Projected Impacts of Climate and Land Use Change on Endemic Plant Distributions in a Mediterranean Island Hotspot: The Case of Evvia (Aegean, Greece)

Abstract

Anthropogenic climate and land use change pose major threats to island floras worldwide, yet few studies have integrated these drivers in a single vulnerability assessment. Here, we examine the endemic flora of Evvia, the second-largest Aegean island in Greece and an important biodiversity hotspot, as a model system to address how these disturbances may reshape species distributions, community composition, and phylogenetic diversity patterns. We used species distribution models under the Ensemble of Small Models and the ENphylo framework, specifically designed to overcome parameter uncertainty in rare species with inherently limited occurrence records. By integrating climate projections and dynamic land use data, we forecasted potential range shifts, habitat fragmentation, and biodiversity patterns for 114 endemic taxa through the year 2100. We addressed transferability uncertainty, a key challenge in projecting distributions under novel conditions, using the Shape framework extrapolation analysis, thus ensuring robust model projections. Our findings reveal pronounced projected range contractions and increased habitat fragmentation for all studied taxa, with more severe impacts on single-island endemics. Our models demonstrated high concordance with established IUCN Red List assessments, validating their ecological relevance despite the sample size limitations of single-island endemics. Current biodiversity hotspots, primarily located in mountainous regions, are expected to shift towards lowland areas, probably becoming extinction hotspots due to projected species losses, especially for Evvia’s single-island endemics. Emerging hotspot analysis identified new biodiversity centres in lowland zones, while high-altitude areas showed sporadic hotspot patterns. Temporal beta diversity analysis indicated higher species turnover of distantly related taxa at higher elevations, with closely related species clustering at lower altitudes. This pattern suggests a homogenisation of plant communities in lowland areas. The assessment of protected area effectiveness revealed that while 94.6% of current biodiversity hotspots are within protected zones, this coverage is projected to decline by 2100. Our analysis identified conservation gaps, highlighting areas requiring urgent protection to preserve future biodiversity. Our study reveals valuable information regarding the vulnerability of island endemic floras to global change, offering a framework applicable to other insular systems. Our findings demonstrate that adaptive conservation strategies should account for projected biodiversity shifts and serve as a warning for other insular biodiversity hotspots, urging immediate actions to maintain the unique evolutionary heritage of islands.