Content uploaded by Jörn Theuerkauf
Author content
All content in this area was uploaded by Jörn Theuerkauf on Oct 23, 2017
Content may be subject to copyright.
What Drives Wolves: Fear or Hunger? Humans, Diet, Climate
and Wolf Activity Patterns
Jo
¨rn Theuerkauf
Museum and Institute of Zoology, Polish Academy of Sciences, Warsaw, Poland
Introduction
Optimisation models predict that the evolutionary
process favours characteristics that maximise the fit-
ness of an individual (Krebs & McCleery 1984). An
optimal behaviour should therefore follow the eco-
nomic principle: maximising benefits to the animals
(e.g. the exploitation of food resources) while mini-
mising costs, such as the risk of predation (Lima &
Dill 1990). Man has long been the most important
predator of large carnivores. It is therefore likely that
centuries of persecution have favoured wolves
(Canis lupus) that avoided humans (Linnell et al.
2002). In fact, wolves seem to fear humans in most
regions with the exception of high arctic regions
where they have little or no contact with man
(Mech 1988).
Wolves can adopt two different strategies to mini-
mise contact with humans: temporal or spatial
avoidance. The latter has been shown by previous
studies (Thurber et al. 1994; Theuerkauf et al.
2003a,b, 2007; Kaartinen et al. 2005; Chavez & Gese
2006; Habib & Kumar 2007), although spatial avoid-
ance might be correlated with wolves selecting habi-
tats that are far from human settlements (e.g. large
forest tracks, mountains). Temporal avoidance of
humans is more difficult to prove because many fac-
tors may affect the time budget of wolves. Temporal
avoidance is nonetheless suggested as the reason
why wolves in Italy (Ciucci et al. 1997) and Spain
(Vila
`et al. 1995) are nocturnal. However, there is
some evidence that wolves do not change their daily
activity patterns to avoid humans: the activity pat-
terns of wolves in northeast Poland under different
Correspondence
Jo
¨rn Theuerkauf, 45 rue Maurice Herzog,
98800 Noume
´a, New Caledonia.
E-mail: jtheuer@miiz.waw.pl
Received: December 18, 2008
Initial acceptance: February 13, 2009
Final acceptance: March 1, 2009
(J. Schneider)
doi: 10.1111/j.1439-0310.2009.01653.x
Abstract
Activity patterns of animals depend on environmental and intrinsic fac-
tors. Studies undertaken across the current wolf (Canis lupus) range sug-
gested a number of variables that may be correlated with activity
patterns of wolves. These factors vary locally and there has been no
attempt so far at defining those that ubiquitously impact wolf behav-
iour. I compared 11 studies (from Alaska to Israel) to assess the influ-
ence of (1) public road density, (2) human population density, (3)
human-caused mortality, (4) proportion of domestic animals in wolf
diet, (5) proportion of forest, (6) latitude and (7) mean annual tempera-
ture on nocturnal wolf activity and movements. Nocturnal activity was
mainly correlated to the proportion of domestic animals in the diet and
the density of public roads, whereas nocturnal movements were mainly
correlated to latitude. The importance of latitude indicates that sun peri-
odicity might represent an important signal (‘Zeitgeber’) for circadian
rhythms in wolves. Environmental constraints such as high tempera-
tures during the day and a higher hunting success in crepuscular periods
probably limit the ability of wolves to avoid humans by nocturnal
behaviour. I therefore suggest that in regions where wolves hunt wild
prey, they experience a trade-off between predation risk by humans and
increased hunting success during twilight hours.
Ethology
Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH 649
levels of anthropogenic pressure were similar (The-
uerkauf et al. 2003c) and wolves in an area of high
human density in southeast Poland were not noctur-
nal (Theuerkauf et al. 2007).
Optimisation models predict that a species’ forag-
ing strategy is an important determinant of its activ-
ity rhythms and home range use. If wolves had
lower hunting success at night, then this would
probably affect their reproduction rates. On the
other hand, diurnal wolves might be more at risk
from human predation. Wolves should therefore
attempt to balance the conflicting strategies of maxi-
mising food intake and minimising encounters with
humans (Theuerkauf et al. 2003b). Spatio-temporal
avoidance of humans is obviously easier for wolves
living in forests as they can hide behind cover, but
more difficult for those that move and forage over
open ground, especially in daylight. Nocturnal activ-
ity might therefore be a strategy for wolves living in
open areas.
In most regions, wolves prefer to prey on large
ungulates such as red deer (Cervus elaphus), white-
tailed deer (Odocoileus virginianus), moose (Alces alces),
reindeer (Rangifer tarandus), roe deer (Capreolus capre-
olus) and wild boar (Sus scrofa) (e.g. Fritts & Mech
1981; Mattioli et al. 1995; Mech et al. 1995; Ballard
et al. 1997; Je˛drzejewski et al. 2002; Gula 2004).
Wolves living close to human populations also use
anthropogenic food sources such as domestic animals
and carrion (e.g. Meriggi et al. 1996; Mech et al.
2000; Hovens & Tungalaktuja 2005; Gula 2008) and
sometimes garbage (Meriggi & Lovari 1996).
