Species-specific patterns of aggregation of wild fish around fish farms

Fish-farming structures are widespread in coastal waters and are highly attractive to wild fish. Several studies have estimated that tons to tens of tons of wild fish aggregate around fish farms. These estimates assumed that the majority of wild fish are concentrated immediately beneath farms, although this assumption has never been explicitly tested. We tested the hypothesis that abundances of wild fish would be greatest immediately beneath farms and progressively diminish with distance at 4 full-scale coastal salmon (Salmo salar) farms in Norway. At each farm, fish were counted with a video-camera system at 5 different distances from the cages (farm = 0 m, 25, 50, 100 and 200 m) throughout the water column on three separate days. Combined across all locations and times, the total abundance of wild fish was 20 times greater at the farm than at the 200 m sampling distance. Saithe (Pollachius virens) dominated assemblages at all 4 farms and were consistently significantly more abundant at the farm than at the 25–200 m distances. This ‘tight aggregation’ around farms corresponds to the reliance of saithe on waste feed when they school near farms. In contrast, patterns of distribution of both cod (Gadus morhua) and poor cod (Trisopterus minutus) varied among farms, with either highest abundances at the farm or a more even distribution of abundance across all 5 distances sampled. No specific pattern of aggregation was evident for the bottom-dwelling haddock (Melanogrammus aeglefinus). Our results suggest that the present 100 m no-fishing zone around salmon farms protects the greatest proportion of farm-aggregated saithe and cod from fishing during the daytime. However, whether this reduces their overall susceptibility to fishing requires further research regarding nighttime distribution and movements.     

Sediment infiltration traps: their use to monitor salmonid spawning habitat in headwater tributaries of the Cascapédia River,Québec

Sediment infiltration can clog salmon nests and reduce egg survival. As a countermeasure, environmental managers often deploy infiltration traps to monitor sediment infiltration. Traps provide a repeatable means of measuring infiltration and enable comparisons to be made between sites. Results from infiltration rates measured with traps have also been used to estimate infilling rates into salmon nests. Application of these data is questionable, as the composition of the bed and the amount of fine sediment within the bed is known to affect infiltration rates. Thus, infiltration rates measured with infiltration traps may differ from the infiltration rates occurring in redd and riffle gravels. To examine how relationships between sediment infiltration rates varied between four watersheds, we continuously monitored suspended sediment transport, shear stress and infiltration rates at four sites over 5 months. We also compared infiltration rates measured with infiltration traps with changes in the hydraulic conductivity and subsurface grain size distribution of adjacent artificially constructed salmon nests and natural riffle gravels. Among the four watersheds, clear differences in sediment infiltration rates were observed. The differences correlated with the subsurface silt content but no strong relationship existed between land‐use or basin physiography/geology. Despite observing an average of 30 kg m⁻² of sediment finer than 2 mm being deposited in the infiltration traps during the study, no change in redd or riffle substrate was observed. If the deposition rates measured with the traps reflect the processes in redds, enough sediment would have been deposited to inhibit egg emergence. However, no reduction in egg survival to the eyed stage was observed. In summary, our results show that infiltration traps with clean gravels can be used to detect intersite differences in sediment transport regimes. Extrapolations of sediment infiltration rates measured with such collectors to estimate infiltration rates in redds or riffles is, however, flawed. Copyright © 2005 John Wiley & Sons, Ltd.     

Modeling the effects of riparian planting strategies on stream temperature: Increasing suitable habitat for endangered Formosan Landlocked Salmon in Shei‐Pa National Park,Taiwan

Reducing or stabilizing the stream temperature of ChiChiaWan Creek is a crucial work for Formosan Landlocked Salmon because ChiChiaWan Creek is the only one habitat for this endangered species. Planting trees in the riparian zone would be one of the alternatives. The purpose of this study was to evaluate the effects of several planting strategies on daily maximum stream temperature along the river. The results showed the effective vegetative shading angles should be more than 50° along ChiChiaWan Creek to reduce the direct solar radiation heating effectively. Upstream planting with 70° vegetative shading angle could be the most effective way among all the scenarios. However, this planting strategy could not improve the worst situations in summer because of the large solar elevation angles. The upstream planting in ChiChiaWan Creek was strongly recommended because the canopies could be easier to extend to totally cover the narrow width of river producing the most effective shades. Practicing the upstream planting with 90° vegetative shading angle can increase more than 1 km long suitable habitats for the endangered Salmon in summer. Alternatively, the west‐side planting scenario was the second effective way for temperature reduction. Our result provided a useful suggestion for the authorities in charge of saving the Formosan Landlocked Salmon, particularly under the stress of global warming. Copyright © 2011 John Wiley & Sons, Ltd.     

