Speothos Venaticus Across the Amazon and Other Biomes

How rare is rare? Quantifying and assessing the rarity of the bush dog Speothos venaticus across the Amazon and other biomes

T ADEU G. DE O LIVEIRA,FERNANDA M ICHALSKI,ANDRÉ L. M. BOTELHO L INCOLN J. MICHALSKI,ARMANDO M. CALOURO and A RNAUD L. J. DESBIEZ

Abstract The bush dog Speothos venaticus is a medium- Keywords Amazon, bush dog, camera trap, conservation, sized Neotropical canid. It is considered to be rare and its group size, rarity, relative abundance, Speothos venaticus biology and population parameters are still poorly under- To view supplementary material for this article, please visit stood. The Amazon is one of the main strongholds of this https://doi.org/./S species and is important for maintaining viable populations, as the region still holds extensive tracts of pristine habitat. We gathered field data from camera-trap studies throughout the Brazilian Amazon to estimate the relative abundance Introduction of the species and gain an understanding of its rarity, and how this compares with estimates from other vegetative for- he bush dog Speothos venaticus is a medium-sized mations and for sympatric hypercarnivores. We focused on T(c.  kg) Neotropical canid that lives in packs. The spe- three pristine or partially disturbed sites and one fragmen- cies’ biology and population characteristics are poorly ted site. The estimated relative abundance of the species was understood, and it is one of the least known carnivores in  –   . . individuals per trap-days, confirming that South America (Eisenberg & Redford, ; Zuercher ’ the species is rare. The bush dog s abundance in the et al., ; DeMatteo & Loiselle, ). Most information Amazon is equivalent to that in all other areas outside the available on the species in Brazil comes from opportunistic   Basin. The mean group size recorded was c. . individuals. observations (e.g. Peres, ; Silveira et al., ; Rocha There were no differences in group sizes between forests in et al., ; Michalski et al., a), and it has only recently the Amazon and in other regions of Central America; how- become the subject of longer-term studies (Lima et al., , ever, there were significant differences between forests and , ). open habitats. A combination of competition/predation, The bush dog is recognized as being a rare carnivore habitat structure/integrity, and disease may be acting syner- (Zuercher et al., ; Oliveira, ; DeMatteo et al., gistically in determining the abundance and rarity of bush ), and is categorized as Near Threatened on the IUCN dogs. Red List (DeMatteo et al., ) and Vulnerable in Brazil (Jorge et al., ). The Amazon Basin represents the species’ main geographical range and is important for its long-term conservation (e.g. Zuercher et al., ; Oliveira, ). Currently, the Amazon has sufficient suitable habitat to TADEU G. DE OLIVEIRA* (Corresponding author) Departamento de Biologia, Universidade Estadual do Maranhão, Cidade Universitária Paulo VI – Caixa maintain a genetically viable population of bush dogs Postal 09, São Luís, Maranhão, 65055-000, Brazil. E-mail [email protected] (Oliveira, ).

FERNANDA MICHALSKI† Instituto Pró-Carnívoros, Atibaia, São Paulo, Brazil The Amazon Basin is the largest tract of lowland rainfor-

ANDRÉ L. M. BOTELHO‡ Mestrado em Ecologia e Manejo de Recursos Naturais, est, although many areas are under pressure from ongoing Campus Universitário Rodovia, Rio Branco, Acre, Brazil habitat loss and fragmentation as a result of human en-

LINCOLN J. MICHALSKI Faculdade de Itaituba, Itaituba, Pará, Brazil croachment and development activities, especially along  ARMANDO M. CALOURO Universidade Federal do Acre, Centro de Ciências the so-called deforestation arc (Nogueira et al., ; Silva Biológicas e da Natureza, Campus Universitário–Rodovia, Rio Branco, Acre, et al., ). To date, .% of the original forest cover has Brazil been lost in the Brazilian Amazon alone (INPE, ) and ARNAUD L. J. DESBIEZ Royal Zoological Society of Scotland, Murrayfield, further decline is predicted (e.g. a reduction to % of the Edinburgh, UK original cover by ; Soares-Filho et al., ). The *Also at: Instituto Pró-Carnívoros, Parque Edmundo Zanoni, Atibaia, São Paulo, Brazil Amazon provides a safe haven for species that range within †Also at: Laboratório de Ecologia e Conservação de Vertebrados, Universidade its boundaries (Rodrigues & Oliveira, ), by a combin- Federal do Amapá, Macapá, Amapá, Brazil ation of extensive tracts of primary forest and a network ‡Also at: Instituto Federal de Educação, Ciência e Tecnologia do Acre. Campus Avançado Baixada do Sol, Rio Branco, Acre, Brazil of protected areas. However, given the current rate of defor- Received  January . Revision requested  April . estation and the number of existing and proposed hydro- Accepted  June . First published online  October . electric dams (Tundisi et al., ; INPE, ) there is

Oryx, 2018, 52(1), 98–107 © 2016 Fauna & Flora International doi:10.1017/S0030605316000624 Downloaded from https://www.cambridge.org/core. IP address: 170.106.202.8, on 01 Oct 2021 at 00:35:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605316000624 How rare is rare? 99

