Habitat suitability curves for brown trout (Salmo trutta fario L.) in the River Adda,Northern Italy: comparing univariate and multivariate approaches

Habitat suitability of brown trout (Salmo trutta fario) was studied in the upper portion of the Adda River, Northern Italy. Measurements were made for 528 individuals distributed in two life‐stage classes, adult and juvenile, based on body length. In order to provide basic biological information for the physical habitat simulation (PHABSIM) system of the instream flow incremental methodology (IFIM) in the Italian regulated rivers, habitat suitability curves (HSCs) have been developed with respect to several microhabitat riverine parameters. Initially, current velocity, water depth, substrate class size and cover were analysed with an univariate approach, then bivariate habitat suitability models were developed from depth and velocity data. The comparison of experimental univariate HSCs with those from the literature outlined some differences that can essentially be explained by characteristics of the investigated river, confirming the necessity of using site‐specific curves in relation to each experimental study area. To compare the univariate and bivariate approaches, the weighted usable area (WUA)–discharge relationships were calculated using both types of HSCs. Response curves obtained from the two approaches turned out to be quite different. In PHABSIM habitat modelling, HSCs univariate functions need to be aggregated to produce the WUA–discharge relationship. A multiplicative criterion is generally used for the combined suitability factor; by means of this aggregation criterion all variables have equal weight. According to bivariate models, depth is much more important than velocity in defining habitat suitability requirements. Copyright © 2001 John Wiley & Sons, Ltd.     

Evaluating Methods to Establish Habitat Suitability Criteria: A Case Study in the Upper Delaware River Basin,USA

Defining habitat suitability criteria (HSC) of aquatic biota can be a key component to environmental flow science. HSC can be developed through numerous methods; however, few studies have evaluated the consistency of HSC developed by different methodologies. We directly compared HSC for depth and velocity developed by the Delphi method (expert opinion) and by two primary literature meta‐analyses (literature‐derived range and interquartile range) to assess whether these independent methods produce analogous criteria for multiple species (rainbow trout, brown trout, American shad, and shallow fast guild) and life stages. We further evaluated how these two independently developed HSC affect calculations of habitat availability under three alternative reservoir management scenarios in the upper Delaware River at a mesohabitat (main channel, stream margins, and flood plain), reach, and basin scale. In general, literature‐derived HSC fell within the range of the Delphi HSC, with highest congruence for velocity habitat. Habitat area predicted using the Delphi HSC fell between the habitat area predicted using two literature‐derived HSC, both at the basin and the site scale. Predicted habitat increased in shallow regions (stream margins and flood plain) using literature‐derived HSC while Delphi‐derived HSC predicted increased channel habitat. HSC generally favoured the same reservoir management scenario; however, no favoured reservoir management scenario was the most common outcome when applying the literature range HSC. The differences found in this study lend insight into how different methodologies can shape HSC and their consequences for predicted habitat and water management decisions. Published 2016. This article is a U.S. Government work and is in the public domain in the USA.     

History and review of the habitat suitability criteria curve in applied aquatic ecology

Hydraulic microhabitat assessment is a category of environmental flow tools (e.g., Physical Habitat Simulation system and other methodologically similar software) that, at its core, uses habitat suitability criteria (HSC) to link values of point hydraulic variables (usually depth, velocity, and substrate/cover) to habitat values for target life stages. Although this assessment tool has been used worldwide for decades, the history of the HSC curve is relatively unknown because the foundational information is predominantly contained in obscure and often unpublished reports. We review the history of the HSC concept in applied aquatic ecology to clarify its scientific pedigree, ensure its proper use, and build a foundation for future research. We begin the review with the formative decades of the 1950's through the 1970's, when consumptive‐based western USA water law conflicted with conservation traditions and natural resource management objectives, although water allocation issues date back at least to the 19th century. By analysing the history of the HSC concept, we aim to establish the biological, hydrologic, and geomorphological conditions that must be met for the HSC concept to be successfully employed. In spite of its documented assumptions and limitations, the HSC concept will likely continue to be a useful tool to help address water resources allocation issues in defined hydrologic and geomorphic settings. We conclude that HSC‐based methodologies should be considered as one of several environmental flow approaches involved in sustainable water resources management.     

