Comparing diel activity patterns of individual bears

advertisement
Kaczensky et al.
1
2
3
4
5
6
7
8
9
10
6 September 2005
Dr. Petra Kaczensky
Department of Wildlife Ecology and Management
Institute of Forest Zoology
University of Freiburg
Tennenbacher Strasse 4
D-79085 Freiburg, Germany
Tel. +49-761-203-3799
Fax +49-761-203-3667
e-mail: petra.kaczensky@wildlife.uni-freiburg.de
11
12
13
RH: Activity patterns of brown bears
14
Activity patterns of brown bears in Slovenia and Croatia
15
16
17
Petra Kaczensky1,*, Djuro Huber2, Felix Knauer3,*, Hans Roth4, Axel Wagner5, and Josip
18
Kusak2
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
1
Institute of Wildlife Biology and Game Management at the Agricultural University of
Vienna, Peter Jordan Straße 76, A-1190 Vienna , Austria.
2
Biology Department, Veterinary Faculty of the University of Zagreb, Heinzelova 55, HR10000 Zagreb, Croatia.
Department of Wildlife Ecology and Management, Institute of Forest Zoology, University of
Freiburg
3
4
Centro Studi Ecologici Appenninici, Parco Nazionale d´Abruzzo, Viale S. Lucia, I-67032,
Italy.
5
Institute of Evolutionary Biology and Ecology, University of Bonn, An der Immenburg 1, D53121 Bonn, Germany.
*
Current address: Department of Wildlife Ecology and Management, Institute of Forest Zoology, University of
Freiburg, Tennenbacher Strasse 4, D-79085 Freiburg, Germany. E-mail: petra.kaczensky@wildlife.unifreiburg.de
1
Kaczensky et al.
1
ABSTRACT
2
3
In most of Europe true wilderness areas do not exist and brown bears (Ursus arctos) generally
4
have to cope with human disturbance and infrastructure. The few studies in Europe that
5
investigated brown bear activity demonstrated a predominantly nocturnal and “shy” behaviour
6
in bears. There is still quite a debate whether the shy, nocturnal bears of Europe are the result
7
of centuries of persecution by men (genetically fixed trait) or if hunting and the high
8
disturbance potential in the multi-use landscapes are the driving force (individual learnt trait).
9
We analysed the activity pattern of 16 individual bears monitored for 3,372 hours between
10
May and October 1982-1998 in the Dinaric Mountains of Slovenia and Croatia. The data were
11
collected via time sampling and basically analysed with two approaches: a general linear
12
model (GLM) with seasonal component to delineate the most important variables influencing
13
the activity pattern and level and cluster analysis to group bears according to their 24-hour
14
activity pattern. Time of the day and age were the most important variables predicting
15
activity. Although individual variation in the activity pattern was high among individual
16
bears, in general, yearlings were more diurnal and had a less distinct difference between day-
17
and night time activity levels than adult bears. Subadults were somewhat intermediate to
18
adults and yearlings. We believe that nocturnal behaviour is most likely driven through
19
negative experiences with humans, giving space for much individual variation. More research
20
is needed to prove the causal relationship of nocturnal behaviour and the degree of
21
disturbance an individual bear is exposed to.
22
23
24
Key words: activity pattern; brown bear; circadian rhythm; Croatia; disturbance; multi-use
landscape, Slovenia; Ursus arctos.
25
2
Kaczensky et al.
1
INTRODUCTION
2
3
The activity pattern of animals is determined by internal and external factors. Most
4
vertebrates are either active during the day (diurnal) or during the night (nocturnal), often with
5
peaks of high activity around dawn or dusk (Ashby, 1972 cited in Halle & Stenseth, 1994,
6
Enright 1970). Light and temperature act as external synchronizer for a certain activity pattern
7
(Nielsen, 1983), while the availability of resources, competition, predation (Geffen &
8
MacDonald, 1993) or disturbance (Liddle, 1997) may alter the genetically fixed and
9
physiologically regulated circadian rhythm.
10
Few studies in Europe have addressed activity patterns in brown bears (Ursus arctos) due
11
to the difficulties of observing and following a far-ranging, forest dwelling species (Swenson
12
et al., 2000). The few studies in Europe that investigated brown bear activity demonstrated a
13
predominantly nocturnal activity pattern (Roth, 1980; Roth & Huber, 1986; Clevenger, Purroy
14
& Pelton, 1990; Wabakken & Maartmann 1994; Rauer & Gutleb, 1997). In addition, bears are
15
known to avoid people and are considered “shy” (avoid people, e.g. by running upon first
16
sight of humans) and un-aggressive animals (Metz, 1990; Swenson, et al. 1996; Swenson et
17
al., 1999; Swenson, 1999; Zedrosser et al., 2001).
18
In North America, on the other hand, brown bears seem largely diurnal (Klinka & Reichen,
19
2002) and more aggressive than their European counterparts (Swenson et al., 1996; Swenson
20
et al., 1999). However, contrary to North America, brown bears in Europe co-exist with
21
humans in densely settled, multi-use landscapes (Mattson, 1990; Swenson et al, 2000; Linnell
22
et al., 2002). Bear habitat in Europe is almost exclusively restricted to forested area (Knauer,
23
2000; Knauer, Kaczensky & Rauer, 2000), but these forests are heavily managed by foresters
24
and hunters and are used by the general public for camping, hiking, mountain biking, and
25
berry- and mushroom picking. Because almost all of the human activities are confined to
26
daytime hours, bears can largely avoid encounters with humans by being nocturnal.
3
Kaczensky et al.
1
In North American diurnal activity levels of brown bears may also vary with the intensity
2
of human presence. In areas with low intensity of human utilization, bears are largely diurnal,
3
while in areas with high intensity of human utilization or during periods of frequent human
4
access, bears shift to nocturnal behaviour (Gunther, 1990; Klinka & Reichen, 2002;
5
MacHutchon et al., 1998; Olson, Squibb & Gilbert, 1998). There are also situations when in
6
absence on aggressive human behaviour (hunting) and with access to abundant forage, bears
7
may become human habituated and shift back to a diurnal activity pattern (Smith, Herrero &
8
DeBruyn, 2005).
9
In Europe the pressure on bears to behave inconspicuously and to avoid humans is high.
10
Bears that do not avoid people run a high risk of being shot in the hunted populations, and to
11
become food conditioned and subsequently removed as “problem bears” in protected
12
populations (Rauer, Kaczensky & Knauer, 2003). In addition, diurnal bears have a higher risk
13
for surprise encounters with humans, which occasionally have resulted in human injury or
14
death (Adamič, 1996; Swenson et al., 1996; Swenson et al., 1999) and subsequently in the
15
killing of the involved bears. Public acceptance of brown bears is greatly challenged by such
16
accidents and may hinder or even stop bear conservation efforts (Kaczensky, 2000a+b;
17
Kaczensky, Blazic & Gossow, 2004). Therefore, shy and nocturnal behavior in European
18
brown bears is preferable and deserves more intensive studies.
19
There is still quite a debate whether the shy, nocturnal bears of Europe are the result of
20
centuries of persecution by men (genetically fixed trait) or if hunting and the high disturbance
21
potential in the multi-use landscapes are the driving force (individual learnt trait; also see
22
Swenson, et al. 1996; Swenson, 1999). This question could be tested by comparing activity
23
patterns and behaviours of brown bears in wilderness environments versus those in a multi
24
use landscapes. Whereas these comparisons are possible in North America and data suggest
25
the importance of individual learning; in densely settled Europe large wilderness areas do not
4
Kaczensky et al.
