Effects of habitat loss on populations of the eastern green lizard <em>Lacerta viridis</em> at the core and periphery of its distribution range

Abstract

Habitat loss is the main threat to biodiversity. Modified landscapes resulting from habitat loss are characterized by a reduced amount and patchy distribution of habitat, and by changes in landscape composition and landscape configuration (spatial relations between landscape elements). Populations inhabiting such landscapes face reduced connectivity and gene flow, population decline, higher vulnerability to stochastic processes, and eventually, extinction. Ecological specialization is one of the main traits predicting vulnerability of species to habitat loss. It has been mainly studied at the species level, with insights at the intraspecific level being barely addressed in the habitat loss literature. The Kühnelt principle of regional stenoecy states that the range of colonizable habitats and suitable conditions is wider at the core of the distribution range of species compared to the periphery. Then, populations living at the core are habitat generalist, and those in the periphery habitat specialist. This implies that peripheral populations would have a higher sensibility to habitat loss compared to populations at the core, and also that different conservation measures are necessary to protect the species in each region. The study of occupancy patterns of habitat patches across modified landscapes is one of the most useful tools to study effects of habitat loss on the persistence of populations, and can be applied to test differential sensitivity to habitat loss among populations of the same species in different regions. On the other hand, because this approach depends on the extinction of populations to find patterns and effects, in order to identify vulnerable population before they irreversibly decline, morphological and physiological parameters representative of the status of individuals have also been proposed to be used as early warning indicators of populations’ stress. In this project I combined extensive fieldwork with advanced methods in landscape ecology and statistics to study populations of the eastern green lizard <em>Lacerta viridis</em> located at the core (Bulgaria) and at the northern periphery of its distribution range (Germany, Czech Republic). <em>L. viridis</em> is protected under the EU Habitats Directive and is threatened to extinction in Germany and Czech Republic. The objectives of this research are to test the Kühnelt principle, evaluate if predictions of specialization at the species level regarding effects of habitat loss on occupancy patterns are also met at the intraspecific level, and evaluate the suitability of morphological and physiological parameters as early indicators of populations stress. My results show that peripheral populations of <em>L. viridis</em> are specialist in comparison to core ones, have a smaller niche size and select microhabitats based on different environmental parameters regions (Chapter 3). Specialization predictions of vulnerability to habitat loss were also met at the intraspecific level. In the periphery, in comparison to the core, populations depended more on habitat quality, the effects of individual landscape composition predictors were stronger and overall habitat loss had an impact at smaller scales. Moreover, regions differed in the parameters of landscape structure that affected occupancy patterns the most (Chapter 4). Finally, I identified two parameters of individual status -body condition and fluctuating asymmetry- that are suitable to be used as early indicators of stress for populations of <em>L. viridis</em> inhabiting modified landscapes (Chapter 5). This dissertation is a contribution to the knowledge of the ecology of <em>L. viridis</em> and to the improvement of conservation measures to protect the species across its range. This work also broadens the knowledge about intraspecific regional differences in ecological traits of species and the application of traits at the intraspecific level to predict geographically-dependent populations’ sensibility to habitat loss.