— 381 —

VARIATION IN HOME RANGE SIZE OF FEMALE FALLOW DEERINHABITING A SUB-MEDITERRANEAN HABITAT

Simone CIUTI, Sara DAVINI, Siriano LUCCARINI & Marco APOLLONIO1

RÉSUMÉ

De 1997 à 2000, 13 femelles de Daim (Dama dama) ont été suivies par télémétrie dans laréserve naturelle de San Rossore (située en Toscane, Italie centrale) afin de déterminer lasuperficie et le mode d’occupation de leur domaine vital. La superficie annuelle moyenne dudomaine vital a été estimée à 532,9 ± 226,8 ha par la méthode du Minimum Convex Polygon(95 %) et à 426,8 ± 151,2 ha par la méthode dite de Kernel (95 %). La surface de ces domai-nes vitaux est plus vaste que celles signalées précédemment dans la littérature. L’environne-ment de San Rossore est caractérisé par une distribution fragmentée des ressources trophiqueset ceci pourrait favoriser l’étendue des domaines vitaux. La surface moyenne du domaine vitaldiffère considérablement selon les saisons. Elle est réduite en été à 140,9 ± 112,5 ha(MCP 95 %) ou 126,0 ± 102,4 ha (Kernel 95 %). En automne et au printemps, elle double,s’élevant respectivement à 348,5 ± 170,5 ha et 304,0 ± 175,8 ha (MCP 95 %), ou380,5 ± 178,1 ha et 292,9 ± 177,3 ha (Kernel 95 %). Le domaine vital hivernal est sensible-ment plus large que l’estival, couvrant en moyenne 184,5 ± 119,8 ha (MCP 95 %), ou195,2 ± 122,1 ha (Kernel 95 %). Cette étude démontre que des facteurs écologiques et la pré-sence de jeunes réduisent la surface du domaine vital pendant les mois de mai à août. Enrevanche, la distribution de la nourriture, le temps du rut en automne et la floraison au prin-temps déterminent une augmentation de la surface du domaine vital.

SUMMARY

From 1997 to 2000 13 female Fallow Deer (Dama dama) were monitored using teleme-try in the San Rossore Estate (Tuscany, central Italy) to determine home range size and pat-terns of range development. Mean annual home range size was estimated using the MinimumConvex Polygon method (95%) over 532.9 ± 226.8 ha and the Kernel method (95%) over426.8 ± 151.2 ha. These home range sizes are larger than those previously reported in the lit-erature: San Rossore is an environment characterized by a fragmented distribution of trophicresources and this may lead to an increase in range size. The average size of home ranges dif-fered considerably between seasons, the smallest ranges being found during summer, averag-ing 140.9 ± 112.5 ha (MCP 95%) and 126.0 ± 102.4 ha (Kernel 95%). The autumn and springranges amounted to twice as much, with an average size of 348.5 ± 170.5 ha and304.0 ± 175.8 ha, respectively, using MCP 95%, and 380.5 ± 178.1 ha and 292.9 ± 177.3 ha,

1 Dipartimento di Zoologia ed Antropologia Biologica, Università degli Studi di Sassari, Via Muroni25, I-07100 Sassari (Italy). Fax: +39079-228665. E-mail address: marcoapo@uniss.it

Rev. Écol. (Terre Vie), vol. 58, 2003.

brought to you by COREView metadata, citation and similar papers at core.ac.uk

provided by I-Revues

— 382 —

using Kernel 95%. Winter ranges were slightly larger than summer ones averaging184.5 ± 119.8 ha (MCP 95%), and 195.2 ± 122.1 ha (Kernel 95%). This study showed thatecological factors and the presence of young reduce home range size during the May-Augustperiod. Food distribution and the rutting season during autumn, and green-up during spring,caused an increase in home range size.

INTRODUCTION

Home range size evaluation has an important role in ecology and allows theunderstanding of the patterns of space use and spatial behaviour of a species. It hasbeen found to be related to physiological, ecological and social factors, that eitheract separately or in conjunction with each other. Habitat structure and distributionand quality of food exert an important influence on the extent of home range. Forexample, home range size of female Roe Deer (Capreolus capreolus) increases withthe visibility in the home range, supporting the hypothesis that cover is importantin reducing the risk of predation and thereby increasing adult survival (Tufto et al.,1996). In addition, females Roe Deer adjust the size of their home range in responseto decreasing food supply (Tufto et al., 1996). As found by Catt & Staines (1987)for Red Deer (Cervus elaphus), home range size is smaller for animals with highproportion of favourable habitats in their range. Among cervids in the temperate cli-mates of the northern hemisphere metabolic rate and food intake decline duringwinter which is associated with weight loss (Clutton-Brock et al., 1982). In theseareas deer conserve energy during winter by reducing activity and movementswithin a restricted home range (Craighead et al., 1973; Moen, 1976; Georgii, 1980;Cederlund, 1981; Lieb, 1981a; Lieb, 1981b; Clutton-Brock et al., 1982; Georgii &Schröder, 1983). The presence of snow especially reduces activity and home range(Rongstad & Tester, 1969; Drolet, 1976; Georgii, 1980; Cederlund, 1982; Georgii& Schröder, 1983). The present study was performed in a Mediterranean ecosystemthat shows a different seasonality in food availability and distribution of resourcesthroughout the year: therefore, we should expect important differences in annualand seasonal home range size with respect to deer populations in temperate clima-tes. Whereas food shortages are to be found during winter months in more tempe-rate climates, in the Mediterranean environment it is the hot and dry summer thatposes this problem. Winters, on the other hand, are characterized by a relativelygreater food supply, as they tend to be wet and cool. A summer restriction in foodavailability is responsible for the differences in body mass between Wild Boars (Susscrofa) from Mediterranean areas versus those from Atlantic and continental areas(Spitz et al., 1998). The size of home range is also related to the combination ofother parameters such as age (Bideau et al., 1983; Cederlund & Sand, 1992), matingsystems (Clutton-Brock & Harvey, 1978), population density (Vincent et al., 1995),predation and human disturbance (Jeppesen, 1987; Van Dyke & Klein, 1996). Asthese parameters vary according to habitat and season of the year, so usually doesthe home range size of a species. The aim of the present paper was to analyze theuse of space by a number of radio-tagged female Fallow Deer (Dama dama) in asub-Mediterranean environment to understand the area used throughout the year,also considering the lack of field studies on this species. The European Fallow Deerhas presently a wide distribution throughout much of Europe, North and SouthAmerica, and Oceania (Chapman & Chapman, 1997). However, despite its exten-sive geographical distribution, broadly larger than those reported for the other twowestern european cervid species, Red Deer and Roe Deer, very few studies on the

