The Third International Tree Squirrel Colloquium and The Seventh European Squirrel Workshop

Ford Castle, Northumberland, UK
26th May to 30th May 2003

Welcome address Dear colleagues and friends, welcome to Ford Castle. On behalf of the organising committee I would like to welcome you to the Third International Colloquium of Tree Squirrels and the Seventh European Squirrel Workshop. We are particularly pleased to bring together meetings from two very successful series of international conferences on tree squirrels that have spanned the last 15 years. The First International Tree Squirrel Colloquium was held at Powdermill Biological Station of the Carnegie Museum of Natural History, Pennsylvania in 1994, and the second in Sublimity, Oregon in 2000. The European Workshops go back to 1992 when the first one was held in Portovaltravaglia, Italy. This was followed by Workshops in Southampton, England (1995), the Nationalpark Bayerischer Wald, Germany (1996), Grimsö Wildlife Research Staion, Sweden (1997), Wareham, Dorset, England (1999), and, lastly, Acqui Terme (Alessandria), Italy (2001). Traditionally, these meetings all been characterised by a relaxed and friendly atmosphere as delegates have debated, often in a lively fashion, the latest developments in the field of squirrel biology. We hope that this meeting will be equally rewarding and develop closer contacts and collaborations with colleagues from many different countries from throughout the world. Tree squirrels are fascinating animals. As you will see through the oral and poster presentations; they are increasingly becoming used to test theories about, for example, phylogeny, ecology and behaviour, but that they are also the focus of conservation or, conversely, control management strategies. We also hope you will enjoy the social events including the excursion to the Farne Islands. We would like to thank our sponsors and the many other people who have contributed to setting up this meeting. We wish you an enjoyable and fruitful time.

26th May 2003, CSLM, University of Newcastle-upon-Tyne

A preliminary study on some taxonomical, ecological and biological characteristics of Sciurus anomalus in Turkey by Irfan Albayrak (Kirikkale University, Science and Art Faculty, Biology Department, 71450 Yahsihan, Kirikkale, Turkey) and Arslan Atilla (University of Selcuk, Faculty of Science and Arts, Department of Biology, 42031, Konya, Turkey)

This study is based on a total of 62 Sciurus anomalus specimens obtained from four provinces in the middle of Turkey between 1998 and 2002. As some ecological and biological features, habitat, feeding, nest, fur colour, baculum and phallus were investigated. Karyological analyses indicated that the number of choromosomes in this species was 40. Comparisons were made taxonomically with those data given for the species in relevant literature and discussed at the subspecies level.

Red squirrel ecology in alpine forests: space, seeds and fungi by Bertolino Sandro (University of Turin, DI.VA.P.R.A. – Laboratory of Zoology Email: )

Eurasian red squirrel (Sciurus vulgaris) populations occurring in the alpine conifer forests might constitute the only stronghold against the spreading of the introduced eastern grey squirrels (Sciurus carolinensis) in northern Italy. In 2000 we started a study on the ecology of red squirrels in subalpine conifer forests (Gran Paradiso National Park, Italy), investigating the annual variation in food resources (conifer seeds, fungi), seasonal and annual changes in squirrel population density, and spacing behaviour. In summer and autumn 2001 and 2002 we radio-tracked 18 and 13 squirrels respectively. In 2001, after a poor cone crop in 2000, home ranges of red squirrels were very large (94.1± 72.2 ha in males, 78.9 ± 51.5 ha in females) and an unusual high core-area overlap was observed, suggesting a breakdown of the normal social organisation and spacing pattern of this species. This high level of overlap suggested that animals concentrated their activity in patches where spruce seeds were still available or where fungi were abundant.In 2002, all resident squirrels apparently reorganised themselves in stable home ranges, foraging intensively in small core-areas (mean home range size: 20.6 ± 7.9 ha in males, 14.5 ± 5.3 ha in females, core-areas from 4.5 to 9.5 ha in males, 3 to 6 ha in females). The core-areas of adult females were exclusive between one another, showing the typical intrasexual territoriality.

Comparative energetics of red and grey squirrels and the constraints of habitat? by R.M.Bevan (School of Biology, University of Newcastle, Newcastle upon Tyne, NE1 7RU, UK)& P.W.W.Lurz (Centre for Life Science Modelling, School of Biology, University of Newcastle, Newcastle upon Tyne, NE1 7RU, UK)

One area of squirrel biology that has not received much attention is their energy usage, even though the energy expenditure of any animal is central to our understanding of many aspects of its ecology. In the UK, the decline in the red squirrel (Sciurus vulgaris) distribution has been attributed mainly to the spread of the introduced grey squirrel (S. carolinensis). Nevertheless, there are ‘poor quality’ habitats, such as spruce dominated forests, where the red squirrel is able to coexist with or even out compete the grey squirrel. This is thought to be due mainly to the constraints of the higher daily energy expenditure (DEE) of the larger grey squirrel. We used the doubly labelled water technique to measure the DEE of red and grey squirrels in different habitats: (i) red squirrels in typical conifer plantation habitat, (ii) grey squirrels in a typical conifer habitat, (iii) grey squirrels in good quality deciduous habitat. The study was conducted during the breeding season as this is when the females will be most energetically stressed. Lactating females of both species had a DEE that was 2-2.5 times higher than that of non-reproductive individuals. The DEE of red squirrels was 422 ± 141 kJ.d-1 (mean ± SD), whereas the grey squirrels had a significantly higher DEE of 1080 ± 387 kJ.d-1. The DEE of grey squirrels was greater in the coniferous (1127 ± 460 kJ.d-1) than in the deciduous habitat (961 ± 119 kJ.d-1) and may reflect an increased foraging effort in this habitat.

Research on distribution and ecology of red squirrel (Sciurus vulgaris) in Hungary by Szilvia Bosze (Department of Zoology and Ecology, Szent István University, H-2103 Gödöllo, Páter K. u. 1., Hungary, ), Botond Bakó (Department of Zoology and Ecology, Szent István University, H-2103 Gödöllo, Páter K. u. 1., Hungary, ) & Gábor Csorba (Mammal Collection, Hungarian Natural History Museum, H-1081 Budapest, Ludovika tér 2., Hungary,)

Thus far, the Hungarian science did not have any comprehensive data on ecological, population-biological and distributional peculiarities of red squirrel (Sciurus vulgaris) populations in the country. Our aim is to establish a nation-wide distribution database, to describe the habitat preference relationships and to maintain the conditions of long-term monitoring of the species.Firstly, we interpreted red squirrel distribution data from the technical literature, public collections and species lists gained from owl-pellet analysis on 10*10 grid U.T.M. maps. Secondly, we contrasted our distribution map with the arboreal woodland cover map of Hungary. Thirdly, we collected data from places we still do not have but are potentially suitable for red squirrels. We worked out a so called ‘Squirrel watching’ programme in 2001 to involve wide social strata (school-children, foresters, nature conservationists, etc.) in the collection of distribution data. We asked for locality, coat colour and behaviour peculiarities of specimens observed, and also for local habitat composition. We appraised habitats, based on the origin of it. We differentiated natural forests, anthropogenic biotopes and arboreal woodland locations of invasive species like coniferous, acacia, etc. woodlands. We ranked the data based on the trustworthiness of its origin; whether it is from a school-child or a national park guard for example. We tested our programme in 2002 in a small area of Hungary with attendance of 42 schools. We widened the programme to the whole country in 2003. We interpret data collected through ‘Squirrel watching’ programme as well on 10*10 grid U.T.M. map. We compare the two distribution maps and the map of arboreal woodland cover.

