Post by warsaw on Dec 3, 2011 2:33:49 GMT -9
Andrés Ordiz · Ole-Gunnar Støen · Miguel Delibes ·
Jon E. Swenson
Abstract Prey usually adjust anti-predator behavior to
subtle variations in perceived risk. However, it is not clear
whether adult large carnivores that are virtually free of natural
predation adjust their behavior to subtle variations in
human-derived risk, even when living in human-dominated
landscapes. As a model, we studied resting-site selection by
a large carnivore, the brown bear (Ursus arctos), under
diVerent spatial and temporal levels of human activity. We
quantiWed horizontal and canopy cover at 440 bear beds
and 439 random sites at diVerent distances from human settlements,
seasons, and times of the day. We hypothesized
that beds would be more concealed than random sites and
that beds would be more concealed in relation to humanderived
risk. Although human densities in Scandinavia are
the lowest within bear ranges in Western Europe, we found
an eVect of human activity; bears chose beds with higher
horizontal and canopy cover during the day (0700–
1900 hours), especially when resting closer to human settlements,
than at night (2200–0600 hours). In summer/fall
(the berry season), with more intensive and dispersed
human activity, including hunting, bears rested further from
human settlements during the day than in spring (pre-berry
season). Additionally, day beds in the summer/fall were the
most concealed. Large carnivores often avoid humans at a
landscape scale, but total avoidance in human-dominated
areas is not possible. Apparently, bears adjust their behavior
to avoid human encounters, which resembles the way
prey avoid their predators. Bears responded to Wne-scale
variations in human-derived risk, both on a seasonal and a
daily basis.
Introduction
In order to maximize their Wtness, prey modify their habitat
use and movement patterns in response to predators, as they
must balance risks (mainly predation) and potential beneWts
(e.g., foraging; Lima and Dill 1990). In a humanized biosphere
(e.g., Vitousek et al. 1986), man has become a universal
predator. Human activity and developments induce
mortality, and human recreation disturbs the dynamics or
the eco-ethological characteristics of populations of many
animal taxa (Blanc et al. 2006). Evolutionarily adaptive
behaviors have likely arisen after a long coexistence
between predators and prey, and it has been argued that disturbance
stimuli could even be analogous to predation risk
from an evolutionary perspective (Frid and Dill 2002).
Species with large spatial requirements, such as large
carnivores, use multiple-use landscapes at a large scale
(Noss et al. 1996), and in human-dominated environments a
very high proportion of their adult mortality is humaninduced
(WoodroVe and Ginsberg 1998). Thus, large carnivores
may alter their behavior to reduce encounters with
humans, e.g., by becoming more nocturnal, avoiding areas
with high human activity, or hiding in dense vegetation
(Boydston et al. 2003). It is well documented that prey can
reduce their perceived risk and fear via behavioral and morphological
modiWcations to the same degree that predator
behavior can increase the perception of risk in prey (see
Stankovich and Blumstein 2005), and prey are able to
adjust their anti-predator behavior to Wne-degree variations
in the perceived risk (e.g., Martín and López 2004). However,
it is not clear whether adult large carnivores that are
virtually free of natural predation, therefore lack anti-predator
behavior, and only recently (in an evolutionary perspective)
have had to cope with humans, are able to Xexibly
adjust their behavior to subtle variations in human-derived
risk.
To address this issue, we used the brown bear (Ursus
arctos) as a model species. Bears are threatened by humancaused
mortality, habitat fragmentation, and habitat loss,
both in North America and Europe (e.g., Servheen et al.
1999). Brown bears need large areas of habitat with suYcient
availability of food and cover to satisfy their lifetime
requirements (e.g., Swenson et al. 2000). The persistence of
brown bear populations depends on habitat quality, human
density, and human behavior (e.g., Mattson et al. 1996).
Indeed, brown bears tend to avoid human activity throughout
their range (e.g., Mace et al. 1996). In a large area that
included our study site, Nellemann et al. (2007) found that,
for comparable habitat and terrain types, bear density
increased substantially with increasing distance to towns
and resorts. Bears also avoid humans by altering their temporal
use of areas with high recreational activities (Rode
et al. 2006a).
To avoid encounters, detection, and capture, prey have
developed behavioral defenses against predators. As a metric
of wariness, much attention has been paid to the distance
at which an individual prey initiates Xight when
approached by a predator (e.g., Blumstein 2006). Other surrogates
of perceived risk include the amount of activity in
the presence versus absence of predators (Stoks et al.
