Climate Change Expands the Spatial Extent and Duration of Preferred Thermal Habitat for Lake Superior Fishes

Climate change is expected to alter species distributions and habitat suitability across the globe. Understanding these shifting distributions is critical for adaptive resource management. The role of temperature in fish habitat and energetics is well established and can be used to evaluate climate change effects on habitat distributions and food web interactions. Lake Superior water temperatures are rising rapidly in response to climate change and this is likely influencing species distributions and interactions. We use a three-dimensional hydrodynamic model that captures temperature changes in Lake Superior over the last 3 decades to investigate shifts in habitat size and duration of preferred temperatures for four different fishes. We evaluated habitat changes in two native lake trout (Salvelinus namaycush) ecotypes, siscowet and lean lake trout, Chinook salmon (Oncorhynchus tshawytscha), and walleye (Sander vitreus). Between 1979 and 2006, days with available preferred thermal habitat increased at a mean rate of 6, 7, and 5 days per decade for lean lake trout, Chinook salmon, and walleye, respectively. Siscowet lake trout lost 3 days per decade. Consequently, preferred habitat spatial extents increased at a rate of 579, 495 and 419 km² per year for the lean lake trout, Chinook salmon, and walleye while siscowet lost 161 km² per year during the modeled period. Habitat increases could lead to increased growth and production for three of the four fishes. Consequently, greater habitat overlap may intensify interguild competition and food web interactions. Loss of cold-water habitat for siscowet, having the coldest thermal preference, could forecast potential changes from continued warming. Additionally, continued warming may render more suitable conditions for some invasive species.     

Dynamics of an age‐structured population drawn from a random numbers table

I constructed age‐structured populations by drawing numbers from a random numbers table, the constraints being that within a cohort each number be smaller than the preceding number (indicating that some individuals died between one year and the next) and that the first two‐digit number following 00 or 01 ending one cohort’s life be the number born into the next cohort. Populations constructed in this way showed prolonged existence with total population numbers fluctuating about a mean size and with long‐term growth rate (r) ≈ 0. The populations’ birth rates and growth rates and the females’ per capita fecundity decreased significantly with population size, whereas the death rates showed no significant relationship to population size. These results indicate that age‐structured populations can persist for long periods of time with long‐term growth rates of zero in the absence of negative‐feedback loops between a population’s present or prior density and its birth rate, growth rate, and fecundity, contrary to the assumption of density‐dependent regulation hypotheses. Thus, a long‐term growth rate of zero found in natural populations need not indicate that a population’s numbers are regulated by density‐dependent factors.     

Production of juvenile salmonids in small Norwegian streams is affected by agricultural land use

1. We estimated the biomass and production of juvenile anadromous brown trout (Salmo trutta) and Atlantic salmon (Salmo salar) (parr) in 12 streams in the Skagerrak area of Norway to identify controlling environmental factors, such as land‐use and water chemistry. 2. Production estimates correlated positively with fish density in early summer, but not with the size of the catchment. The summer biomass of age‐0 brown trout and Atlantic salmon was smaller than that of age‐1 and constituted 27.4 and 25.7%, respectively, of the total biomass of the two groups. 3. Mean production of brown trout from July to September varied between streams, but in most cases it was below 2 g 100 m^?2 day^?1. Yearly cohort production from age‐0 in July to age‐1 in July was 10 g m^?2 or less, with mean annual production of 1.32 g 100 m^?2 day^?1, equivalent to 4.8 g m^?2 year^?1. The corresponding annual cohort production of Atlantic salmon was 0.38 g 100 m^?2 day^?1 or 1.4 g m^?2 year^?1. Annual production to biomass ratio (P/B) for brown trout of the same cohort in the various streams was between 1.47 and 4.37; the overall mean (±SD) for all streams was 2.25 ± 0.94. Mean turnover rate of Atlantic salmon was 2.73 ± 0.24. 4. Production of 0+ brown trout during the summer correlated significantly with the percentage of agricultural land and forest/bogs in the catchment, with maxima at 20 and 75%, respectively. Age‐0 brown trout production also correlated with concentration of nitrogen and calcium in the water, with maxima at 2.4 and 14 mg L^?1, respectively. 5. The results support the hypothesis that brown trout parr production reflects the quality of their habitat, as indicated by the dome‐shaped relationship between percentage of agricultural land and the concentration of nitrogen and calcium in the water.     