Curio (1976) suggested that predators take on the
activity patterns of their prey. Activity patterns of
wolves feeding on wild prey might therefore be similar
to those of their main prey species. Red deer (Georgii &
Schro
¨der 1983; Kamler et al. 2007), roe deer
(Cederlund 1981; Jeppesen 1989), white-tailed deer
(Montgomery 1963; Kammermeyer & Marchinton
1977) and moose (Geist 1963) are all important wolf
prey, with usually bimodal activity patterns peaking
around sunrise and sunset. The activity patterns of
wolves preying on these ungulates would therefore be
predicted to be also bimodal.
There is a wide range of factors that might influ-
ence the behaviour of wolves, whether separately or
simultaneously. However, wolves occur over a broad
range of climates and conditions so the importance
of any given factor might vary with location. Vary-
ing local influences might explain why different fac-
tors have been suggested to be important in shaping
wolf activity patterns. If this is the case, studies
undertaken in a particular region would not reveal
the range of factors that can influence the activity
patterns of wolves. Core factors shaping the activity
and movements of wolves would be best determined
by comparing wolf behaviour over a wide range of
environmental conditions. To identify important fac-
tors shaping the activity patterns of wolves, I there-
fore reviewed and compared the results of 11 studies
undertaken in different geographic areas and
selected seven variables that were previously found,
or suggested, to influence wolf activity patterns.
Methods
I searched the wolf literature for studies on daily
activity or movement patterns of wolves. Eleven
studies provided sufficient data for inclusion in this
paper (Table 1). I excluded the study by Kolenosky
& Johnston (1967) because of insufficient informa-
tion on environmental factors and small sample size.
Based on tables or figures in these publications, I
estimated, for each study, the time active and the
distance travelled per hour of night (sunset to sun-
rise) and day (sunrise to sunset). Most studies used
changes in signal strength to estimate activity, some
used activity sensors, and each had a different radio-
tracking interval and duration. As the estimation of
time active varies with the method used to estimate
activity (Theuerkauf & Je˛drzejewski 2002) and the
estimation of distance travelled depends on the sam-
pling interval (Rouys et al. 2001), I standardised data
by calculating nocturnal activity (or movement) indi-
ces as the time active (or distance travelled) at night
divided by the total time active (or distance travelled)
during the whole day. As the mean annual length of
night is 12 h (amplitude being the only variable that
changes with latitude), using the proportion of time
active to estimate yearly nocturnal activity indices is
equivalent to using the duration of activity. Noctur-
nal activity indices represented the proportion of
time wolves were active (e.g. moving, feeding) at
night, thus an index of 0.5 would mean that half of
the activity took place at night and half in the day.
Nocturnal movement indices represented the dis-
tance they covered (e.g. marking tours, searching
for prey) at night, thus if wolves in an area had a
nocturnal activity index of 0.5 and a nocturnal
movement index of 0.6, this means that wolves were
as active at night than in the day but that they
covered relatively larger distances at night. If sunset
and sunrise was not given, I used mapsource 6.13.7
(Garmin, Olathe, KS, USA) to estimate the mean
time of sunrise and sunset for the period and location
of the respective study.
What Drives Wolves: Fear or Hunger? J. Theuerkauf
650 Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH
Three studies did not provide activity patterns
and three others no data on distance travelled. I
estimated nocturnal activity or movement indices
for these studies. Nocturnal activity indices are
lower than nocturnal movement indices because
wolves cover longer distances per hour at night
than during the day (Theuerkauf et al. 2003c). In
the five studies that provided both activity and
movement patterns, nocturnal activity indices were
90 8% (95% confidence interval) of nocturnal
movement indices. In the three studies that did not
provide activity data, I estimated the nocturnal
activity index as 90% of the nocturnal movement
index. In the three studies that did not provide
movement data, I estimated the nocturnal move-
ment index as 111% of the nocturnal activity
index. This might have introduced some circularity
in the analyses but excluding three of 11 studies
for the activity and movement analysis would have
biased the result towards the other studies and not
have allowed to assess if the seven factors affected
activity and movement patterns differently. Addi-
tionally, I classified the general pattern of activ-
ity ⁄movements as either bimodal (peaks in the
morning and evening), nocturnal (peak in the mid-
dle of the night) or diurnal (peak in the middle of
the day). A more detailed analysis of the deviation
from crepuscular bimodality was unfortunately not
possible because of the varying data presentation in
the different studies.
I considered seven variables that might influence
nocturnal behaviour of wolves in the analyses: (1)
the proportion of domestic animals or garbage in
the wolf diet (diet) which represents the degree to
which wolves depend on hunting wild prey, (2)
public paved road density in km ⁄km
2
(road) as a
measure of the intensity of human presence, (3)
human population density per km
2
(humans) as a
measure of the number of potential predators, (4)
proportion of land covered by forest or dense shrub
(forest) which represents the available cover from
human sight or the sun, (5) degree of latitude (lati-
tude) that influences the timing of sunrise and sun-
set as well as the intensity of sunshine, (6) mean
annual temperature in C in the study area (tem-
perature) which represents xand extreme tempera-
tures that might restrict wolf activity and (7) the
proportion of wolf mortality that was caused by
humans (mortality) as a measure of the evolution-
ary pressure to reduce predation risk by humans.
