Paviolo et al 2009 Protection affect puma abundance and activity pattern .pdf
Nombre del archivo original: Paviolo et al 2009 Protection affect puma abundance and activity pattern.pdf
Este documento en formato PDF 1.2 fue generado por 3B2 Total Publishing System 7.51n/W / Apogee Create Series3 v1.0, y fue enviado en caja-pdf.es el 28/01/2015 a las 22:13, desde la dirección IP 200.112.x.x.
La página de descarga de documentos ha sido vista 1117 veces.
Tamaño del archivo: 358 KB (9 páginas).
Privacidad: archivo público
Descargar el documento PDF
Vista previa del documento
Journal of Mammalogy, 90(4):926–934, 2009
PROTECTION AFFECTS THE ABUNDANCE AND ACTIVITY
PATTERNS OF PUMAS IN THE ATLANTIC FOREST
AGUSTI´N PAVIOLO,* YAMIL E. DI BLANCO, CARLOS D. DE ANGELO,
MARIO S. DI BITETTI
Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas de Argentina (CONICET), Yapeyu´, 23, CP 3370 Puerto
Iguazu´, Misiones, Argentina (AP, CDDA, MSDB)
Asociacio´n Civil Centro de Investigaciones del Bosque Atla´ntico (CeIBA), Yapeyu´, 23, CP 3370 Puerto Iguazu´, Misiones,
Argentina (AP, CDDA, MSDB, YEDB)
Knowing the factors that affect the abundance and activity patterns of pumas (Puma concolor) in South
American forests may help in their conservation. Using camera traps, we conducted 4 surveys in 3 areas with
different levels of protection against poaching and logging within the biggest continuous fragment of the Upper
Parana Atlantic Forest. We used capture–mark–recapture population models to estimate the density of pumas
for each area. The core area of Iguazu´ National Park, with low poaching pressure and no logging for .60 years,
had the highest density of pumas (between 1.55 and 2.89 individuals/100 km2). Yabotı´ Biosphere Reserve, an
area with the highest poaching and logging pressure, showed the lowest density (between 0.3 and 0.74
individuals/100 km2). Areas with intermediate levels of poaching and logging pressure had densities between
0.66 and 2.19 individuals/100 km2. Puma activity peaked during the 1st hours of morning in the most protected
area, but became more crepuscular and nocturnal in areas with less protection. The lower abundance of pumas
in the more degraded areas may be related to lower prey abundance. Differences in activity patterns of pumas
among areas with different poaching pressures may be a direct response to poaching or to changes in the
availability and activity patterns of primary prey. Conservation efforts should focus on decreasing poaching and
logging pressures within protected areas to benefit pumas and other endangered species in the Atlantic Forest.
Key words: activity pattern, Atlantic Forest, camera traps, density estimate, jaguar, logging, poaching, prey abundance,
protection, Puma concolor
The puma (Puma concolor) inhabits most of the American
continent (Young and Goldman 1946). Although an extensive
amount of information about the ecology of this species exists,
90% of the published studies were conducted in North
America (Laundre 2005). Most of the existing information
from South America focuses on the trophic ecology of pumas
(Sunquist and Sunquist 2002). Studies related to other
biological aspects affecting puma ecology are scarce and
were conducted in temperate semidesert habitats (Franklin et
al. 1999; Novaro and Walker 2005) or savannahs (Schaller and
Crawshaw 1980; Scognamillo et al. 2003).
Important environmental and socioecological differences
exist between North and South American countries. As a
result, the management and conservation problems that pumas
face are different in these 2 regions (Laundre 2005). Habitat
loss and degradation are major threats to natural habitats in
South America. The Upper Parana Atlantic Forest is a
dramatic example of this process, with only 7% of its original
surface remaining in isolated fragments (Di Bitetti et al. 2003).
The biggest fragment of this ecoregion is known as the Green
Corridor (about 10,000 km2) and is located in Misiones
Province of Argentina and neighboring areas of Brazil (Di
Bitetti et al. 2003).
Most of these forest remnants suffered timber extraction of
different intensities and reflect different states of degradation
(Campanello et al. 2007). Habitat degradation caused by forest
overexploitation in the Green Corridor has been identified as
one of the possible causes of population decline in other
predators such as ocelots (Leopardus pardalis—Di Bitetti et
al. 2008a) and jaguars (Panthera onca—Paviolo et al. 2008).
Puma populations also may be affected by this factor.
In addition to forest degradation by logging, these forests
also are affected by poaching. In the Atlantic Forest, poaching
is a common activity (Giraudo and Abramson 2000) and it
negatively affects the abundance and behavior of some prey
species of pumas (Chiarello 2000; Cullen et al. 2000; Di
* Correspondent: firstname.lastname@example.org
E 2009 American Society of Mammalogists
PAVIOLO ET AL.—DENSITY AND ACTIVITY OF PUMAS
Bitetti et al. 2008b; Paviolo 2002). Therefore, variation in
protection efforts against poaching and logging may affect the
abundance and behavior of the primary prey species of pumas,
and in turn puma abundance and behavior.
Kelly et al. (2008) found that the density of pumas is very
low at Yabotı´ Biosphere Reserve in the Green Corridor,
suggesting that it may be related to the high poaching pressure
and intense logging activity suffered in the area. However,
their study compared densities among areas located in
different regions (Argentina, Bolivia, and Belize) where
factors other than poaching may affect the abundance of
pumas. The Green Corridor presents a variety of forest areas in
different states of conservation, providing an ideal situation to
test the hypothesis that human activities, such as poaching and
logging, negatively affect the abundance of pumas.