The survival of the red squirrel in Wales – an update on the situation. by Sarah Cartmel (Cymerau Isaf, Blaenau Ffestiniog, Gwynedd. LL41 4BN UK) & John Gurnell (School of Biological Sciences, Queen Mary, University of London E1 4NS UK)

It has been nearly three years since the end of the red squirrel project at Clocaenog Forest in North Wales. The results of that survey suggested that red squirrel’s could potentially survive in the conifer plantations of Wales (despite the presence of the grey squirrels) on condition that (in broad terms) the right tree species and age classes are present and that there are sufficient links throughout the forest to enable the squirrels to travel between blocks of suitable habitat (i.e. that the forest has not become too fragmented). These theories have been applied to a group of forests in mid Wales that are known to contain a population of red squirrels – although the size and ‘health’ of the red squirrel population is not known. The current state of the forests was determined with regard to their suitability for red squirrels using available information and ground truthing, and their predicted suitability according to the current forest design plans was also explored. This gave an insight into the potential size of the red squirrel population in mid Wales and to their future size over the next 50 years. This study then attempted to come up with some simple solutions to improve the habitat for red squirrels with as little economic loss as possible in order to be acceptable to the forest managers of these mid Wales forests. Finally there will be an update on the work in Clocaenog, which has begun this winter.

The effects of forest fire on the squirrel and tree shrew community dynamic in southern Sumatra by Asri A. Dwiyahreni (Wildlife Conservation Society Indonesia Program, PO Box 311 Bogor 16003, Indonesia)

A study to investigate the effect of forest fire on the dynamic of squirrel and tree shrew community was conducted in the south of Bukit Barisan Selatan National Park (BBSNP), Lampung, Indonesia. The park is surrounded by agricultural and plantation areas but contains some of the largest intact tracts of lowland forest remaining on the island which undergoing the most rapid forest conversion. The study focused on 8 squirrel (Ratufa bicolor, R. affinis, Callosciurus notatus, Sundasciurus hippurus, Lariscus insignis C. nigrovittatus, S. lowii and S. tenuis; Mammalia: Rodentia) and 3 tree shrew species (Tupaia tana, T. glis and T. minor; Mammalia: Scandentia) and described how forest structure and pre- and post-fire conditions of the habitat affect the density, distribution and niche partitioning among squirrel and tree shrew species in the study area. Data were collected before the fire in 1997, then each year subsequently after that until 2001. The population density of each animal species was assessed using line transect and calculated using DISTANCE program. The study showed that density was significantly lower after the fire and in the burned area. The ground dwelling species were the most suffered species as fire destroyed their preferred disturbed habitat more severely than the undisturbed ones. There was also an interesting change on the average height choice of each species in the burned and unburned areas after the fire. All ground dwelling species moved slightly higher in the burned compare to the unburned area. This was probably due to more fallen logs available in the unburned area to sit and scan their surrounding to avoid predators. The animals living in the burned area might be more profound to predators due to the openness of the canopy. In the other hand, the tree-dwelling species tend to move to a lower stratum in the burned area probably to avoid predators such as eagle. Analysis of substrate choice among the species showed that height is a more important factor in the choosing of stratum compare to substrate. Species seemed to just use the available substrate in the associated height although it was not the right substrate for them. The study has showed that fire caused changes in habitats and moreover would change the structure of the animal community. Species with high adaptability tend to have higher survival rates to the fire and might drive other weaker species to a temporary or even permanent extinction from their habitats. In the long run, it seemed that fire could increase the abundance of damaging species whilst further decline the abundance of other species that could counteract those damaging species.

Space use of sympatric endangered Mount Graham red squirrels and introduced Abert’s squirrels by Andrew J. Edelman (Wildlife & Fisheries Sciences, School of Renewable Nat. Res., Univ. of Arizona, Tucson, AZ 85721), John L. Koprowski (ildlife & Fisheries Sciences, School of Renewable Nat. Res., Univ. of Arizona, Tucson, AZ 85721 and Mount Graham Red Squirrel Monitoring Project, School of Renewable Nat. Res., Univ. of Arizona, Tucson, AZ 85721) & Sarah R. B. King (Mount Graham Red Squirrel Monitoring Project, School of Renewable Nat. Res., Univ. of Arizona, Tucson, AZ 85721)

The introduction of non-native species into ecosystems can have serious impacts on native biodiversity. Abert’s squirrels (Sciurus aberti) are an introduced species that is believed to contribute to the current decline of the endangered Mount Graham red squirrel (Tamiasciurus hudsonicus grahamensis) through resource competition. The Mount Graham red squirrel is endemic to mesic forests of the isolated Pinaleño Mountains, Arizona. Introduced Abert’s squirrels are sympatric with Mount Graham red squirrels throughout the latter’s habitat. As part of an ongoing effort to examine the ecological relationship between these sympatric tree squirrels, we initiated a space use study in June 2002. We used radio-tracking to determine characteristics of intra- and interspecific space use. Home ranges of Abert’s squirrels were much larger than red squirrel home ranges. Intraspecific overlap of home ranges was higher among Abert’s squirrels than red squirrels. Abert’s squirrel home ranges overlap with more conspecifics than red squirrel home ranges. The percent overlap and number of Abert’s squirrels overlapping red squirrel home ranges were similar to intraspecific overlap of red squirrels. These results demonstrate that red squirrels and Abert’s squirrels differ greatly in space use characteristics. In addition, reduced interspecific overlap may indicate that Abert’s squirrel space use is affected by red squirrel territoriality. Avoidance of red squirrel home ranges by Abert’s squirrels could reduce interspecific competition and allow coexistence between these species.

Activity and Time Budgets of the Mt. Graham Red Squirrel by Vicki L. Greer & John L. Koprowski, Mt. Graham Biology Programs (School of Renewable Natural Resources, University of Arizona, Tucson, Arizona 85721, USA)

The endangered Mt. Graham red squirrel (Tamiasciurus hudsonicus grahamensis) is found only in the Pineleño (Graham) mountains of southeast Arizona, and is the southernmost population of red squirrels in North America. Since 1989, as part of a long-term monitoring program, we have studied squirrels around an astro-physical complex in the north-central part of the mountain range. From September 1989 to October 1991, we collected over 3000 h of behavior observations at territory centres (middens), in both spruce-fir (above 3110 m elev.) and mixed conifer (below 3110 m elev.) habitats. Observations consisted of an observer sitting quietly near the midden center for two hours and recording the activities of all squirrels (to the nearest 0.25 min) within a 10 m radius of the midden center. Activities were tallied into nearly 20 different behaviour categories. Vocalizations within and nearby the midden were also recorded. The original focus of the observations was to detect potential differences in behaviour between squirrels near construction of the astro-physical complex and squirrels in non-construction areas. A previous examination of these data demonstrated no consistent, distinguishable differences in the behaviour of squirrels related to construction activity; however, we re-examine the data with respect to possible differences in sex, age, habitat (both forest type and “micro-habitat” variables around the midden center), weather, and food availability. Preliminary analysis indicates that squirrels in both habitats spent more time in food related (feeding, foraging, and caching) than in passive (grooming, nest building, basking) or defensive (territorial chases and responses to predators or humans) behaviours.

