Evaluation of ecological network effectiveness by means of gene flow analysis

The aim of this project is to study the fragmentation in broad-leaved forests, so the species selected were mainly linked to this ecosystem. Almost all amphibian species are model candidates for studies of fragmentation effects on connectivity: in fact most amphibian species occur as metapopulations; in addition, they are a taxon particularly susceptible to isolation, as they generally have low dispersal capabilities and are rather philopatric to breeding sites. These characteristics often lead to high genetic differentiation, even at restricted scales. In particular we choose the Fire Salamander (Salamandra salamandra, AMPHIBIA, URODELA) that is strongly linked to broad-leaved forest ecosystems, depends on ecosystem processes (such as those that allow the development of an adequate litter structure); is strongly affected by specific resources (such as hydrology, some chemical and physical water parameters); and it is also limited by a low dispersal capability. However not only amphibian species are affected by habitat fragmentation, also arboreal mammals, such tree squirrels and dormice, are often more threatened than other species due to their low dispersal capability in absence of structural connection between habitat patches. In European regions, various single-species studies have been carried out, suggesting similar responses in this guild of mammals. Among them, the Hazel Dormouse (Muscardinus avellanarius, RODENTIA, GLIRIDAE) is an arboreal rodent living in woodland and hedgerows, whose presence is negatively affected by fragmentation. For its poor conservation status and negative long-term trend in almost all Europe, the species is included in the Annex IV of the 92/43/ECC Directive (Habitat Directive), concerning species that require a strict protection regime that must be applied across their entire natural range within the EU. For this reason we sampled also the Hazel Dormouse that can be defined as a focal species sensu Lambeck (1997) at least for three out four ecological traits (they are limited by resources, dispersal capability, and ecological processes). We choose two different areas of study, both affected by the habitat fragmentation: Lombardy Region (Northern Italy, where we worked on the Fire Salamander), and Latium Region (central Italy, where we sampled the Dormouse). In both regions we sampled a fragmented forest area and a continuous one, our control.