Data from a nine-year study of benthic macroinvertebrate response to Elwha River dam removal

The Elwha River dam removal project—one of the largest in history—provided a rare opportunity to track watershed-scale ecological change. To assess benthic macroinvertebrate (BMI) responses, we sampled floodplain and tributary habitats for three years within each of the pre removal (2004-2006), during removal (2012-2014), and post removal (2015-2017) periods. Sampling sites were grouped according to their position relative to the former dams (downstream of Elwha Dam, between the two dams, upstream of Glines Canyon Dam, and tributaries). During and after dam removal, additional sites located within the former Lake Aldwell and Lake Mills reservoir footprints were incorporated. Within each reach, sampling locations were selected near side channel complexes and tributary junctions to allow efficient sampling of multiple habitat types. The number of sites sampled each year varied due to changes in flow, site accessibility, reservoir drawdown, and other dynamic conditions associated with dam removal. Although specific riffles were not always revisited annually due to shifting habitat, we consistently targeted shallow (<0.3 m), low gradient riffles with gravel to cobble sized substrates whenever possible. Over the nine year monitoring period, we collected BMI from 451 site-sample-events. BMI samples were collected from shallow riffles (3–30 cm depth) using a Slack sampler (500 µm mesh, 0.25 m² frame). At each site, except where otherwise noted, five individual samples were taken and composited into a single pooled sample. In side channels and tributaries, samples were drawn from five consecutive riffles; in mainstem reaches, the five subsamples were distributed evenly along a single long riffle. Samples were preserved in ethanol and processed in the laboratory. A subsample of 500–600 individuals from each composite was identified to the lowest practical taxonomic level (typically species or genus) and converted to density based on sampling area and the proportion of the sample processed. Taxa were also assigned to functional feeding groups (FFGs), voltinism categories, and other relevant life history traits. To characterize changes in sediment composition, we measured substrate embeddedness within each Slack sample and conducted pebble counts along transects spanning shallow riffles at each BMI site. Embeddedness—the extent to which coarse particles are surrounded or buried by fine sediment—was visually estimated and assigned a rating from 1 (<5% fines) to 5 (>75% fines) following Platts et al. (1983). Sediment size distributions were further assessed with pebble counts by measuring the intermediate (b axis) dimension of 100 randomly selected particles along Wolman (1954) transects positioned upstream of invertebrate sampling locations.