pe- sessions conducted after four survey days. Finally, riod of fi ve consecutive months ranging from June the percentage of sites where the playback correctly to October. Those fi ve months presented no signif- attested the presence of the birds (defi ned by point icant difference between each other along the given counts results) was calculated. To verify how many period (G = 3.807; P> 0.25). Responses from June An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 635 TABLE II Percentages of detected positive responses to playback according to each tested period of the day and month for the five studied species. B. leucoblepharus B. cinerea C. caudata P. leucoptera T. surrucura Periods Sunrise 77.8 27.8 61.1 39.8 53.7 of the Noon 72.2 25.9 61.1 28.7 45.4 day Sunset 58.3 27.8 41.7 36.1 25.9 January 48.1 14.8 59.3 40.7 25.9 February 70.4 2.,2 55.6 44.4 22.2 March 81.5 3.7 44.4 33.3 55.6 April 70.4 29.6 51.9 22.2 55.6 May 66.7 22.2 59.3 29.6 44.4 June 70.4 48.1 77.8 29.6 59.3 Months July 70.4 40.7 33.3 25.9 51.9 August 59.3 25.9 33.3 33.3 18.5 September 88.9 40.7 48.1 48.1 25.9 October 66.7 48.1 70.4 33.3 55.6 November 70.4 14.8 63.0 40.7 40.7 December 70.4 14.8 59.3 37.0 44.4 TABLE III Mean latency values (in minutes) and playback positive (pos) and negative (neg) responses according to the period of the day for the five studied species (N= 108 for each species and period) inside the Morro Grande Forest Reserve. P: significance level. Playback responses G test a posteriori χ Species Latency N* Sunrise Noon Sunset sr × nn × st (sr + nn) × st sr × nn 2 2 (sr) (nn) (st) G (P) χ (P) χ (P) (2) c(1) c(1) pos neg pos neg pos neg 2.0±1.9 10.09 8.65 0.62 B. leucoblepharus 84 24 78 30 63 45 225 (P<0.01) (P<0.01) (P>0.25) 8.3±8.1 0.12 B. cinerea 30 78 28 80 30 78 88 (P>0.9) 2.6±2.6 10.99 10.21 0.02 C. caudata 66 42 66 42 45 63 177 (P<0.01) (P<0.01) (P>0.99) 3.6±4.1 3.08 P. leucoptera 43 65 31 77 39 69 113 (P>0.1) 4.2±5.1 18.58 15.56 1.18 T. surrucura 58 50 49 59 28 80 135 (P<0.01) (P<0.01) (P>0.9) * The numbers of observations (N) are unequal between species because latency times could only be accounted after positive responses. An Acad Bras Cienc (2006) 78 (4) 636 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD Fig. 2 – Distribution of positive responses (gray dots) to the playback during one year of surveys and Rayleigh’s test results. A: B. cinerea; B: B. leucoblepharus; C: T. surrucura; D: C. caudata; E: P. leucoptera. N: number of playback sessions; r: concentration level, P: signifi cance level. to October were also compared to responses from occurrence only when IPA was less than 0.05. the rest of the year, confi rming that this was the Logistic regression was calculated only for T. period of highest playback effi ciency to detect B. surrucura since it was the only species not present cinerea in the forest (McNemar’s test, χ = 5.1127; at all sampled sites. One single survey day was c1 P< 0.025). enough to incur into a signifi cant regression, indi- cating a strong relation between the two tested vari- DETECTION EFFICIENCY ACCORDING TO THE ables (Table V). ABUNDANCE OF THE BIRDS INSIDE The correlation analyses showed contrasting FOREST FRAGMENTS results. For three species, B. cinerea, C. caudata and P. leucoptera, almost no correlation between Three species were detected in all sampled sites playback response rate and IPA were observed for by both the playback and the point counts (Table any of the 16 possible combinations (Table VI). For IV). The two methods presented similar species oc- currence results, except for two fragments where B. leucoblepharus and T. surrucura, nearly all pos- T. surrucura and P. leucoptera were detected only sible combinations had signifi cant positive correla- tions, showing a strong influence of the abundance by point counts but not by playback (Table IV). In on the detection by playback. both cases, species abundance (IPA) was low, indi- For all species, the percentage of sites where cating that the playback was defi cient in detecting An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 