As the mean annual temperature is not indepen-
dent of latitude, I did not analyse both variables
simultaneously. The three human-related variables
Table 1: Data from the 11 studies on which the analyses in the current study were based, with information on latitude (latitude), mean annual temperature in C (temperature), human population
density per km
2
(humans), proportion of wolves that were killed by humans (mortality), public road density in km ⁄km
2
(road), proportion of land covered by forest or dense shrub (forest), and
proportion of domestic animals or garbage in the wolf diet (diet). Data in the last two columns were not included in the regression models
Region Latitude Temperature Humans Mortality Road Forest Diet
Nocturnal
activity index
Nocturnal
movement index
No. wolves, radio-tracking
interval and length of study General pattern
Alaska
a
67 )4 0 0.42 0.00 0.80 0.00 0.48 0.53* 23 wolves, variable interval, 4 yr Diurnal (winter), nocturnal (summer)
Finland
b
64 1 2 0.10 0.40 0.80 0.00 0.68 0.75* 1 wolf, 5-min interval, Mar.–Oct. Nocturnal (summer)
Norway
c
62 4 1 1.00 0.30 0.80 0.00 0.54* 0.60 2 wolves, 4-h and 15-min intervals, 1 yr Bimodal
NE Poland
d
53 7 7 0.50 0.10 0.97 0.005 0.54 0.60 11 wolves, 15-min interval, 5 yr Bimodal
SE Poland
e
50 5 44 0.33 0.64 0.62 0.08 0.53 0.59 3 wolves, 15-min interval, 4 yr Bimodal
NW Minnesota
f
48 5 2 0.44 1.00 0.20 0.15 0.59* 0.65 7 wolves, 45-min interval, 2 yr Nocturnal ⁄dawn active
W Minnesota
g
46 5 11 0.63 1.40 1.00 0.00 0.76* 0.84 9 wolves, 15-min to 3-h intervals, 2 yr Nocturnal
Croatia
h
44 13.5 22 0.50 0.49 0.37 0.70 0.60 0.66* 3 wolves, 15-min interval, 2 yr Nocturnal ⁄bimodal
Italy
i
42 8 26 1.00 0.50 0.48 1.00 0.82 0.85 1 wolf, 10-min interval, Jun.–Mar. Nocturnal
Spain
j
42 12 25 0.58 0.40 0.20 0.75 0.74 0.77 4 wolves, 30-min interval, 1 yr Bimodal
Israel
k
33 18 30 0.57 0.26 0.10 0.47 0.67 0.84 13 wolves, >3-h interval, 3 yr Bimodal
Sources:
a
Fancy & Ballard 1995; Ballard et al. 1997; W.B. Ballard, pers. comm.;
b
Kojola 2002; Kojola et al. 2004; Kaartinen et al. 2005;
c
Wabakken et al. 2001; Sand et al. 2005; Eriksen 2006;
d
Je˛drzejewski et al. 2002; Theuerkauf et al. 2003c;
e
Theuerkauf et al. 2007; Gula 2008;
f
Chavez & Gese 2005, 2006; A. Chavez, pers. comm. ;
g
Merrill 2000; Merrill & Mech 2003; L.D. Mech, pers.
comm.;
h
Kusak et al. 2005; J. Kusak, pers. comm.;
i
Ciucci et al. 1997; L. Boitani, pers. comm.;
j
Vila
`et al. 1995; Blanco & Corte
´s 2007;
k
Reichmann & Saltz 2005; A. Reichmann, pers. comm.
*Estimated as described in Methods.
J. Theuerkauf What Drives Wolves: Fear or Hunger?
Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH 651
(road, humans, mortality) were not correlated (all
r < 0.07, p > 0.8, post hoc power > 0.88, calculated
with G*Power 3.0.10 by Faul et al. 2007).
Most studies did not distinguish the activity pat-
terns of breeding and non-breeding individuals. I did
not include breeding as a variable in the analyses
because it is seasonal and likely to influence wolves
of different regions similarly (Vila
`et al. 1995; The-
uerkauf et al. 2003c; Schmidt et al. 2008; Tsunoda
et al. 2009). As only two studies provided environ-
mental information on all seven variables, I searched
for other studies in the respective region to provide
the information. When data were still lacking I con-
tacted authors directly for unpublished data.
As all data were continuous and normally distrib-
uted (Kolmogorov–Smirnov test, all p > 0.35), I used
multiple linear regression models (calculated with
spss for Windows 11.0) and AIC (Burnham & Ander-
son 2002) to assess the factors that most influence
nocturnal behaviour of wolves. I tested approx. 50
biological meaningful models and ranked them by
Akaike weights (Burnham & Anderson 2002). As
the sample size was 11 studies, I retained only the
10 highest ranking models.
Only two of the 11 study areas had a wolf hunting
season: in Alaska from Aug. to Apr., and in Finland
from Nov. to Mar. However, poaching or poisoning
was common in most places (or occasional culling).
I estimated mortality rates for most studies based on
the human-caused mortality of radio-tracked wolves.
The chance of a wolf being killed by humans during
radio-tracking is directly proportional to the length
of tracking. To exclude the possibility that biased
mortality rates influenced the results, I tested for a
relationship between mortality rates and either the
length of the study or the number of wolves studied.