In the Atlantic Forest, pumas are often in conflict with
humans because they prey on domestic cattle (Conforti and
Azevedo 2003; Mazzolli et al. 2002) or are potentially
dangerous to humans, as was sadly confirmed by a fatal
puma attack on a child at the visitor’s area of Iguazu´ National
Park in 1997. Information on patterns of puma abundance and
activity might help to mitigate conflicts with humans, and to
establish a baseline for the elaboration of conservation
strategies for this species (Cougar Management Guidelines
Working Group 2005).
The goal of this study was to compare the abundance and
activity patterns of pumas in areas under different management and degradation conditions within the Green Corridor
and assess the effect of these management practices on the
ecology and behavior of the species.
MATERIALS AND METHODS
Study area.—We carried out this study in 3 areas of the
Green Corridor. This region is characterized by a semideciduous subtropical forest with no discernible dry season
(Cabrera and Willink 1980). Average temperatures are around
22uC and 17uC during the warmest and the coldest months,
respectively. Average annual precipitation is around
2,000 mm with 2 peaks in the spring and autumn (Crespo
One of the study sites was in Yabotı´ Biosphere Reserve
(2,600 km2; 27uS, 54uW). The surveyed area included part of
Esmeralda Provincial Park (300 km2; logged until 1990) and
several private properties. At the time of the study, these
private properties were being intensely exploited by logging
companies with the exception of Miot’s property, where
logging was less intense (Di Bitetti et al. 2008a). Some of the
results of the survey conducted at Yabotı´ Biosphere Reserve
were presented by Kelly et al. (2008).
Another surveyed area was Urugua-ı´ (25u589S, 54u069W).
This area included Urugua-ı´ Wildlife Reserve (32 km2), part
of Urugua-ı´ Provincial Park (840 km2), and Campo de los
Palmitos (300 km2), a property belonging to a logging
company. The area was subject to selective timber extraction
The Iguazu´ area (25u409S, 54u309W) was surveyed twice,
1st in 2004 and again between 2006 and 2007. During the 1st
survey we covered the central area of Iguazu´ National Park
(670 km2) of Argentina. This park was subjected to selective
logging until 1934 (Dimitri 1974). During the 2nd survey we
expanded the study area, adding the western portion of Iguazu´
National Park, San Jorge Forest Reserve (174 km2), and the
western area of Iguac¸u National Park of Brazil (1,850 km2).
Iguac¸u´ National Park of Brazil was selectively logged until the
decade of 1930 and the San Jorge Reserve until the end of the
1980s. A map of the study areas and surveys can be found in
Paviolo et al. (2008).
Measurement of poaching intensity.—Hunting is an illegal
activity in Misiones; therefore, we used indirect evidence to
assess its intensity. We collected information on the evidence
of poaching activities during our fieldwork. We recorded
encounters with armed poachers or dogs, photographic records
of dogs or people, hunting campsites, artificial salt licks,
poaching platforms, gunshots heard, hunting trails, spent
shotgun cartridges, and camera-trap stations robbed or
destroyed. A detailed list of evidence of poaching intensity
in the study areas can be found in Paviolo et al. (2008) and Di
Bitetti et al. (2008b).
Poaching pressure was variable among the areas and
depended mostly on the effort dedicated to controlling it and
on the accessibility to different areas by poachers (Paviolo et
al. 2008). Yabotı´ Biosphere Reserve suffered very high
poaching pressure, although the pressure in Esmeralda
Provincial Park and Miot’s property was lower than in the
rest of the surveyed area (Di Bitetti et al. 2008a; Paviolo et al.
2008). The Urugua-ı´ area suffered a medium to high poaching
pressure (Paviolo et al. 2008). Iguazu´ National Park suffered
the lowest poaching pressure in the central area where we
conducted the 1st survey (2004) but an intermediate poaching
pressure in the areas added in the 2006–2007 survey (Paviolo
et al. 2008).
Camera-trapping surveys.—We used records obtained by
camera traps in combination with closed capture–mark–
recapture population models to estimate animal densities
(Karanth 1995; Karanth and Nichols 2002). Individuals were
identified in the photographs by distinct pelage markings
(Karanth 1995; Silver et al. 2004; Trolle and Kery 2003).
Recently, Kelly et al. (2008) demonstrated that it is possible to
identify individual pumas using photographs, which allows the
estimation of the density of this species using this methodology if applied with caution and following certain protocols
to evaluate the degree of confidence in the results.
Between 2003 and 2007, we conducted 4 surveys to
estimate the absolute density of jaguars, pumas, and ocelots in
different areas of the Green Corridor. At each study site, we
placed between 34 and 47 sampling stations (Table 1). Each
sampling station consisted of a pair of camera traps facing
each other and operating independently. The stations were
located on infrequently used dirt roads or small trails opened
in the forest and were distributed at regular intervals with the
purpose of evenly covering the entire surveyed area. We used
Vol. 90, No. 4
JOURNAL OF MAMMALOGY
TABLE 1.—Dates and sampling effort of the different camera-trap surveys of pumas (Puma concolor) conducted in the Green Corridor of
Misiones Province, Argentina.
April 2006–January 2007
May 2003–February 2004
Full survey duration (days) Full survey effort (trap-days) Total survey effort (trap-days)a
Pilot + full surveys.
camera-traps of different brands and models. The equipment
consisted of 2 Camtrakker (Camtrakker, Watkinsville, Georgia), 50 Leaf Rivers Trail Scan Model C-1 (Vibra Shine,
Taylorsville, Mississippi), 30 TrailMACs 35mm Standard
Game (Trail Sense Engineering, LLC, Middletown, Delaware), and 20 Trapacamera (CIETEC, Sa˜o Paulo, Brazil)
scouting cameras. Prior to the full survey period, we
conducted pilot surveys with the purpose of identifying the
best sites for the locations of the stations (Table 1). The full
surveys consisted of a period of 90–96 days (Table 1).
