sign in sign up
<<
Home , Cutthroat trout

Transactions of the American Fisheries Society 125:768-779. 1996 © Copyright by the American Fisheries Society 1996

Spatial Variatio Spawninn i g Habita f Cutthroato t Troua n ti Sediment-Rich Stream Basin JAMES P. MAGEE' AND THOMAS E. McMAHON2 Biology Department, FishWildlifeand Program State University, Bozeman, Montana 59717,USA RUSSELL E THUROW U.S. Forest Service, Intermountain Research Station 316 East Myrtle Street. Boise, 83702,USA

Abstract.—We examined distributio habitad nan t characteristic f spawnino s g site f cutthroaso t clarki t varioua s spatial scale asseso st s effect f sedimentatioso n withi nlarga e basin in Montana. Redd density varied widely across the basin; nearly all (99%) of the 362 redds observed occurred in two high-elevation headwater . Redd density at the reach scale s positivelwa y correlate 0.001= 0.72= 2 P ,d(r ) with abundanc f spawnino e g gravels. Other habitat variables, such as gradient, width, depth, embeddedness, bank stability, and percent riffle, were not significantly correlated to redd density. Taylor Fork redds contained some of the highest proportions of fine sediments (<6.35 mm, mean = 41.6%; <0.85 mm, 17.9%) observed in egg pocket salmonif so d Rocke reddth n si y Mountain region. Cache Creek highl,a y disturbed subbasin, d significantlha y greater proportion f fino s e sediments smaller tha reddn i n 0.8m s m 5tha e nth undisturbed Wapiti Creek subbasin. High fine-sediment level vero reddestimaten t w i s d ylo sle d embryo survival (mean, 8.5%) t sedimentatiobu , t appeano limid o t r di n t recruitment r datOu a. suggest that compensatory juvenile survival and high embryo survival in the small proportion of redds with good substrate quality may buffer the effects of the high sediment levels in the basin.

Predictin effecte gth f sedimentatioso n resulting 1987). Dominant geological formatione th n i s from lan activitiee dus s beesha researcna h pri- range of cutthroat trout Oncorhynchus clarki in orit r fisheryfo y biologist e pas 0 yearth 6 tr fo s Montana include highly erosive granitic serief o s (Chapman 1988). e researcMuc- th fo f s ho ha h the Idaho Batholith, cobble-producing Belt series, cused on effects of fine sediment on emergence varioud an s fine silt-producing volcani sedd can - succes f salmonio s y (e.g.fr d , Chapman 1988; imentary series (AlHyndmad an t n 1986). Much Young et al. 1991; Weaver and Fraley 1993). In of the research on sediment effects on salmonids their review of sediment studies, Everest et al. Rocke inth y Mountain regio focuses nha gran do - (1987) concluded that studies of population-level nitic soils and sand-sized particles of the Idaho effect needee ar s examino dt sedimentatiow eho n Batholith (Bjorn . 1977al t ne ; Platt. 1989)al t e s . affects fish populations within entire drainage ba- To evaluate the effects of smaller sediment parti- sins. However foune publishe,w w dfe d studies that cles, we sampled streams draining sedimentary ge- have adopted this approach. Exception studiee sar s ology. by Shepar . (1984a)al t e d , Cederhol Reid man d Recent studies indicate that spawning habitat of (1987) Scrivened an , Brownled an r e (1989) that Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 stream-dwelling salmonids may be more spatially assessed population effect f sedimentatioo s y nb variable across drainag play e basinma ya d an s combining data on numbers of recruits, relations larger rol determininn ei g pattern f distributioso n between fine sediment d egg-to-fran s y survival, of juveniles and adults than previously thought d amounan f fino t e sediments within redds con- (Bear Carlind dan e 1991; Boze Rahed kan l 1991). tributed by land use activities. To properly characterize such high spatial vari- Geologic characteristic drainaga f so e basi- nex abilit documenhabitad n yi an e tus t populatio- nef strona t er g influenc degren eo f sedimentatioeo n fects of habitat change, sampling designs that em- and sensitivit lano yt d disturbance (Everes. al t e t ploy a hierarchial series of spatial scales ranging froe entirmth e watershe o microhabitatt d e ar s needed (Frissell et al. 1986). 1 Present address: Montana Fish, Wildlife, and Parks, 0 Nort73 h Montana Avenue, Dillon, Montana 59725, The objectives of our study were to (1) char- USA. acterize spawning habitats of cutthroat trout across 2 To whom correspondence should be addressed. larga e sediment-rich examin) (2 , e 768 CUTTHROAT TROUT SPAWNING HABITAT 769