Neither the length of the study (r = )0.324, n = 11,
p = 0.330, power = 0.724) nor the number of wolves
(r = )0.184, n = 11, p = 0.588, power = 0.786) were
correlated to the mortality rate.
Results
Increasing proportion of domestic animals in the
wolf diet (diet) and public road density (road) best
explained increased nocturnal activity in wolves
(Table 2). Diet was the highest ranking factor with a
sum of Akaike weights of 0.942, followed by road
with a sum of 0.853. All other factors had lower
sums than 0.3 (mortality 0.257, latitude 0.174,
humans 0.135, forest 0.073, temperature 0.073). Lat-
itude (sum of Akaike weights of 0.708) appeared to
be the main cause of nocturnal movements
(Table 3); the other variables had scores lower than
0.3 (road 0.275, diet 0.245, temperature 0.128,
humans 0.101, forest 0.101, mortality 0.075). Most
high ranking models included only 1–2 variables.
The correlations provided similar results; diet was
most highly correlated with nocturnal activity and
latitude with nocturnal movements (Fig. 1). For the
general activity pattern, five sites were bimodal
(Norway, NE Poland, SE Poland, Spain and Israel)
and only the site in Italy was clearly nocturnal
(Table 1). At the subarctic sites in Alaska and Fin-
land, wolves were diurnal in winter (data only for
Alaska) and nocturnal in summer.
Discussion
Wolf activity patterns have previously been
explained by human (Vila
`et al. 1995; Ciucci et al.
1997; Kusak et al. 2005) or prey activity (The-
uerkauf et al. 2003c). However, when I compared
studies carried out throughout the Holarctic, latitude
was the variable that best correlated with nocturnal
Table 2: The proportion of domestic animals in wolf diet and public
road density are the most important factors in the top 10 linear mod-
els that influence the nocturnal wolf activity index ranked by Akaike
weights (w). The five other factors have lower rankings
Rank Parameters in model w DAIC
C
AIC
C
1 Diet, road 0.283 0.00 )50.05
2 Diet, road, forest 0.172 1.00 )49.05
3 Diet, road, humans 0.092 2.24 )47.81
4 Diet 0.089 2.33 )47.72
5 Diet, road, mortality 0.073 2.70 )47.35
5 Diet, road, latitude 0.073 2.70 )47.35
5 Diet, road, temperature 0.073 2.70 )47.35
8 Latitude 0.058 3.16 )46.89
9 Diet, road, latitude, forest 0.043 3.79 )46.26
9 Diet, road, humans, forest 0.043 3.79 )46.26
Table 3: Latitude is the most important factor in the top 10 linear
models that influence the nocturnal wolf movement index ranked by
Akaike weights (w). The six other factors have lower rankings
Rank Parameters in model w DAIC
C
AIC
C
1 Latitude 0.242 0.00 )47.72
2 Latitude, road 0.109 1.60 )46.13
3 Latitude, humans 0.101 1.75 )45.97
3 Latitude, forest 0.101 1.75 )45.97
3 Road, diet 0.101 1.75 )45.97
6 Latitude, diet 0.080 2.21 )45.52
7 Latitude, mortality 0.075 2.35 )45.37
8 Road, temperature 0.065 2.64 )45.08
9 Diet 0.063 2.68 )45.04
9 Temperature 0.063 2.68 )45.04
What Drives Wolves: Fear or Hunger? J. Theuerkauf
652 Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH
movements of wolves, whereas latitude little
correlated with their nocturnal activity. This implies
that the daytime movements of wolves in southern
countries were more restricted than daytime activity.
As nocturnal movements were little correlated to
human-related factors, it seems unlikely that wolves
reduced their daytime movements exclusively
because of humans. The reason for a lower daytime
activity might be that wolves need shelter from the
sun when travelling, especially in open areas. The
most likely interpretation would therefore be that
daytime heat restricted the movements of wolves.
0.4
0.5
0.6
0.7
0.8
0.9
30 40 50 60 70
Latitude (ºN)
0.4
0.5
0.6
0.7
0.8
0.9
–5 0 5 10 15 20
Temperature (ºC)
0.4
0.5
0.6
0.7
0.8
0.9
0 10203040
Human density (per km2)
Nocturnal index
0.4
0.5
0.6
0.7
0.8
0.9
0 0.5 1 1.5
Public road density (km/km2)
0.4
0.5
0.6
0.7
0.8
0.9
0% 50% 100%
Mortality through humans
R2 = 0.372
0.4
0.5
0.6
0.7
0.8
0.9
0% 50% 100%
Domestic animals in diet
0.4
0.5
0.6
0.7
0.8
0.9
0% 50% 100%
Proportion of forest
R2 = 0.434
R2 = 0.317
R2 = 0.092 R2 = 0.277
R2 = 0.157
R2 = 0.159
R2 = 0.177
R2 = 0.069
R2 = 0.058
R2 = 0.097
R2 = 0.094
R2 = 0.055
R2 = 0.283
Fig. 1: Nocturnal activity (open circles, dot-
ted regression line) and movement (closed cir-
cles, continuous regression line) indices are
most correlated with latitude and the propor-
tion of domestic animals in the wolf diet, and
to a lesser degree with five other environmen-
tal factors across 11 wolf studies.
