No abstract

For an estuarine restoration project to be successful it must reverse anthropogenic effects and restore lost ecosystem functions. Restoration projects that aim to rehabilitate endangered species populations make project success even more important, because if misjudged damage to already weakened populations may result. Determining project success depends on our ability to assess the functional state or "performance" and the trajectory of ecosystem development. Mature system structure is often the desired "end point" of restoration and is assumed to provide maximum benefit for target species; however, few studies have measured linkages between structure and function and possible benefits available from early recovery stages. The Salmon River estuary, Oregon, U.S.A., offers a unique opportunity to simultaneously evaluate several estua-rine restoration projects and the response of the marsh community while making comparisons with a concurring undiked portion of the estuary. Dikes installed in three locations in the estuary during the early 1960s were removed in 1978, 1987, and 1996, creating a "spacefor-time substitution" chronosequence. Analysis of the marsh community responses enables us to use the development state of the three recovering marshes to determine a trajectory of estuarine recovery over 23 years and to make comparisons with a reference marsh. We assessed the rate and pattern of juvenile salmon habitat development in terms of fish density, available prey resources, and diet composition of wild juvenile Oncorhynchus tshawytscha (chinook salmon). Results from the outmigration of 1998 and 1999 show differences in fish densities, prey resources, and diet composition among the four sites. Peaks in chinook salmon densities were greatest in the reference site in 1998 and in the youngest (1996) site in 1999. The 1996 marsh had higher densities of chironomids (insects; average 864/m 2 ) and lower densities of amphipods (crustaceans; average 8/m 3 ) when compared with the other sites. Fauna differences were reflected in the diets of juvenile chinook with those occupying the 1978 and 1996 marshes based on insects (especially chironomids), whereas those from the 1987 and reference marshes were based on crustaceans (especially amphipods). Tracking the development of recovering emergent marsh ecosystems in the Salmon River estuary reveals significant fish and invertebrate response in the first 2 to 3 years after marsh restoration. This pulse of productivity in newly restored systems is part of the trajectory of development and indicates some level of early functionality and the efficacy of restoring estuarine marshes for juvenile salmon habitat. However, to truly know the benefits consumers experience in recovering systems requires further analysis that we will present in forthcoming publications.

This study evaluated estuarine habitat use, life-history composition, growth and survival of four successive broods of coho salmon Oncoryhnchus kisutch in Salmon River, Oregon, U.S.A. Subyearling and yearling O. kisutch used restored and natural estuarine wetlands, particularly in the spring and winter. Stream-reared yearling smolts spent an average of 2 weeks in the estuary growing rapidly before entering the ocean. Emergent fry also entered the estuary in the spring, and some resided in a tidal marsh throughout the summer, even as salinities increased to >20. A significant portion of the summer stream-resident population of juvenile O. kisutch migrated out of the catchment in the autumn and winter and used estuary wetlands and adjacent streams as alternative winter-rearing habitats until the spring when they entered the ocean as yearling smolts. Passive integrated transponder (PIT) tag returns and juvenile life-history reconstructions from otoliths of returning adults revealed that four juvenile life-history types contributed to the adult population. Estuarine-associated life-history strategies accounted for 20-35% of the adults returning to spawn in the four brood years, indicating that a sizable proportion of the total O. kisutch production is ignored by conventional estimates based on stream habitat capacity. Juvenile O. kisutch responses to the reconnection of previously unavailable estuarine habitats have led to greater life-history diversity in the population and reflect greater phenotypic plasticity of the species in the U.S. Pacific Northwest than previously recognized.

1. Protection of places important for aesthetic, ecological, and cultural values has been a goal of conservationists for over 150 years. Cornerstones of place-based conservation include legal designations, international agreements, and purchase by public or non-profit organizations.2. In the Salmon River catchment, Oregon, protections were initially developed in the 1930s for the freshwater riparian corridor and forestry research in the uplands. Over time, additional protections in the estuary and nearshore marine environments were added, motivated by local desire to protect and restore habitats and fish populations.3. Removal of three levees in the Salmon River estuary occurred over three consecutive 9-year time-steps, and provided the opportunity for research on tidal marsh recovery in the framework of a space-for-time chronoseries. Elevation, channel morphology, and vegetation all exhibited trajectories toward reference conditions. Fish and macroinvertebrates also served as indicators of tidal marsh recovery, although their recovery patterns were not strictly related to the chrono-series trajectories. The extent of restoration provided a novel opportunity to measure a significant response of biotic indicators at the site and catchment scales.4. Salt marsh restoration augmented protected freshwater habitats by expanding rearing habitats for juvenile salmonids and increasing expression of life-history diversity for both Chinook and coho salmon. This finding highlights linkages between freshwater and marine habitats and populations, and has the potential to influence important policy advances and changes in management of Pacific salmon. 5. Restoration promoted collaborations among stakeholders, community involvement, and inspiring educational opportunities that enabled more comprehensive research than any single sponsor could have accomplished.6. Protected status designations have fostered a wealth of opportunities that were not specifically envisioned when the protections were first put in place. In particular, dedicated scientific investigation of landscape-scale change did not occur by design, but was pieced together as funding opportunities arose over time. Figure 3. Extensive levees, marsh drainage, and infrastructure were present at CHSRA, Oregon, prior to.estuary rehabilitation. As of 1975 (A), only the reference marsh remained intact as a large, naturally.functioning estuarine system. Rehabilitation of marshes through removal of levees, structures, and fill took decades (B-years refer to the timing of marsh rehabilitation), but resulted in the recovery of approximately 70% of original estuarine marsh habitat to natural tidal inundation regimes. R. L. FLITCROFT ET AL.44

Recent genetic studies, meta-analyses, and retrospective analyses have documented reduced productivity of wild salmon and steelhead Oncorhynchus mykiss that interbreed with hatchery-reared fish, raising concerns about the longterm viability and recovery of at-risk stocks. In 2007, the Oregon Department of Fish and Wildlife discontinued a Coho Salmon Oncorhynchus kisutch hatchery program at the Salmon River to support recovery of a wild Coho Salmon population in the Oregon Coast Evolutionarily Significant Unit. This decision constituted a unique management "experiment," allowing for direct measurement of the wild population's response after the discontinuation of a decades-old hatchery program. We used a before-after, control-impact design to examine whether selected viability metrics of the naturally produced population in the Salmon River changed after the hatchery program ended. We compared metrics for the 2006-2013 broods, representing periods after the hatchery program ended, to those for the 1995-2005 broods, when the hatchery program was still releasing 200,000 smolts annually. We also examined neighboring populations during similar time periods to account for changes or variation due to other factors. Although hatchery-origin spawners previously had accounted for most of the adults returning to the Salmon River, the naturally produced population did not collapse, and two viability metrics improved significantly after the Coho Salmon hatchery program ended: (1) adult abundance increased and (2) spawn timing expanded and moved closer to the historical timing. Recruits-to-spawner ratios in the Salmon River, although initially low, are now approximately equal to those of neighboring populations. The results indicate that hatchery closure can be an effective strategy to promote wild population recovery. However, considerable variability in population trends and environmental conditions will require continued monitoring to verify the long-term resilience and viability of the wild population.