Canadian Journal of Zoology

Density, habitat use, and activity patterns of a vulnerable population of solitarius (Vieillot, 1819) in a Brazilian Atlantic Forest fragment

Journal: Canadian Journal of Zoology

Manuscript ID cjz-2017-0153.R1

Manuscript Type: Article

Date Submitted by the Author: 05-Sep-2017

Complete List of Authors: Ferreguetti, Átilla; Universidade do Estado do Rio de Janeiro, Ecology Pereira-Ribeiro,Draft Juliane; Universidade do Estado do Rio de Janeiro, Ecology Tomas, Walfrido; Embrapa Pantanal, Wildlife laboratory Bergallo, Helena; Universidade do Estado do Rio de Janeiro, Ecology Rocha, Carlos Frederico; Universidade do Estado do Rio de Janeiro, Ecology

Detectability, Distance sampling, Line-transect, Occupancy modelling, Keyword: Reserva Natural Vale, Solitary , Tinamus solitarius

https://mc06.manuscriptcentral.com/cjz-pubs Page 1 of 29 Canadian Journal of Zoology

Density, habitat use, and activity patterns of a vulnerable population of Tinamus

solitarius (Vieillot, 1819) in a Brazilian Atlantic Forest fragment

1* 1 2 Átilla Colombo Ferreguetti , Juliane PereiraRibeiro , Walfrido Moraes Tomas ,

Helena Godoy Bergallo 1 and Carlos Frederico Duarte Rocha 1

1 Department of Ecology, Rio de Janeiro State University, Rua São Francisco Xavier, nº

524, PHLC, sala 220. Maracanã, CEP: 20550019. Rio de Janeiro, RJ, Brazil. Phone:

55(21)23340260.

2 Wildlife Laboratory, Embrapa Pantanal, Rua 21 de Setembro,n° 1.880, Bairro Nossa

Senhora de Fátima,CEP:79320900. Corumbá, Mato Grosso do Sul, Brazil.

Draft

*Corresponding author: [email protected]

1

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 2 of 29

Density, habitat use, and activity patterns of a vulnerable population of Tinamus solitarius (Vieillot, 1819) in a Brazilian Atlantic Forest fragment

ACF, JPR, WMT, HGB, and CFDR

Abstract

We present the 1st estimates of density of the (Tinamus solitarius

(Vieillot, 1819)) in the Reserva Natural Vale (RNV), one of the two largest remnants of the Atlantic Rainforest in the State of Espírito Santo, Brazil. We also modeled the spatial distribution and habitat use of the using occupancy and detectability modeling. We used a linetransect survey to estimate density and abundance, and cameratraps to estimate habitat use and activity pattern. We conducted the survey within a 13month period (April 2013 to May 2014). Estimated density was 9 ± 2 individuals / km 2 and estimated populationDraft size for the entire RNV was 2115 ± 470 individuals. Occupancy was best described by the: distance to forest edge; distance between trees; poaching intensity; and understorey cover. Detectability was affected by the: distance between trees and poaching intensity. We conclude that in the RNV, the nearthreatened T. solitarius is a diurnal species preferring intact closedcanopy forest but tending to avoid areas near to forest edges or areas under comparatively high intensity of poaching. These results reinforce the importance of RNV as a source of resource to T. solitarius reproduction at Brazilian Atlantic forest remnants.

Key Words: Detectability; Distance sampling; Linetransect; Occupancy modelling;

Reserva Natural Vale; Solitary Tinamou; Tinamus solitarius

2

https://mc06.manuscriptcentral.com/cjz-pubs Page 3 of 29 Canadian Journal of Zoology

Introduction

The Atlantic forest spreading along Brazil, northeastern Argentina, and eastern

Paraguay, is one of the world’s top five biodiversity hotspots (Myers et al. 2000) and

one of South America’s highest priorities for conservation (Stotz et al. 1996;

Stattersfield et al. 1998). In Brazil, only about 11% of the original area covered by the

Atlantic Forest still remains (Ribeiro et al. 2009). species that occurs in this biome

are subject to different threats such as high grade logging (Aleixo 1999), impact of

roads (Kuitunen et al. 1998; Develey and Stouffer 2001; Clevenger et al. 2003),

poaching (Chiarello 2000; Hill et al. 2003; Nunes et al. 2012), habitat loss and habitat

fragmentation (Anjos 2006; LouresRibeiro et al. 2011), all of which have led to local

extirpations of formerly common speciesDraft such as the Drabbreasted Bambootyrant

(Hemitriccus diops (Temminck, 1822)) and the Pheasant Cuckoo (Dromococcyx

phasianellus (Spix, 1824)) (Willis 1979; Christiansen and Pitter 1997; Ribon et al.

2003).

The Tinamidae family, the only component of the Tinamiformes , has

approximately 50 species, which occur from northwestern Mexico to southern South

America (Cabot 1992; Sick 1997). The Solitary Tinamou, Tinamus solitarius (Vieillot,

1819), is the largest tinamid outside the area of the Amazon basin and can reach up to

52 cm in length, with males with mass between 1.2 and 1.5 kg and females between 1.3

and 1.8 kg (Bokermann 1991; Sick 1997). Originally the T. solitarius had a wide

distribution, from the state of Pernambuco to the north of Rio Grande do Sul, including

Minas Gerais, west of São Paulo, Misiones (Argentina) and regions adjacent to

Paraguay (BirdLife International 2016). However, there are few information on the

ecology of the T. solitarius . Excepting the study of Kuhnen et al. (2013), information

3

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 4 of 29

available on the species are restricted to anecdotal data found in bird guides and in handbooks, with only few studies concerning aspects of its ecology ( e. g. Bokermann

1991; Sick 1997; Brooks et al. 2004; Schelsky 2004; Sigrist 2009).