J. Theuerkauf What Drives Wolves: Fear or Hunger?
Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH 653
Indeed, in Poland, wolves became less active when
temperatures reached over 20C (Theuerkauf et al.
2003c), which might explain why wolves studied in
southern countries were more nocturnal than in
temperate countries. Nocturnal activity, on the other
hand, was mostly correlated to diet and road density.
Thus, while heat seems to reduce daytime move-
ments of wolves, it does not seem to be generally
correlated to daytime activity besides travelling.
Dependence on anthropogenic food resources and
the potential presence of humans in the wolf range
therefore seem to reduce the daytime activity of
wolves.
Most studies undertaken in temperate and sub-
tropical sites revealed bimodal patterns (activity
peaks at dawn and dusk), even though activity was
higher at night than during the day. Merrill & Mech
(2003) found that wolves in Minnesota had a noc-
turnal activity pattern, but suggested that their
method might not have been appropriate to detect a
bimodal pattern. I suggest that wolves usually tend
to be active during crepuscular phases, resulting in a
bimodal pattern in most regions. Subarctic sites offer
an interesting case of the impact of daylight on wolf
activity as light periods are extended to most of sum-
mer and winters are mostly dark. Wolves in these
regions were also mainly active during crepuscular
phases, which coincide with summer nights and
winter days (see Table 1). Studies by Ciucci et al.
(1997) in Italy and Kojola (2002) in Finland, in
which wolves were mainly nocturnal, were shorter
than a year. These studies might therefore not have
sampled the complete activity patterns of wolves in
their respective areas, because wolf activity patterns
are very variable (Eggermann et al. 2009).
Wolf howling, often associated with departure for
hunting, is usually most frequent around dawn
(Harrington & Mech 1982) or dusk (Nowak et al.
2007). Furthermore, both the temporal distribution
of wolf kills and their activity patterns were bimodal
with peaks at dawn and dusk in northeast Poland
(Theuerkauf et al. 2003c). The peaks of wolf activity
coincided with the periods when red deer, their
main prey in the Białowie_
za Forest (Je˛drzejewski
et al. 2002), are most active (Kamler et al. 2007).
This would support Curio’s (1976) suggestion that
the activity patterns of predators follow those of
their prey. Eriksen (2006), however, did not find
correlations between moose and wolf activity pat-
terns in Norway. Hunting strategies have evolved to
maximise kill rates, so the killing success of wolves
might be higher if they hunt at times when they
have maximum performance than if they synchro-
nise their activity patterns with those of prey species.
The vision of wolves is best adapted to crepuscular
light (Kavanau & Ramos 1975; Roper & Ryon 1977),
accordingly they probably find it easier to locate and
kill wild prey around sunrise and sunset. Good
vision is particularly important when attacking prey,
not only to ensure that a kill is made but also to pre-
vent fatal injuries to the predator (Asa & Mech
1995). Therefore, temporal selection of crepuscular
phases for hunting might be more important than
spatial selection as wolves do not seem to kill prey at
high density locations (Theuerkauf & Rouys 2008).
Nocturnal activity to avoid humans or the sun might
be an advantageous option for wolves that do not
rely on hunting in crepuscular light (e.g. because of
anthropogenic food resources). This might be the
case for the wolves in Italy that fed mostly on gar-
bage at dump sites (Ciucci et al. 1997).
Activity patterns are adaptations that enable ani-
mals to exploit their environment efficiently (Daan
& Aschoff 1982). From an evolutionary perspective,
it should be disadvantageous for wolves to adjust
their temporal activity patterns to avoid humans
unless the risk of being killed by direct persecution
is high. By reducing their daytime activity, wolves
foraging in the wild may fail to effectively exploit
the available food resources. Activity patterns of
wolves, abstraction made of the physiological restric-
tions at high temperatures, should therefore be
organised to both maximise food intake and
minimise the risk of being killed by humans.
Surprisingly, human-caused mortality was similarly
high in most studies considered in this paper despite
different activity patterns across sites. Of the other
variables directly related to humans that I analysed,
road density seemed more important than human
population density. A given population density
would probably have a higher impact on wolves if
people are distributed evenly over the wolf range
than if people are concentrated in a small area (e.g.
larger towns). Road density may therefore be a
better indicator of human-caused mortality because
it represents the degree of human penetration into
the wolf range. If it is, it would support the hypothe-
sis that human persecution can cause nocturnal
behaviour in wolves, as shown for coyotes, Canis
latrans (Kitchen et al. 2000). In this study, coyotes
adjusted their activity patterns towards nocturnal
behaviour during a period of human persecution but
changed their activity patterns after persecution
ceased. The impact of human-caused mortality on
wolf activity patterns may therefore vary through
the years according to reflect changes in human
What Drives Wolves: Fear or Hunger? J. Theuerkauf
654 Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH
attitudes toward wolves. I suggest that wolves are
probably adapted to hunting in crepuscular periods
because this activity pattern generally provides
greater foraging success, unless they feed on carrion
(Ciucci et al. 1997) or when prey density is very
high (Merrill & Mech 2003).
The reliability of a meta-analysis depends on
consistent and comparable data across studies.
I found several combinations of variables associated
with the activity and movement patterns of wolves.
