Nest site selection by common dormouse (Muscardinus avellanarius) in a coastal evergreen scrub forest of Central Italy

Even though the common dormouse (Muscardinus avellanarius) has been observed in a variety of woodland types, yet appears to have specific habitat requirements. Deciduous woodlands with a developed understorey are considered to be the best habitats for this species. Nevertheless, his preferences have been studied only in North Europe, and there is no information about the optimal habitat for dormice in southern European countries. Common dormice build arboreal globular nests that conceal the animals from predator sight and protect them from environmental conditions. These nests are supported by understorey branches or constructed inside of tree hollows. Moreover, they can be constructed inside nest boxes which represent artificial substitutes of tree hollows, and are used to monitor dormouse populations. Several researchers described the general features of the habitat surrounding arboreal nests of dormice, while no study has been focused to quantify the habitat features of the immediate neighbourhoods of the nests. We hypothesized that the specific characteristics of the habitat could induce them to select an arboreal site. The objectives of this study were (1) to identify the specific habitat features that affected the selection of arboreal nest site by common dormice in a typical coastal Mediterranean scrub forest of central Italy and (2) to verify if nest box occupation rate differed between sexes and age classes. The study was carried out in a thermophilous evergreen scrub forest dominated by holm oak (Quercus ilex), within the Natural Reserve of Castelporziano, along the Tyrrhenian coast of Latium (41°44'N, 12°24'E). Fifty nest boxes were set up in the study site and were inspected monthly from May 2002 to December 2004. Trapped dormice were weighted, sexed, aged and marked. In order to examine nest-site selection, we compared habitat features at nest boxes occupied by reproductive females to those at nest boxes never occupied. Several habitat variables have been measured for three vegetation layers (overstorey: DBH &#8805; 7.5 cm, > 2 m tall; understorey: DBH &#8804; 7.5 cm, 2 m tall; shrub: < 2 m tall), within a 3m-radius circular plot, a 10m-radius circular plot and two perpendicular transects (20 x 2 m) centered on nest-box. Nest boxes were used more frequently by females than males (ratio: 2.37:1), and this result depended on a different nesting behaviour between adult females and adult males. In fact, juvenile/adult ratio and juvenile-male/juvenile-female ratio were both 1:1. Moreover, we observed into nest boxes a lot of females with offspring (60.52%). We suggested that breeding females and juveniles (both sexes) preferred nesting into boxes because these offered them a high protection from climatic conditions and predators, affecting reproductive success and survival. On the contrary, it is possible that adult males were more interested to defend their ranges than boxes. In particular, we may suppose a territorial and polygynous structure of our common dormouse population with more females living in a restricted part of male’s range. In relation to nest site selection, common dormice showed to prefer sites with a high tree density and high branch overlapping. In particular, they showed to prefer sites characterised by: (1) a high understorey density, (2) a high necromass density of Erica arborea, (3) a high number of contacts between wood branches and (4) a high number of contacts between climbing and woody vegetati