Because of the longevity and length of territory tenure of
pumas, we assumed that a survey of this duration fulfilled the
assumptions of a closed population (Karanth and Nichols
2002; Kelly et al. 2008).
We identified pumas following the protocol proposed by
Kelly et al. (2008). Three of the authors independently
classified the photographs of individuals, noting the distinguishing characteristics of each animal. After independent
classifications, the 3 authors compared results and discussed
their reasons for each classification, correcting discrepancies
in cases when 1 of the authors could find evidence that the
classification was incorrect. When the evidence was not clear
the authors maintained their independent classifications. After
this, we estimated the density of pumas using the classification
of the 3 authors.
We estimated puma abundance using the program CAPTURE (Rexstad and Burnham 1991), which provides
population estimates using several models (Otis et al. 1978;
White et al. 1982). We present the results of the model Mh
using jackknife estimates that assume heterogeneity in the
capture probability among individuals. This model is the most
appropriate because of the varying accessibility to the stations
among individuals, product of the social structure of the
population, and the location of the stations within each
individual’s home range (Karanth and Nichols 2002). We
divided the survey into capture occasions of 6 consecutive
days with the purpose of obtaining a capture probability .0.1
(Otis et al. 1978; White et al. 1982). Cubs (,1 year old) were
not included in this analysis because their capture probability
is related to the capture probability of their mothers (Karanth
and Nichols 2002). Consequently, our density estimates refer
to the population of adults and subadults.
To estimate density it is necessary to calculate the area
surveyed. Most authors suggest that the area surveyed must be
estimated by adding a buffer width equal to one-half the
average of the maximum distance between captures of the
individuals captured more than once during the survey (mean
maximum distance moved [MMDM]) to each camera or the
polygon that includes all the cameras (Karanth 1995; Silver et
al. 2004; Trolle and Kery 2003). However, Maffei and Noss
(2007) suggest that if the surveyed area covers ,4 mean home
ranges of the studied species, MMDM may be underestimated
and in turn the area surveyed may be underestimated. In these
situations, the appropriate buffer should be between one-half
MMDM and MMDM (Maffei and Noss 2007). Because we
lacked estimates of the size of puma home ranges for our study
areas, we estimated density using 2 different calculations of
the surveyed area: 1 was obtained by applying to each
sampling station a buffer of one-half MMDM, and the other by
applying a full MMDM buffer. We deducted those areas that
are not suitable habitats for pumas, such as cities, annual
crops, and airports. The value of MMDM was estimated as the
average of the maximum distance of recapture for individuals
captured at .1 station (Karanth 1995; Karanth and Nichols
2002), according to each investigator’s classification. The
values of MMDM and the surveyed areas were estimated
using the program ArcView (version 3.2; Environmental
Systems Research Institute, Inc., Redlands, California).
Some researchers have suggested that the photographic rate
of a species is correlated with its absolute abundance (Carbone
et al. 2001), especially when controlling for some confounding
factors (Di Bitetti et al. 2008a). In order to validate the
patterns observed using the density estimates, we compared
different indices of relative abundance among surveys and the
study areas. We used the recording rate of pumas (number of
photographs of pumas/1,000 trap-days), the mean number of
individuals recorded per station, and the percentage of stations
with puma presence as relative abundance indices. Because
the indices varied widely between roads and trails (see
‘‘Results’’), and because the number of stations located on
trails at Yabotı´ (only 1) was insufficient to make a bifactorial
analysis including this variable, we compared the abundance
indices using only the values obtained from the stations
located on roads. In the Iguazu´ 2006–2007 and Yabotı´ surveys
we compared the relative abundance indices of pumas
between the best-protected and the least-protected subareas.
In addition, we compared the indices between the Iguazu´ 2004
survey and the same area of the Iguazu´ 2006–2007 survey to
determine whether differences between years existed. Because
the relative abundance data were not normally distributed, we
used nonparametric statistics for these comparisons.
Activity pattern analysis.—To describe the activity pattern
of pumas, we used the time printed on the photographs
obtained during the pilot and full surveys (Table 1). We
PAVIOLO ET AL.—DENSITY AND ACTIVITY OF PUMAS
considered as independent records only those that were .1 h
apart at the same station. We compared the activity pattern of
pumas between the stations located in the best- and leastprotected areas within the Iguazu´ 2006–2007 survey. We did
not perform this analysis for Yabotı´, because the number of
records in the least-protected area was very low (n 5 11).
Additionally, we performed the same analysis considering the
stations of all the surveys together (the well-protected central
area of Iguazu´ National Park versus the rest of the areas).
Finally, we compared the activity pattern in the central area of
Iguazu´ National Park between the 2004 and 2006–2007 surveys
to test whether there were differences between years. For these
analyses we used the Mardia–Watson–Wheeler test (Batschelet
1981). During all procedures we followed guidelines approved
by the American Society of Mammalogists for the use of wild
animals in research (Gannon et al. 2007).
Puma abundance.—At Yabotı´ we obtained 45 photographs
of pumas during the survey, of which 5 were discarded
because of their poor quality. The 3 investigators independently classified these photos as 6 or 7 different individuals
and the MMDM value varied between 12,486 m and
13,986 m. The area surveyed varied between 1,082 and
2,006 km2 according to the different methods and values of
MMDM applied. In turn, density estimates for this area were
between 0.3 and 0.74 individuals/100 km2, respectively.
During the full survey at Urugua-ı´, we obtained 16
photographs of pumas that corresponded to 3 or 4 individuals
according to the identification by the 3 investigators. The
MMDM was 6,854 m and was the same for all investigators.