1
exist any more (EUROPARC & IUCN, 2000) and human population densities are high, even
2
in the remaining bear areas (Mattson, 1990; Swenson et al., 2000). In addition, almost all
3
protected areas are smaller than an individual bears’ home range (Huber & Roth, 1993;
4
Linnell et al., 2002), hunting is an integral part of game and bear management in most areas
5
(Swenson et al., 2000), and human access to forested areas is facilitated by a dense network of
6
forest roads (Kaczensky, 2000b). Unfortunately, data on brown bears from wilderness areas in
7
Asia is largely anecdotal (Swenson, 1999).
8
Thus the question whether nocturnal behaviour in European brown bears is genetically
9
fixed or learnt can only be approached indirectly. If the nocturnal behaviour is genetically
10
fixed, we would expect the same nocturnal activity pattern for all bears with a low degree of
11
individual variation. Though the activity pattern might vary among age classes because of
12
maturity effects, there should be little individual variation within the same age class.
13
If, on the other hand nocturnal activity is a result of individual learning and experience, we
14
would expect a high level of individual variation. Because age is a surrogate for other, more
15
germane effects, more difficult to measure - in particular, the cumulative effects of learning
16
and life experience, as well as somatic changes that include increasing size - we would expect
17
age to be a key variable that determines the activity pattern in bears.
18
In order to study this question, we analysed the activity pattern of 16 individual bears from
19
the continuous brown bear population in the Dinaric mountain range of Slovenia and Croatia.
20
Our expectation was that nocturnal activity is a result of individual learning and experience
21
and we expected to see:
22
(1)
Age is a key variable explaining differences in the activity pattern of individual
bears
23
24
(2)
Young, un-experienced bears are be less nocturnal than adult, experienced bears
25
(3)
Individual variation is high among and between different age classes
5
Kaczensky et al.
1
STUDY AREA
2
3
All three study areas were located in the Dinaric Mountain range in Slovenia (Menisija),
4
and Croatia (Gorski Kotar and Plitvice Lakes) (Fig. 1). They are within the range of the
5
contiguous Dinara-Pindus bear population stretching from Slovenia in the north into Albania
6
and Greece in the south. The total number is estimated at about 2,800 bears, of which 300-500
7
are believed to live in Slovenia and about 600-1,000 in Croatia ( Dečak et al., 2004; Servheen,
8
Herrero & Peyton, 1998; Swenson et al., 2000; Zedrosser et al., 2001). In both countries bears
9
are hunted between 1 October and 30 April after a quota system (Huber & Frkovic, 1993;
10
11
Simonic 1994).
The relief shows typical karst phenomena, water sink holes (dolines), steep canyons, caves
12
and shallow soils. Surface water is rare as water run off is largely underground. Periodical
13
lakes (poljes) and rivers that submerge after short distances are typical landscape features.
14
Elevations range from 300m to 1200m in the Menisija region, from 600m to 1500m in Gorski
15
Kotar and from 500m to 1200m in the Plitvice Lakes area. Bear habitat consists of mixed,
16
uneven aged forest stands. The most common forest community in our study areas (Abieti-
17
Fagetum dinaricum) is dominated by beech (Fagus silvatica) and fir (Abies alba) -
18
intermingled with varying amounts of spruce (Picea abies), maple (Acer pseudoplatanus) and
19
elm (Ulmus spec.). Only selective cutting is allowed, resulting in a dense network of forest
20
roads (1.5-2.0 km roads per km²; Kusak & Huber, 1998; Kaczensky, 2000b), most of them
21
opened for public use. Overall forest cover is high and varies between 66% in Gorski Kotar
22
and 74% and 75% in the Menisija and Plitvice Lakes area, respectively.
23
Human population density is low to moderate for European standards and ranges from 13
24
inhabitants per km² in the Plitvice Lakes area to 27 and 42 inhabitants per km² in the Gorski
25
Kotar and Menisija area, respectively. Concerning the intensity of human use, the Plitvice
26
Lakes area is famous for Plitvice Lakes National Park (200 km2), which is a major tourist
6
Kaczensky et al.
1
attraction. The Menisija area (1,500 km²) is located only 30 km from Ljubljana, the capital of
2
Slovenia and is heavily used for recreation. The Gorski Kotar area (1,500 km2) is known for
3
Risnjak National Park (30 km2). Even though there is no bear hunting within Risnjak and
4
Plitvice Lakes National Park, movements of all monitored bears covered ranges larger than
5
the parks (Huber & Roth, 1993) and it can be assumed that bears in all three areas are exposed
6
to hunting pressure at least in part of their range. Additional disturbance arises from forestry
7
operations, collections of plant parts, and various recreational uses (camping, hiking,
8
mountain-biking etc.).
9
10
A more detailed description of the study areas can be found in Huber & Roth (1993), Roth
& Huber (1986), Kusak & Huber (1998), and Kaczensky (2000b).
11
12
MATERIAL AND METHODS
13
14
Capture and radio marking
15
16
We captured bears with Aldrich foot snares at bait sites and chemically immobilized them
17
with either Tiletamin and Zolazepam or a mixture of Ketamin and Xylazine. Traps were
18
checked each morning in the Croatin study area, and continuously monitored with trap
19
transmitters in the Slovenian study area. Trapping, chemical restrain and radio-marking
20
procedures followed methods described by Huber, Kusak & Radisic (1996) and Kaczensky et
21
al. (2002). A rudimentary first premolar tooth was extracted for age estimation (Stoneberg &
22
Jonkel 1966).
23
Bears were fitted with different types of radio collars (Croatia: Advanced Telemetry
24
Systems (ATS), USA, AVM Instruments Company, USA and Telonics, USA; Slovenia:
25
MOD-600, MOD-400 Telonics, USA) or ear tag transmitters (EL-2(42), Holohil, Canada).
7
Kaczensky et al.
1
Each collar was fitted with a drop off system made of corroding wires or cotton spacers to
2
prevent the life long wearing of the device.
3
4
Activity monitoring
5
6
Bear activity was monitored via time sampling (Tyler, 1979). Only clearly audible signals
7
were used to determine the activity status of monitored bears. We used analogue receivers
8
with a meter for signal strength that can be adjusted to reception strength (Croatia: AVM
9
Instruments Company, USA; Slovenia: YAESU, Wagener, Germany).
10
In Croatia, bear activity was recorded at sampling intervals of 15 minutes. Observers
11
listened to at least 30 signal pulses (“beeps”) and classified the bear “active” if at least 4 out
12
of 30 signal pulses clearly differed in signal strength, else “inactive” (for details refer to:
13
Roth, 1980; Roth & Huber, 1986). Visual observations during wake-up and random
14
observations during field work showed that ttypically an active animal showed more than 4
15
fluctuation of signal volume over 30 beeps (D. Huber, unpubl. Data). In this way we realized
16
a maximum of 4 activity samples per bear and hour during each monitoring session.