Assessing Habitat Suitability for Juvenile Atlantic Salmon in Relation to In‐Stream Restoration and Discharge Variability

In‐stream restoration often aims at increasing the availability of the stream habitat suitable for salmonid fishes, thus creating potential for increased fish abundance. We assessed the success of in‐stream restoration of River Kiiminkijoki, northern Finland, by combining River2D habitat hydraulic modeling and fish density monitoring at the same sites, with data from multiple restored and reference reaches for 3 years both before and after restoration. We modeled the effects of restoration on the area suitable (weighted usable area, WUA) for juvenile Atlantic salmon from post‐hatching to age‐1 fish. Wetted width in the restored reaches increased by 8.1% on average compared with only ?0.2% change in the reference reaches. Habitat time series across 10 years showed significant increases in the amount of suitable habitat under summer conditions for both age‐0 and age‐1 salmon. However, improvement of overwintering habitats was marginal or nonexistent. Densities of age‐1 salmon showed no response to restoration. Low river discharge during the winter was correlated with low salmon densities the following summer. It thus appears that variability in wintertime discharge, and associated high interannual variation of WUA values, overrode the almost 20% increase in average post‐ versus pre‐restoration summertime WUA. Our study shows that the combination of hydraulic modeling and biological monitoring is a promising approach to stream restoration assessment.     

Changes in seasonal climate outpace compensatory density‐dependence in eastern brook trout

Understanding how multiple extrinsic (density‐independent) factors and intrinsic (density‐dependent) mechanisms influence population dynamics has become increasingly urgent in the face of rapidly changing climates. It is particularly unclear how multiple extrinsic factors with contrasting effects among seasons are related to declines in population numbers and changes in mean body size and whether there is a strong role for density‐dependence. The primary goal of this study was to identify the roles of seasonal variation in climate driven environmental direct effects (mean stream flow and temperature) vs. density‐dependence on population size and mean body size in eastern brook trout (Salvelinus fontinalis). We use data from a 10‐year capture‐mark‐recapture study of eastern brook trout in four streams in Western Massachusetts, USA to parameterize a discrete‐time population projection model. The model integrates matrix modeling techniques used to characterize discrete population structures (age, habitat type, and season) with integral projection models (IPMs) that characterize demographic rates as continuous functions of organismal traits (in this case body size). Using both stochastic and deterministic analyses we show that decreases in population size are due to changes in stream flow and temperature and that these changes are larger than what can be compensated for through density‐dependent responses. We also show that the declines are due mostly to increasing mean stream temperatures decreasing the survival of the youngest age class. In contrast, increases in mean body size over the same period are the result of indirect changes in density with a lesser direct role of climate‐driven environmental change.     

Short‐term Costs and Benefits of Habitat Complexity for a Territorial Fish

An increase in habitat complexity is thought to decrease visibility and the territory size of visually oriented animals. Hence, the addition of physical structure has been viewed as a useful restoration technique to increase the density of territorial species, particularly in stream fishes. However, a decrease in territory size may have a negative effect on the fitness of individual organisms. We attempted to quantify some of the positive and negative effects of increasing habitat structure on the behaviour and growth rate of wild young‐of‐the‐year (YOY) Atlantic salmon. Fish were exposed to one of two habitat treatments in mesh enclosures in Catamaran Brook, New Brunswick: a gravel substrate (low complexity) or a gravel substrate with boulders added (high complexity). Wild‐caught individuals were tagged, weighed and measured before being stocked at densities of 1·m^?2 for 7‐d trials. While fish from high‐complexity treatments benefited from lower rates of aggression, they also had lower foraging rates and smaller territories compared to those in low‐complexity treatments. Specific growth rate, however, did not differ significantly between treatments. While the addition of structure to a habitat may be beneficial at the population level in terms of an increase in population density, our results suggested that individual fish may pay some short‐term costs in these environments. Further research is needed to evaluate the longer term costs and benefits of adding structure to improve the habitat quality for stream salmonids.     