— 383 —

ecology of this species were carried out. Little work has been published on homerange size of female Fallow Deer: to date, studies have dealt with British popula-tions (Putman, 1986; Chapman & Chapman, 1997) and a New Zealand population(Nugent, 1994), but have remained scarce, in general. Some works done in theMediterranean basin have investigated diet (Bruno & Apollonio, 1991) and habitatselection (Braza, 1975; Braza & Alvarez, 1987; Apollonio et al., 1998) of this spe-cies, but do not include data about spatial use in Fallow Deer in a Mediterraneanand sub-Mediterranean habitat, i.e. the original habitat of this species (Chapman &Chapman, 1997).

STUDY AREA AND POPULATION

The study was performed in the San Rossore Estate in central Italy (43˚43’ N;10˚ 19’ E). The Preserve is an area of 4,596 ha bounded by the Tyrrhenian Sea tothe West, the Serchio and Arno rivers to the North and South, and a fence to theEast. It is characterized by a sub-Mediterranean climate and a plain orography. Theground is mainly sandy.

Its vegetation is composed of pine woods (1,736 ha of Pinus pinea, 285 ha ofPinus pinaster), deciduous woods (1,153 ha of Fraxinus spp., Populus alba, Popu-lus canescens, Alnus glutinosa and Quercus spp.), marshes and meadows.There arealso over 700 ha of fenced agricultural land. Ten different types of vegetationalcover are present at San Rossore, indicating a heterogeneous environment. Thisdiversity is confirmed by a Shannon index value of 2.69 (max. possiblevalue = 3.32). Some cover types are more represented, especially mixed deciduousforest and domestic pine, two habitats fragmented throughout the study area, as areothers. The high overall number of fragments for all cover types (137) gives furthersupport to the observation of a non-uniform distribution of resources in this envi-ronment. The average annual temperature was 15.8 ˚C (± 6.4 ˚C). July-August wasthe hottest bimonth (mean = 25.0 ˚C) and January-February the coldest one(mean = 8.2 ˚C). Most precipitation fell from September to April (averaging306.2 mm, 181.2 mm, and 152.9 mm in autumn, winter, and spring, respectively),while lower values were recorded during summer (averaging 96.8 mm). Apart fromFallow Deer, Wild Boars are the only wild ungulates in the area. Large predators areabsent. Only a small section of the study area is open to the public at week-ends. TheFallow Deer population probably originated from the XVI th century, and certainlyfrom the beginning of the XVIII th century (Simoni, 1910). Population size was esti-mated with spring censuses. The census took place in the first 10 days of April whendeer were counted from high seats in pasture areas and counts were repeated fourtimes at dawn and morning. Deer density within the study area was about 26 deer/100 ha. In San Rossore, the Fallow Deer population adopts a mixed mating system:some males defend a territory on a lek, others a single, isolated territory.

MATERIAL AND METHODS

Thirteen female Fallow Deer were studied by discontinuous radio-tracking(Swihart & Slade, 1985; Harris et al., 1990) from April 1997 to March 2000.

— 384 —

Because some died or ceased transmitting during the study, 9 were monitored for24 months, 7 for 32 months and 5 for 36 months.

