Animal Behaviour 92 (2014) 93e100
Contents lists available at ScienceDirect
Animal Behaviour
journal homepage: www.elsevier.com/locate/anbehav
A penny saved is a penny earned: lean season foraging strategy of an
alpine ungulate
Mayank Kohli a, *, Mahesh Sankaran b, Kulbhushansingh R. Suryawanshi c, d,
Charudutt Mishra c, d
a
Postgraduate Program in Wildlife Biology and Conservation, Wildlife Conservation Society e India Program, National Center for Biological Sciences,
Bangalore, India
b
Ecology and Evolution Group, National Center for Biological Sciences, Bangalore, India
c
Nature Conservation Foundation, Mysore, India
d
Snow Leopard Trust, Seattle, WA, U.S.A.
a r t i c l e i n f o
Article history:
Received 26 August 2013
Initial acceptance 17 October 2013
Final acceptance 12 March 2014
Published online
MS. number: 13-00707R
Keywords:
blue sheep
grazing
herbivore
mountain ungulate
optimal foraging
Pseudois nayaur
trans-Himalaya
Lean season foraging strategies are critical for the survival of species inhabiting highly seasonal environments such as alpine regions. However, inferring foraging strategies is often difficult because of
challenges associated with empirically estimating energetic costs and gains of foraging in the field. We
generated qualitative predictions for the relationship between daily winter foraging time, body size and
forage availability for three contrasting foraging strategies including time minimization, energy intake
maximization and net energy maximization. Our model predicts that for animals employing a time
minimization strategy, daily winter foraging time should not change with body size and should increase
with a reduction in forage availability. For energy intake maximization, foraging time should not vary
with either body size or forage availability. In contrast, for a net energy maximization strategy, foraging
time should decrease with increase in body size and with a reduction in forage availability. We contrasted proportion of daily time spent foraging by bharal, Pseudois nayaur, a dimorphic grazer, across
different body size classes in two high-altitude sites differing in forage availability. Our results indicate
that bharal behave as net energy maximizers during winter. As predicted by the net energy maximization
strategy, daily winter foraging time of bharal declined with increasing body size, and was lower in the
site with low forage availability. Furthermore, as predicted by our model, foraging time declined as the
winter season progressed. We did not find support for the time minimizing or energy intake maximizing
strategies. Our qualitative model uses relative rather than absolute costs and gains of foraging which are
often difficult to estimate in the field. It thus offers a simple way to make informed inferences regarding
animal foraging strategies by contrasting estimates of daily foraging time across gradients of body size
and forage availability.
Ó 2014 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Alpine and temperate regions are characterized by a short
summer with abundant nutrient-rich forage, and a severe, often
long, winter when much of the forage is of poor quality (post leaf
senescence), and often inaccessible because of snow (Goodson &
Stevens, 1991; Parker, Barboza, & Gillingham, 2009). Ruminants
inhabiting such regions often face food-related stresses during the
lean season (Parker et al., 2009). Furthermore, the energetic costs of
foraging associated with thermoregulation and locomotion increase owing to low ambient temperatures and snow, making
* Correspondence: M. Kohli, F-21, MSc Wildlife Office, National Center for Biological Sciences, GKVK Campus, Bellary Road, Bangalore 560065, Karnataka, India.
E-mail address: [email protected] (M. Kohli).
foraging relatively costly (Dailey & Hobbs, 1989; Murray, 1991;
Sabine et al., 2002). Animals often lose body condition during the
winter months and have to rely on body reserves built over the
productive summer season to see the lean winter season through
(Parker et al., 2009; Taillon, Sauvé, & Côté, 2006). Lean season
foraging strategies supplement body reserves built over summer
and are thus crucial for the survival of species inhabiting highly
seasonal environments.
Despite having received much attention, there is still disagreement over lean season foraging strategies of ungulates. While some
suggest that ruminants should behave as time minimizers to
reduce thermal exposure and minimize predation risk (Bergman,
Fryxell, Gates, & Fortin, 2001), others support an energy intake
maximization (feeding time maximization) strategy (Belovsky,
http://dx.doi.org/10.1016/j.anbehav.2014.03.031
0003-3472/Ó 2014 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
1986; Schmitz, 1990, 1991). Others, still, have found no evidence for
either (kudu, Tragelaphus spp.: Owen-Smith, 1994; musk ox, Ovibos
moschatus: Forchhammer & Boomsma, 1995), and suggest that
during the lean season, ruminants in extreme environments should
employ an energy saving or net energy maximization strategy (Kie,
1996; Loe et al., 2007; Murray, 1991).