Tinamus solitarius is considered as a Near Threatened species according to the

IUCN Red List of Threatened Species, but the T. solitarius populations could be a threshold for classification as Vulnerable because a moderately rapid and ongoing population decline is suspected (BirdLife International 2016). Tinamus solitarius currently survives in few forest fragments mostly due to poaching and habitat loss impacts (Dinerstein et al. 1995), and has been the subject of reintroduction projects in the wild, such as those in the state of Minas Gerais (Bokermann 1991; Sick 1997;

BirdLife International 2016). Tinamus solitarius is mostly found in lowland humid forest up to 1,200 m, preferring intactDraft closedcanopy forest with little undergrowth

(Cabot et al. 1992; Parker et al. 1996). It is supposedly susceptible to forest fragmentation, but there is Paraguayan population surviving in a small area of degraded forest, and the species has also been seen regularly in secondary forest in one site in

Argentina (Cabot et al . 1992; Chebez 1996; Lowen et al.1996). According to BirdLife

International (2016) the population status of the species in its geographical distribution remains a gap, and Stotz et al. (1996) described this species as 'uncommon'.

To provides additional tools for the conservation of T. solitarius , our study presents the 1st estimates of density of the species in the Reserva Natural Vale, one of the two largest remnants of the Atlantic Rainforest in the State o Espírito Santo, located in Brazil. By using occupancy and detectability modeling, we could model the spatial distribution and habitat use of the species, from which we predicted the direction of response of six covariates based on prior knowledge of T. solitarius ecology (Bokerman

1991; Cabot et al. 1992; Lowen et al. 1996; Sick 1997). We tested the hypothesis that

4

https://mc06.manuscriptcentral.com/cjz-pubs Page 5 of 29 Canadian Journal of Zoology

occupancy probability would be higher in sites with primary forest cover. We also

hypothesized that detectability would be higher in sites with lower influence of

poaching and greater distance from forest edges.

Material and Methods

Study area. This study was conducted in the Reserva Natural Vale (RNV), a

protected area of 23,500 ha belonging to the Vale Company. The Reserve is in the

neighboring municipalities of Linhares and Jaguaré (19°06’–19°18’ S and 39°45’–

40°19’ W), in northeastern Espírito Santo, Brazil (Fig. 1).

The RNV was established through a gradual process of land acquisition, which

started in 1955, when Vale bought its 1 st properties in the region, and reached its current

limits in 1973. The Reserve is composedDraft of one main block of forest (approximately

98% of the total area), and a much smaller fragment, known as Biribas Reserve, in the

southwest of the main block.

The RNV is inserted within the Atlantic Rainforest biome, and composed of a

mosaic of habitats with four main vegetation types (adapted from Jesus 1987; Kierulff

et al. 2014): coastal plain, riparian, and sandy soil forests, and natural grassland. The

evergreen coastal plain forest has two or more upper strata and high densities of lianas

and epiphytes, and covers approximately 68% of the total area of the reserve. The

riparian forest, which covers some 4% of the Reserve, is a mixed type of vegetation

associated with streams, characterized by widelyspaced trees and a predominance of

palms. The sandy soil forest (covering approximately 8% of the RNV) is a type of

woody vegetation on sandy soils, physiognomically similar to the coastal plain forest at

an early or intermediate stage of regeneration. The natural grasslands occur as enclaves

within the forest, which were once the sites of ponds in the geological past, and cover

5

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 6 of 29

ca. 6% of the area of the reserve. In addition to these vegetation types, approximately

8% of the reserve is covered by wetlands (swamps) and streams (Fig. 1). The remaining

6% is composed of administrative structures of the Reserve.

Linetransect surveys. To estimate density and abundance of T. solitarius , we established 4 5km long line transects in the RNV (Buckland et al. 2001), using the

RAPELD protocol (Magnusson et al. 2005). RAPELD is a technique that seeks to standardize the collection of biological data and aims to develop methods for longterm ecological research (PELD) and allow rapid inventories (RAP Rapid Assessment

Program). Transects were separated from one to another by a minimum of 4 km of distance.

During a 13month period (April 2013 to May 2014), transects were surveyed using distance sampling techniquesDraft (Buckland et al. 2001). Each month, transects were surveyed twice by a single observer starting at sunrise (between 0530 and 0630 h). The observer waited 3 h before starting the afternoon survey, between 1300 and 1400 h. The observer walked transects at a speed of approximately 1 km/h. Transects were surveyed twice a month with the order being alternated each month.

For each sighting, we recorded the perpendicular distance of the from transect using a measuring tape, the length of transect walked to that point, date, timing of the record, and transect number. We did not consider the detections by hearing due to the lack of precision to measure the perpendicular distance. We did not used any approach to avoid double counting of T. solilarius individuals. Thus, some individuals may have been counted twice, therefore biasing some of the estimates.

