Bias in Diet Determination: Incorporating Traditional Methods in Bayesian Mixing Models

摘要: There are not “universal methods” to determine diet composition of predators. Most traditional methods biased because their reliance on differential digestibility and the recovery hard items. By relying assimilated food, stable isotope Bayesian mixing models (SIMMs) resolve many biases methods. SIMMs can incorporate prior information (i.e. proportional composition) that may improve precision in estimated dietary composition. However few studies have assessed performance with without informative priors study predators’ diets. Here we compare compositions South American fur seal sea lions obtained by scats analysis SIMMs-UP (uninformative priors) assess whether (SIMMs-IP) from scat improved compared SIMMs-UP. According SIMM-UP, while pelagic species dominated seal’s lion’s did a clear dominance any prey. In contrast, SIMM-IP’s diets were same preys as analyses. When influenced SIMMs’ estimates, incorporating at risk inducing estimates. If isotopic data allow discriminating preys’ contributions diets, should lead more precise but unbiased Just estimates critically interpreted biases, care must be exercised when interpreting SIMMs-IP. The best approach obtain near-complete view involve simultaneous consideration different sources partial evidence (traditional methods, SIMM-UP SIMM-IP) light natural history predator so reliably ascertain weight yielded each method.