The area surveyed was between 228 and 454 km2 and the
density of pumas was between 0.66 and 2.19 individuals/
During the Iguazu´ 2004 survey, we obtained 73 photographs
of pumas, of which 5 were discarded because of their poor
quality. The different investigators classified the photos as
either 10 or 11 individuals. The MMDM was 8,100 m and did
not vary among the investigators. The area surveyed was
between 450 and 774 km2 and puma densities were between
1.55 and 2.89 individuals/100 km2.
During the Iguazu´ 2006–2007 survey, we obtained 78
photographs of pumas, of which only 1 was eliminated
because of poor quality. The investigators identify between 11
and 16 different individuals. The estimates of MMDM varied
between 7,800 and 9,154 m. In turn, the area surveyed varied
between 750 and 1,295 km2 and the population density was
from 1 to 2.4 individuals/100 km2.
The recording rate and the mean number of individuals
recorded per station were higher on roads than on small trails
(Mann–Whitney 1-tailed U-test, recording rate: U 5 2,074, P
, 0.0001; mean number of individuals: U 5 2,127, P ,
0.0002). The recording rate and the mean number of
individuals recorded at stations located on roads were
statistically higher for Iguazu´ 2004 than for the Urugua-ı´
and Yabotı´ surveys. For the Iguazu´ 2006–2007 survey, these
indices also were significantly higher than for the Yabotı´
survey but were not statistically different from those from
Urugua-ı´ and Iguazu´ 2004 surveys. Finally, the indices were
not statistically higher for Urugua-ı´ than for Yabotı´ (Kruskal–
Wallis and all-pair comparisons test, recording rate: H 5 23.4,
P , 0.0001; mean number of individuals: H 5 23.81, P ,
In the Iguazu´ 2006–2007 survey, the recording rate was
higher in the best-protected area than in the least-protected one
(Mann–Whitney 1-tailed U-test, U 5 42, P 5 0.009; Fig. 1a),
as was the number of individuals per station (Mann–Whitney
1-tailed U-test, U 5 52, P 5 0.033; Fig. 1b) and the
probability of a station to record pumas (Fisher exact 1-tailed
test, x2 5 6.17, d.f. 5 1, P 5 0.017; Fig. 1c). On the other
hand, the abundance indices for the surveys of Iguazu´ in 2004
and for the same area of the Iguazu´ in 2006–2007 were not
different (Mann–Whitney 1-tailed U-test, recording rate: U 5
81, P 5 0.89; mean number of individuals: U 5 69.5, P 5
0.46), nor was the probability of a station to photograph pumas
(Fisher exact test, x2 5 2.1, d.f. 5 1, P 5 0.265).
The comparison between areas with different protection
levels in Yabotı´ showed that the recording rate and the number
of individuals recorded by station had a tendency to be higher
in the best-protected area, but not statistically so (Mann–
Whitney 1-tailed U-test, recording rate: U 5 167.5, P 5 0.06;
Fig. 1a; mean number of individuals: U 5 170, P 5 0.07;
Fig. 1b). Nevertheless, the probability of a station to
photograph a puma was statistically higher in the bestprotected compared to the least-protected area (Fisher exact 1tailed test, x2 5 5.31, d.f. 5 1, P 5 0.022; Fig. 1c).
Activity patterns.—In all the areas studied, pumas showed
some level of activity around the clock. Nevertheless, pumas
were more active during the 1st hours of the day in the wellprotected area, whereas in the least-protected areas they
showed 2 main activity peaks, 1 in the early morning and the
other in the 1st hours of the night, remaining active during the
night (Figs. 2a and 2b). These results were obtained when we
considered the sampling stations of all the surveys together
(Mardia–Watson–Wheeler test, x2 5 9.33, d.f. 5 2, P ,
0.011; Fig. 2a) and when we considered only the stations of
the Iguazu´ 2006–2007 survey (Mardia–Watson–Wheeler test,
x2 5 6.85, d.f. 5 2, P , 0.05; Fig. 2b). On the other hand, the
activity patterns in the well-protected area of Iguazu´ were not
different between the 2004 and 2006–2007 surveys (Mardia–
Watson–Wheeler test, x2 5 0.96, d.f. 5 2, P 5 0.607;
The abundance and behavior of pumas varied among areas
with different levels of protection within the Green Corridor.
Puma abundance was higher in the better-protected areas than
in areas with less protection, and this was observed using
indices of relative abundance and density estimates from
capture–recapture population models.
JOURNAL OF MAMMALOGY
Vol. 90, No. 4
FIG. 2.—Activity patterns of pumas (Puma concolor) in areas with
different levels of protection: a) including records from all study sites
(n 5 196 protected sites and n 5 121 less-protected sites); and b)
including records from the surveys Iguazu´ 2004 and Iguazu´ 2006–
2007 in protected and less protected areas (n 5 142, n 5 54, and n 5
24, respectively). The survey of Iguazu´ in 2004 included the same
area as the survey of the protected area in Iguazu´ in 2006–2007.
FIG. 1.—Indices of the relative abundance of pumas (Puma
concolor) in areas with different levels of protection in Iguazu´ 2006–
2007 and Yabotı´ surveys: a) recording rate (6SD), b) mean number
of individuals (6SD), and c) percentage of stations with
This correlation between the abundance of pumas and the
level of protection could result from several different factors.
One of them is prey abundance, because in general the
abundance of pumas depends mainly on the abundance of its
prey (Logan and Sweanor 2001; Pierce et al. 2000). Three of
the most important prey animals of pumas in this region are
red brocket deer (Mazama americana), agoutis (Dasyprocta
azarae), and collared peccaries (Pecari tajacu—Azevedo
2008; Crawshaw 1995). The relative abundance of these
species was lower in less-protected areas as a consequence of
poaching activity and habitat degradation due to the logging
activities (Di Bitetti et al. 2008b; Paviolo et al., in press),
which is consistent with the hypothesis that lower abundance
of pumas in those areas could be caused by the lack of prey.