17
In Slovenia, bear activity was recorded slightly different. Activity was checked at sampling
18
intervals of 10 minutes and observers listened for a one-minute sampling period. Like in
19
Croatia, we considered the bear active if the strength in the signal pulse was clearly different
20
in ≥4 beeps. This activity criterion was previously determined from Zoo experiment, where
21
we had simultaneously radio-monitored and observed a radio-collared bear (Kaczensky,
22
Wagner & Walzer, 2004). In Slovenia we realized a maximum of 6 activity samples per bear
23
and hour during each monitoring session. Previous tests with different sampling intervals
24
have shown only minor differences for sampling intervals of 10 versus 15 minutes (I.
25
Reinhardt unpubl. Data).
8
Kaczensky et al.
1
Each activity sample resulted in information on the activity status, coded as a dichotomous
2
variable and the time of the day (middle European standard time (MEZ)). To compare
3
different bears we only used data collected between 1 May and 31 October, the time when all
4
bears were active outside their winter dens. We grouped bears according to three age-classes:
5
yearlings (1 year), subadults (2-3 years) and adults (> 4 years). All yearlings monitored had
6
already separated from the mother. All adult females monitored were without offspring.
7
For analysis we only used bears that were monitored for at least 48 hours evenly
8
distributed over all hours (equally covering day- and nighttime hours), which means a
9
minimum of 192 activity samples for the Croatian dataset and 288 activity samples for the
10
Slovenian dataset. In the Croatian study areas, several bears were monitored over several
11
years. The data were pooled if the bear did not change its age class status (3 adult bears), or
12
otherwise subdivided according to age class (1 bear). For all multi-individual analysis, each
13
bear was only used in one age class and the one bears that changed the age class was assigned
14
to the age class with the smaller sample size.
15
For analysis we only used hours with ≥3 activity measures and calculated the average
16
activity for each hour a bear was monitored on a given day. In total we had 3,372 hours of 16
17
individual bears available for analysis (Table 1). For age class specific comparison of activity
18
patterns we additionally used 170 hours from 13 individual bears (each with less than 200
19
activity samples) pooled by age class (Table 2).
20
21
Triangulation of monitored bears
22
23
In the Slovenian study area, we continuously followed bear movements during 24-hour
24
monitoring sessions by car or on foot. Due to a dense network of forest roads, the distance
25
between observer and bear was generally less than 1,000m. After each activity sample we
9
Kaczensky et al.
1
checked whether the bear had changed its position. If so, we determined the new position by
2
triangulation taking successive bearings by a single observer. The accuracy of the position
3
was estimated by the observers from the angle between the different bearings, the signal
4
strength and the topography, and was classified as:
5
(1) Location error < 50m: bear circled on close range and/or radio signal close to maximal
6
(2) Location error < 250m: bear only partly circled or circled at a longer distance and/or
7
8
9
topography limits the maximum distance between bear and observer
(3) Location error < 500m: bear not circled, or circled on long distance (>1 km), and/or
azimuth between most distant bearings less than 120° apart
10
11
Accuracy of locations was opportunistically confirmed when searching for daybeds,
12
feeding signs, or scats the following day (D. Huber & P. Kaczensky, unpubl. Data). We
13
classified bears as travelling (travelling activity), when the distance between two locations
14
was more than the expected location error, as active on the same area (stationary activity) if
15
the change was within the expected location error; these information were only available for
16
the Slovenian data set.
17
18
Standardization of the variable time
19
20
We compared activity among bears based on average activity per hour. However, the data
21
were collected between May and October and thus sunset and sunrise as well as their centre,
22
local noon and local midnight, change over the season. To compare hours with the same light
23
regime we standardized the variable time to the average day of the observation period. On
24
average sunrise was at 4:53 and sunset at 19:10. On days with later (earlier) sunset and earlier
25
(later) sunrise than on the average day, the time between sunrise and sunset was relatively
10
Kaczensky et al.
1
stretched (shortened, respectively). We separately conducted a linear transformation for each
2
day to fit the time intervals to an average day:
3
t st  MPst 
MPr  t r MPst  Sst
, where
4
MPr  S r
tst is the transformed time,
5
tr the measured time (Central European time),
6
MPst the standardized midpoint: local noon (12.00) or midnight (0.00 after midnight or 24.00
7
before midnight),
8
MPr the real midpoint,
9
Sst the standardized sunrise/sunset and
10
Sr the real sunrise/sunset.
11
Noon was used as midpoint (reference) for the time between sunrise and sunset, else
12
midnight. Sunrise was used for the time between midnight and noon, else sunset. This
13
resulted in four different formulas (combinations of different midpoints and sunrise/sunset).
14
Local noon and midnight change over the year due to the elipsoid trajectory of the earth
15
around the sun. By using this transformation long and short days can be directly compared.
16
Although, the real time available during night- and daytime is slightly changed, this does not
17
seem to be problematical, because season had no significant influence on bear activity (see
18
Table 3).
19
20
General linear model
21
22
In order to test for the influence of different variables on activity we used a general linear
23
model (GLM) with age, sex, season, and study area as factor variables. From a first visual
24
analysis of activity patterns, it seemed that most bears roughly followed a monophasic or 24-
25
hour activity rhythm and to a less extent a biphasic or 12-hour rhythm. Due to this periodic
11
Kaczensky et al.
1
behaviour and the circular structure of the time we fitted sine curves with 24 hours and 12
2
hours rhythms (seasonal component) into the model (Zar 1999). As time series data are
3
usually autocorrelated we used time lagged variables of the dependent variable as explanatory
4
variables. The periodic structure of the data results in the absence of a linear trend and thus
5
we did not have to de-trend the data, as done usually in time series analysis.
6
For the GLM we used the following independent variables:
7
Act_n: activity of n hours before - covariate
8
Age: Age class (1=yearling, 2=subadult, 3=adult) - factor
9
Sex: Sex (1=male, 2=female) – factor
10
Season: Month (1=May, 2=June, …. 6=October) - factor
11
Area: Study area (1=MN, 2=GK, 3=PL) - factor
12
24-hour rhythm - covariate
13
12-hour rhythm - covariate
14
and interactions of these variables.
15
We selected variables stepwise in a backwards fashion, removing those that failed to be
16
significant at the 0.05 significance level. For the final model we tested the residuals for
17
normal distribution (which was roughly the case) and confirmed that they were not
18
autocorrelated.
19
20
Comparing diel activity patterns of individual bears
21
22
To describe similarities in the diel activity pattern we run a cluster analysis using squared
23
Euclidean distances between all bears based on average activity level of each hour. Bears
24
were attributed to clusters using the between group linkage - this method uses the average
12
Kaczensky et al.
1
distance between all samples in a cluster to determine the distance to a new cluster, this way
2
considering all samples of a cluster.
3
4
Data from bears with small datasets of activity samples
5
6
We pooled activity samples of an additional 13 bears that did not have the minimum
7
number of activity measures required for individual analysis, first by bear and monitored
8
hours and than by age class (Table 2). We visually compared activity patterns and mean
9
activity levels with the results from the individual analysis.
10
All statistical analysis was done in SPSS 12.0.1 (SPSS Inc., Chicago, USA).