likely to be a major shift in the ecological landscape towards tributaries (George et al., ). The Amapá National many isolated and fragmented habitats; for example, .  Forest, in the north-east, is a sustainable-use area compris- million ha of forest are expected to be flooded as a result ing continuous tropical rainforest vegetation, predominant- of the construction of hydroelectric dams (Fearnside, ly never-flooded terra firme forest, with some areas of ), representing % of the total area of the Brazilian flooded forest, bamboo and rocky outcrops (Michalski Amazon (FAO, ), which will contribute directly to et al., b). The Alta Floresta site is located along the de- widespread extinction of vertebrates (Benchimol & Peres, forestation arc of the Amazon and includes terra firme forest ). fragments of varying sizes, shapes and degrees of connectiv- The detection and study of cryptic or elusive species has ity (Michalski et al., ). become easier and more reliable with the advent of camera trapping. Although the technique has drawbacks for certain species, such as arboreal or very small species, it has been Methods utilized effectively to study several mammalian carnivores and has proved a reliable source of data for estimating rela- Defining rarity tive abundance and population density (Carbone et al., ; A rare species is one that is uncommon, scarce or encoun- O’Brien et al., ;O’Connell et al., ; Oliveira, ; tered infrequently (Gaston, ). For the purpose of this Kasper et al., ). Although the rarity of bush dogs is wide- study we define a rare species as one for which the number ly acknowledged, there is no specific quantitative measure of of camera-trap records is low, which is an inherent reflec- abundance, nor has the abundance of bush dogs been com- tion of low density and small population size (Rabinowitz, pared with that of other sympatric hypercarnivores. The ap- ). This definition has been used in other studies of parent rarity and distribution of this canid could be a mammals (Cofré & Marquet, ; Yu & Dobson, ; product of competitive interactions with the puma Puma Harcourt et al., ). We assessed camera-trap data concolor, jaguar Panthera onca, and possibly ocelot (photographic records) from our study sites (Fig. ) and Leopardus pardalis (Oliveira & Pereira, ), as they for several uncommon and elusive species in the share similar prey species (Oliveira, ; Zuercher et al., Neotropics (e.g. Martins et al., ; Tobler et al., ; ; Lima et al., ; Oliveira et al., ). To address Negrões et al., ; Pimenta, ; Botelho, ; these issues we estimated the relative abundance of bush Oliveira et al., ; Borges et al., ). Difficult-to-detect dogs in the Brazilian Amazon and compared it with those species such as the short-eared dog Atelocynus microtis, in other biomes and potential sympatric competitors/preda- giant armadillo Priodontes maximus and giant anteater tors to assess the long-term conservation and persistence Myrmecophaga tridactyla were all recorded at rates of no prospects of the species in this area. more than . records per  trap-days. Thus, we considered species with , . records per  trap-days to be rare. Study area

The study focused on four sites in the south-west, central, Bush dog abundance south-central, and north-east of the Brazilian Amazon (Fig. ). Three of these were mostly pristine (with slight dis- The location and arrangement of cameras was targeted turbance; i.e. , % alteration by human activity, such as fire towards mammalian carnivores. All cameras were spaced or logging) and one was fragmented. The study site at Alto .– km apart and typically positioned at – cm height Tarauacá Extractive Reserve, in the south-west, is in an area along forest trails, riparian forest/stream banks, and at other of open and closed rainforest with alluvial seasonally sites where carnivores were likely to be detected. The proto- flooded forest and some degree of disturbance related to col used for camera placement was based on the objective of the latex extraction industry. The sites surveyed in the gathering data on the abundance of carnivores and other Reserve included areas where subsistence hunting had low mammals, and cameras were placed at optimum distances and high impact (Botelho, ). The mid-Tapajós/ from the focal area to capture all mammals, and particularly Amazonia National Park area in central Amazonia is char- bush dogs (mean shoulder height – cm; Zuercher et al., acterized by a humid equatorial climate and hilly terrain ). This protocol was also followed to capture data on with a mosaic of rainforest vegetation types. We used smaller (,  kg) mammals, such as small rodents and mar- study areas both within and outside the park. The area is supials. Thus, we predicted the camera-trap sampling proto- dominated by tall terra firme forest (– m) interspersed col would provide a robust representation of how common by areas of lower forest and lianas (– m). There is also an or rare carnivores were in these areas. As bush dogs are gre- area of seasonally flooded riverine forest (igapó) occupying a garious, we considered the total number of individuals cap- narrow strip along the banks of the Tapajós River and its tured on video or in photographic records to represent

FIG. 1 Locations of the four study sites in the Amazon Basin where camera-trap surveys of the bush dog Speothos venaticus were conducted. Other camera-trap studies have been conducted at () Los Amigos Conservation Concession, Peru (Tobler et al., ), () Chandless State Park (Borges et al., ), () Amanã Sustainable Development Reserve (Rocha et al., ), () Piagaçu-Purus Sustainable Development Reserve (Pimenta, ), () Balbina Hydroelectric Dam (Benchimol & Peres, ), () Caxiuanã National Forest (Martins et al., ), () Carajás National Forest (Bergallo et al., ), and () Cantão State Park (Negrões et al., ).

relative abundance, reported as individuals per  trap- abundance and number of records with trap effort. The sig- days. Consecutive photographic records of the same species nificance level was considered to be α = .. were defined as independent occurrences if the individual(s) could be distinguished unambiguously or if the interval between records was .  minutes (O’Brien et al., ). Bush Results dog group size was determined from the number of individuals recorded on camera or observed by researchers at the Rarity study sites during fieldwork. We assessed bush dog abun-   dance in relation to the abundance of other sympatric hy- Trapping effort at our four study sites was , trap-days   ± percarnivores and the bush dog’s main prey species, in total, which yielded a mean relative abundance of .     –  agouti Dasyprocta spp., paca Cuniculus paca and armadillo SD . bush dogs per trap-days (range . . ;    ± Dasypus spp. (Zuercher et al., , ; Lima et al., ). Table ). In pristine areas mean abundance was . SD . individuals per  trap-days, more than double that found at the fragmented Alta Floresta site (. individuals per  trap-days). The species’ rarity was confirmed Analytical procedures by a lack of records in several other similar camera-trap studies across the Amazon, which accounted for an add- We present data as mean ± standard deviation per sampling itional , trap-days (Table ). Thus, considering the unit. We conducted normality tests to determine whether to other sites where the focal species was observed (Table ), use parametric or non-parametric statistics. To compare the the relative abundance of bush dogs in the Amazon was es- abundance of bush dogs with that of their potential intra- timated to be .–. individuals per  trap-days guild predators or competitors we used ANOVA on ranks, (mean . ± SD .; Negrões et al., ; Bergallo and Tukey’s test for all pairwise multiple comparisons. We et al., ; Rocha et al., ; this study). The mean abun- used ANOVA to compare bush dog group size among vari- dance in fragmented and non-fragmented areas across the ous vegetative formations, and the Holm–Sidak method for Amazon Basin was . and . ± SD . individuals all pairwise multiple comparisons. We applied t-tests to per  trap-days, respectively. The mean for all camera- compare bush dog abundance in the Amazon with all trap sites evaluated in the Amazon (including non-detection other areas in the species’ range as well as abundance in frag- sites) was . ± SD . individuals per  trap-days mented and non-fragmented sites. We used a Mann– (range –.; Martins et al., ; Tobler et al., ; Whitney rank sum test to compare group size from camera- Negrões et al., ; Pimenta, ; Benchimol & Peres, trap data and visual observations, and conducted regression ; Borges et al., ; Rocha et al., ; this study). analyses to determine the relationship between bush dog Bush dog abundance in the Amazon was equivalent to