We generated predictions regarding foraging time for the time
minimization, energy intake maximization and net energy maximization foraging strategies. In particular, we examined how lean
season foraging time is influenced by body size and forage availability, factors that are known to explain within- and across-species
differences in foraging behaviour. Body size plays a critical role in
mediating the foraging behaviour of animals as it determines the
energy required to maintain basal metabolic rate (Illius & Gordon,
1987) and the intake rate of food (Gross, Hobbs, & Wunder, 1993;
Shipley, Gross, Spalinger, Hobbs, & Wunder, 1994). For ruminants,
body size assumes even more significance as rumen volume and
gut capacity, factors that determine the amount of food a ruminant
can eat and how well it can digest it, are linked closely with body
size (Clauss, Schwarm, Ortmann, Streich, & Hummel, 2007; Illius &
Gordon, 1987). Although the role of body size in shaping foraging
behaviour has been investigated across species (Mysterud, 1998;
Owen-Smith, 1992), its role within species, even highly dimorphic
ones, has rarely been explored (but see Pelletier & Festa-Bianchet,
2004).
CONCEPTUAL MODELS AND PREDICTIONS
Time Minimization
Animals seeking to minimize foraging time should maximize
short-term intake rate. Given that minimum energy required (E) to
maintain basal metabolic rate (BMR) scales with body weight (W)
as W0.75 (Clarke, Rothery, & Isaac, 2010; Kleiber, 1932), while
maximum intake rate (I) of dry matter scales as W0.71 (Shipley et al.,
1994), time to meet minimum energy requirements,
.
Tmin fE=I ¼ pW 0:75 qW 0:71 ¼ aW 0:04
(1)
where ‘p’ can be conceptualized as energy required per unit body
weight, ‘q’ as intake rate of energy per unit of body weight and ‘a’ as
the amount of time required to meet energy requirements of unit
body weight. Intake rate of energy per unit of body weight, ‘q’, will
depend on forage availability and increase with increasing forage
availability. Hence ‘a’ will increase linearly as forage availability
decreases (see Andersen & Saether, 1992).
For a fixed availability of forage, Tmin scales very gradually with
body size (equation (1), Fig. 1) especially for within-species size
differences. Although our model predicts that foraging time should
scale with body size with an exponent of 0.04, it is likely that this
exponent is indistinguishable from zero based on empirical data
(Shipley et al., 1994). At the level of a species, this implies almost
equal foraging times across body size classes. For a given body size,
however, as forage availability decreases, the time required to meet
minimum energy requirements increases (Fig. 1). Therefore, a time
minimization strategy predicts almost equal foraging times for all
body size classes, and an increase in foraging time for all size classes
with decreasing forage availability.
Energy Intake Maximization
Energy intake maximizers are likely to be constrained by the
daily time available for foraging (Belovsky, 1981; Forchhammer &
Boomsma, 1995; but also see Owen-Smith, 1994) when either
More forage
Less forage
T min
94
Body size
Figure 1. Time required to meet minimum energy requirement, Tmin, as a function of
body size. Body size varies over two orders of magnitude for ease of interpretation. For
a fixed body size, Tmin increases as forage availability decreases.
encounter rate of food is low or ingestion time is high (owing to
prolonged cropping and chewing processes). Ruminants may also
be constrained by digestive processes (rumen volume/turnover
rate). Foraging time for ruminants should then be determined by
which of the two processes, ingestive or digestive, is more limiting.
When forage availability is limited and/or forage quality is poor,
animals will be limited by the total time available for foraging
(Fortin, Fryxell, & Pilote, 2002). As our study is restricted to the
winter season, we limit our discussion to the latter scenario. In this
case, animals seeking to maximize energy intake are likely to forage
for as much time as possible. Thus, foraging time should not vary
with body size or forage availability, and should approximate the
total time available for foraging in a day.
Net Energy Maximization
Gross energy accumulated, Ea, in time T spent foraging in a patch
is a type II functional response (Laca, Distel, Criggs, & Demment,
1994; but also see Searle, Hobbs, & Shipley, 2005) whereas energy spent foraging or the cost of foraging, Ec, is a linear function of
the time spent foraging (Fig. 2). The net energy gained, Enet, is, then,
the difference between gross energy gained and energy spent, i.e.,
Enet ¼ Ea Ec ¼ b 1 ecT dT;
(2)
where b, c and d are functions of body size and the environment
(Fig. 2). Specifically, ‘b’ reflects the total energy contained in a patch
(the asymptote in Fig. 2) and increases as forage availability increases (Laca et al., 1994). For a fixed availability of forage, ‘c’ reflects
the rate at which the patch is depleted (asymptote is reached), and
increases with body size; ‘d’ is the cost of foraging (locomotory and
thermoregulatory) per unit foraging time and increases with body
size (Murray, 1991). Also, ‘d’ increases with decreasing forage
availability as the costs of locomotion increase because of an increase in searching effort (Murray, 1991).
Graphically, Enet is the difference between the gain and cost
curves and the solution for optimum foraging time, Tf, lies where
this difference is greatest (vertical lines in Fig. 2c, d).
