Conservation biogeography of large Mediterranean islands. Butterfly impoverishment, conservation priorities and inferences for an ecologica “island paradigm”


  • Leonardo Dapporto,

  • Roger L. H. Dennis

L. Dapporto (, Istituto Comprensivo Materna Elementere Media Convenevole da Prato via 1° Maggio 40, IT-59100, Prato, Italy. – R. L. H. Dennis, NERC centre for Ecology and Hydrology, Monks Wood, Abbots Ripton, Huntingdon, Cambridgeshire PE28 2LS, UK, and Inst. For Environment, Sustainability and Regeneration, Mellor Building, Staffordshire Univ., College Road, Stoke on Trent ST4 2DE, UK.


Conservation biogeography is considered the Cinderella of biological conservation. Nevertheless biogeography provides the basis for establishing species distributions over space and time, therefore conservation priorities among areas and individual species. We demonstrate that there is no need to simplify analyses by using subsets of species (rare species, endemics) as surrogates. In doing so, we apply strict biogeographical techniques to determine butterfly impoverishment on three of the west Mediterranean's largest islands (Sardinia, Corsica and Sicily). The analyses performed on species, both collectively and individually, reveal that regional species richness in the Mediterranean zone can be largely predicted by latitude, altitude and latitudinal range (maximum minus minimum latitude), but that Sardinia and Corsica have clearly impoverished faunas. Logistic regression at individual species level demonstrates that several species, predicted to be present in these islands on the basis of their continental distributions, are actually absent. When compared with species that are present in these islands, such missing species are disclosed as having ecological traits which reduce their colonization capability. Probabilities of occurrence are calculated for each species on each island; they reflect the potential for each butterfly species to migrate to and colonise each island, and can be considered as a measure of conservation value. As such, species present on islands but having low immigration probabilities are predicted to represent isolated populations from the mainland that are unlikely to re-colonize the islands in the case of extinction. Island endemic species and races are shown to have lower occurrence probabilities compared to widespread species occurring on islands and illustrate the usefulness of occurrence probabilities for identifying isolated populations in need of conservation attention.