D5

LEGEND Traps Thermographs Reach Boundaries

2km

FIGURE I.—Location f spawninso g reaches (stipple coded dan d areas), weirs thermographd an , Taylon si r Fork basin, Montana.

factors influencing redd distribution at different drainage. Headwater sections (above 2,300 m) of

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 spatial scales (basin, subbasin ) reach)d (3 an , d an , the major third-order tributaries (Cache, Wapiti, evaluate effects of sedimentation on redd substrate and Lightning creeks) meander through alpine compositio estimated nan d emergence successe .W meadows. Midelevation sections (2,100-2,300 m) use the data to address the question, Is fine sedi- have steeper gradients, more constrained valley ment limiting recruitmen e cutthroath f o t t trout floors, and are bordered by coniferous forest. Low- populatio Tayloe th n ni r Fork basin? er sections (below 2,100 m) and the main-stem Taylor Fork drain wide valley floors vegetated by Study Area sagebrush Artemisia spp. and willow Salix spp. Taylor Fork is a fourth-order to the Taylor Fork drains sedimentary soils predomi- upper Gallatin in southwest Montana (Figure nantly consisting of fine silt (Snyder et al. 1978). 1). The drainage encompasses 161 km2 and con- This material is highly erosive and produces large tains abou strea0 10 t m kilometers. Peak flow- oc s quantities of fine sediment; many of the stream- cur from snowmelt during late May to mid-June. bankbasie th n i s were rated unstabl Snydey eb r Three major elevational land types exist in th . (1978)e al t e . Habitat condition basie th nn si range 770 MAGE. AL T EE

TABLE 1.—Genetic statu f Tayloso r Fork cutthroat trout morphology, or the presence of a tributary junction s determinea proteiy db n electrophoresi . Learys(R , Uni- (Frissell et al. 1986). In 1992 we counted redds in versity of Montana, personal communication). eac July1 h o t reac. y h Ma frod ever5 m2 4 y3- Percent genetic compositio: nof We counted a redd only if we observed fish - Westslope Yellowslone ing; cleaned gravel sites reliabl e coulb t dno y iden- cutthroat Rainbow cutthroat tifie s redda d s becaus f theieo r small size, lacf ko Location trout trout trout periphyton hige th hd fine-sedimenan , te loath n di Cache Creek 87 11 2 stream e markeW . d individual redd y placinb s g Upper Wapiti Creek 57 43 numbered stakes on the streambank and mapped Deadhorse Creek 85 10 5 Upper Taylor Fork0 93 2 5 probable egg pocket locations by observing where a fish spawned. Redd densit- ex r reacs pe y wa h Near Lightning Creek. pressed as number of redds/100 m2 of wetted stream area. Thermographs recorded hourly tem- from pristin highlo et y altered. Splash dams were peratur t fivea e location basie th n si (Figure 1). operate late th e n main-stede 1800i th n so m Taylor We also employed weirs in the lower (Little Fork to transport logs downstream (Snyder et al. Wapiti Creek), middle (Deadhorse Creek)- up d an , 1978). Livestock graze allotment uppen i s r Cache per (Cache Creek) portions of the basin (Figure 1) and middle Wapiti creeks, and areas of the Cache to assess whed wheran n e spawning occurred. Creek drainage have been recently logged. In the Weirs spanne e streath d m channe d capturean l d Cache Creek subbasin, there is a predominance of upstream and downstream migrants. Weirs were bare, unstable banks and the incised channels and checked twice dail yJul1 fro yo t m 1992y 19Ma . encroachmen sagebrusf to h characteristi heavif co - Fish were weighed, measured for fork length, and ly grazed riparian zones (Snyder et al. 1978). In tagged with a visible implant tag behind the left contrast, upper Wapiti Creek, whic s similahi n i r eye. Species, sex, and spawning condition were size, elevation d geomorphologan , o t uppey r also noted. Cache Creek, has no known history of logging, Spawning habitat.—Habitat features were mea- grazing r roao , d building. sured in 11 of the 17 spawning reaches to evaluate remaininw e basife Th e n th contain f go e on s potential physical factors associated with variation populations of westslope cutthroat trout O. clarki in redd density. We were unable to measure habitat lewisi in the Gallatin River drainage (Liknes 1984). feature e othe th spawninx si n ri s g reachea s a s Other fishes inhabiting the basin include mountain resul f earl o te cover ic y . Sampled reaches repre- whitefish Prosopium williamsoni, mottled sente fule dth l rang observef eo d habitat conditions Cottus bairdi, nonnative O. mykiss, and redd densities. We measured habitat according trutta, d Yellowstonan e cut- to the procedure of Overton et al. (in press), a throat trout O. c. bouvieri. Rainbow trout were modificatio e systematith f o n c habitat-sampling stocke Taylon di r Fork from late 192th e o 81980st . procedure described Hankiy b Reeved nan s (1988). Yellowstone cutthroat trout were stocke smaln di l Proceeding upstream describee w , d each channel headwater subsequentld an s y invaded down- unit (pool, riffle, etc.d measurean ) s lengthit d , stream reaches a result s A . , westslope cutthroat mean wetted width, and mean and maximum trout are introgressed with rainbow trout and Yel- depths. The area of potential spawning gravel was