Human-induced mortality is another factor that may
diminish puma populations (Hornocker 1970; Logan et al.
1986). In Florida, vehicle collisions were an important source
of mortality (Maehr 1997). Nevertheless, most records of
pumas killed on roads in the Green Corridor (at least 6 in the
last 10 years) came from within the Iguazu´ area, which
presents the highest densities of paved routes and pumas. On
the other hand, the less-protected areas are crossed by few dirt
roads and we do not have records of roadkills in those areas.
Therefore, roadkills could not explain the differences in
abundance among study areas.
Another cause of mortality is the sport and control hunting
of pumas by humans. This is the main cause of puma mortality
in areas where these kinds of hunting are allowed (Logan and
Sweanor 2001; Sunquist and Sunquist 2002). In the Green
Corridor pumas are occasionally killed because they prey on
domestic animals, but puma attacks are usually attributed to
jaguars (Conforti and Azevedo 2003). Unlike jaguars, pumas
PAVIOLO ET AL.—DENSITY AND ACTIVITY OF PUMAS
are not locally considered a trophy by poachers and are
considered to be less dangerous (Conforti and Azevedo 2003).
Therefore, pumas are not as systematically persecuted.
However, lack of information on the number of pumas
poached in our study areas prevents us from discarding this
factor as a possible influence on puma abundance.
Another factor that could be limiting the population of
pumas is the presence of competitor species. Interactions
between feline species have been suggested as a possible
cause for the decline of some cat species (Caro and Stoner
2003; Donadio and Buskirk 2006). In the Green Corridor,
pumas live in sympatry with jaguars and are approximately
one-third smaller. Some authors have suggested that jaguars
can exclude pumas by competition (Crawshaw and Quigley
2002; Schaller and Crawshaw 1980). However, at present the
abundance of jaguars is very low in the region (Paviolo et al.
2008), with jaguars being between 1.4 and 7 times less
abundant than pumas in our study sites. In the Iguazu´ area, where
the relative abundance of jaguar signs was higher than that of
pumas some years ago (Crawshaw 1995; Crespo 1982), the
situation has been reversed. This suggests that pumas are probably
tolerating better some pressures that have decimated the jaguar
population. On the other hand, jaguars, pumas, and ocelots present
the same pattern of abundance across study sites, with lower
densities in less-protected areas (Di Bitetti et al. 2006, 2008a;
Paviolo et al. 2008), suggesting that the 3 predators are more
affected by other factors than by competition among each other.
We believe that the differences in puma abundance among
areas with different levels of protection in the Green Corridor
are mainly caused by differences in prey availability.
However, the absence of areas where poaching and logging
were separate did not allow us to evaluate the relative direct
and indirect effects of these 2 factors.
As suggested by Kelly et al. (2008), the cause of the low
density of pumas found at Yabotı´ is likely related to high
poaching pressure and intense logging activities. Puma density
in this area is among the lowest reported in the literature
(Anderson 1983; Sunquist and Sunquist 2002). On the other
hand, densities in well-protected areas of the Green Corridor
are similar to those found in the tropical forest of Belize and
the places with high densities in North America (Hornocker
1970; Kelly et al. 2008; Logan and Sweanor 2001; Sunquist
and Sunquist 2002), but lower than densities in the Bolivian
Chaco (Kelly et al. 2008).
Activity patterns.—Pumas showed differences in their
activity pattern in areas with different levels of protection.
The same pattern was found when we analyzed data from all
the surveys together and when we compared 2 areas with
different levels of protection in the same year (Iguazu´ 2006–
2007). Also, in the area for which we have data from .1 year,
the activity pattern did not vary between surveys, which
suggests that the observed patterns are not caused by
interannual variation in ecological conditions.
Three hypotheses may explain these differences in the
activity patterns of pumas. The 1st hypothesis is that pumas
change their activity pattern to avoid periods when jaguars are
more active. Some authors suggest that jaguars and pumas
partition temporal and spatial activity (Emmons 1987) or that
pumas actively avoid encounters with jaguars (Schaller and
Crawshaw 1980). In Misiones, jaguars are predominantly
nocturnal and more abundant at Iguazu´ than any other area in
the Green Corridor (Paviolo et al. 2008). In Iguazu´, the
activity pattern of these 2 species is complementary,
suggesting that time partitioning exists. On the other hand,
jaguars live at very low densities at Urugua-ı´ and Yabotı´
(Paviolo et al. 2008), so we would expect pumas could be
more nocturnal in these areas because the probability of
encounter with a jaguar is lower. Nevertheless, in the leastprotected areas of the Iguazu´ 2006–2007 survey, jaguars were
relatively abundant (Paviolo et al. 2008) and pumas also
showed a more nocturnal pattern. The activity of pumas
overlapped with that of jaguars, contradicting the hypothesis
of temporal partitioning and suggesting that coexistence
between jaguars and pumas may be altered by anthropogenic
impacts, as proposed by Haines (2006).
Another hypothesis is that pumas are more nocturnal
because they avoid periods of higher human activity. This
has been observed in North America, where pumas were more
nocturnal in areas with logging activity even years after these
activities had ceased (Van Dyke et al. 1986). In other areas of
the Atlantic Forest with cattle, pumas attacked domestic
animals in hours of low human activity (Mazzolli et al. 2002).
In our study, pumas were more nocturnal even in areas where
logging activity had ceased more than 15 years previously.
Nevertheless, in those areas where poachers were active
during the day, pumas may have altered their activity pattern
to avoid encounters with poachers and their dogs.
