ARTICLE

Are brown trout replacing or displacing bull trout populationsin a changing climate?Robert Al-Chokhachy, David Schmetterling, Chris Clancy, Pat Saffel, Ryan Kovach, Leslie Nyce,Brad Liermann, Wade Fredenberg, and Ron Pierce

Abstract: Understanding how climate change may facilitate species turnover is an important step in identifying potentialconservation strategies. We used data from 33 sites in western Montana to quantify climate associations with native bull trout(Salvelinus confluentus) and non-native brown trout (Salmo trutta) abundance and population growth rates (�). We estimated � usingexponential growth state-space models and delineated study sites based on bull trout use for either spawning and rearing (SR) orforaging, migrating, and overwintering (FMO) habitat. Bull trout abundance was negatively associated with mean August streamtemperatures within SR habitat (r = −0.75). Brown trout abundance was generally highest at temperatures between 12 and 14 °C.We found bull trout � were generally stable at sites with mean August temperature below 10 °C but significantly decreasing, rare,or extirpated at 58% of the sites with temperatures exceeding 10 °C. Brown trout � were highest in SR and sites with temperaturesexceeding 12 °C. Declining bull trout � at sites where brown trout were absent suggest brown trout are likely replacing bull troutin a warming climate.

Résumé : Il importe de comprendre comment le climat pourrait faciliter le renouvellement des espèces pour cerner desstratégies de conservation potentielles. Nous avons utilisé des données de 33 sites de l’ouest du Montana pour quantifier lesassociations climatiques avec l’abondance et les taux de croissance de populations (�) d’ombles a tête plate (Salvelinus confluentus)indigènes et de truites brunes (Salmo trutta) non indigènes. Nous avons estimé � en utilisant des modèles d’espaces d’états decroissance exponentielle et délimité les sites étudiés selon leur utilisation par l’omble a tête plate soit comme habitat de frai etd’alevinage (SR) ou d’approvisionnement, de migration et d’hivernage (FMO). L’abondance des ombles a tête plate était néga-tivement associée aux températures moyennes des cours d’eau en août dans les habitats SR (r = −0,75). L’abondance de la truitebrune était généralement maximum a des températures entre 12 et 14 °C. Nous avons constaté que les � des ombles a tête plateétaient généralement stables aux sites présentant une température moyenne en août inférieure a 10 °C, mais qu’il diminuaitsignificativement, l’espèce y étant rare ou disparue, dans 58 % des sites où cette température dépasse 10 °C. Les � des truitesbrunes étaient maximums dans les habitats SR et les sites caractérisés par des températures supérieures a 12 °C. Des � en baissedes ombles a tête plate dans des sites exempts de truites brunes donnent a penser que ces dernières remplacent probablementles ombles a tête plate dans un climat en réchauffement. [Traduit par la Rédaction]

IntroductionClimate change is likely to substantially alter stream ecosystems

with pronounced effects for cold-water fishes such as salmonids(Jonsson and Jonsson 2009; Williams et al. 2009; Elliott and Elliott2010). Salmonid life histories, vital rates, and demographics arestrongly tied to factors influenced by climate, including thermaland hydrologic regimes (Elliott 1994; Lobon-Cervia 2004; Crozieret al. 2008; Warren et al. 2012). With different thermal tolerancesand life-history expressions, the effects of changing climatic con-ditions are likely to differ across species. Understanding how cli-mate, among other factors, may be influencing populations iscritical for identifying the potential for effective managementand restoration scenarios.

Non-native salmonids are an additional concern for the conser-vation of native salmonids across North America (Dunham et al.2002). Widespread introductions for recreation have resulted innaturally producing populations of non-native salmonids in manystreams. The mechanistic threats of non-native species to native

salmonids, particularly in the context of climate change, are oftennot well understood (e.g., Rahel and Olden 2008; Lawrence et al.2014). This uncertainty stems partly from the paucity of situationswhere changing climatic conditions have been empirically linkedwith salmonid population and demographic data (Kovach et al.2016). Likewise, delineating between non-native displacement(i.e., declines in native salmonids due to negative interactions withnon-natives) or replacement (i.e., declines in native salmonids dueto factors unrelated to non-natives) is an inherent challenge inspecies turnover studies (Dunham et al. 2002).

Refining our understanding of the influences of climate changeand non-native species on extant populations of bull trout (Salvelinusconfluentus) is essential in designing conservation strategies to en-hance long-term persistence. Bull trout are currently listed as “Threat-ened” in the United States under the Endangered Species Act andranked “Of Special Concern” or “Threatened” for three of four geo-graphic populations by the Committee on the Status of EndangeredWildlife in Canada. Bull trout are extremely temperature-sensitive

Received 10 June 2015. Accepted 31 December 2015.

R. Al-Chokhachy. US Geological Survey, Northern Rocky Mountain Science Center, 2327 University Way, Suite 2, Bozeman, MT 59715, USA.D. Schmetterling, P. Saffel, B. Liermann, and R. Pierce. Montana Fish, Wildlife and Parks, 3201 Spurgin Road, Missoula, MO 59804, USA.C. Clancy and L. Nyce. Montana Fish, Wildlife and Parks, 1801 North 1st Street, Hamilton, MO 59840, USA.R. Kovach. US Geological Survey, Northern Rocky Mountain Science Center, Glacier Field Station, West Glacier, MT 59936, USA.W. Fredenberg. US Fish and Wildlife Service, Creston Fish & Wildlife Center, Kalispell, MT 59901, USA.Corresponding author: Robert Al-Chokhachy (email: ral-chokhachy@usgs.gov).Copyright remains with the author(s) or their institution(s). Permission for reuse (free in most cases) can be obtained from RightsLink.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

1

Can. J. Fish. Aquat. Sci. 73: 1–10 (2016) dx.doi.org/10.1139/cjfas-2015-0293 Published at www.nrcresearchpress.com/cjfas on 23 February 2016.

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

relative to other salmonids (Selong et al. 2001), and distributionsduring summer months are strongly tied to ambient stream temper-atures (Dunham et al. 2003; Wenger et al. 2011). However, bull troutexhibit complex life histories, and adults and subadults can season-ally demonstrate large upstream and downstream movements toaccess foraging and overwintering habitat (Swanberg 1997; Howellet al. 2010; Starcevich et al. 2012).

Brook trout (Salvelinus fontinalis) and lake trout (Salvelinus namaycush)have been identified as considerable threats to bull trout populationsthrough predation (Martinez et al. 2009; Hansen et al. 2010;Fredenberg 2014), competition (Guy et al. 2011; Warnock andRasmussen 2014), and nonintrogressive hybridization resulting ingametic wastage (Leary et al. 1993; DeHaan et al. 2010). Recently,however, there has been growing concern regarding the effects ofintroduced brown trout (Salmo trutta) on bull trout populations inwestern Montana, USA, particularly in light of recent evidence sug-gesting expansion of brown trout into historic bull trout habitat(USFWS 2015). Similar to bull trout, brown trout are a fall-spawningspecies and considered a top-level predator within streams; however,bull trout have substantially lower thermal tolerances than browntrout (Elliott 2009) and, as a result, often occupy headwater streamsupstream of the distribution of brown trout.

Recent observations of expansion in brown trout abundanceand distribution (P. Saffel, personal observation, 2011) elevatedconcerns about the effects of brown trout and, more specifically,whether brown trout are displacing or replacing bull trout. Here,we used long-term monitoring data from locations in the Colum-bia River headwaters of western Montana to evaluate (i) popula-tion status and trends of bull trout and brown trout; (ii) how

changing climatic conditions are associated with population trendsand abundance; and (iii) the putative threat brown trout may repre-sent to extant bull trout populations.

