Abstract
Fish have one of the highest occurrences of individual specialization in trophic strategies among Eukaryotes. Yet, few studies characterize this variation during trophic niche analysis, limiting our understanding of aquatic food web dynamics. Stable isotope analysis (SIA) with advanced Bayesian statistics is one way to incorporate this individual trophic variation when quantifying niche size. However, studies using SIA to investigate trophodynamics have mostly focused on species- or guild-level (i.e., assumed similar trophic strategy) analyses in settings where source isotopes are well-resolved. These parameters are uncommon in an ecological context. Here, we use Stable Isotope Bayesian Ellipses in R (SIBER) to investigate cross-guild trophodynamics of 11 reef fish species within an oceanic atoll. We compared two- (δ (15)N and δ (13)C) versus three-dimensional (δ (15)N, δ (13)C, and δ (34)S) reconstructions of isotopic niche space for interpreting guild-, species-, and individual-level trophic strategies. Reef fish isotope compositions varied significantly among, but also within, guilds. Individuals of the same species did not cluster together based on their isotope values, suggesting within-species specializations. Furthermore, while two-dimensional isotopic niches helped differentiate reef fish resource use, niche overlap among species was exceptionally high. The addition of δ (34)S and the generation of three-dimensional isotopic niches were needed to further characterize their isotopic niches and better evaluate potential trophic strategies. These data suggest that δ (34)S may reveal fluctuations in resource availability, which are not detectable using only δ (15)N and δ (13)C. We recommend that researchers include δ (34)S in future aquatic food web studies.