that in all other areas outside the Basin (mean . ± SD . individuals per  trap-days; t = .,df=, P=.; Table ). Additionally, no difference in abun- Dec. 2008 – dance was found when comparing fragmented/disturbed and non-fragmented areas at any of the study sites (t = ., df = ,P=.). As abundance was equivalent across all vegetation formations, we conducted regression analyses of bush dog abundance against camera-trap effort across all vegetation types. The expected relative abundance in re- Dec. 2004; Oct. 2007

– lation to camera-trap effort was determined as follows: bush   −  effort  

and their potential intraguild compe- dog abundance = . ( . )(r = . ,        R = . , F,  = . ,P= . ). Considering the number of records, we expected the following: bush dog photograph-  ic records = . +(.effort)(r = ., R = .,     F,  = . ,P= . ). Speothos venaticus

Oct. 2014 June 2003 Group size and activity –

All photographic records and some of the visual observations of bush dogs were from terra firme forest. Bush dogs were mostly recorded along trails (Plate ) but were also sighted at road crossings, on stream banks and in igapó (flooded for-   trap-days) of bush dogs est). The majority of both visual (n = , %) and camera- trap (n = , .%) records occurred during the day. There  was no significant difference between group sizes estimated from camera-trap and visual records (T = .,P=.; Fig. ). We found no significant differences between sizes of bush dog groups in Amazonian forests and in Central Nov. 2013 Oct. 2013 ).       – America (T = . ,P=. ; Table ). However, there were significant differences between forested environments Fig. and open formations of the Pantanal floodplains and savan-     nahs (F,  = . ,P= . ).

Species interactions

The mean relative abundances of jaguars, pumas and ocelots in the study areas were . ± SD ., . ± SD . and . ± SD ., respectively (Table ; Supplementary Table S). All of these species, which are potential intraguild

Dec. 2012 June 2012 competitors or predators of the bush dog, were significantly – more abundant than the bush dog (H = .,df=, Alto Tarauacá Extractive Reserve Mid-Tapajós/Amazonia National Park Amapá National Forest Alta Floresta 0.3651.21610.418 1.2 1.1 3.8 1.7 1.0 2.0 0.208 1.131 3.050 0.081P= 0.675.). Pairwise comparisons of relative 1.6 abundances be- 0.253 tween hypercarnivores and bush dogs were also significantly different (P , .). In addition, relative abundances of the main prey species of the bush dog were relatively high at the four study sites (Table ). spp. 1.784 2.48 1.1 7.008 spp. 16.133 10.775 15.7 4.612 ) 43 106 25 2,580

2 Discussion

Dasypus Bush dog rarity Dasyprocta Leopardus pardalis Panthera onca Puma concolor 1 Details of our camera trap study to estimate the relative abundance (number of individuals per Cuniculus paca The relative abundances of bush dogs in the four study ABLE Sampling effort (trap-days)Camera stationsSurvey area (km 2,467 59 4,000 140 2,700 30 6,721 49 Study period May T titors/predators and prey species at four sites throughout the Brazilian Amazon ( Ocelot Prey species Agouti Paca Armadillo Forest typeBush dog abundance No. of individuals per 100No. trap-days of photographs per 100No. trap-days of 0.122 sightings 0.081Intraguild competitors/predators Jaguar Puma Pristine/lightly disturbed 2 Pristine/lightly disturbed 0.100areas 0.025 were low, Pristine/lightly disturbed Highly fragmented corroborating 6 0.185 previous 0.074 observations 0.060 0.030 and

TABLE 2 Relative abundance of bush dogs recorded in camera-trap studies in major vegetation formations of the Amazon and Atlantic rainforests, the Pantanal floodplains and savannahs in Brazil, and the Central American forests in Panama, with sampling effort, no. of individuals per  trap days, habitat integrity, and data source.

Sampling Individuals per Photographs per Site effort (days) 100 trap-days 100 trap-days Habitat integrity Source Amazon (rainforest) Alto Tarauacá Extractive Reserve 2,467 0.122 0.081 Pristine/lightly disturbed This study Mid-Tapajós/Amazonia National 4,000 0.100 0.025 Pristine/lightly disturbed This study Park Amapá National Forest 2,700 0.185 0.074 Pristine/lightly disturbed This study Alta Floresta 6,721 0.060 0.030 Highly fragmented This study Amanã Sustainable Development 4,894 0.163 0.061 Pristine/lightly disturbed Rocha et al. (2015) Reserve (Amazonas, Brazil) Carajás National Forest (Pará, 3,572 Data 0.084 Pristine/lightly disturbed Bergallo et al. (2012) Brazil) unavailable Cantão State Park (Tocantins, 7,929 0.013 0.013 Pristine/lightly disturbed Negrões et al. (2011) Brazil) Caxiuanã National Forest (Pará, 2,838 * * Pristine/lightly disturbed Martins et al. (2007) Brazil) Chandless State Park (Acre, 3,213 * * Pristine/lightly disturbed Borges et al. (2015) Brazil) Piagaçu-Purus Sustainable Use 3,743 * * Pristine/lightly disturbed Pimenta (2012) Reserve (Amazonas, Brazil) Balbina Hydroelectric Dam 12,420 * * Highly fragmented Benchimol & Peres (Amazonas, Brazil) (2015) Los Amigos Conservation 2,340 * * Pristine/lightly disturbed Tobler et al. (2008) Concession (Peru) Atlantic Forest Carlos Botelho State Park (São 4,818 0.042 0.021 Pristine/lightly disturbed Beisiegel (2009) Paulo, Brazil) Natural Reserves complex 4,112 0.097 0.073 Lightly disturbed but Fusco-Costa & (Paraná, Brazil) with fragments Ingberman (2013) Pantanal floodplains/forest mosaic Nhumirim Ranch (Mato Grosso 2,238 0.179 0.045 Lightly–moderately Bianchi (2009) do Sul, Brazil) disturbed Cerrado (savannah) Veredas do Peruaçu State Park 6,000 0.017 0.017 Secondary savannah Ferreira et al. (2015) (Minas Gerais, Brazil) with old Eucalyptus plantation Central American forests (Panama) Pirre 3,119 0.160 0.096 Pristine/lightly disturbed Meyer et al. (2015) Nusagandi 830 0.120 0.120 Pristine/lightly disturbed Meyer et al. (2015) Donoso 24,641 0.024 0.020 Pristine/lightly disturbed Meyer et al. (2015) Santa Fé 3,165 0.095 0.063 Pristine/lightly disturbed Meyer et al. (2015)

