Extended larval development compensates for sublethal effects of fish predation in a mayfly population (Rhithrogena semicolorata, Ephemeroptera)

Many predation experiments in streams are carried out in enclosures. Hence, the relevance of their results to predict population dynamics is often unclear due to the relatively small spatial and temporal scale of the experiments. To enhance the transferability of experimental results on the ecosystem scale the impact of fish predators on a prey population was observed in a reach scale approach over 2 years in a natural stream. A 400-m reach inhabited by the small benthivorous fishes gudgeon (Gobio gobio) and stone loach (Barbatula barbatula) was compared with a fishless reference reach.It was shown that fish predation may affect the population of the grazing mayfly Rhithrogena semicolorata on the ecosystem scale. Although the larvae grew slower in the fish reach than in the fishless reach, the adults reached the same size and fecundity because they emerged 2–3 weeks later. By this compensation, the prey species avoided a reduction of their individual fecundity. On the other hand, the extended exposure to the fish predators resulted in an enhanced mortality and a reduced density of adult mayflies. Thus, there was obviously a trade-off between maximising fecundity and minimising mortality from fish predation.The observed differences were almost certainly caused be fish predation and not by natural differences of the reaches. This was concluded from results gained after eliminating all benthivorous fish from the former fish reach.With the help of scenario analyses based on our empirical data and simple model assumptions we could demonstrate that compensating the potential loss in fecundity by extending development time led to higher average fitness of the prey population than emerging at an earlier fixed time to avoid additional predation losses. Therefore, we concluded that this strategy was adaptive in the presence of benthivorous fish.