11
12
RESULTS
13
14
Main variables influencing activity
15
16
The outputs of the GLM showed that the activity in the hour before, the underlying 12- and
17
24-hour activity rhythm, and the 24-hour activity rhythm modified by age class were the most
18
important factors predicting activity or non-activity of a given hour (Table 3). Less important
19
were age class alone and the 24-hour activity rhythm modified by sex. Sex by itself was not a
20
significant variable, nor was season or study area. The model explains 45% of the variation
21
(Table 3) in activity and shows differences between age classes in the activity rhythm as well
22
as the overall activity level (Table 3). Differences in overall activity levels were significantly
23
different between adults (52%) and yearlings (62%) with subadults ranging somewhat in-
24
between (57%; post-hoc t-test, p<0.01). Yearlings have a distinct break in activity from 23:00
25
to 3:00, which is comparable to their mid-day break (12-hour or biphasic pattern). Adults, on
13
Kaczensky et al.
1
the other hand, have a more distinct and extended break during the day, but are basically
2
active all night (monophasic or 24-hour rhythm). Subadults were somewhat intermediate
3
(post-hoc t-test, p<0.01 for 24-hour-rhythm and p<0.01 for 12-hour rhythm). All age classes
4
show a combination of both patterns.
5
Bivarate comparisons among age classes showed a significant difference between yearlings
6
and adults only for daytime activity levels (adults 39%, subadults 52%, yearlings 64%;
7
ANOVA P=0.02, Padult-yearling<0.01; Padult-subadult=0.50; Pyearling-subadult=0.48 ), but not for
8
nighttime activity levels, mainly due to higher individual variation (adults 72%, subadults
9
68%, yearlings 60%; ANOVA P=0.32).
10
One bear (LILI) changed the age class during the monitoring period and showed a
11
reduction in the daytime activity level with increasing age. As a yearling (LILI82) she was
12
active 66% during the day and 78% during the night, whereas as a subadult (LIL83) she was
13
active 46% during the day and 77% during night (T-test Pday(yearling-subadult)<0.01, Pnight(yearling-
14
subadult)=0.85;
Fig. 3).
15
16
Comparing diel activity patterns of individual bears
17
18
The results of the cluster analysis grouped bears into three distinct groups. Although the
19
grouping did not sharply separate bears by age, adults tended to be in group 1 (five adults,
20
two subadults and one yearling) or group 3 (two adults) and yearlings in group 2 (one adult,
21
one subadult and three yearlings; Fig. 4).
22
The main differences between groups were that bears in group 2 have a less distinct
23
difference between day and night activity levels. They show a more biphasic activity pattern
24
(12-hour cycle) as compared to the largely monophasic activity pattern (24-hour cycle) of
25
bears in group 1 and group 3 (Fig. 5). However, both groups had peak activities in the early
14
Kaczensky et al.
1
morning, a depression around noon, a second peak in the early evening and another
2
depression around midnight. However, contrary to bears in group 1 and 3 minimal activity
3
levels of bears in group 2 are higher during the day than during the night (Fig. 5). One bear
4
from group 2 (females LUCIA) did not follow the general pattern and showed a peak at noon
5
(Fig. 5). The two adult males in group 3, like bears in group 1, had a distinct depression in
6
their activity level during the day. However, contrary to bears in group 1, the activity
7
gradually sloped down from peak activity levels in the morning to a low at 16:00 and than
8
steeply rose again to high activity levels during the night.
9
10
Stationary versus travelling activity
11
12
The difference in the activity patterns between adult and younger bears was even more
13
distinct when comparing the 24-hours distribution of travelling activity. While three adult
14
females hardly travelled at all between 8:00 and 17:00, three yearlings frequently travelled
15
during daytime (Fig. 6). One female yearling (VERA) even showed a peak of travelling
16
activity around noon. Of the subadults, one male (SRECKO) showed a similar pattern than the
17
adult females, whereas a female (LUCIA) almost exclusively travelled during the day, with a
18
small depression around noon. This pattern of almost no activity during the day is much less
19
distinct in adult bears, when looking at stationary activity (activity without displacement). The
20
adult female MAJA, for reasons unknown, even showed a relative high level of stationary
21
activity during the day, similar to the pattern observed in the yearlings (Fig. 6).
22
Although mean travelling activity levels were lowest for adults during the day (adults 9%,
23
subadults 20%, yearlings 22%) and highest during the night (adults 27%, subadults 25%,
24
Yearlings 18%) differences were not significant (ANOVA Pday=0.54, Pnight=0.12), nor were
25
differences in mean stationary activity levels (day: yearlings 41%, subadults 35%, adults
15
Kaczensky et al.
1
21%, ANOVA P=0.21; night: yearlings 37%, subadults 35%, adults 31%, ANOVA P=0.66).
2
However sample sizes in each age group were very small.
3
4
Data from bears with less than 200 activity samples
5
6
Twenty-four-hour activity patterns of an additional 13 bears pooled by age classes, also
7
produced the expected difference between the age classes, that is low activity levels for adults
8
and peaks of high activity for yearlings throughout the day (Fig. 7). Again, subadults were
9
somewhat in between adults and yearlings.
10
Whereas at night all age groups had significantly different activity levels, during the day
11
only adults and yearlings differed significantly (day: adults 43%, subadults 48%, yearlings
12
60%, ANOVA P<0.01, Padult-yearling=0.05, Padult-subadult=0.85, Psubadult-yearling=0.10; night: adults
13
81%, subadults 60%, yearlings 38%, ANOVA P=0.02, Padult-yearling<0.01, Padult-subadult=0.08,
14
Psubadult-yearling=0.02).
15
16
DISCUSSION
17
18
Diurnal young bears and nocturnal adult bears
19
20
Our results showed a clear difference in the activity patterns of yearlings and adult bears,
21
with subadults being somewhat in between. The general pattern was that adults are mainly
22
nocturnal, whereas yearlings could be found active at any time. Individual variation was quite
23
large and not all bears followed this general pattern. Our results support the assumption that
24
the nocturnal activity pattern observed in European brown bears is more likely the result of
25
individual learning than a genetically fixed trait.
16
Kaczensky et al.
1
Difference between adults and yearlings were even more pronounced when comparing
2
travelling activity only. Adults hardly travelled at all during daytime, whereas yearlings and
3
some subadults could also be found travelling during the day. Consequently their chances to
4
encounter people were much higher. In the few cases (<20 occasions) where we saw bears, it
5
was either a yearling or the subadult female LUCIA. However, contrary to the yearlings,
6
LUCIA did not seem to be afraid of people and appeared to be habituated to human presence.
7
This data corresponds well with anecdotal observations by hunters and foresters in Slovenia
8
and Croatia, who also claim to mainly see small bears (D. Huber & P. Kaczensky, unpubl.
9
Data).
10
For one adult female (MAJA) with a relatively high daytime activity level, this activity was
11
largely confined to stationary activity that is activity in or around the daybed area. As daybeds
12
were mainly located in inaccessible areas (high cover or steep slopes, Kusak & Huber, 1998;
13
Kaczensky, 2000b) the chances for bear-humans encounters were minimal. However, as can
14
be expected when dealing with individual learning and behaviour that is believed to be shaped
15
by individual experience, not all bears strictly followed the same pattern. One adult bear (the
16
male HANS) showed a high level of daytime activity. Unfortunately we did not have data to
17
distinguish between stationary and travelling activity, nor did we have any information about
18
this experience with humans to better explain this behaviour.