637 TABLE IV Number of positive responses for each study site obtained after twelve playback sessions per species and abundance values (IPA) in parenthesis (Uezu et al. 2005). Cases where IPA and playback results disagreed are marked with *. Fragments B. leucoblepharus B. cinerea C. caudata P. leucoptera T. surrucura Mioko5 (0.71) 7 (0.73) 9 (1.03) 5 (0.14) 1 (0.11) Takimoto 5 (0.89) 3 (0.53) 9 (1.31) 8 (0.26) 6 (0.23) Pedro 8 (1.43) 6 (0.70) 10 (1.60) 5 (0.34) 3 (0.11) Pedroso 4 (1.69) 6 (0.30) 5 (1.14) 2 (0.37) 1 (0.34) Zezinho 2 (0.31) 5 (0.30) 10 (1.03) 3 (0.26) 4 (0.23) Lila 3 (0.34) 5 (0.13) 8 (1.03) 3 (0.17) 0 (0.00) Dito-André 6 (0.89) 9 (0.63) 9 (0.77) 5 (0.20) 0 (0.00) Beto 6 (0.96) 9 (0.50) 7 (1.18) 5 (0.32) 0 (0.00) Lacerda 4 (0.25) 10 (0.38) 9 (1.46) 0 (0.04)* 0 (0.00) Godoy9 (1.00) 6 (0.50) 8 (1.18) 3 (0.43) 0 (0.00) Nelson 9 (1.69) 8 (0.43) 8 (0.63) 7 (0.54) 0 (0.00) Reizinho 5 (1.43) 6 (0.37) 5 (0.94) 3 (0.29) 0 (0.00) Agostinho 11 (1.40) 3 (0.47) 10 (0.89) 6 (0.23) 0 (0.03)* ReserveA10 (2.09) 4 (0.23) 7 (1.11) 2 (0.34) 6 (0.23) ReserveB6 (1.34) 1 (0.27) 8 (1.40) 1 (0.31) 8 (0.51) ReserveC8 (1.37) 5 (0.23) 10 (1.26) 8 (0.26) 9 (0.17) TABLE V Logistic regression between species occurrences according to playback results and to abundance index (IPA) for T. surrucura. “Combination” indicates which survey days (a, b, c, d) were utilized in each test. P: significance level. B0 is the estimated logistic coefficient. Quantity of survey days Combination Logistic regression B0 P a – 3.40 0.02 b – 2.31 < 0.01 c – 2.25 < 0.01 d – 3.08 < 0.01 a+b – 2.31 < 0.01 a+c – 2.51 < 0.01 a+d – 3.08 < 0.01 b+c – 6.12 < 0.01 b+d – 2.51 < 0.01 c+d – 6.12 < 0.01 a+b+c – 6.12 < 0.01 a+b+d – 2.31 < 0.01 a+c+d – 6.12 < 0.01 b+c+d – 6.12 < 0.01 4 a+b+c+d – 6.12 < 0.01 An Acad Bras Cienc (2006) 78 (4) 638 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD TABLE VI Pearson correlations coefficients (r ) between the efficiency of the playback and the abundance index values (IPA) for all species and all possible combinations of survey days (a, b, c, d). Quantity of B. leucoble- Combinations B. cinerea C. caudata P. leucoptera T. surrucura survey days pharus a 0.39* 0.03 0.05 0.00 0.26* b 0.23 0.10 0.08 0.34* 0.38* c 0.47* 0.05 0.10 0.00 0.36* d 0.05 0.03 0.01 0.00 0.47* a+b 0.50* 0.16 0.11 0.17 0.49* a+c 0.61* 0.05 0.00 0.00 0.36* a+d 0.34* 0.06 0.06 0.00 0.40* b+c 0.42* 0.16 0.00 0.09 0.57* b+d 0.16 0.07 0.09 0.15 0.48* c+d 0.25* 0.07 0.02 0.00 0.52* a+b+c 0.60* 0.14 0.02 0.07 0.54* a+b+d 0.37* 0.12 0.12 0.08 0.50* a+c+d 0.48* 0.07 0.00 0.00 0.45* b+c+d 0.29* 0.12 0.00 0.06 0.57* 4 a+b+c+d 0.47* 0.13 0.03 0.05 0.54* *P< 0.05. the playback correctly attested the bird presence in- derived from their complex courting behaviour, in- creased with the sampling effort (Fig. 3). For two dividuals shall be more prone to react to intruders species, B. leucoblepharus and B. cinerea, maxi- (in this case the broadcasted song) during this pe- mum success (100% of correct presence detection) riod, leading to higher method effi ciency. was obtained with three survey days. Two days were On the other hand, some passive point counts enough to obtain maximum success for C. caudata. studies tend to disagree with this result, consider- The remaining two species did not reach 100% of ing sunrise as the best period for bird surveys, fol- accuracy: the highest success for P. leucoptera was lowed by sunset and at last noon (Järvinen et al. around 95% and for T. surrucura around 90%. 