Materials and methods

Study areaOur study area included 33 long-term sampling sites in the Clark

Fork, Bitterroot, and Blackfoot river drainages in western Mon-tana (Fig. 1). Lands within the study area include a mixture ofmostly private lands in the valley bottoms and river corridors andstate and federal ownership in the tributaries and headwaters.Native vegetation within riparian zones varies elevationally acrosssites, but reflects montane forests commonly found in the region.Riparian communities commonly includes a mixture of grasses andsedges (Carex spp.), shrub-dominated taxa (e.g., Salix spp.), mixedconifer (Pinaceae; e.g., Abies spp.), and poplars (Populus spp.). Cli-mate within the study area is characterized by relatively cold, wetwinter and spring months and relatively warm, dry summers.Streamflows are typical for snowmelt-influenced streams in thenorthern Rocky Mountains, with high spring flow events duringMay and June and declining flow throughout the summer andearly winter.

The sample sites occurred on first- to fourth-order streams (Strahler1952) ranging in elevation from 712 to 1893 m. In addition to bulltrout and brown trout, other species commonly found in samplesurveys included native westslope cutthroat trout (Oncorhynchusclarkii lewisi), mountain whitefish (Prosopium williamsoni), sculpin

Fig. 1. Major rivers, locations of the long-term monitoring sites (grey circles with numbers distinguishing sample sites; see Table 1), andstream gauges (triangles) in western Montana, USA. Inset denotes location of the study area relative to the western portion of the USA.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

2 Can. J. Fish. Aquat. Sci. Vol. 73, 2016

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

(Cottus spp.), longnose dace (Rhinichthys cataractae), and non-nativerainbow trout (Oncorhynchus mykiss) and brook trout.

Sampling dataOur dataset included relatively long-term monitoring sites sam-

pled by Montana Fish, Wildlife and Parks and US Forest Servicebiologists. We constrained our analyses to all sites with at least5 years of data within watersheds currently occupied by browntrout (fifth code hydrologic unit codes; www.nhd.usgs.gov/) andwithin the historical range of bull trout. Sampling methods variedacross sites and included a mixture of backpack electrofishing,bank electrofishing, and boat electrofishing. Despite differencesin approaches across sites, however, the sampling methods weregenerally consistent through time at any given sampling site.

Survey dates varied from early April through mid-September(see below) and the length of stream sampled varied according tostream size (see online supplemental material, Table S11). We in-cluded only counts of individuals collected from the first-passelectrofishing survey (hereinafter relative abundance), given thedifficulties of estimating abundance for species at low densities(Al-Chokhachy et al. 2009). While single-pass data are known tounderestimate trout abundance, correlations between single andmultiple passes are typically high (Bateman et al. 2005), suggest-ing such difference are unlikely to drastically alter our results.

Stream temperature dataYear-specific stream temperature data were lacking for all sites

and all years. As such, we relied on stream temperature predic-tions from a spatially explicit stream temperature model. TheNorWeST stream temperature model incorporates spatial statisti-cal models to provide estimates of August mean temperature at a1 km scale (Isaak et al. 2010). We initially evaluated how well theyear-specific stream temperature predictions from NorWeST matchedobservations from those locations where long-term empiricaltemperature data were available. We found a wide range of corre-lations between empirical and modeled annual data (r = 0.75;range = 0.47–0.95). Given this variability across sites and the factthat interannual NorWeST predictions were relatively inaccurateat several sites, we focused only on mean August temperaturepredictions (1993–2011) from the NorWeST model in our analyses.Mean NorWeST predictions across years were highly correlatedwith mean August temperature estimated from empirical data(r = 0.98).

AnalysesWe delineated the sites in our analyses according to designa-

tions of bull trout occupancy used by the US Fish and WildlifeService (USFWS) in the Bull Trout Recovery Plan (Table 1; USFWS2014), thereby aligning our results with ongoing management

1Supplementary data are available with the article through the journal Web site at http://nrcresearchpress.com/doi/suppl/10.1139/cjfas-2015-0293.

Table 1. Site numbers (corresponding with location in Fig. 1), stream, latitude, longitude, classification of habitat typefor bull trout (Recovery classification), range of years in analyses, the number of years of data (n), and the mean Augusttemperature predicted from the NorWeST stream temperature model.

Site Stream Latitude (°N) Longitude (°W)Recoveryclassification Year range n

Mean Augusttemp. (°C)

1 Bertie Lord Creek 45.90909 113.783 SRa 1990–2012 12 11.12 Bitterroot River 46.46943 114.116 FMO 1982–2009 9 14.93 Bitterroot River 46.83103 114.054 FMO 1989–2009 10 17.44 Bitterroot River 46.01525 114.165 FMO 2005–2013 8 18.15 Blackfoot River 46.91069 113.691 FMO 1988–2012 14 15.76 Blackfoot River 47.02225 113.308 FMO 1985–2013 16 16.47 Camp Creek 45.81539 113.954 SRa 2005–2012 11 13.08 Clark Fork River 47.02166 114.402 FMO 1980–2012 14 16.59 Clark Fork River 46.86888 113.935 FMO 1980–2013 17 18.010 Clark Fork River 46.83817 113.847 FMO 1988–2013 11 20.411 Daly Creek 46.17873 113.900 SR 1989–2012 13 9.412 EF Bitterroot River 45.90939 113.709 FMO 1992–2011 9 11.113 EF Bitterroot River 45.91686 114.103 FMO 1995–2013 13 14.714 EF Bitterroot River 45.86007 114.026 FMO 1998–2012 10 15.215 EF Bull River 48.11330 115.775 SR 2001–2008 8 11.316 EF Bull River 48.12529 115.723 SR 2001–2007 7 12.117 Fishtrap Creek 47.76384 115.075 SR 1999–2011 9 9.718 Gold Creek 46.94177 113.669 SR 1996–2011 11 12.219 Martin Creek 45.93969 113.731 SR 1992–2013 10 11.220 Meadow Creek 45.84822 113.821 SR 1989–2010 7 8.721 Meadow Creek 45.82944 113.802 SR 1989–2011 15 9.222 Moose Creek 45.93667 113.717 SR 1991–2011 11 10.723 NF Blackfoot River 46.97979 113.100 SR 1989–2001 7 11.924 Rock Creek 46.57256 113.686 FMO 1980–2013 15 14.325 Rock Creek 46.42300 113.721 FMO 1980–2013 15 14.526 Rye Creek 45.99271 114.068 SRa 1990–2011 13 11.927 Skalkaho Creek 46.16379 113.899 SR 1989–2013 25 9.928 Sleeping Child Creek 46.10994 114.004 SR 1989–2013 25 12.829 Thompson River 47.73717 115.022 SRa 1985–2012 13 14.630 Thompson River 47.66886 115.108 SRa 1984–2012 17 12.231 Warm Springs Creek 45.82392 114.064 SR 1992–2009 10 11.532 WF Thompson River 47.66100 115.193 SR 2001–2010 7 8.633 WF Thompson River 47.70250 115.207 SR 1999–2010 9 9.3

Note: SR, spawning and rearing; FMO, foraging, migrating, and overwintering.aSites not originally characterized as SR habitat by the USFWS but subsequently designated owing to known bull trout use.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Al-Chokhachy et al. 3

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

strategies. Sites were delineated as either spawning and rearing (SR)or foraging, migrating, and overwintering (FMO) habitat based onbull trout life history and the habitats corresponding to specificlife stages. For migratory bull trout, the SR habitat largely consistsof headwater tributaries, with the FMO habitat occurring down-stream in mainstem (fourth-order or larger) rivers and lakes. Forresident bull trout, which typically occupy second- and third-order streams, there is no clear delineation in SR and FMO habitat.Not all streams within the historical range of bull trout weredelineated, and five of our sites occurred on streams undesignatedin the Bull Trout Recovery Plan. Given that these five streams allrepresent tributary habitat and (or) use by bull trout for spawningand rearing, we aggregated these with the SR sites. Sampling datesgenerally occurred during April–June or September for sites inFMO habitat and July and August for sites in SR habitat. By delin-eating our results for FMO (n = 13) and SR (n = 20), this approachallowed us to gain insight into population trends for differentlife-history stages (e.g., migratory components).