Density-dependent growth rate in an age-structured population: a field study on stream-dwelling brown trout Salmo trutta

A field experiment during autumn, winter and spring was performed in a small stream on the west coast of Sweden, aiming to examine the direct and indirect consequences of density-dependent intercohort competition in brown trout Salmo trutta . Individual growth rate, recapture rate and site fidelity were used as response variables in the young-of-the-year (YOY) age class, experiencing two different treatments: presence or absence of yearlings and over-yearlings (age ≥ 1+ year individuals). YOY individuals in stream sections with reduced density of age ≥ 1+ year individuals grew significantly faster than individuals experiencing natural cohort structure. In the latter, growth rate was negatively correlated with density and biomass of age ≥ 1+ year individuals, which may induce indirect effects on year-class strength through, for example, reduced fecundity and survival. Movement of YOY individuals and turnover rate ( i.e. proportion of untagged individuals) were used to demonstrate potential effects of intercohort competition on site fidelity. While YOY movement was remarkably restricted (83% recaptured within 50 m from the release points), turnover rate was higher in sections with reduced density of age ≥1+ year individuals, suggesting that reduced density of age ≥1+ year individuals may have released favourable microhabitats.     

Context‐specific influence of water temperature on brook trout growth rates in the field

1. Modelling the effects of climate change on freshwater fishes requires robust field‐based estimates accounting for interactions among multiple factors. 2. We used data from an 8‐year individual‐based study of a wild brook trout (Salvelinus fontinalis) population to test the influence of water temperature on season‐specific growth in the context of variation in other environmental (i.e. season, stream flow) or biotic factors (local brook trout biomass density and fish age and size) in West Brook, a third‐order stream in western Massachusetts, U.S.A. 3. Changes in ambient temperature influenced individual growth rates. In general, higher temperatures were associated with higher growth rates in winter and spring and lower growth rates in summer and autumn. However, the effect of temperature on growth was strongly context‐dependent, differing in both magnitude and direction as a function of season, stream flow and fish biomass density. 4. We found that stream flow and temperature had strong and complex interactive effects on trout growth. At the coldest temperatures (in winter), high stream flows were associated with reduced trout growth rates. During spring and autumn and in typical summers (when water temperatures were close to growth optima), higher flows were associated with increased growth rates. In addition, the effect of flow at a given temperature (the flow‐temperature interaction) differed among seasons. 5. Trout density negatively affected growth rate and had strong interactions with temperature in two of four seasons (i.e. spring and summer) with greater negative effects at high temperatures. 6. Our study provided robust, integrative field‐based estimates of the effects of temperature on growth rates for a species which serves as a model organism for cold‐water adapted ectotherms facing the consequences of environmental change. Results of the study strongly suggest that failure to derive season‐specific estimates, or to explicitly consider interactions with flow regime and fish density, will seriously compromise our ability to predict the effects of climate change on stream fish growth rates. Further, the concordance we found between empirical observations and likely energetic mechanisms suggests that our general results should be relevant at broader spatial and temporal scales.     

Static habitat partitioning and dynamic selection by sympatric young Atlantic salmon and brown trout in south-west England streams

Direct underwater observation of micro‐habitat use by 1838 young Atlantic salmon Salmo salar [mean L_T 7·9 ± 3.1(s.d.) cm, range 3·19] and 1227 brown trout Salmo trutta (L_T 10·9 ± 5·0 cm, range 3·56) showed both species were selective in habitat use, with differences between species and fish size. Atlantic salmon and brown trout selected relatively narrow ranges for the two micro‐habitat variables snout water velocity and height above bottom, but with differences between size‐classes. The smaller fishes <7 cm held positions in slower water closer to the bottom. On a larger scale, the Atlantic salmon more often used shallower stream areas, compared with brown trout. The larger parr preferred the deeper stream areas. Atlantic salmon used higher and slightly more variable mean water velocities than brown trout. Substrata used by the two species were similar. Finer substrata, although variable, were selected at the snout position, and differences were pronounced between size‐classes. On a meso‐habitat scale, brown trout were more frequently observed in slow pool‐glide habitats, while young Atlantic salmon favoured the faster high‐gradient meso‐habitats. Small juveniles <7 cm of both species were observed most frequently in riffle‐chute habitats. Atlantic salmon and brown trout segregated with respect to use of habitat, but considerable niche overlap between species indicated competitive interactions. In particular, for small fishes <7 cm of the two species, there was almost complete niche overlap for use of water depth, while they segregated with respect to water velocity. Habitat suitability indices developed for both species for mean water velocity and water depth, tended to have their optimum at lower values compared with previous studies in larger streams, with Atlantic salmon parr in the small streams occupying the same habitat as favoured by brown trout in larger streams. The data indicate both species may be flexible in their habitat selection depending on habitat availability. Species‐specific habitat overlap between streams may be complete. However, between‐species habitat partitioning remains similar.     

