Host plant-mediated effects of climate change on the occurrence of the Alcon blue butterfly (Phengaris alcon)

Among the expected consequences of climate change are shifts in species’ ranges. Most of current methods to predict such shifts in species distributions consider changes in suitability of climatic conditions for existence. With these models, it is possible to indicate the potential distribution of species that would arise under spatial conditions that cause unlimited landscape accessibility and habitat suitability. At the regional scale, however, detailed predictions of changes in species distributions and performance are pivotal for conservation planning. This study aims to predict species occurrences at the regional scale, incorporating demographic processes and dispersal to assess habitat accessibility and suitability in detail. We investigated a system with trophic dependence: the Alcon blue butterfly (Phengaris alcon) is fully dependent on the occurrence of its host plant species marsh gentian (Gentiana pneumonanthe). We applied a model chain, consisting of a soil and biomass model, a plant species occurrence and dispersal model and a butterfly metapopulation model. We investigated the effect of future climate change, both under affected and unaffected habitat conditions as determined by host plant occurrence. Our modelled results show that the butterflies perform best when habitat conditions remain unaffected by climate change. However, when climate change does affect the occurrence of its host plant species, butterfly distribution and performance will be deteriorated. This implies that detailed predictions of changes in species distributions and performance should incorporate dispersal, demographic processes and biotic interactions explicitly. Our approach allows for the identification of locations that are potentially suitable for the measures increasing network robustness for P. alcon.

► Current envelope models consider changes in suitability of climatic conditions. ► We project species occurrences incorporating demographic processes and dispersal. ► We applied a model chain to a trophic system (soil – host plant – butterfly). ► Detailed projections should incorporate processes and trophic relations explicitly.