We used counts of bull trout and brown trout at sites for mea-sures of relative abundance and scaled this to linear stream length(fish·m−1) as survey length differed across sites. We investigatedpotential autocorrelation among sites using pairwise, interannualcorrelations in abundance for sites within each river basin (Bitter-root, Blackfoot, upper Clark Fork, lower Clark Fork). Given re-gional shifts in climate since 2000 (e.g., supplemental Fig. S11;Pederson et al. 2010; Al-Chokhachy et al. 2013), we assessed differ-ences in relative abundance (pre-2000 and 2000 and later) using aMann–Whitney rank sum test.

We estimated bull trout and brown trout population growthrate (�) using exponential growth state space (EGSS) models withrestricted maximum likelihood (Humbert et al. 2009). The EGSSmodel is a flexible approach to estimate population trends fromabundance data that can span short time periods (i.e., 10 years)and have unequal time steps, which commonly occur in popula-tion monitoring data.

The EGSS model provides point estimates of population growthrate that are highly concordant with other approaches for esti-mating � in an exponential growth model (diffusion approxima-tion or linear regression). However, the confidence intervals aregenerally more accurate compared with simple, exponential trendanalyses (Humbert et al. 2009; Meyer et al. 2014), as this approachalso accounts for observation error and process variation. TheEGSS model assumes that process and observation errors are nor-mally distributed and that process noise is independent of obser-vation error, and the model more naturally accommodates intrinsicautocorrelation present in time-series data (Kery and Schaub 2012).We considered the exponential growth model appropriate for ourdata given the relatively low abundances of bull trout, the factthat most observations of brown trout expansions were relativelyrecent, and the difficulty in understanding the appropriate density-dependent structure across sites.

Our analyses included datasets with unequal sample sizes andrange of years with sampling data, as the number of sample yearsvaried across sites (x = 12.2; range = 7–25 years). However, we foundlittle evidence of correlations between bull trout or brown troutpopulation trends with range of years of sample data at sites(r < 0.40). Abundances of bull trout and brown trout were low, atleast during some time periods, at many sites, and relative abun-dance values of zero were not uncommon. The EGSS estimates areperformed on a logarithmic scale where values of zero are unde-fined; thus, we added a value of one to all observations. We usedan alpha value of 0.10 for all analyses given the conservationimplications of type II error and considered trends significantwhere confidence intervals did not include one (i.e., � = 1, stablepopulation trend).

Results

Relative abundanceWe observed no clear patterns in correlations in bull trout

(r = −0.02) or brown trout (r = 0.16) relative abundance across allsites. In FMO sites, relative abundance of bull trout (mean =0.004 fish·m−1; SD = 0.009 fish·m−1) was substantially lower thanbrown trout (mean = 0.023 fish·m−1; SD = 0.031 fish·m−1), andbrown trout relative abundance exceeded bull trout at all FMOsites except one (Site 12; Fig. 1). Bull trout and brown trout relativeabundance were weakly correlated at sites within bull trout SRhabitat (r = −0.34) and across all sites (r = −0.27).

There were considerable changes in bull trout and brown troutabundance through time. Brown trout relative abundance wassignificantly higher than bull trout at FMO sites during the early(pre-2000; P = 0.05) and late (2000 and later; P < 0.001) sampleyears, and this difference increased between the sample periods(supplemental Fig. S11). In contrast, bull trout relative abundance(x = 0.058 fish·m−1; SD = 0.071 fish·m−1) was two times higher thanbrowntrout relativeabundance (x =0.026fish·m−1; SD = 0.051 fish·m−1)in SR sites; bull trout relative abundance exceeded brown trout at14 of the 20 SR sites (brown trout > bull trout at Sites 15, 18, 23, 29,and 30; Fig. 1). Bull trout relative abundance exceeded brown troutduring the early (P < 0.001) and late periods (P = 0.022), but thedifference decreased between the sample periods.

Mean bull trout relative abundance was negatively associatedwith August stream temperatures when considering all sites (r =−0.65; Fig. 2) and sites within SR habitat alone (r = −0.75; Fig. 2). Onthe contrary, brown trout relative abundance demonstrated noapparent correlation with August temperatures at all sites (r =0.10) and only a weak positive correlation at sites within bull troutSR habitat (r = 0.45). Brown trout relative abundance was generallylowest at temperatures below 12 °C, highest between 12 and 14 °C,and generally decreased above 15 °C (Figs. 2 and 3). Despite thesepatterns, brown trout were frequently detected at sites with tem-peratures as low as 8.5 °C.

Population trendsFor extant bull trout populations, we found only one site with

increasing population trends (Site 6; Fig. 3a), with all other sitesdemonstrating stable or decreasing trends. In FMO sites, bulltrout populations have been extirpated or occur at extremely lowabundance levels (i.e., trends inestimable, median relative abun-dance = 0) at 31% of the sites and demonstrated significant decreas-ing trends at 23% of the sites and stable population trends at 38%

Fig. 2. A per survey plot of the mean number per metre of bulltrout (open) and brown trout (grey) from single-pass electrofishingsurveys and mean August stream temperature at long-termmonitoring sites within bull trout spawning and rearing (SR; circles)and foraging, migrating, and overwintering (FMO; squares) habitatin western Montana, USA.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

4 Can. J. Fish. Aquat. Sci. Vol. 73, 2016

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

of the sites. No bull trout populations exhibited increasing popu-lation trends in SR sites, with one currently extirpated, and 38% ofthe sites indicating significant decreasing trends.

Brown trout illustrated considerably different patterns in ouranalyses, with only one site indicating significant decreasing trends(Fig. 3b). Brown trout population trends were stable or signifi-cantly increasing at all other sites. Where estimates were possible,brown trout population trends were significantly increasing at31% of the FMO sites and 50% of the SR sites.

Brown trout and bull trout population trends were stronglyassociated with August stream temperatures. Similar to observedpatterns for relative abundance, brown trout population trendswere stable at locations with temperatures between 11 and 12 °C,increasing between 12 and 15 °C, and stable at temperatures above15 °C (Fig. 4). In SR habitat, we found bull trout population trendsto be generally stable below 10 °C but significantly decreasing,rare (i.e., trend not estimable owing to extremely low abundance),or extirpated at 58% of the sites with temperatures exceeding 10 °C(Fig. 5a). Bull trout population trends were significantly decreas-ing at multiple sites where brown trout have not yet colonized.

Brown trout population trends were highest in SR and sites withtemperatures exceeding 12 °C (Fig. 5b).

DiscussionTemperature is a critical component controlling the physiol-

ogy, fitness, and life-history expressions of salmonids (Fry 1971;Elliott 1976; Jonsson and Jonsson 2009; Warren et al. 2012). Chang-ing thermal regimes can alter the distributions and, thus, inter-actions among species (McMahon et al. 2007; Finstad et al. 2011).With substantially colder thermal tolerances than most salmonids,bull trout are likely to be increasingly stressed under expectedchanges in climate (Rieman et al. 2007; Wenger et al. 2011). Under-standing the effects of sympatric, non-native species is an impor-tant step in developing robust climate adaptation measures toenhance the persistence of native salmonids such as bull trout(Peterson et al. 2013).

