Articles Salmon runs in the Pacific Northwest have been declining for decades, so much so that many runs are threatened or endangered; others have been completely extirpated (Nehlsen et al. 1991). This "salmon crisis" looms large in the public eye, because it has serious and wideranging economic, cultural, and ecological repercussions. Billions of dollars have gone into industrial and agricultural projects that alter regional rivers in ways that, often unintentionally, make them inaccessible or unsuitable for salmon. Recently, billions more have been spent in largely unsuccessful attempts to restore the languishing salmon runs (Lichatowich 1999). Moreover, enormous nonmonetary resources have been expended in assigning and denying responsibility for failed runs and debating the possible efficacy of various remedies.As resources that are devoted to reversing declining runs of salmon have increased, scientists and resource managers have been expanding our understanding of the ecological role of salmon and other anadromous fishes, which return from the sea to spawn in fresh water. We have known for years that spawning salmon serve as a food resource for wildlife species (e.g., Shuman 1950) and, when they die after spawning (as most Pacific salmon do), their carcasses provide nutrients (e.g., carbon [C], nitrogen [N], phosphorus [P]) to freshwater systems (e.g., Juday et al. 1932). More recently, scientists have documented that these "salmon-derived nutrient" subsidies may have significant impacts on both freshwater and riparian communities and on the life histories of organisms that live there (Willson et al. 1998, Cederholm et al. 1999.Because of the burgeoning interest in salmon, growing indications of their ecological importance, and recent calls for management to consider the role of salmon in aquatic and terrestrial ecosystems (e.g., Larkin and Slaney 1997), we take this opportunity to review what is understood about the function of salmon as key elements of ecological systems. Our objectives are twofold. First, we expand on previous reviews of salmon (Willson et al. 1998, Cederholm et al. 1999) to include recent research that has amplified and modified earlier ideas about the contribution of salmon to ecosystem processes. In doing so, we describe the composition, magnitude, and distribution of marine inputs to freshwater and terrestrial systems via salmon. We use an expanding group of studies pertaining to stream nutrient budgets and salmon physiology to construct a schematic that illustrates salmon-derived products and the pathways by which they enter and are retained in aquatic and terrestrial food webs. We then consider the ecological variation associated with salmonid ecosystems and how this may influence the ecological response to the salmon input. Second, we consider how this variation in ecosystem response may influence management and conservation efforts.

No abstract

Headwater streams make up a large proportion of the total length and watershed area of fluvial networks, and are partially characterized by the large volume of organic matter (large wood, detritus, and dissolved organic matter) and invertebrate inputs from the riparian forest, relative to stream size. Much of those inputs are exported to downstream reaches through time where they potentially subsidize river communities. The relative rates, timing, and conversion processes that carry inputs from small streams to downstream reaches are reasonably well quantified. For example, larger particles are converted to smaller particles, which are more easily exported. Also, dissolved organic matter and surface biofilms are converted to larger particles which can be more easily intercepted by consumers. However, the quality of these materials as it affects biological activity downstream is not well known, nor is the extent to which timing permits biological use of those particles. These ecological unknowns need to be resolved. Further, land uses may disrupt and diminish material transport to downstream reaches by removing sources (e.g., forest harvest), by affecting transport and decomposition processes (e.g., flow regulation, irrigation, changes in biotic communities), and by altering mechanisms of storage within headwaters (e.g., channelization). We present conceptual models of energy and nutrient fluxes that outline small stream processes and pathways important to downstream communities, and we identify informational gaps that, if filled, could significantly advance the understanding of linkages between headwater streams and larger rivers. The models, based on empirical evidence and best professional judgment, suggest that navigable waters are significantly influenced by headwater streams through hydrological and ecological connectivities, and land use can dramatically influence these natural connectivities, impacting downstream riverine ecosystems.

Terrestrial-derived invertebrate (TI) inputs into streams and predation on them by salmonids (40-180 mm fork length) were measured in six coastal Alaska stream reaches from April through October 1993-1994; riparian habitat of three stream reaches contained conifer-dominated old-growth (no timber harvesting) and three were alder-dominated young-growth (31 years postclearcutting). Data from pan-traps placed on stream surfaces showed that TI biomass and nitrogen inputs averaged up to 66 and 6 mg ·m-2 ·day-1, respectively, with no significant difference between habitats. Stomach contents from coho salmon (Oncorhynchus kisutch), cutthroat trout (O. clarki), and Dolly Varden (Salvelinus malma) revealed that TI and aquatic-derived invertebrates (AI) were equally important prey. Additionally, salmonids from young-growth systems ingested a greater TI proportion than those from old-growth systems. There were trends but no significant differences between habitats of TI and AI biomass ingested; however, statistical power was <0.30. These results showed that TI were important juvenile salmonid prey and that a riparian overstory with more alder and denser shrub understory may increase their abundance. Riparian vegetation management will likely have important consequences on trophic levels supporting predators, including but not limited to fishes.

No abstract