Finally, a 3rd hypothesis is that pumas change their activity
patterns to improve their hunting success. Predators in general
follow the activity period of their main prey (Curio 1976), a
relationship reported for pumas in other areas (Beier et al.
1995; Maehr et al. 1990; Sunquist and Sunquist 2002). In our
study sites, we found that red brocket deer were more
nocturnal in less-protected areas (Di Bitetti et al. 2008b),
presenting an activity pattern similar to that shown by pumas.
Agoutis were active during the 1st hours of the day and in
the afternoon and were very abundant at Iguazu´ area but
scarce at Urugua-ı´ and Yabotı´. The change in activity pattern
of red brocket deer and the scarcity of agoutis in the lessprotected areas are probably contributing to the behavioral
change in the activity pattern of pumas. However, this
hypothesis does not exclude the previous ones, and the change
in activity pattern in less-protected areas may bring several
benefits for pumas.
Conservation of pumas in the Green Corridor.—The
differences in density of pumas in the Green Corridor means
that the 500 km2 in the center of Iguazu´ National Park is
supporting as many pumas as the entire Yabotı´ Biosphere
Reserve of 2,600 km2. In the Green Corridor, there is an
extensive network of areas with some level of protection
(nearly 6,000 km2), but the areas also receive a high impact
from poaching and logging and a great pressure from
Vol. 90, No. 4
JOURNAL OF MAMMALOGY
economic activities and urban areas. Under these conditions,
we consider that the best strategy for conserving pumas in this
region depends on strengthening the implementation of the
existing protected areas through more effective protection
against poaching activities and illegal logging, and consolidating corridors among those areas to allow the interarea
exchange of individuals.
In the Green Corridor, pumas are present in a total area of
20,000 km2 (De Angelo 2009). If we extrapolate our density
values for areas with different levels of protection, we estimate
a population of between 150 and 400 adult and subadult
individuals. According to a general population viability model
for pumas (Beier 1993), the population of pumas in the Green
Corridor would not be threatened by extinction in the short
term. Nevertheless, pumas, like other top predators, play a key
role in the environment by regulating the populations of their
prey and structuring the entire community (Logan and
Sweanor 2001). If we consider that in areas with deficient
protection other predators such as jaguars and ocelots also are
at very low densities (Di Bitetti et al. 2006, 2008a; Paviolo et
al. 2008), predation by top predators may be almost absent,
with unpredictable consequences for the future of the Green
Conocer los factores que pueden afectar la abundancia y los
patrones de actividad del puma (Puma concolor) en los
bosques de Sudame´rica es importante para la conservacio´n de
la especie. Utilizando ca´maras-trampa realizamos 4 muestreos
en 3 a´reas con distinto nivel de proteccio´n contra la caza
furtiva y explotacio´n forestal en el mayor remanente continuo
del Bosque Atla´ntico del Alto Parana´. Utilizamos modelos
poblacionales de captura–marcado–recaptura para estimar la
densidad de pumas en cada una de las a´reas. El a´rea central del
Parque Nacional Iguazu´, que tienen baja presio´n de caza
furtiva y no ha sido explotado forestalmente por .60 an˜os,
tuvo la mayor densidad de pumas (entre 1,55 y 2,89
individuos/100 km2). La Reserva de Bio´sfera Yabotı´ que
sufre una alta presio´n de caza furtiva y fuerte explotacio´n
forestal tuvo la menor densidad de pumas (entre 0,3 y 0,74
individuos/100 km2). Las a´reas con niveles intermedios de
caza furtiva y explotacio´n forestal tuvieron densidades entre
0,66 y 2,19 individuos/100 km2. Los pumas tuvieron el pico
de actividad durante las primeras horas de la man˜ana en las
a´reas mejor protegidas mientras que en las a´reas con menor
proteccio´n mostraron mayor actividad crepuscular y nocturna.
La menor abundancia de pumas en las a´reas ma´s degradadas
podrı´a estar relacionada con una menor abundancia de presas.
Las diferencias en el patro´n de actividad en a´reas con distintos
niveles de proteccio´n podrı´a ser una respuesta directa a la
presio´n de caza o a cambios en la abundancia y el patro´n de
actividad de sus presas principales. Los esfuerzos de
conservacio´n se deberı´an concentrar en disminuir los niveles
de caza furtiva y explotacio´n forestal lo que beneficiara´ al
puma y otras especies amenazadas del Bosque Atla´ntico.
We are very grateful to all the volunteers and park rangers that
helped us with fieldwork. We acknowledge the support and permits
provided by the Ministry of Ecology, Natural Resources and Tourism
of Misiones Province (MERNRT) and the National Parks Administration of Argentina. We thank A. Ricieri and A. Bertand for their
help and permission to develop the survey at Iguac¸u National Park.
We are grateful to Fundacio´n Vida Silvestre Argentina and the
property owners for their support and permission to conduct this
work. Financial support was provided by CONICET, Fundacio´n Vida
Silvestre Argentina, World Wildlife Fund–USA, World Wildlife
Fund–International, World Wildlife Fund–Switzerland, Lincoln Park
Zoo, Fundacio´n Antorchas, Wildlife Conservation Society, Idea
Wild, Rufford Foundation, and the Eden Project through a grant from
the Darwin Initiative. We also thank A. Noss, L. Montti, M. Kelly, A.
Green, A. Bertrand, D. Maehr, and an anonymous reviewer for help
and comments on the manuscript.
ANDERSON, A. E. 1983. A critical review of literature on puma (Felis
concolor). Colorado Division of Wildlife, Fort Collins, Colorado,
Special Report 54:1–91.
AZEVEDO, F. C. C. 2008. Food habits and livestock depredation of
sympatric jaguars and pumas in the Iguacu National Park area,
south Brazil. Biotropica 40:494–500.