1977, Robbins 1978, Bibby et al. 1993). G.S. Be- tini (unpublished data) detected that T. surrucura, P. leucoptera and Basileuterus culicivorus, a spe- DISCUSSION cies closely related to B. leucoblepharus, were PLAYBACK DETECTION EFFICIENCY:TIME OF THE spontaneously more active during early morning in DAY AND SEASON INFLUENCES contrast to the rest of the day, when their vocal ac- tivity was lower. For three species the best periods of the day to conduct playback census were just after sunrise and These results suggested that for these species at noon. This is in compliance with the daily peak the playback increased the time period when bird of spontaneous activity of some species, such as C. census should be conducted to acquire the high- caudata. This species presents a period of higher est effi ciency from sunrise to the entire morning, spontaneous activity between sunrise and 15:00, up to noon. Otherwise, the playback may induce the main period of sexual activity (Foster 1981). Be- birds to become detectable at any time, maintain- cause of the strong vocal interaction of the males, ing the quantity of contacted individuals equal along An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 639 Fig. 3 – Percentages (±SD) of sites where the playback correctly detected the presence of each species according to the number of survey days. Sites where IPA=0.0 were omitted. the day, in contrast to the census without playback. The only species with signifi cant seasonal vari- This was observed for P. leucoptera at the present ation in playback response was B. cinerea, which study and for Dendroica caerulescens in Jamaica should be surveyed between June and October to (Sliwa and Sherry 1992), which leads to the pattern acquire maximum census effi ciency. This is in ac- of no daily variation in bird detection presented by cordance with the pattern found by A. Uezu (unpub- these species. Of course, in this case the playback lished data), who recorded a more intense peak of had also increased the survey period to further than spontaneous vocal activity of this species between merely sunrise. July and September. The annual period of greater An Acad Bras Cienc (2006) 78 (4) 640 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD playback effi ciency may be linked to the reproduc- those cases, playback will be effi cient to identify tive biology of the species. Many authors use the only close birds, since the playback can only reach reproductive season to proceed with avifaunal sur- nearby individuals. But, even if the song functional veys (Dawson 1981, Mayfi eld 1981, Greenwood et structure is preserved (Aubin et al. 2002), allowing al. 1995, Clergeau and Burel 1997, Ratcliffe et al. birds farther away to recognize their specifi c vocal- 1998, Drapeau et al. 2000), when individuals are ization, the observer may not have the same ability, reinforcing their dominance or establishing new ter- decreasing the quantity of contacts proportionally ritories and consequently are more active. For this to the abundance. reason, they may be more prone to confront intrud- The signifi cant relationship found for Trogon ers, increasing the chances of response to playback. surrucura is probably an effect of the rarity of the However, contrary to the expected, for four of species. According to several researchers (Willis our species there was no seasonal variation in play- 1979, Christiansen and Pitter 1997, Marini 2001, back effi ciency. Even in non-reproductive months, Marsden et al. 2001), trogons are highly affected the birds kept responding with the same intensity by patch size, being present only in large fragments to the playback, producing no signifi cant variation or reserves and in low abundance, as detected in in the quantity of contacts along the tested year. the current study (Uezu et al. 2005). Usually, big This suggests that playback may be used as a sur- canopy frugivorous birds are rare and need large vey method for these bird species all year long and areas to search for food (Goerck 1997, Price et. al. not only during the reproductive season. This might 1999). The size of each individual territory was thus be related to the active stimulation of the birds or presumably large in relation to the distance between may come from a strong bond of the individuals to sample points. As the abundance increases, territo- their territories. ries might overlap or decrease in size (Sherman and Eason 1998) and contact with the birds may become PLAYBACK DETECTION EFFICIENCY ACCORDING TO more frequent, increasing detection rates along with BIRD ABUNDANCE