* Species present but not detected by camera traps

conclusions that the species is rare (e.g. Zuercher et al., ; issue for this species. Previous studies of the species based Oliveira, ; DeMatteo et al., ; Jorge et al., ). on radio tracking have showed a trend of increasing home Although the abundance estimated from camera-trap re- range with increasing degree of fragmentation of habitat cords at the mid-Tapajós/Amazonia National Park site (Lima et al., , ; E. Lima, pers. comm.). We found was lower compared to the other undisturbed areas, the a similar trend at our sites, with higher relative abundances number of visual records at this site was higher than at of bush dogs in pristine or only slightly disturbed areas the other sites. Thus, relative abundance was similar across compared with fragmented areas. However, we found no the study sites, except for the fragmented Alta Floresta area, differences when comparing camera-trap records from frag- where it was approximately half that at the other sites. Even mented/disturbed and non-fragmented areas at any of the with the limited number of records of the bush dog, it is study sites. Given the generally low number of records of clear that habitat fragmentation is a major conservation the species, caution is warranted regarding this issue.

; Meyer et al., ; Rocha et al., ; this study). Of all the carnivores that occur in the Amazon Basin only the so- called phantom species, the Amazon weasel Mustela africana and the northern tiger cat Leopardus tigrinus,may be rarer than the bush dog (Oliveira et al., ; Rodrigues, ). This general pattern of rarity observed for the bush dog is not restricted to the Amazon Basin but is characteristic of the species throughout its range. Abundance estimates from the Amazon are similar to those from the Atlantic and Central American forests and the Pantanal floodplains and savannahs. Thus, the species is rare in all its habitats, irrespective of the area, biome or method of detection (cam- PLATE 1 A bush dog Speothos venaticus pack at the disturbed  era trap, radio telemetry, detection dogs). This rarity is and fragmented study site of Alta Floresta, Brazil (Fig. ). ’ Photograph by Fernanda Michalski. probably related to the species inherently low population density, large home range and secretive behaviour (Zuercher et al., ; Beisiegel, ; DeMatteo et al., ; Lima et al., , ; Fusco-Costa & Ingberman, ; Ferreira et al., ; this study). Camera-trap records are considered to be a consistent and reliable means of estimating the relative abundance of cryptic mammals (Carbone et al., ; Goulart et al., ; Negrões et al., ), although a linear relationship between the rate of photographic records and true abundance is assumed, as is a constant detection probability across space and time (O’Connell et al., ). Thus, it is possible that low abundance estimates reflect low detectability (Lima et al., ; Ferreira et al., ), in which case alter- native methods, such as detection dogs, may be more cost- efficient for detecting rare species, including the bush dog (Long et al., ; DeMatteo et al., ). However, detec-

FIG. 2 The total number of bush dogs recorded by camera traps, tion dogs may not be appropriate for population assessment sightings, and both methods combined, at the four study sites of all species or in all areas, especially as there are no data on combined (Alto Tarauacá Extractive Reserve, Mid-Tapajós/ the use of dogs to detect Neotropical carnivores with which Amazonia National Park, Amapá National Forest and Alta to make comparisons. Nevertheless, camera-trapping stud- Floresta; Fig. ). ies of carnivores, even rare ones, have proven to be effective for detecting and estimating abundance when compared to The inherent rarity of bush dogs was also highlighted by other techniques, such as telemetry, especially when trap- the fact that despite a large sampling effort (,–, ping effort is high (e.g. . , trap-days; Soisalo & trap-days) the species was not recorded in %( of )of Cavalcanti, ; T.G. Oliveira, , unpubl. data; L.P. other systematic and robust camera-trap studies across the Meira et al., unpubl. data; this study). Additionally, bush Amazon (Martins et al., ; Tobler et al., ; Pimenta, dog behaviour in the study areas and elsewhere did not sug- ; Benchimol & Peres, ; Borges et al., ), most of gest avoidance of camera traps or camera locations. The which were conducted in areas with good habitat integrity. number of records from our study areas also supports this Nevertheless, it is important to highlight that absence of de- argument. We therefore concur with previous, similar stud- tection by camera-trapping does not prove that the species is ies and conclude that bush dogs are rare. absent in an area, only that the low numbers make it difficult In this study we considered any estimates of abundance to detect. It has been shown that the use of detection dogs is , . individuals per  trap-days to be indicators of rar- more accurate for locating this rare canid (DeMatteo et al., ity. We are confident that this is an accurate indicator when ). compared to camera-trap records for other Neotropical car- Relative abundance estimates were from areas where nivores and rare or uncommon Neotropical mammals that bush dogs were considered to be relatively common were accurately sampled using the same technique (e.g. (based on visual and/or other evidence). As such, this spe- Tobler et al., ; Negrões et al., ; Botelho, ; cies is rare even when recorded in greater numbers (Bianchi, Oliveira et al., ; L.P. Meira et al., unpubl. data).

TABLE 3 Group sizes of bush dogs recorded in various vegetation formations throughout the species’ range.