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 lowstone cutthroat trout to varying degrees measured wit meteha r stic expressed kan pers da - throughout the basin (Table 1). We assume a neg- cen f totao t l reach surface area. Base obsern do - ligible influence of this introgression on the effects vations of 362 completed redds, we defined po- of sedimentation. tential spawning grave s patchea l f substratso t ea least 0.25 maren i 2 a with particlesn i 2-3 m 5m Methods diameter. The same observer made all estimates of Spawner and redd distribution.—The entire ba- spawning grave o reduct l e bias t everA . y 10th sin was surveyed in summer 1991 to identify po- channel unit measuree w , d gradient wit clinomha - tential spawnin ge presenc th site y b s f reddso e , eter and visually estimated percent bank stability suitable spawning gravel, and newly emerged fry. and substrate embeddedness, using rating criteria Potential spawning sites were then stratified into described Platt. y (1983)b al t se poon I . l tailouts 17 reaches (Figure 1), rangin lengtn gi h fro4 m32 and riffles, we measured substrate composition to 3,02 (mea0m n . Reac lengthm) 4 h 91 bound, - with the pebble count technique (Wolman 1954), aries were delineated by changes in gradient, geo- using 100 data points at each site in streams wider CUTTHROAT TROUT SPAWNING HABITAT 771