Stream temperature, abundance, and trendsIn our study, linkages between summer stream temperatures

and bull trout and brown trout abundance were apparent, a pattern

Fig. 3. Estimates of population growth rate (�) for bull trout (open; panel a) and brown trout (grey; panel b) at sites (see Table 1 for list of sitenumbers) within bull trout foraging, migrating, and overwintering (FMO; squares) and spawning and rearing (SR; circles) habitat in westernMontana, USA. The horizontal line is a reference indicating stable population trend, the × symbols indicate sites where bull trout have beenextirpated, triangles indicate sites where brown trout have not yet colonized, and diamonds indicate where bull trout or brown trout areinfrequent.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Al-Chokhachy et al. 5

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

consistent with recent distributional studies (Isaak et al. 2015). Wefound the highest brown trout abundances and generally increas-ing trends at sites where temperatures were within the optimumrange for brown trout from lab and field studies (Hari et al. 2006;Elliott 2009). Brown trout were consistently observed where meanAugust temperatures were above 11 °C up to 20 °C (the highesttemperatures at our sites). While the general absence of browntrout at temperatures below 11 °C suggests thermal regimes maybe limiting expansion (sensu Isaak et al. 2015), distributions withintheir native range can occur at thermal regimes as low as 3 °C(Hari et al. 2006), suggesting that colonization of these colder sitesis possible. The limited brown trout expansion at colder sites mayultimately be driven by temperature-mediated competitive ad-vantages for bull trout (sensu Taniguchi and Nakano 2000), amechanism warranting further research.

Indications of bull trout at or near extirpation (i.e., occurringat extremely low relative abundance) at some sites and decliningtrends in bull trout abundance within FMO and SR habitat high-light concerns for the long-term persistence of bull trout withinthis region. Bull trout abundance was highest and populationswere stable where mean August temperatures were below 10 °C(Fig. 5), again indicating corroboration of results with distribution-only studies (Isaak et al. 2015). There was not a negative relation-ship between population growth and relative abundance for bulltrout, indicating stability or declining growth was not density-dependent (Fig. 6). However, many sites demonstrated relativelylow abundance (e.g., ≤0.05 bull trout·m−1) and could be suscepti-ble to depensatory growth (i.e., Allee effects with a positive rela-tion between population growth and abundance). Persistence islikely tenuous for these low abundance populations with decreas-ing or even stable population trends (Stacey and Taper 1992;Rieman and Allendorf 2001). While trend analyses indicated con-siderable variability in associations with temperature, the rela-tively low bull trout abundance at warmer sites (Figs. 2 and 6) andinherent difficulties of detecting declines in such small popula-tions (Taylor and Gerrodette 1993) suggests extirpation may beinevitable without dramatic intervention (sensu Tilman et al.1994).

Displacement or replacement?Whereas we expect the distribution of fishes to shift in accor-

dance with changing climatic conditions (Almodovar et al. 2012;Comte et al. 2013), it is often unclear whether such changes are

Fig. 4. Estimates of brown trout population growth rates from allsample sites in western Montana, USA (circles; 90% CI), locationswhere brown trout have been detected infrequently (allowing for noestimate of population growth rate; diamonds), and where browntrout have never been detected (triangles) against mean Auguststream temperatures. The horizontal line at � = 1.0 indicates a stablepopulation growth rate.

Fig. 5. Estimates of bull trout (a) and brown trout (b) populationgrowth rates (�) at sample sites within bull trout spawning andrearing (SR) habitat with different mean August streamtemperatures in western Montana, USA. The horizontal line is areference indicating stable population growth rate, the × symbolsindicate sites where bull trout have been extirpated, trianglesindicate sites where brown trout have not yet colonized, anddiamonds indicate where either bull trout or brown trout areinfrequent.

Fig. 6. The relationships between bull trout population growthrates (�) and relative abundance (mean number of bull trout) atbull trout spawning and rearing (SR) sites with August streamtemperatures <10 °C (open circles), SR sites with August streamtemperatures >10 °C (black circles), and bull trout foraging,migrating, and overwintering (FMO) sites (grey squares) in westernMontana, USA.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

6 Can. J. Fish. Aquat. Sci. Vol. 73, 2016

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

due to changes in climate-related attributes (e.g., temperature) orexacerbated by interspecific interactions. Expansion of brown troutappears to be in part driven by regional air temperature and en-suing stream temperature warming during the 21st century(Pederson et al. 2010; Isaak et al. 2012; Fig. S21). Conversely, de-clines in bull trout appear to be at least partly thermally driven, asevidenced by multiple populations demonstrating significant de-clines in the absence of brown trout. However, we do acknowl-edge that our datasets were not randomly chosen but were thebest data available in our region and thus restrict our inference tothe sites included herein. Further analyses using larger datasetsare needed to allow for formal statistical testing of factors associ-ated with such patterns. In lieu of these shortcomings, the appar-ent transition from initial bull trout declines followed byincreases in brown trout abundance, rather than concurrenttrends, provides considerable support for species replacement(Dunham et al. 2002).

Further contraction of thermally suitable habitat for bull troutis anticipated in the future for western Montana (Jones et al. 2014;Isaak et al. 2015; USFWS 2015). Despite indications of bull troutdeclines in allopatry as a result of potential thermal constraints,there is uncertainty in how the presence of brown trout and (or)increasing trends of brown trout may exacerbate the effects orwarming climate where populations occur in sympatry. Withinstream ecosystems, both bull trout and brown trout demonstrateincreasing piscivory with size and are commonly top-level preda-tors (Elliott and Hurley 2000; Lowery 2009; Budy et al. 2013), sug-gesting potential competition for space and forage resources.Other mechanisms for negative interactions include redd super-imposition (e.g., Essington et al. 1998; Storaasli and Moran 2015),as both species spawn during the fall–winter, with brown troutcommonly spawning after the cessation of bull trout spawning(Wood and Budy 2009). Finally, increasing numbers of brown troutmay exacerbate declining bull trout populations through preda-tion, particularly at early life stages, thus limiting recruitment(Budy et al. 2013). An interplay of interactions may also occur(Persson et al. 2013), making it challenging to specifically identifythe strength of bull trout – brown trout interactions across differ-ent life stages without targeted research.

Beyond thermal relationshipsWe acknowledge that our analyses have largely focused on ther-

mal relationships, yet additional factors may have contributed tothe observed population trends. In particular, hydrologic regimescan also act to regulate populations of fall-spawning salmonidssuch as bull trout and brown trout (Tonina et al. 2008). Includingdischarge in our analyses was not possible, as site-specific esti-mates of discharge were not available. A post hoc analyses ofexisting gauging data (www.usgs.gov) suggests the number ofhigh discharge events during November through March, the pe-riod when both species eggs are typically in the gravel, decreasedby nearly 75% during the 2000s (Fig. S3a1). The apparent decreasein high discharge events during the winter is likely driven byreduced snowpack at lower elevations (Mote 2003), such that win-ter warming events do not result in rain-on-snow freshets. De-creases in the number of high discharge events likely reducedscouring of redds and contributed to the increasing populationgrowth rates of brown trout (Cattaneo et al. 2002; Daufresne andRenault 2006; Unfer et al. 2011). Brown trout populations can alsobe limited by high discharge during periods of emergence, thusreducing young-of-year survival (Lobon-Cervia 2014). However, wefound little evidence of changes in the frequency of high dis-charge events during the period of likely emergence (Fig. S3b1).

Additionally, changes in brown trout abundance may havebeen influenced by indirect effects of the exotic whirling disease(Myxobolus cerebralis). Whirling disease was detected in the mid-1990s and subsequently spread throughout the lower elevationsportions of our study area (Baldwin et al. 1998; Granath et al.

