Inferring quantitative species interactions of seeds and seed-feeding carabid beetles from ecological survey data

Here, we develop a trait-based approach suitable for creating quantitative networks, i.e. with varying interaction strengths. We applied this method to existing ecological survey data from an arable field of carabid ground beetles (Coleoptera: Carabidae) from pitfall traps and plant seeds from seed rain traps. We used existing data in the literature to predict a per-individual interaction cost index from carabid and seed size, based on frequency-dependent prey selection and the energetic intake of seeds by carabids. This was scaled up to the population level to create predicted inferred weighted networks using the sampled abundance of carabids and seeds, energetic intake rates in the literature, and assuming bottom up control. From this we calculated a novel predation pressure ratio which was the predicted seed predation (the sum of interaction strengths) relative to seed abundance. This is made available as an R markdown file with associated data files.