Camera trapping. We selected 39 sampling sites using a systematic random design stratified by vegetation type to ensure that all 4 of the principal vegetation types found in the RNV were represented. This scheme was designed to model occupancy

6

https://mc06.manuscriptcentral.com/cjz-pubs Page 7 of 29 Canadian Journal of Zoology

probability of the RNV by the T. solitarius , as well as to register its respective circadian

activity. We placed a random grid over a digital map of the Reserve and identified the

sampling points by selecting grid cells.

First, in each selected grid cell, we located randomly one camera trap per grid

cell. We did not use bait to attract individuals of T. solitarius . This approach resulted in

a relatively even distribution of points within the RNV, while maintaining independence

among points, which were separated from one another by a distance of more than one

km (Magnusson et al. 2005). At each site, we installed one passive infrared Bushnell®

camera trap (Bushnell Outdoor Products, Overland Park, Kansas) in picture function,

approximately 40–50 cm above the ground, for continuous surveying throughout the

study (from May 2013 to June 2014). All stations were examined every 2025 days to

change batteries, when necessary. DraftTraps were programmed to operate for 24 h/d.

Occupancy Model Covariates — We used the six covariates to model the occupancy

probability and detectability of T. solitarius : distance to forest edge (in meters; edge);

distance between trees (in meters; tree); poaching intensity (in n/hectare; poach see

below for details of estimation); canopy cover (in percent; canopy); understorey cover

(in percent; under); distance to water resources (in meter; water).

At each sampling point, we established four plots (30 x 50 m) arranged by the

cardinal compass points (north, south, east, and west). In each plot, we measured the

diameter at breast height (DBH) of each tree and counted the number of large trees

(DBH > 50 cm). For all the plots, canopy cover was estimated with a convex spherical

densiometer, and using digital camera images. In addition, ocular estimates were used in

the comparison. Understory cover was measured along the central (longitudinal) line of

each plot, considering 5 m on each side of the line (transect width of 10 m). Understory

cover was measured every 10 m using a 2.0 x 0.5 m sighting frame (each 0.5 m² portion

7

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 8 of 29

representing 25% visibility). Understory cover within each transect was estimated by the mean visibility of the sighting frame at the 5 points sampled within the transect.

Three spatial covariates – distances from forest edge, water resource, and road were quantified for each of the 39 sample sites using the ArcGIS software

(ESRI*ArcMap 10.1, Redlands, California).

Intensity of poaching within the study area was calculated using the georeferenced database of a 10year period in which poaching events were recorded by the reserve’s security guards (source: Reserva Natural Vale). The density of poaching

(records per km²) was calculated for each grid cell in which a camera trap had been installed by dividing the number of records by the grid cell area.

Data analysis

Density and population size. DensityDraft and population size were estimated by total number of individuals observed along each trail through the program DISTANCE 6.2

(Buckland et al. 2001). Distance program 6.2 uses the perpendicular distances (animal track) to estimate effective strip width (ESW) in the study area and model the detection function that best suits the probability of detection of an animal at a given distance track

(Laake et al. 1994; Buckland et al. 2001). The best detection model is selected by the

Akaike Information Criterion (AIC), which originates from the minimization of information (or distance) of KullbackLeibler (KL) as a basis for the selection of models

(Akaike 1973). The KL information is a measure of distance between the true model and a candidate model. The real model is an abstraction often being desirable to obtain a good model to satisfactorily represent reality. Burnham and Anderson (2002) recommend using the AIC to select models only when the number of observations is greater than or equal to 60. This minimum number of observations allows for accurate estimates.

8

https://mc06.manuscriptcentral.com/cjz-pubs Page 9 of 29 Canadian Journal of Zoology

Occupancy and detectability models . We classified the total camera trapping data

(200 days) using 40 occasions consisting of intervals of 5 consecutive days each based

on the Mackenzie et al. (2006) approach to constructed a reliable detection history. We

estimated site occupancy (Ψ) and detection probability ( p) for the species, with three

possible outcomes: (1) the site was occupied and the species was detected (Ψ x p); (2)

the species was present but not detected (Ψ x [1– p]), and (3) the species was not

present and therefore was not detected (1– Ψ).

We estimated detection probabilities based on detections obtained from each site

in 40 occasions. The probability was the parameter projected by a maximum likelihood

estimation of the proportion of sites occupied (Ψ) during the sample period.

In our occupancy analysis, we assessed the covariates that might affect

occupancy and detectability, to identifyDraft a pattern of habitat use for the T. solitarius . We

constructed a set of candidate models for the species, which were selected by a priori

hypotheses based on three different approaches: (1) considering occupancy probability

and detectability as constant across all sites, (2) considering the variation in occupancy

as a function of covariates, and (3) considering both the variation in occupancy and

detectability as a function of covariates. This allowed us to evaluate if the differences in

habitat use were determined by a single covariate or a set of covariates, which would

contribute to an improvement in the model’s performance. All models were screened for

collinearity using a correlation coefficient of 0.5.

Using the cameratrap data, singlespecies, singleseason occupancy models

were created using the UNMARKED package in Program R (Fiske and Chandler 2011).

Top models were selected using Akaike's information Criterion adjusted for small

sample size (AICc). All models with a AICc value < 2 were equivalent. To choose the

best model for test our hypotheses, we also used the weight (AICwt) for each model,

9

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 10 of 29

which corresponds to the amount of evidence in favor of a given model. We performed

2000 bootstraps to assess the adjustment fit ( z) and the overdispersion parameter ( ĉ).