than 1 m and 50 data points in streams narrower comparison test (Zar 1984) usee W .d rank corre- classifiee W . tham 1 nd substrate siz s sand-silea t latio examino nt e associations between redd den- diameter)n i m m grave a 2 pe , (< l (2-16 mm), grav- d measuresitan y f substrato s e composition (F/; el (17-75 mm), rubble (76-150 mm), cobble (151- proportion of particle sizes less than 6.35, 2.36, 300 mm), or boulder (>300 mm). We used Spear- 0.85 mm). Stepwise multiple regressio uses nwa d man \s rank correlation and simple linear regression proportioe asseso th t w sho different-sizef no d fine (Zar 1984 o assest ) s associations between redd particles influenced variatio F/ n nMann-Whiti .A - density and reach habitat features. We were unable ney IMest was used to compare redd substrate to statistically compare habitat features among composition amon pockeg g eg non-eg d an t g pock- subbasins because onlreace samples yon hwa n di et cores and to compare redd substrate composition three th tw f eo o subbasins . among subbasins havin (Wapitw glo i Creekd an ) Redd substrate composition.—We used a McNeil high (Cache Creek) land disturbance. hollow-core sampler (Platts et al. 1983) to exca- Fry production.—We predicte emergency fr d e vate substrate from 21 redds in upper Cache Creek success for each cored redd using the equation (five reaches or reach groups) and 15 redds in upper developed by Weaver and Fraley (1993) for west- Wapiti Creek (three reaches). In Cache Creek, we slope cutthroat trout: grouped north fork (reaches 8-11 soutd )an h fork % emergence = -0.7512 (arcsine % SP^js) (reaches 13-17) spawning reaches together for eas f samplingeo . Thre seveo et n redds were ran- + 39.67; domly chosen for coring in each of these eight % Sp6.35 = percentage of substrate particles designated reaches or reach groups. To quantify smaller than 6.35 mm. substrate conditions near tim f emergenceeo e w , collected cores during a 2-week period (23 July to Total fry production per reach was estimated by 6 August) following our first sighting of emergent combining dat n totao a l redd number, estimated fry (22 July). We used the redd maps to position egg deposition, and average substrate composition cores over egg pockets. Cores were taken to a of cored redds accouno T . r effectfo t f femalo s e depth of 10 cm to mimic egg pocket depths ob -depositiong eg siz n eo dividee w , length-free dth - serve cutthroan di t trout redd . May(B s , Gallatin quency distribution of females captured in the National Forest, personal communication)- la e W . Cache Creek spawning trap into 25-mm intervals beled pocke g coreeg s sa t sample observee w f si d and calculated egg deposition (£) for each size- class using a length-fecundity relation for west- eggs or alevins in the core. 0007958L Oven-dried core samples were weighed after be- slope cutthroat trout: E = 82.63e° , where ing mechanically shaken through sieves of 50.8, s forLi k lengt millimetern hi . Down. B (C s d san 25.4, 12.4, 9.5, 6.35, 2.36, 0.85, and 0.074 mm Shepard, Montana State University, personal com- (see Magee 199 r furthe3fo r detail) collectee W . d munication). We assumed all mature females had an average of 3.6 kg of substrate per core. As equan a l likelihoo f spawnindo apportioned gan d recommended by Chapman (1988) and Young et depositiog eg percentage reddo nth t y - b s fe f eo . (1991)al expressee w , d substrate compositioy nb males within each size-class (25%, 150-17; 4mm two methods: the percentage of fine substrate 50%, 175-199 mm; 16%, 200-224 mm; 6%, 225- smaller tha givena n size (6.35, 2.36 0.8) d ,an 5 mm , >253% 0d mm)an productioy ; Fr . mm 9 24 n ni

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 and by a measure of central tendency, the fredle a reach was estimated by multiplying total esti-

index F/ (F/ = DIS\ D is the geometric mean mated egg deposition by estimated emergence suc-

0 g g cess. particle diamete a sortin Sd s i 0 an r g coefficient; see Young et al. 1991). As in previous studies Results (Grost et al. 1991; Thurow and King 1994), we calculated excludiny Fb / g particle sizes greater Spawner and Redd Distribution tha nminimizo t 50.m 8m e bias. (Particles greater The majority (271, 94%) of trout trapped during than 50.rangem 8m y weighd7 b 1 fro % n mi 0 t upstream spawning migration were capturee th n di redds to 3.3-37.5% in 19 redds.) This procedure Cache Creek weir; only 10 trout were captured in also allowe comparo t s du r resulteou s with other Deadhors n Littli e 6 Creee d Wapitan k i Creek studies that used truncated data. weirs. Cutthroat trout accounted for 99.6% of fish We compared substrate composition among caught in the upper basin weir at Cache Creek; spawning reaches with a Kruskal-Wallis nonpara- rainbow trout constituted a larger proportion of the metric analysis of variance and Tukey's multipl o lower-elevatioetw e catcth t a h n weirs (10%n i 772 MAGEE ET AL.