Mating strategy and natal dispersal in the Siberian flying squirrel, Pteromys volans by Ilpo K. Hanski & Vesa Selonen (Department of Ecology and Systematics, PO Box 65, FIN-00014 University of Helsinki, Finland)

Siberian flying squirrel is a nocturnal, strictly arboreal species living in coniferous forests in northern Eurasia. We have studied mating behaviour, social spacing behaviour and natal dispersal of radio-tagged flying squirrels in Finland in 1996-2002. Paternity of offspring was analysed using 7 DNA microsatellite loci. Females of the Siberian flying squirrel are larger than males and they occupy separate home ranges 4-10 ha in size. The male ranges are 5-10 times larger than female ranges, and overlap with each other. One male range may include several female ranges. Before mating period (late March) males spend time in the same nest with one or a few females, and during mating period may visit several females. After a short, one-day receptive period of the female, the sexes separate. Both sexes are promiscuous in their copulation behaviour, both having multiple mates. The DNA analyses revealed that about 40% of litters were sired by several fathers. A half of the males potentially available in the area did not father any offspring. Natal dispersal is female biased. All young females dispersed > 500 m from the natal home range, whereas about 40% of males stayed. The dispersal distances did not differ between sexes. The average distance was less than 3 km, but maximally as long as 9 km.

Mt-dna haplotype dynamics in two populations of Japanese squirrels (Sciurus lis) in continuous and isolated forests by Fumio Hayashi & Noriko Tamura (Email: )

The Japanese squirrel (Sciurus lis) is distributed on Honshu, Shikoku, and Kyushu Islands in Japan. Recently, populations on south-western Honshu and Shikoku decreased and those on Kyushu disappeared. One of the factors affecting to local extinction of this species seems to be forest fragmentation by human activities. In the present study, we compared genetic heterogeneity between the populations living in the continuous forest in Mt.Takao, western Tokyo, and the isolated forest, ca.70 ha in area, which is neighbouring to the former but interrupted by high way. In the both study sites, we established a 50 ha trapping area arranged 20 live-traps. Trapping was conducted at 1-3 month intervals from June 1999. A small piece of skin was sampled from the tip of ears for genetic analysis and collars for individual identification were attached when the squirrels were first captured. A total of 8 haplotypes in the mt-DNA D-Loop region were ascertained from 57 individuals obtained from the both study sites (6 types in the continuous forest and 4 types in the isolated forest). By repeating captures by traps, the period persisting in the study site was estimated for individuals. The mean number of haplotypes existing per year was 2.25 in the isolated forest while 4.00 in the continuous forest. The number of haplotypes was significantly smaller in the isolated forest than in the continuous forest (Mann-Whitney U-test, U=1, P<0.05). Thus, forest isolation caused temporal decreasing in genetic diversity of the Japanese squirrel in respect of the mt-DNA D-Loop region.

Patterns of genetic diversity in the red squirrel (Sciurus vulgaris L.): footprints of biogeographical history and artificial introductions by Hale, M. & Lurz, P. W. W. (School of Biology, University of Newcastle upon Tyne, NE1 7RU Email: )

Landscape management and conservation practices which alter the degree of habitat fragmentation can significantly impact upon the genetic structure of animal populations. In a microsatellite DNA survey we found that British red squirrels use ‘stepping stone’ patches of habitat to move considerable distances through a fragmented habitat. Over the past few decades, the planting of a large conifer forest has connected groups of forest fragments in the north of England with those in southern Scotland. This defragmentation of the landscape has resulted in substantial genetic mixing of Scottish and Cumbrian genes in squirrel populations up to 100km from the site of the new forest. In addition, a mitochondrial DNA survey showed that the majority of extant populations of British S. vulgaris are of continental European ancestry, many with a very recent (last 40 years) Scandinavian ancestry. The Scandinavian haplotype has rapidly become the most dominant haplotype in the northeastern British populations, despite not appearing in northern English populations until 1966. This suggests that these squirrels may have an adaptive advantage in the non-native spruce dominated conifer plantations of northern England. Initial results suggest that squirrels recolonised Scandinavia from northern or eastern glacial refugia rather than southern refugia.

Reproductive profiles of tree-nesting and ground-nesting squirrels by V. Hayssen, S. Soss, & A. Vandevusse (Department of Biological Sciences, Smith College, Northampton, MA 01063)

Do ecological aspects of a species niche influence aspects of reproduction? Ground-nesting and tree-nesting species have different nest predators and thus differences in the causes of juvenile mortality. Adult mortality may also differ as food availability may also differ for ground- vs tree-nesting species. These theoretical differences in mortality patterns may influence reproduction. This prediction was tested by comparing the reproductive patterns of over 30 species of tree-nesting squirrels with those of over 20 species of ground-nesting squirrels. In general tree-nesting squirrels have smaller litter sizes but longer lengths of gestation and lactation. However, overall litter mass does not differ when compared to female mass. Thus, the number of offspring and the time used for reproduction vary between the groups but the amount of energy invested at the time of birth does not. More simply, tree-nesting squirrels take longer to produce fewer, but slightly larger, young than ground-nesting squirrels. For squirrels, therefore, ecological aspects of a species niche (e.g. nesting site) have influenced the duration of reproductive investment and litter size but allometric constraints affect the energetic investment into a litter at birth.

An automated dispenser for feeding and logging individual squirrels by Ben Kenward & Robert Kenward (Email: )

I present an automatic system for feeding free-ranging squirrels (Sciurus carolinensis) according to arbitrary schedules which could differ between individuals. The system requires trapping wild squirrels in the surrounding area and implanting them with subcutaneous PIT tags. These tags are used to allow individuals selective access to an inner chamber with a joystick that may be programmed to require arbitrary movements to cause the delivery of a hazelnut or similar reward. In field tests squirrels learnt to operate the access system well but were reluctant to use the joystick, suggesting that a different manipulandum may be more appropriate. The apparatus would be useful for any study where food access of different free-ranging individuals is to be controlled.