2007). Brown trout are considered to be more resilient to whirlingdisease than other salmonids (Baldwin et al. 2000; Pierce et al.2014). The higher prevalence of whirling disease at lower-elevation,warmer streams (Baldwin et al. 2000) may have contributed toenhancing some source populations of brown trout through re-ductions in sympatric species abundance (e.g., rainbow trout) andthe subsequent increase in brown trout abundance in FMO habi-tat leading to the spread of brown trout to bull trout SR streams.However, considerable differences in brown trout populationtrends between adjacent tributaries (e.g., Bitterroot River) suggestthe importance of local climatic conditions at sites.

Declining bull trout populationsWhile stream temperature and discharge are likely driving the

observed patterns of bull trout population trends, a host of otherfactors may have contributed to our results. For example, migra-tory bull trout within the lower Bitterroot River and lower ClarkFork River have been largely extirpated or severely reduced forover three decades. Large, migratory fish have considerably higherfecundity than smaller resident females (Al-Chokhachy and Budy2008), which may help offset competitive interactions with non-natives such as brown trout. Large pulses of recruitment frommigratory bull trout spawning may also allow populations to re-cover from pulsed disturbances (e.g., Bohlin et al. 2001), as resident-only populations may lack resiliency over a period of multiplegenerations.

In addition, bull trout throughout our study area have beendirectly or indirectly affected by thermal constraints, dewatering,agricultural use, development, and factors associated with increas-ing human population growth (e.g., increased angling; Al-Chokhachyet al. 2008). While each factor in isolation may not drive bull troutpopulation trends, the synergistic effects of multiple stressorsalong with changing climatic conditions may ultimately result ina tipping point for bull trout (sensu Nelson et al. 2009). Continuedmonitoring within higher-resolution climate data (e.g., annualtemperature data) may allow for analyses to disentangle sucheffects.

Our results indicating relatively low abundance and stable ordeclining population trends at a large portion of our sites arecontrary to recent findings for populations of bull trout in Idaho,USA, that show stable or increasing trends (Meyer et al. 2014).Aside from different sampling methodologies (e.g., redd counts,snorkeling, etc.), the observed differences between populationtrends in Montana and Idaho require further study but may stemfrom the scale of analyses. We specifically delineated sites byhabitat designations under the US Fish and Wildlife Service’s BullTrout Recovery Plan and considered trends at the local populationlevel (i.e., panmictic breeding group; Whitesel et al. 2004), whileMeyer et al. (2014) grouped local populations at a larger “core”population level (Whitesel et al. 2004). Given the scale at whichconnectivity between local populations occurs varies across prox-imate landscapes (Whiteley et al. 2006), we chose to evaluate rel-ative abundance and trends at a finer resolution. Analyses at thelocal population level also allowed for inferences across differenthabitat types and life-history expressions, a broader understand-ing of differences in trends within larger populations (sensuHanski 1999), and, more specifically, linkages with site-specificclimatic attributes. Indeed, a comprehensive analysis consideringhow inference changes across scales is warranted.

Management opportunitiesIdentifying effective management strategies to enhance persis-

tence of existing bull trout populations is expected to becomeincreasingly valuable (Falke et al. 2015). Concomitantly, browntrout represent an important sport fishery in Montana and occupya niche as a large piscivore that was formerly occupied by bulltrout — a species that is increasingly scarce and one that currentlyoffers limited angling opportunities. With little empirical infor-

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Al-Chokhachy et al. 7

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

mation describing bull trout – brown trout interactions, uncer-tainty exists in the effectiveness of controlling brown troutpopulations where sympatric with declining bull trout popula-tions.

Our results clearly suggest streams warming to levels optimalfor brown trout (12 to 15 °C) are the most vulnerable to futureinvasions and major increases in brown trout abundance. Ob-served declines in allopatric bull trout populations suggest remov-ing brown trout where sympatric may not recover populations ofbull trout to historical abundance. However, reducing or remov-ing brown trout where sympatric with bull trout may facilitatebull trout persistence under changing climatic conditions (e.g.,Peterson et al. 2008). An important next step may be to specificallyevaluate bull trout – brown trout interactions at streams ap-proaching 12 °C (e.g., Sites 1, 12, 15, and 19; Table 1), while concom-itantly quantifying the effectiveness of targeted brown troutremoval efforts on the persistence of extant bull trout popula-tions.

Despite being increasingly used in salmonid conservation tocombat non-native species, we urge caution in the use of barriersto control brown trout expansion, given the vulnerability of iso-lated populations and likely elimination of migratory bull trout(Falke et al. 2015). Even infrequent returns of large, fecund adultscan substantially enhance persistence (Morita and Yamamoto 2002;Al-Chokhachy et al. 2015), suggesting the importance of includingselective passage where barriers are implemented. Additionally,isolation will exacerbate the loss of genetic diversity within bulltrout populations, thereby reducing adaptive potential and in-creasing inbreeding load, a problematic scenario for a fish alreadystressed by other human-induced and climatic stressors (Kovachet al. 2015). Certainly more information is needed to understandthe benefits of barrier installations, particularly across a gradientof local conditions (e.g., stream productivity). Prioritizing loca-tions more resilient to climate change may also be warranted(Palmer et al. 2009; Peterson et al. 2013).

In addition to species management considerations, there is needto consider options to enhance the climate resilience of streamssupporting bull trout populations. Enhancing riparian conditionsto allow for increased shading and water storage (Caissie 2006) inaddition to novel approaches to increase groundwater contribu-tions (Pollock et al. 2014) should be evaluated as options for miti-gating warming air temperatures. Preventing or prolonging streamtemperature increases may ultimately allow for bull trout to be-haviorally or physiologically adapt to changing climatic condi-tions (e.g., Alvarez et al. 2006; but see Jonsson and Jonsson 2009).

AcknowledgementsFunding for this project was provided by the US Geological

Survey – USFWS Quick Response Program and Montana Fish Wild-life and Parks. We thank Z. Shattuck (Montana Fish, Wildlife andParks) for reviewing and commenting on an earlier draft of thismanuscript and D. Isaak (US Forest Service) for assistance withNorWeST temperature predictions. Any use of trade, product, orfirm names is for descriptive purposes only and does not implyendorsement by the US Government.

ReferencesAl-Chokhachy, R., and Budy, P. 2008. Demographic characteristics, population

structure, and vital rates of a fluvial population of bull trout in Oregon.Trans. Am. Fish. Soc. 137(6): 1709–1722. doi:10.1577/T07-247.1.

Al-Chokhachy, R., Fredenberg, W., and Spalding, S. 2008. Surveying professionalopinion to inform bull trout recovery and management decisions. Fisheries,33(1): 18–28. doi:10.1577/1548-8446-33.1.18.

Al-Chokhachy, R., Budy, P., and Conner, M. 2009. Detecting declines in theabundance of a bull trout (Salvelinus confluentus) population: understandingthe accuracy, precision, and costs of our efforts. Can. J. Fish. Aquat. Sci. 66(4):649–658. doi:10.1139/F09-026.

Al-Chokhachy, R., Alder, J., Hostetler, S., Gresswell, R., and Shepard, B. 2013.Thermal controls of Yellowstone cutthroat trout and invasive fishes underclimate change. Global Change Biol. 19(10): 3069–3081. doi:10.1111/gcb.12262.

Al-Chokhachy, R., Moran, S., McHugh, P.A., Bernall, S.R., Fredenberg, W., andDosSantos, J.M. 2015. Consequences of actively managing a small bull troutpopulation in a fragmented landscape. Trans. Am. Fish. Soc. 144: 515–531.doi:10.1080/00028487.2015.1007162.

Almodovar, A., Nicola, G.G., Ayllon, D., and Elvira, B. 2012. Global warmingthreatens the persistence of Mediterranean brown trout. Global Change Biol.18(5): 1549–1560. doi:10.1111/j.1365-2486.2011.02608.x.