Circadian activity patterns . We used the set of times that the species was photographed to estimate the circadian activity patterns of T. solitarius. We used a conditional circular kernel density function to estimate the circadian activity (Oliveira

Santos et al. 2013). This circular kernel had the same features as the home range kernel estimator, including a smoothing parameter and a conditional density isopleth, defined as the threshold of probability that specifies the section of the function that accounts for a given proportion of the entire probability function (OliveiraSantos et al. 2013).

Analyses were conducted in the “Circular” package in R with associated analytic packages (Lund and Agostinelli 2007). Circular summaries (Lund and Agostinelli 2007) were used to determine the mean overallDraft timing of the activity as recorded by camera traps.

Results

Density and population size. 908 km of the four selected transects were surveyed 196 times in this study (Table 1). Throughout the surveys, Tinamus solitarius was sighted

206 times.

Estimated density for T. solitarius was 9 ± 2 individuals / km 2 and estimated population size for the entire RNV was 2115 ± 470 individuals. The effective strip width (ESW) was 4.45 ± 0.41 m with records obtained from 0 to 18 m from the line of the transect. The model which best fitted our data was a Half Normal Key with cosine adjustment. The coefficient of variation for both parameters was 10.17%.

10

https://mc06.manuscriptcentral.com/cjz-pubs Page 11 of 29 Canadian Journal of Zoology

Occupancy and detectability models. A total of 7,020 trapdays were conducted

during the study. We obtained 580 independent detections of T. solitarius and observed

the species in 19 of the 39 sites, which resulted in a naïve occupancy of 0.487 and a

detectability of 0.28. From the 32 occupancy models produced (Table 2 showing the 20

top models), the best model to predict occupancy and detectability was

“Ψ(edge;tree;poach;under;canopy); p(tree;poach)”.

Occupancy was best described by four covariates: 1) distance to forest edge

(edge), which had a positive relationship in which occupancy increased with higher

distances (Ψ = 0.4 to 0.79; Fig. 2A); 2) Distance between trees with DBH > 50 cm

(tree), which had a positive relationship, in which occupancy by the T. solitarius

increased as the distance between trees increased (Ψ = 0 to 0.95; Fig. 2B); 3) Poaching

intensity (poach), which predicted Drafta negative relationship in which occupancy decrease

in site with higher poaching intensity (dropping to Ψ = 0.99 to 0; Fig. 2C); 4)

Understorey cover (under), which predicted a negative relationship with occupancy

(dropping to Ψ = 0.62 to 0.2; Fig. 2D); 5) Canopy cover (canopy), predicting a positive

relationship in which occupancy increases in sites with higher canopy coverage (Ψ = 0

to 0.95; Fig. 2E).

Detectability was affected by two covariates: 1) distance between trees with

DBH > 50 cm (tree), which had a positive relationship in which detectability increased

in sites with higher distances (p = 0 to 0.42; Fig. 3A); 2) Poaching intensity (poach),

which predicted a negative relationship with higher detectability in sites with lower

poaching records ( p = 0.34 to 0; Fig. 3B).

Circadian activity patterns. Tinamus solitarius showed two peaks of activity in the

RNV, with records of cameratrap throughout the 24 hours cycle (Fig. 4). The species

11

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 12 of 29

was active at dawn and early in the morning between 0400 to 0930 h, and at the late afternoon to dusk between 1630 to 1800 h.

Discussion

The T. solitarius density, estimated for the Reserva Natural Vale (RNV) through line transects, is the first estimate for the Atlantic Forest and showed a coefficient of variation below 20%, which is the maximum value recommended to obtain reliable estimates (Buckland et al. 2001). This means that the density estimated serves as a starting point for monitoring T. solitarius in the RNV and other areas that the species still remains. The density estimated in our study area (9 individuals / km 2) is higher when compared to the study of SãoBernadoDraft (2004) in a protected area in São Paulo, Brazil that estimated a density of 1.2 individuals / km 2. However, the density here estimated appears to be lower when compared with the densities of other members of the Tinamus of similar size to that of the T. solitarius (Schelsky 2004; Negret et al. 2015). For example, the Black Tinamou ( T. osgoodi hershkovitzi Conover, 1949) is relatively abundant (13.5 ± 2.5 individuals / km 2) in a protected area in the eastern

Andes of in the upper Amazon Basin (Negret et al. 2015), and the Great

Tinamou (T. major (Gmelin, 1789)) is comparatively more abundant (up to 17 individuals / km 2) in the mature floodplain and latesuccessional secondary forest in

Manu National Park located in Peru (Schelsky 2004). All three species present a susceptibility to hunting (Wege and Long 1995; BirdLife International 2016; 2017). Of these three species, T. major has the wider distribution within the Neotropics, with the distribution of T. solitarius being 3 times less, and the T. osgoodi hershkovitzi with the most restricted distribution among these species (Cabot 1992; Sick 1997). Tinamus

12

https://mc06.manuscriptcentral.com/cjz-pubs Page 13 of 29 Canadian Journal of Zoology

solitarius is a Near Threatened species but have less individuals (9 ind/km 2) and T.

major with a density 2 times higher (17 ind/km 2) that T. solitarius is also classified as

Near Threatened, while T. osgoodi hershkovitzi have 13 individuals per km 2 and it is

Vulnerable. This comparison draws our attention because could contribute to

reevaluation of T. solitarius conservation status in Brazil. At Brazilian Atlantic forest,

there are reported cases of local extinction of T. solitarius (Ribon et al. 2003; Mendonça

et al. 2009), however, the remaining forests of the RNV still hold an important

population of this species along its restrict distribution.