The timing of sunrise and sunset, temperature, prey
availability and possibly human-caused mortality
were the most important variables associated with
nocturnal activity and movements of wolves. Future
studies of wolf activity patterns should be more
standardised and provide information on at least the
seven variables that I used to allow future
meta-analyses to be carried out. Replication in
wildlife research is crucially important to obtain
reliable results (Johnson 2002). Despite agreeing
with the above statement, I additionally postulate
that the most sophisticated locally undertaken study
might not reveal the most important factor that
influences species behavioural patterns. Local studies
have come to sometimes contradictory conclusions
about the factors that influence wolf activity
patterns, because it is often difficult to distinguish
the influence of different factors. A meta-analysis,
however, enables to remove this discrimination
because the effects of factors vary with studies. The
results of this study point to a general tendency to
underestimate the impact of heat on wolves and to
overestimate the impact of humans. I suggest that
studies from different areas should be taken into
account and reviewed before drawing definite con-
clusions about the behavioural ecology of wolves, or
any other species with a broad distribution.
Acknowledgements
I thank W. B. Ballard, L. Boitani, A. Chavez,
J. Kusak, L. D. Mech and A. Reichmann for provi-
ding me with additional information, as well as
R. Gula, G. R. Hunt, S. Rouys and anonymous
reviewers for comments and corrections.
Literature Cited
Asa, C. S. & Mech, L. D. 1995: A review of the sensory
organs in wolves and their importance to life history.
In: Ecology and Conservation of Wolves in a Changing
World (Carbyn, L. N., Fritts, S. H. & Seip, D. R., eds).
Canadian Circumpolar Institute, Alberta, Occasional
Publication No. 35, pp. 287—292.
Ballard, W. B., Ayres, L. A., Krausman, P. R., Reed, D. J.
& Fancy, S. G. 1997: Ecology of wolves in relation to a
migratory caribou herd in northwest Alaska. Wildl.
Monogr. 135, 3—47.
Blanco, J. C. & Corte
´s, Y. 2007: Dispersal patterns, social
structure and mortality of wolves living in agricultural
habitats in Spain. J. Zool. 273, 114—124.
Burnham, K. P. & Anderson, D. R. 2002: Model Selection
and Multi-Model Inference: A Practical Information –
Theoretic Approach, 2nd edn. Springer-Verlag,
New York.
Cederlund, G. 1981: Daily and seasonal activity pattern
of roe deer in a boreal habitat. Viltrevy 11, 315—353.
Chavez, A. S. & Gese, E. M. 2005: Food habits of wolves
in relation to livestock depredations in northwestern
Minnesota. Am. Midl. Nat. 154, 253—263.
Chavez, A. S. & Gese, E. M. 2006: Landscape use and
movements of wolves in relation to livestock in a wild-
land-agriculture matrix. J. Wildl. Manage. 70,
1079—1086.
Ciucci, P., Boitani, L., Francisci, F. & Andreoli, G. 1997:
Home range, activity and movements of a wolf pack in
central Italy. J. Zool. 243, 803—819.
Curio, E. 1976: The Ethology of Predation. Springer-
Verlag, Berlin.
Daan, S. & Aschoff, J. 1982: Circadian contributions to
survival. In: Vertebrate Circadian Systems: Structure
and Physiology (Aschoff, J., Daan, S. & Groos, G. A.,
eds). Springer-Verlag, Berlin, pp. 305—321.
Eggermann, J., Gula, R., Pirga, B., Theuerkauf, J., Tsu-
noda, H., Brzezowska, B., Rouys, S. & Radler, S. 2009:
Daily and seasonal variation in wolf activity in the
Bieszczady Mountains, SE Poland. Mammal. Biol. 74,
159—163.
Eriksen, A. 2006: Temporal and spatial activity patterns
of predator and prey – wolf and moose in south-
eastern Norway. MSc Thesis. Norwegian Univ. of Life
Science, A
˚s, 35 pp.
Fancy, S. G. & Ballard, W. B. 1995: Monitoring wolf
activity by satellite. In: Ecology and Conservation of
Wolves in a Changing World (Carbyn, L. N., Fritts,
S. H. & Seip, D. R., eds). Canadian Circumpolar
Institute, Alberta, Occasional Publication No. 35, pp.
329—333.
Faul, F., Erdfelder, E., Lang, A.-G. & Buchner, A. 2007:
G*Power 3: a flexible statistical power analysis program
for the social, behavioral, and biomedical sciences.
Behav. Res. Methods 39, 175—191.
Fritts, S. H. & Mech, L. D. 1981: Dynamics, movements,
and feeding ecology of a newly protected wolf popula-
tion in northwestern Minnesota. Wildl. Monogr. 80,
1—79.
J. Theuerkauf What Drives Wolves: Fear or Hunger?
Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH 655
Geist, V. 1963: On the behaviour of the North American
moose (Alces alces andersoni Peterson, 1950) in British
Columbia. Behaviour 20, 377—416.
Georgii, B. & Schro
¨der, W. 1983: Home range and activ-
ity patterns of male red deer (Cervus elaphus L.) in the
Alps. Oecologia 58, 238—248.
Gula, R. 2004: Influence of snow cover on wolf predation
patterns in Bieszczady Mountains, Poland. Wildl. Biol.
10, 17—23.
Gula, R. 2008: Wolf depredation on domestic animals in
the Polish Carpathian Mountains. J. Wildl. Manage.