For a fixed availability of forage, the cost of foraging increases
with body size, that is, the slope of the cost curve is steeper for
larger animals (Fig. 2a). At the same time, larger animals deplete
patches faster which means that their returns reach an asymptote
sooner relative to smaller animals (Fig. 2a). Thus, the optimal
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
(b)
(c)
(d)
Net energy
Energy
(a)
95
Time spent foraging
Figure 2. Gross energy gained (type II curve) and spent (straight line) and net energy gained with time spent foraging for (a, c) varying body size and (b, d) varying forage
availability. The thicker lines represent larger size classes (a, c) or increased forage availability (b, d). In (a), foraging cost increases with size and larger animals also deplete the patch
sooner. Optimum foraging time (indicated by vertical lines), where net energy is maximized, decreases with increasing size (c). In (b), with increasing availability of forage, foraging
cost decreases while energy accumulated in time T increases. Optimum time increases with increase in forage availability (d).
foraging time, Tf, decreases as body size increases (Fig. 2c). For a
fixed body size, an increase in forage availability leads to an increase in the amount of energy gained per unit time. It also leads to
a decline in the searching time which means that the cost of
foraging per unit time decreases (Fig. 2b). Accordingly, the optimal
foraging time, Tf, will increase with increasing forage availability
(Fig. 2d).
The three different strategies thus lead to contrasting predictions regarding how time spent foraging should vary with body
size and forage availability. A time minimization strategy predicts
that foraging time remains largely invariant across body sizes but
increases as forage quality decreases (Fig. 3). In contrast, an energy
intake maximization strategy predicts that foraging time is not
influenced by either body size or forage availability, while a net
energy maximization strategy predicts a decrease in foraging time
with increasing body size and with decreasing forage quality
(Fig. 3).
We tested these predictions on bharal, Pseudois nayaur, a
medium-sized (mean adult body weight 55 kg; Mishra, Van
METHODS
Study Species
Adult male bharal show considerable variation in size with age,
with the youngest adult males nearly the size of adult females
while the largest ones grow up to twice this size (Schaller, 1998).
Morphologically, the species is graze-adapted (specialized to feed
on graminoids), and while its summer diet comprises almost
(b)
(c)
Foraging time
(a)
Wieren, Heitkönig, & Prins, 2002) grazing ruminant, in a highaltitude (>4000 m) trans-Himalayan region of India. The species
shows high sexual dimorphism in body size with adult males
weighing 60e75 kg and adult females 35e45 kg (Schaller, 1998;
Weckerly, 1998). Specifically, we examined how time spent
foraging by bharal in winter varied as a function of body size and
forage availability. In addition, we quantified how time spent
foraging within sites changed over the course of the winter as
forage availability decreased.
Body size
Figure 3. Graphical summary of predictions from the three strategies showing relationship between foraging time and body size of the animal and availability of forage: (a) time
minimization, (b) energy intake maximization and (c) net energy maximization. Dashed lines represent conditions of lower forage availability.
96
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
entirely of graminoids, it has a mixed diet in winter (Mishra, Van
Wieren, Ketner, Heitkönig, & Prins, 2004; Suryawanshi,
Bhatnagar, & Mishra, 2010).
Study Area
The study was carried out in the Kibber region of Spiti Valley
(32 N, 78 E), Himachal Pradesh, India. The area is a cold, mountainous desert characterized by a mosaic of rolling and steep hills,
rocky cliffs and outcrops with the altitude varying from 3800 to
5000 m. Temperatures during winters drop below 35 C and
vegetation growth is restricted to a short springesummer window
from May to August. Precipitation is predominantly in the form of
winter snow most of which falls during December to March.
The vegetation of this region is classified as dry alpine steppe.
Much of the landscape is bare rock and scree slopes where no
vegetation grows. Dominant species include shrubs such as Caragana versicolor, Lonicera spinosa, Eurotia (short shrub) and Rosa
webbiana. Graminoids found here include species of Stipa, Carex,
Poa, Elymus and Festuca. Commonly occurring herbs include Scorzonera and Bupleurum.
Native herbivores in the study area include bharal, ibex, Capra
sibirica, and woolly hare, Lepus oiostolus, while the domestic
component comprises sheep, Ovis aries, goat, Capra hircus, horse,
Equus caballus, donkey, Equus asinus, cow, Bos indicus, yak, Bos
grunniens, and ‘dzomo’ which is a coweyak hybrid. Large tracts of
the study area are utilized by both wild herbivores and domestic
livestock in the summer. However, in some areas, as part of a
conservation initiative, villagers are compensated for keeping domestic livestock off tracts of land which are thus ‘reserved’ for wild
herbivores (Mishra, Prins, & Van Wieren, 2003). Areas outside the
reserve that are extensively grazed by livestock in summer tend to
be characterized by lowered forage availability for bharal in the
winter, whereas ‘reserve’ areas typically support higher forage
biomass (Suryawanshi et al., 2010). To investigate how variation in
forage availability affects bharal foraging in the lean winter months,
we chose two study sites: one within the ‘reserve’ area and another
in an area intensively grazed by livestock. The two sites were
similar in altitude and aspect.
Estimating Forage Availability
Difference in forage availability between sites was quantified
between 26 December 2011 and 4 January 2012 by harvesting
above-ground herbaceous biomass from 37 and 38 randomly
located 3 3 m plots in the high and low forage sites, respectively.