the number of present birds in the forest. On the other hand, there was a lack of corre- Playback was a very effi cient technique to detect lation between the response rate and IPA for three the presence of these species in all sampled sites species. In the case of C. caudata, the lack of cor- where the point counts detected the occurrence of relation may be an effect of the sexual behaviour of the birds. It incurred in correct presence/absence the species. Male manakins are constantly engaged detection in at least 98% of the studied cases. How- in vocal duels for dominance in the hierarchy of the ever, there was not always a direct relationship be- group (Foster 1981) and present a natural aggres- tween the abundance of the birds and the detection sive behaviour that makes the individuals extremely effi ciency. responsive, increasing the detection rate after play- For B. leucoblepharus and T. surrucura there back. So, even in low abundance sites, the quantity was a signifi cant positive relationship between the of contacts shall be high. response rate and IPA, indicating that playback For B. cinerea, the lack of correlation was not capacity to identify sites where the species is pre- due to the behaviour of the individuals but to the sent increased with the abundance. According to properties of its vocalization. This species song Schieck (1997) vocalizations with frequencies is very loud and the responsive individual was fre- higher than 2 kHz may suffer strong attenuation quently far away from the playback source, indicat- in the forest. Aubin et al. (2002, 2004) stated ing that the song could travel long distances in the that the song of B. leucoblepharus is particularly forest and still be recognizable for the receiver. As high-pitched and susceptible to a strong attenua- a result, individuals might still be able to identify tion when propagating through dense vegetation. In An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 641 and respond to the broadcasted song even if they tion of correct detection did not reach 100% after are further than 100 m from the sound source. Be- four visits due to fragments with very low abun- cause of that, even in places with low abundance, dance. There was a threshold of approximately 0.05 it is possible to locate distant individuals, creating of IPA, below which the bird detection through the non-correlation pattern. playback is improbable for a sampling effort of four In the case of P. leucoptera no results pointed days. An increase of the sampling effort could solve out to any clear cause to the lack of correlation be- this problem. tween detection effi ciency and abundance. The de- tected pattern of response may be related not only CONCLUSION to bird abundance. For instance, if for any reason The results of the present study pointed out to a the access to food resources decline at a certain site, large variation in Atlantic Forest bird species re- individual birds may tend to become more aggres- sponse to playback and to the importance of method- sive in order to defend their territories (Armstrong ological studies in order to standardize census meth- 1991, Carrascal et al. 1998) and therefore more re- ods. Future researchers should try to understand sponsive to playback. In places of higher resource each species behaviour before applying the method. availability, even if the species is more abundant, If known, the latency time and the variation of play- there is no need to defend each territory so strongly back responses according to the time of the day, and the response rate may then decrease. Therefore, season of the year, abundance and sampling effort the abundance of a species did not appear to be the could help to reduce the problem of “ false absence” sole or main factor to influence the occurrence de- detections and maximize the effi ciency of the play- tection effi ciency of these birds. According to each back to collect occurrence data, assuring good re- species, other factors may modify this relationship. sults for quick access surveys. When