Activity budget and foraging behaviour of red squirrels

J . Zool., Lond. ( 1992) 227, 7 1-86

Activity budget and foraging behaviour of red squirrels (Sciuvus vulgaris) in coniferous and deciduous habitats L. WAUTERS*,C . SWINNEN A N D A. A. DHONDT University of Antwerp, Uniuersitaire Instelling Antwerpen, Department Biology, Universiteitsplein I , B-2610 Wilrijk, Belgium (Accepted 19 April 1991)

(With 4 figures in the text) Activity pattern, activity budget and foraging behaviour are compared between radiotagged squirrels in a coniferous and a deciduous habitat. Differences between habitats are explained in terms of differences in food availability, food choice and feeding rate.

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Activity rhythm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Activity budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Habitat use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Foraging behaviour and food-choice . . . . . . . . . . . . . . . . . . . . . . Rate of energy intake . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Page 71 72 74 74 77 79 81 83 83 85

86

Introduction

The seasonal changes in activity pattern of tree squirrels appear to result from a balance between resting and sleeping in the nest to conserve energy on the one hand, and foraging and feeding to obtain energy on the other (Tonkin, 1983; Reynolds, 1985; Gurnell, 1987; Wauters & Dhondt, 1987). Hence, although the 24-hour activity rhythm of red squirrels (Sciurus vulgaris Linnaeus 1758) appears to have an endogenous basis, as is the case in most other mammals (Aschoff, 1963), environmental factors, such as daylength, temperature and food availability, have a great effect on the annual variation in the time spent active per day (Saint-Girons, 1966; Pulliainen, 1973; Purroy & Rey, 1974; Zwahlen, 1975; Tonkin, 1983; Wauters & Dhondt, 1987). General information about food preference of red squirrels available from field observations (Shorten, 1962; Pulliainen, 1973; Tittensor, 1977; Moller, 1983; Tonkin, 1983; Wauters & Dhondt, 1987), or from stomach-contents analysis of shot animals (Degn, 1974; Tittensor, 1977; Gronwall, 1982), show that their foraging is

* Address for correspondence 71

0 1992 The Zoological Society of London

100

0. RIEPPEL

Kluge, A. G. (1962). Comparative osteology of the eublepharid lizard genus Coleonyx. J. Morph. 1 1 0 299-332. Mathur, J. K. & Goel, S. C. (1976). Patterns of chondrogenesis and calcification in the developing limb of the lizard, Calotes versicolor. J . Morph. 149 401-420. Miiller, G. B. & Alberch, P. (1990). Ontogeny of the limb skeleton in Alligator mississippiensis: developmental invariance and change in the evolution of archosaur limbs. J . Morph. 203 151-154. Peabody, F. E. (1951). The origin of the astragalus of reptiles. Evolution 5 339-344. Raynaud, A. (1985). Development of limbs and embryonic limb reduction. In Biology ofthe Reptiliu 15: 59-148. Cans, C. & Billett, F. (Eds). New York: John Wiley & Sons. Renous-Lecuru, S. (1973). Morphologie comparee du carpe chez les lepidosauriens actuels (rhynchocephales, lacertiliens, amphisbeniens). Gegenbaurs morph. Jb. 1 1 9 727-766. Rieppel, 0. (1984~).The upper temporal arcade of lizards: an ontogenetic problem. Rev. suisse Zool. 91: 475482. Rieppel, 0. (19846). The structure of the skull and jaw adductor musculature in the Gekkota, with comments on the phylogenetic relationships of the Xantusiidae (Reptilia: Lacertilia). Zool. J . Linn. Soc. 8 2 291-318. Rieppel, 0.(1987). The development of the trigeminal jaw adductor musculature and associated skull elements in the lizard Podarcis sicula. J. Zool., Lond. 212 13 1-150. Rieppel, 0. (1988). The development of the trigeminal jaw adductor musculature in the grass snake Nutrix nutrix. J . Zool., Lond. 216 743-770. Rieppel, 0.(1989~).Helveticosuurus zollingeri Peyer (Reptilia, Diapsida): Skeletal paedomorphosis, functional anatomy and systematic affinities. Palueontographicu ( A ) 208 (4-6): 123-152. Rieppel, 0. (1989b). A new pachypleurosaur (Reptilia: Sauropterygia) from the Middle Triassic of Monte San Giorgio, Switzerland. Phil. Trans. R. Soc. [B) 323: 1-73. Rieppel, 0.(1990). The structure and development of the jaw adductor musculature in the turtle Chelydruserpentina. Zool. J . Linn. SOC.98: 27-62. Robinson, P. L. (1975). The functions of the hooked fifth metatarsal in lepidosaurian reptiles. Coll. int. C.N.R.S. 218: 461483. Romer, A. S . (1956). Osteology of the reptiles. Chicago: The University of Chicago Press. Russell, A. P. (1975). A contribution to the functional analysis of the foot of the tockay, Gekko gecko (Reptilia: Gekkonidae). J . Zool., Lond. 176 437-476. Sander, P. M. (1989). The pachypleurosaurids (Reptilia: Nothosauria) from the Middle Triassic of Monte San Giorgio (Switzerland) with the description of a new species. Phil. Trans. R. SOC.( B ) 325 561-670. Sewertzoff, A. N. (1908). Studien iiber die Entwicklung der Muskeln, Nerven und des Skeletts der Extremitaten der niederen Tetrapoda. Bull. Soc. Imp. Nut. Moscou (N.S.) 21: 1430. Shubin, N. H. & Alberch, P. (1986). A morphogenetic approach to the origin and basic organization of the tetrapod limb. Evol. B i d . 2 0 319-387. Smith, M. M. & Hall, B. K. (1990). Development and evolutionary origins of vertebrate skeletogenic and odontogenic tissues. Biol. Rev. 65: 277-373.