Shrubs were not clipped but their percentage cover in plots was
recorded. Harvested biomass was air-dried, separated into grass
and forb components, and weighed to within 10 mg using an
electronic scale.
Quantifying Bharal Foraging Behaviour
In each of the two study sites, bharal time activity budgets were
constructed through scan samples of herds (St-Louis & Côté, 2012).
Bharal herds were observed using a 45e60 spotting scope and
8 40 binoculars from a distance of 20e150 m. During each scan,
the ageesex class of each individual of the herd and its activity
were noted. A scan of the herd was usually completed within 2 min.
For larger herds, or where herds were far off, scans were restricted
to a maximum of 5 min. Successive scans were separated by 10 min
intervals. The number of scans in a day varied from 10 to 47. Agee
sex classification was done following Mishra et al. (2004). Animals
were classified as kid (<12 months old), yearling (12e24 months),
adult female (>24 months), Class 1 male (24e48 months), Class 2
male (48e72 months), Class 3 male (72e96 months) and Class 4
male (>96 months old). Size classes were later grouped into four
major classes reflecting a size gradient, namely kids (kids þ yearlings), females (adult females), small males (Class 1 þ Class 2 males)
and large males (Class 3 þ Class 4 males). Females and small males
represented the same size class but were kept independent to
ensure that body size-related patterns were not confounded by sex
differences.
Bharal activities were classified into feeding, resting, walking,
social interactions, standing and vigilance. ‘Feeding’ included time
spent biting and chewing food and moving from one plant to
another within a feeding patch. This movement from one plant to
another while feeding was different from ‘walking’ as the former
was usually slow and deliberate with the head held below the
shoulder whereas the latter was faster with the head held above
the shoulder. Animals that were lying down were classified as
‘resting’. An animal was recorded as ‘standing’ when it was
standing but not vigilant. ‘Vigilance’ (most often while standing;
rarely while resting) involved the animal being alert with the head
and shoulders tensed and could be easily differentiated from
‘standing’. Any activity involving more than one animal was classified as a ‘social interaction’, including maleemale, maleefemale
and femaleefemale interactions.
A total of 632 and 730 5 min scans were performed and averaged daily over 27 and 26 sampling days in the high and low forage
sites, respectively. Care was taken to ensure similar sampling effort
across all times of day, from first light to a little after sunset, and to
sample across a range of group sizes in both sites. Data were
collected on sequential days as far as weather permitted. The
observation team was split into two to collect data from the two
sites. The two teams spent sufficient time together to calibrate
classification of animals and activities. Importantly, team members
were rotated across sites to avoid any observer bias.
Analyses
Proportion of daily time spent foraging (and resting) by bharal in
each site was estimated as the proportion of individuals within
each size class in a scan that were foraging (and resting), averaged
over all scans in a day. We used a general linear model (GLM) to
examine how proportion of daily time spent foraging by bharal
varies over time as the winter progresses as a function of site and
body size class . We included days since the start of the study as a
predictor variable since forage depletion with the advance of the
winter season is likely to affect foraging time. Although our
response variable is a proportion, we used a GLM with normal errors and identity link function since inspection of residuals indicated homogeneity of variance with normally distributed errors.
Starting with a global model that included all interaction terms, we
sequentially dropped nonsignificant model terms using F tests until
a minimum adequate model was achieved. All statistical analyses
were done using the statistical software R version 2.14.2 (R
Development Core Team, http://www.R-project.org).
RESULTS
Forage Availability
Mean above-ground graminoid biomass in the ‘reserve’ area was
higher (3.23 0.41 g/m2; hereafter ‘high-forage’ site) than that in
the livestock-grazed area (0.56 0.15 g/m2; hereafter ‘low-forage’
site). In contrast, mean above-ground biomass of forbs was similar
in both areas (reserve: 4.41 0.59 g/m2; livestock grazed:
4.19 0.80 g/m2). Mean percentage cover of shrubs, which are
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
typically avoided by bharal, was 37% (median 30%) and 16% (median
5%) in the reserve and grazed sites, respectively.
Table 2
ANOVA table showing significant main effects of body size class, forage availability
(site) and advance of winter season (day) on daily resting time
Time Activity Budget
Our final minimum adequate model included the main effects of
time, forage availability and body size, but did not include any
interaction terms (see Table 1 for foraging time, Table 2 for resting
time). Proportion of daily time spent foraging was highest for kids,
followed by small males, females and large males in that order
(Appendix Table A1, Fig. 4). Kids spent a significantly higher proportion of time foraging compared with all other size classes
(P < 0.05, Table A1). Small males spent significantly more time
foraging than large males (P < 0.05). There were no significant
differences between small males and females, which represent the
same body size class, or between females and large males. For each
size class, the proportion of daily time spent foraging was 6.8%
higher (P < 0.001; Table A1, Fig. 4) in the high-forage ‘reserve’ site
than in the low-forage availability site, and declined by 0.3% per day
with advance of the winter season in both sites (P < 0.001;
Table A1, Fig. 4).
