Urbanisation of floodplain ecosystems: Weight-of-evidence and network meta-analysis elucidate multiple stressor pathways

Wendy A Monk; Zacchaeus G Compson; Catherine B Choung; Kathryn L Korbel; Natalie K Rideout; Donald J Baird

doi:10.1016/j.scitotenv.2019.02.253

Urbanisation of floodplain ecosystems: Weight-of-evidence and network meta-analysis elucidate multiple stressor pathways

Sci Total Environ. 2019 Sep 20:684:741-752. doi: 10.1016/j.scitotenv.2019.02.253. Epub 2019 Feb 22.

Authors

Wendy A Monk¹, Zacchaeus G Compson², Catherine B Choung², Kathryn L Korbel³, Natalie K Rideout⁴, Donald J Baird²

Affiliations

¹ Environment and Climate Change Canada @ Canadian Rivers Institute, Department of Biology, University of New Brunswick, Fredericton, New Brunswick E3B 5A3, Canada; Faculty of Forestry and Environmental Management, University of New Brunswick, Fredericton, New Brunswick E3B 5A2, Canada. Electronic address: wmonk@unb.ca.
² Environment and Climate Change Canada @ Canadian Rivers Institute, Department of Biology, University of New Brunswick, Fredericton, New Brunswick E3B 5A3, Canada.
³ Department of Biological Sciences, Macquarie University, New South Wales 2109, Australia.
⁴ Canadian Rivers Institute, Department of Biology, University of New Brunswick, Fredericton, New Brunswick E3B 5A3, Canada.

PMID: 30827674
DOI: 10.1016/j.scitotenv.2019.02.253

Abstract

Freshwater floodplains are dynamic, diverse ecosystems that represent important transition zones between terrestrial, riparian, subsurface and aquatic habitats. Given their historic importance in human development, floodplains have been exposed to a variety of pressures, which in combination have been instrumental in driving changes within these ecosystems. Here, we present an evidence-based framework to explore direct and indirect effects of pressures and stressors on floodplain ecosystems and test this structure within the urban landscape. Evidence was obtained from peer-reviewed scientific literature, focusing on effects of key pressures and stressors on receptors, including species composition (e.g., species presence-absence, diversity) and ecosystem function (e.g., biomass, decomposition). The strength of direct and indirect effects of individual and multiple stressors on biological receptors was quantified using two separate analyses: an evidence-weighted analysis and a quantitative network meta-analysis using data extracted from 131 studies. Results demonstrate the power of adopting a systematic framework to advance quantitative assessment of floodplain ecosystems affected by multiple stressors. While direct pathways were generally stronger and provided the core network skeleton, there were many more significant indirect pathways indicating evidence gaps in our mechanistic understanding of these processes. Indeed, the importance of indirect pathways (e.g. increase in impervious surface → increase in the accumulation rate of sediment nutrients) suggests that embracing complexity in network meta-analysis is a necessary step in revealing a more complete snapshot of the network. Results from the weight-of-evidence approach generally mirrored the direct pathway structure and demonstrated the strength of incorporating study quality alongside data sufficiency. Networks illustrated novel disturbance pathways (e.g., decrease in habitat structure → decrease in structure and function of aquatic and riparian assemblages) that can be used for hypothesis generation for future scientific enquiries. Our results highlight the broader applicability of adopting the proposed framework for assessing complex environments, such as floodplains.

Keywords: Bioassessment; EcoEvidence; Ecological complexity; Information discovery; Literature review; Netmeta.