Squeezing the most out of existing literature: a systematic re-analysis of published evidence on ecological responses to altered flows



  1. Human-induced changes in river flow regimes are ubiquitous worldwide. Although numerous case studies have identified negative ecological impacts of changes in different aspects of flow regimes (e.g. magnitude, timing), there have been few attempts to systematically review this literature to derive general relationships regarding ecological responses to changes in flow regimes.
  2. Systematic literature reviews can inform science and management in ecologically complex systems not amenable to experimentation. However, such analysis of existing literature is often limited by inconsistent study design and data reporting. To attempt to overcome these difficulties, we used the recently developed Eco Evidence method and software to analyse 165 studies of ecological responses to changes in river flow regimes.
  3. Eco Evidence provides a rule set and standardised list of terms to assist reviewers to interpret consistently the results of disparate studies. The companion software assists with the synthesis of this information to reach transparent and repeatable conclusions regarding cause–effect hypotheses of ecological responses to environmental drivers.
  4. We compared our results to those of a recent, informal systematic review of the same studies, which is proving extremely influential. Stronger conclusions are reached when evidence is weighted, classified and combined according to the rules in Eco Evidence. Compared to the original review, we reached informative conclusions for a larger number of flow–response hypotheses, found that hypotheses for which the most evidence was available returned inconsistent results, addressed hypotheses at levels of conceptual resolution relevant to management and identified where insufficient evidence exists to reach a conclusion.
  5. Analyses conducted at several levels of conceptual resolution found strong support for many hypotheses regarding ecological impacts. We found a consistent sensitivity to changes in flow regime for both fish and riparian vegetation across a variety of performance metrics. While macroinvertebrate responses varied among performance metrics (e.g. abundance was negatively affected by increases or decreases in flows, diversity was only negatively affected by flow decreases, and assemblage structure was affected by neither), they were largely consistent within these metrics.
  6. We thus conclude that the Eco Evidence approach allowed us to extract more knowledge from the data set than was possible in the original review. Eco Evidence can improve synthesis of the burgeoning ecological literature and improve our general understanding in ecology. Amid widespread calls for ‘evidence-based’ environmental management, this powerful tool provides managers with a means of using research to help inform complex environmental decision-making.