Individuals of T. solitarius in this study was only recorded in cameratraps

located in primary forested areas, presenting a strong relationship with covariates that

were directly related with forested habitats and avoiding areas near to forest edges. Our

results corroborate that the speciesDraft prefers intact closedcanopy forest with little

understorey coverage and more sparsed trees (Cabot et al. 1992; Parker et al. 1996).

Conversely, Kuhnen and collaborators (2013) studying habitat use by the species in

Serra do Tabuleiro State Park in the state of Santa Catarina, Brazil, found that the

species uses areas of secondary forest, but maintained its dependency of areas with

relatively higher canopy coverage. Kuhnen and collaborators (2013) argued that the

higher occurrence of T. solitarius in secondary than in primary forest may be explained

by the presence of a thick carpet of bromeliads on the primary forest floor, which could

possibly hamper the search for food, which it is not the case of the RNV. The primary

forests of the RNV presents a dense canopy cover and open understorey cover.

Understorey birds, such as species, are highly sensitive to forest

fragmentation, and seem to avoid areas under edge effect (Anjos 2006; LouresRibeiro

et al. 2011). Furthermore, increases in forest edge areas may change the movements of

the understorey birds, especially of those living in their interior (Hansbauer et al. 2008),

13

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 14 of 29

which tend to avoid forest edges and clearings. Our data suggest that changes in local conditions may have been preponderant concerning the distribution of the T. solitarius .

Many bird species are intolerant to high levels of light and tend to avoid edge areas

(Laurance et al. 2004). Among Amazonian understorey birds, factors, such as the evolutionary history of species, structural change of environment and trafficrelated disturbances, are determinants of edge avoidance (Laurance et al. 2004).

In addition to sensitivity to forest fragmentation, T. solitarius avoided areas with high intensity of poaching in the RNV. In fact, poaching is known as one of the major impacts on this species (BirdLife International 2016). Tinamus solitarius is poached for subsistence consumption and bushmeat market (Chiarello 2000; Hill et al. 2003;

Kierulff et al. 2014), and for illegal trafficking of species (Nunes et al. 2012). In the

Atlantic Forest of Paraguay, T. solitariusDraft was the most poached bird species in a period of 7 years (Hill et al. 2003). Poaching impact on vertebrate populations is imminent, and this tend to lead to a defaunation scenario that may result in unbalanced ecosystems

(Dirzo et al. 2014; Young et al. 2016), and this impact may be even higher in populations of endemic and fragmented species (Hall et al. 2008), such as T. solitarius .

The individuals of T. solitarius showed a diurnal activity in the RNV. The daily activity of T. solitarius in the RNV for the present study was similar of that of previous estimation carried out in the same locality by Dias et al. (2016), and similar to that obtained in the study conducted at the Serra do Tabuleiro State Park (Kuhnen et al.

2013).

We conclude that in the RNV, the nearthreatened T. solitarius is a diurnal species preferring intact closedcanopy forest which tending to avoid areas near to forest edges or areas under comparatively high intensity of poaching. Our data reinforces the importance of RNV as an important reservoir of the species. Therefore, the results

14

https://mc06.manuscriptcentral.com/cjz-pubs Page 15 of 29 Canadian Journal of Zoology

presented herein can be a starting point to support future action plans for the species,

making predictions regarding the ecosystem and management and conservation of T.

solitarius . Furthermore, the results can be used as a surrogate for other regions in which

the species occurs, because many locations may be affected by the same covariates used

herein (i.e. poaching and habitat fragmentation).

Acknowledgment

This study is portion of the results of the “Programa de Pesquisas em Biodiversidade da

Mata Atlântica (PPBio Mata Atlântica Program)" of Ministério da Ciência, Tecnologia,

Inovação e Comunicação (MCTIC) and was supported by Conselho Nacional de Desenvolvimento Científico e TecnológicoDraft (CNPq) (Process No. 457458/2127). The authors benefitted from grants provided to HGB (process 307781/20143) and to CFDR

(304791/20105; 470265/20108; 302974/20156) from Conselho Nacional do

Desenvolvimento Científico e Tecnológico (CNPq) and through “Cientistas do Nosso

Estado” Program from FAPERJ to CFDR (process No. E26/102.765.2012 E

26/202.920.2015) and to HGB (process E26.201.267.2014). ACF and JPR received

PhD fellowships from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior.

The ICMBio provided the permit for the development of the study (No. 463274) and

the Reserva Natural Vale for the Research in the Reserve.

Literature cited

Akaike, H. 1973. Maximum likelihood identification of Gaussian autoregressive

moving average models. Biometrika 60 : 255265.

15

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 16 of 29

Aleixo, A. 1999. Effects of selective logging on a bird community in the Brazilian

Atlantic Forest. The Condor 101 :537548.

Anjos, L. D. 2006. Bird Species Sensitivity in a Fragmented Landscape of the Atlantic

Forest in Southern Brazil. Biotropica 38 (2): 229234.

