The multiscale effects of stream restoration on water quality

Stream restoration is often considered as an effective watershed management tool to reduce riverine loads of nitrogen, phosphorus, and suspended sediments, and meet government-mandated water quality goals. However, despite the billions of dollars which have been spent on stream restoration, questions remain about its effectiveness for improving water quality, as many studies report either mixed success or lack the adequate methodological framework to detect water quality improvements. In this study, we measured fluxes of nutrients and sediment in an eroded stream before and after restoration by filling the eroded channel with a mixture of sand, gravel, and woodchips stabilized with rock weirs at intervals along the channel. Our monitoring design used a before-after-control-impact (BACI) approach at two spatial scales, one at the reach-scale, and one farther downstream to detect whether reach-scale changes in nutrient and sediment loads propagated downstream. At the reach scale, we found that the restoration enhanced stream function, removing 44.8% of the phosphate, 45.8% of the total phosphorus, 48.3% of the ammonium, 25.7% of the nitrate, 49.7% of the total nitrogen, and 73.8% of the suspended sediment. However, due to hydrological variance, monitoring stations farther downstream suggested no detectable changes at the larger spatial scale relative to a reference stream, which highlights the challenges of detecting watershed-scale responses to small-scale stream restoration projects. This study provides a methodological framework for evaluating the effect of stream restoration on water quality at multiple scales and shows that reach-scale improvements may not be detectable at watershed-scales.