Validation and implications of a growth model for brown trout, Salmo trutta, using long-term data from a small stream in north-west England

Summary 1. The objectives were: (i) to check the validity of a new growth model; (ii) to examine the relationship between population density and both mean mass and mean growth rate and (iii) to discover if compensatory growth occurred. First (0+) and second (1+) year‐old juvenile sea‐trout were sampled by electrofishing at the beginning and end of the summer from 1967 to 2000. Additional samples were taken in some years in winter and in the critical period for survival when the fry first emerge from the gravel. The trout left the stream as pre‐smolts in May, soon after their second birthday. 2. A growth model ( Elliott, Hurley & Fryer, 1995 ) estimated the mean mass of the trout over the 2 years spent in fresh water. The date and mean mass at the start of the growth period were defined as the median date for fry emerging from the gravel and their mean mass at emergence, both being estimated from individual‐based models ( Elliott & Hurley, 1998a, b ). 3. The variation in mean mass among year‐classes was small for newly‐emerged fry (CV = 6.2%), maximum at the start of the first summer of the life cycle (CV = 38.1%), and then decreased gradually for successive life‐stages to a low value for pre‐smolts (CV = 10.8%). Mean mass was not related to population density and, therefore, mean growth rate was density‐independent. Growth in the first, but not the second, winter of the life cycle was lower than model prediction, but when it was assumed in the model that there was no first‐winter growth, there was good agreement in most year‐classes between model estimated values and observed mean mass. Exceptions were that mean masses and growth rates for 0+ trout after four summer droughts were lower than expected, but compensatory growth followed, so that observed and expected masses were similar for 1+ trout. 4. Pre‐smolt mean mass on 30 April measured total growth achieved in the freshwater phase of the life cycle. This was significantly related to mean mass at the end of the first and second summers of the life cycle, but not to the emergence date and mean mass of emerging fry. 5. These juvenile sea‐trout were growing at their maximum potential in most year‐classes but when this was not achieved, compensatory growth soon restored their mass to values expected from the model. This ensured a low variation in the mean mass of pre‐smolts just before they migrated to the sea. However, the latter mass was higher in more recent year‐classes (1987–98) than in previous ones (1967–86), demonstrating the effect of slightly higher stream temperature. This study has shown the importance of developing realistic growth models in order to detect departure from maximum potential growth, and the more subtle effects of temperature change, possibly due to the effects of climate change.     

Life History,Reproduction, Growth,Population Dynamics and Production of Gammarus aequicauda (Crustacea: Amphipoda) at Extremely Low Salinities in a Mediterranean Lagoon

Aspects of the biology of Gammarus aequicauda were studied at extremely low salinities (0.3–5.7 psu) in Monolimni Lagoon, N. Aegean Sea. Samples were collected monthly from February 1998 to February 1999. Breeding occurred continuously but peaked in late spring, late summer and autumn and three cohorts were produced. The spring and summer cohorts showed fast growth (0.15 mm d^–1), accelerated maturity and life span of about three and seven months respectively, while the overwintering cohort showed a life span of about nine months. The largest individual had a body length of 23.6 mm. Mean brood size was 54.5 early embryos, while the intramarsupial loss was 46%. Population density sharply increased in late spring, summer and autumn. Secondary production calculated by Hynes' method gave a mean annual density of 1077.4 ind. m^–2, a mean annual crop (B) of 2.93 g DW m^–2, an annual production (P) of 35.03 g DW m^–2 and a P: B ratio of 11.96. Gammarus aequicauda showed a life‐history pattern similar to those previously reported for this species at higher salinity environments with comparatively large final body length and high growth rate during summer, brood size and P: B ratio, but with high embryo loss as well. (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Density‐dependent offspring interactions do not explain macroevolutionary scaling of adult size and offspring size

Most life forms exhibit a correlated evolution of adult size (AS) and size at independence (SI), giving rise to AS–SI scaling relationships. Theory suggests that scaling arises because relatively large adults have relatively high reproductive output, resulting in strong density‐dependent competition in early life, where large size at independence provides a competitive advantage to juveniles. The primary goal of our study is to test this density hypothesis, using large datasets that span the vertebrate tree of life (fishes, amphibians, reptiles, birds, and mammals). Our secondary goal is to motivate new hypotheses for AS–SI scaling by exploring how subtle variation in life‐histories among closely related species is associated with variation in scaling. Our phylogenetically informed comparisons do not support the density hypothesis. Instead, exploration of AS–SI scaling among life‐history variants suggests that steeper AS–SI scaling slopes are associated with evolutionary increases in size at independence. We suggest that a positive association between size at independence and juvenile growth rate may represent an important mechanism underlying AS–SI scaling, a mechanism that has been underappreciated by theorists. If faster juvenile growth is a consequence of evolutionary increases in size at independence, this may help offset the cost of delayed maturation, leading to steeper AS–SI scaling slopes.     

