Projecting fish community responses to dam removal – Data-limited modeling

Abstract

Modeling fish community responses to dam removal is an emerging field of study as dam removals become more common, but uncertainties concerning recovery time and community stability remain. In Europe, an EU-wide biodiversity strategy plans to restore around 25,000 km of rivers to free-flowing status, which emphasizes the importance of being able to predict fish community responses after dam removal. We developed a multi-species size spectrum model for a fish community in the Mörrum River in Sweden to identify possible outcomes after a dam was removed in 2020. Electrofishing monitoring before the dam removal was used to calibrate the model. We projected multiple scenarios into the future to explore patterns of community stability, individual species responses, and recovery time while varying parameters related to dam removal mortality, base resource rate change, and maximum recruitment change. We created 30 hypothetical scenarios using an abrupt change perspective (parameters are step-based) and 30 scenarios using a gradual change perspective (parameters are smooth). In both perspectives, dam removal mortality and a decreasing resource rate reduced community biomass and delayed recovery time compared to pre-dam removal conditions. Our results demonstrate that recovery from a dam removal scenario is not necessarily a benefit for all species. In scenarios where dam removal practices or dam failures cause high mortality events and sustained impacts on base trophic level resources, recovery of pre-removal biomass may take decades, while community stability may be unstable for twice that time-period. Our study shows that size spectrum models can be applied to dam removal scenarios to explore potential recovery outcomes, particularly from a risk avoidance perspective. A benefit of using such an approach is the relatively low data requirements needed to perform projections (e.g., present species, fish growth rates, relative fish abundance). Implementing this model in other river systems, particularly at the reach scale, can help river restoration and management assess tradeoffs associated with different habitat restoration approaches prior to committing to a dam removal plan.