1 | INTRODUCTION
Biological invasions are regarded as one of the largest threats to native biodiversity (Mack et al., 2000), including in freshwater (Sala et al., 2000, Caffrey et al., 2014). Alien species may display invasive properties inducing serious ecological consequences through both direct and indirect ecological interactions (Kumschick et al., 2015), as well as major economic impacts (Dittel & Epifanio, 2009). Such alien invasive species may spread rapidly and cause problems that escalate from local to regional following establishment. Early targeted eradication actions are often therefore the only effective management method. This approach may be particularly effective when species are required to move through environmental corridors, such as in freshwater ecosystems (Hulme, 2009). However, options for cost-effective and reliable measures are often limited since it may become virtually impossible to eradicate invasive aquatic species in anything but small, enclosed waterbodies (Holdich et al., 1999).
In aquatic ecosystems, one of the most widely used measures for eradication of invasive species is rotenone. This is a piscicide which is highly toxic to fish and to certain invertebrates but has low toxicity to birds and mammals (Ling, 2003). Rotenone is a botanical compound originating from plant species in the Leguminosae family and have been used for centuries as a mean of capturing fish in areas where these plants are naturally occurring (Brooks & Price, 1961, Meadows, 1973). It has also been used as a fish management tool for almost a century (Finlayson et al., 2010), either with the purpose of quantifying fish abundances, manipulating fish populations to maintain sport fisheries or treatment of rearing ponds (McClay, 2000, 2005). It is one of the few, or the only, cost-effective and widely accepted measure for eradicating populations of invasive fish. In addition, rotenone has been widely used for eradication of fish parasites through culling of host populations.
Use of rotenone is controversial. This is largely due to its indiscriminate impacts on the wider aquatic fauna. Short-term impacts on non-target organisms, like benthic invertebrates is well documented (e.g. Koksvik & Aagaard, 1984, Melaas et al., 2001, Dalu et al., 2015). The observed effects often include reduction in the total abundance of benthic invertebrates comparted to pre-treatment levels. However, there are large variations ranging from little or no reduction (Cook & Moore, 1969, Dudgeon, 1990, Bellingan et al., 2019) to up to 95% reduction (Hamilton et al., 2009). Part of the variation may be explained by taxa-specific tolerance to rotenone. Ephemeroptera, Plecoptera and Trichoptera, which are abundant in oxygen rich lotic habitats, are generally reported to be rotenone sensitive (Mangum & Madrigal, 1999, Arnekleiv et al., 2001, Eriksen et al., 2009), whereas Coleoptera, Odonata and Gastropoda, which are often abundant in lentic habitats, are considered rotenone tolerant (Chandler & Marking, 1982, Holocombe et al., 1987, Arnekleiv et al.,1997, Kjærstad & Arnekleiv, 2011). There are also species-specific differences in tolerance among closely related taxa. For example, species of gill breathing Ephemeroptera are particularly sensitive to rotenone (Arnekleiv et al. 2001). Recovery of the total benthic invertebrate density can occur within a year (Kjærstad & Arnekleiv, 2004, Pham et al., 2018). However, recovery of single taxa may potentially take several years (Arnekleiv et al., 1997, Mangum & Madrigal, 1999).
Long-term consequences of failing to act towards continuous spread of invasive fish or parasites may be severe to the ecosystem and ecosystem services. Short-term disturbances to the ecosystem are expected, and to certain degree accepted, as appropriate collateral damage. However, long-term consequences of rotenone treatment may be less accepted. It is currently a lack of studies assessing long-term ecosystem effects caused by rotenone treatments (Vinson et al., 2010). Therefore, this study focused on the long-term consequences of an intensive rotenone treatment on lotic and lentic benthic invertebrate communities. Specifically, we tested for impact of rotenone treatment on the long-term abundance and temporal turnover of key taxa by comparing time series from a rotenone treated watercourse and a nearby control watercourse.