During the yearly deer control operations on the Estate, deer were caught incircular pens into which they were driven by game wardens. Selected deer were cor-nered in a narrow raceway, restrained, and blindfolded. Each animal was then agedby tooth eruption and wear (Lowe, 1967; Chapman & Chapman, 1997), weighed,and measured. Animals were then fitted with radiocollars and released. All trans-mitter frequencies were on the 151 MHz band. Collars included an activity monitor.The animal was considered to be inactive if no change was heard in the pulse rateafter 1 min of clear continuous reception. Televilt transmitters with a battery in anepoxy resin and attached to PVC-plastic collars in different colours were used. Theassembly weighed 400 g and was adjustable to the animals’ neck size. TeleviltRX 8910 HE receivers and four element hand-held Yagi antenna were used. Loca-tions were calculated by triangulation using bearings obtained from three differentreference points (Springer, 1979; White & Garrot, 1990) employing the “loudestsignal” technique (Springer, 1979), and plotted onto a 1: 10,000 digitized scale mapof the study area (Springer, 1979; Kenward, 1987). Each month, twelve locationsper animal were made, uniformly distributed over the 24 hours of the day (one fixfor each of the twelve time bands). Additional fixes were collected randomly. Addi-tional data, such as date, time, weather conditions, and visual sightings were recor-ded. Accuracy of fixes was determined in the field by placing test trasmitters invarious habitats and taking fixes on these (Harris et al., 1990). The error box wasno more than 50

× 50 m (0.25 ha) when located from < 0.5 km, which was the moreusual distance at which we collected data. The minimum time interval between suc-cessive locations was 12 hours to allow any radio-tagged deer to cross its homerange several times. A special session of continuous telemetry was organized toobtain an independent time interval that was set at 4 hours (Schoener, 1981; Swihart& Slade, 1985). Therefore, our fixes can be considered biologically (Lair, 1987)and statistically (Swihart & Slade, 1985) independent. Using Ranges V software(Kenward & Hodder, 1996), both the Minimum Convex Polygon (MCP: Mohr,1947; Southwood, 1966; Kenward, 1990) and the Kernel (Worton, 1989) methodswere used to calculate home range size (using 95% of available locations). Homerange overlaps (using MCP 95% polygons) were calculated between seasons usingthe same software; overlaps were also considered for annual and bimonthly ranges.Home ranges were determined on an annual, seasonal, and bimonthly basis.Seasons were defined as follows: winter (Dec.-Feb.), spring (Mar.-May), summer(June-Aug.) and autumn (Sept.-Nov.): because the monitored period started fromApril 1997, summer 1997 was the first season available. Bimonths were defined asfollows: Mar.-Apr., May-June, July-Aug., Sept.-Oct., Nov.-Dec. and Jan.-Feb.Centres of activity were calculated using the Kernel method (Kenward & Hodder,1996). Site fidelity, i.e. the tendency of animals to remain within the same area foran extended period (White & Garrot, 1990), was evaluated for each individual bymeasuring the distances between the centres of activity in successive annual, seaso-nal, and bimonthly periods. Kernel and MCP data are presented in all instances: theMCP method has been widely used in other studies and is therefore likely to be use-ful for making comparisons between studies (Harris et al., 1990). The SPSS 8.0 pro-gram (SPSS inc. 1989-1997) was used for statistical analysis. We first analyzeddata using General Linear Model analysis (GLM) with home range size as thedependent variable, year (3-modalities), and season (4-modalities) as the two fixedfactors and individual considered as a random effect. The same procedure wasadopted with year and bimonth (6-modalities) as the two factors. Consequently, we

— 385 —

used Bonferroni post hoc multiple comparisons test to check for differencesbetween different seasons (and bimonths). Friedman tests were used to test varia-tion of annual home ranges sizes and of distances between centres of activity. In alltests significance was set at p

≤ 0.05.

RESULTS

ANNUAL HOME RANGES

A total of 23 annual home range sizes was estimated (9 females in 1997-1998,9 in 1998-1999, and 5 in 1999-2000). In 1997-1998, the average size of the annualhome range was 582.88 ± 260.45 ha (MCP 95%) and 435.02 ± 155.02 ha (Kernel95%). For 1998-1999, the doe’s mean annual home range measured498.28 ± 210.57 (MCP 95%) and 412.31 ± 164.7 (Kernel 95%). For 1999-2000, themean annual home range was 505.68 ± 222.10 ha (MCP 95%) and438.33 ± 150.67 ha (Kernel 95%). Annual home ranges were not significantly dif-ferent throughout the three years (Friedman test: n = 5, df = 2, Chi-square = 1.60,and p = 0.449 using MCP; n = 5, df = 2, Chi-square = 2.80, and p = 0.247 usingKernel). The annual home range of each individual was overlapped on each of theother two years to determine the degree of range fidelity shown by the females.Mean overlaps between annual home ranges proved to be 81.9 ± 14.7%.

The mean distance between annual centres of activity measured547.6 ± 956.1 m. This value was influenced by one female which moved to a diffe-rent area at the end of the first year of study (mean = 254.5 ± 266.60 m without thisfemale). No significant differences were detected in these distances throughout thethree years (Friedman test: n = 5, df = 1, Chi-square = 0.200, p = 0.655).

SEASONAL HOME RANGES

The size of seasonal home ranges varied throughout the year, as analyses ofboth the whole sample of animals and the intra-individual comparison of the seaso-nal home ranges showed (Fig. 1). The smallest ranges were found during summer(n = 28), averaging 140.91 ± 112.55 ha (MCP 95%) and 126.04 ± 102.40 ha (Ker-nel 95%). The autumn (n = 26) and spring (n = 18) ranges measured twice as much,with an average size of 348.58 ± 170.25 ha and 304.04 ± 175.85 ha respectively,using MCP 95%, and 380.51 ± 178.19 ha and 292.94 ± 177.39 ha, using Kernel95%. Winter ranges were larger than summer ones (n = 25; 184.54 ± 119.85 haMCP 95%, 195.27 ± 122.15 ha Kernel 95%). The model derived from the GLManalysis showed a significant effect of factors considered and their interactions: inparticular significant effects were found for seasons (F = 7.669, df = 3, p = 0.001using MCP 95%; F = 9.423, df = 3, P = 0.000 using Kernel 95%).