2003), shifts in habitat choice (Kotler et al. 1991), frequency
of vigilance or use of alarm signals (Blumstein
2007), and group size (Heard 1992). Most of these
approaches are not strictly applicable for large carnivores,
which are usually scarce, solitary, elusive, and nocturnal.
However, radio-tracking techniques can allow the identiWcation
of responses of large carnivores to human presence,
because large carnivores usually avoid humans at a large
scale. For example, the location of wolf (Canis lupus)
breeding and rendezvous sites depends more on the distribution
of villages and roads than on habitat types (Theuerkauf
et al. 2003), and Scandinavian bears try to avoid
large human settlements and resorts (Nellemann et al.
2007) and select winter dens where human disturbance
could be minimized (Elfström et al. 2008).
However, total avoidance of people in human-dominated
landscapes is not possible. Cover is a key habitat factor that
may lower the risk of mortality by reducing the chance of
detection and hindering attacks (Mysterud and Østbye
1999), and it is important for many animal groups, including
carnivores. For instance, American black bears (Ursus
americanus) select for a mosaic of habitat types that provide
cover near food resources (Lyons et al. 2003); European
badgers (Meles meles) need cover for the selection of
diurnal resting dens, even in a highly protected national
park where disturbance by people is expected to be low
(Revilla et al. 2001); spotted hyenas (Crocuta crocuta) are
able to persist in areas with increasing livestock pressure by
relying on dense cover (Boydston et al. 2003); and the
availability of cover for resting during daytime is probably
the limiting habitat requirement for European lynx (Lynx
lynx) in human-dominated areas (Sunde et al. 1998). Also,
the availability of cover is important in determining how
human activities inXuence brown bear habitat use (e.g.,
Suring et al. 2006). Bears may adapt to living in humandominated
landscapes by choosing denser habitat when
closer to people. Bears in the Italian Alps avoid areas of
major human activity and forage most intensively in areas
with a high degree of cover (Preatoni et al. 2005).
Prey sensitivity to the risk of predation while resting is
shown by the diVerent tactics frequently used, e.g., hiding
cryptically in the safest possible areas, orienting the detection
senses toward the most likely direction of approach by
a predator, or forming groups (see Semeniuk and Dill
2005). Improving our knowledge on animals’ adaptive
behaviors, such as foraging, patch use, and habitat selection,
can be useful for conservation (Morris et al. 2009).
For large carnivores, activities such as feeding (e.g., diet
analyses) or movement (e.g., home-range estimations) are
better documented than resting-site selection, although animals
devote much time to resting. For instance, Scandinavian
brown bears rest »12 h a day, with a short (»3 h)
resting period in the night and the longest (»9 h) in the day
(Moe et al. 2007). Legal hunting, which takes place during
the day, is the single most important cause of mortality for
brown bears in Sweden, with human-caused mortality
accounting for 86.5% of conWrmed deaths of radio-marked
animals between 1984 and 2006 (Bischof et al. 2009).
Thus, it can be expected that large carnivores are very
selective when choosing their daily resting sites, which
makes the study of this selection of prime importance for
their ecology and conservation. Technical limitations have
likely hindered such Wne-scale studies until recently.
We have used GPS radio-tracking combined with Weldwork
to estimate the “fear” (i.e., the perceived risk; sensu
Stankovich and Blumstein 2005) of bears towards humans
at a Wne scale, by measuring the amount of concealment
around their temporary resting sites (beds). We analyzed
the concealment, i.e., horizontal and canopy cover, at beds
of GPS-collared bears at diVerent distances from human
settlements, diVerent times of the day, and in relation to
seasonal shifts in human activities throughout the non-denning
season. We aimed to understand the very Wne-scale
behavior and requirements of a large carnivore when
choosing resting sites in relation to human-derived risk. If
Wne-scale adjustment in bed selection were a response to
human-derived risk, we would expect to Wnd that beds were
(1) more concealed than random sites, (2) more concealed
during the day, when humans are more active, than during
the night, (3) more concealed close to human settlements
than further away, and (4) more concealed in late summerearly
fall (when human activity outdoors is most common
and includes hunting) than in spring.
