EcoDiet

EcoDiet implements a hierarchical Bayesian framework to estimate diet matrices and infer food-web topology and consumer diets by integrating literature data, stomach content analyses, and biotracer mixing-model (MM) data.

Key Features:

• Integration of Multiple Data Sources: Combines literature data as priors, frequencies of prey occurrence from stomach content analyses, and biotracer data incorporated via a mixing model (MM).
• Hierarchical Bayesian Modeling: Uses a hierarchical Bayesian framework to jointly estimate food-web topology and diet compositions across all consumers while quantifying uncertainty.
• Model Validation and Comparison: Employs simulated datasets and direct comparisons with classical network mixing-model (MM) approaches to evaluate performance.

Scientific Applications:

• Diet matrix estimation: Produces quantitative diet matrices to inform prey–predator interaction strengths.
• Food-web topology inference: Infers consumer–resource links and trophic structure within ecosystems.
• Method benchmarking: Enables comparison and validation of mixing-model approaches against hierarchical Bayesian outputs using simulations.
• Celtic Sea ecosystem analysis: Applied to the Celtic Sea to reconstruct diet matrices and characterize trophic interactions in that system.

Methodology:

EcoDiet applies a hierarchical Bayesian model integrating literature priors, stomach-content prey occurrence frequencies, and biotracer data via a mixing model (MM), and uses simulated datasets and comparisons with classical network MM approaches for evaluation.