Geographical variation in British red squirrels, Sciurus vulgaris by Andrew C. Kitchener (National Museums of Scotland , Dept. of Geology and Zoology, Chambers Street, Edinburgh EH1 1JF. Tel: 0131-247 4240 e-mail: ) Gayle Peacock, John M. Lynch, & John Gurnell>/p>

The British red squirrel is often regarded as a distinct subspecies, Sciurus vulgaris leucourus, on the basis of its tail and ear tufts, which fade to white seasonally. However, British red squirrel populations may have become corrupted by introductions of red squirrels from Europe. More than 300 red squirrels from the Britain were examined for geographical variation in pelage coloration, seasonal fading of tail and ear tufts, and skull measurements. Four main populations were compared with presumed differing origins; Jersey (introduction from France), Formby (possible introduction from Europe), Cumbria (indigenous) and the Highlands of Scotland (re-introduced from England, possible Swedish introductions). The results of this study contradict genetic studies done so far, and provide some answers to the question of the subspecific distinctiveness of British red squirrels.

Sequence diversity of mitochondrial control region in Red Squirrels, Sciurus vulgaris L., from two most disjunct populations of Korea and England by Koh, Hung Sun, Shin Dong Sun, & Ryu Mi Hyun (Department of Biology, Chungbuk University, Cheongju 361-763, South Korea)

Partial sequences of mitochondrial DNA control region in red squirrel (Sciurus vulgaris vulgaris) from Korea in Far East Asia were analysed. These haplotypes were compared with those of red squirrel from England in west Europe (S. v. leucourus), obtained from Genbank. It was revealed that all haplotypes of red squirrels from Korea and England were intermingled without the formation of any major clade, although they are the most distantly distributed populations. It was concluded that this species is one of mammals rapidly dispersed in the late Pleistocene after the last glaciation, and that conservation efforts have to be conducted at international levels, rather than national scale.

(p>Conservation of Tree squirrels: future directions for the study of arboreal squirrels by John L. Koprowski (Wildlife and Fisheries Science, School of Renewable Natural Resources, University of Arizona, Tucson, AZ 85721 USA. )

Conservation of biological diversity is a major endeavour of many present day ecologists. Management, stewardship, and conservation of forest resources is a major responsibility of natural resource conservation agencies and ecologists be they in the temperate or tropical regions of our planet. Tree squirrels are obligates of mature forests that produce substantial seed crops and nest sites. These ecological requirements suggest that tree squirrels can serve as the ‘miner’s canary’ with respect to issues of forest health and biodiversity. Herein, I discuss the potential for a new paradigm of tree squirrel research by reviewing the limits of our knowledge and the sources of these limits. Fully 77% of the species of holarctic arboreal-dwelling squirrels in the genera Sciurus, Tamiasciurus, and Glaucomys are of questionable conservation status in some or all of their range. Furthermore, 68% of the 31 species of arboreal dwelling sciurids in these genera are virtually unstudied and are without detailed publications on even the most basic of life history traits or ecological requirements. Our knowledge of species in other related genera from South America, Africa, and Asia is extraordinarily poor. Basic descriptive ecological studies are needed to fill the gaps of knowledge and permit informed conservation and management decisions. Comparative approaches within and outside the tree squirrels are needed to synthesise current states of knowledge, elucidate patterns, and highlight shortcomings. Collaborative ventures are necessary to link expertise and opportunity as well as to design useful experiments or monitoring programs. The charge that I issue is for each of us to think >10 years down the road (beyond tenure decisions and retirements) and thereby provide direction to research and management programs.

Social interactions are related to cache placement in grey squirrels by Lisa Leaver (School of Psychology University of Exeter, Exeter EX4 4QG, UK)

Dominance interactions among grey squirrels (Sciurus carolinensis) at artificial feeders were measured and their cache locations were mapped. Position in the hierarchy was negatively related to distribution of scattered caches. More dominant squirrels ‘clumped’ caches whereas more submissive squirrels distributed their caches more widely.

A Modelling Study of Red Squirrels on the Island of Jersey, Channel Islands: A Guide to Conservation Management by Louise Magris (Environment Department, States of Jersey), John Gurnell(Queen Mary College, University of London), Steve Rushton (Centre for Life Sciences Modelling, University of Newcastle), & Peter Lurz, (Centre for Life Sciences Modelling, University of Newcastle)

Between 1994 and 1998, intensive studies were carried out on the population of red squirrels (Sciurus vulgaris) that exist among many woodland fragments on the Channel Island of Jersey. About 400-600 squirrels persist in an inhospitable environment with little woodland cover (approx 650 hectares) separated by agricultural and suburban landscapes. The woodlands are extremely small and fragmented and links between them rely on some 340km of hedgerows. Mortality from road deaths are common and because the squirrels receive a high degree of supplemental feeding from the public, predation by domestic cats can be high. Using data from the intensive study we used a Spatially Explicit Population Model and local habitat information to model the viability of the squirrel populations into the future. We present data on predicted population trends under different management scenarios such as hedgerow planting and variation in supplemental food availability. Moreover, we place these predictions in the context of variations in the natural seed supply. The results showed the importance of woodland quality as well as hedgerow continuity for dispersal success and Island wide population viability. The information the models provided on local population persistence have allowed management regimes to be planned as part of the Island’s ongoing species recovery programme.

Ecology of native and exotic urban populations of Eastern gray squirrels by Tommy S. Parker and Charlie Nilon (University of Missouri—Columbia, 302 ABNR Columbia, MO, USA 65201-7240)

The eastern gray squirrel (Sciurus carolinensis) is one of the many species of tree squirrels found in North America. This species was introduced at 30 locations throughout Great Britain from 1876 to 1929. Resulting from the urbanisation of the species, gray squirrels are abundant in urban areas in both North America (native) and the United Kingdom (exotic). Manski et al., (1980) developed hypotheses on the effects of urbanisation of wildlife species. Because of human activity, urban areas are more susceptible to introductions of exotic fauna; however, there have been no investigations on the efficacy of these hypotheses on exotic species in urban areas. Therefore, six study sites will be established in both metropolitan areas of Baltimore, Maryland and London, England. Comparisons of population densities, intraspecific aggression, wariness, and activity patterns of native and exotic urban populations of gray squirrels will be done to test hypotheses on the effects of urbanisation on wildlife species.

Space use of Chiricahua fox squirrels relative to fire history by Bret S. Pasch & John L. Koprowski (Wildlife and Fisheries Science, School of Renewable Natural Resources, University of Arizona, Tucson, AZ 85721 USA)

Impacts of fire on biotic communities are of growing interest due to recent concerns regarding the efficacy of suppression regimes. Tree squirrels are model organisms for studying the effects of fire due to their dependence on mature forests for food and nest sites. Chiricahua fox squirrels (Sciurus nayaritensis chiricahuae), a subspecies of Mexican fox squirrels, are found only in the Chiricahua Mountains of southeastern Arizona. S. n. chiricahuae are considered one of the rarest mammals in Arizona yet the basic ecology of this sky island endemic is poorly understood. We initiated a space use study in May 2002 as part of a study investigating the population ecology of Chiricahua fox squirrels. We radiotracked squirrels across a continuum of historical fire regimes that included fire-suppressed canyon bottoms, plots of prescribed burns, and remnants of an 80 ha wildfire. Home ranges of males were greater than females (10.7 ha vs. 7.2 ha; 85% fixed kernel). Animals outside fire-impacted areas had greatly inflated ranges (males = 24.7 ha, females = 12.0 ha) relative to conspecifics within areas of prescribed fire. Squirrels did not use the area impacted by the wildfire. Squirrels appear to respond to condition-dependent characteristics of fire regimes and may respond best to prescribed fires; such fires may burn cooler and maintain edible seed banks and fungi or cover that better meet the requirements of Chiricahua fox squirrels. Squirrel use of fire-impacted areas will enable managers to assess the impact of fire on a mature forest specialist and develop informed conservation strategies.