More
www.bearproject.info/uploads/publications/2011%20Ordiz%20et%20al%20Beds%20Oecologia.pdf
Jon E. Swenson
Abstract Prey usually adjust anti-predator behavior to
subtle variations in perceived risk. However, it is not clear
whether adult large carnivores that are virtually free of natural
predation adjust their behavior to subtle variations in
human-derived risk, even when living in human-dominated
landscapes. As a model, we studied resting-site selection by
a large carnivore, the brown bear (Ursus arctos), under
diVerent spatial and temporal levels of human activity. We
quantiWed horizontal and canopy cover at 440 bear beds
and 439 random sites at diVerent distances from human settlements,
seasons, and times of the day. We hypothesized
that beds would be more concealed than random sites and
that beds would be more concealed in relation to humanderived
risk. Although human densities in Scandinavia are
the lowest within bear ranges in Western Europe, we found
an eVect of human activity; bears chose beds with higher
horizontal and canopy cover during the day (0700–
1900 hours), especially when resting closer to human settlements,
than at night (2200–0600 hours). In summer/fall
(the berry season), with more intensive and dispersed
human activity, including hunting, bears rested further from
human settlements during the day than in spring (pre-berry
season). Additionally, day beds in the summer/fall were the
most concealed. Large carnivores often avoid humans at a
landscape scale, but total avoidance in human-dominated
areas is not possible. Apparently, bears adjust their behavior
to avoid human encounters, which resembles the way
prey avoid their predators. Bears responded to Wne-scale
variations in human-derived risk, both on a seasonal and a
daily basis.
Introduction
In order to maximize their Wtness, prey modify their habitat
use and movement patterns in response to predators, as they
must balance risks (mainly predation) and potential beneWts
(e.g., foraging; Lima and Dill 1990). In a humanized biosphere
(e.g., Vitousek et al. 1986), man has become a universal
predator. Human activity and developments induce
mortality, and human recreation disturbs the dynamics or
the eco-ethological characteristics of populations of many
animal taxa (Blanc et al. 2006). Evolutionarily adaptive
behaviors have likely arisen after a long coexistence
between predators and prey, and it has been argued that disturbance
stimuli could even be analogous to predation risk
from an evolutionary perspective (Frid and Dill 2002).
Species with large spatial requirements, such as large
carnivores, use multiple-use landscapes at a large scale
(Noss et al. 1996), and in human-dominated environments a
very high proportion of their adult mortality is humaninduced
(WoodroVe and Ginsberg 1998). Thus, large carnivores
may alter their behavior to reduce encounters with
humans, e.g., by becoming more nocturnal, avoiding areas
with high human activity, or hiding in dense vegetation
(Boydston et al. 2003). It is well documented that prey can
reduce their perceived risk and fear via behavioral and morphological
modiWcations to the same degree that predator
behavior can increase the perception of risk in prey (see
Stankovich and Blumstein 2005), and prey are able to
adjust their anti-predator behavior to Wne-degree variations
in the perceived risk (e.g., Martín and López 2004). However,
it is not clear whether adult large carnivores that are
virtually free of natural predation, therefore lack anti-predator
behavior, and only recently (in an evolutionary perspective)
have had to cope with humans, are able to Xexibly
adjust their behavior to subtle variations in human-derived
risk.
To address this issue, we used the brown bear (Ursus
arctos) as a model species. Bears are threatened by humancaused
mortality, habitat fragmentation, and habitat loss,
both in North America and Europe (e.g., Servheen et al.
1999). Brown bears need large areas of habitat with suYcient
availability of food and cover to satisfy their lifetime
requirements (e.g., Swenson et al. 2000). The persistence of
brown bear populations depends on habitat quality, human
density, and human behavior (e.g., Mattson et al. 1996).
Indeed, brown bears tend to avoid human activity throughout
their range (e.g., Mace et al. 1996). In a large area that
included our study site, Nellemann et al. (2007) found that,
for comparable habitat and terrain types, bear density
increased substantially with increasing distance to towns
and resorts. Bears also avoid humans by altering their temporal
use of areas with high recreational activities (Rode
et al. 2006a).
To avoid encounters, detection, and capture, prey have
developed behavioral defenses against predators. As a metric
of wariness, much attention has been paid to the distance
at which an individual prey initiates Xight when
approached by a predator (e.g., Blumstein 2006). Other surrogates
of perceived risk include the amount of activity in
the presence versus absence of predators (Stoks et al.
