Stock structure and vulnerability of commercially exploited fishes in Brazil

Sustainable management of fish stocks is critical to the conservation of marine biodiversity, but is often hindered by knowledge gaps, which include the spatial distribution of fish stocks, how this distribution is influenced by environmental characteristics and the consequences of connectivity to genetic variation. In this thesis I describe the conservation status, effective population size, genetic structure and connectivity of commercially important marine fishes in Brazil. The first chapter presents an analysis of catch data for 132 species over 61 years. The results indicate that the magnitude of population decline is influenced by the type of exploitation and complex interactions between climatic and biological factors. The second chapter analyzes mitochondrial DNA (mtDNA) at COI and CytB and describes the genetic structure and historical demography for 17 species of commercially exploited marine fish. Here, the genetic structure implies a lack of environmental barriers to gene flow along the Brazilian coast for 15 of these species. The exceptions were the mtDNA structure identified for Pomatomus saltatrix and Cynoscion jamaicensis, which indicate genetic discontinuity due to temperature changes and isolation by distance, respectively. The goal of the third chapter was to, through the analysis of single nucleotide polymorphisms (SNPs), characterize the effective population size of two reef species - Lutjanus jocu and Sparisoma axillare - and examine the influence of past climatic variability and habitat suitability on changes on the effective population size. The population of Sparisoma axillare expanded until five thousand years ago due to increased habitat suitability, and has been mostly stable since then. The population of L. jocu has been stable in the last one million years, although an expansion was detected prior to this period. Although L. jocu currently finds less suitable habitats than in the past, this change does not seem related to population size estimates. The distribution of L. jocu, from south Brazil to Caribe, is larger than that of S. axillare, which is endemic to Brazil, so the first is likely less susceptible to regional climate change. In the fourth chapter, I identified which seascape characteristics explain patterns of genetic connectivity for S. axillare, by analyzing samples collected at six locations on the Brazilian coast and also at the islands of Fernando de Noronha, Abrolhos and Trindade, which are located at various distances from the coast. The results indicate that only individuals at Trindade Island, the most distant one, are genetically distinct. I found that seascape factors, such as bathymetry and oceanic currents are better at explaining genetic variation than geographical distance. The results of this thesis suggest that some environmental variables affect species conditions: temperature can positively affect some stocks, but species vulnerable to warmer temperatures have a higher chance of collapse; Sparisoma axillare, a species with vulnerable characteristics, seems to be sensitive to past changes in habitat suitability, whereas depth and oceanic currents influence its population connectivity, increasing the risks of its isolated population -- summary.