TABLE 2.—Numbe densitd ran y (redds/100 m f trou2)o t the main stem of Taylor Fork or in Eldridge Creek, redds by stream order in Wapiti (W), Little Wapiti (LW), Meadow Creek, Left Fork Creek, lower Cache Deadhorse (D), and Cache (C) creeks. Creek, middle Wapiti Creek, or Alp Creek. Small Number of Redd numbers of newly emerged fry were observed in Reach Lengt ) h(m redds density lower Wapiti Creek, Little Wapiti Creek, and First order Lightning Creek. Wl 2,262 57 1.7 Most redds were foun first n second-ordedi d an - r C8 500 1 O.I tributaries t redbu ,d density varied widely, even C9 900 25 3.3 among adjacent reaches (Table 2). Cache Creek CIO 843 34 3.4 Cll 859 14 1.4 reaches 15 and 16 contained three redds each C13 585 13 2.0 (0.6/100 m2), whereas adjacent reach 17 contained C14 656 9 1.1 redd9 3 s (3.8/100 m2). Similarly, reac containeh8 d C15 464 3 0.6 2 C16 493 3 0.6 1 redd (0.01/100 m ), whereas adjoining reaches 2 C17 1,025 39 3.redd 5 8containe9 2 d d s an an (3.3/10 7 9 d3 0 m ), Tota r meao l n 8,587 198 1.8 respectively. Second order Spawning Habitat W2 472 17 2.4 W3 1.258 32 0.8 Cutthroat trout spawne reachen di smalf so l (1- LW4 575 1 0.1 3-m-wide) tributarie moderato st witw lo h e gra- D5 324 3 0.7 C7 833 39 3.3 dients (0.5-3.8%) and suitable spawning gravels CI2 477 24 2.8 (Tabl . Redd3) e s were typicall t observeno y t da Tota r meao l n 3.939 116 1.7 gradients greater than 4%. Streambank stabilitf yo spawning reaches averaged less than 50% stable, Third order C6 3.020 48 0.5 and no reaches had streambanks classified as great- stabler tha% e n75 . Substrate embeddedness swa Grand total or mean 15,585 362 1.7 high, averaging about 50%. Measured habitat features generally varied little among reaches that supported redds. Features- in , Deadhors Littln i e% e CreeWapit70 d kan i Creek). cluding gradient, width, depth, embeddedness, Mature male cutthroat trout averaged 189 mm (N bank stability, percent riffle, and substrate com- 156= forn N i )( k= 282m femaled m an ) 1 19 s position, wer t significantlno e y correlated with length. redd density (Tabl . Onl eproportio3) e yth - po f no Ninety percent of upstream migrants entered tential spawning grave highls wa l y correlated with Cache Creek from 19 May to 4 June at mean daily redd density (Figure 2; Table 3). Proportion of water temperatures of 7-9°C. Fish were observed potential spawning gravel varied widely, ranging on redds in Cache Creek from 25 May to 19 June fro25%o t m3 , even among nearby reachese W . and in Wapiti Creek from 8 to 22 June when mean foun associatioo dn 0.1-0.6= P n( ) between abun- daily temperatures were near 8°C. Peak spawning dance of spawning gravel and reach gradient, pro- in Cache Creek occurred from 3 to 5 June, when portion of pea gravel, or proportion of gravel sub- 7 redd7 s (40 f %totalo ) were observed. Peak strate.

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 spawning in Wapiti Creek occurred about 1 week later. Water temperatures in the main-stem Taylor Redd Characteristics Fork were simila thoso t r Cachn ei e Creek. Mean We foun redd6 3 d e eggsth cored1f n si 1o e W . daily temperatures in Lightning Creek did not foun significano dn t differences (Latest 0.47> ,P ) reach 8°C until mid-July. Females remained in in percentages of substrate smaller than 6.35, 2.36, spawning tributaries an average of 11 d; males and 0.85 mm or in F/ values between cores with remaine averagn . a d r 9 d fo f eo and without eggs therefore W . ecor6 3 use l edal redd2 36 se Mos counteth f Tayloe to th n di r Fork samples to characterize substrate composition of basi n 199ni 2 were clustere areao tw sn di (Tabl e redds. 2). Nearly all redds were observed in the upper Cutthroat trout redds consisted primarilf o y reache f Cacho s e Creek (252, 70% Wapitd an ) i small gravels (75% less tha diameternn i 25. m 4m ) Creek (106, 29%). Despite large numbers of redds and contained a high proportion of particles less in these areas redo n , d superimpositio- ob s nwa than 6.3(Figurm 5m . Forte3) y percen sube th -f o t served. In 1991, no redds or fry were observed in strate was smaller than 6.35 mm and 18% was CUTTHROAT TROUT SPAWNING HABITAT 773