Nest-site selection by red squirrels, Sciurus vulgaris by J. Piqué, & J.D. Rodríguez Teijeiro (Email:)

Squirrels can select different measurements in relation to trees where they located the nests (Farentinos, 1972; Pulliainen, 1973; Tittensor, 1970; Rothwell, 1979; Nixon et al., 1984), and in relation to location nest in relationship trunk tree (Farentinos, 1972b; Pulliainen, 1973; Tittensor, 1970; Rothwell, 1979; Nixon et al., 1984). Perhaps they select this measurements because is more easy to have acces to the nests or, because they search most protection in relation raining or wind (Halloran & Bekoff, 1994). This measurements can be modificated when the age of individual change and when they learn manufacturing techniques for make the nests (Holm, 1987), and consequently they supose advantages in squirrel survival (Rothwell, 1979).The aims are know tree measurements where squirrels located the different types of nests, and know the situation of the nests in relation to trunk tree. We would know: (a) if there are differences between sex and between individuals in location nests for sleeping at nigth or resting during the day, (b) know if are individual differences in breeding nests location and if squirrels select entrance hole orientation or select nest orientation in relation to trunk tree, (c) describe the nests which females moving her youngs during their infancy, the “new lactation nests”, and (d) know if are individual differences in relation to orientation of the nests for sleping during nigth and resting during the day. Two sites in Collserola Mountains (province of Barcelona, NE Spain,) with NE-SW orientation, without forestal explotation. Urban Parc of 11 ha situated in SE area, and mediterranean woodland of 38 ha situated in NE area. Location nests are recorded between march 1991 and December 1993. They are used by radio-tagged squirrels (Biotrack) intensively monitored (Urban Parc = 20 squirrels, 8 males and 12 females; Mediterranean woodland = 26 squirrels, 9 males and 17 females), between 1 hour before dawn and 1 hour after dusk . Each squirrel use 6.7 nests in tree months winter period (from 3 squirrels intensively monitored). Also we used a tape measure (25 meters), compass Suunto KB-14/360R and Clinometer Suunto PM-5/360.


  1. Are there any difference between sexes in any habitat, in relation to measurements of trees and in relation to the situation of the nest on the tree, in relation to the nests used by sleeping during the night or by resting during the day, situated in Pinus halepensis and Quercus ilex?
  2. Are there statistical differences between habitats in relation to the distance between nest and tree trunk (P. halepensis) with higher values in Urban Parc. We suggest they are associated with advantages in relation to thermoregulation. The sun warm the nest in two biological and very important day hours for the squirrel: before the night and, especially after all the night, in sunset. Perhaps the squirrel can situated the nest in a optimal situation in exterior branches of the trees, without aerial predators.
  3. Are there individual differences (adults, Urban Parc) in relation to the variables of the tree (trunk perimeter, altitude, and distance between the floor to first branch), and situation of the nest in the tree (altitude and distance to tree trunk) in relation to the nests used by sleeping during the night or by resting during the day situated in P. halepensis. We suggest each squirrel select the trees for located the nests, among all trees available in their home range.
  4. Does the dimension of the tree (P. halepensis) condition the utilisation of the nest (adults, Urban Parc): are directly related between altitude of tree and altitude of nest , directly relation between perimeter of the trunk and distance nest to trunk, directly relation trend (P = 0.0827) between altitude (where squirrels located the nest) and distance nest to trunk.
  5. There are not have individual differences in relation to the variables of the breeding nests.
  6. The nests which females moving her young during their infancy, the “new lactation nests” are situated in perennial trees. They are nests that they are first nests o secondary nests (used during the night or during the night and day) in pre-breeding time.
  7. The nests by sleeping during the night and resting during the day, used for males and females, and located in P. halepensis (Urban Parc and together habitats) are orientated to SE quadrant.
  8. There are not individual variations (Urban parc) in nest orientation (sleeping night nest and resting nest during day) in relation to tree trunk, but they are orientated in adults in SE quadrant. When more months old are the squirrel, more concentrated are located the nests by sleeping during night and resting during day, in SE quadrant.
  9. The breeding nest are orientated in different tree species in relation of the tree trunk (Urban Parc and together habitats), to SE quadrant.

Associated forest structure and geographic distribution of two sympatric flying squirrels in southern India by Nandini Rajamani (Department of Biological Sciences, Auburn University, Auburn, AL 36849)

Documentation of the geographic distribution and basic ecological requirements of a species is a prerequisite for the formation of effective conservation programs. The distribution of the endemic Travancore flying squirrel (Petinomys fuscocapillus fuscocapillus) and the large brown flying squirrel (Petaurista philippensis) was documented in 47 sites across a variety of forest types in the Western Ghats of southern India. One hundred and twelve trails totaling 320 km were sampled between 1900 and 0100 hrs and spotlighting was the principal method used for detection of species. Vegetation was characterized at all sampling locations to determine habitat parameters associated with the occurrence of flying squirrels. P.f.fuscocapillus was mainly restricted to evergreen forests, and occurred across a narrower range of altitudes, suggesting it was more of a specialist thanP.philippensisP.philippensis was encountered most in moist deciduous forests and plantations. Both species preferentially used tall, large trees within stands. Such trees are better resources for food and nest cavities and enable longer and possibly more economical glides. Stands where the largerP.philippensis was encountered were taller and less dense than stands where it was not encountered. It is suggested that the preferences for stands of different tree densities by the two species of flying squirrels might be related to their gliding abilities. The presence of flying squirrels at certain sites also seemed to be regulated by anthropogenic influences such as subsistence hunting by local communities.