Using habitat guilds to develop habitat suitability criteria for a warmwater stream fish assemblage

The diversity of fish species found in warmwater stream systems provides a perplexing challenge when selecting species for assessment of instream flow needs from physical habitat analyses. In this paper we examined the feasibility of developing habitat suitability criteria (HSC) for the entire fish community of a warmwater stream using habitat guilds. Each species was placed a priori into a guild structure and habitat data were collected for depth, velocity, Froude number, distance to cover, embeddedness and dominant and subdominant substrate. Correct guild classification was tested with linear discriminant analysis for each species. Correct classification based on habitat‐use data was highest for riffle and pool‐cover guilds, whereas the fast‐generalist and pool‐run classes, the broader niche guilds, were more frequently misclassified. Variables most important for discriminating guilds were Froude number, velocity and depth in that order. Nonparametric tolerance limits were used to develop guild suitability criteria for continuous variables and the Strauss linear index was used for categorical variables. We recommend the use of a wide array of variables to establish more accurate habitat analysis. Additionally, guild HSC can be developed with similar effort to that needed to develop HSC for a small number of individual species. Results indicate that a habitat guild structure can be successfully transferred to another river basin and that habitats for a diverse fish assemblage can be adequately described by a small number of habitat guilds. This approach represents an alternative for incorporating entire fish assemblages into habitat analyses of warmwater stream systems. Copyright © 2010 John Wiley & Sons, Ltd.     

INFLUENCE OF THE MORPHOLOGICAL AND HYDRAULIC CHARACTERISTICS OF MOUNTAIN STREAMS ON FISH HABITAT SUITABILITY CURVES

Microhabitat preferences of adult brown trout (Salmo trutta m. fario) were monitored for the purpose of determining design parameters for river restoration. The habitat preferences were evaluated during the summer period of minimum flows. Since 1995, field measurements have been performed in 52 reaches in 43 mountain and piedmont streams. The relationship between hydraulic characteristics and the values of maximum habitat suitability derived from velocity and depth habitat suitability curves (HSCs) was statistically determined. Trout in natural stream reaches showed a strong degree of dependence on depths, but in regulated streams, they were dependent on velocities. The representative habitat suitability curves for four depth intervals were extrapolated. From these outputs, the optimum depths of a microhabitat for river restoration measures and/or assessment of the influence of water withdrawals can be derived. The influence of geological regions on the shape of HSCs has not been proved; therefore, it is conceivable that after verification, the generalized HSCs may also be valid in other mountain and piedmont regions. Copyright © 2011 John Wiley & Sons, Ltd.     

PREDICTING HABITAT RESPONSE TO FLOW USING GENERALIZED HABITAT MODELS FOR TROUT IN ROCKY MOUNTAIN STREAMS

Dams and water diversions can dramatically alter the hydraulic habitats of stream ecosystems. Predicting how water depth and velocity respond to flow alteration is possible using hydraulic models, such as Physical Habitat Simulation (PHABSIM); however, such models are expensive to implement and typically describe only a short length of stream (10² m). If science is to keep pace with development, then more rapid and cost‐effective models are needed. We developed a generalized habitat model (GHM) for brown and rainbow trout that makes similar predictions to PHABSIM models but offers a demonstrated reduction in survey effort for Colorado Rocky Mountain streams. This model combines the best features of GHMs developed elsewhere, including the options of desktop (no‐survey) or rapid‐survey models. Habitat–flow curves produced by PHABSIM were simplified to just two site‐specific components: (i) Q95h (flow at 95% of maximum habitat) and (ii) Shape. The Shape component describes the habitat–flow curves made dimensionless by dividing flow increments by Q95h and dividing habitat (weighted usable area) increments by maximum habitat. Both components were predicted from desktop variables, including mean annual flow, using linear regression. The rapid‐survey GHM produced better predictions of observed habitat than the desktop GHM (rapid‐survey model explained 82–89% variance for independent validation sites; desktop 68–85%). The predictive success of these GHMs was similar to other published models, but survey effort to achieve that success was substantially reduced. Habitat predicted by the desktop GHM (using geographic information system data) was significantly correlated with the abundance of large brown trout (p < 0.01) but not smaller trout. Copyright © 2013 John Wiley & Sons, Ltd.     