High‐resolution isotope analysis of young alewife Alosa pseudoharengus otoliths: assessment of temporal resolution and reconstruction of habitat occupancy and thermal history

Otoliths of age 0 year alewife Alosa pseudoharengus collected in different Lake Michigan habitats were microsampled, and carbon and oxygen isotope ratios (δ¹⁸O_otolith and δ¹³C_otolith) of resulting microsamples were quantified. To assess the temporal resolution of the method, age and otolith growth rates were also estimated by counting otolith daily growth increments. Core and outer intra‐otolith samples averaged 36 and 23 days, respectively. Because of the accretionary nature of otolith growth, a habitat switch by a larva occurring between 0 and 18 days post‐hatch may not be recognized by this approach. Taking this temporal resolution into account, A. pseudoharengus habitat occupancy and thermal history in nearshore Lake Michigan, and a connecting drowned river‐mouth lake were documented. Comparisons between δ¹⁸O_otolith and δ¹³C_otolith profiles, and isotope values of Lake Michigan habitats suggested that movements by individual fish between a nearshore area of Lake Michigan proper and drowned river‐mouth lake habitats were rare. Some individuals evidently moved between habitats, and such movements occurred during different periods of ontogeny. Thermal reconstructions, based on δ¹⁸O_otolith values suggested that during early life (e.g. first month of life) young A.pseudoharengus appeared to inhabit microhabitats with temperatures greater than mean epilimnetic temperatures. This study demonstrates not only the utility of intra‐otolith geochemical analysis to describe the complexity of fish behaviour in fresh water but also identifies limitations of the present approach.     

Habitat Partitioning in a Patchy Environment: Considering the Role of Intraspecific Competition

Coexistence of many size groups of conspecifics in habitat patches may complicate resource partitioning and increase intraspecific interactions. The objectives of my study were to determine partitioning of habitat among age groups of rainbow trout, Oncorhynchus mykiss, coexisting in pool habitat of a headwater stream, and to determine the role of intraspecific competition for such resource partitioning. The trout population showed size and age specific patterns of habitat use, and trout selected locations based on depth and longitudinal position. This habitat use pattern decreased intraspecific overlap among the trout age groups for use of pool space. I used a removal experiment to determine if two-year old trout constrained habitat use by the smaller conspecifics. Although the experimental results imply that recent intraspecific competition was not present, the absence of competitive exclusion was not clearly demonstrated because of low experimental power. While this study identified habitat partitioning among the trout age groups, it remains unclear whether biotic interactions or size specific requirements were causing the habitat use patterns.     

Non‐additive effects of density‐dependent and density‐independent factors on brown trout vital rates

Interactions between density‐dependent and density‐independent processes can lead to variation in both growth and survival rates. Detecting such effects, however, will often require sampling on an individual level and at the appropriate spatial and temporal scale. This study documents substantial variation in survival and growth of stream‐dwelling brown trout Salmo trutta from a small Norwegian stream. The data is based on seasonal capture–recaptures of individually marked trout on fixed stations during eight years. The fish were small‐sized, rarely reaching sizes larger than 20 cm and ages older than seven years. Density varied between 0.2–0.8 fish m^?2. Variation in survival and recapture probabilities was analysed using program MARK. Apparent survival (the probability of being alive and present within the study area) generally decreased with increasing trout density and increasing drought level (measured as lowest observed water flow) during both winter and summer. Further, there was a significant interaction effect between density and water flow, indicating that density‐dependent effects on survival predominated when environmental conditions were benign (no drought), while density‐independent processes were most important under harsh environmental conditions (drought). Observed length‐at‐age during autumn indicated a more or less linear growth trajectory throughout life, and no effect of density, water flow or temperature was found. However, using the individual‐based capture–recapture data to estimated specific growth rate, significant positive effects of water flow and temperature and a negative effect of density were identified. Thus, the capture–recapture data suggest a strong potential for population regulation at the rather low densities found in this stream, and regulation may occur both through effects on survival and growth.     