BATSCHELET, E. 1981. Circular statistics in biology. Academic Press,
BEIER, P. 1993. Determining minimum habitat areas and habitat
corridors for cougars. Conservation Biology 7:94–108.
BEIER, P., D. CHOATE, AND R. H. BARRET. 1995. Movement patterns of
mountain lions during different behaviors. Journal of Mammalogy
CABRERA, A. L., AND A. WILLINK. 1980. Biogeografı´a de Ame´rica Latina.
Organizacio´n de los Estados Americanos. Serie Biologı´a 13:1–122.
CAMPANELLO, P. I., M. G. GATTI, A. ARES, L. MONTTI, AND G.
GOLDSTEIN. 2007. Tree regeneration and microclimate in a liana and
bamboo-dominated semideciduous Atlantic Forest. Forest Ecology
and Management 252:108–117.
CARBONE, C., ET AL. 2001. The use of photographic rates to estimate
densities of tigers and other cryptic mammals. Animal Conservation 4:75–79.
CARO, T. M., AND C. J. STONER. 2003. The potential for interspecific
competition among African carnivores. Biological Conservation
CHIARELLO, A. G. 2000. Density and population size of mammals in
remnants of Brazilian Atlantic Forest. Conservation Biology
CONFORTI, V. A., AND F. C. C. AZEVEDO. 2003. Local perceptions of
jaguars Panthera onca and pumas Puma concolor in the Iguacu
National Park area, south Brazil. Biological Conservation
COUGAR MANAGEMENT GUIDELINES WORKING GROUP, eds. 2005. Guı´a de
manejo del puma. Wildfuture, Bainbridge, Washington.
CRAWSHAW, P. G., JR. 1995. Comparative ecology of ocelot Felis
pardalis and jaguar Panthera onca in a protected subtropical forest
in Brazil and Argentina. Ph.D. dissertation, University of Florida,
CRAWSHAW, P. G., JR., AND H. B. QUIGLEY. 2002. Jaguar and puma
feeding habits. Pp. 223–235 in El jaguar en el nuevo milenio. Una
evaluacio´n de su estado, deteccio´n de prioridades y recomenda-
PAVIOLO ET AL.—DENSITY AND ACTIVITY OF PUMAS
ciones para la conservacio´n de los jaguares en Ame´rica (R. A.
Medellı´n, et al., eds.). Universidad Nacional Auto´noma de Me´xico
and Wildlife Conservation Society, Distrito Federal, Me´xico.
CRESPO, J. A. 1982. Ecologı´a de la comunidad de mamı´feros del Parque
Nacional Iguazu´, Misiones. Revista MACN, Ecologı´a 3:45–162.
CULLEN, L., R. E. BODMER, AND C. VALLADARES-PADUA. 2000. Effects
of hunting in habitat fragments of the Atlantic forests, Brazil.
Biological Conservation 95:49–56.
CURIO, E. 1976. The ethology of predation. Springer-Verlag, New
York, New York.
DE ANGELO, C. 2009. El paisaje del Bosque Atla´ntico del Alto Parana´ y
sus efectos sobre la distribucio´n y estructura poblacional del jaguar
(Panthera onca) y el puma (Puma concolor). Ph.D. dissertation,
Universidad de Buenos Aires, Buenos Aires, Argentina.
DI BITETTI, M. S., C. DE ANGELO, A. PAVIOLO, AND Y. DI BLANCO.
2008a. Local and continental correlates of the abundance of a
neotropical cat, the ocelot (Leopardus pardalis). Journal of
Tropical Ecology 24:1–12.
DI BITETTI, M. S., A. PAVIOLO, AND C. DE ANGELO. 2006. Density,
habitat use, and activity patterns of ocelots Leopardus pardalis in
the Atlantic Forest of Misiones, Argentina. Journal of Zoology
DI BITETTI, M. S., A. PAVIOLO, C. FERRARI, C. DE ANGELO, AND Y. DI
BLANCO. 2008b. Differential responses to hunting in two sympatric
species of brocket deer (Mazama americana and Mazama nana).
DI BITETTI, M. S., G. PLACCI, AND L. A. DIETZ. 2003. A biodiversity
vision for the Upper Parana´ Atlantic Forest eco-region: designing a
biodiversity conservation landscape and setting priorities for
conservation action. World Wildlife Fund, Washington, D.C.
DIMITRI, M. J. 1974. La flora arbo´rea del Parque Nacional Iguazu´.
Anales de Parques Nacionales 12:1–180.
DONADIO, E., AND S. W. BUSKIRK. 2006. Diet, morphology, and
interspecific killing in Carnivora. American Naturalist 167:524–536.
EMMONS, L. H. 1987. Comparative feeding ecology of felids in a
neotropical rainforest. Behaviour, Ecology and Sociobiology
FRANKLIN, W. L., W. E. JOHNSON, R. J. SARNO, AND J. A. IRIARTE. 1999.
Ecology of the Patagonia puma Felis concolor patagonica in
southern Chile. Biological Conservation 90:33–40.
GANNON, W. L., R. S. SIKES, AND THE ANIMAL CARE AND USE COMMITTEE
OF THE AMERICAN SOCIETY OF MAMMALOGISTS. 2007. Guidelines of
the American Society of Mammalogists for the use of wild
mammals in research. Journal of Mammalogy 88:809–823.
GIRAUDO, A. R., AND R. R. ABRAMSON. 2000. Diversidad cultural y
usos de la fauna silvestre por los pobladores de la selva misionera:
¿Una alternativa de conservacio´n? Pp. 233–243 in La situacio´n
ambiental Argentina 2000 (C. Bertonatti and J. Corcuera, eds.).
Fundacio´n Vida Silvestre, Buenos Aires, Argentina.
