Projeto de pesquisa

This study will encompass fish assemblages associated to reef of both biogenic (e.g., coral) and rocky origin, and will be based on the analysis of two extensive datasets. A first dataset comprises reef fish species occurrences (checklist) extracted from published literature (Kulbicki et al. 2013; Parravicini et al. 2013)and online repositories. The second dataset is composed of standardized underwater visual transects over 31 sites distributed across the Atlantic Ocean. Most of the data has been collected by the applicant (Quimbayo et al. 2018)and other collaborators. Other datasets have been provided to the applicant by close collaborators who have granted full permission to for their use in the analyses proposed here [Michecl Kulbicki (IRD-France), Sergio Ricardo Floeter (UFSC-Brazil), Carlos Eduardo L. Ferreira(UFF-Brazil), Sonia Bejarano (ZMT, Germany)]. Two abiotic factors will be used as predictors of functional structure throughout Atlantic Ocean. These predictors include annual mean sea surface temperature (SST) and primary productivity (PP) estimated from mean surface chlorophyll a values, for each locality in the EP. Both variables will be estimated from satellite data averaged between 2002 and 2009 from BIO-oracle (Tyberghein et al. 2012).

Each species is arranged in a multidimensional functional space according to its functional niche (Villéger et al. 2008). The functional niche - defined by the set of biological attributes of the species - is used to position the species in relation to the others that constitute the community in question. Since the attributes that define the functional groups of reef fish are not necessarily continuous, a functional space must be constructed (Villéger et al. 2011). The first step corresponds to the estimation of Gower distance between pairs of species of the communities based on species attributes (Villéger et al. 2008). The second step is the Principal Coordinate Analysis (PCoA), applied in the dissimilarity matrix (Villéger et al., 2008, 2011), to generate coordinates for the positioning of species in the multidimensional space.

Once species are distributed in their functional space, estimates of the proportion of this space that is occupied by different communities (Villéger et al., 2008) will be obtained. Focus will be given to species belonging to different functional groups, and groups with variable redundancy. This space is measured through the volume within the convex polygon involving the species (Villéger et al., 2011). The convex polygon and therefore the functional space of reef fish communities will be estimated using the geometry package available in the R software, and specifically the conve-hull function.