The size differences between the four types of seasonal home ranges werehighly significant: autumn home ranges proved larger than summer ranges (Bonfer-roni test: p = 0.000 using MCP 95%, p = 0.000 using Kernel 95%) and winter ran-ges (Bonferroni test: p = 0.001 using MCP 95%, p = 0.000 using Kernel 95%),

— 386 —

while spring home ranges were larger than summer ones (Bonferroni test: p = 0.002using MCP 95%, p = 0.002 using Kernel 95%).

The overlap of each seasonal home range on the other three was calculated todetermine whether the areas occupied by females do vary throughout the year(Fig. 2). Low values of overlap were obtained for every seasonal range on summerones (spring 37.5 ± 25.2%; autumn 27.4 ± 25.4%; winter 33.2 ± 31.0%) becauseareas used by females in summer were different and smaller than those used in otherseasons.

The distances between seasonal centres of activity were calculated, the greaterones being found between summer and autumn (1854.1 ± 1229.4 m) and betweenwinter and summer (1772.1 ± 1291.2 m). On the contrary, the smallest distancesbetween centres of activity were found between autumn and winter centres(889.8 ± 1084.0 m).

The seasonal home ranges of each individual were overlapped on the sameseason of different years and high values of overlap were recorded. Mean overlapsbetween spring home ranges measured 67.7 ± 15.9%, between summer home ran-ges 56.3 ± 31.3%, between autumn home ranges 67.6 ± 20.0%, and between winter

Figure 1. — Seasonal home ranges of female Fallow Deer.

— 387 —

61.9 ± 25.2%. Distances between the centre of activity of the same season of eachindividual throughout the three years were confronted using Friedman test (sum-mer: n = 7, df = 2, Chi-square = 2.571, p = 0.276; autumn: n = 7, df = 2, Chi-square = 0.963, p = 0.618; winter: n = 6, df = 2, Chi-square = 1.333, p = 0.513),confirming that these distances did not vary during the monitored period: the lowvalues obtained, especially for subsequent summers and winters, suggest that theseranges were quite close to one another (summer 495.6 ± 445.2 m; winter608.6 ± 782.8 m).

Figure 2. — Seasonal home range overlaps (female Fallow Deer n. 015).

— 388 —

BIMONTHLY HOME RANGES

The model derived from the GLM analysis showed a significant effect of fac-tors considered and their interactions: especially significant effects were found forbimonths (F = 8.089, df = 5, p = 0.000 using MCP 95%; F = 8.918, df = 5,p = 0.000 using Kernel 95%). The smallest ranges were found during July-Augustand May-June, while the largest ones were during September-October (Figs. 3-4).September-October ranges proved significantly larger than those of May-June(Bonferroni test: p = 0.000 using MCP; p = 0.000 using Kernel), July-August(Bonferroni test: p = 0.000 using MCP; p = 0.000 using Kernel), November-December (Bonferroni test: p = 0.012 using MCP, p = 0.049 using Kernel),January-February (Bonferroni test: p = 0.000 using MCP; p = 0.000 using Kernel),and March-April (Bonferroni test: p = 0.005 only using Kernel). On the otherhand, July-August ranges were smaller than those of November-December (Bon-ferroni test: p = 0.013 only using MCP), and March-April (Bonferroni test:p = 0.011 only using MCP).

Considering the overlap of each bimonthly home range on the other five, aclearer difference was found between areas used by females in July-August andMay-June than those used in the other four bimonths. July-August and May-Juneranges shared more than half of their areas (60.7 ± 25.0%), while their degree ofoverlap with the others four bimonths proved lower (28.7 ± 27.9% and27.6 ± 30.5%, respectively).

The analysis of the distances between bimonthly centres of activity confirmedthe previous observations: that is, May-June and July-August ranges both proved tobe far away from the other bimonthly centres of activity (1798.2 ± 1158.7 m and1637.1 ± 1284.4 m, respectively). On the other hand, low values were recordedbetween May-June and July-August (588.4 ± 734.1 m) and between November-December and January-February (472.6 ± 623.4 m).

For each deer, bimonthly home ranges of different years were overlapped. Thelargest mean overlaps between home ranges were found in May-June, averaging64.9 ± 24.5%. However, all the other overlaps were above the value of 50%.Bimonthly site fidelity was confirmed by the small distance between centres of acti-vity (< 1 km), especially for May-June (456.2 ± 296.2 m) and July-August(659.6 ± 485.8 m), the ranges showing smaller values.