72, 283—289.
Habib, B. & Kumar, S. 2007: Den shifting by wolves in
semi-wild landscapes in the Deccan Plateau, Maharash-
tra, India. J. Zool. 272, 259—265.
Harrington, F. H. & Mech, L. D. 1982: Patterns of home-
site attendance in two Minnesota wolf packs. In:
Wolves of the World: Perspectives of Behavior,
Ecology, and Conservation (Harrington, F. H. &
Paquet, P. C., eds). Noyes Publications, Park Ridge, NJ,
pp. 81—105.
Hovens, J. P. M. & Tungalaktuja, K. 2005: Seasonal fluc-
tuations of the wolf diet in the Hustai National Park
(Mongolia). Mammal. Biol. 70, 210—217.
Je˛drzejewski, W., Schmidt, K., Theuerkauf, J., Je˛drze-
jewska, B., Selva, N., Zub, K. & Szymura, L. 2002: Kill
rates and predation by wolves on ungulate populations
in Białowie_
za Primeval Forest (Poland). Ecology 83,
1341—1356.
Jeppesen, J. L. 1989: Activity patterns of free-ranging roe
deer (Capreolus capreolus) at Kalø. Dan. Rev. Game Biol.
13, 1—32.
Johnson, D. H. 2002: The importance of replication
in wildlife research. J. Wildl. Manage. 66, 919—
932.
Kaartinen, S., Kojola, I. & Colpaert, A. 2005: Finnish
wolves avoid roads and settlements. Ann. Zool. Fenn.
42, 523—532.
Kamler, J. F., Je˛drzejewska, B. & Je˛drzejewski, W. 2007:
Activity patterns of red deer in Białowie_
za National
Park, Poland. J. Mammal. 88, 508—514.
Kammermeyer, K. E. & Marchinton, R. L. 1977: Seasonal
changes in circadian activity of radio-monitored deer.
J. Wildl. Manage. 41, 315—317.
Kavanau, J. L. & Ramos, J. 1975: Influences of light on
activity and phasing of carnivores. Am. Nat. 109,
391—418.
Kitchen, A. M., Gese, E. M. & Schauster, E. R. 2000:
Changes in coyote activity patterns due to reduced
exposure to human persecution. Can. J. Zool. 78,
853—857.
Kojola, I. 2002: Worldwide First: GPS-GSM Collar on a
Finnish Wolf. Finnish Game and Fisheries Research
Institute and Environmental Studies. Available at:
http://www.environmental-studies.de/projects/2/gps-
gsm_collar_wolf-finland.html.
Kojola, I., Huitu, O., Toppinen, K., Heikura, K., Heikki-
nen, S. & Ronkainen, S. 2004: Predation on European
wild forest reindeer (Rangifer tarandus) by wolves (Canis
lupus) in Finland. J. Zool. 263, 229—235.
Kolenosky, G. H. & Johnston, D. 1967: Radio-tracking
timber wolves in Ontario. Am. Zool. 7, 299—303.
Krebs, J. R. & McCleery, R. H. 1984: Optimization in
behavioural ecology. In: Behavioural Ecology: An Evo-
lutionary Approach (Krebs, J. R. & Davies, N. B., eds).
Blackwell Science, Oxford, pp. 91—121.
Kusak, J., Skrbins
ˇek, A. M. & Huber, D. 2005: Home
ranges, movements, and activity of wolves (Canis lupus)
in the Dalmatian part of Dinarids, Croatia. Eur. J.
Wildl. Res. 51, 254—262.
Lima, S. L. & Dill, L. M. 1990: Behavioural decisions
made under the risk of predation. Can. J. Zool. 68,
619—640.
Linnell, J. D. C., Andersen, R., Andersone, Z., Balciaus-
kas, L., Blanco, J. C., Boitani, L., Brainerd, S.,
Breitenmoser, U., Kojola, I., Liberg, O., Løe, J.,
Okarma, H., Pedersen, H. C., Promberger, C., Sand, H.,
Solberg, E. J., Valdmann, H. & Wabakken, P. 2002:
The fear of wolves: a review of wolf attacks on
humans. NINA Oppdragsmelding 731, 1—65.
Mattioli, L., Apollonio, M., Mazzarone, V. & Centofanti,
E. 1995: Wolf food habits and wild ungulate availabil-
ity in the Foreste Casentinesi National Park, Italy. Acta
Theriol. 40, 387—402.
Mech, L. D. 1988: The Arctic Wolf: Living with the Pack.
Yoyageur Press, Stillwater, MN.
Mech, L. D., Meier, T. J., Burch, J. W. & Adams, L. G.
1995: Patterns of prey selection by wolves in Denali
National Park, Alaska. In: Ecology and Conservation of
Wolves in a Changing World (Carbyn, L. N., Fritts, S. H.
& Seip, D. R., eds). Canadian Circumpolar Institute,
Alberta, Occasional Publication No. 35, pp. 231—244.
Mech, L. D., Harper, E. K., Meier, T. J. & Paul, W. J.
2000: Assessing factors that may predispose Minnesota
farms to wolf depredation on cattle. Wildl. Soc. Bull.
28, 623—629.