FORAGING BEHAVIOUR OF RED SQUIRRELS

13

(b) Behaviour: behaviour was assigned to one of the following categories: 1. Foraging, including searching, handling and recovering stored food; 2. Food-caching behaviour; 3. Travelling, not related to foraging; 4. Interactions with other squirrels; two or more squirrels foraging in close proximity ( < 2 m) without any signs of aggression was regarded as a non-aggressive interaction; 5. Grooming; 6. Resting outside the drey; 7. Mating-chase; 8. Other behaviour; 9. Not active, when resting o r sleeping in the nest. (c) Feeding time: in case a squirrel was foraging it was recorded how long it spent feeding on one food item from the moment it found the food item until it dropped the remains. (d) Position: all locations were mapped using X, Y coordinates. Mean monthly maximum temperatures and daylength are derived from the nearest station of the Koninklijk Metereologisch Instituut at Gee1 (1 5 km to Herenthout).

Data analysis Although data were recorded on a one-minute basis, the time-unit was further subdivided when necessary. Therefore data could be treated as if monitored continuously.

Foraging place The time spent on each tree species or on the ground was expressed as a percentage of the total hours of active behaviour observed in each month (Wauters & Dhondt, 1987). An Ivlev’selectivity index (Krebs, 1989) was calculated in order to investigate preference or avoidance of various tree species in different seasons (Table I).

Activity The daily activity pattern of each month was deduced from all data collected that month using both radiotracking methods. A detailed description of the calculation of the mean number of hours of activity per day is given elsewhere (Wauters & Dhondt, 1987). As regards the activity budget of the squirrel, the time allocated to each of the behavioural categories was expressed as a percentage of the total time spent active each month (Wauters & Dhondt, 1987).

Foraging behaviour Since foraging was quantitatively the most important type of behaviour, it was examined more closely. The percentage of time spent searching for and handling different food items was estimated on a monthly basis, SO that changes in the diet could be detected (Wauters, 1986; Wauters & Dhondt, 1987).

