Elsevier

Water Research

Volume 196, 15 May 2021, 116981
Water Research

Making Waves
Making waves. Bridging theory and practice towards multiple stressor management in freshwater ecosystems

https://doi.org/10.1016/j.watres.2021.116981Get rights and content

Highlights

  • The management of multiple stressor interactions (MSI) in fresh waters is uncommon.

  • Empirical modelling using monitoring data can be used for the detection of MSIs.

  • Evidence of MSI effects during degradation and recovery is urgently needed.

  • Recommendations are provided on management responses for MSI scenarios.

Abstract

Despite advances in conceptual understanding, single-stressor abatement approaches remain common in the management of fresh waters, even though they can produce unexpected ecological responses when multiple stressors interact. Here we identify limitations restricting the development of multiple-stressor management strategies and address these, bridging theory and practice, within a novel empirical framework. Those critical limitations include that (i) monitoring schemes fall short of accounting for theory on relationships between multiple-stressor interactions and ecological responses, (ii) current empirical modelling approaches neglect the prevalence and intensity of multiple-stressor interactions, and (iii) mechanisms of stressor interactions are often poorly understood. We offer practical recommendations for the use of empirical models and experiments to predict the effects of freshwater degradation in response to changes in multiple stressors, demonstrating this approach in a case study. Drawing on our framework, we offer practical recommendations to support the development of effective management strategies in three general multiple-stressor scenarios.

Introduction

Freshwater ecosystems are commonly exposed to multiple anthropogenic stressors, which can interact and produce ecological surprises (Ormerod et al., 2010). While conceptual understanding and experimental demonstration of these interactions are now well established (Schäfer and Piggott, 2018), a major challenge remains to develop approaches to detect, quantify and manage stressor interactions in the real world (Feld et al., 2016). To inform this development, various attempts have been made to assess the frequency of stressor interactions across a broad range of freshwater ecosystems (Birk, 2019). These endeavours have identified issues that limit our capacity to generalise and predict undesirable ecological responses to single stressor reduction strategies. More conspicuously, very few published studies have demonstrated the successful management of single or multiple stressors, where interactions and hierarchies have first been quantified.

This inability to generalise poses a problem for ecosystem management, which has historically focussed on abating individual stressors (Schindler et al., 2016). Well-informed multiple-stressor management could offer opportunities to offset effects of large-scale stressors that are hard to manage locally, including anthropogenic warming and changes in precipitation patterns associated with climate change (Moss et al., 2011) or the widespread proliferation of synthetic chemicals (Bernhardt et al., 2017) and toxic substances from industrial and domestic sources (Walters et al., 2020). There is an urgent need to develop methods to diagnose multiple stressor interactions and assess responses of ecological indicators to them across both degradation and recovery pathways. These methods must be applicable to data gathered at different scales and resolutions (Blair et al., 2019).

Here, we demonstrate how empirical data on fresh waters can underpin effective management of ecosystems subject to multiple stressors. Specifically, we explore how theory on multiple-stressor interactions and ecological responses is relevant to empirical data, particularly from national monitoring schemes such as those stipulated by the EU Water Framework Directive (WFD; European Commission, (2000)) or the USA Federal Water Pollution Control Act (2002), ‘The Clean Water Act’). We argue, however, for greater integration of understanding from such monitoring data with outcomes of experiments and modelling. Finally, we build on this understanding to develop practical recommendations for integrating the assessment and management of multiple stressors into future freshwater management and biodiversity protection strategies, highlighting limitations that remain to be addressed.

Conceptual models describing forms and directions of stressor interactions have predominantly focused on quantifying and classifying deviations from additive effects models (Piggott et al., 2015a). Effects are defined as additive when an ecological response is equal to the sum of the effects of the individual stressors. Synergistic interactions occur when ecological responses are greater than the sum of the additive effects, and antagonistic interactions where ecological responses are less than the sum of the additive effects (Fig. 1). Additive effects indicate that stressors act independently of one another, and so control of any one stressor should result in exactly proportional ecological responses. Under such a scenario, gradual changes in ecological response should be detected in monitoring data (Hillebrand et al., 2020). Such data may reveal ecological improvements that are greater than expected when stressors producing synergistic interactions are mitigated. In contrast, reduction of an antagonistic stressor could result, counter-intuitively, in the detection of further ecological degradation through monitoring. Piggott et al. (2015b) extended this basic model by considering the cumulative magnitude and direction of effects. This revealed cross-over interactions where combined stressor effects cancel each other and can lead to effects opposite to those of the individual effects. This phenomenon has been called mitigating synergism (Piggott et al., 2015b) or reversal (Jackson et al., 2016).