Alvarez, D., Cano, J.M., and Nicieza, A.G. 2006. Microgeographic variation inmetabolic rate and energy storage of brown trout: countergradient selectionor thermal sensitivity? Evol. Ecol. 20(4): 345–363. doi:10.1007/s10682-006-0004-1.

Baldwin, T.J., Peterson, J.E., McGhee, G.C., Staigmiller, K.D., Motteram, E.S.,Downs, C.C., and Stanek, D.R. 1998. Distribution of Myxobolus cerebralis insalmonid fishes in Montana. J. Aquat. Anim. Health, 10(4): 361–371.

Baldwin, T.J., Vincent, E.R., Silflow, R.M., and Stanek, D. 2000. Myxobolus cerebralisinfection in rainbow trout (Oncorhynchus mykiss) and brown trout (Salmo trutta)exposed under natural stream conditions. J. Vet. Diagn. Invest. 12(4): 312–321.doi:10.1177/104063870001200403. PMID:10907859.

Bateman, D.S., Gresswell, R.E., and Torgersen, C.E. 2005. Evaluating single-passcatch as a tool for identifying spatial pattern in fish distribution. J. Freshw.Ecol. 20(2): 335–345. doi:10.1080/02705060.2005.9664974.

Bohlin, T., Pettersson, J., and Degerman, E. 2001. Population density of migratoryand resident brown trout (Salmo trutta) in relation to altitude: evidence for amigration cost. J. Anim. Ecol. 70(1): 112–121. doi:10.1046/j.1365-2656.2001.00466.x.

Budy, P., Thiede, G.P., Lobon-Cervia, J., Gonzalez Fernandez, G., McHugh, P.,McIntosh, A., Vollestad, L.A., Becares, E., and Jellyman, P. 2013. Limitationand facilitation of one of the world’s most invasive fish: an intercontinentalcomparison. Ecology, 94(2): 356–367. doi:10.1890/12-0628.1. PMID:23691655.

Caissie, D. 2006. The thermal regime of rivers: a review. Freshw. Biol. 51(8):1389–1406. doi:10.1111/j.1365-2427.2006.01597.x.

Cattaneo, F., Lamouroux, N., Breil, P., and Capra, H. 2002. The influence ofhydrological and biotic processes on brown trout (Salmo trutta) populationdynamics. Can. J. Fish. Aquat. Sci. 59(1): 12–22. doi:10.1139/f01-186.

Comte, L., Buisson, L., Daufresne, M., and Grenouillet, G. 2013. Climate-inducedchanges in the distribution of freshwater fish: observed and predicted trends.Freshw. Biol. 58(4): 625–639. doi:10.1111/fwb.12081.

Crozier, L.G., Hendry, A.P., Lawson, P.W., Quinn, T.P., Mantua, N.J., Battin, J.,Shaw, R.G., and Huey, R.B. 2008. Potential responses to climate change inorganisms with complex life histories: evolution and plasticity in Pacificsalmon. Evol. Appl. 1(2): 252–270. doi:10.1111/j.1752-4571.2008.00033.x. PMID:25567630.

Daufresne, M., and Renault, O. 2006. Population fluctuations, regulation andlimitation in stream-living brown trout. Oikos, 113(3): 459–468. doi:10.1111/j.2006.0030-1299.14295.x.

DeHaan, P.W., Schwabe, L.T., and Ardren, W.R. 2010. Spatial patterns of hybrid-ization between bull trout, Salvelinus confluentus, and brook trout, Salvelinusfontinalis in an Oregon stream network. Conserv. Genet. 11(3): 935–949. doi:10.1007/s10592-009-9937-6.

Dunham, J.B., Adams, S.B., Schroeter, R.E., and Novinger, D.C. 2002. Alien inva-sions in aquatic ecosystems: toward an understanding of brook trout inva-sions and potential impacts on inland cutthroat trout in western NorthAmerica. Rev. Fish Biol. Fish. 12(4): 373–391. doi:10.1023/A:1025338203702.

Dunham, J., Rieman, B., and Chandler, G. 2003. Influences of temperature andenvironmental variables on the distribution of bull trout within streams atthe southern margin of its range. N. Am. J. Fish. Manage. 23(3): 894–904.doi:10.1577/M02-028.

Elliott, J.M. 1976. The energetics of feeding, metabolism and growth of browntrout (Salmo trutta L.) in relation to body weight, water temperature andration size. J. Anim. Ecol. 45: 923–948. doi:10.2307/3590.

Elliott, J.M. 1994. Quantitative ecology and the brown trout. Oxford UniversityPress, New York.

Elliott, J.M. 2009. Validation and implications of a growth model for browntrout, Salmo trutta, using long-term data from a small stream in north-westEngland. Freshw. Biol. 54(11): 2263–2275. doi:10.1111/j.1365-2427.2009.02258.x.

Elliott, J.M., and Elliott, J.A. 2010. Temperature requirements of Atlantic salmonSalmo salar, brown trout Salmo trutta and Arctic charr Salvelinus alpinus: pre-dicting the effects of climate change. J. Fish Biol. 77(8): 1793–1817. doi:10.1111/j.1095-8649.2010.02762.x. PMID:21078091.

Elliott, J.M., and Hurley, M.A. 2000. Daily energy intake and growth of piscivo-rous brown trout, Salmo trutta. Freshw. Biol. 44(2): 237–245. doi:10.1046/j.1365-2427.2000.00560.x.

Essington, T.E., Sorensen, P.W., and Paron, D.G. 1998. High rate of redd super-imposition by brook trout (Salvelinus fontinalis) and brown trout (Salmo trutta)in a Minnesota stream cannot be explained by habitat availability alone. Can.J. Fish. Aquat. Sci. 55(10): 2310–2316. doi:10.1139/f98-109.

Falke, J.A., Flitcroft, R.L., Dunham, J.B., McNyset, K.M., Hessburg, P.F., andReeves, G.H. 2015. Climate change and vulnerability of bull trout (Salvelinusconfluentus) in a fire-prone landscape. Can. J. Fish. Aquat. Sci. 72(2): 304–318.doi:10.1139/cjfas-2014-0098.

Finstad, A.G., Forseth, T., Jonsson, B., Bellier, E., Hesthagen, T., Jensen, A.J.,Hessen, D.O., and Foldvik, A. 2011. Competitive exclusion along climate gra-dients: energy efficiency influences the distribution of two salmonid fishes.Global Change Biol. 17(4): 1703–1711. doi:10.1111/j.1365-2486.2010.02335.x.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

8 Can. J. Fish. Aquat. Sci. Vol. 73, 2016

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

Fredenberg, C. 2014. Efficacy of suppressing non-native lake trout in an isolatedbackcountry lake in Glacier National Park. Masters thesis, Montana StateUniversity, Bozeman, Montana.

Fry. 1971. The effect of environmental factors on the physiology of fish. In Fishphysiology. Edited by W.S. Hoar. Academic Press, New York. pp. 1–98.

Granath, W.O., Gilbert, M.A., Wyatt-Pescador, E.J., and Vincent, E.R. 2007.Epizootiology of Myxobolus cerebralis, the causative agent of salmonid whirl-ing disease in the rock creek drainage of west-central Montana. J. Parasitol.93(1): 104–119. doi:10.1645/GE-948R.1. PMID:17436949.

Guy, C.S., McMahon, T.E., Fredenberg, W.A., Smith, C.J., Garfield, D.W., andCox, B.S. 2011. Diet overlap of top-level predators in recent sympatry: bulltrout and nonnative lake trout. J. Fish Wildl. Manage. 2(2): 183–189. doi:10.3996/012011-JFWM-004.

Hansen, M.J., Schill, D., Fredericks, J., and Dux, A. 2010. Salmonid predator–preydynamics in Lake Pend Oreille, Idaho, U.S.A. Hydrobiologia, 650(1): 85–100.