Estimating energy intake Of Scots pine 50 cones were taken, covering the whole range of cone size. The cones were cut open and all viable seeds were extracted. The mean number of seeds per cone was 27.7k7.8. Since squirrels are able to extract 89% of all seeds (pers. obs.), an average of 25 seeds are consumed per cone eaten. One hundred viable seeds of Scots pine, oak and beech were dried for 48 hours at 80°C. Average dry weight per seed was calculated for each seed species (Table 11) and for the amount of nutritive material. The weight of the nutritive material per seed was multiplied by the caloric value of the seeds (Grodzinski & Sawicka-Kapusta, 1970), to give a measure of energy content per seed (Table 11). The mean feeding time (n= 50) per seed species was estimated by observing feeding behaviour and hence the amount of energy intake per unit time could be separately calculated for each seed species (Table 11).

L. WAUTERS, C . S W I N N E N A N D A . A. D H O N D T

74

TABLE I1 Dry weight of’total seed, of the nutrltioe material, caloric value of seeds (after Grodzinskr & Suwicka-Kupusta, 1970). energy content per seed, mean handling time and energy intake per time unit for different seedspeciest pine cone, beechnut, acorn

Seed Weight Nutritive Energy Energy per Feeding Food intake species (g) material (g) (kJ/g) seed (kJ) time (sec) (kJjmin) Pine Beech Acorn

0.180

0.186 1.613

0.131 0.124 1.379

27.03 29.13 18.64

3.54 3.61 25.70

120+37 33+7 248k43

1.77 6.56 6.22

To calculate energy intake per day in winter, 4 different squirrels were observed for their whole active period, on different days, in both study areas (December 1986). The total time spent feeding on various seed species was noted. Combining these data with the average energy intake per unit of time for different seed species allowed us to calculate an average energy intake per day and per hour of activity, in both study areas.

Results Actiuity rhythm

In both habitats the activity rhythm varied considerably over the year (Fig. 1). Squirrels were exclusively day-active, and the start of activity was closely related to sunrise (Fig. 1). From December to February the number of hours spent active per day was less than six (Table 111). In the coniferous habitat activity peaked during the first hours after sunrise, and squirrels returned to their drey as the afternoon progressed (Fig. 1). Each individual squirrel had only one active period per day. In the deciduous area the overall activity pattern combined for all animals did not show any clear pattern (Fig. 1) due to the start of activity being unsynchronized for different squirrels. When, however, an individual squirrel was considered on any day in the December-February period, the activity pattern was always bimodal, with a first peak soon after sunrise, a resting period after about two and a half hours of activity and a second peak early afternoon (Fig. 2). In March the activity rhythm in the coniferous habitat changed from a unimodal to a bimodal pattern. In the deciduous habitat squirrels had two active periods per day during the whole month (Fig. 1). From April to November the activity pattern was bimodal in both habitats. As spring progressed, the number of hours spent active per day increased (Table 111). The longest active periods were recorded between May and September, squirrels spending 8-1 1 hours outside the nest. In the period May-August the percentage of afternoon activity was generally higher than in other months (Table 111). The length of the active period decreased again during the autumn, with decreasing daylength. There was no difference in the length of the active period between years, when the number of hours spent active per month was compared (Wilcoxon matched pairs signed-ranks test: coniferous 1985-1986: N=7; T = 10; N.S.; deciduous 1986-1987: N = 12; T=22; N.S.). In a multiple regression analysis daylength and temperature were significantly correlated to the duration of the active period and monthly variation in daylength and temperature together explained 90% of the variation in time spent active (coniferous area R2= 0.895; deciduous area

0

4

r-

z

r'

&I

_----

0

4

0 00

2

0 00

---__

-P

P

___---

-h-k

4

0 0 0 0 00 w

4

0 0 0

3

u

-A

0 0 0

4

go

A

0 0 0

K

-F----- IT

b-

O 0 0

-I=+--

0 0

4

0

A

0 00

Percentage activity 2

0 8 0

P

;3 P

L. WAUTERS, C . S W I N N E N A N D A. A . D H O N D T

76

TABLE 111 Squirrel actitity in coniferous and deciduous habitat in relation to daylength and maximum temperature. Mean time spent active per day per squirrel ( T ) , percentage of daylength spent actire and afternoon activity as a percentage of rotal daily activity

Number of hours active per day Coniferous Month

Temp. Daylength (“C)