Meriggi, A. & Lovari, S. 1996: A review of wolf predation
in southern Europe: does the wolf prefer wild prey to
livestock? J. Appl. Ecol. 33, 1561—1571.
Meriggi, A., Brangi, A., Matteucci, C. & Sacchi, O. 1996:
The feeding habits of wolves in relation to large prey
availability in northern Italy. Ecography 19, 287—295.
Merrill, S. B. 2000: Road densities and gray wolf, Canis
lupus, habitat suitability: an exception. Can. Field Nat.
114, 312—313.
Merrill, S. B. & Mech, L. D. 2003: The usefulness of GPS
telemetry to study wolf circadian and social activity.
Wildl. Soc. Bull. 31, 947—960.
What Drives Wolves: Fear or Hunger? J. Theuerkauf
656 Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH
Montgomery, G. G. 1963: Nocturnal movements and
activity rhythms of white-tailed deer. J. Wildl. Manage.
27, 422—427.
Nowak, S., Je˛drzejewski, W., Schmidt, K., Theuerkauf, J.,
Mysłajek, R. W. & Je˛drzejewska, B. 2007: Howling
activity of free-ranging wolves (Canis lupus) in the
Białowie_
za Primeval Forest and the Western Beskidy
Mountains (Poland). J. Ethol. 25, 231—237.
Reichmann, A. & Saltz, D. 2005: The Golan wolves: the
dynamics, behavioral ecology, and management of an
endangered pest. Isr. J. Zool. 51, 87—133.
Roper, T. J. & Ryon, J. 1977: Mutual synchronisation of
diurnal activity rhythms in groups of red wolf ⁄coyote
hybrids. J. Zool. 182, 177—185.
Rouys, S., Theuerkauf, J. & Krasin
´ska, M. 2001: Accuracy
of radio-tracking to estimate activity and distances
walked by European bison in the Białowie _
za Forest,
Poland. Acta Theriol. 46, 319—326.
Sand, H., Zimmermann, B., Wabakken, P., Andre
`n, H. &
Pedersen, H. C. 2005: Using GPS technology and GIS
cluster analyses to estimate kill rates in wolf-ungulate
ecosystems. Wildl. Soc. Bull. 33, 914—925.
Schmidt, K., Je˛drzejewski, W., Theuerkauf, J.,
Kowalczyk, R., Okarma, H. & Je˛drzejewska, B. 2008:
Reproductive behaviour of wild-living wolves in
Białowie_
za Primeval Forest (Poland). J. Ethol. 26,
69—78.
Theuerkauf, J. & Je˛drzejewski, W. 2002: Accuracy of
radiotelemetry to estimate wolf activity and locations.
J. Wildl. Manage. 66, 859—864.
Theuerkauf, J. & Rouys, S. 2008: Habitat selection by
ungulates in relation to predation risk by wolves and
humans in the Białowie_
za Forest, Poland. For. Ecol.
Manage. 256, 1325—1332.
Theuerkauf, J., Rouys, S. & Je˛drzejewski, W. 2003a:
Selection of den, rendezvous, and resting sites by
wolves in the Białowie_
za Forest, Poland. Can. J. Zool.
81, 163—167.
Theuerkauf, J., Je˛drzejewski, W., Schmidt, K. & Gula, R.
2003b: Spatiotemporal segregation of wolves from
humans in the Białowie_
za Forest (Poland). J. Wildl.
Manage. 67, 706—716.
Theuerkauf, J., Je˛dzejewski, W., Schmidt, K., Okarma,
H., Ruczyn
´ski, I., S
´nie_
zko, S. & Gula, R. 2003c: Daily
patterns and duration of wolf activity in the Białowie_
za
Forest, Poland. J. Mammal. 84, 243—253.
Theuerkauf, J., Gula, R., Pirga, B., Tsunoda, H., Egger-
mann, J., Brzezowska, B., Rouys, S. & Radler, S. 2007:
Human impact on wolf activity in the Bieszczady
Mountains, SE Poland. Ann. Zool. Fenn. 44, 225—231.
Thurber, J. M., Peterson, R. O., Drummer, T. D. & Thom-
asma, S. A. 1994: Gray wolf response to refuge bound-
aries and roads in Alaska. Wildl. Soc. Bull. 22, 61—68.
Tsunoda, H., Gula, R., Theuerkauf, J., Rouys, S., Radler,
S., Pirga, B., Eggermann, J. & Brzezowska, B. 2009:
How does parental role influence the activity and
movements of breeding wolves? J. Ethol. 27,
185—189.
Vila
`, C., Urios, V. & Castroviejo, J. 1995: Observations on
the daily activity patterns in the Iberian wolf. In: Ecol-
ogy and Conservation of Wolves in a Changing World
(Carbyn, L. N., Fritts, S. H. & Seip, D. R., eds). Cana-
dian Circumpolar Institute, Alberta, Occasional Publica-
tion No. 35, pp. 335—340.
Wabakken, P., Sand, H., Liberg, O. & Bja
¨rvall, A. 2001:
The recovery, distribution, and population dynamics of
wolves on the Scandinavian peninsula, 1978–1998.
Can. J. Zool. 79, 710—725.
J. Theuerkauf What Drives Wolves: Fear or Hunger?
Ethology 115 (2009) 649–657 ª2009 Blackwell Verlag GmbH 657