Hanski, I. 1999. Metapopulation ecology. Oxford University Press, New York.Hari, R.E., Livingstone, D.M., Siber, R., Burkhardt-Holm, P., and Guttinger, H.

2006. Consequences of climatic change for water temperature and browntrout populations in Alpine rivers and streams. Global Change Biol. 12(1):10–26. doi:10.1111/j.1365-2486.2005.001051.x.

Howell, P.J., Dunham, J.B., and Sankovich, P.M. 2010. Relationships betweenwater temperatures and upstream migration, cold water refuge use, andspawning of adult bull trout from the Lostine River, Oregon, U.S.A. Ecol.Freshw. Fish, 19(1): 96–106.

Humbert, J.-Y., Mills, L.S., Horne, J.S., and Dennis, B. 2009. A better way toestimate population trends. Oikos, 118: 1940–1946. doi:10.1111/j.1600-0706.2009.17839.x.

Isaak, D.J., Luce, C.H., Rieman, B.E., Nagel, D.E., Peterson, E.E., Horan, D.L.,Parkes, S., and Chandler, G.L. 2010. Effects of climate change and wildfire onstream temperatures and salmonid thermal habitat in a mountain rivernetwork. Ecol. Appl. 20(5): 1350–1371. doi:10.1890/09-0822.1. PMID:20666254.

Isaak, D.J., Wollrab, S., Horan, D.L., and Chandler, G. 2012. Climate changeeffects on stream and river temperatures across the northwest U.S. from1980–2009 and implications for salmonid fishes. Clim. Change, 113(2): 499–524. doi:10.1007/s10584-011-0326-z.

Isaak, D.J., Young, M.K., Nagel, D.E., Horan, D.L., and Groce, M.C. 2015. Thecold-water climate shield: delineating refugia for preserving salmonid fishesthrough the 21st century. Global Change Biol. 21(7): 2540–2553. doi:10.1111/gcb.12879.

Jones, L.A., Muhlfeld, C.C., Marshall, L.A., McGlynn, B.L., and Kershner, J.L. 2014.Estimating thermal regimes of bull trout and assessing the potential effectsof climate warming on critical habitats. River Res. Appl. 30(2): 204–216. doi:10.1002/rra.2638.

Jonsson, B., and Jonsson, N. 2009. A review of the likely effects of climate changeon anadromous Atlantic salmon Salmo salar and brown trout Salmo trutta,with particular reference to water temperature and flow. J. Fish Biol. 75(10):2381–2447. doi:10.1111/j.1095-8649.2009.02380.x. PMID:20738500.

Kery, M., and Schaub, M. 2012. Population analysis using WinBUGS, a hierarchi-cal perspective. Academic Press, Waltham, Mass.

Kovach, R.P., Al-Chokhachy, R., Muhlfeld, C.C., Dunham, J., Letcher, B.H., andKershner, J.L. 2016. Impacts of climatic variation on trout: a global synthesisand path forward. Rev. Fish Biol. Fish. 26(2): 135–151. doi:10.1007/s11160-015-9414-x.

Kovach, R.P., Muhlfeld, C.C., Wade, A.A., Hand, B.K., Whited, D.C., DeHaan, P.W.,Al-Chokhachy, R., and Luikart, G. 2015. Genetic diversity is related to climaticvariation and vulnerability in threatened bull trout. Global Change Biol. 21:2510–2524. doi:10.1111/gcb.12850.

Lawrence, D.J., Stewart-Koster, B., Olden, J.D., Ruesch, A.S., Torgersen, C.E.,Lawler, J.J., Butcher, D.P., and Crown, J.K. 2014. The interactive effects ofclimate change, riparian management, and a nonnative predator on stream-rearing salmon. Ecol. Appl. 24(4): 895–912. doi:10.1890/13-0753.1. PMID:24988784.

Leary, R.F., Allendorf, F.W., and Forbes, S.H. 1993. Conservation genetics of bulltrout in the Columbia and Klamath River drainages. Conserv. Biol. 7(4): 856–865. doi:10.1046/j.1523-1739.1993.740856.x.

Lobon-Cervia, J. 2004. Discharge-dependent covariation patterns in the popula-tion dynamics of brown trout (Salmo trutta) within a Cantabrian river drain-age. Can. J. Fish. Aquat. Sci. 61(10): 1929–1939. doi:10.1139/f04-118.

Lobon-Cervia, J. 2014. Recruitment and survival rate variability in fish popula-tions: density-dependent regulation or further evidence of environmentaldeterminants? Can. J. Fish. Aquat. Sci. 71(2): 290–300. doi:10.1139/cjfas-2013-0320.

Lowery, E.D. 2009. Trophic relations and seasonal effectsof predation on Pacificsalmon by fluvial bull trout in a riverine food web, School of Aquatic andFishery Sciences, University of Washington, Seattle, Wash.

Martinez, P.J., Bigelow, P.E., Deleray, M.A., Fredenberg, W.A., Hansen, B.S.,Horner, N.J., Lehr, S.K., Schneidervin, R.W., Tolentino, S.A., and Viola, A.E.2009. Western lake trout woes. Fisheries, 34(9): 424–442. doi:10.1577/1548-8446-34.9.424.

McMahon, T.E., Zale, A.V., Barrows, F.T., Selong, J.H., and Danehy, R.J. 2007.Temperature and competition between bull trout and brook trout: A test ofthe elevation refuge hypothesis. Trans. Am. Fish. Soc. 136(5): 1313–1326. doi:10.1577/T06-217.1.

Meyer, K.A., Garton, E.O., and Schill, D.J. 2014. Bull trout trends in abundanceand probabilities of persistence in Idaho. N. Am. J. Fish. Manage. 34(1): 202–214. doi:10.1080/02755947.2013.869280.

Morita, K., and Yamamoto, S. 2002. Effects of habitat fragmentation by dam-ming on the persistence of stream-dwelling charr populations. Conserv. Biol.16(5): 1318–1323. doi:10.1046/j.1523-1739.2002.01476.x.

Mote, P.W. 2003. Trends in snow water equivalent in the Pacific Northwest andtheir climatic causes. Geophys. Res. Lett. 30(12). doi:10.1029/2003GL017258.

Nelson, K.C., Palmer, M.A., Pizzuto, J.E., Moglen, G.E., Angermeier, P.L.,Hilderbrand, R.H., Dettinger, M., and Hayhoe, K. 2009. Forecasting the com-bined effects of urbanization and climate change on stream ecosystems:from impacts to management options. J. Appl. Ecol. 46(1): 154–163. doi:10.1111/j.1365-2664.2008.01599.x.

Palmer, M.A., Lettenmaier, D.P., Poff, N.L., Postel, S.L., Richter, B., and Warner, R.2009. Climate change and river ecosystems: protection and adaptationoptions. Environ. Manage. 44(6): 1053–1068. doi:10.1007/s00267-009-9329-1.PMID:19597873.

Pederson, G.T., Graumlich, L.J., Fagre, D.B., Kipfer, T., and Muhlfeld, C.C. 2010. Acentury of climate and ecosystem change in Western Montana: what dotemperature trends portend? Clim. Change, 98(1–2): 133–154. doi:10.1007/s10584-009-9642-y.

Persson, L., Amundsen, P.A., De Roos, A.M., Knudsen, R., Primicerio, R., andKlemetsen, A. 2013. Density-dependent interactions in an Arctic char – browntrout system: competition, predation, or both? Can. J. Fish. Aquat. Sci. 70(4):610–616. doi:10.1139/cjfas-2012-0175.