The proportion of daily time spent resting showed the opposite
trend: it was almost 5% less in the high-forage ‘reserve’ site
(P ¼ 0.018; Table A2, Fig. 5) and increased by 0.25% per day with
advance of the winter season (P < 0.0001; Table A2, Fig. 5). Bharal
kids rested the least followed by small males, females and large
males in that order (Table A2, Fig. 5). Proportion of daily time spent
resting by kids was lower than for other classes (P < 0.05 for differences with females and large males, P ¼ 0.08 for kids and small
males), while small males spent less time resting than large males
(P < 0.05). Time spent resting by females was not significantly
different from that for either small or large males.
DISCUSSION
Our results suggest that bharal behave as net energy maximizers
during winter. In accordance with the predictions of this strategy,
the mean proportion of daily time spent foraging decreased with
increasing body size and was lower in the site where forage availability was lower. The fact that bharal fed less where there was less
forage available, and that this trend was consistent across body size
classes, was not in accordance with the predictions for the time
minimization strategy, wherein bharal should have compensated
for lower forage availability by feeding longer. Similarly, our results
did not support energy intake maximization, where feeding time
was expected to be similar across size classes, and close to the
maximum possible in a day regardless of forage availability.
Resting, an energy-conserving activity, rather than other energyconsuming activities (e.g. social interaction and vigilance)
appeared to compensate for reduced foraging. All animals rested
more in the grazed site with lower forage availability and smaller
animals rested less than larger ones. An important result that
emerged from our study was that small adult males differed from
large adult males but not from adult females, which represent the
same body size class, in their daily foraging or resting times. This
Table 1
ANOVA table showing significant main effects of body size class, forage availability
(site) and advance of winter season (day) on daily foraging time
Site
Class
Day
Residuals
df
Sum of squares
Mean square
F
Pr(>F)
1
3
1
188
0.3862
0.3774
0.3360
3.4198
0.38623
0.12579
0.33596
0.01819
21.2324
6.9151
18.4687
<0.001
<0.001
<0.001
97
Site
Class
Day
Residuals
df
Sum of squares
Mean square
F
Pr(>F)
1
3
1
188
0.2219
0.3294
0.2512
3.6471
0.221912
0.109802
0.251168
0.019399
11.4392
5.6601
12.9472
<0.001
<0.001
<0.001
strongly suggests that observed differences between winter activity
budgets were driven by body size differences rather than sexrelated reasons.
Earlier work in the study area has shown that forage quality
during winters falls below maintenance level (mean crude protein
content for graminoids ¼ 2.1%, Suryawanshi et al., 2010) and forage
availability is generally low in comparison with other grasslands in
the world (Mishra, 2001). Like many other temperate and alpine
ungulates (e.g. black-tailed deer, Odocoileus hemionus: Parker,
Gillingham, Hanley, & Robbins, 1993; white-tailed deer, Odocoileus virginianus: DelGiudice, Mech, Kunkel, Gese, & Seal, 1992; red
deer, Cervus elaphus: Loison, Langvatn, & Solberg, 1999; moose,
Alces alces: Milner, van Beest, Solberg, & Storaas, 2013; but see
Couturier, Côté, Huot, & Otto, 2008), bharal lose body condition as
winter advances. This implies that average daily net energy intake
over winter is lower than the energy required to meet the basal
metabolic rate, and hence time minimization, which implies
curtailment of foraging once the minimum energy requirement is
met (Hixon, 1982), would be an unachievable strategy. Instead of
minimizing foraging time, animals may maximize daily energy
intake during winter to minimize their energy deficit (e.g. whitetailed deer: Schmitz, 1990, 1991). However, if the costs of foraging
are high and gains low, maximizing feeding time (hence maximizing intake of energy) could increase the daily energy deficit
instead of reducing it (Murray, 1991). Bharal seem to balance
foraging and resting according to body size and forage availability
such that net energy gained daily is maximized and, hence, daily
energy deficit minimized. Although this still means a daily loss in
body condition, the strategy can potentially minimize the rate of
this loss. This could be a critical overwintering strategy of bharal.
During winters, death caused by loss of body reserves is perhaps
the greatest risk facing temperate and alpine ungulates and this risk
can be reduced by minimizing daily energy deficit (Parker et al.,
2009).
While it is accepted that foraging strategies form a continuum
between end points represented by time minimization and energy
maximization, few studies have explored this continuum or proposed a framework to understand and model them (Bergman et al.,
2001). In the qualitative framework presented here, net energy
maximization offers an explanation for such a continuum. As it is
responsive to changes in both the energetic costs and gains of
foraging, a change in foraging conditions can cause the optimum
foraging time (where net energy is maximized) to vary. Factors that
alter the energetic gains and/or cost of foraging (e.g. forage
depletion, snow cover, extreme weather, large group sizes) will
affect the optimum solution and lead to a gradient in predicted
foraging time (see Fig. 2). Natural selection would favour such a
strategy (Bergman et al., 2001) rather than a restrictive choice between a minimum (as per time minimization) or a maximum (as
per energy intake maximization). This is an important aspect as in
many conditions neither time minimization nor energy intake
maximization may be optimal for the animal. For example, northern ungulates reduce their foraging time when harassed by biting
flies (e.g. Toupin, Huot, & Manseau, 1996), but unless the selection
on this reduction is very strong, animals may not benefit by
98
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
(a)
Proportion of time spent foraging
1
0.8
(b)
a
b
bc
0.6 c
0.4
0.2
Kids
Females
Small males
Large males
0
0
10
20
30
40
50
60
0
10
20
30
40
Days from start
Figure 4. Proportion of time spent foraging by bharal in winter in relation to the advance of the season (sampling day). (a) The low-forage ‘grazed’ site. (b) The high-forage ‘reserve’
site. Different letters over regression lines indicate significant difference between intercepts.
