Drivers of vertebrates'richness and diversity in Baixo Vouga Lagunar region

• Autor(a) principal: Sales, Manuela Cordeiro
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/25777
• topic: Baixo Vouga Lagunar; Hotspots; Diversity measures; Ecological modelling; Mosaic landscape; Vertebrates; Wetlands

The identification of the factors driving species richness and diversity variations in a landscape has key role in ecology and in biodiversity conservation plans. In this study, we focused our attention on different vertebrates’ taxa present in the Baixo Vouga Lagunar (BVL) and aimed to identify biodiversity hotspots in this Portuguese region. To achieve that goal, we reviewed ecological studies that targeted terrestrial vertebrates inhabiting the BVL and used their datasets as a basis for our study. Then, we collected ecologically relevant environmental variables, established three ecological hypotheses encompassing those variables (H1: Landscape composition; H2: Anthropic disturbance; H3: Water presence), and tested them to assess which drivers can restrain or promote the BVL vertebrates’ biodiversity patterns (species richness and diversity).The test of those hypotheses was implemented using an ecological modeling approach (Generalized Linear Mixed Models - GLMM) and later, based on the identified drivers, we forecasted species richness and diversity values to the study area. We expected to be able to identify vertebrates’ richness and diversity hotspots in the BVL, which would be a crucial tool to efficiently manage this area and assure the maintenance of its biodiversity values. Unfortunately, we faced data limitations, which affected model building robustness, resulting in low accuracy, and consequently in limited feasibility of the models’ predictive capacity. Therefore, in our forecasted map we could not define specific areas as hotspots, which is unrealistic. In the future, a well-defined sample design and more intense sample effort is needed in order to encompass a broader range of sampling points and promote a more robust model.