No. of Area/formation Mean ± SD Range observations Source Amazon forest This study Camera trap 2.286 ± 1.113 1–47 Sighting 2.75 ± 2.188 1–88 Combined 2.533 ± 1.727 1–815 Central American forests 2.421 ± 1.712 1–8 19 Meyer et al. (2015) Northern Pantanal floodplains/ 2.75 ± 1.07 1–7 24 Lima et al. (2009) forest/pastureland mosaic Savannahs & Pantanal 4.556 ± 2.833 1–10 9 Silveira et al. (1998), Lima et al. (2012, 2015), Teribele et al. floodplains (2012), Ferreira et al. (2015), M. Buonato (unpubl. data)

Group size and activity Species interactions

All photographic records were from terra firme forest; how- The abundance of bush dogs at the study sites was low, and  ever, we also recorded sightings and track records in other up to orders of magnitude lower compared to intraguild areas, including at highway crossings, on stream banks and competitors and predators. However, the highest abundance in igapó. Lima et al. (, ) reported that radio-tracked estimates for large felids in the study area are for Amapá groups of bush dogs did not utilize roads, whereas National Forest, where the highest abundance of bush Fusco-Costa & Ingberman () observed that they did. dogs was also recorded. Conversely, in the Tarauacá In any case, at our study sites cameras were not placed Extractive Reserve, where bush dog abundance was almost along roads, only on narrow trails. Similar to Rocha et al. as high as in Amapá, large felids were uncommon. (), the majority of our records suggest that this canid Although this appears to be contradictory, we suggest that is predominantly diurnal. However, in the savannahs of our data can neither confirm nor deny the existence of in- Mato Grosso state (Brazil) a radio-tracked pack showed traguild interactions, which have been demonstrated for  higher levels of activity both early in the morning and at smaller Neotropical felids (Oliveira et al., ; Oliveira,  night (Lima et al., ). ). Negative impacts of dominant competitors on subor- Typical groups of bush dogs are estimated to comprise dinate competitors include decreases in niche breadth, re- – individuals, and most observations suggest group striction of habitat to less favourable areas, changes in sizes of – (Zuercher et al., ). Previous camera-trap activity patterns, and localized extinction, in addition to studies of the bush dog recorded groups of – individuals being preyed upon. Thus, subordinates tend to have smaller (Beisiegel, ; Bianchi, ; Negrões et al., ; population sizes than dominant competitors (e.g. Miquelle    Fusco-Costa & Ingberman, ; Meyer et al., ; Rocha et al., ; St-Pierre et al., ; Saleni et al., ; Oliveira   et al., ), which is similar to our results. Mean group et al., ; Oliveira, ). Thus, bush dog rarity may be a sizes recorded in the Amazon and Panamanian forests result of competitive interactions with jaguars, pumas and were c. . individuals (Meyer et al., ). In a Pantanal perhaps ocelots, as they share a similar prey base    floodplain savannah/forest/pastureland mosaic the mean (Oliveira, ; Zuercher et al., ; Lima et al., ;   group size (.) was only slightly larger than in forested Oliveira et al., ; Oliveira & Pereira, ). Predation areas (Lima et al., ). However, limited data from by domestic dogs could also be a threat in some areas  other open habitats in seasonally flooded floodplains and (Ferreira et al., ; E. Lima, pers. comm.). ’ savannahs suggest larger and significantly different group Relative abundances of the bush dog s main prey items sizes than those in forested areas (Lima et al., , ; (i.e. agouti, paca and armadillo) were relatively high at all Teribele et al., ; Ferreira et al., ; Meyer et al., ; of our study sites. Although prey availability is a defining this study). Differences in group size between savannahs factor of carnivore abundance, we suggest it may not be and forests may reflect differences in the availability or ac- the limiting factor determining inherent bush dog rarity, cessibility of the bush dog’s main prey species between these as this would not explain their low abundance at our  vegetative formations. The predominantly small group size study sites or elsewhere (Ferreira et al., ; this study). may be explained in part by the more solitary behaviour of  Neotropical hypercarnivores (Oliveira & Pereira, ), Conservation issues with the formation of groups being related to the need to subdue larger prey in some habitats (Zuercher et al., The expansion of hydroelectric energy projects in the , ). Amazon Basin, where most of the viable populations of

bush dogs occur, will continue and possibly accelerate in the Park area. Fieldwork there depended on the assistance of future. Current estimates vary from  to  additional dam many colleagues; we are especially indebted to Fábio projects planned for Amazonia (Tundisi et al., ). This Mazim, Gilberto Nascimento, Jean Pierre Santos, and José will introduce large-scale changes to the hydrological Bonifácio Soares. We thank Instituto Chico Mendes de cycle, inland connectivity and ecosystem services, which Conservação da Biodiversidade and the Federal University will in turn affect the persistence and maintenance of several of Amapá for logistical support at the Amapá National species populations. This future scenario for the Amazon Forest study site, and the Programa de Pesquisa em Basin in Brazil is of particular concern (Soares-Filho et al., Biodiversidade Biodiversity Monitoring Programme for ). The population viability assessment conducted for providing the grid system used during field activities Amazonia National Park showed a % probability of ex- there. We thank Cremilson Marques, Érico Kauano and tinction of the bush dog in ,  years if populations be- Sueli dos Santos for their assistance during the field work. come isolated (Godoy et al., ). The overall scenario for The United Nations Development Programme (UNDP), this area is predicted to worsen as the population size and Coordenação de Aperfeiçoamento de Pessoal de Nível carrying capacity decrease as a result of the suppression of Superior (CAPES) and Fundação de Tecnologia do Estado flooded riverine forest following the installation and do Acre funded the work in Acre. Rosenil Oliveira and operation of a hydroelectric dam (Godoy et al., ). Cláudia Cunha provided field assistance in Acre. Simulations indicate that isolation, which has already Fundação de Amparo à Pesquisa e ao Desenvolvimemto begun in some places (Nogueira et al., ; Silva et al., Científico e Tecnológico do Maranhão (FAPEMA) provided ), reduces the likelihood that even forest reserves of additional support. Nuno Negrões and several anonymous  . , km would be adequate to maintain viable popu- reviewers provided important and insightful comments. lations of bush dogs and to provide the conditions necessary for the long-term conservation of this rare canid. Author contributions