Movement and summer habitat of brown trout (Salmo trutta) below a pulsed discharge hydroelectric generating station

Radiotelemetry was used to investigate detailed movement and summer habitat of brown trout Salmo trutta (size range 157–488 mm TL, n=18) in the Kananaskis River, Alberta. Flows in the Kananaskis River respond to pulsed daily discharge from an upstream hydroelectric generating facility (range 0.15–25 m³ s⁻¹). Wetted area available for brown trout doubled during periods of high flow. Fluctuating river levels did not appear to influence the degree to which brown trout moved within the study site. However, there was evidence that brown trout used cover and pools more as discharge increased. During high flow conditions, brown trout used similar depths (63 cm), and significantly lower surface water velocities than during low flow conditions. Brown trout also moved closer to shore into interstitial spaces among woody debris and root complexes during high flow. Pool habitats were used most often compared with all other habitat types combined. Pools with large woody debris accounted for 75% of all habitat observations. Woody debris was used more often than all other cover types. Results of the study indicate that the effects of river regulation on brown trout appear to have been moderated by woody debris in pools and along river banks, which provided refuge from high water velocities during periods of high flow. Copyright © 1999 John Wiley & Sons, Ltd.     

Mesohabitat use by brown trout (Salmo trutta) in a small groundwater‐dominated stream

Groundwater‐dominated streams have particular flow regimes that commonly support populations of trout. Meso‐ and micro‐habitat surveys were carried out on a reach of the river Tern that drains a Triassic sandstone aquifer in the English West Midlands, to investigate brown trout (Salmo trutta) habitat use with varying flows. Mesohabitats were mapped over a range of summer and autumn flows and coupled with direct underwater observation (snorkelling) of fish locations together with point measurements of velocity and depth. The number of habitat types recorded was low and dominated by glides, runs, and backwaters. Brown trout showed a strong association with glides and runs with adults being more associated with runs and parr with glides. General habitat use curves showed brown trout to favour depths between 0.30 and 0.40 m and velocities below 0.40 m s^?1. A clear preference was shown for sand and gravel bed materials. However, the differentiation of hydraulic habitats was weak and there was no trend in mesohabitats or change in trout use of mesohabitats with discharge. The study raises limitations of the mesohabitat survey approach when linking fish ecology, flow and physical habitat in small streams with low flow variability and low habitat diversity. In these situations, other factors (especially cover features) appear to strongly influence brown trout distribution. Copyright © 2010 John Wiley & Sons, Ltd.     

Effect of aquatic mosses on juvenile fish density and habitat use in the regulated River Suldalslågen,western Norway

Two morphologically distinct moss communities were found in the River Suldalslågen. The liver moss community consists of species which form a dense mat on the bottom, while the river moss (Fontinalis) community forms long tufts. Moss growth has increased since hydropower regulations due to reduced floods and increased winter flows. Increased moss cover affects the bottom structure, as well as intra‐gravel and near‐bottom hydraulics. We studied densities of juvenile Atlantic salmon (Salmo salar) and brown trout (Salmo trutta) by electrofishing and habitat selection by direct underwater observation, in areas with natural moss cover compared with areas where mosses were experimentally removed. Areas with dense mats of liver mosses held lower densities of young of year (YoY) and older salmon parr than areas where liver moss had been removed. No differences in densities of YoY salmon were found between areas with and without Fontinalis. For older salmon, parr results were inconclusive. In some samples more and in others fewer fish were found in areas with Fontinalis moss removed. For trout, densities were higher in areas with Fontinalis, while results for liver moss were inconclusive. No major differences were found with regard to microhabitat selection between areas with and without river moss, suggesting that habitat quality in these areas was similar during summer, except with respect to substrate. Salmon held more exposed positions in areas without liver moss, but this is mainly attributed to different habitat availabilities. It is concluded that the relative increase in liver mosses in the River Suldalslågen has a negative impact on juvenile Atlantic salmon fish density. Copyright © 2002 John Wiley & Sons, Ltd.     