Peterson, D.P., Fausch, K.D., Watmough, J., and Cunjak, R.A. 2008. When eradi-cation is not an option: modeling strategies for electrofishing suppression ofnonnative brook trout to foster persistence of sympatric native cutthroattrout in small streams. N. Am. J. Fish. Manage. 28(6): 1847–1867. doi:10.1577/M07-174.1.

Peterson, D.P., Wenger, S.J., Rieman, B.E., and Isaak, D.J. 2013. Linking climatechange and fish conservation efforts using spatially explicit decision supporttools. Fisheries, 38(3): 112–127. doi:10.1080/03632415.2013.769157.

Pierce, R., Podner, C., Marczak, L., and Jones, L. 2014. Instream habitat restora-tion and stream temperature reduction in a whirling disease-positive springcreek in the Blackfoot River Basin, Montana. Trans. Am. Fish. Soc. 143(5):1188–1198. doi:10.1080/00028487.2014.925972.

Pollock, M.M., Beechie, T.J., Wheaton, J.M., Jordan, C.E., Bouwes, N., Weber, N.,and Volk, C. 2014. Using beaver dams to restore incised stream ecosystems.Bioscience, 64(4): 279–290. doi:10.1093/biosci/biu036.

Rahel, F.J., and Olden, J.D. 2008. Assessing the effects of climate change onaquatic invasive species. Conserv. Biol. 22(3): 521–533. doi:10.1111/j.1523-1739.2008.00950.x. PMID:18577081.

Rieman, B.E., and Allendorf, F.W. 2001. Effective population size and geneticconservation criteria for bull trout. N. Am. J. Fish. Manage. 21(4): 756–764.doi:10.1577/1548-8675(2001)021<0756:EPSAGC>2.0.CO;2.

Rieman, B.E., Isaak, D., Adams, S., Horan, D., Nagel, D., Luce, C., and Myers, D.2007. Anticipated climate warming effects on bull trout habitats and popu-lations across the interior Columbia River basin. Trans. Am. Fish. Soc. 136(6):1552–1565. doi:10.1577/T07-028.1.

Selong, J.H., McMahon, T.E., Zale, A.V., and Barrows, F.T. 2001. Effect of temper-ature on growth and survival of bull trout, with application of an improvedmethod for determining thermal tolerance in fishes. Trans. Am. Fish. Soc.130(6): 1026–1037. doi:10.1577/1548-8659(2001)130<1026:EOTOGA>2.0.CO;2.

Stacey, P.B., and Taper, M. 1992. Environmental variation and the persistence ofsmall populations. Ecol. Appl. 2(1): 18–29. doi:10.2307/1941886.

Starcevich, S.J., Howell, P.J., Jacobs, S.E., and Sankovich, P.M. 2012. Seasonalmovement and distribution of fluvial adult bull trout in selected watershedsin the mid-Columbia River and Snake River basins. Plos ONE, 7(5): e37257.doi:10.1371/journal.pone.0037257. PMID:22655037.

Storaasli, J., and Moran, S. 2015. Lower Clark Fork River, Montana-Avista ProjectArea 2014 annual bull and brown trout redd survey report. Avista Corpora-tion, Noxon, Mont.

Strahler, A.N. 1952. Hypsometric (area–altitude) analysis of erosional topogra-phy. Geol. Soc. Am. Bull. 63: 1117–1142. doi:10.1130/0016-7606(1952)63[1117:HAAOET]2.0.CO;2.

Swanberg, T.R. 1997. Movements of and habitat use by fluvial bull trout in theBlackfoot River, Montana. Trans. Am. Fish. Soc. 126(5): 735–746. doi:10.1577/1548-8659(1997)126<0735:MOAHUB>2.3.CO;2.

Taniguchi, Y., and Nakano, S. 2000. Condition-specific competition: implica-tions for the altitudinal distribution of stream fishes. Ecology, 81(7): 2027–2039. doi:10.1890/0012-9658(2000)081[2027:CSCIFT]2.0.CO;2.

Taylor, B.L., and Gerrodette, T. 1993. The uses of statistical power in conservationbiology: the vaquita and northern spotted owl. Conserv. Biol. 7(3): 489–500.doi:10.1046/j.1523-1739.1993.07030489.x.

Tilman, D., May, R.M., Lehman, C.L., and Nowak, M.A. 1994. Habitat destructionand the extinction debt. Nature, 371(6492): 65–66. doi:10.1038/371065a0.

Tonina, D., Luce, C.H., Rieman, B., Buffington, J.M., Goodwin, P., Clayton, S.R.,Ali, S.M., Barry, J.J., and Berenbrock, C. 2008. Hydrological response to timberharvest in northern Idaho: implications for channel scour and persistence ofsalmonids. Hydrol. Process. 22(17): 3223–3235. doi:10.1002/hyp.6918.

Unfer, G., Hauer, C., and Lautsch, E. 2011. The influence of hydrology on the

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Al-Chokhachy et al. 9

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.

recruitment of brown trout in an Alpine river, the Ybbs River, Austria. Ecol.Freshw. Fish, 20(3): 438–448. doi:10.1111/j.1600-0633.2010.00456.x.

USFWS. 2014. Revised draft recovery plan for teh coterminous United Statespopulation of bull trout (Salvelinus confluentus). US Fish and Wildlife Service,Portland, Ore.

USFWS. 2015. Draft Columbia Hadwaters Rcovery Uit Imlementation Plan forBull Trout (Salvelinus confluentus) Recovery Plan. US Fish and Wildlife Service,Helena, Mont.

Warnock, W.G., and Rasmussen, J.B. 2014. Comparing competitive ability andassociated metabolic traits between a resident and migratory population ofbull trout against a non-native species. Environ. Biol. Fishes, 97(4): 415–423.doi:10.1007/s10641-013-0161-3.

Warren, D.R., Robinson, J.M., Josephson, D.C., Sheldon, D.R., and Kraft, C.E. 2012.Elevated summer temperatures delay spawning and reduce redd construc-tion for resident Brook Trout (Salvelinus fontinalis). Global Change Biol. 18(6):1804–1811. doi:10.1111/j.1365-2486.2012.02670.x.

Wenger, S.J., Isaak, D.J., Dunham, J.B., Fausch, K.D., Luce, C.H., Neville, H.M.,

Rieman, B.E., Young, M.K., Nagel, D.E., Horan, D.L., and Chandler, G.L. 2011.Role of climate and invasive species in structuring trout distributions in theinterior Columbia River Basin, USA. Can. J. Fish. Aquat. Sci. 68(6): 988–1008.doi:10.1139/f2011-034.

Whiteley, A.R., Spruell, P., Rieman, B.E., and Allendorf, F.W. 2006. Fine-scalegenetic structure of bull trout at the southern limit of their distribution.Trans. Am. Fish. Soc. 135(5): 1238–1253. doi:10.1577/T05-166.1.

Whitesel, T., Brostrom, J., cummings, T., Delavergne, J., Fredenberg, W.,Schaller, H., Wilson, P., and Zydlewski, G. 2004. Bull trout recovery planning:a review of the science associated with population structure and size ScienceTeam Report 2004-01. US Fish and Wildlife Service, Portland, Ore.

Williams, J.E., Haak, A.L., Neville, H.M., and Colyer, W.T. 2009. Potential conse-quences of climate change to persistence of cutthroat trout populations.N. Am. J. Fish. Manage. 29(3): 533–548. doi:10.1577/M08-072.1.

Wood, J., and Budy, P. 2009. The role of environmental factors in determiningearly survival and invasion success of exotic brown trout. Trans. Am. Fish.Soc. 138(4): 756–767. doi:10.1577/T08-123.1.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

10 Can. J. Fish. Aquat. Sci. Vol. 73, 2016

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y U

SGS

LIB

RA

RY

on

06/0

3/16

For

pers

onal

use

onl

y.