TGO, FM, ALMB and LJM collected and analysed the data, Conclusions and co-wrote the article with AMC and ALJD. Extensive tracts of undisturbed forest with low levels of human impact dominated the study areas, except for Alta References Floresta, yet even in pristine areas the bush dog is naturally

rare and threatened by predation and disease. Sarcoptic BEISIEGEL, B.M. () First camera trap record of bush dogs in the mange in particular is one of the main threats to the species state of São Paulo, Brazil. Canid News, ., –. in all Brazilian biomes (e.g. Jorge et al., ; Godoy et al., BENCHIMOL,M.&PERES, C.A. () Widespread forest vertebrate ). Disease can cause the loss of species from some extinctions induced by a mega hydroelectric dam in lowland Amazonia. PLoS ONE, (), e. areas for prolonged periods, as observed for short-eared BERGALLO, H.G., CARVALHO, A.S., REIS, M.L. & COSTA, F.G. ()  dogs in Peru (Leite-Pitman & Williams, ). Diseases af- Mamíferos de médio e grande porte. In Fauna da Floresta Nacional fecting wild canids tend to be transmitted from domestic de Carajás: estudos sobre vertebrados terrestres (eds F.D. Martins, dogs, which are commonly brought into protected and un- A.F. Castillo, J. Campos, F.M. Hatano & S.G. Rolim), pp. –. protected forested areas by local and indigenous people Nitro Imagens, São Paulo, Brazil. BIANCHI, R.C. () Ecologia de mesocarnívoros em uma área no (Leite-Pitman et al., ; Leite-Pitman & Williams, ). Pantanal Central. PhD thesis. Universidade Federal do Mato Grosso However, it is unlikely that disease alone could pose a sim- do Sul, Campo Grande, Brazil. ultaneous threat across the entire range of the bush dog, or BORGES, L.H.M., CALOURO, A.M. & DE SOUSA, J.R.D. () Large and be the sole cause of its rarity. Rather, the inherent rarity of medium-sized mammals from Chandless State Park, Acre, Brazil.  – the bush dog (defined here as , . individuals per  Mastozoología Neotropical, , .  trap-days) appears to be attributable to a combination of BOTELHO, A.L.M. ( ) Caça de subsistência e os mamíferos da Resex do Alto Tarauacá, Acre. MSc thesis. Universidade Federal do Acre, competitive pressure and predation (intraguild or other- Rio Branco, Brazil. wise), habitat structure and integrity, and disease, acting CARBONE, C., CHRISTIE, S., CONFORTI, K., COULSON, T., FRANKLIN, synergistically (but not separately) to determine the abun- N., GINSBERG, J.R. et al. () The use of photographic rates to dance and degree of rarity in any particular area. estimate densities of tigers and other cryptic mammals. Animal Conservation, , –. COFRÉ,H.&MARQUET, P.A. () Conservation status, rarity, and geographic priorities for conservation of Chilean mammals: an Acknowledgements assessment. Biological Conservation, , –. DEMATTEO, K.E. & LOISELLE, B.A. () New data on the status and We thank CNEC WorleyParsons Engenharia S.A. and distribution of the bush dog (Speothos venaticus): evaluating its Eletrobras, who provided logistical and financial support quality of protection and directing research efforts. Biological for the fieldwork in the mid-Tapajós/Amazonia National Conservation, , –.

