Pacific salmon as vectors of environmental contaminants: An experimental test confirms synoptic surveys in natural streams.

Pacific salmon transfer large quantities of material to tributaries during their spawning migrations, including carcass tissue and labile nutrients but also persistent organic pollutants (POPs) and heavy metals. We conducted a Before-After-Control-Intervention experiment by adding salmon carcasses and eggs to a Michigan (USA) stream that had never received inputs from non-native salmon to understand the bioaccumulation and persistence of biotransported contaminants. Our experimental outcomes were compared to previous studies using meta-analysis. Coincident with the introduction of salmon, the PCB and DDE burden of resident trout significantly increased. However, we did not observe changes in total mercury (Hg). Two years after the salmon addition experiment concluded, resident trout POP concentrations had returned to pre-addition levels, with no difference between the treatment and control reaches. Analysis of effect sizes suggested that the contaminant response observed in our experiment is consistent with field survey observations. Our study suggested that the consumption of salmon eggs drove the increase in POP burden of resident trout while Hg bioaccumulation was influenced by watershed sources. Critically, our study suggests that ecosystems are capable of quickly recovering from POP inputs from species migrations if contaminant sources are removed.

Using a dynamic bioenergetics-bioaccumulation model to understand mechanisms of uptake and bioaccumulation of salmon-derived contaminants by stream-resident fish.

Ecosystem linkages created by migratory organisms such as Pacific salmon (Oncorhynchus spp.) facilitate the transfer of ecologically beneficial resource subsidies and environmentally damaging contaminants to recipient food webs. In the Laurentian Great Lakes, introduced Pacific salmon accumulate large contaminant burdens that they disperse to streams during spawning in the form of carcass and gametic tissue, with uncertain consequences for stream food webs. Here, we describe a coupled bioenergetics-bioaccumulation model parameterized using empirical and literature-sourced data to predict the dual effect of Pacific salmon on stream-resident brook trout (Salvelinus fontinalis) growth and contaminant bioaccumulation. Within the model, we developed four unique scenarios to ascertain how the (1) trophic pathway to contamination, (2) level of salmon egg consumption, (3) intensity and duration of salmon exposure, and (4) age of first exposure to salmon, affected growth and contaminant bioaccumulation in brook trout. Our model demonstrated that salmon egg consumption increased brook trout growth and PCB bioaccumulation while reducing Hg tissue concentrations. Other trophic pathways, including direct carcass consumption and an indirect food web pathway, did not strongly influence growth or contaminant bioaccumulation. Our model also demonstrated that variation in the magnitude and temporal duration of salmon egg consumption mostly strongly influenced the growth and contaminant concentration of younger brook trout. Overall, our model highlighted that Pacific salmon transfer energy and contaminants but this balance is dictated by the food web pathway and plasticity in the diet of stream-resident fish. Our mechanistic, model-based evaluation of salmon contaminant biotransport can be extended to predict the impact of other migratory fishes on recipient food webs.

Evaluating the consequences of salmon nutrients for riparian organisms: Linking condition metrics to stable isotopes.

Stable isotope ratios (δ13C and δ15N) have been used extensively to trace nutrients from Pacific salmon, but salmon transfer more than carbon and nitrogen to stream ecosystems, such as phosphorus, minerals, proteins, and lipids. To examine the importance of these nutrients, metrics other than isotopes need to be considered, particularly when so few studies have made direct links between these nutrients and how they affect riparian organisms. Our study specifically examined δ13C and δ15N of riparian organisms from salmon and non-salmon streams in Idaho, USA, at different distances from the streams, and examined whether the quality of riparian plants and the body condition of invertebrates varied with access to these nutrients. Overall, quality and condition metrics did not mirror stable isotope patterns. Most notably, all riparian organisms exhibited elevated δ15N in salmon streams, but also with proximity to both stream types suggesting that both salmon and landscape factors may affect δ15N. The amount of nitrogen incorporated from Pacific salmon was low for all organisms (<20%) and did not correlate with measures of quality or condition, probably due to elevated δ15N at salmon streams reflecting historical salmon runs instead of current contributions. Salmon runs in these Idaho streams have been declining, and associated riparian ecosystems have probably seen about a 90% reduction in salmon-derived nitrogen since the 1950s. In addition, our results support those of other studies that have cautioned that inferences from natural abundance isotope data, particularly in conjunction with mixing models for salmon-derived nutrient percentage estimates, may be confounded by biogeochemical transformations of nitrogen, physiological processes, and even historical legacies of nitrogen sources. Critically, studies should move beyond simply describing isotopic patterns to focusing on the consequences of salmon-derived nutrients by quantifying the condition and fitness of organisms putatively using those resources.

Congener Patterns of Persistent Organic Pollutants Establish the Extent of Contaminant Biotransport by Pacific Salmon in the Great Lakes.