Interacting effects of temperature and density on individual growth performance in a wild population of brown trout

相似文献   

Population Dynamics and Spatial Distribution of Coral Reef Fishes: Comparison Between Continuous and Isolated Habitats

Recent studies have shown that there are high degrees of spatial and temporal stability in coral reef fish assemblage structures in a continuous habitat, in contrast to results of observations in isolated habitats. In order to determine the reason for the difference in temporal stability of fish assemblage structures in a continuous habitat site and an isolated habitat site, population dynamics and spatial distributions of coral reef fishes (six species of pomacentrids and two species of apogonids) in the two habitat site were investigated over a 2-year period in an Okinawan coral reef. The population densities of pomacentrid and apogonid species increased in juvenile settlement periods at both sites, but the magnitude of seasonal fluctuation in population density was significantly greater at the isolated habitat site, indicating that the rate of juvenile settlement and mortality rate in the isolated habitat were greater than those in the continuous habitat. The magnitude of aggregation of fishes, which affects density-dependent biological interactions that modify population density such as competition and predation, was also significantly greater at the isolated habitat site, especially in the juvenile settlement season. Most of the fishes at the isolated habitat site exhibited more generalized patterns of microhabitat selection because of less coral coverage and diversity. The seasonal stability in the species composition of fishes was greater at the continuous habitat site than that at the isolated habitat. Our findings suggest that the relative importance of various ecological factors responsible for regulation of the population density of coral reef fishes (e.g., competition, predation, microhabitat selection and post-settlement movement) in a continuous habitat site and the isolated habitat site are different.     

Density dependence forces divergent population growth rates and alters occupancy patterns of a central place foraging Antarctic seabird

Density‐dependent regulation is an important process in spatio‐temporal population dynamics because it can alter the effects of synchronizing processes operating over large spatial scales. Most frequently, populations are regulated by density dependence when higher density leads to reduced individual fitness and population growth, but inverse density dependence can also occur when small populations are subject to higher extinction risks. We investigate whether density‐dependent regulation influences population growth for the Antarctic breeding Adélie penguin Pygoscelis adeliae. Understanding the prevalence and nature of density dependence for this species is important because it is considered a sentinel species reflecting the impacts of fisheries and environmental change over large spatial scales in the Southern Ocean, but the presence of density dependence could introduce uncertainty in this role. Using data on population growth and indices of resource availability for seven regional Adélie penguin populations located along the East Antarctic coastline, we find compelling evidence that population growth is constrained at some locations by the amount of breeding habitat available to individuals. Locations with low breeding habitat availability had reduced population growth rates, higher overall occupancy rates, and higher occupancy of steeper slopes that are sparsely occupied or avoided at other locations. Our results are consistent with evolutionary models of avian breeding habitat selection where individuals search for high‐quality nest sites to maximize fitness returns and subsequently occupy poorer habitat as population density increases. Alternate explanations invoking competition for food were not supported by the available evidence, but strong conclusions on food‐related density dependence were constrained by the paucity of food availability data over the large spatial scales of this region. Our study highlights the importance of incorporating nonconstant conditions of species–environment relationships into predictive models of species distributions and population dynamics, and provides guidance for improved monitoring of fisheries and climate change impacts in the Southern Ocean.     

The distribution of salmonids in upland streams in relation to depth and gradient

The distribution of various age classes of salmon and trout was assessed in upland streams by electrofishing. Water depths and site gradients were measured and correlated to fish densities. The fry of both species were significantly more abundant in shallow water; up to 75·3% of salmon fry and 72·2% of trout fry were captured in sites of mean depth < 20 cm. Older trout were found mainly in the deeper areas, with a maximum of 7·4% captured in sites < 20 cm mean depth. Yearling fish were found in all the depth-ranges sampled, but with a tendency for higher numbers in mid-range depths. There were similar correlations in the abundance of each age class with the actual areas of shallow, mid-range and deep water habitat available within sites. Correlations of fish density with gradient indicated that trout were limited in their distribution to areas of lower flow, whereas salmon were not. Since depth and gradient were significantly negatively correlated, there was an apparent preference of trout for slightly deeper habitats than the equivalent year classes of salmon. The observed habitat segregation is discussed in terms of competition and selection.