DEMATTEO, K., MICHALSKI,F.&LEITE-PITMAN, M.R.P. () study of bush dogs: home range, activity and habitat selection. Speothos venaticus. The IUCN Red List of Threatened Species : Wildlife Research, , –. e.TA. Http://dx.doi.org/./IUCN.UK.-. LIMA, E.S., JORGE, R.S.P. & DALPONTE, J.C. () Habitat use and RLTS.TA.en [accessed  October ]. diet of bush dogs, Speothos venaticus, in the Northern Pantanal, DEMATTEO, K.E., RINAS, M.A., SEDE, M.M., DAVENPORT, B., Mato Grosso, Brazil. Mammalia, , –. ARGÜELLES, C.F., LOVETT,K.&PARKER, P.G. () Detection LIMA, E.S., JORGE, M.L.S.P., JORGE, R.S.P. & MORATO, R.G. () The dogs: an effective technique for bush dog surveys. The Journal of bush dog Speothos venaticus: area requirement and habitat use in Wildlife Management, , –. cultivated lands. Oryx, , –. EISENBERG, J.F. & REDFORD, K.H. () Mammals of the Neotropics. LONG, R.A., DONOVAN, T.M., MACKAY, P., ZIELINSKI, W.J. & Volume . Ecuador, Peru, Bolivia, Brazil. University of Chicago BUZAS, J.S. () Effectiveness of scat detection dogs for Press, Chicago, USA. detecting forest carnivores. The Journal of Wildlife Management, FAO () Computerized Data Gathering and Networking as a , –. Control and Monitoring System for the Improvement of and MARTINS, S.S., SANDERSON, J.G. & SILVA-JÚNIOR, J.S. () Reporting on Forest Management in the Amazon: The Case of Brazil. Monitoring mammals in the Caxiuanã National Forest, Brazil–First Forest Management Working Papers, Working Paper . Forest results from the Tropical Ecology, Assessment and Monitoring Resources Development Service, Forest Resources Division, FAO, (TEAM) program. Biodiversity and Conservation, , –. Rome, Italy. MEYER, N., MORENO, R., VALDES, S., MÉNDEZ-CARVAJAL,P., FEARNSIDE, P.M. () Dams in the Amazon: Belo Monte and BROWN,E.&ORTEGA,J.() New records of bush dog in Brazil’s hydroelectric development of the Xingu River Basin. Panama. Canid Biology & Conservation, , –. Environmental Management, , –. MICHALSKI, F., PERES, C.A. & LAKE, I.R. () Deforestation FERREIRA, G.B., OLIVEIRA, M.J.R., DE PAULA, R.C., RODRIGUES, F.H. dynamics in a fragmented region of southern Amazonia: evaluation G. & CARMO, E.D.C. () Regionally extinct species rediscovered: and future scenarios. Environmental Conservation, , –. the bush dog Speothos venaticus in Minas Gerais, south-eastern MICHALSKI, L.J., de OLIVEIRA, T.G. & MICHALSKI,F.(a) New Brazil. Oryx, , –. record for bush dog in Amapá State, Eastern Brazilian Amazonia. FUSCO-COSTA,R.&INGBERMAN,B.() Records of the bush dog Canid Biology & Conservation, , –. Speothos venaticus in a continuous remnant of coastal Atlantic MICHALSKI, L.J., NORRIS, D., de OLIVEIRA, T.G. & MICHALSKI,F. Forest in southern Brazil. Oryx, , –. (b) Ecological relationships of meso-scale distribution in  GASTON, K.J. () Rarity. Chapman & Hall, London, UK. Neotropical vertebrate species. PLoS ONE, (), e. GEORGE, T.K., MARQUES, S.A., DE VIVO, M., BRANCH, L.C., GOMES, MIQUELLE, D.G., STEPHENS, P.A., SMIRNOV, E.N., GOODRICH, J.M., N. & RODRIGUES,R.() Levantamento de mamíferos do Parna– ZAUMYSLOVA, O.J. & MYSLENKOV, A.E. () Tigers and wolves in Tapajós. Brasil Florestal, , –. the Russian far east: competitive exclusion, functional redundancy, GODOY, S., DESBIEZ, A.L.J. & MILLER, P. (eds) () Bush Dog and conservation implications. In Large Carnivores and the Population Viability Workshop. IUCN/SSC Conservation Breeding Conservation of Biodiversity (eds J.C. Ray, K.H. Redford, Specialist Group (CBSG), Atibaia, Brazil. R.S. Steneck & J. Berger), pp. –. Island Press, Washington, GOULART, F.V.B., CÁCERES, N.C., GRAIPEL, M.E., TORTATO, M.A., DC, USA. GHIZONI,JR, I.R. & OLIVEIRA-SANTOS, L.G.R. () Habitat NEGRÕES, N., REVILLA, E., FONSECA, C., SOARES, A.M.V.M., JÁCOMO, selection by large mammals in a southern Brazilian Atlantic Forest. A.T.A. & SILVEIRA,L.() Private forest reserves can aid in Mammalian Biology, , –. preserving the community of medium and large-sized vertebrates in HARCOURT, A.H., COPPETO, S.A. & PARKS, S.A. () Rarity, the Amazon arc of deforestation. Biodiversity and Conservation, , specialization and extinction in primates. Journal of Biogeography, –. , –. NOGUEIRA, E.M., YANAI, A.M., FONSECA, F.O.R. & FEARNSIDE, P.M. INPE () Levantamento das áreas desflorestadas da Amazônia () Carbon stock loss from deforestation through  in Legal no período –: resultados. Http://www.obt.inpe.br/ Brazilian Amazonia. Global Change Biology, , –. prodes/prodes__n.htm [accessed  January ]. O’BRIEN, T.G., KINNAIRD, M.F. & WIBISONO, H.T. () Crouching JORGE, R.P.S., BEISIEGEL, B.M., LIMA, E.S., JORGE, M.L.S.P., LEITE- tigers, hidden prey: Sumatran tiger and prey populations in a PITMAN, M.R.P. & PAULA, R.C. () Avaliação do estado de tropical forest landscape. Animal Conservation, , –. conservação do cachorro-vinagre Speothos venaticus (Lund, ) O’CONNELL, A.F., NICHOLS, J.D. & KARANTH, K.U. (eds) () no Brasil. Biodiversidade Brasileira, , –. Camera Traps in Animal Ecology: Methods and Analyses. Springer, KASPER, C.B., MAZIM, F.D., SOARES, J.B.G. & de OLIVEIRA, T.G. New York, USA. () Density estimates and conservation of Leopardus pardalis OLIVEIRA, T.G. () Ecología comparativa de la alimentación del southernmost population of the Atlantic Forest. Iheringia, Série jaguar y del puma en el neotrópico. In El Jaguar en el Nuevo Milenio Zooologia, , –. (eds R.A. Medellín, C. Equihua, C.L.B. Chetkiewicz, P.G. Crawshaw LEITE-PITMAN, M.R.P., NIETO,F.V.&DAVENPORT,L.() Jr., A. Rabinowitz, K.H. Redford et al.), pp. –. Fondo de Amenaza de enfermedades epidémicas a la conservación de Cultura Económica/Universidad Nacional Autónoma de México/ carnívoros silvestres en la Amazonía peruana. In Alto Purús: Wildlife Conservation Society, Mexico City, Mexico. Biodiversidad, Conservación y Manejo (eds M.R.P. Leite Pitman, OLIVEIRA, T.G. () Distribution, habitat utilization and N.C.A. Pitman & P.C. Alvarez), pp. –. Center for Tropical conservation of the Vulnerable bush dog Speothos venaticus in Conservation and INRENA, Lima, Peru. northern Brazil. Oryx, , –. LEITE-PITMAN, M.R.P. & WILLIAMS, R.S.R. () Atelocynus OLIVEIRA, T.G. () Ecologia e conservação de pequenos felinos no microtis. The IUCN Red List of Threatened Species : Brasil e suas implicações para o manejo. PhD thesis. Universidade e.TA. Http://dx.doi.org/./IUCN.UK.-. Federal de Minas Gerais, Belo Horizonte, Brazil. RLTS.TA.en [accessed  October ]. OLIVEIRA, T.G. & PEREIRA, J.A. () Intraguild predation and LIMA, E.S., DEMATTEO, K.E., JORGE, R.S.P., JORGE, M.L.S.P., interspecific killing as structuring forces of carnivoran communities DALPONTE, J.C., LIMA, H.S. & KLORFINE, S.A. () First telemetry in South America. Journal of Mammalian Evolution, , –.