19
The fact that subadult bears are more active during the day and expose themselves more
20
frequently to humans as compared to adults is frequently described also from North America
21
(Ison, Gilbert & Squibb, 1997). It is often assumed that subadult bears are generally less
22
affected by humans because of greater habituation (MacHutchon et al., 1998). But other
23
authors noted that subadults do not show the same tolerance than habituated adults (Braaten,
24
1988 cited in Olson, Squibb & Gilbert, 1998). The lower wariness is often explained with
25
food competition or the avoidance of aggressive adult bears. In our study areas yearling bears
17
Kaczensky et al.
1
did not actually shift from a nocturnal to a diurnal activity pattern, but rather showed equally
2
high activity levels during day and night and thus did not really avoid the time most adult
3
bears are active. Contrary to other regions (e.g. for Sweden see Swenson et al., 1997), we had
4
no evidence of interspecific killings, possibly because young bears grow rapidly and yearling
5
body masses in early spring often exceeded 55 kg (Kaczensky et al., 2002).
6
We speculate that young bears have a rather uniform activity pattern, with activity bouts
7
throughout the day and night, which through individual negative experience is changed into a
8
predominately nocturnal pattern in adults. Anecdotal evidence from Austria and Croatia
9
suggests that adult females are more diurnal in years with cubs than in years without cubs
10
(Rauer & Gutleb, 1997; Rauer, Kaczensky & Knauer, 2003, D. Huber, unpubl. Data).
11
Possibly this is the result of nocturnal adult females having to cope with relatively high
12
diurnal cub activity levels. Cubs of a wary, nocturnal mother probably have few encounters
13
with humans and thus lack negative experience with people. After family break-up at 1 ½
14
years, young bears (yearlings) are no longer forced into a nocturnal activity pattern, but rather
15
will be equally active during day and night.
16
Diurnal activity might allow young bears to access food bonanzas, like high productivity
17
berry patches or bait sites, which may be monopolized by large adults at night. This would be
18
in accordance with the fact that it is mainly small bears that are observed at bait sites during
19
daytime hours in Croatia and Slovenia (D. Huber & P. Kaczensky, unpubl. Data). Klinka &
20
Reimchen (2002) also speculate that the higher daytime foraging activity of females with cubs
21
and subadult bears on spawning salmon on Knight Inlet, coastal British Columbia was mainly
22
due to avoid large nocturnal males.
23
Perhaps young bears first consider other bears more dangerous than humans. However,
24
their high daytime activity will result in frequent encounters with humans and will in
25
combination with the fast gain in body mass (other bears are not a threat any more) lead to a
18
Kaczensky et al.
1
shift in the activity pattern in order to avoid people. That this is a process of individual
2
learning is also suggested by the fact that bears can become habituation to humans when a
3
negative stimulus is missing or even food conditioning when “not running from people” is
4
positively enforced by food (Rauer, Kaczensky & Knauer, 2003).
5
Because we were not able to measure human activity or the degree of disturbance an
6
individual bear was exposed to, we are unable to prove the causal relationship of nocturnal
7
behaviour (used as a surrogate to avoiding humans) and age (used as a surrogate of
8
experience). All bear populations in Europe are exposed to humans and in addition, the
9
observed behaviour might well be a result of past experiences that would be impossible to
10
assess. Even a bear that lives in a remote area might have dispersed from an area of high
11
human impact, as especially males are known to disperse long distances (Taberlet et al., 1994;
12
Knauer, 2000; Knauer, Kaczensky & Rauer, 2000). However, all bears live in landscapes
13
inhabited by people and in all areas hunting occurs, consequently we assume that all bears
14
were exposed to some negative experiences with humans.
15
We believe that hunting is one important negative stimulus to keep bears shy and
16
nocturnal. Many game animals have adopted a nocturnal behaviour due to human persecution
17
(Georgii & Schröder, 1983) and readily switch back to a diurnal activity when protected (e.g.
18
Kitchen, Gese & Schauster, 2000), also suggesting a high degree of individual learning. The
19
first major selection against diurnal behaviour in bears of the Dinaric Mountains takes place
20
during the fall hunting season (Frković et al., 1987). By this time most yearlings have masses
21
around 70-100 kg (Kaczensky, 2000b) and are considered reasonable trophies. There is a
22
quota system, but within a hunting unit there are often several hunters that are interested in
23
shooting a bear. Often different hunters have to take turns at a bait site and thus seize the first
24
opportunity to shoot a bear. The earlier a bear comes to a feeding site, the higher the chances
25
are to get shot. The radio collared female bear LUCIA was the only bear we monitored in
19
Kaczensky et al.
1
Slovenia that was active during the day and did not show fear of people. She was frequently
2
seen by hunters at bait sites, but was spared because of her radio collar.
3
However, removing a diurnal bears does not have much to do with individual learning at
4
least not for the bear shot dead. But bears are missed by bullets or get wounded: a collared
5
bear shot in Croatia in 1997 had old fractures caused by bullets in the jaw and pelvis, and the
6
skeletons of two other bears had old bullets imbedded in bones (D. Huber, unpubl. Data). In
7
addition, bears often come to established bait sites in groups: siblings, several males courting
8
the same female in oestrous, or females with cubs (D. Huber & P. Kaczensky, unpubl. Data).
9
Normally only one bear gets shot at any given bait site thus providing a negative stimulus to
10
the spared bears. Even if bears are not the target of a hunt and do not get physically harmed,
11
they are often accidentally chased by hunters and their dogs during drive hunts for other
12
game. In North America, it has been shown that black bears (Ursus americanus) become
13
more nocturnal during the hound training season, possibly in an attempt to avoid people and
14
their dogs (Bridges, Vaughan & Klenzendorf, 2004).
15
In Austria, where bears are strictly protected, several young bears started to be active
16
during the day and did not show much fear of people. They were frequently observed by
17
people with no negative consequences; on the contrary, some might have even been fed by
18
people to allow for photographs and video sequences (Rauer & Gutleb, 1997; Rauer,
19
Kaczensky & Knauer, 2003). At least three of these bears were females and two had already
20
raised cubs. Especially when accompanied by cubs these females were often observed at close
21
range and cubs never learnt to avoid people, predisposing them to come into trouble.
22
Counteracting this process is a major concern of the Austrian bear conservation program and
23
several attempts were made to aversively condition these bears (Rauer, Kaczensky & Knauer,
24
2003; Zedrosser, Gerstl & Rauer, 1999). One of the pre-requisites for aversive conditioning is
25
that behaviour can be changed through a negative stimulus. Because in Europe shy behaviour
20
Kaczensky et al.
1
is closely linked to a nocturnal activity pattern, aversive conditioning can only be successful if
2
the activity pattern can be changed through individual learning. Based on the results of this
3
study, this seems highly likely.
4
We believe that maintaining nocturnal behaviour in bears is one important prerequisite for
5
the coexistence of brown bears and people in the multi-use landscapes of Europe. For this
6
coexistence it is essential to separate bears and people, either in space or time. As it is almost
7
impossible to restrict human access into bear habitat and because a viable bear population
8
needs a huge area, a separation in time reduces the probability of encounters between bears
9
and humans. We are well aware that this will not solve all problems. Access to human
10
provided food and livestock needs to be minimized and people need to be educated not to feed
11
bears intentionally or accidentally. Especially in small protected bear populations this is a
12
difficult task. People have lost the experience of living with bears and a bear that becomes
13
visible will either be perceived as a threat to live and property or as a sensation and will
14
attracts heaps of people. Unfortunately the most likely outcome for both scenarios is a dead
15
bear. How much and what negative feedback is necessary to maintain shy, nocturnal
16
behaviour in protected population is unknown and should be the focus of further studies.