compared to passive methods, the play- EFFECTS OF SAMPLING EFFORT back expanded the favourable time to conduct avi- According to Bibby et al. (2000) the chances of faunal surveys for these Atlantic Forest birds from detecting new individuals should increase along the fi rst hours of the day to the entire morning, up with the sample effort, independently of the sam- to noon. Also the season of the year is not an im- pling method. That was the case for all species, pediment to census these birds with playback. Four since the proportion of correct presence detection of the fi ve studied species might be surveyed with through playback increased continuously with the the same effi ciency of detection all year long. Bar- number of survey days. The main question here is ring when the abundance is too low, the playback how large should the sampling effort be to obtain an technique can detect with only three survey days accurate incidence data set, without wasting more per site the occurrence pattern of these birds as ac- time than necessary at a single site, in order to be curately as the passive point counts. These patterns able to quickly sample a large number of sites? Con- may also be applied to other bird species not studied sidering a confi dence threshold of 90%, one visit in the present work. was enough to accurately detect C. caudata, while two visits were necessary for B. leucoblepharus and ACKNOWLEDGMENTS B. cinerea, and three for P. leucoptera.For T. sur- rucura, four visits were necessary to obtain a con- This research was part of the thematic project fi dence value close to 90%. For a confi dence of “ Biodiversity conservation in fragmented land- 100%, two visits were necessary for C. caudata and scapes at the Atlantic Plateau of São Paulo” and was three for B. leucoblepharus and B. cinerea. In the fi nancially supported by the Programa de Pesquisas case of P. leucoptera and T. surrucura, the propor- em Caracterização, Conservação e Uso Sustentável An Acad Bras Cienc (2006) 78 (4) 642 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD da Biodiversidade da Fundação de Amparo à Pes- REFERENCES quisa do Estado de São Paulo (BIOTA-FAPESP) ARMSTRONG DP. 1991. Aggressiveness of breeding (n. 99/05123-4; master fellowship for DB territorial honeyeaters corresponds to seasonal n. 00/01358-6). We would like to thank Ilse Storch, changes in nectar availability. Behav Ecol Sociobiol Wesley R. Silva and José Carlos Motta Jr who re- 29: 103– 111. viewed previous versions of this manuscript and AUBIN T, MATHEVON N, VIELLIARD JME AND SILVA José Roberto Nali along with all the staff from Com- ML. 2002. How an inadequate acoustic signal suc- panhia de Saneamento Básico do Estado de São ceeds in the extreme environment of a tropical for- Paulo (SABESP) and the Morro Grande Forest est? The song of the White-browed Warbler. Beijing, Reserve for their hospitality and support. China: Abstracts XXIII Int Ornith Congr. AUBIN T, MATHEVON N, SILVA ML, VIELLIARD JME RESUMO AND SEBE F. 2004. How a simple and stereotyped A técnica de play-back é muito útil para a detecção de acoustic signal transmits individual information: the aves, mas este método geralmente não é padronizado. Sua song of the White-browed Warbler Basileuterus leu- efi ciência em atestar a ocorrência de cinco espécies de coblepharus. An Acad Bras Cienc 76: 335– 344. aves da Mata Atlântica (Pula-pula-assobiador Basileute- BIBBY CJ, BURGESS ND AND HILL DA. 1993. Bird rus leucoblepharus, Batará Batara cinerea, Tangará Chi- Census Techniques. San Diego: Academic Press. roxiphia caudata, Olho-de-fogo Pyriglena leucoptera e BIBBY CJ, BURGESS ND, HILL DA AND MUSTOE nd Surucuá-de-barriga-vermelha Trogon surrucura) foi ana- SH. 2000. Bird Census Techniques. 