Nest site selection by the Mexican gray squirrel, Sciurus aureogaster by Nicolás Ramos (Zoología. Instituto de Biología, Universidad Nacional Autónoma de México, UNAM, México, D.F., MEXICO,)

I studied nest site selection by Mexican gray squirrel (Sciurus aureogaster) in an oak-pine forest located south to the city of Pátzcuaro, Michoacán, in Mexico during 2001. Both the biology and ecology of the species is poorly understood, particularly concerning its habitat requirements. Thus in order to understand more about its nesting behavior, I quantified 20 habitat characteristics and six nest characteristics for a total of 90 nest sites, employing 10×10 m quadrats, with the nest tree at the center of each quadrat. These sites were compared with 90 random sites (availability), and 90 sites taken likewise randomly in patches of the study area where nests were never found throughout the research (non-used sites). Even though squirrels nested in eight of the 10 species of trees located in the forest (Clethra mexicana, Cornus disciflora, Quercus candicans, Q. crassipes, Q. laurina, Q. obtusata, Pinus pseudostrobus, Styrax ramirezii, Symplocos prionophylla, Ternstroemia pringlei), the oaks Q. candicans, Q. crassipes, and Q. laurina were the only used over their proportions taken randomly. To build their nests, squirrels selected on average the tallest trees of their habitat, with the largest trunks, the first major branching higher, the crowns interlocked with those of other trees, in sites with the canopy open or close to it, in steeper slopes, and with dominance of Q. obtusata. Even though the Principal Component Analysis (PCA) explained just 64.91% of the whole variance, it was consistent with the univariate analysis, revealing that nest tree characteristics were the most influential in the selection for nest sites, followed by surrounding vegetation and finally by slope and distance to the nearest road or trail. On the other hand, when nest sites and non-used sites were compared, it was found that the latter are limiting the nesting capacity of squirrels apparently because these maintain on average a lower density of trees than the former. As far as nests are concerned, these were built at the highest parts of the trees, preferably at the upper part of the crowns, in the forks of branches, close to the main trunk, and oriented mostly from southeast to southwest. Likewise, of the six nest characteristics quantified, five were found significantly correlated with certain characteristics of the habitat where squirrels built their nests. When the use of habitat characteristics was analyzed for seasonal differences, it was found that Q. candicans was the species of tree most widely used for nesting during the wet season, whereas Q. laurina was the most widely used in the dry season, apparently in relation to their periods of seed production. On the contrary, nest characteristics never exhibited significant seasonal differences probably because of the slight variations recorded in temperature during the research.

Everything you wanted to know about modelling squirrels, but were too afraid to ask by Rushton, S. P., Lurz, P. W. W., Shirley, M. D. F. and Gurnell, J.

We discuss how modelling can be used to investigate the biology and ecology of squirrels. We illustrate our talk with examples that encompass a wide range of spatial and temporal scales. We consider how modelling can be used to generate theory and show how these approaches can be used tactically in the conservation of endangered species and the management of pests.

A comparison of locomotor performance in eutherian and metatherian gliding mammals. by John S. Scheibe, Liz Flaherty, Winston Smith, Jill Bassham, Ross Goldingay, and Jim Robins(/p>

We used laboratory and field data to evaluate the locomotor performance and cost of transport for the southern flying squirrel (Glaucomys volans), northern flying squirrel (G. sabrinus), sugar glider (Petaurus breviceps), and squirrel glider (P. norfolcensis). Performance of the flying squirrels was markedly different than that of the marsupials. The squirrels exhibited smaller glide angles and faster running and climbing speeds than the marsupials. The cost of gliding transport for the squirrels was less than that of quadrupedal transport after distances of about 10 m, and this was shorter than the mean glide distance. The marsupials climbed tree trunks relatively slowly, and consequently incurred a high cost of gliding transport. Both species of marsupials had cost effective glide distances that were considerably longer than their mean glide distances. Our results suggest that the evolution of gliding locomotion is not a result of selection for reduced cost of transport, but may instead be a response to foraging economics. (/p>

Spacing behaviour of the Siberian flying squirrel – effects of landscape structure by Vesa Selonen & Ilpo K. Hanski (Department of Ecology and Systematics, P.O. Box 65, FIN-00014 University of Helsinki, Finland Email: )

In the present theory of spatial ecology, movement is a key determinant of population dynamics of species. During the last decade the empirical understanding of animal movement has increased, but too little effort has been put into the study of how animals respond to landscape structure. We have studied the space use and movements of the Siberian flying squirrel (Pteromys volans) and the effect of landscape structure on those. This has been done by examining with radio telemetry the movements of individual flying squirrels within their life cycle and by examining with microsatellite variation the genetic differentiation between different areas. Movements and dispersal of the flying squirrel were affected in several ways by the landscape structure and individual behaviour. However, in the landscape of our study areas, the movement of flying squirrels did not actually seem to be much restricted due to the good movement abilities of the flying squirrel. Still, the genetic differentiation between different areas seemed to be high maybe due to low density of suitable forest patches for flying squirrel. Flying squirrels were able to move in wide variety of habitats if there were trees present, did not avoid crossing edges between habitats, and did not need forest corridors for moving between patches. In addition flying squirrels seemed to slightly prefer some types of edges. Results on juvenile dispersal indicated that there are behavioural differences between dispersers. By using modelling approach to combine our data and landscape structure, it should in the future be possible to form predictions on flying squirrel responses to landscape structure in the population level.

Searching for effects of tree squirrel caching behaviour on the distribution of oak seedlings by Peter D. Smallwood (Department of Biology, University of Richmond, Richmond, VA 23173), William Terzaghi (Department of Biology, Wilkes University, Wilkes-Barre, PA 18766), Amy McEuen (Department of Biology, Wilkes University, Wilkes-Barre, PA 18766), John E. Carlson (School of Forest Resources, Pennsylvania State University, University Park, PA 16802. 4Department of Biology, Western Illinois University, Macomb, IL 61455) Eric Ribbens, Tom Contreras (Department of Biology, Wilkes University, Wilkes-Barre, PA 18766) & Michael A. Steele (Department of Biology, Wilkes University, Wilkes-Barre, PA 18766)

Previous work has shown that squirrels (and other mammals) disperse viable acorns from the red oak species group far more often than those from the white oak group. Assuming animal dispersal is important in determining the distribution of oak seedlings, we predict that red oak seedlings should be widely distributed, with a greater mean dispersal distance from parent trees than for white oak seedlings. We are testing this prediction by i) mapping oak stems in forest study plots (potential parent trees and seedlings), and ii) using DNA fingerprinting to establish parent-seedling pairs, allowing us to measure dispersal distances directly. We have established five study plots in four states of the USA (IN, PA, MD, VA), ranging in size from 1.5 to 8 hectares. At this writing, we have isolated DNA from over 1500 seedlings, and over 1200 adult trees. We have parent-offspring matches for two of the five study plots. The data so far do not support our prediction. The average distance between parent trees and their seedlings for white oaks is the same as that for red oaks on these study plots. Further, the vast majority of seedlings appear to have come from adult trees beyond the bounds of our study plots. The results suggest that acorns –especially white oak acorns– are dispersed further from parent trees than we expected. We compare these results with those of the RECRUITS model for estimating seed shadows. We discuss possible explanations for these surprising results.