2003), shifts in habitat choice (Kotler et al. 1991), frequency
of vigilance or use of alarm signals (Blumstein
2007), and group size (Heard 1992). Most of these
approaches are not strictly applicable for large carnivores,
which are usually scarce, solitary, elusive, and nocturnal.
However, radio-tracking techniques can allow the identiWcation
of responses of large carnivores to human presence,
because large carnivores usually avoid humans at a large
scale. For example, the location of wolf (Canis lupus)
breeding and rendezvous sites depends more on the distribution
of villages and roads than on habitat types (Theuerkauf
et al. 2003), and Scandinavian bears try to avoid
large human settlements and resorts (Nellemann et al.
2007) and select winter dens where human disturbance
could be minimized (Elfström et al. 2008).
However, total avoidance of people in human-dominated
landscapes is not possible. Cover is a key habitat factor that
may lower the risk of mortality by reducing the chance of
detection and hindering attacks (Mysterud and Østbye
1999), and it is important for many animal groups, including
carnivores. For instance, American black bears (Ursus
americanus) select for a mosaic of habitat types that provide
cover near food resources (Lyons et al. 2003); European
badgers (Meles meles) need cover for the selection of
diurnal resting dens, even in a highly protected national
park where disturbance by people is expected to be low
(Revilla et al. 2001); spotted hyenas (Crocuta crocuta) are
able to persist in areas with increasing livestock pressure by
relying on dense cover (Boydston et al. 2003); and the
availability of cover for resting during daytime is probably
the limiting habitat requirement for European lynx (Lynx
lynx) in human-dominated areas (Sunde et al. 1998). Also,
the availability of cover is important in determining how
human activities inXuence brown bear habitat use (e.g.,
Suring et al. 2006). Bears may adapt to living in humandominated
landscapes by choosing denser habitat when
closer to people. Bears in the Italian Alps avoid areas of
major human activity and forage most intensively in areas
with a high degree of cover (Preatoni et al. 2005).
Prey sensitivity to the risk of predation while resting is
shown by the diVerent tactics frequently used, e.g., hiding
cryptically in the safest possible areas, orienting the detection
senses toward the most likely direction of approach by
a predator, or forming groups (see Semeniuk and Dill
2005). Improving our knowledge on animals’ adaptive
behaviors, such as foraging, patch use, and habitat selection,
can be useful for conservation (Morris et al. 2009).
For large carnivores, activities such as feeding (e.g., diet
analyses) or movement (e.g., home-range estimations) are
better documented than resting-site selection, although animals
devote much time to resting. For instance, Scandinavian
brown bears rest »12 h a day, with a short (»3 h)
resting period in the night and the longest (»9 h) in the day
(Moe et al. 2007). Legal hunting, which takes place during
the day, is the single most important cause of mortality for
brown bears in Sweden, with human-caused mortality
accounting for 86.5% of conWrmed deaths of radio-marked
animals between 1984 and 2006 (Bischof et al. 2009).
Thus, it can be expected that large carnivores are very
selective when choosing their daily resting sites, which
makes the study of this selection of prime importance for
their ecology and conservation. Technical limitations have
likely hindered such Wne-scale studies until recently.
We have used GPS radio-tracking combined with Weldwork
to estimate the “fear” (i.e., the perceived risk; sensu
Stankovich and Blumstein 2005) of bears towards humans
at a Wne scale, by measuring the amount of concealment
around their temporary resting sites (beds). We analyzed
the concealment, i.e., horizontal and canopy cover, at beds
of GPS-collared bears at diVerent distances from human
settlements, diVerent times of the day, and in relation to
seasonal shifts in human activities throughout the non-denning
season. We aimed to understand the very Wne-scale
behavior and requirements of a large carnivore when
choosing resting sites in relation to human-derived risk. If
Wne-scale adjustment in bed selection were a response to
human-derived risk, we would expect to Wnd that beds were
(1) more concealed than random sites, (2) more concealed
during the day, when humans are more active, than during
the night, (3) more concealed close to human settlements
than further away, and (4) more concealed in late summerearly
fall (when human activity outdoors is most common
and includes hunting) than in spring.
More
www.bearproject.info/uploads/publications/2011%20Ordiz%20et%20al%20Beds%20Oecologia.pdf