17
18
ACKNOWLEDGEMENTS
19
20
We would like to thank Hartmut Gossow, Blaž Krže, Wolfgang Schröder and Miha
21
Adamič for scientific, organizational and logistic support for the Slovenian part of this field
22
project. Field work for activity monitoring was very time consuming and especially tiring at
23
night. Without the help of employees, students and volunteers it would not have been possible
24
to collect that many data. Special thanks for data collection in Slovenia go to: Mateja Blazic,
25
Matjaz Prosen, Evi Tschunko, Thomas Speierl, Christian Leitenberger, Ilka Reinhardt, Ferry
26
Pickavet, Jennifer Clarke, Janez Adamič, Antonio Di Croce, Alessia Gallastroni, Chiara
21
Kaczensky et al.
1
Braschi, Susanne Falkenstett, Marcus Regelmann, Sandra Heyer, Philipp Ferstere and Sonja
2
Sinnmayer and for data collection in Croatia to: numerous veterinary students from the
3
University of Zagreb including Sanja Morić, Violeta Dabanović, and Andrea Poljak. Special
4
thanks also to Helmut Küchenhoff from the Statistical Advice Office at the University of
5
Munich and to his students Regina Schwarz, Brigitte Dorsch and Christina Bindl for help with
6
the statistical analysis.
7
Funding for this study came from the Austrian Science Foundation (FWF project: P 11529-
8
BIO), the Austrian Federal Ministry of Science and Research (project: G.Z. 30.435/-23/92),
9
the Croatian Ministry of Science and Technology (Project 053026), the Donors Association
10
for the Promotion of Science and Humanities in Germany, the Forest Faculty at the University
11
of Munich in Germany, The National Geographic Society, EURONATURE, the Slovenian
12
Hunters Association and the Bevins Memorial Fundation of the International Bear
13
Association.
14
15
LITERATURE CITED
16
17
Adamič, M. (1996). Decision making in the conservation management of problem wildlife
18
species in Slovenia, based on the knowledge gained through current research projects.
19
Gozd 54:297-306. [In Slovenian with English abstract]
20
Bridges, A.S., Vaughan, M.R. & Klenzendorf, S. (2004). Seasonal activity in American black
21
bear Ursus amercanus activity patterns: qualification via remote photography. Wildlife
22
Biology 10:277-284.
23
Clevenger, A. P., Purroy, F. J. & Pelton M. R. (1990). Movement and activity patterns of a
24
European brown bear in the Cantabrian Mountains, Spain. International Conference on
25
Bear Research and Management 8:205-211.
22
Kaczensky et al.
1
Dečak, D., Frković, A. Grubešić, M., Huber, D., Iviček, B., Kulić, B., Sertić, D. & Štahan, Ž.
2
(2004). The Brown Bear Management Plan for The Republic of Croatia. Ministry of
3
Agriculture, Forestry and Water Management, Department for Hunting Ministry of
4
Culture, Department for Nature Protection, Zagreb, Croatia.
5
6
7
Enright, J.T. (1970). Ecological aspects of endogenous rhythmicity. Annual Reviews 1:221238.
EUROPARC & IUCN. (2000). Guidelines for protected area management categories -
8
interpretation and application of the protected area management categories in Europe.
9
EUROPARC and WCPA, Grafenau, Germany.
10
Frković, A., Ruff, R. L., Cicnjak, L. & Huber D. (1987). Brown bear mortality during 1946-
11
85 in Gorski Kotar, Yugoslavia. International Conference on Bear Research and
12
Management 7:87-92.
13
14
15
16
Geffen, E., & MacDonald, D. W. (1993). Activity and movement patterns of Blanford´s
foxes. Journal of Mammalogy 74:455-463.
Georgii, B. & Schröder. W. (1983). Home range and activity patterns of male red deer
(Cervus elaphus L.) in the Alps. Oecologia 58:238-248.
17
Gunther, K. A. (1990). Visitor impact on grizzly bear activity in Pelican valley, Yellowstone
18
National Park. International Conference on Bear Research and Management 8:73-78.
19
Halle, S., & Stenseth, N. C. (1994). Mictotine ultradian rhythm of activity: an evaluation of
20
21
22
23
24
different hypotheses on the triggering mechanism. Mammal Review 24:17-39.
Huber, D. & Frkovic, A. (1993). Brown bear management in Croatia, IUGB Congress. 21:
287-292, Halifax.
Huber, D., & Roth, H. (1993). Movements of European brown bears in Croatia. Acta
Theriologica 38:151-159.
23
Kaczensky et al.
1
2
3
4
Huber. D., Kusak, J. & Radisić, B. (1996). Analysis of efficiency in live-capture of European
brown bears. Journal of Wildlife Research 1:162-166.
Ison, T. L., Gilbert, B. K. & Squibb, R. C. (1997). The effects of increasing human activity on
brown bear use of an Alaskan River. Biological Conservation 82:95-99.
5
Kaczensky, P. (2000a). Bear politics in Slovenia. International Bear News 9(2):9.
6
Kaczensky, P. (2000b). Co-existence of brown bears and men in Slovenia. PhD thesis
7
8
9
Technische Universität München, Munich, Germany.
Kaczensky, P., Wagner, A. & Walzer, C. (2004). Activity monitoring of a brown bear - a
model approach to test field methods. Mammalian Biology 69(3):1-5.
10
Kaczensky, P., Knauer, F. Jonozovic, M. Walzer, C. and Huber, T. (2002). Experiences with
11
trapping, chemical immobilization, and radiotagging of brown bears in Slovenia. Ursus
12
13:347-356.
13
14
15
Kaczensky, P., Blazic, M. & Gossow, H. (2004). Public attitude towards brown bears (Ursus
arctos) in Slovenia. Biological Conservation 118:661-674.
Klinka, D. R. & Reimchen T.E. (2002). Nocturnal and diurnal foraging behaviour of brown
16
bears (Ursus arctos) on a salmon stream in coastal British Columbia. Canadian Journal of
17
Zoology 80:1317-1322.
18
19
20
Kitchen, A. M., Gese, E. M. & Schauster, E. R. (2000). Changes in coyote activity patterns
due to reduced exposure to human persecution. Canadian Journal of Zoology 8:853-857.
Knauer, F. (2000). Dispersal und Ausbreitung von Braunbären in die Ostalpen. PhD thesis
21
Technische Universität München, Munich, Germany. [Dispersal and expansion model for
22
brown bears in the eastern Alps; In German]
23
24
Knauer, F., Kaczensky, P. & Rauer, G. (2000). A habitat model for bear expansion from
Slovenia into the eastern Alps. In Co-existence of brown bears and men in the cultural
24
Kaczensky et al.
1
landscape of Slovenia. Kaczensky, P. (Ed.). Unpublished report of the Institute of Wildlife
2
Biology and Game Management at the Agricultural University of Vienna, Austria.
3
4
5
Kusak, J., & Huber, D. (1998). Brown bear habitat quality in Gorski Kotar, Croatia. Ursus
10:281-291.