Controlling grey squirrels for red squirrel conservation in conifer forests by Janie Steele (Forest Enterprise, Thetford District Office, Santon Downham, Brandon, Suffolk IP27 0TJ UK) & John Gurnell (School of Biological Sciences, Queen Mary, University of London E1 4NS )

A three year study on controlling grey squirrels for red squirrel conservation was carried out in Thetford Forest in East England between 1998 and 2001. Multi-capture, ground-placed cage traps were used to capture and remove grey squirrels within the 4600ha study area. The cost of control was based on approximately 0.5 person-years (1.09E-04 person-years ha-1 in year 1 and 1.26E-04 person-years ha-1 in years 2 and 3). To assist in the interpretation of the results, the terms trapping efficiency and control efficiency have been defined. The number of grey squirrels captured each year increased from 688 in Year 1, to 716 in Year 2 and to 805 in Year 3. As expected, the number of animals captured declined during each trap week but squirrels were still being trapped at the end of the week. The results show that the control effort was too low at all stages of the study to reduce grey squirrel numbers significantly, or indeed to maintain them at low numbers, in the centre of the control area. Thus, control efficiency was poor. Additional studies in the summers of 2000 and 2001 involving the use of hair tubes, feeding transects, live trapping and radiotracking before during and after control was carried out in one part of the study area, demonstrated that many grey squirrels were present in the vicinity of the control traps but were not captured during control operations. The findings are discussed and further, specific studies are recommended.

Food-hoarding decisions by tree squirrels: Behavioural mechanisms and ecological and evolutionary consequences by Michael A. Steele, Peter Smallwood, William B. Terzaghi, John E. Carlson, Thomas Contreras & Amy McEuen (Department of Biology Wilkes University, Wilkes-Barre PA 18766, Department of Biology, University of Richmond, Richmond VA 23173, School of Forest Resources, Pennsylvania State University, 113 Ferguson Building, University Park, PA 16802)

Despite evidence that many food-hoarding animals disperse plants, few studies have considered how specific seed characteristics may influence these interactions. We review a decade of experimental studies on the behavioral decisions of tree squirrels in response to seed characteristics and the potential for such decisions to influence the dispersal, establishment, and regeneration of oaks. We first show how four factors (seed size, germination schedules, handling time, and seed chemistry) each independently influence feeding and hoarding decisions and in turn the dispersal of acorns. We further demonstrate how these behavioral decisions collectively result in the selective dispersal of acorns of red oak species over those of white oak. Secondly, we introduce the Habitat Structure Hypothesis, an alternative to the Optimal Density Model (Stapanian and Smith). Our model suggests that gray squirrels regularly risk predation to cache larger more profitable seeds in open habitats as compared with smaller less profitable seeds. Finally we review the ecological and evolutionary implications and consequences of these interactions.

Differential responses of tree squirrels and other small mammals to acorn masting: an evaluation of the predator satiation hypothesis by Steele, M.A., Contreras, T.A., McEuen A.B., Sherick, M., Reed, S. & P.D. Smallwood (Wilkes University, Wilkes-Barre, PA 18766 & University of Richmond, Richmond, VA 23173)

The predator satiation hypothesis suggests that masting evolved in response to seed predators and that seed survival and seedling establishment will only occur during high mast years. Oak species may, however, differ in the conditions required to satiate predators because of specific acorn-hoarding decisions of small mammals. Specifically, small mammals in temperate oak forests are known to selectively disperse and cache acorns of the red oak subgenus (Eythrobalanus) in contrast to acorns from the white oak subgenus (Quercus) which are preferentially eaten. We therefore predicted that oak species would show differential responses to oak masting depending on the degree of synchrony of acorn production across species and subgenera. We hypothesized that caching and recruitment of red oaks will occur in all years of high acorn production and that white oaks should establish only when they mast alone. We tested this prediction by monitoring acorn production and following the fate of tagged acorns (9000/year) at three sites in NE Pennsylvania over four years. As predicted, caching and establishment of northern red oak (Quercus rubra) increased in a year of high red oak and white oak production. White oaks (Quercus alba) failed to mast without red oak in the years of our study and, as predicted, caching for this species was extremely low across all years and sites. We suggest that the conditions necessary for establishment of white oak species are far more limited than for species of red oak. We also argue that an understanding of the behavioral decisions of small mammals may be key to explaining patterns of oak recruitment within temperate forests.

Population dynamics and expansion of the Formosan squirrel introduced to Japan by Noriko Tamura (Email:  )

Formosan squirrels (Callosciurus erythraeus) had been introduced on Izu-oshima Island, Japan in 1930’s. After that, this species has been established in the several parts of Japan. In Kanagawa Prefecture, one of such places, the distribution area expanded to 300 km2during the past 50 years. Japanese squirrels (Sciurus lis), a native species, are also distributed in neighbouring mountain forests in Kanagawa Prefecture. In order to avoid competition between the two species of squirrels, expansion of the alien species have to be suppressed immediately. In the present study, expansion process of the Formosan squirrel is analyzed in respect of population dynamics. In the Formosan squirrel, reproductive females have exclusive home ranges each other. Daughters establish their home ranges adjacent to their mothers’ one. Therefore, the distribution area of this species is expected to have a positive relation with the number of females. The population parameters, such as reproductive and survival rates, were estimated by the mark-recapture study for 7 years. The mean litter size was 1.32 + 0.47sd (n=47) and the mean number of litters per year was 1.16 + 0.68 sd (n=38). The annual survival rate of the females was 0.60 for less than 1 year, 0.79 for 1year old, 0. 58 for 2years old, and 0.09 for 3 years old. Thus, the intrinsic rate of increase was calculated as 0.0862. When we assume an exponential population growth, distribution areas in log- translated forms increase with time after invasion with a slope of the intrinsic rate of increase (0.0862). In fact, expansion of distribution areas well fits to the predicted line of this exponential population growth model.

Seeing the forest for the trees: nest-site selection of the endangered Mount Graham red squirrels (Tamiasciurus hudsonicus grahamensis) by S.R.Taylor & J.L. Koprowski (School of Renewable Natural Resources, The University of Arizona, College of Agriculture and Life Sciences, Tucson, AZ.),/p>

The dependence of tree squirrels on mature forests includes their need for appropriate forest structure to provide nest cavities and/or sites for drey construction. Red squirrels build dreys or make use of existing cavities for rearing young, sleeping and protection. The nesting ecology of many species is poorly known due to the difficulty of detecting the exact location of active nests. We studied the nesting behaviours of the federally endangered Mount Graham red squirrel (Tamiasciurus hudsonicus grahamensis) from June 2001 to April 2003 in the Pinaleno Mountains of southeast Arizona. Various factors were examined using intensive observations and radio-telemetry. Standard methods were employed to assess focal tree, nest, and habitat characteristics of each plot. Cavity nests were found in taller and larger trees than drey nests, and trees that housed drey nests had more interdigitated branches than trees with cavities. Dreys were located in a higher percentages of corkbark fir (Abies lasiocarpa) and Engelmann spruce (Picea engelmanni) relative to abundance, while cavities were found more frequently in quaking aspen (Populus tremuloides). Nest entrances tended to face west. A number of animals were found to use more than one nest over the observation period and a few pairs were found nesting together in the same location. Although nesting habits of North American red squirrels have not been extensively explored previously, this information may have significant management implications for manipulations of red squirrel habitat, especially in a habitat isolate like that of the Pinaleno Mountains.