DISCUSSION

The annual home range sizes in female Fallow Deer at San Rossore are largerthan those reported in the literature. Nugent (1994) reported mean annual homeranges of 117.24 ha (8 adult females; MCP 100%) for the Blue Mountains of NewZealand, but this value is influenced by one female which moved to a completelydifferent area following excessive disturbance, greatly increasing the mean of thesample (mean = 58.7 ha without this female). Home range sizes for female FallowDeer in the New Forest, England, calculated over a six-month period from Mayto October varied in size from 48 to 89 ha (MCP 100%), with a mean value of69.2 ha (Rand in Putman, 1986). Chapman & Chapman (1997) estimated mean

— 389 —

Figure 3. — Cumulative bimonthly home ranges of female Fallow Deer. Box-plots represent theinterquartile range, which contains the 50% of values. The whiskers represent the highest and lowestvalues containing the 80% of values, excluding outliers (o) and extreme values (+). The line across the

box indicates the median.

— 390 —

ranges slightly greater than 40 ha in the South Weald, England. The annual homerange sizes in female Fallow Deer at San Rossore are more comparable to thoserecorded for female Fallow Deer in the Maremma Regional Park (Tuscany, Italy),also a Mediterranean environment, averaging 407.0 ± 169.3 ha (MCP 95%; Luc-carini & Apollonio, unpublished data). For Red Deer, Staines (1974) observedthat differences in home range sizes within a species are generally due to diffe-rences in habitat structure with respect to food and shelter distribution. Thiswould appear to be true in Fallow Deer as well. When food supply is uniformlydistributed in the environment, home range sizes in cervid species shoulddecrease (Clutton-Brock & Harvey, 1978). San Rossore, however, is a diverseenvironment characterized by a fragmented distribution of trophic resources, andthis leads to an increase in range size, as individuals must travel from one patchto the next. Range sizes recorded at San Rossore are more comparable to thosefound by Catt & Staines (1987) in female Red Deer in Glenbranter Forest, whereranges varied from 406 to 1,008 ha. They are, however, larger than those repor-ted, again for Red Deer by Carranza (1991) in Monfrague, Spain(mean = 258.4 ha), Clutton-Brock et al. (1982) on the island of Rum(mean = 200 ha) and Jeppesen (1987) at Oksbøl State Forest, Denmark(mean = 257 ha). However, those studies refer to deer characterized by a greaterbody weight and living in different habitats. As Harestad & Bunnell (1979) sug-gested, size of home range among herbivorous mammals may increase with bodyweight, confirming McNab’s (1963) findings. Therefore, even if it is well-esta-

Figure 4. — Bimonthly home ranges of female Fallow Deer in each year.

— 391 —

blished that individuals of large species have larger home range than those ofsmall species (Peters, 1983), home range sizes in female Fallow Deer in a sub-Mediterranean environment resulted larger than those recorded for deer characte-rized by a greater body weight, but living in rich habitats.

The average size of home ranges recorded in San Rossore differed considera-bly between seasons, and showed a different pattern than those reported in the lite-rature. Home range sizes reported in the Blue Mountains, New Zealand (Nugent,1994), had this trend in size variation: summer < autumn < winter < spring. FallowDeer at Salzau, Germany, showed greater movements between May and December,when they preferred forest and meadows, while they reduced their home rangesbetween January and April, when they used only forest because meadows werecovered by snow (Heidemann, 1973). In San Rossore, annual home range sizeswere obviously influenced by the large autumn and spring home ranges. Duringautumn, especially in September-October, habitat structure and distribution of foodsupply lead to an enlargement of the home range. The rainfalls, lacking during sum-mer, promote growth of grass in the pasture areas present in the eastern and westernparts of the Preserve (Apollonio et al., 1998). Females were observed and monito-red while travelling to the eastern pasture at dusk, where they remained for mostpart of the night to feed. At dawn, they returned further West, some even reachingthe western pasture. This movement was probably influenced by food distributionand day-time human disturbance, which is restricted to the eastern side of the Pre-serve. The fall home range size enlargement was also influenced by onset of thebreeding season, when males concentrated in the general proximity of one lek andfemale movements were influenced as well. The lek is located on the western sideof San Rossore, far from the monitored female home ranges. While males occupythis area from September to November, females reach it only during the oestrusdays.

Winter home range reduction in female Fallow Deer in our study area appearedmainly due to the fact that animals can find a large quantity of food concentratedwithin localized feeding areas. These areas are the deciduous woods, which repre-sent a predictable source of food.

It is reasonable to assume that new vegetational growth and greater nutritionaldemands faced by females as they prepare for parturition caused the expansion ofspring ranges with respect to winter ones. The same behaviour was observed in RoeDeer (Cederlund, 1983) but not in Fallow Deer at the New Forest, where home ran-ges during spring proved smaller (Rand in Putman, 1986). Definitely, at San Ros-sore, spring and fall ranges may be mostly considered an enlargement of the winterones.