HAINES, A. M. 2006. Is there competition between sympatric jaguar
Panthera onca and puma Puma concolor? Acta Zoologica Sinica
HORNOCKER, M. G. 1970. An analysis of mountain lion predation upon
mule deer and elk in the Idaho Primitive Area. Wildlife
KARANTH, K. U. 1995. Estimating tiger Panthera tigris populations
from camera trap data using capture–recapture models. Biological
KARANTH, K. U., AND J. D. NICHOLS. 2002. Monitoring tigers and their
prey: a manual for researchers, managers and conservationists in
tropical Asia. Centre for Wildlife Studies, Bangalore, India.
KELLY, M. J., ET AL. 2008. Estimating puma densities from camera
trapping across three study sites: Bolivia, Argentina, Belize.
Journal of Mammalogy 89:408–418.
LAUNDRE, J. W. 2005. Prefacio en espan˜ol. Pp. xiii–xiv in Guı´a de
manejo del puma (Cougar Management Guidelines Working
Group, eds.). Wildfuture, Bainbridge, Washington.
LOGAN K. A, L. L. IRWIN, AND R. SKINNER. 1986. Characteristics of a
hunted mountain lion population in Wyoming. Journal of Wildlife
LOGAN, K. A., AND L. L. SWEANOR. 2001. Desert puma: evolutionary
ecology and conservation of an enduring carnivore. Island Press,
MAEHR, D. S. 1997. The Florida panther, life and death of a vanishing
carnivore. Island Press, Washington, D.C.
MAEHR, D. S., E. D. LAND, J. C. ROOF, AND J. W. MCCOWN. 1990. Day
beds, natal dens, and activity of Florida panthers. Proceedings of
the Annual Conference of the Southeastern Association of Fish and
Wildlife Agencies 44:310–318.
MAFFEI, L., AND A. J. NOSS. 2007. How small is too small? Camera
trap survey areas and density estimates for ocelots in the Bolivian
Chaco. Biotropica 40:71–75.
MAZZOLLI, M., M. E. GRAIPEL, AND N. DUNSTONE. 2002. Mountain
lion depredation in southern Brazil. Biological Conservation
NOVARO, A. J., AND R. S. WALKER. 2005. Human-induced changes in
the effect of top carnivores on biodiversity in Patagonia. Pp. 268–
288 in Large carnivores and the conservation of biodiversity: does
conserving one save the other? (J. C. Ray, et al., eds.). Island Press,
OTIS, D. L., K. P. BURNHAM, G. C. WHITE, AND D. R. ANDERSON. 1978.
Statistical inference from capture data on closed animal populations. Wildlife Monographs 62:1–135.
PAVIOLO, A. 2002. Abundancia de presas potenciales de yaguarete´
(Panthera onca) en a´reas protegidas y no protegidas de la Selva
Paranaense, Argentina. Degree in Biological Sciences thesis,
Universidad Nacional de Co´rdoba, Co´rdoba, Argentina.
PAVIOLO, A., C. DE ANGELO, Y. DI BLANCO, AND M. S. DI BITETTI.
2008. Jaguar population decline in the Upper Parana´ Atlantic
Forest of Argentina and Brazil. Oryx 42:554–561.
PAVIOLO, A., C. DE ANGELO, Y. DI BLANCO, AND M. S. DI BITETTI.
In press. Efecto de la caza furtiva y el nivel de proteccio´n en la
abundancia de los grandes mamı´feros del Bosque Atla´ntico de
Misiones. In Contribuciones para la conservacio´n y manejo
del Parque Nacional Iguazu´ (B. Carpinetti and M. Garciarena,
eds.). Administracio´n de Parques Nacionales, Buenos Aires,
PIERCE, B. M., V. C. BLEICH, AND R. T. JENKINS. 2000. Social organization of mountain lions: does a land tenure system regulate
population size? Ecology 81:1533–1543.
REXSTAD, E., AND K. P. BURNHAM. 1991 User’s guide for interactive
program CAPTURE. Abundance estimation of closed populations.
Colorado State University, Fort Collins.
SCHALLER, G. B., AND P. G. CRAWSHAW, JR. 1980. Movement patterns
of jaguar. Biotropica 12:161–168.
SCOGNAMILLO, D., I. E. MAXIT, M. E. SUNQUIST, AND J. POLISAR. 2003.
Coexistence of jaguar (Panthera onca) and puma (Puma concolor)
in a mosaic landscape in the Venezuelan llanos. Journal of Zoology
SILVER, S. C., ET AL. 2004. The use of camera traps for estimating
jaguar Panthera onca abundance and density using capture/
recapture analysis. Oryx 38:148–154.
JOURNAL OF MAMMALOGY
SUNQUIST, M., AND F. SUNQUIST. 2002. Wild cats of the world.
University of Chicago Press, Chicago, Illinois.
TROLLE, M., AND M. KERY. 2003. Estimation of ocelot density in the
Pantanal using capture–recapture analysis of camera-trapping data.
Journal of Mammalogy 84:607–614.
VAN DYKE, S. G., R. H. BROCKE, H. G. SHAW, B. B. ACKERMAN, T. P.
HEMKER, AND F. G. LINDZEY. 1986. Reactions of mountain lions to
logging and human activities. Journal of Wildlife Management
Vol. 90, No. 4
WHITE, G. C., D. R. ANDERSON, K. P. BURNHAM, AND D. L. OTIS. 1982.
Capture–recapture and removal methods for sampling closed populations. Los Alamos National Laboratory, Los Alamos, New Mexico.
YOUNG, S. P., AND E. A. GOLDMAN. 1946. The puma, mysterious
American cat. American Wildlife Institute, Washington, D.C.
Submitted 18 April 2008. Accepted 8 January 2009.
Associate Editor was Rodrigo A. Medellı´n.