TABLE 3.—Mean f habitao s t feature spawninn i s g reache f Wapito s i (W), Deadhorse (D)d Cach an ) ,creeks (C e .

Spearman's rank correlation with redd density (Table 1) shown as r. Asterisk denotes significance at P < 0.01. s

Reach Percent by area or Gra- corre- dient Width Depth Embed- Bank Spawning Pea lation ) (% (m) (cm) dedness3 stability5 gravel Riffle Silt gravel Gravel Rubble W3 1.2 3.1 25 2.8 1.9 14.1 39 17 24 52 8 D5 0.5 1.3 15 2.6 2.5 9.9 42 20 9 59 7 C5 0.5 3.2 27 2.9 3.4 8.5 34 18 14 46 16 C7 0.5 .4 19 3.1 2.7 17.3 38 16 22 58 3 C8 4.0 .1 6 3.2 2.2 3.2 50 23 11 44 15 C9 2.5 .1 6 3.7 2.7 16.2 42 20 14 46 11 CIO 3.0 .2 7 4.4 3.4 13.9 46 19 18 40 15 CI2 0.5 .8 9 2.8 2.1 18.0 44 22 16 50 10 C13 1.0 .1 12 2.6 2.9 13.3 43 24 24 42 4 C14 3.8 .2 27 2.9 3.4 8.5 34 16 8 37 28 C17 1.5 1.0 7 3.2 2.5 25.4 45 25 18 54 4

rs 0.05 -0.38 -0.36 0.47 0.15 0.80* 0.25 0.14 0.48 0.06 -0.39 3 Rate scala percenr n fo do e5 f substratfroo o tt m1 e embedded1 : >75% 50-75%= ;2 25-49%= ;3 5-24%= ;4 (Plait % =;5 <5 s et al. 1983). b Rated on a scale from 1 to 4 for percent stable banks: 1 = <25%; 2 25-49%; 3 = 50-75%; 4 = >75% (Ptatts et al. 1983).

smaller than 0.85 mm. When substrate samples among the eight spawning reaches. Other redd sub- were truncated to remove particles larger than 50.5 strate measures (f/, proportions smaller than 6.35 mm, the percentage of substrate smaller than 6.35 and 2.63 mm) were not significantly different. Co- mm ranged 73%o frot percentage 6 m2 th , e smaller efficients of variation (SD/mean) in redd substrate than 0.85 mm ranged from 8 to 40%, and Ff values variables differed widely both amon withid gan n ranged from 0.4 s negativel5.14o 8t e F wa / Th . y reaches, being low (less than 15%) in some reaches correlated with percentage of substrate smaller (e.g., C6 and C8-11) and high (up to 106%) in than 6.35, 2.36 0.8(rd m = -0.8,an m 5, o 6t others (e.g., W3 and C7). We found no significant -0.97 0.001)< P , . However, whe threl nal e mea- associations (P > 0.16) between redd density and sure f fineso s were regressed against multiplFy /b e various measure f redso d substrate composition. regression, only percentag f substrato e e smaller Both the low-disturbance subbasin (Wapiti than 0.8accountem 5m significana r dfo t propor- Creek) and the high-disturbance subbasin (Cache tion of the variation in F/ (R2 = 0.75, P = 0.001). Creek) had high percentages of fines smaller than e proportioTh f substrato n e smaller than 0.85 6.35 mm (means, 42.6 and 44.6%). However, the mm was the only substrate measure that differed percentag f substrateo e smaller thas n wa 0.8m 5m significantly (Kruskal-Wallis: H = 14.8, P = 0.04)