During summer, especially during July-August, when heat and lack of rain-fall cause poor grass quality, we should expect an increase in range size as fema-les move to reach new feeding areas. Yet, summer ranges were the smallest of allfour seasons. Their small size can be attributed both to climatic and social factors.The high temperatures reached during the day limited movement, and deer weremore sedentary, as happens in Spain, during summer, where heat and lack of rain-fall drastically reduce food availability, and local populations of Red Deer limittheir movements in the forest (Carranza et al., 1991; Carranza & Valencia, 1992).According to the seasonal and bimonthly range analyses, areas occupied by fema-les during summer, more in general between May-August, were essentially diffe-rent from those exploited in the other months. These areas, marshes and wet mea-dows located in the western part of San Rossore, were sometimes used also during

— 392 —

spring and fall, but were the only ones used during the period May-August(marshes represent only 5.78% of the whole study area, limited to 5 fragmentsaveraging 54.2 ha). The presence of water guaranteed adequate food supply alsoduring the hot and dry season. The presence of predictable sources of food makesthese restricted wet areas of fundamental importance during parturition and lac-tation when the energetic requirement of female ungulates increases (ca. 40%during late gestation and 150% during lactation; Loudon, 1985). Monitored fema-les occupied these areas during parturition and then their movement patterns wererestricted by the presence of fawn, born in late May or early June. Females seldommove further than 200 meters away from young in the first 2-3 months followingparturition (Chapman & Chapman, 1997). Reductions in home range size forabout 6 weeks or longer following parturition have been documented in femaleWhite-tailed Deer Odocoileus virginianus (Hawkins & Klimstra, 1970; Ozoga etal., 1982; Gavin et al., 1984; McCullough et al., 1989; Schwede et al., 1993; Ber-trand et al., 1996) and in several other studies on cervids (Nelson & Mech, 1981;Vincent et al., 1983; Maublanc, 1986; Jeppesen, 1990; Chapman et al., 1993; SanJosé & Lovari, 1998).

Interseasonal site-fidelity proved to be remarkably high in our study area, com-parable to what found by Nugent (1994). Cervids appear to have a strong attach-ment to their home range, confirming the large element of tradition in the move-ments of deer recorded by Staines (1974, 1977). Home range fidelity, in fact,appears to be a common phenomenon among ungulates like Roe Deer (Linnell &Andersen, 1995, 1998), Moose Alces alces (Sweanor & Sandergren, 1989; Ander-sen, 1991) and White-tailed Deer (Nixon et al., 1992). Several studies reported thatfemale Red Deer use the same seasonal home range in successive years (Georgii,1980), and they also use the same range throughout the seasons and from year toyear (Catt & Staines, 1987). Stability of home range has been also reported formature Roe Deer females (Strandgaard, 1972; Janeau et al., 1981; Vincent et al.,1983) and, more in general, for groups of female ungulates (Brown, 1992; Wec-kerly, 1993).

In conclusion, Fallow Deer female’s use of space proved to be strongly affec-ted by several ecological factors, as the analyses of the home range pattern throu-ghout three years showed.

ACKNOWLEDGEMENTS

We are grateful to the Segretariato della Presidenza della Repubblica, the Tuscany Region and to theGame Keepers of the San Rossore Estate. We thank J.M. Gaillard and an anonymous referee who madevaluable comments on this paper and gave statistical advice. A. Meriggi gave an appreciated help instatistical data processing.

RÉFÉRENCES

ANDERSEN, R. (1991). — Habitat deterioration and the migratory behaviour of moose (Alces alces) inNorway. J. Appl. Ecol., 28: 102-108.

APOLLONIO, M., FOCARDI, S., TOSO, S. & NACCI, L. (1998). — Habitat selection and group formation pat-tern of fallow deer Dama dama in a submediterranean environment. Ecography, 21: 225-234.

— 393 —

BERTRAND, M.R., DENICOLA, A.J., BEISSINGER, S.R. & SWIHART, R.K. (1996). — Effects of parturition onhome ranges and social affiliations of female white-tailed deer. J. Wildl. Manage., 60: 899-909.

BIDEAU, E., VINCENT, J.-P., QUÉRÉ, J.-P. & ANGIBAULT, J.-M. (1983). — Occupation de l’espace chez lechevreuil (Capreolus capreolus L., 1789). I. Cas de mâles. Acta Oecol. Applic., 4: 163-184.

BRAZA, F. (1975). — Censo de gamo (Dama dama) en Doñana. Nat. Hisp., 3: 5-27.BRAZA, F. & ALVAREZ, F. (1987). — Habitat use by red deer and fallow deer in Doñana National Park. Misc.

Zool., 11: 363-367.BROWN, C.G. (1992). — Movement and migration patterns of mule deer in southeastern Idaho. J. Wildl.

Manage., 56: 246-253.BRUNO, E. & APOLLONIO, M. (1991). — Seasonal variations in the diet of adult male fallow deer in a sub-

mediterranean coastal area. Rev. Ecol. (Terre Vie), 46: 349-362.CARRANZA, J. (1991). — Space use by red deer in a Mediterranean ecosystem as determined by radio trac-

king. Appl. Anim. Behav. Sci., 30: 363-371.CARRANZA, J., HIDALGO DE TRUCIOS, S.J., MEDINA, R., VALENCIA, J. & DELGADO, J. (1991). — Space use

by red deer in a Mediterranean ecosystem as determined by radio-tracking. Appl. Anim. Behav.Sci., 30: 363-371.

CARRANZA, J. & VALENCIA, J. (1992). — Organización social del ciervo en habitat mediterraneo. Misc.Zool., 16: 223-232.

CATT, D.C. & STAINES, B.W. (1987). — Home range use and habitat selection by red deer (Cervus elaphus)in a Sitka spruce plantation as determined by radio-tracking. J. Zool., Lond., 211: 681-693.

CEDERLUND, G. (1981). — Daily and seasonal activity pattern of roe deer in a boreal habitat. Viltrevy,11:315-353.