Linnell, J. D. C., D. Steuer, J. Odden, P. Kaczensky, and J. E. Swenson. 2002. European
6
brown bear compendium. Safari Club International Foundation: Wildlife Conservation
7
Issues-Technical Series, 004A. Available from:
8
http://www.lcie.org/Publicaciones/BB/European%20Brown%20Bear%20Compendium.pdf
9
[last accessed August 2005]
10
11
12
13
14
15
Liddle, M. (1997). A classification of disturbance to animals. In Recreation ecology: 347-442.
Liddle, M. (Ed.). Chapman and Hall, University Press, Cambridge, Great Britain.
MacHutchon, A. G., Himmer, S. Davis, H. & Gallagher, M. (1998). Temporal and spatial
activity patterns among coastal bear populations. Ursus 10:539-546.
Mattson, D. J. 1990. Human impacts on bear habitat use. International Conference on Bear
Research and Management 8:33-56.
16
Metz, C. 1990. Der Bär in Graubünden. Desertina Verlag, Disentis, Switzerland. [The brown
17
bear in the Kanton of Graubünden, In German]
18
Nielsen, E. T. (1983). Relation of behavioural activity rhythms to the changes of day and
19
20
21
22
23
night. A revision of views. Behavior 89:147-173.
Olson, T. L., Squibb, R. C. & Gilbert, B. K. (1998). Brown bear diurnal activity and human
use: a comparison of two salmon streams. Ursus 10:547-555.
Rauer, G., & Gutleb, B. (1997). Der Braunbär in Österreich. Umweltbundesamt,
Monographie 88:64pp. [in German]
24
Rauer, G., Kaczensky, P. & Knauer, F. (2003). Experiences with aversive conditioning of
25
habituated brown bears in Austria and other European countries. Ursus 14(2):215-224.
25
Kaczensky et al.
1
2
3
Roth, H. (1980). Diel activity of a remnant population of European brown bears. International
Conference on Bear Research and Management 5:223-229.
Roth, H., & Huber, D. (1986). Diel activity of brown bears in Plitvice Lakes National Park,
4
Yugoslavia . International Conference on Bear Research and Management 6:177-181.
5
Servheen, C., Herrero, S. & Peyton, B. (1998). Conservation action plan for the world bears.
6
7
IUCN, Gland, Switzerland.
Simonic, A. (1994). The legal protection of the brown bear in Slovene territory-past and
8
present, and some suggestions for the future. In Braunbär in den Ländern Alpen-Adria: 43-
9
75. Proceedings of a workshop, Ljubljana, Slovenia.
10
11
12
13
Smith, T. S., Herrero, S. & DeBruyn, T. (2005). Alaskan brown bears, humans, and
habituation. Ursus 16: 1-10.
Stoneberg R. P. & Jonkel C. J. (1966). Age determination of black bears by cementum layers.
Journal of Wildlife Management 30: 411-414.
14
Swenson, J. E., Sandegren, F. Heim, M. Brunberg, S. Sorensen, O. J. Söderberg, A. Bjärvall,
15
A. Franzen, R. Wikan, S. Wabakken, P. & Overskaug, K. (1996). Er den skandinaviske
16
bjornen farlig? NINA Oppdragsmelding 404:1-26. [in Swedish]
17
18
19
20
21
Swenson, J. E. 1999. Does hunting affect the behavior of brown bears in Eurasia? Ursus
11:157-162.
Swenson, J. E., Sandegren, F. Söderberg, A. Bjärvall, A. Franzen, R. & Wabakken, P. (1997).
Infanticide caused by hunting of male bears. Nature 386:450-451.
Swenson, J. E., Sandegren, F. Söderberg, A, Heim, M. Sorensen, O. J. Bjärvall, A. Franzen,
22
R. Wikan, S. & Wabakken, P. (1999). Interactions between brown bears and humans in
23
Scandinavia. Biosphere Conservation 2:1-9.
24
Swenson, J. E., Gerstl, N. Dahle, B. & Zedrosser, A. (2000). Action Plan for the conservation
25
of the brown bear in Europe. Convention on the Conservation of European Wildlife and
26
Kaczensky et al.
1
Natural Habitats (Bern Convention) , Nature and Environment 114, Council of Europe
2
Publishing, Strasbourg, France.
3
Taberlet, P., Swenson, J. E. Sandegren, F. & Bjärvall, A. (1994). Localization of a contact
4
zone between two highly divergent mitochondrial DNA lineages of the brown bear Ursus
5
arctos in Scandinavia. Conservation Biology 9:1255-1261.
6
Tyler, S. (1979). Time-sampling: a matter of convention. Animals Behavior 27:801-810.
7
Wabakken, P. & Maartmann, E. (1994). Sluttrapport for bjorn - sauprosjektet i Hedmark
8
1990-93. NINA forskningsrapport 58: 41pp. [in Norwegian with English abstract]
9
10
11
12
13
14
Zar, J. H. (1999). Biostatistical analysis. Prentice-Hall International, Inc., Simon &
Schuster/A Viacom Company, Upper Saddle River, New Jersey, USA.
Zedrosser, A., Gerstl, N. & Rauer, G. (1999). Brown bears in Austria. Federal Environment
Agency, Volume M-117:42pp, Vienna, Austria.
Zedrosser, A., Dahle, B. Swenson, J.E. & Gerstl, N. (2001). Status and Management of the
brown bear in Europe. Ursus 12:9-20.
15
27
Kaczensky et al.
1
Table 1: Dataset used for individual analysis of brown bear activity in Menisija (MN),
2
Slovenia, and Plitvice Lakes (PL) and Gorski Kotar (GK), Croatia.
Bear
Age
Sex
Activity monitoring
Activity
Total
Study
Start
readings
hours1
Area2
End
Adults:
FRANJO
5
male
21.06.1982 09.10.1982
268
6
male
14.05.1983 04.10.1983
254
7
male
20.05.1984 09.09.1984
713
8
male
26.05.1985 07.09.1985
6
HAL
6
male
16.05.1983 19.09.1983
260
HANS
5
male
18.05.1985 28.10.1985
422
7
male
10.05.1987 05.09.1987
209
DADO
4
male
16.05.1986 06.10.1986
197
GABI
5
female
14.06.1986 12.10.1986
512
6
female
21.05.1987 19.09.1987
136
9
female
01.05.1997 19.06.1997
600
103
MN
adult
female
14.05.1998 09.10.1998
1,982
338
MN
5
female
15.05.1998 24.10.1998
2,259
385
MN
LILI83
2
female
09.05.1983 30.07.1983
500
108
PL
INGA
3
female
27.05.1987 10.09.1987
220
43
GK
LUCIA
2
female
04.05.1997 08.10.1997
3,334
564
MN
SRECKO
3
male
04.05.1997 09.10.1997
2,634
448
MN
(LILI822
1
female
01.05.1982 17.10.1982
798
144
PL)
PEPI
1
male
10.07.1987 19.09.1987
414
91
GK
VANJA
1
female
01.05.1997 11.08.1997
1,615
277
MN
VERA
1
female
07.05.1997 28.05.1997
590
97
MN
DUSAN
1
male
10.05.1997 09.07.1997
718
125
MN
18,641
3,372
MAJA
ANCKA
POLONA
PL
285
PL
PL
49
135
45
135
PL
PL
PL
GK
Subadults:
Yearlings:
total
3
1
Monitoring hours with ≥3 activity samples
4
2
Was used to compare with data from LILI83, but not in multi-bear comparisons
28
Kaczensky et al.