high, optimal foraging time for net energy maximizers may lie
beyond the total time available for foraging in a day (see Fig. 2). In
such a situation, net energy maximizers are likely to forage for all
the time available and may, therefore, appear to behave as energy
intake maximizers. With the onset and advance of the lean season
(and therefore of forage depletion), the energetic costs of foraging
increase and gains decrease. While energy intake maximization
would still predict that animals forage for all of the time available,
net energy maximization will predict a continuous decline in
adopting a time minimization strategy at the cost of extra energy
intake. Within the framework of net energy maximization,
increased costs of foraging because of biting flies would mean that
animals must reduce their foraging time without adopting the
extreme time minimization solution.
Although we have dealt only with winter foraging across two
sites differing in forage availability, our qualitative model can provide insights into foraging behaviour across seasons. In the productive season, when costs of foraging are relatively low and gains
Proportion of time spent resting
0.8
(a)
Kids
Females
Small males
Large males
(b)
0.6
0.4
c
b
b
0.2
a
0
0
10
20
30
40
50
60
0
10
20
30
40
Days from start
Figure 5. Proportion of time spent resting by bharal in winter in relation to the advance of the season (sampling day). (a) The low-forage ‘grazed’ site. (b) The high-forage ‘reserve’
site. Different letters over regression lines indicate significant difference between intercepts.
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
foraging time and time minimization a continuous increase. Our
results show proportion of time spent foraging and resting by
bharal decreased and increased, respectively, through our study
period (Fig. 4), providing further support to the net energy maximization hypothesis. The extent of depletion of forage (or advance
of the lean season) at which predictions from energy intake and net
energy maximization strategies diverge will depend on the environment and species concerned.
While the patterns we have observed suggest that body size
plays an important role in foraging behaviour, there are other differences between individuals and groups that must be inspected in
future work. Body condition at the start of winter is one such
important factor that is likely to affect winter foraging behaviour in
combination with body size (Monteith et al., 2013). Animals are
likely to regulate body reserves during the lean season in accordance with the amount of somatic reserves available at the start of
the lean season and this may vary between individuals as a result of
differences in body size, physiological state and other factors. For
example, large rutting males may enter winter in the poorest
nutritional condition owing to the energetic costs of mating
(Barboza, Hartbauer, Hauer, & Blake, 2004). Including such factors
that operate at the level of an individual may explain the observed
patterns better. Although other studies have found evidence for net
energy maximization during the lean season (reindeer, Rangifer
tarandus: Loe et al., 2007; black-tailed deer: Kie, 1996; kudu: OwenSmith, 1994), a more widespread application of this concept in
ungulate foraging studies has presumably been limited owing to
difficulty in estimating energetic costs and gains of foraging for
free-ranging wild herbivores which makes it difficult to generate
testable quantitative predictions. A qualitative framework, such as
the one presented here, using ‘quasi experimental’ differences in
forage availability and the natural variation in body size, generates
contrasting predictions which can be easily tested in the field. The
challenge for the future is to incorporate energetic costs other than
locomotion and thermoregulation, and effects of group foraging in
our model of net energy maximization and to test it in a broad array
of foraging conditions. Finally, changes in foraging behaviour of
animals can have significant population-level impacts mediated
through changes in energy intake. Bharal are not migratory and
inhabit more-or-less the same area throughout the year. It is thus
likely that animals in areas of more forage availability enter the
winter season in better body condition. Bharal in areas grazed by
livestock may, therefore, be doubly disadvantaged; first, they enter
winter in poorer body condition and, second, are less able to supplement their body reserves during winter. This may affect the
productivity of bharal herds in areas with very poor forage
conditions.
Acknowledgments
We thank the Department of Science and Technology, Ministry
of Science and Technology, India for funding the field work and the
Forest Department, Himachal Pradesh for providing necessary
work permits. The National Centre for Biological Sciences provided
crucial institutional, financial and infrastructural support. The Nature Conservation Foundation helped with logistics in the field and
we are grateful to the Whitley Fund for Nature for support. We
thank Dr Ajith Kumar and Dr Anindya Sinha for their valuable
suggestions at various stages of the study, Dr Kavita Isvaran for
much help with statistical analysis, and Dr Sumanta Bagchi and two
anonymous referees for comments on the manuscript. We thank
Tenzin Thukten, Lobzang Chhupil, Tandup Chhering, Rinchen
Tobge, and Chhunit Kesang and other villagers of Kibber, Spiti for
their immense help with field work.
