Comparison of predator–prey interaction models for fish assemblages from the neotropical region

Abstract Predation is one of the main fish assemblages structuring force in many aquatic ecosystems. Finding a functional relationship between predator and prey that improves our understanding of this process is a challenge to ecologists. In order to evaluate this interaction, several models have been created, each one of them representing a specific biological mechanism. Although, these models have not been intensively confronted against empirical data for some group of organisms, such as fish, they are the scientific base to predator–prey systems. In this paper, some models, with distinctive assumptions, were fitted to the same fish assemblage data set from an isolated lagoon of the upper Parana River floodplain using non-linear procedure. Then, they were compared in order to explore the mechanisms (represented by model assumptions) that could be acting on that assemblage. We used the Lotka–Volterra model and its modifications to consider carrying capacity, saturation effect, ratio-dependence predator–prey trophic function, and environmental heterogeneity. To fit the models, all prey species were grouped as prey and the same was done to predators. Result of this approach allowed us to make four suggestions about predator–prey relationship for the environment studied: (i) predators were efficient in controlling prey populations; (ii) the best fitted model was found with the sigmoidal functional response; (iii) density-dependence presented better fit than ratio-dependence predator–prey trophic function; (iv) refuges did not present intense interference in the predator–prey relationship. Therefore, comparisons of models that represent different assumptions showed to be a good tool in evaluating main mechanisms acting on the relationship between predators and prey in a fish assemblage.