1
Table 2: Dataset of bears with less than 200 activity samples, used for analysis pooled by age
2
class.
Area1
Activity monitoring
Bear
Age Sex
Start
End
Activity Surveyed
readings
Adults:
hours
309
51
VIOLETA
10 female
10.07. 1987 05.09. 1987
87
18
GK
NIVA
13 female
02.05. 1990 13.10. 1990
71
11
GK
BOB
5
male
16.05. 1986 29.08. 1986
6
1
PL
VLADO
5
male
16.05. 1986 17.05. 1986
5
1
PL
NENO
12 male
30.07. 1986 19.10. 1986
140
20
GK
380
60
Subadults:
JURICA
2
female
03.05. 1985 31.05. 1985
160
24
PL
JURA
3
male
18.05. 1983 30.07. 1983
28
6
PL
GORAN
3
male
06.07. 1986 17.09. 1986
49
6
GK
FRKO
3
male
27.05. 1987 03.08. 1987
143
24
GK
446
59
Yearlings:
LINDA
1
female
07.07. 1990 12.12. 1990
69
9
GK
DARKO
1
male
03.05. 1985 11.09. 1985
152
24
PL
MIKI
1
male
11.06. 1990 12.10. 1990
84
7
GK
NEJC
1
male
12.06. 1998 14.07. 1998
141
19
MN
1,135
170
total
3
1
MN=Menisija (SLO), GK=Gorski Kotar (HR), PL=Plitvice Lakes (HR)
4
2
Hours monitored with ≥3 activity samples
5
6
29
Kaczensky et al.
1
Table. 3: Effects of different variables on the activity pattern of bears using a general linear
2
model (GLM) with seasonal component (12-hour and 24-hour rhythm). Most effect on activity
3
(according to the sum of squares) has the activity of the hour before (Act_1), followed by the
4
periodic activity rhythms (12 and 24-hour rhythm) and interactions of them with age and to a
5
lesser degree with sex. Age alone has only a slight effect and sex alone has no significant
6
effect. Note, that age and sex affect more the pattern of the daily activity, than the level.
Sum of
Mean
Parameter
square
df
square
F
P value
Corrected model
188.19
18
10.46
116.26
<0.01
51.08
1
51.08
568.04
<0.01
Age
1.35
2
0.67
7.48
<0.01
Sex
0.05
1
0.05
0.58
0.45
82.18
1
82.18
913.81
<0.01
24-hour rhythm
5.17
2
57.46
<0.01
12-hour rhythm
5.63
2
62.63
<0.01
Age * 24-hour rhythm
4.77
4
26.51
<0.01
Age * 12-hour rhythm
0.75
4
4.18
<0.05
Sex * 24-hour rhythm
2.34
2
25.97
<0.01
Error
227.79
2533
Corrected total variation
415.99
2551
Constant
Act_1
0.09
R² = 0.452
7
8
30
Kaczensky et al.
1
Fig.1: Location of the three study areas in Slovenia and Croatia (MN = Menisija, GK =
2
Gorski Kotar, PL = Plitvice Lakes).
3
4
Fig. 2: Activity level during day- (white) and night time (gray) hours for adult, subadult and
5
yearling bears.
6
7
Fig. 3: Activity patterns of one bear that were monitored during different seasons and
8
changed from the yearling to the subadult age class.
9
10
Fig. 4: Dendrogram using Average Linkage (Euclidean distances between Groups) of the
11
cluster analysis comparing diel activity patterns between individual bears. Age composition of
12
group 1: 5 adults, 2 subadults and 1 yearling, group 2: 1 adult, 1 subadult and 3 yearlings, and
13
group 3: 2 adults.
14
15
Fig. 5: Activity per hour of all bears within the 3 groups identified by cluster analysis.
16
17
Fig. 6: Travelling and stationary activity per hour for 8 bears in three age groups. These data
18
were only available for the Slovenian study area.
19
20
Fig. 7: Activity per hour for an additional 13 bears pooled by age class.
31
Kaczensky et al.
1
32
Adults
Subadults
G
A8
7
LI
83
PE
PI
SA
N
VE
R
A
JA
0.90
D
U
VA
N
C
IA
LU
IN
LI
SR
EC
KO
KA
D
AD
O
G
AB
I
M
AJ
A
H
AN
S
AN
C
H
AL
AN
J
PO O
LO
N
A
1
FR
Day- and nighttime activity levels
Kaczensky et al.
1.00
night
day
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
Yearlings
33
Kaczensky et al.
1
Bear
bear:
lili82
LILI82
LILI83
lili83
1.00
1,00
Activity level
activity level
0.80
0,80
0.60
0,60
0.40
0,40
0.20
0,20
0,00
0.00
22
20
18
16
14
12
10
8
6
4
2
0
0
2 4
6 8 10 12 14 16 18 20 22
hours
Hour
34
Kaczensky et al.
Rescaled Distance Cluster Combine
1
CASE
Label
Num
FRANJO
SRECKO
POLONA
ANCKA
GABI
MAJA
LILI83
VANJA
INGA
LUCIA
PEPI
HANS
DUSAN
VERA
DADO
HAL
4
14
13
1
5
11
9
15
8
10
12
7
3
16
2
6
0
5
10
15
20
25
35
Kaczensky et al.
1
1.00
1.00
Group 1
0.90
0.90
0.80
0.80
0.60
0.60
activity level
Activity level
0.70
0.70
0.50
0.50
0.40
0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0.00
1.00
1.00
1
2
3
4
5
6
7
8
9
10
11
12
13
14
0.90
0.90
15
16
17
18
19
20
21
22
23
24
17
18
19
20
21
22
23
Group 2
hour
0.80
0.80
0.60
0.60
activity level
Activity level
0.70
0.70
0.50
0.50
0.40
0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0.00
1.00
1.00
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
hour
Group 3
0.90
0.90
0.80
0.80
0.60
0.60
activity level
Activity level
0.70
0.70
0.50
0.50
0.40
0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0.00
0
0
11
22 33 44
55 66
77
88
9
9
10 11
11 12
12 13
13 14
14 15
15 16
16 17
17 18
19 20
20 21
21 22
23
10
18 19
22 23
hour
Hour
36
Kaczensky et al.
Stationary activity
1
Traveling activity
Adults:
1.00
ANCKA
MAJA
POLONA
Activity level
0.80
0.60
0.40
0.20
subadult
1,0
0,8
0.80
0,8
activity level
0,6
0.60
0,4
0.40
0,0
0.00
0,0
lucia
LUCIA
srecko
SRECKO
22
20
18
16
14
12
10
8
6
4
2
0
22
20
18
16
14
12
10
8
6
4
2
0
0.80
Subadults:
0,4
0,2
hour
lucia
srecko
0,6
0,2
0.20
1.00
Activity level
subadults
1,0
1.00
activity level
Activity level
0.00
hour
Yearlings:
DUSAN
VANJA
VERA
0.60
0.40
0.20
0.00
37
Kaczensky et al.
1
1.00
2
adults
subadults
yearlings
0.90
0.80
Activity
level
activity level
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
hours
Hour
38
Download