Contrasting temperatures,waterflows, and light: seasonal habitat selection by young Atlantic salmon and brown trout in a boreonemoral river

Habitat use and habitat selection by young Atlantic salmon and brown trout were investigated by direct underwater observation. We sampled during winter and summer water temperatures (low: 3–7°C; high: 9–12°C) coinciding with low and high waterflows (12–20 and 60–80 m³ s^?1), and during day and night in winter, and on six selected stations in the river. Observations of 396 salmon and 120 trout indicated a distinct seasonal pattern in behaviours and habitat selection. Feeding was the dominant behaviour at high water temperatures during summer. In winter, there was a diurnal pattern in behaviour; both species sheltered in interstitial spaces in the substrate during daylight, but during night held positions on or close to the substrate in slower flowing stream areas. Coarse substrate providing cover was therefore an important habitat factor during daylight at low water temperatures, while slow‐flowing water was important during night. Although spatial niche overlap was considerable both in summer and winter, salmon and trout segregated with respect to meso‐ and microhabitat selection, and relatively more at low temperatures. Both species changed their use of mesohabitats towards more slow‐flowing glide/flat habitats in winter. Irrespective of season, trout preferred in general more slow‐flowing water than salmon did, but the difference was more pronounced in winter. Salmon used a wider range of water depths and in particular water velocities, than did trout. Both species were less tolerant of high water velocities at low water temperatures. The seasonal and diurnal pattern in habitat selection reported have important implications for habitat research and habitat‐hydraulic modelling. Copyright © 2001 John Wiley & Sons, Ltd.     

Habitat Suitability Criteria via Parametric Distributions: Estimation,Model Selection and Uncertainty

Previous methods for constructing univariate habitat suitability criteria (HSC) curves have ranged from professional judgement to kernel‐smoothed density functions or combinations thereof. We present a new method of generating HSC curves that applies probability density functions as the mathematical representation of the curves. Compared with previous approaches, benefits of our method include (1) estimation of probability density function parameters directly from raw data, (2) quantitative methods for selecting among several candidate probability density functions, and (3) concise methods for expressing estimation uncertainty in the HSC curves. We demonstrate our method with a thorough example using data collected on the depth of water used by juvenile Chinook salmon (Oncorhynchus tschawytscha) in the Klamath River of northern California and southern Oregon. All R code needed to implement our example is provided in the appendix. Published 2015. This article is a U.S. Government work and is in the public domain in the USA.     

Evaluating Uncertainty in Physical Habitat Modelling in a High‐Gradient Mountain Stream

Predictions of habitat‐based assessment methods that are used to determine instream flow requirements for aquatic biota are uncertain, but instream flow practitioners and managers often ignore those uncertainties. Two commonly recognized uncertainties arise from (i) estimating the way in which physical habitat within a river changes with discharge and (ii) the suitability of certain types of physical habitat for organisms. We explored how these sources of uncertainty affect confidence in the results of the British Columbia Instream Flow Methodology (BCIFM), which is a commonly used transect‐based habitat assessment tool for small‐scale water diversions. We calculated the chance of different magnitudes of habitat loss resulting from water diversion using a high‐gradient reach of the North Alouette River, BC, as a case study. We found that uncertainty in habitat suitability indices for juvenile rainbow trout generally dominated uncertainty in the results of the BCIFM when large (>15) numbers of transects were used. In contrast, with small numbers of transects, variation in physical habitat among sampled transects was the major source of uncertainty in the results of the BCIFM. Presentations of results of the BCIFM in terms of probabilities of different amounts of habitat loss for a given flow can help managers prescribe instream flow requirements based on their risk tolerance for fish habitat loss. Copyright © 2015 John Wiley & Sons, Ltd.     

UNCERTAINTY IN PREDICTING THE FISH‐RESPONSE TO TWO‐DIMENSIONAL HABITAT MODELING USING FIELD DATA

Both water managers and researchers have the same goal when it comes to fish conservation, namely, to sustain, to improve or to restore aquatic habitat. To this aim, two‐dimensional (2D) hydrodynamic models have been widely used in aquatic habitat studies because they simulate flow with high accuracy and can predict habitat dynamics. The River2D model is able to integrate the habitat suitability curves for fish life stages with the simulated depth and velocity fields and the riverbed characteristics of substrate and cover, thereby estimating the corresponding weighted usable area, and thus predicting the potential distribution of fish species in the river. However, little is known about the in situ variability associated with such predictions both for hydraulic and biological data, whereas ecological responses are known to be driven by variability. Moreover, when calculating habitat availability, differences can be found by considering in the weighted usable area formulation substrate or cover or even both. To test the level of predictive accuracy of hydraulic and biological simulations, we modelled the habitat use by two fish species, the Iberian barbel Luciobarbus bocagei and the Iberian straight‐mouth nase Pseudochondrostoma polylepis, according to their requirements for depth, velocity, substrate and cover and then compared measured and simulated hydraulic and biological outcomes using the River2D model. Results indicate that 2D simulation depends on data collection, especially the density and location of bed topography points. Substantial differences were found in the biological responses. Results may differ when choosing different habitat availability variables. Similarly, habitat use may also be influenced by other biotic and abiotic interactions occurring in ecosystems, and restoration planning should be aware of such variability. Copyright © 2012 John Wiley & Sons, Ltd.     