BirdLife International. 2016. Tinamus solitarius . The IUCN Red List of Threatened

Species 2016: e.T22678139A92757280. http://dx.doi.org/10.2305/IUCN.UK.2016

3.RLTS.T22678139A92757280.en. Downloaded on 19 February 2017 .

Bokermann, W. C. A. (Editor ) 1991: Observações sobre a biologia do macuco, Tinamus solitarius (Aves, Tinamidae). Brooks, D. M., PandoVasquez, L.,Draft OcminPetit, A., and TejadaRenjifo, J. 2004. Resource separation in a NapoAmazonian tinamou community. Ornitol. Neotrop. 15

(Suppl.): 323–328.

Buckland, S. T., Anderson, D. R., Burnham, K. P., Laake, J. L., Borchers, D. L., and

Thomas, L. (Editors ). 2001. Introduction to distance sampling: estimating abundance of biological population. Oxford University Press, Oxford.

Burnham, K. P., and Anderson, D. R. (Editors ). 2002. Model selection and multimodel inference: a practical informationtheoretic approach, Springer, New York.

Cabot, J. 1992. Family Tinamidae. In Handbook of the birds of the world. Ostrich to ducks. Edited by J. del Hoyo, A. Elliot, and J. Sargatal. Lynx Editions, Barcelona. Pp.

112–138.

Chebez, J. C. 1996. Fauna misionera: catalógo sistemático y zoogeográfico de los vertebrados de la provincia de Misiones, Argentina, n. 5, p. 106.

16

https://mc06.manuscriptcentral.com/cjz-pubs Page 17 of 29 Canadian Journal of Zoology

Chiarello, A. G. 2000. Influência da caça ilegal sobre mamíferos e aves das matas de

tabuleiro do norte do estado do Espírito Santo. Bol. Mus. Biol. Prof. Mello Leitao. 11 :

229247.

Christiansen, M. B., and Pitter, E. 1997. Species loss in a forest bird community near

Lagoa Santa in southeastern Brazil. Biol. Conserv. 80 (1): 2332.

Clevenger, A. P., Chruszcz, B., and Gunson, K. R. 2003. Spatial patterns and factors

influencing small vertebrate fauna roadkill aggregations. Biol. Conserv. 109 (1): 1526.

Develey, P. F., and Stouffer, P. C. 2001. Effects of roads on movements by understory

birds in mixed ‐species flocks in central Amazonian Brazil. Conserv. Biol. 15 (5):1416

1422. Draft Dias, L. C. S., Bernardo, C. S. S., and SrbekAraujo, A. C. 2016. Daily and Seasonal

Activity Patterns of the Solitary Tinamou ( Tinamus solitarius ) in the Atlantic Forest of

Southeastern Brazil. Wilson. J. Ornithol. 128 (4): 885894.

Dinerstein, E., Olson, D. M., Graham, D. J., Webster, A. L., Pimm, S. A., Bookbinder,

M. A., and Ledec, G. 1995. A conservation assessment of the terrestrial Ecoregions of

Latin America and the Caribbean. The World Bank, Washington, D.C.

Dirzo, R., Young, H. S., Galetti, M., Ceballos, G., Isaac, N. J., and Collen, B. 2014.

Defaunation in the Anthropocene. Science 345(6195): 401406.

Fiske, I., and Chandler, R. 2011. Unmarked: An R package for fitting hierarchical

models of wildlife occurrence and abundance. J. Stat. Softw. 43 (10): 1–23.

Hall, R. J., Milner ‐Gulland, E. J., and Courchamp, F. 2008. Endangering the

endangered: the effects of perceived rarity on species exploitation. Conserv. Lett. 1(2):

7581. 17

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 18 of 29

Hansbauer, M. M., Storch, I., Leu, S., NietoHolguin, J. P., Pimentel, R. G., Knauer, F., and Metzger, J. P. W. 2008. Movements of neotropical understory passerines affected by anthropogenic forest edges in the Brazilian Atlantic rainforest. Biol. Conserv.

141 (3): 782791.

Hill, K., McMillan, G., and Fariña, R. 2003. Hunting ‐Related Changes in Game

Encounter Rates from 1994 to 2001 in the Mbaracayu Reserve, Paraguay. Conserv.

Biol. 17 (5): 13121323.

Jesus, R. M. 1987. Mata Atlântica de Linhares: Aspectos Florestais. In

Desenvolvimento Econômico e Impacto Ambiental em Áreas de Trópico Úmido

Brasileiro A experiência da CVRD. Edited by Secretaria Estadual de Meio Ambiente, Idaho Water Resource Board, and DraftCompanhia Vale do Rio Doce. Rio de Janeiro. Pp. 3571.

Kierulff, M. C., Avelar, L. H. S., Ferreira, M. E. S. Povoa, K. F., and Bernils, R. S.

2014. Reserva Natural Vale: História e Aspectos Físicos. Ciência & Ambiente 49 : 7

40.

Kuitunen, M., Rossi, E., and Stenroos, A. 1998. Do highways influence density of land birds?. Environ. Manage. 22 (2): 297302.

Kuhnen, V. V., de Lima, R. E. M., Santos, J. F., and MachadoFilho, L. C. P. 2013.

Habitat use and circadian pattern of Solitary Tinamou Tinamus solitarius in a southern

Brazilian Atlantic rainforest. Bird. Conserv. Int. 23 (01): 7882.

Laake, J., Buckland, S., Anderson, D., and Burnham, K. (Editors ). 1994. Distance user’s guide. Colorado State University Press, Colorado.