CEDERLUND, G. (1982). — Mobility response of roe deer (Capreolus capreolus) to snow depth in a borealhabitat. Viltrevy, 12: 37-68.

CEDERLUND, G. (1983). — Home range dynamics and habitat selection by roe deer in a boreal area in cen-tral Sweden. Acta Theriol., 28: 443-460.

CEDERLUND, N.G. & SAND, H.K. (1992). — Dispersal of subadult moose (Alces alces) in a non migratorypopulation. Can. J. Zool., 70: 1309-1314.

CHAPMAN, N.G., CLAYDON, K., CLAYDON, M., FORDE, P.G. & HARRIS, S. (1993). — Sympatric popula-tions of muntjac (Muntiacus reevesi) and roe deer (Capreolus capreolus): a comparative analysisof their ranging behaviour, social organization and activity. J. Zool., Lond., 229: 623-640.

CHAPMAN, D. & CHAPMAN, N. (1997). — Fallow deer: their history, distribution and biology. Second Edi-tion. Coch-y-bonddu books, Machynlleth.

CLUTTON-BROCK, T.H. & HARVEY, P.H. (1978). — Mammals, resources and reproductive strategies.Nature, 273: 191-195.

CLUTTON-BROCK, T.H., GUINNESS, F. & ALBON, S.D. (1982). — Red deer: behavior and ecology of twosexes. Chicago University press, Chicago.

CRAIGHEAD, J.J, CRAIGHEAD, F.C. JR, RUFF, R.L. & O’GARA, B.W. (1973). — Home ranges and activitypatterns of non migratory elk of the Madison Drainage herd as determined by radiotelemetry. Wil-dlife Monogr., 33: 1-50.

DROLET, C.A. (1976). — Distribution and movements of white-tailed deer in southern New Brunswick inrelation to environmental factors. Can. Field-Nat., 90: 123-136.

GAVIN, T.A., SURING, L.H., VOHS, P.A. JR & MESLOW, E.C. (1984). — Population characteristics, spatialorganization and natural mortality in the Columbian white-tailed deer. Wildlife Monogr., 91: 1-41.

GEORGII, B. (1980). — Home range patterns of female red deer (Cervus elaphus L.) in the Alps. Oecologia,47: 278-285.

GEORGII, B. & SCHRÖDER, W. (1983). — Home range and activity patterns of male red deer (Cervus elaphusL.) in the Alps. Oecologia, 58: 238-248.

HARESTAD, A.S. & BUNNELL, F.L. (1979). — Home range and body weight – A reevaluation. Ecology, 60:389-402.

HARRIS, S., CRESSWELL, W.J., FORDE, P.G., TREWHELLA, W.J., WOOLLARD, T. & WRAY, S. (1990). —Home-range analysis using radio-tracking data.- A review of problems and techiques particularlyas applied to the study of mammals. Mammal Rev., 20: 97-123.

HAWKINS, R.E. & KLIMSTRA, W.D. (1970). — A preliminary study of the social organization of white-taileddeer. J. Wildl. Manage., 34: 407-419.

HEIDEMANN, G. (1973). — Zur Biologie des Damwildes (Cervus dama Linné, 1758). Parey, MammaliaDepicta, Hamburg und Berlin.

— 394 —

JANEAU, G. L., QUÉRÉ, J.-P., SPITZ, F. & VINCENT, J.-P. (1981). — Etude par radiotracking des variationssaisonnières de l’étendue du domaine vital de cinq chevreuils (Capreolus capreolus L.). Biol.Behav., 6: 291-304.

JEPPESEN, J.L. (1987). — Impact of human disturbance on home range, movements and activity of red deer(Cervus elaphus) in a Danish environment. Danish Rev. Game Biol., 13: 1-38.

JEPPESEN, J.L. (1990). — Home range and movements of free-ranging roe deer at Kalø. Danish Rev. GameBiol., 14: 2-13.

KENWARD, R. (1987). — Wildlife radio tagging: Equipment, Field Techniques and Data Analysis. AcademicPress, London.

KENWARD, R. (1990). — Ranges IV. Software for analysis animal location data. Institute of Terrestrial Eco-logy, Wareham U.K.

KENWARD, R. & HODDER, R.H. (1996). — Range V. An analysis system for biological location data. Insti-tute of Terrestrial Ecology, Wareham U.K.

LAIR, H. (1987). — Estimating the location of the focal centre in red squirrel home ranges. Ecology, 64:1092-1101.

LIEB, J.W. (1981a). — Activity, heart rate, and associated energy expenditure of elk in western Montana.Ph.D. Dissertation. University of Montana, Missoula.

LIEB, J. W. (1981b). — Activity of free-ranging elk as determined by a motion-sensing biotelemetry system.Proc. 3rd Int. Conf. Wildl. Biotelemetry: 1-12. ICWB.

LINNELL, J.D.C. & ANDERSEN, R. (1995). — Site tenacity and logging disturbance in roe deer. Wildlife Soc.B., 23: 31-36.

LINNELL, J.D.C. & ANDERSEN, R. (1998). — Territorial fidelity and tenure in roe deer bucks. Acta Theriol.,43: 67-75.

LOWE, V.P.W. (1967). — Teeth indicators of age with special reference to red deer (Cervus elaphus) ofknown age of Rhum. J. Zool., Lond., 152: 137-153.

