Can a piscicide treatment alter stream ecosystem functioning through trophic cascading effects on benthic invertebrates?

Abstract

Processing of detritus is an important ecosystem function in freshwaters. In rivers and streams, the activity of shredding invertebrates play an important role in processing of coarse-particulate organic matter such as leaf litter. In stream food webs, fish may act as top predators and control activity of consumers such as shredders top down. This top-down effect may be of a directly consumptive nature, or indirect and mediated by species traits. Most studies on trophically cascading effects in stream ecosystems are done in small scale, either in small laboratory setups, or in in-stream experimental units such as cages or flow through channels. These setups are very useful in exploring specific effects and relationships, but might not be adequate to document effects on whole stream or catchment scale. I utilized a management-imposed rotenone treatment of the sub-arctic watercourse Skibotn catchment in Troms, northern Norway, as a setup for a large scale field experiment. Using pairs of coarse- and fine-meshed plastic litter bags filled with dry birch ( Betula pubescens ) leaf litter, I measured decomposition rates ( kd − 1 ), and collected and identified leaf litter colonizing invertebrates in riffle habitats in the autumn one year before (2014), and one year after (2017) the treatment. Nordkjos catchment, an untreated catchment in the adjacent area, was studied simultaneously. Shredding invertebrates contributed to litter decomposition in both years in all but one stream, but I found no significant change in invertebrate-mediated decomposition between the two years on catchment scale. There was no marked change in density or diversity of invertebrates in the litter bags between the two years, while fish had a diverse diet dominated by Baetid mayflies. However, young of the year salmonids were present already in a few sites after the treatment. Results of this field experiment indicate that fish did not have a strong top-down effect on shredding invertebrates in Skibotn catchment. The most likely explanation is that fish densities were low and their diets were not dominated by the most important shredder species. The studied streams are heterogenic environments and subject to natural stochasticity that might outweigh any small ecological effects, emphasizing the importance of good background data when performing before- after- impact control studies.