In the Great Lakes, introduced Pacific salmon (Oncorhynchus spp.) can transport persistent organic pollutants (POPs), such as polychlorinated biphenyls (PCBs) and polybrominated diphenyl ethers (PBDEs), to new environments during their spawning migrations. To explore the nature and extent of POP biotransport by salmon, we compared 58 PCB and 6 PBDE congeners found in spawning salmon directly to those in resident stream fish. We hypothesized that stream fish exposed to salmon spawners would have congener patterns similar to those of salmon, the presumed contaminant source. Using permutational multivariate analysis of variance (PERMANOVA) and nonmetric multidimensional scaling (NMDS), we found that POP congener patterns of Pacific salmon varied among regions in the Great Lakes basin (i.e., Lake Huron, Lake Michigan, or Lake Superior), tissue type (whole fish or eggs), and contaminant type (PCB or PBDE). For stream-resident fish, POP congener pattern was influenced by the presence of salmon, location (i.e., Great Lakes Basin), and species identity (i.e., brook trout [Salvelinus fontinalis] or mottled sculpin [Cottus bairdii]). Similarity in congener patterns indicated that salmon are a source of POPs to brook trout in stream reaches receiving salmon spawners from Lake Michigan and Lake Huron but not from Lake Superior. Congener patterns of mottled sculpin differed from those of brook trout and salmon, suggesting that brook trout and mottled sculpin either use salmon tissue to differing degrees, acquire POPs from different dietary sources, or bioaccumulate or metabolize POPs differently. Overall, our analyses identified the important role of salmon in contaminant biotransport but also demonstrated that the extent of salmon-mediated POP transfer and uptake in Great Lakes tributaries is location- and species-specific.

Resident fishes display elevated organic pollutants in salmon spawning streams of the great lakes.

Pacific salmon (Oncorhynchus spp.) can transport bioaccumulated organic pollutants to stream ecosystems where they spawn and die. We quantified PCBs, DDE, and PBDEs in resident fishes from 13 Great Lakes tributaries to assess biotransport of pollutants associated with introduced Pacific salmon. Resident fishes sampled from salmon spawning reaches had higher mean pollutant concentrations than those from upstream reaches lacking salmon (93.5 and 4.1 µg x kg(-1) [PCB], 24.0 and 3.1 µg x kg(-1) [DDE], 8.5 and 1.0 µg x kg(-1) [PBDE], respectively), but differences varied substantially among lake basins. In Lake Michigan tributaries, PCB concentrations in resident fishes from salmon reaches were over four times higher than those from salmon reaches in Lake Huron and over 30 times higher than those from Lake Superior. Moreover, resident fish pollutant concentrations were better explained by pollutant inputs from salmon (µg x m(-2); R(2) = 0.76 [PCB], 0.64 [DDE], 0.64 [PBDE]) than by land development/agriculture, watershed area, resident fish species, body length, or lipid content. These results suggest that pollutant dispersal to stream ecosystems via biotransport is an often overlooked consequence of salmon stocking and historical food web contamination in the Great Lakes. Our findings have implications for Great Lakes management, including dam removal and wildlife conservation.

Pacific salmon effects on stream ecosystems: a quantitative synthesis.

Pacific salmon (Oncorhynchus spp.) disturb sediments and fertilize streams with marine-derived nutrients during their annual spawning runs, leading researchers to classify these fish as ecosystem engineers and providers of resource subsidies. While these processes strongly influence the structure and function of salmon streams, the magnitude of salmon influence varies widely across studies. Here, we use meta-analysis to evaluate potential sources of variability among studies in stream ecosystem responses to salmon. Results obtained from 37 publications that collectively included 79 streams revealed positive, but highly inconsistent, overall effects of salmon on dissolved nutrients, sediment biofilm, macroinvertebrates, resident fish, and isotopic enrichment. Variation in these response variables was commonly influenced by salmon biomass, stream discharge, sediment size, and whether studies used artificial carcass treatments or observed a natural spawning run. Dissolved nutrients were positively related to salmon biomass per unit discharge, and the slope of the relationship for natural runs was five to ten times higher than for carcass additions. Mean effects on ammonium and phosphorus were also greater for natural runs than carcass additions, an effect attributable to excretion by live salmon. In contrast, we observed larger positive effects on benthic macroinvertebrates for carcass additions than for natural runs, likely because disturbance by live salmon was absent. Furthermore, benthic macroinvertebrates and biofilm associated with small sediments (<32 mm) displayed a negative response to salmon while those associated with large sediments (>32 mm) showed a positive response. This comprehensive analysis is the first to quantitatively identify environmental and methodological variables that influence the observed effects of salmon. Identifying sources of variation in salmon-stream interactions is a critical step toward understanding why engineering and subsidy effects vary so dramatically over space and time, and toward developing management strategies that will preserve the ecological integrity of salmon streams.

Timber harvest transforms ecological roles of salmon in southeast Alaska rain forest streams.

Although species commonly modify habitats and thereby influence ecosystem structure and function, the factors governing the ecological importance of these modifications are not well understood. Pacific salmon have repeatedly been shown to positively influence the abundance of benthic biota by annually transferring large quantities of nutrients from marine systems to the nutrient-poor freshwaters in which they spawn. Conversely, other studies have demonstrated that salmon can negatively influence the abundance of freshwater biota, an effect attributed to bioturbation during upstream migration and nest construction. The factors determining which of these contrasting ecological effects predominates are unknown, including how human activities, such as land use, influence ecological responses to salmon. We sampled a key basal food resource, sediment biofilm, in seven southeast Alaskan streams impacted to varying degrees by timber harvest. Biofilm abundance (measured as chlorophyll a and ash-free dry mass) was positively related to timber-harvest intensity prior to salmon arrival. However, during the salmon run, an inverse relationship emerged of more abundant biofilm in less-harvested watersheds. Among-stream variability in biofilm response to salmon was largely explained by sediment particle size, which was larger in less-harvested watersheds. Collectively, these results suggest that, by altering stream sediment size, timber harvest transformed the dominant effect of salmon from nutrient enrichment to physical disturbance, thus modifying nutrient linkages between marine and freshwater ecosystems.