18

https://mc06.manuscriptcentral.com/cjz-pubs Page 19 of 29 Canadian Journal of Zoology

Laurance, S. G. W., Stouffer, P. C., and Laurance, W.F. 2004. Effects of road clearings

on movement patterns on the understory rainforest birds in Central Amazonia. Conserv.

Biol. 18 : 1099–1109.

LouresRibeiro, A., Manhães, M. A., and Dias, M. M. 2011. Sensitivity of understorey

bird species in two different successional stages of the lowland Atlantic Forest,

Brazil. An. Acad. Bras. Cienc. 83 (3): 973980.

Lowen, J. C., Baertrina, L., Clay, R. P., and Tobias, J. A. (Editors ). 1996. Biological

surveys and conservation priorities in eastern Paraguay. CSB Conservation

Publications, Cambridge, UK.

Lund, U., and Agostinelli, C. 2007. Circular: circular statistics. Available from

http://cran.rproject.org/package5circularDraft (accessed January 2017).

Mackenzie, D., Nichols, J., Royle, J., Pollock, K., Bailey, L., and Hines, J. (Editors ).

2006. Occupancy estimation and modeling: Inferring patterns and dynamics of species

occurrence. Elsevier Publishing.

Magnusson, W. E., Lima, A. P., Luizão, R., Luizão, F., Costa, F. R. C., Castilho, C. V.,

and Kinupp, V. F. 2005. RAPELD: Uma modificação do método de Gentry para o

levantamento da biodiversidade em áreas. Biota Neotropica 5.

Mendonça, L. B., Lopes, E. V., and Anjos, L. 2009. On the possible extinction of bird

species in the Upper Paraná River floodplain, Brazil. Braz. J. Biol. 69 (2): 747755.

Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A., and Kent J. 2000.

Biodiversity hotspots for conservation priorities. Nature 403 (6772): 853858.

19

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 20 of 29

Negret, P. J., Garzón, O., Stevenson, P. R., and Laverde, O. 2015. New ecological information for the Black Tinamou ( Tinamus osgoodi hershkovitz i). The Auk 132 (3):

533539.

Nunes, P. B., Barreto, A. S., and Franco, E. Z. 2012. Subsídios à ação fiscalizatória no combate ao tráfico de aves silvestres e exóticas em Santa Catarina. Ornithologia 5(1):

2633.

OliveiraSantos, L. G. R., Zucco, C. A., and Agostinelli, C. 2013. Using conditional circular kernel density functions to test hypotheses on animal circadian activity. Anim.

Behav. 85 (1): 269280.

Parker III, T. A., Stotz, D. F., and Fitzpatrick, J. W. 1996. Ecological and distributional databases. In Neotropical birds: ecologyDraft and conservation. Edited by D. F. Stotz, J. W.

Fitzpatrick, T. A. Parker III, and D. K. Moskovits. The University of Chicago Press,

Chicago. pp. 115131.

Ribeiro, M. C., Metzger, J. P., Martensen, A. C., Ponzoni, F. J., and Hirota, M. M.

2009. The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation. Biol. Conserv. 142 (6): 11411153.

Ribon, R., Simon, J. E., and de Mattos, G. T. 2003. Bird extinctions in Atlantic forest fragments of the Viçosa region, southeastern Brazil. Conserv. Biol. 17 (6): 18271839.

SãoBernardo, C. S. 2004. Abundancia, Densidade e Tamanho Populacional de Aves e

Mamíferos Cinegéticos no Parque Estadual Ilha Cardoso, SP, Brasil. Universidade de

São Paulo, Piracicaba, Estado de São Paulo, Brazil.

Schelsky, W. M. 2004. Research and conservation of forestdependent tinamou species in Amazonia Peru. Ornitol. Neotrop. 15 : 317321.

20

https://mc06.manuscriptcentral.com/cjz-pubs Page 21 of 29 Canadian Journal of Zoology

Sick, H. (Editor ). 1997. Ornitologia Brasileira.—Ed. Nova Fronteira SA, Rio de

Janeiro, Brazil.

Sigrist, M. S., and Carvalho, C. J. B. D. 2009. Historical relationships among areas of

endemism in the tropical South America using Brooks Parsimony Analysis

(BPA). Biota Neotropica 9(4): 7990.

Stattersfield, A. J., Crosby, M. J., Long, A. J., and. Wege, C. (Editors ). 1998. Global

directory of endemic bird areas. BirdLife International, Cambridge, United Kingdom.

Stotz, D. F. (Editor ). 1996. Neotropical birds: ecology and conservation. University of

Chicago Press. Willis, E. O. 1979. The compositionDraft of avian communities in remanescent woodlots in southern Brazil. Pap. Avulsos. Zool. 33: 125.

Young, H. S., McCauley, D. J., Galetti, M., and Dirzo, R. 2016. Patterns, causes, and

consequences of anthropocene defaunation. Annu. Rev. Ecol. Evol. Syst. 47 : 333358.

21

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 22 of 29

List of Figures

Figure 1. Habitat mosaics inside the Reserva Natural Vale, Espírito Santo, Brazil. Black triangles represent the location of each camera trap, and the red lines represent the survey transects.

