| Resumo: | The effect of scale on species distribution has been highlighted by several authors and indicates the necessity of incorporate multi-scale approaches in conservation analyses. Several populations of European wildcat (Felis silvestris silvestris) are suffering considerable threats and thus conservation measures are vital. Conservation decisions have been based on the availa-ble information. However, most of wildcat studies have been conducted at fine scales, and thus may only represent part of the full picture. This thesis aims to provide the first insights on eco-logical factors constraining wildcat distribution in Scotland at local and national level. Several models based on food, land cover, disturbance, topographic and climatic variables were built to explain wildcat distribution at both scales. The results show that wildcat presence is influenced by the interaction of different types of environmental variables. At broad-scale, rabbit presence and rodent diversity showed to be the strongest factors influencing the species occurrence. However, a negative association was detected between rodent diversity and wildcat locations at finer scale, but conversely locations were associated with higher rodent abundance. Rabbit abundance was not detected as an influent factor at this level. Regarding land cover features, whatever the scales considered, wildcats benefited from higher habitat diversity. Wildcat pres-ence at both scales was related to grassland patches and negatively associated with heather moorland. At local-level wildcats seem to be associated with woodland patches and riparian habitats. Disturbance influence was not detected, at least directly, at broader scale, but human settlements were positively associated with wildcats at finer scale, while closer distance to roads was related to lower wildcat occurrence. A direct influence of climatic conditions on the target species was not detected. However, areas theoretically associated with smoother climatic fea-tures were related to wildcat presence at broad scale. According with our main findings, the definition of special areas for wildcat conservation in Scotland should take into account multi scale approaches. At broader scales, areas should incorporate food rich patches (rabbits and rodents), specific land cover (e.g. woodland and grasslands) and less mountainous areas (topo-graphic characteristics), while at the individual level areas showing higher levels of prey availa-bility, heterogeneous land cover and few disturbance features should be the main target for as-suring wildcat long term persistence. |