Combining molecular and landscape tools for targeting evolutionary processes in reserve design: An approach for islands.

The importance of targeting ecological and evolutionary processes in reserve design has been widely acknowledged in the literature but rarely implemented on islands. Using Socotran reptiles as models, we aim to relate richness of widespread and restricted-range species directly with landscape variables and to compare the impact of setting conservation targets for lineages versus species. Socotra Island is a UNESCO Natural World Heritage Site, containing high levels of endemism in relation to its area, especially of reptiles, the vertebrates with the most comprehensive available genetic data. We predicted the occurrences of reptile species using distribution models and used a novel approach to interpolate maps of spatial phylogenetic patterns. Patterns of intra and interspecifc diversity and differences between spatial outputs of lineage and species richness were related to eco-geographic variables. We evaluated differences in target achievement for each conservation unit within protected areas (PAs) under the current Zoning Plan (ZP) using gap and reserve design analyses. Although intraspecific richness was strongly correlated with interspecific richness, differences in their spatial distribution reached ~30% in some areas. Differences were more pronounced for wide-ranging than restricted-range taxa. Gap analysis indicates that most conservation units are under-represented in sanctuaries and that intra and interspecific richness were significantly higher outside PAs. This work will guide local-scale conservation planning as the ZP is due to be re-evaluated. This is one of the few studies on islands using genetic data from an entire class of vertebrates to incorporate lineage diversity in reserve design. This study provides an alternative methodological framework for supporting the use of landscape and genetic tools in reserve design, circumventing the use of phylogenetic distances and deterministic spatial interpolation of lineage diversity that can be widely applied to other systems.