A review of statistical methods for the evaluation of aquatic habitat suitability for instream flow assessment

Habitat models serve three main purposes: First, to predict species occurrences on the basis of abiotic and biotic variables, second to improve the understanding of species‐habitat relationships and third, to quantify habitat requirements. The use of statistical models to predict the likely occurrence or distribution of species based on relevant variables is becoming an increasingly important tool in conservation planning and wildlife management. This article aims to provide an overview of the current status of development and application of statistical methodologies for analysing the species‐environment association, with a clear emphasis on aquatic habitat. It describes the main types of univariate and multivariate techniques available for analysis of species‐environment association, and specifically focuses on the assessment of the strengths and weaknesses of the available statistical methods to estimate habitat suitability. A second objective of this article is to propose new approaches using existing statistical methods. A wide array of habitat statistical models has been developed to analyse habitat‐species relationship. Generally, physical habitat is dependent on more than one variable (e.g. depth, velocity, substrate, cover) and several suitability indices must be combined to define a composite index. Multivariate approaches are more appropriate for the analysis of aquatic habitat as they inherently consider the interrelation and correlation structure of the environmental variables. Ordinary multiple linear regression and logistic regression are popular methods often used for modelling of species and their relationships with environment. Ridge regression and Principal component regression are particularly useful when the independent variables are highly correlated. More recent regression modelling paradigms like generalized linear models (GLMs) present advantages in dealing with non‐normal environmental variables. Generalized additive models (GAMs) and artificial neural networks are better suited for analysis of non‐linear relationships between species distribution and environmental variables. The fuzzy logic approach presents advantages in dealing with uncertainties that often exist in habitat modelling. Appropriate methods for analysis of multi‐species data are also presented. Finally, the few existing comparative studies for predictive modelling are reviewed, and advantages and disadvantages of different methods are discussed. Copyright © 2006 John Wiley & Sons, Ltd.     

THE INFLUENCE OF VARIABLE HABITAT SUITABILITY CRITERIA ON PHABSIM HABITAT INDEX RESULTS

The Physical Habitat Simulation System (PHABSIM) still probably remains as the most widespread habitat method used to establish inflow standards or to link habitat temporal variations with fish population dynamics. However, statistical uncertainties around the PHABSIM main output, the weighted usable area (WUA) over discharge curves, are usually ignored. Here, we assess the uncertainty in WUA curves and derived habitat duration curves induced by the variability around the PHABSIM biological model, the habitat suitability criteria, using brown trout Salmo trutta as the model species. Bootstrap analyses showed that the uncertainty around the WUA curves was rather high when bootstrap sample (BS) size was low and differed among age classes, being generally lower for young‐of‐the‐year (YOY). Width of 95% confidence intervals for maximum WUA magnitude increased with decreasing BS size, ranging from 19.3% for YOY trout at the largest BS size (40 transects, 270 habitat use observations) to 146% for juveniles at the smallest BS size (nine transects, 60 habitat use observations). The uncertainty arose primarily from the construction of the channel index variable. Nevertheless, results showed that the uncertainty in WUA values could be reduced down to acceptable levels by using general functional channel index categories. Likewise, the shape of WUA curves was also highly variable when BS was small. These patterns resulted in habitat duration curves being highly uncertain, much more in their amplitude than in their shape. Uncertainty about the flows corresponding to different habitat exceedance values increased with decreasing probability of exceedance. Width of peak flow confidence intervals ranged from 3.3% for YOY trout at the largest BS size to 226% for adults at the smallest BS size. Yet such levels of uncertainty do not necessarily entail critical errors in the decision‐making process because large variability in flow peak does not necessarily lead to large variability in WUA magnitude.     