Bedform morphology of salmon spawning areas in a large gravel-bed river

While the importance of river channel morphology to salmon spawning habitat is increasingly recognized, quantitative measures of the relationships between channel morphology and habitat use are lacking. Such quantitative measures are necessary as management and regulatory agencies within the Pacific Northwest region of the USA, and elsewhere, seek to quantify potential spawning habitat and develop recovery goals for declining salmon populations. The objective of this study was to determine if fall Chinook salmon (Oncorhynchus tshawytscha) spawning areas in the Snake River, Idaho, USA, were correlated with specific bedform types at the pool–riffle scale. A bedform differencing technique was used to objectively quantify the longitudinal riverbed profile into four distinct pool–riffle units that were independent of discharge. The vertical location of thalweg points within these units was quantified with a riffle proximity index. Chinook salmon spawning areas were mapped and correlated with the pool–riffle units through the use of cross-tabulation tables. The results indicate that 84% of fall Chinook salmon spawning areas were correlated with riffles (χ² = 57.5, df = 3, p < 0.001), with 53% of those areas located on the upstream side of riffle crests. The majority of Snake River fall Chinook salmon spawning occurred at elevations greater than 80% of the difference in elevation between the nearest riffle crest and pool bottom. The analyses of bedform morphology will assist regional fish managers in quantifying existing and potential fall Chinook salmon spawning habitat, and will provide a quantitative framework for evaluating general ecological implications of channel morphology in large gravel-bed rivers.     

Breeding,settlement and survival of barnacles at artificially modified shore levels at Leigh,New Zealand

The breeding and settlement seasons of the intertidal barnacles Chamaesipho brunnea, C. columna and Elminius plicatus were determined, and experiments using cages on the shore to examine barnacle survival and growth rate at four levels were carried out at Leigh during 1964–66.

Chamaesipho brunnea bred in spring and summer; C. columna and E. plicatus bred throughout the year, but settled intermittently. When protected from predators and from overgrowth by algae, all three species survived at levels lower than normal. At higher levels C. columna, E. modestus, Balanus trigonus and Tetraclita purpuraseens survived for significant periods (C. brunnea occurs naturally higher on the shore than the highest experimental level).     

Diet of landlocked sockeye salmon (Oncorhynchus nerka) and trout in the Waitaki lakes,New Zealand

Landlocked sockeye salmon (Oncorhynchus nerka), ranging in fork length (FL) from 105 to 313 mm, were captured in fine‐mesh gill nets set in the limnetic zone of the Waitaki hydro lakes (44° 30′ S, 170° 10’ E) in the South Island, New Zealand. A total of 443 stomachs was examined and the frequency of occurrence, volume and weight of prey items calculated. In the Ahuriri Arm of Lake Benmore the principal food (54% by weight) was zooplankton (Boeckella dilatata) whereas in the Haldon Arm of Lake Benmore it was larval and juvenile common bullies (Gobiomorphus cotidi‐anus) (73% by volume). In Lake Waitaki in winter, salmon had eaten insects (43% by volume) with smaller amounts of snails (Potamopyrgus antipo‐darum, 23%) and bullies (24%). In Lake Ohau adult insects may be an important food. There were also variations in diet with season and fish size. The stomachs of 147 brown trout (Salmo trutta) and 181 rainbow trout (S. gairdnerii) caught in the same gill nets were also examined. In contrast to sockeye salmon stomachs they contained negligible amounts of zooplankton (< 1% by weight) and large amounts of aquatic insects (50–58% by weight in the Ahuriri Arm of Lake Benmore). Comparisons with juvenile sockeye salmon and kokanee in North American lakes are made. The impact of introductions of sockeye salmon into other New Zealand lakes is discussed.     

Methods for locating the proper position of a planned fishway entrance near a hydropower tailrace

The spatial distribution of upstream migrating Atlantic salmon (Salmo salar) spawners was studied in 2008 and 2009 in the surroundings of the tailrace from a hydropower station in the River Umeälven. This area is problematic because the fish have difficulties finding their way into the original riverbed, which prompted an investigation on the feasibility of adding a fishway in the area. Echo sounding was used in 2008 to investigate the spatial use of fish in the tailrace channel during the time of salmon migration. Presence of other fish species was so low that all echo sounding detections were assumed to be salmon. In 2009, data on wild radio-tagged salmon (n = 94) was collected in the same tailrace by an array of underwater antennas with a detection radius of approximately 10 m, to validate the results from the echo sounding. Both types of surveys showed aggregations of fish in one part of the tailrace. As a final step, Computational Fluid Dynamics (CFD) modeling was performed to analyze hydrodynamics. This CFD modeling showed a coincidence of fish detections in areas with turbulence intensities between 0.6 and 0.8, which may be an indication that the fish are holding in these areas to reduce energy expenditure during migration. A high proportion of the radio-tagged salmon were observed in the tailrace area (a median of 21 days between their first and last detection), indicating that the salmon are delayed in their upstream migration by the attraction to the tailrace in this area. The overall data on fish behavior in the tailrace led us to suggest a location for a new fishway where the fish aggregation was most pronounced. The number of detections from either technology had similar distributions over the tailrace, suggesting that the results are comparable. Thus, the split beam echo sounder can preferably be used to investigate which location is most appropriate for locations of entrances to new fishways since many more individual fishes are covered by this method compared to telemetry. Furthermore, there is no need to handle the fish in the echo sounding studies as is required in telemetry studies.