99
References
Andersen, R., & Saether, B. E. (1992). Functional response during winter of a herbivore, the moose, in relation to age and size. Ecology, 73, 542e550.
Barboza, P. S., Hartbauer, D. W., Hauer, W. E., & Blake, J. E. (2004). Polygynous mating
impairs body condition and homeostasis in male reindeer (Rangifer tarandus
tarandus). Journal of Comparative Physiology B, 174(4), 309e317.
Belovsky, G. E. (1981). Optimal activity times and habitat choice of moose. Oecologia, 48, 22e30.
Belovsky, G. E. (1986). Generalist herbivore foraging and its role in competitive
interactions. American Zoologist, 26, 51e69.
Bergman, C. M., Fryxell, J. M., Gates, C. C., & Fortin, D. (2001). Ungulate foraging
strategies: energy maximizing or time minimizing? Journal of Animal Ecology,
70(2), 289e300.
Clarke, A., Rothery, P., & Isaac, N. J. (2010). Scaling of basal metabolic rate with body
mass and temperature in mammals. Journal of Animal Ecology, 79(3), 610e619.
Clauss, M., Schwarm, A., Ortmann, S., Streich, W. J., & Hummel, J. (2007). A case of
non-scaling in mammalian physiology? Body size, digestive capacity, food
intake, and ingesta passage in mammalian herbivores. Comparative Biochemistry
and Physiology Part A: Molecular & Integrative Physiology, 148(2), 249e265.
Couturier, S., Côté, S. D., Huot, J., & Otto, R. D. (2008). Body-condition dynamics in a
northern ungulate gaining fat in winter. Canadian Journal of Zoology, 87(5),
367e378.
Dailey, T. V., & Hobbs, N. T. (1989). Travel in alpine terrain: energy expenditures for
locomotion by mountain goats and bighorn sheep. Canadian Journal of Zoology,
67, 2368e2375.
DelGiudice, G. D., Mech, L. D., Kunkel, K. E., Gese, E. M., & Seal, U. S. (1992).
Seasonal patterns of weight, hematology, and serum characteristics of freeranging female white-tailed deer in Minnesota. Canadian Journal of Zoology,
70(5), 974e983.
Forchhammer, M. C., & Boomsma, J. J. (1995). Foraging strategies and seasonal diet
optimization of muskoxen in West Greenland. Oecologia, 104, 169e180.
Fortin, D., Fryxell, J. M., & Pilote, R. (2002). The temporal scale of foraging decisions
in bison. Ecology, 83(4), 970e982.
Goodson, N., & Stevens, D. (1991). Effects of snow on foraging ecology and nutrition
of bighorn sheep. The Journal of Wildlife Management, 55, 214e222.
Gross, J. E., Hobbs, N. T., & Wunder, B. A. (1993). Independent variables for predicting intake rate of mammalian herbivores: biomass density, plant density, or
bite size? Oikos, 68, 75e81.
Hixon, M. A. (1982). Energy maximizers and time minimizers: theory and reality.
American Naturalist, 119, 596e599.
Illius, A., & Gordon, I. J. (1987). The allometry of food intake in grazing ruminants.
Journal of Animal Ecology, 56, 989e999.
Kie, J. G. (1996). The effects of cattle grazing on optimal foraging in mule deer
(Odocoileus hemionus). Forest Ecology and Management, 88, 131e138.
Kleiber, M. (1932). Body size and metabolism. Hilgardia, 6, 315e353.
Laca, E. A., Distel, R. A., Criggs, T. C., & Demment, M. W. (1994). Effects of canopy
structure on patch depression by grazers. Ecology, 75(3), 706e716.
Loe, L. E., Bonenfant, C., Mysterud, A., Severinsen, T., Øritsland, N. A., Langvatn, R.,
et al. (2007). Activity pattern of arctic reindeer in a predator-free environment:
no need to keep a daily rhythm. Oecologia, 152(4), 617e624.
Loison, A., Langvatn, R., & Solberg, E. J. (1999). Body mass and winter mortality in
red deer calves: disentangling sex and climate effects. Ecography, 22(1), 20e30.
Milner, J. M., van Beest, F. M., Solberg, E. J., & Storaas, T. (2013). Reproductive success
and failure: the role of winter body mass in reproductive allocation in Norwegian moose. Oecologia, 172(4), 995e1005.
Mishra, C. (2001). High-altitude survival: Conflicts between pastoralism and wildlife in
the Trans-Himalaya (Unpublished doctoral dissertation). Wageningen, The
Netherlands: Wageningen University.
Mishra, C., Prins, H. H., & Van Wieren, S. E. (2003). Diversity, risk mediation, and
change in a Trans-Himalayan agropastoral system. Human Ecology, 31(4), 595e
609.
Mishra, C., Van Wieren, S. E., Heitkönig, I., & Prins, H. H. (2002). A theoretical
analysis of competitive exclusion in a Trans-Himalayan large-herbivore
assemblage. Animal Conservation, 5(3), 251e258.