Development of habitat suitability criteria for trout in small streams

We observed 2863 trout in the wild to determine habitat utilization in small streams of the Kings River basin in California's Sierra Nevada mountains. The habitat utilization data were used to develop habitat suitability functions that provide input variables to the instream flow incremental methodology (IFIM) habitat assessment model of the U.S. Fish and Wildlife Service. Observations of habitat utilization of rainbow trout (Salmo gairdneri), brown trout (Salmo trutta), and brook trout (Salvelinus fontinalis) were obtained during the summer months of 1983 and 1984. The observations were made in small streams with discharges ranging from 0.7 m³ s^?1 to 0.03 m³ s^?1. The streams are at elevations of 1250 to 2530 m. Equal effort was applied to observing undisturbed trout in all habitat types. Snorkeling proved to be the most effective method of observation. Individual trout of all species and life stages were most often observed in the lower half of the water column, utilizing low-velocity currents of less than 3.0 cm s^?1. From the depth and velocity utilization data, several forms of habitat suitability functions were developed and evaluated:

• 1 Univariate depth and velocity functions derived from frequency histogram analysis.
• 2 Univariate depth and univariate velocity exponential polynomial models.
• 3 Bivariate depth and velocity exponential polynomial models.

Univariate exponential polynomial models provided the best fit to the data for each species, based on visual comparisons of response surfaces and contour plots, and comparisons of computed sums of squared errors. Bivariate models fitted to the data resulted in greater sums of squared errors than multiplicative aggragation of univariate models, and frequently predicted utilization at zero depth. The habitat suitability functions derived from the univariate exponential polynomial models provided the best input to the IFIM habitat assessment models.     

TRANSFERABILITY OF HYDRAULIC PREFERENCE MODELS FOR AQUATIC MACROINVERTEBRATES

Quantifying the transferability across stream reaches and sampling occasions of models relating the density of aquatic taxa to microhabitat hydraulics are important for increasing our understanding of the mechanisms explaining microhabitat selection. We revisited data collected at multiple occasions in two regions, that is, in nine reaches of seven German streams and in four reaches of a large French river, to provide detailed information by taxon on the transferability of hydraulic preferences of aquatic macroinvertebrates. In each region and for each taxon, we tested the statistical significance of a regional preference model relating density to microhabitat hydraulics simultaneously in all surveys (i.e. reach–occasion combinations). The transferability of hydraulic preferences was assessed using cross‐validation procedures within each region. The regional models showed significant associations with microhabitat hydraulics for 56% of the 151 taxa collected in Germany and for 67% of 66 taxa in France. Cross validations indicated that regional models significantly explained density variations within independent surveys for 60% of the cases in Germany and 54% in France. We conclude that both regional and survey‐specific models are useful in specific studies, depending on the taxa considered and the spatial extent and objectives of the application. Graphs of all raw data and models are supplied as supporting information. Copyright © 2012 John Wiley & Sons, Ltd.     

Spatial distribution and geomorphic condition of fish habitat in streams: an analysis using hydraulic modelling and geostatistics

Reach‐scale physical habitat assessment scores are increasingly used to make decisions about management. We characterized the spatial distribution of hydraulic habitat characteristics at the reach and sub‐reach scales for four fish species using detailed two‐dimensional hydraulic models and spatial analysis techniques (semi‐variogram analyses). We next explored whether these hydraulic characteristics were correlated with commonly used reach‐scale geomorphic assessment (RGA) scores, rapid habitat assessment (RHA) scores, or indices of fish biodiversity and abundance. River2D was used to calculate weighted usable areas (WUAs) at median flows, Q₅₀, for six Vermont streams using modelled velocity, depth estimates, channel bed data and habitat suitability curves for blacknose dace (Rhinichthys atratulus), brown trout (Salmo trutta), common shiner (Notropis cornutus) and white sucker (Catostomus commersoni) at both the adult and spawn stages. All stream reaches exhibited different spatial distributions of WUA ranging from uniform distribution of patches of high WUA to irregular distribution of more isolated patches. Streams with discontinuous, distinct patches of high score WUA had lower fish biotic integrity measured with the State of Vermont's Mixed Water Index of Biotic Integrity (MWIBI) than streams with a more uniform distribution of high WUA. In fact, the distribution of usable habitats may be a determining factor for fish communities. A relationship between predicted WUAs averaged at the reach scale and RGA or RHA scores was not found. Future research is needed to identify the appropriate spatial scales to capture the connections between usable patches of stream channel habitat. Copyright © 2008 John Wiley & Sons, Ltd.