Model-based riverscape genetics: disentangling the roles of local and connectivity factors in shaping spatial genetic patterns of two Amazonian turtles with different dispersal abilities

Genetic patterns are shaped by the interaction of different factors such as distance, barriers, landscape resistance and local environment. The relative importance of these processes may vary for species with different ecological traits. Here we compared two related Amazonian riverine turtle species (Podocnemis erythrocephala and Podocnemis sextuberculata) with distinct dispersal abilities to assess how differently local and connectivity variables influence their genetic patterns. We used a total of 609 genetic samples to estimate mitochondrial (mtDNA) genetic diversity and differentiation for each locality. We applied model selection on models associating genetic diversity to local variables representing hypotheses of climate and primary productivity, water level variation, hunting pressure and downstream increase in genetic diversity. We modeled the relationship of genetic differentiation with connectivity variables representing hypotheses of isolation by distance (IBD), isolation by resistance (IBR) and isolation by barrier (IBB). Model selection for genetic diversity was only important (excluded the null model) for the high-dispersal species (P. sextuberculata), with best models including hypotheses of productivity and hunting pressure. Genetic diversity was higher in more productive sites and in sites with higher concentration of villages (opposed to expected). Although a variable importance testing showed low importance for connectivity models, IBB (Amazon River) and IBR (resistance by current and past climatic suitability and river color) models explained more genetic differentiation turnover than IBD (riverway distance). Models explained a higher percentage of genetic differentiation for the low-dispersal species (P. erythrocephala), with Amazon River as main predictor. We show that, although local variables are often overlooked in riverscape genetics studies, they can influence intrapopulacional genetic diversity of aquatic species, even those with high dispersal ability. By applying a resistance-model framework and by using riverscape genetics factors relevant in basin-wide context, we provide a novel approach to investigate genetic patterns of other aquatic vertebrates in fluvial systems.