Main content area

Simulating stream response to floodplain connectivity and revegetation from reach to watershed scales: Implications for stream management

Singh, Nitin K., Wemple, Beverley C., Bomblies, Arne, Ricketts, Taylor H.
The Science of the total environment 2018 v.633 pp. 716-727
ecosystem services, ecosystems, floodplains, infrastructure, land restoration, multivariate analysis, nonprofit corporations, simulation models, streams, variance, water quality, watersheds, Vermont
Natural-infrastructures (e.g., floodplains) can offer multiple ecosystem services (ES), including flood-resilience and water quality improvement. In order to maintain these ES, state and non-profit organizations consider various stream interventions, including increased floodplain connectivity and revegetation. However, the effect of these interventions is rarely quantified. We build a hydraulic model to simulate the influence of above-mentioned interventions on stream power and water depth during 5 yr and 100 yr flood return-intervals for two watersheds in Vermont, USA. Simulated revegetation of floodplains increased water depth and decreased stream power, whereas increasing connectivity resulted in decline of both responses. Combination of increased connectivity and floodplain revegetation showed greatest reduction in stream-power suggesting that interventions may influence stream response in diverse ways. Across all three interventions, 14% and 48% of altered reaches showed increase in stream power and water depth over baseline, indicating that interventions may lead to undesirable outcomes and their apparent effectiveness can vary with the measure chosen for evaluation. Interventions also influenced up to 16% of unaltered reaches (i.e., in which no interventions were implemented), indicating the consequences of interventions can spread both up and downstream. Multivariate analysis showed that up to 50% of variance in stream response to interventions is attributable to characteristics of reaches, indicating that these characteristics could mediate the effectiveness of interventions. This study offers a framework to evaluate the potential ES provided by natural infrastructure. All stream interventions involve tradeoffs among responses and between target and non-target areas, so careful evaluation is therefore needed to compare benefits and costs among interventions. Such assessments can lead to more effective management of stream-floodplain ecosystems both in Vermont and elsewhere.