2 ed., San lisada de acordo com o horário do dia, estação do ano e Diego: Academic Press. abundância das espécies na Reserva Florestal do Morro BLONDEL J, FERRY C AND FROCHOT B. 1970. La Grande (São Paulo, Brasil) e em treze fragmentos flo- méthode des indices ponctuels d’ abondance (I.P.A.) ou des relevés d’ avifaune par “ stations d’ écoute”. restais de uma paisagem adjacente. Gravações das vo- Alauda 38: 55– 71. calizações foram emitidas mensalmente ao amanhecer, CARRASCAL LM, SENAR JC, MOSETICH I, URIBE meio do dia e entardecer, durante um ano. Para B. leu- F AND DOMENECH J. 1998. Interactions among coblepharus, C. caudata e T. surrucura, o amanhecer e o environmental stress, body condition, nutritional sta- meio do dia apresentaram-se mais efi cientes que o entar- tus, and dominance in Great Tits. Auk 115: 727– 738. decer. Batara cinerea apresentou maior efi ciência de ju- CATCHPOLE CK AND SLATER PJB. 1995. Bird Song: lho a outubro. O uso do play-back aumentou o período Biological themes and variations. Cambridge: Cam- de maior efi ciência dos levantamentos de aves em matas bridge University Press. tropicais, comumente restrito ao período da manhã du- CHRISTIANSEN MM AND PITTER E. 1997. Species rante a estação reprodutiva. A técnica utilizada foi efi - loss in a forest bird community near Lagoa Santa ciente na detecção da ocorrência de todas as espécies, in Southeastern Brazil. Biol Conserv 80: 23– 32. salvo quando sua abundância se apresentou muito baixa. CLERGEAU P AND BUREL F. 1997. The role of spatio- Porém, apenas B. leucoblepharus e T. surrucura apre- temporal patch connectivity at the landscape level: sentaram abundância signifi cativa e positivamente rela- an example in a bird distribution. Landscape and cionada com essa efi ciência. O presente estudo propor- Urban Planning 38: 37– 43. ciona uma indicação precisa sobre o melhor horário do dia DAWSON DG. 1981. Counting birds for a relative mea- e época do ano, assim como um intervalo de confi ança, sure (index) of density. In: RALPH CJ AND SCOTT com o intuito de aumentar a efi ciência do play-back em JM (Eds), Estimating the numbers of terrestrial detectar a ocorrência dessas espécies. birds. Studies in Avian Biology 6: 12– 16. Palavras-chave: play-back, aves, fragmentos florestais, DRAPEAU P, LEDUC A, GIROUX JF, SAVARD JPL, desenvolvimento metodológico, Mata Atlântica, Estado BERGERON Y AND VICKERY WL. 2000. Land- de São Paulo. scape-scale disturbances and changes in bird com- An Acad Bras Cienc (2006) 78 (4) EFFICIENCY OF PLAYBACK FOR ASSESSING BIRD OCCURRENCE 643 munities of boreal mixed-wood forests. Ecol Monogr MAYFIELD HF. 1981. Problems in estimating population 70: 423– 444. size through counts of singing males. In: RALPH CJ AND SCOTT JM (Eds), Estimating the numbers of FOSTER MS. 1981. Cooperative behaviour and social terrestrial birds. Studies in Avian Biology 6: 220– organization of the Swallow-tailed Manakin (Chiro- xiphia caudata). Behav Ecol Sociobiol 9: 167– 177. METZGER JP (Coord). 2003. Biodiversity conserva- GLAHN JF. 1974. Study of breeding rails with rec- tion in fragmented landscapes at the Atlantic Plateau orded calls in north-central Colorado. Wilson Bull of São Paulo. São Paulo: Scientifi c report to the 86: 206– 214. BIOTA-FAPESP program. GOERCK JM. 1997. Patterns of rarity in the birds of the MILLIGAN M. 1966. Vocal responses of White-crowned Atlantic Forest of Brazil. Conserv Biol 11: 112– 118. Sparrows to recorded songs of their own and another GREENWOOD JJD, BAILLIE SR, GREGORY RD, species. Anim Behav 14: 356– 361. PEACH WJ AND FULLER RJ. 1995. Some new MØLLER AP. 1992. Interspecifi c response to playback approaches to conservation monitoring of British of bird song. Ethology 90: 315– 320. breeding birds. Ibis 137: 16– 28. MORRELL TE, YAHNER RH AND HARKNESS WL. HANSKI I. 1994. Patch-occupancy dynamics in frag- 1991. Factors affecting detection of Great Horned mented landscapes. Trends Ecol Evol 9: 131– 135. Owls by using broadcast vocalizations. Wildlife Soc JÄRVINEN O, VÄISÄNEN RA AND HAILA Y. 1977. Bull 19: 481– 488. Bird census results in different years, stages of the MOSHER JA AND FULLER MR. 1996. Surveying wood- breeding season and times of the day. Ornis Fennica land hawks with broadcasts of Great Horned Owl 54: 108– 118. vocalizations. Wildlife Soc Bull 24: 531– 536. JOHNSON RR AND DINSMORE JJ. 1986. The use