Section snippets

Moving from theory to practice: detection, prediction & management

The prevalence of interactions across scales and ecosystem types is increasingly recognised. An assessment of more than 100,000 water bodies across Europe, reported under the 2nd WFD River Basin Management cycle (2009–2015) showed that 50% of them were affected by two or more stressors, most commonly, hydromorphological modifications and nutrient pollution (EEA, 2018). Likewise, based on 174 pairwise stressor combinations from experiments and surveys across Europe, Birk et al. (2020) report

Practical recommendations for multiple-stressor management

The current shortcomings of multiple-stressor management outlined above are global in scope. This represents a clear weakness in ecological assessments underpinning, for example, the European WFD (Carvalho et al., 2019). Indeed, nearly all WFD assessment methods have been developed to be responsive to single stressors (Birk et al., 2012). This raises the question, to what extent the currently limited success in restoring water bodies in Europe is the result of targeting only single stressors?

Final considerations

Three final points need brief mention.

First, in a very recent broad synthesis, Hillebrand et al. (2020) found ecological responses to stressors along the degradation pathway are generally gradual. This finding is highly relevant to water management where notable system changes are expected only when thresholds, at times arbitrary or operational thresholds, are surpassed.

Secondly, our current understanding of multiple stressor effects essentially comes from assessing impacts of increasing

Conclusions

  • 1

    The lack of consideration of interactions between multiple stressors represent a potential major limitation in achieving ecological restoration of freshwater ecosystems.

  • 2

    Conceptual models for multiple stressor interactions can be developed to inform novel management approaches, helping practitioners avoid the many pitfalls associated with the detection of interactions.

  • 3

    Outputs from empirical analyses of monitoring data and controlled experiments in realistic settings should be systematically

Authors’ contributions

All authors have approved the submitted version and agree to be accountable for the aspects of the work they conducted.

Declaration of Competing Interest

None.

Acknowledgements

This work was enabled by the MARS project EU FP7, Contract No. 603378; NERC NE/T003200/1 and NE/N00597X/1; the UK SCAPE project; the RePhoKUs project as part of the UK Global Food Security research programme (Grant Nos. BB/R005842/1); the Scottish Government project no. 05946; and Fundação para a Ciência e a Tecnologia PTDC/CTA-AMB/31245/2017, IF/01304/ 2015 and UID/AGR/00239/2013.

References (40)

  • G.S. Blair et al.

    Data science of the natural environment

    Front. Environ. Sci.

    (2019)
  • C.J. Brown et al.

    Managing for interactions between local and global stressors of ecosystems

    PLoS ONE

    (2013)
  • A. Bruder et al.

    The importance of ecological networks in multiple-stressor research and management

    Front. Environ. Sci.

    (2019)
  • J.W. Bull et al.

    Reconciling multiple counterfactuals when evaluating biodiversity conservation impact in socialecological systems

    Conserv. Biol.

    (2020)
  • I.M. Côté et al.

    Interactions among ecosystem stressors and their importance in conservation

    Proc. R. Soc. B

    (2016)
  • D. De Zwart et al.

    Complex mixture toxicity for single and multiple species: proposed methodologies

    Environ. Toxicol. Chem.

    (2005)
  • J.M. Dyste et al.

    Assessing stream channel restoration: the phased recovery framework

    Restor. Ecol.

    (2019)
  • Directive 2000/60/EC of the European Parliament and the Council of 23 October 2000 Establishing A Framework for Community Action in the Field of Water Policy

    OJEC, L

    (2000)
  • European Environment Agency. 2018. European waters Assessment of status and pressures. EEA Report, No 7/2018,...
  • Federal Water Pollution Control Act Amendments of 1972, Pub. L. No. 107-303, 33U.S.C. 1251 et seq., November 27,...
  • Cited by (32)

    View all citing articles on Scopus
    View full text