Estimating Changes in Coho Salmon and Steelhead Abundance from Watershed Restoration: How Much Restoration is Needed to Measurably Increase Smolt Production?

Abstract

AbstractUsing existing data from evaluations of habitat restoration, we estimated the average change in coho salmon Oncorhynchus kisutch and steelhead O. mykiss parr and smolt densities for common in‐channel (culvert removal, large wood placement, boulder placement, and constructed logjams) and floodplain restoration techniques (constructed side channels and reconnected floodplain habitats). We then used these numbers and a Monte Carlo simulation to predict changes in fish numbers in a model watershed for two restoration scenarios: (1) restoration of all accessible habitat within the watershed and (2) restoration of the average amount historically implemented in Puget Sound watersheds (8% of total restorable areas). Mean increases in coho salmon parr or smolt density after restoration ranged from 0.19 to 2.32 parr/m for in‐channel techniques and from 0.34 to 1.70 parr/m2 for floodplain techniques. Increases in steelhead parr or smolt density ranged from −0.06 to 0.71 fish/m and from 0.03 to 0.06 fish/m2 for in‐channel and floodplain techniques, respectively. Under restoration scenario 1, the predicted mean increase in numbers was 1,459,254 (117%) and 285,302 (140%) for coho salmon parr and smolts and 93,965 (65%) and 28,001 (125%) for steelhead parr and smolts. Under scenario 2, the predicted mean increase in parr and smolts was 59,591 (5%) and 15,022 (7%) for coho salmon and 1,733 (1%) and 1,195 (5%) for steelhead. The percentage of floodplain and in‐channel habitat that would have to be restored in the modeled watershed to detect a 25% increase in coho salmon and steelhead smolt production (the minimum level detectable by most monitoring programs) was 20%. However, given the large variability in fish response (changes in density or abundance) to restoration, 100% of the habitat would need to be restored to be 95% certain of achieving a 25% increase in smolt production for either species. Our study demonstrates that considerable restoration is needed to produce measurable changes in fish abundance at a watershed scale.