of MOSHER JA, FULLER MR AND KOPENY M. 1990. tape-recorded calls to count Virginia Rails and Soras. Surveying woodland raptors by broadcast of conspe- Wilson Bull 98: 303– 306. cifi c vocalizations. J Field Ornith 61: 453– 461. JOHNSON RR, BROWN BT, HAIGHT LT AND SIMPSON PRICE OF, WOINARSKI JZ AND ROBINSON D. 1999. JM. 1981. Playback recording as a special avian Very large area requirements for frugivorous birds censusing technique. In: RALPH CJ AND SCOTT in monsoon rainforests of the Northern Territory, JM (Eds), Estimating the numbers of terrestrial birds. Australia. Biol Conserv 91: 169– 180. Studies in Avian Biology 6: 68– 75. RATCLIFFE N, VUGHAN D, WHYTE C AND SHEP- KAUFMANN GW. 1988. The usefulness of taped Spot- HERD M. 1998. Development of playback census less Crake calls as a census technique. Wilson Bull methods for Storm Petrels Hydrobates pelagicus. 100: 682– 686. Bird Study 45: 302– 312. KÖPPEN W. 1948. Climatologia. Mexico City: Ed ROBBINS CS. 1978. Census techniques for forest birds. Fondo Cultura Economica. In: DEGRAED RM (Ed), Management of southern LINDENMAYER DB, M CCARTHY MA AND POPE ML. forest for nongame birds. Atlanta: USDA General 1999. Arboreal marsupial incidence in eucalypt Technical Report SE-GTR 14: 142– 163. patches in south-eastern Australia: a test of Hans- ROSENFIELD RN, BIELEFELDT J, ANDERSON RK ki’s incidence function metapopulation for patch AND SMITH WA. 1985. Taped calls as an aid in occupancy. Oikos 84: 99– 109. locating Cooper’s Hawk nests. Wildlife Soc Bull 13: MARINI MA. 2001. Effects of forest fragmentation on 62– 63. birds of the cerrado region, Brazil. Bird Conserv ROSS JLS AND MOROZ IC. 1997. Mapa Geomorfo- Intern 11: 13– 25. lógico do Estado de São Paulo, escala 1:500.000. Volume 1. São Paulo: Universidade de São Paulo. MARSDEN SJ, W HIFFIN M AND GALETTI M. 2001. Bird diversity and abundance in forest fragments and SCHIECK J. 1997. Biased detection of bird vocaliza- Eucalyptus plantations around an Atlantic forest re- tions affects comparisons of bird abundance among serve, Brazil. Biodiv Conserv 10: 737– 751. forested habitats. Condor 99: 179– 190. An Acad Bras Cienc (2006) 78 (4) 644 DANILO BOSCOLO, JEAN PAUL METZGER and JACQUES M.E. VIELLIARD SHERMAN PT AND EASON PK. 1998. Size determi- UEZU A, METZGER JP AND VIELLIARD JME. 2005. nants in territories with inflexible boundaries: ma- Effects of structural and functional connectivity and nipulation experiments on White-winged Trumpe- patch size on the abundance of seven Atlantic Forest ters’ territories. Ecology 79: 1147– 1159. bird species. Biol Conserv 123: 507– 519. SICK H. 1997. Ornitologia Brasileira. Rio de Janeiro: VELOSO HP, RANGEL FILHO ALR AND LIMA JCA. Editora Nova Fronteira. 1991. Classifi cação da vegetação brasileira adap- tada a um sistema universal. Rio de Janeiro: IBGE. SLIWA A AND SHERRY TW. 1992. Surveying win- tering warbler populations in Jamaica: point counts VIELLIARD J AND SILVA WR. 1990. Nova metodolo- with and without broadcast vocalizations. Condor gia de levantamento quantitativo da avifauna e pri- 94: 924– 936. meiros resultados do interior do Estado de São Pau- lo, Brasil. In: IV ENCONTRO NACIONAL DE ANI- STOTZ DF, FITZPATRICK JM, PARKER III TA AND LHADORES DE AVES, Recife. Universidade Federal MOSKOVITS DK. 1996. Neotropical birds. Ecology Rural de Pernambuco, PE, Brasil, p. 117– 151. and conservation. Chicago: University of Chicago Press. EO. 1979. The composition of avian communi- WILLIS ties in three remanescent woodlots in southern Brazil. SWIFT BL, ORMAN SR AND OZARD JW. 1988. Re- Papéis Avuls Zool 33: 1– 25. sponse of Least Bitterns to tape-recorded calls. Wil- son Bull 100: 496– 499. ZAR JH. 1999. Biostatistical analysis. Upper Saddle River, New Jersey: Prentice Hall International Edi- THOMPSON WL. 2002. Towards reliable bird surveys: tions. accounting for individuals present but not detected. Auk 119: 18– 25. An Acad Bras Cienc (2006) 78 (4)

Journal

Anais da Academia Brasileira de CiênciasUnpaywall

Published: Dec 1, 2006

There are no references for this article.