Effects of habitat fragmentation on red squirrels – what extra information do we get from genetic analyses? by Goedele Verbeylen, Peter Galbusera, Luc De Bruyn & Erik Matthysen (Email:)

In large parts of Western Europe, the habitat of the Red squirrel (Sciurus vulgaris L. 1758) is strongly fragmented. To gain more insight in the functioning of squirrel populations in forest fragments (dispersal, extinctions and colonisations, inbreeding, …) an intensive study was conducted in 1993-2001. The study area (Antwerp, Belgium) contains 58 woodland patches (8 larger than 10 ha) covering +/- 7 % of the total area. Squirrels were (re)captured and radio-tracked regularly in all fragments. DNA fingerprinting was used to study parenthood (to determine the origin of new squirrels that come into the population) and genetic diversity of local populations. In this talk, the emphasis lies on the dispersal data and the extra information provided by the genetic analyses. We used paternity assignment to find out the origin of newly captured squirrels. These results, together with the detailed life history information of each squirrel, allowed us to indicate a much higher amount of squirrels as immigrants than we already knew from the radio-tracking data. This is due to the fact that young squirrels are difficult to capture and radio-collar and often have dispersed already before first capture. The probability of a correct assignment was very low, though. This may be caused by a large influx of squirrels from outside the study area, for which we don’t have DNA samples from their parents. Genetic analyses can make an important contribution to understanding the functioning of a fragmented population, but should be handled with a great deal of caution.

Population regulation in tree squirrels: effects of food abundance and density dependence by Luc A. Wauters (Department of Structural and Functional Biology, University of Insubria, Varese, Via J.H. Dunant 3, I-21100 Varese, Italy)

Theory on population regulation emphasises that both exogenous factors (no dynamic feedback with population density) and endogenous factors (with dynamical feedbacks affecting population numbers) affect population change, and that their relative contribution will vary between different population systems. Regulating factors, by definition, have a density-dependent effect, involving negative feedback in response to changes in population size Consequently, a regulated population shows constrained fluctuations, and the propensity to increase when small and decrease when large. Here I will first review the major results of long-term studies on squirrel population dynamics, presenting evidence of population regulation. Second, I will present data on a 9-year study of two Eurasian red squirrel populations from different high-quality habitats in Belgium. This study had two main goals: (i) investigating the relationships between exogenous and endogenous factors and population dynamics, and to unravel the interactions (within and between sexes) in the context of population regulation; (ii) testing for the effects of food availability and of density of each single sex on changes in density and on seasonal and annual variation in the various population parameters of animals of the same and of the opposite sex. In other words, we investigated whether we can improve our understanding of populations dynamics and regulation by considering how demographic parameters may be influenced differently according to gender. I will discuss the relative importance of tree seed abundance and gender-specific density on various population parameters and how they affect seasonal and annual changes in population growth of male and females red squirrels.

Modelling the role of competition and disease on the replacement of red squirrels by greys by Andrew White (Mathematics, Heriot-Watt University, Edinburgh)& Dan Tompkins (Biological Sciences, University of Stirling and Mike Boots, Biological Sciences, University of Stirling)

A simple mathematical model is developed to investigate the role of parapoxvirus and competition in the replacement of red squirrels by greys. We focus on the time taken to transform the disease-free red squirrel population equilibrium to either the disease-free grey squirrel population equilibrium (when considering competition-mediated replacement) or the grey squirrel population equilibrium with endemic infection (when considering competition/infection-mediated replacement). We also develop a spatial model of the squirrel/virus system by linking a grid of patches containing the temporal model equations by dispersal. We use the best available parameters for the squirrel/virus system and examine whether the spatial model with or without the disease could replicate the dynamics of the well documented expansion of grey squirrels and decline of red squirrels that occurred in Norfolk (England) from 1960-1982, for which data is available (Reynolds 1985).

How small is too small: minimum viable population size in tree squirrels by David Wood & John L. Koprowski (Mount Graham Red Squirrel Monitoring Project, School of Renewable Natural Resources, University of Arizona)

Conservation of species requires knowledge of the ability of a species to recover from low numbers. Minimum viable population sizes therefore can be used in conservation applications as a tool to gauge a population’s likelihood of persistence. By studying the minimum viable populations of tree squirrels, conservation biologists will have a better idea of how to manage these species. Minimum viable population numbers, however, require a myriad data for analysis, thereby making study of endangered species even more difficult. By combining data from a variety of closely related species to overcome a dearth of information, useful estimates can be generated. Using this innovative approach and reducing the complexity of the model, minimum viable population numbers were generated for Sciurus aberti, S. carolinensis, S. niger, S. granatensis, S. vulgeris, and Tamiasciurus hudsonicus. Nearly 75% of tree squirrel species are of precarious conservation status, in some portion of their range. VORTEX simulations were run for three different scenarios of breeding success, with three different levels of variability. S. aberti had the lowest MVP and S. niger the highest MVP with values ranging from 15 to 125 at low variability and 15 to 195 at high variability. Increased variability was associated with a rise in MVP size and correlation was found between MVP size and susceptibility to variability. Although the smallest MVP sizes were found from the relatively simple and optimistic conditions, the results suggest that even small populations might be salvageable under favorable conditions. Empirical evidence from the introductions of tree squirrels supports this result. Many cases of success involved introductions of low numbers similar to the results of this modelling experiment. In addition, with this new approach, looking for patterns among similar species to increase data for MVP estimates, ecologists can strive to generate additional MVP estimates for endangered species.

Life-history tactics of Eurasian red squirrels in subalpine conifer forests by Massimiliano Zaninetti, Luc A. Wauters, Ambrogio Molinari & Guido Tosi (Department of Structural and Functional Biology, University of Insubria, Varese, Via J.H. Dunant 3, I-21100 Varese, Italy)

(p>Eurasian red squirrel (Sciurus vulgaris) populations occurring in the alpine conifer forests might constitute the only stronghold against spreading introduced eastern gray squirrels (Sciurus carolinensis) in Central-Europe. Therefore, We studied the ecology of red squirrels in montane and subalpine conifer forests (Valtellina and Stelvio National Park, Italy), investigating phenotypic adaptations to different conifer habitats, seasonal and annual variation in food resources (conifer seeds, fungi) and possible relationships with squirrel demography and spacing behaviour. Here, we present preliminary results of our data gathered over different time-periods (one to four years) in five study sites that differ in tree species composition and elevation. Food availability was estimated by counting fresh cones directly in the trees in summer (July-August) of each year. The presence/abesnce of squirrels and their relative abundance were monitored using hair-tube surveys in spring and autumn. Population dynamics and space- and habitat use were studied using capture-mark-recapture techniques and radio-tracking. We found that conifer seed production fluctuated annually and differed between forest types, affecting spatio-temporal variation in squirrel density. All population parameters, residency (local survival), reproductive rates and immigration rates, were highly variable both at the temporal and spatial (habitat) scale. We will discuss relationships between phenotypic characteristics, food availability and squirrel dynamics.

Comments are closed.