T

%,

afternoon (YO)

Deciduous

T

‘%

afternoon (%I)

-

1985 June July Aug. Sept. Oct. Nov. Dec.

I6h28 16h06 14h37 12h41 IOh48 9h00 7h58

18-2 22-5 21.0 19.1 14.4

58.5 62.0 66.0 61.7 59.0 71.9 61.7

27.3 48.3 37.8 46.0 40.5 36.0 36.5

1986 Jan. Feh. March April May June July Aug. Sept. Oct. Nov. Dec.

8h26 9h58 1 1h48 133147 15h29 16h28 16h06 143137 12h41 10h48 9h00 7h58

5.0 4h43 56.0 -0.2 4h50 48.4 9.0 7h29 63.4 10.2 8h28 61.5 18.9 10h06 65.2 22.3 10h54 64.7 21.9 8h51 53.0 20.7 93145 65-8 16.3 9h02 72.7 16.0 6hS1 61.1 15.2 5h56 61.7 7.4 5h00 66-0

37.5 28.4 31.4 36.4 40.1 41.7 47.6 43-4 37.5 35.2 36.7 26.2

3h56 4h19 5h40 6h44 9h07 llh00 9h00 81128 7h57 7h34 6h14 4h31

43.1 43.7 47.0 47.8 59.3 67.6 56.0 57.6 61.0 70.0 68.2 56.9

32.6 13.4 22.0 20.3 44.1 40.3 33.8 38.5 32.2 42.2 39.4 33.2

1987 Jan. Feh. March April May June July Aug. Sept. Oct. Nov. Dec.

8h26 91158 1 1h48 13h47 I5h29 16h28 16h06 14h37 12h41 1 Oh48 9h00 7hS8

-0.8 5h21 4.9 5h59 6.4 7h53 16.1 8h41 14.5 lOhl0 18.1 21.7 20.5 19.5 15.2 8.5 6.6

38.4 23.4 32.5 38.9 45.2

5h08 5h26 5h31 7h28 9h01 9h58 9h16 8h50 8hll 73152 5h20 4h41

61.5 54.9 46.3 54.0 58.9 58.8 57.0 59.2 56.4

29.3 28.1 27.7 42.7 46.9 43.0 38.9 41.2 31.4 40.6 27. I 46.0

5.5

7.9

91138 9h59 9h39 73149 6h22 6h28 4h55

63.1 58.4 65.6 62.4 65.7

71.8

57.8 58.2

R2=0.899). Hence squirrels had a longer activity period when days were longer or when temperature was higher. In the coniferous habitat, during part of the year, the active period per month was longer than in the deciduous habitat (Wilcoxon matched-pairs signed-ranks test: N = 17; T = 15; PiO.01; twotailed). In June-July and October-November differences between both areas were very small (Table 111). In the other months squirrels in the coniferous habitat were active for half an hour to two hours (mean lh05) longer than those in the deciduous habitat. There was no significant difference in the start of activity (Wilcoxon test, 1986 data: N = 11; T=21; N.S.), but squirrels in

427 405

596

(b)- 317 - ' 101

-rn

=

44 1 887

I

8

10

I

12

14

16

Time (h) FIG.2. Activity pattern of individual squirrels (for each squirrel activity pattern on a single day) in (a) coniferous and (b) deciduous habitat in December 1986. Shaded bars indicate active periods.

the coniferous woodland ended their activity later than those in the deciduous woodland (Wilcoxon test, 1986 data: N = 12; T = 8; P < 0.02; two-tailed). Actiuity budget In both habitats the major proportion of active time was spent foraging, between 68% and 8 1YO of total activity in the coniferous woodland, and between 54% and 77Oh in the deciduous woodland (Table IV). In the coniferous area there was a difference in the proportion of foraging time between spring-summer (April-September) and autumn-winter (October-March), showing a tendency towards relatively more time spent foraging in the latter period (Table V). In the deciduous area the seasonal difference was not significant (Table V). In autumn-winter squirrels spent relatively more time foraging in the coniferous than in the deciduous habitat (Wilcoxon test: N = 12; T = 12; P