Figure 2. Relationships between the occupancy rate and (A) the distance to forest edge;

(B) distance between tree; (C) poaching intensity; (D) understorey cover; and (E) canopy cover for the Tinamus solitarius . Data was estimated by cameratrapping at the

Reserva Natural Vale between May 2013 and June 2014, grouped in sampling intervals of 5 consecutive days. Draft Figure 3. Relationships between the detectability of Tinamus solitarius and (A) distance between tree and (B) poaching intensity. Data was estimated by cameratrapping at the

Reserva Natural Vale between May 2013 and June 2014, grouped in sampling intervals of 5 consecutive days.

Figure 4. Circadian activity pattern of Tinamus solitarius in the Atlantic rainforest in the

Reserva Natural Vale, Espirito Santo, Brazil, estimated by cameratrapping between

May 2013 and June 2014.

22

https://mc06.manuscriptcentral.com/cjz-pubs Page 23 of 29 Canadian Journal of Zoology

Draft

Figure 1. Habitat mosaics inside the Reserva Natural Vale, Espírito Santo, Brazil. Black triangles represent the location of each camera trap, and the red lines represent the survey transects.

592x419mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 24 of 29

Draft

Figure 2. Relationships between the occupancy rate and (A) the distance to forest edge; (B) distance between tree; (C) poaching intensity; (D) understorey cover; and (E) canopy cover for the Tinamus solitarius. Data was estimated by camera-trapping at the Reserva Natural Vale between May 2013 and June 2014, grouped in sampling intervals of 5 consecutive days.

423x618mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs Page 25 of 29 Canadian Journal of Zoology

Figure 3. Relationships between the detectability of Tinamus solitarius and (A) distance between tree and (B) poaching intensity. Data was estimated by camera-trapping at the Reserva Natural Vale between May 2013 and June 2014, grouped in sampling intervals of 5 consecutive days. Draft 423x209mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 26 of 29

Draft

Figure 4. Circadian activity pattern of Tinamus solitarius in the Atlantic rainforest in the Reserva Natural Vale, Espirito Santo, Brazil, estimated by camera-trapping between May 2013 and June 2014.

423x290mm (300 x 300 DPI)

https://mc06.manuscriptcentral.com/cjz-pubs Page 27 of 29 Canadian Journal of Zoology

TABLE 1. Survey effort for each transect and total of individuals of Tinamus solitarius (Vieillot, 1819) observed in the Reserva Natural Vale, Brazil. Transect ID Survey effort (km walked) Individuals observed T01 212 52 T02 233 48 T03 225 43 T04 238 63

Draft

https://mc06.manuscriptcentral.com/cjz-pubs Canadian Journal of Zoology Page 28 of 29

TABLE 2. Top 20 best single-season occupancy and detectability models for Tinamus solitarius (Vieillot, 1819) in the Reserva Natural Vale, Brazil, estimated by camera-trapping between May 2013 and June 2014, grouped in sampling intervals of 5 consecutive days. Covariates: distance to forest edge (in meters; edge); distance between trees (in meters; tree); poaching intensity (in n/hectare; poach); canopy cover (in percent; canopy); understorey cover (in percent; under); distance to water resources (in meter; water). Ψ = occupancy, p = detectability, AIC w = Akaike weight. Model AIC AIC AICw nºparameters Ψ(edge;tree;poach;under;canopy); p(tree;poach) 629.75 0 0.46 9

Ψ(edge;tree;poach;under); p(tree;poach) 631.86 2.11 0.16 8

Ψ(edge;tree;poach;canopy); p(tree;poach) Draft631.98 2.23 0.16 8

Ψ(edge;tree;poach;under;canopy); p(tree) 635.67 5.92 0.06 8

Ψ(edge;tree;poach;under;canopy); p(poach) 636.05 6.3 0.05 8

Ψ(edge;tree;poach;under;canopy); p(tree;poach;canopy) 647.45 17.7 0.04 10 Ψ(edge;tree;poach;under;canopy); p(tree;poach; under) 648.03 18.28 0.01 10

Ψ(edge;tree;poach;under;canopy); p(tree;canopy) 648.87 19.12 0.01 9

Ψ(edge;tree;poach;under;canopy); p(edge;tree;poach;canopy) 650.32 20.57 0.01 11

Ψ(edge;tree;poach); p(tree;poach) 650.98 21.23 0.01 7

Ψ(edge;tree;under); p(tree;poach) 652.34 22.59 0.01 7

Ψ(edge;tree;canopy); p(tree;poach) 653.09 23.34 0.001 7

Ψ(edge;tree); p(edge;tree;poach;canopy) 654.34 24.59 0.001 8

Ψ(edge;tree;poach); p(edge;tree;poach;canopy) 656.03 26.28 0.001 9

https://mc06.manuscriptcentral.com/cjz-pubs Page 29 of 29 Canadian Journal of Zoology

Ψ(edge;tree); p(poach) 659.45 29.7 0.001 5

Ψ(edge); p(tree;poach) 660.34 30.59 0.001 6

Ψ(edge; water; poach); p(tree;poach) 661.23 31.48 0.001 7

Ψ(edge;tree.water); p(tree;poach) 663.78 34.03 0.001 7

Ψ(edge;tree;canopy;water); p(tree;poach) 664.89 35.14 0.001 8

Ψ(.); p(.) 672.45 42.7 0.001 2

Model fit = 0.12, and c-hat= 1.31 obtained with 2,000 bootstraps Draft

https://mc06.manuscriptcentral.com/cjz-pubs