Mishra, C., Van Wieren, S. E., Ketner, P., Heitkönig, I., & Prins, H. H. (2004).
Competition between domestic livestock and wild bharal Pseudois nayaur in the
Indian Trans-Himalaya. Journal of Applied Ecology, 41(2), 344e354.
Monteith, K. L., Stephenson, T. R., Bleich, V. C., Conner, M. M., Pierce, B. M., &
Bowyer, R. T. (2013). Risk-sensitive allocation in seasonal dynamics of fat and
protein reserves in a long-lived mammal. Journal of Animal Ecology, 82(2),
377e388.
Murray, M. G. (1991). Maximizing energy retention in grazing ruminants. Journal of
Animal Ecology, 60, 1029e1045.
Mysterud, A. (1998). The relative roles of body size and feeding type on activity time
of temperate ruminants. Oecologia, 113, 442e446.
Owen-Smith, R. N. (1992). Megaherbivores: The influence of very large body size on
ecology. Cambridge, U.K.: Cambridge University Press.
Owen-Smith, N. (1994). Foraging responses of kudus to seasonal changes in food
resources: elasticity in constraints. Ecology, 75, 1050e1062.
Parker, K. L., Barboza, P. S., & Gillingham, M. P. (2009). Nutrition integrates environmental responses of ungulates. Functional Ecology, 23(1), 57e69.
Parker, K. L., Gillingham, M. P., Hanley, T. A., & Robbins, C. T. (1993). Seasonal patterns in body mass, body composition, and water transfer rates of free-ranging
100
M. Kohli et al. / Animal Behaviour 92 (2014) 93e100
and captive black-tailed deer (Odocoileus hemionus sitkensis) in Alaska. Canadian Journal of Zoology, 71(7), 1397e1404.
Pelletier, F., & Festa-Bianchet, M. (2004). Effects of body mass, age, dominance and
parasite load on foraging time of bighorn rams, Ovis canadensis. Behavioral
Ecology and Sociobiology, 56, 546e551.
Sabine, D. L., Morrison, S. F., Whitlaw, H. A., Ballard, W. B., Forbes, G. J., & Bowman, J.
(2002). Migration behavior of white-tailed deer under varying winter climate
regimes in New Brunswick. Journal of Wildlife Management, 66, 718e728.
Schaller, G. B. (1998). Wildlife of the Tibetan steppe. Chicago, IL: University of Chicago
Press.
Schmitz, O. J. (1990). Management implications of foraging theory: evaluating deer
supplemental feeding. Journal of Wildlife Management, 54, 522e532.
Schmitz, O. J. (1991). Thermal constraints and optimization of winter feeding and
habitat choice in white-tailed deer. Ecography, 14, 104e111.
Searle, K. R., Hobbs, N. T., & Shipley, L. A. (2005). Should I stay or should I go? Patch
departure decisions by herbivores at multiple scales. Oikos, 111(3), 417e424.
Shipley, L. A., Gross, J. E., Spalinger, D. E., Hobbs, N. T., & Wunder, B. A. (1994).
The scaling of intake rate in mammalian herbivores. American Naturalist, 143,
1055e1082.
St-Louis, A., & Côté, S. D. (2012). Foraging behaviour at multiple temporal scales in a
wild alpine equid. Oecologia, 169, 167e176.
Suryawanshi, K. R., Bhatnagar, Y. V., & Mishra, C. (2010). Why should a grazer
browse? Livestock impact on winter resource use by bharal Pseudois nayaur.
Oecologia, 162(2), 453e462.
Taillon, J., Sauvé, D. G., & Côté, S. D. (2006). The effects of decreasing winter diet
quality on foraging behavior and life-history traits of white-tailed deer fawns.
Journal of Wildlife Management, 70(5), 1445e1454.
Toupin, B., Huot, J., & Manseau, M. (1996). Effect of insect harassment on the
behaviour of the Riviere George caribou. Arctic, 49(4), 375e382.
Weckerly, F. W. (1998). Sexual-size dimorphism: influence of mass and mating
systems in the most dimorphic mammals. Journal of Mammalogy, 79, 33e52.
Appendix
Table A1
Results for the proportion of daily time spent foraging
Intercept
SiteR
Class2small
Class3female
Class4large
Day
Estimate
SE
t
Pr(>jtj)
0.7299
0.0680113
0.0583228
0.0807486
0.1198
0.00285
0.03154
0.02003
0.02711
0.02697
0.0277
0.00066
23.141
3.394
2.151
2.994
4.324
4.298
<0.001
0.001
0.033
0.003
<0.001
<0.001
Table A2
Results for the proportion of daily time spent resting
Intercept
SiteR
Class2small
Class3female
Class4large
Day
Estimate
SE
t
Pr(>jtj)
0.1834205
0.04939
0.04825
0.05735
0.11475
0.002466
0.03257
0.02069
0.02800
0.02785
0.02863
0.00068
5.631
2.387
1.723
2.059
4.008
3.598
<0.001
0.018
0.087
0.041
<0.001
<0.001