The diadromous watersheds-ocean continuum: Managing diadromous fish as a community for ecosystem resilience

Diadromous fishes play important ecological roles by delivering ecosystem services and making crucial connections along the watersheds-ocean continuum. However, it is difficult to fully understand the community-level impacts and cumulative benefits of diadromous fish migrations, as these species are most often considered individually or in small groups. Their interactions at a community level (e.g., interdependencies such as predation, co-migration, and habitat conditioning) and the connections between their ecosystem roles and functions (e.g., cumulative marine-derived nutrient contributions, impacts on stream geomorphology) are yet to be fully understood. Similarly, freshwater, estuarine, and marine ecosystems are often considered as independent parts, limiting understanding of the importance of connections across systems. We argue that not considering the ecosystem interdependence and importance of diadromous fish as a community currently hinders the implementation of the large-scale management required to increase ecosystem resilience and fish productivity across the full range of these species. We developed a conceptual model, the Diadromous Watersheds-Ocean Continuum (DWOC), that uses ecosystem services to promote a more holistic approach to the management of the diadromous community and encourages an integrated understanding of the ecosystem connections made by these species. DWOC provides a framework for discussions that can help identify research and management needs, discuss the trade-offs of different management options, and analyze what pressing questions impede the implementation of large-scale management solutions toward a more ecosystem-based management approach.

Effects of anthropogenic subsidy and glyphosate on macroinvertebrates in streams

Streams and surrounding terrestrial ecosystems are closely linked by numerous resource subsidies including anthropogenic subsidies which are increasingly entering streams due to intensive human activities. Also, streams are threatened by stressors such as glyphosate-the most widely used herbicide worldwide. However, the ecological consequences of anthropogenic subsidies and glyphosate on freshwaters are not fully understood. Here, we deployed leaf litter (Cinnamomum camphora) bags containing neither, either, or both treatments of anthropogenic carrion subsidy (chicken meat) and glyphosate (coated in agar) in four streams, which had different land use (i.e., forest, village, and suburban) in Huangshan, Anhui Province, China. We aimed to investigate the individual and combined effects of anthropogenic carrion subsidy and glyphosate on macroinvertebrates in streams and whether these effects differ with land use change. Macroinvertebrate communities significantly differed among streams: biodiversity index and total taxon richness were highest in village streams and lowest in suburban stream. Overall effects of carrion subsidy and glyphosate on macroinvertebrates were not significant. However, several taxa were affected in one or more streams by the individual or combined effects of carrion subsidy and glyphosate, indicating the importance of local community structure and physical habitats in driving the response of macroinvertebrates to carrion subsidy and glyphosate. Collectively, these results imply that the effects of carrion subsidy and glyphosate on macroinvertebrates are site-specific, and future studies should cover more streams and last longer time to better understand the ecological mechanisms driving such pattern.

Aquatic food-web dynamics following incorporation of nutrients derived from Atlantic anadromous fishes

Changes in the isotopic composition (δ¹³ C and δ¹⁵ N) in biofilm, macro-invertebrates and resident salmonids were used to characterize temporal dynamics of marine derived nutrients (MDNs) incorporation between stream reaches with and without MDN inputs. Five Atlantic rivers were chosen to represent contrasting MDN subsidies: four rivers with considerable numbers of anadromous fishes; one river with little MDN input. Rainbow smelt Osmerus mordax, alewife Alosa pseudoharengus, sea lamprey Petromyzon marinus and Atlantic salmon Salmo salar, were the primary anadromous species for the sampled rivers. Regardless of the spatial resolution or the pathway of incorporation, annual nutrient pulses from spawning anadromous fishes had a positive effect on isotopic enrichment at all trophic levels (biofilm, 1·2-5·4‰; macro-invertebrates, 0·0-6·8‰; fish, 1·2-2·6‰). Community-wide niche space shifted toward the marine-nutrient source, but the total ecological niche space did not always increase with MDN inputs. The time-integrated marine-nutrient resource contribution to the diet of S. salar parr and brook trout Salvelinus fontinalis ranged between 16·3 and 36·0% during anadromous fish-spawning periods. The high degree of spatio-temporal heterogeneity in marine-nutrient subsidies from anadromous fishes lead to both direct and indirect pathways of MDN incorporation into stream food webs. This suggests that organisms at many trophic levels derive a substantial proportion of their energy from marine resources when present. The current trend of declining anadromous fish populations means fewer nutrient-rich marine subsidies being delivered to rivers, diminishing the ability to sustain elevated riverine productivity.

A Critical Assessment of the Ecological Assumptions Underpinning Compensatory Mitigation of Salmon-Derived Nutrients

We critically evaluate some of the key ecological assumptions underpinning the use of nutrient replacement as a means of recovering salmon populations and a range of other organisms thought to be linked to productive salmon runs. These assumptions include: (1) nutrient mitigation mimics the ecological roles of salmon, (2) mitigation is needed to replace salmon-derived nutrients and stimulate primary and invertebrate production in streams, and (3) food resources in rearing habitats limit populations of salmon and resident fishes. First, we call into question assumption one because an array of evidence points to the multi-faceted role played by spawning salmon, including disturbance via redd-building, nutrient recycling by live fish, and consumption by terrestrial consumers. Second, we show that assumption two may require qualification based upon a more complete understanding of nutrient cycling and productivity in streams. Third, we evaluate the empirical evidence supporting food limitation of fish populations and conclude it has been only weakly tested. On the basis of this assessment, we urge caution in the application of nutrient mitigation as a management tool. Although applications of nutrients and other materials intended to mitigate for lost or diminished runs of Pacific salmon may trigger ecological responses within treated ecosystems, contributions of these activities toward actual mitigation may be limited.

Spatial and temporal variability of macroinvertebrates in spawning and non-spawning habitats during a salmon run in Southeast Alaska

Spawning salmon create patches of disturbance through redd digging which can reduce macroinvertebrate abundance and biomass in spawning habitat. We asked whether displaced invertebrates use non-spawning habitats as refugia in streams. Our study explored how the spatial and temporal distribution of macroinvertebrates changed during a pink salmon (Oncorhynchus gorbuscha) spawning run and compared macroinvertebrates in spawning (riffle) and non-spawning (refugia) habitats in an Alaskan stream. Potential refugia included: pools, stream margins and the hyporheic zone, and we also sampled invertebrate drift. We predicted that macroinvertebrates would decline in riffles and increase in drift and refugia habitats during salmon spawning. We observed a reduction in the density, biomass and taxonomic richness of macroinvertebrates in riffles during spawning. There was no change in pool and margin invertebrate communities, except insect biomass declined in pools during the spawning period. Macroinvertebrate density was greater in the hyporheic zone and macroinvertebrate density and richness increased in the drift during spawning. We observed significant invertebrate declines within spawning habitat; however in non-spawning habitat, there were less pronounced changes in invertebrate density and richness. The results observed may be due to spawning-related disturbances, insect phenology, or other variables. We propose that certain in-stream habitats could be important for the persistence of macroinvertebrates during salmon spawning in a Southeast Alaskan stream.