Disruptive effects of chlorpyrifos on predator-prey interactions of <i>Ceratophrys ornata</i> tadpoles: Consequences at the population level using computational modeling

Abstract

Large-scale ecotoxicological studies have technical and ethical limitations, both related to the need to expose large numbers of individuals to potentially harmful compounds. The computational modeling is a complementary useful and predictive tool that overcomes these limitations. Considering the increasing interest in the effects of pesticides on behavioral traits, the aim of this study was to evaluate the effects of chlorpyrifos (CPF) on intra- and inter-specific interactions of anuran tadpoles, complementing traditional ecotoxicological tools with a theoretical analysis verified by computational simulations. Experiments were developed under two consecutive phases: a first phase of exposure (treated and control group), and a second phase of interactions. The second phase consisted of evaluating the effects of CPF on intra- and inter-specific interactions of exposed C. ornata (Co) tadpoles acting as predators and unexposed Rhinella fernandezae (Rf) tadpoles acting as prey (Experiment I), under different predator-prey proportions (0/10 = 0Co-10Rf, 2/8, 4/6, 6/4, 8/2, 10/0). Also, intraspecific interactions of three Co tadpoles under different conditions of exposure were evaluated (Experiment II: 3 exposed Co, 2 exposed Co/1 non-exposed, 1 exposed Co/2 non-exposed). During the exposure phase, chlorpyrifos induced significant mortality from 48 h (48 h: p < 0.05, 72 h-96 h: p < 0.001), irregular swimming, tail flexure, and the presence of subcutaneous air. Also, it induced effects on the sounds emitted after 96 h of exposure, registering a smaller number of pulses and higher dominant frequencies, and altered intra- and inter-specific interactions. During the interaction phase, the larvae continued to show sound effects, however, the antipredator mechanism continued to be operating and efficient. Finally, it was possible to model the behavior of the larvae under the effects of chlorpyrifos. We conclude that experimental data and computational modeling matched. Therefore, computational simulation is a valuable ecotoxicological tool that provides new information and allows prediction of natural processes.