OLIVEIRA, T.G., TORTATO, M.A., ALMEIDA, L.B., CAMPOS, C.B. & SOARES-FILHO, B.S., NEPSTAD, D.C., CURRAN, L.M., CERQUEIRA, BEISIEGEL, B.M. () Avaliação do risco de extinção do G.C., GARCIA, R.A., RAMOS, C.A. et al. () Modelling gato-do-mato Leopardus tigrinus (Schreber, ) no Brasil. conservation in the Amazon basin. Nature, , –. Biodiversidade Brasileira, , –. SOISALO, M.K. & CAVALCANTI, S.M.C. () Estimating the density OLIVEIRA, T.G., TORTATO, M.A., SILVEIRA, L., KASPER, C.B., MAZIM, of a jaguar population in the Brazilian Pantanal using camera-traps F.D., LUCHERINI, M., et al. () Ocelot ecology and its effect on and capture–recapture sampling in combination with GPS the small-felid guild in the lowland Neotropics. In Biology and radio-telemetry. Biological Conservation, , –. Conservation of Wild Felids (eds D.W. Macdonald & A.J. Loveridge), ST-PIERRE, C., OUELLET, J.P. & CRÊTE,M.() Do competitive pp. –. Oxford University Press, Oxford, UK. intraguild interactions affect space and habitat use by small OLIVEIRA, T.G., VIEIRA, O.Q., KASPER, C.B., TREVELIN, L.C. et al. () carnivores in a forested landscape? Ecography, , –. Mastofauna. In Relatório de impacto ambiental: AHE São Luiz do TERIBELE, R., CONCONE, H.V.B., GODOI, M.N., BIANCHI, R.C., Tapajós; Estudo de impacto ambiental, aproveitamento hidrelétrico SANTOS, J.C.C., MAURO, R.A. et al. () New records for bush dog São Luiz do Tapajós. Volume . pp. –. CNEC (Consórcio in Mato Grosso do Sul, Brazil. Canid News, , –. Nacional dos Engenheiros Consultores), São Paulo, Brazil. TOBLER, M.W., CARRILLO-PERCASTEGUI, S.E., PITMAN, R.L., MARES, PERES, C.A. () Observations on hunting by small-eared dog R. & POWELL,G.() An evaluation of camera traps for (Atelocynus microtis) and bush dog (Speothos venaticus)in inventorying large- and medium-sized terrestrial rainforest central-western Amazonia. Mammalia, , –. mammals. Animal Conservation, , –. PIMENTA, C.S. () Uso do habitat e ocupação por carnívoros em uma TUNDISI, J.G., GOLDEMBERG, J., MATSUMURA-TUNDISI,T.& Reserva de Uso Sustentável na Amazônia Central, Brasil. MSc thesis. SARAIVA, A.C.F. () How many more dams in the Amazon? Instituto Nacional de Pesquisa da Amazônia, Manaus, Brazil. Energy Policy, , –. RABINOWITZ,D.() Seven forms of rarity. In The Biological Aspects YU,J.&DOBSON, F.S. () Seven forms of rarity in mammals. of Rare Plant Conservation (ed. H. Synge), pp. –. Wiley, Journal of Biogeography, , –. New York, USA. ZUERCHER, G.L., GIPSON, P.S. & CARILLO,O.() Diet and habitat ROCHA, D.G., RAMALHO, E.E., ALVARENGA, G.C., GRÄBIN, D.M. & associations of bush dogs Speothos venaticus in the Interior Atlantic MAGNUSSON, W.E. () Records of the bush dog (Speothos Forest of eastern Paraguay. Oryx, , –. venaticus) in Central Amazonia, Brazil. Journal of Mammalogy, , ZUERCHER, G.L., SWARMER, M., SILVEIRA,L.&CARRILLO,O.() –. Bush dog Speothos venaticus (Lund, ). In Canids: Foxes, Wolves, RODRIGUES, L.A. () Avaliação do risco de extinção da Jackals and Dogs: Status Survey and Conservation Action Plan (eds doninha-Amazônica Mustela africana (Desmarest, ) no Brasil. C. Sillero-Zubiri, M. Hoffmann & D.W. Macdonald), pp. –. Biodiversidade Brasileira, , –. IUCN Gland, Switzerland, and Cambridge, UK. RODRIGUES, F.H.G. & OLIVEIRA, T.G. () Unidades de conservação e seu papel na conservação de carnívoros brasileiros. In Manejo e conservação de carnívoros neotropicais (eds R.G. Morato, Biographical sketches F.H.G. Rodrigues, E. Eizirik, P.R. Mangini, F.C.C. Azevedo & J. Marinho-Filho), pp. –. Edições IBAMA, Brasília, Brazil. T ADEU DE O LIVEIRA’s main research interests are the ecology and SALENI, P., GUSSET, M., GRAF, J.A., SZYKMAN, M., ALTERS,M.& conservation of Neotropical carnivores. F ERNANDA M ICHALSKI’sre- SOMERS, M.J. () Refuges in time: temporal avoidance of search interests are in conservation biology and the ecological conse- interference competition in endangered wild dogs Lycaon pictus. quences of habitat fragmentation on large and mid-sized vertebrates. Canid News, ., –. A NDRÉ B OTELHO is a wildlife ecologist and is interested in the ecology SILVA, M.E.S., PEREIRA, G. & da ROCHA, R.P. () Local and remote and conservation of medium-to-large sized mammals. L INCOLN climatic impacts due to land use degradation in the Amazon “Arc of M ICHALSKI is a wildlife ecologist whose main research interests are Deforestation”. Theoretical and Applied Climatology, , –. in the ecology and conservation of mammals. A RMANDO C ALOURO SILVEIRA, L., JACOMO, A.T.A., RODRIGUES, F.H.G. & DINIZ-FILHO, is interested in the effects of anthropogenic perturbations on mammals J.A.F. () Bush dogs (Speothos venaticus), in Emas National Park, in the south-western Brazilian Amazon. A RNAUD D ESBIEZ’s research Central Brazil. Mammalia, , –. focuses on sustainable use of natural resources and species conservation.