MAUBLANC, M.-L. (1986). — Utilisation de l’espace chez le chevreuil en milieu ouvert. Gib. Faune Sauv., 3:297-311.

MCCULLOUGH, D.R., HIRTH, D.H. & NEWHOUSE, S.J. (1989). — Resource partitioning between sexes inwhite-tailed deer. J. Wildl. Manage., 53: 277-283.

MCNAB, B.K. (1963). — Bioenergetics and the determination of home range size. Am. Nat., 97: 133-140.MOEN, A.N. (1976). — Energy conservation by white-tailed deer in the winter. Ecology, 57: 192-198.MOHR, C.O. (1947). — Table of equivalent populations of North American small mammals. Am. Midl. Nat.,

37: 223-249.NELSON, M.E. & MECH, L.D. (1981). — Deer social organization and wolf predation in Northeastern Min-

nesota. Wildlife Monogr., 77: 1-53.NIXON, C.M., HANSEN, L., BREWER, P.A. & CHELSVIG, J.E. (1992). — Stability of white-tailed doe parturi-

tion ranges on a refuge in east-central Illinois. Can. J. Zool., 70: 968-973.NUGENT, G. (1994). — Home range size and its development for fallow deer in the Blue Mountains, New

Zealand. Acta Theriol., 39: 159-175.OZOGA, J.J., VERME, L.S. & BIENZ, C.S. (1982). — Parturition behaviour and territoriality in white-tailed

deer: impacts on neonatal mortality. J. Wildl. Manage., 46: 1-11.PETERS, R.H. (1983). — The Ecological Implications of Body Size. Cambridge University Press.PUTMAN, R.J. (1986). — Grazing in temperate ecosystems: large herbivores and the ecology of the New

Forest. Croom Helm Ltd, Kent. RONGSTAD, O.J. & TESTER, J.R. (1969). — Movements and habitat use of white-tailed deer in Minnesota. J.

Wildl. Manage., 33: 366-379.SAN JOSÉ, C. & LOVARI, S. (1998). — Ranging movements of female roe deer: do home-loving does roam

to mate? Ethology, 104: 721-728.SCHOENER, T.W. (1981). — An empirically based estimate of home range. Theor. Popul. Biol., 20: 281-325.SCHWEDE, G., HENDRICHS, H. & MCSHEA, W. (1993). — Social and spatial organization of female white-

tailed deer, Odocoileus virginianus, during the fawning period. Anim. Behav., 45: 1007-1017.SIMONI, D. (1910). — San Rossore nella storia. Leo S. Olschki, Firenze.SOUTHWOOD, T.R.E. (1966). — Ecological methods with particular reference to the study of insect popula-

tions. Chapman and Hall, London.SPITZ, F., VALET, G. & BRISBIN, I.L. (1998). — Variation in body mass of wild boars from southern France.

J. Mammal., 79: 251-259.

— 395 —

SPRINGER, J.T. (1979). — Some sources of bias and sampling error in radio triangulation. J. Wildl. Manage.,43: 926-935.

STAINES, B.W. (1974). — A review of factors affecting deer dispersion and their relevance to management.Mammal Rev., 4: 79-91.

STAINES, B.W. (1977). — Factors affecting the seasonal distribution of red deer (Cervus elaphus) at GlenDye, north-east Scotland. Ann. Appl. Biol., 87: 495-512.

STRANDGAARD, H. (1972). — Roe deer (Capreolus capreolus) at Kalø and the factors regulating its size.Danish Rev. Game Biol., 7: 1-205.

SWEANOR, P.Y. & SANDEGREN, F. (1989). — Winter-range philopatry of seasonally migratory moose. J.Appl. Ecol., 26: 25-33.

SWIHART, R.K & SLADE, N.A. (1985). — Testing for independence of observations in animal movements.Ecology, 66: 1176-1184.

TUFTO, J., ANDERSEN, R. & LINNELL, J. (1996). — Habitat use and ecological correlates of home range sizein a small cervid: the roe deer. J. Anim. Ecol., 65: 715-724.

VAN DYKE, F. & KLEIN, W.C. (1996). — Response of elk to installation of oil wells. J. Mammal., 77: 1028-1041.

VINCENT, J.-P., BIDEAU, E., QUÉRÉ, J.-P. & ANGIBAULT, J.-M. (1983). — Occupation de l’espace chez lechevreuil (Capreolus capreolus L.). II. Cas des femelles. Acta Oecol. Applic., 4: 379-389.

VINCENT, J.-P., BIDEAU, E., HEWISON, J.M. & ANGIBAULT, J.-M. (1995). — The influence of increasingdensity on body weight, kid production, home range and winter grouping in roe deer (Capreoluscapreolus). J. Zool., Lond., 236: 371-382.

WECKERLY, F.W. (1993). — Intersexual resource partitioning in black-tailed deer: a test of the body sizehypothesis. J. Wildl. Manage., 57: 475-494.

WHITE, G.C. & GARROTT, R.A. (1990). — Analysis of wildlife radio-tracking data. Academic Press Inc.,San Diego.

WORTON, B.J. (1989). — Kernel method for estimating the utilization distribution in home range studies.Ecology, 70: 164-168.