Eelgrass Enhancement and Restoration in the Lower Columbia River Estuary, Period of Performance: Feb 2008-Sep 2009.

Abstract

Executive SummaryThe purpose of this study was to evaluate the ability to enhance distribution of eelgrass (Zostera marina) in the Columbia River Estuary to serve as refuge and feeding habitat for juvenile salmon, Dungeness crab, and other fish and wildlife. We strongly suspected that limited eelgrass seed dispersal has resulted in the present distribution of eelgrass meadows, and that there are other suitable places for eelgrass to survive and form functional meadows.Funded as part of the Bonneville Power Administration's call for Innovative Projects, we initiated a multistage study in 2008 that combined modeling, remote sensing, and field experimentation to:1. Spatially predict habitat quality for eelgrass.2. Conduct experimental plantings. Evaluate restoration potential.Baseline in-situ measurements and remote satellite observations were acquired for locations in the Lower Columbia River Estuary (LCRE) to determine ambient habitat conditions. These were used to create a habitat site-selection model, using data on salinity, temperature, current velocity, light availability, wave energy, and desiccation to predict the suitability of nearshore areas for eelgrass. Based on this model and observations in the field, five sites that contained no eelgrass but appeared to have suitable environmental conditions were transplanted with eelgrass in June 2008 to test the appropriateness of these sites for eelgrass growth. We returned one year after the initial planting to monitor the success rate of the transplants. During the year after transplanting, we carried out a concurrent study on crab distribution inside and outside eelgrass meadows to study crab usage of the habitat.One year after the initial transplant, two sites, one in Baker Bay and one in Young's Bay, had good survival or expansion rates with healthy eelgrass. Two sites had poor survival rates, and one site had a total loss of the transplanted eelgrass. For submerged aquatic vegetation (SAV) restoration projects, these are reasonable success results and represent a small net gain in eelgrass in the LCRE. Crabs used both the eelgrass and unvegetated substrate, though in neither were there great abundance of the young-of-the-year crabs. During the field assessment of 12 potential transplant sites, divers discovered one site in southern Young's Bay that contained a previously undocumented eelgrass bed.This integrated project developed the first predictive maps of sites suitable for eelgrass and other SAV in the lower estuary. In addition, techniques developed for this project to assess light levels in existing and potential submerged habitats have great potential to be used in other regions for nearshore and coastal monitoring of SAV.Based on these preliminary results, we conclude that eelgrass distribution could likely be expanded in the estuary, though additional information on current eelgrass locations, usage by species of interest, and monitoring of current conditions would help develop a baseline and verify benefit. Our recommendations for future studies i...

How do methodological choices influence estimation of river metabolism?

“…Models including the BLAM (Julian et al 2008) can incorporate topographic shading alongside a range of hydrogeomorphic variables to estimate light at the water surface or at depth, but require much additional effort and only averaged accurate within 39% over more than a week of use. After losing several PAR sensors to theft, we looked to alternatives including placing sensors at nearby and more protected basecamps, as well as by calculating diurnal PAR using geographic location, as included in and recommended by the StreamMetabolizer package (Appling et al 2017) This model is widely used (Appling et al 2018;Judd et al 2009). Modeling light assumes that the location of the sensor is in an unobstructed reach on a clear day, as clouds can dramatically change PAR.…”

“…After losing several PAR sensors to theft, we looked to alternatives including placing sensors at nearby and more protected basecamps, as well as by calculating diurnal PAR using geographic location, as included in and recommended by the StreamMetabolizer package (Appling et al 2017) This model is widely used (Appling et al 2018; Judd et al 2009). Modeling light assumes that the location of the sensor is in an unobstructed reach on a clear day, as clouds can dramatically change PAR.…”