100.0 6 3 N= Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013

0 2 4 6 8 10 12 14 16 18 20 22 24 26 Percent spawning gravel 0.074 0.85 2.36 6.35 9.5 12.4 25.4 50.8 Particle size (mm) FIGURE 2.—Relationship between redd density (num- ber/100 m2) and percentage of the substrate consisting FIGURE 3.—Mea ) particlSE + n( 2 e size distribution of potential spawning gravel in 11 reaches of Taylor of cutthroat trout redds sampled in Cache and Wapiti Fork, 1992. creeks, 1992. 774 MAGE. AL T EE

TABLE 4.—Numbe f reddo r s observe coredd dan , mean Trout species composition and density varied fredle index (F/) of substrate composition of cores (range with elevatio appeared nan contributo dt difo et - in parentheses), estimated egg deposition, predicted mean ferences observe numben di f reddro spawnd san - emergence success, and estimated fry production per spawning reach in Cache (C) and Wapiti (W) creeks. s withier basine nth . Segregatio elevatioy nb - nbe tween native cutthroat trout and nonnative rainbow Number of trout and brown trout is common where the species redds Emer- gence co-occur (Bozek and Hubert 1992). Spawning cut- Reach or Ob- Egg suc- Number throat trout used high-elevation tributaries almost subbasin served Cored F, deposition cess of fry exclusively t loweA . r elevations, cutthroat trout Reach numbers declined and the proportion of rainbow WI 57 7 2.8 (1.4-4.8) 20,606 11.6 2.392 trout increased. Ireland (1993) found that juvenile W2 17 3 1.3 (0.7-1.8) 6,679 6.8 455 W3 32 5 2.3(0.5-5.1) 11,565 7.9 916 and adult cutthroat trout were abundant (20 C6 39 4 1.0(0.5-1.6) 13,927 6.8 940 fish/10 reachen i ) 2 0m s immediately downstream C7 39 3 1.6 (0.8-2.7) 13,927 8.3 1.149 of Cache Creek and Wapiti Creek spawning trib- C8-11 74 6 1.0 (0.7-1.4) 26,536 7.8 2,080 C12 33 4 2.0(1.3-2.5) 11,907 10.5 1,254 utarie t wer basine sbu th rese e f rarth o Th t. n ei C13-17 67 4 1.0(0.5-1.6) 24,174 7.9 1,931 close association between redd distributiod an n Subbasin density of juvenile and adults has been reported Wapiii 106 15 2.3(0.7-5.1) 38,850 8.2 3,763 by other investigators (Bear Carlind dan e 1991). Cache 252 21 1.3 (0.5-2.7) 90,471 9.4 7,354 Conversely, rainbo browd wan n trout were found Totals 358 36 1.4 129,321 8.5 11,117 e main-steonlth n i y m Taylor Forlower-eld kan - evation tributarie densitiew lo t sa fish/102 s( 0 m2). significantly highe Cachn ri e Creek (21.6%, versus Ireland (1993) attributed the low densities of rain- 17.1% in Wapiti Creek; U = 2.57, P = 0.01) and trou w browd bo an t n main-ste e trouth n i t m Taylor the FJ was significantly lower in Cache Creek (Ta- Fork to poor habitat conditions. 2.32 = 0.02)= U P bl, ; e4 . Some higher-elevation subbasins, such as Light- ning Creek havy ma e, bee colo extensivr nto d fo e Fry Production spawnin recruitmenty fr d gan . Peak spawning tem- Predicted emergence success was low, averag- perature of 8°C occurred there 1 month later than in gl redd al 8.5 r s%fo (Tabl . Estimate4) e g deg in Cache and Wapiti creeks. We observed newly deposition ranged from 292 to 699 eggs/redd. Of emerge Lightninn i y dfr g Cree earln ki y September the total estimated depositio f 129,32no 1 eggst i , s comparea d with late Jul o earlt y y Augusn i t s estimatewa d that 11,11 wery 7fr e produced- av , Cach Wapitd ean i creeks. Such late-emerginy gfr eraging 24-42 fry/redd. Estimated numbery fr f so may have poor overwinter survival (Smith and produced per redd were similar in Cache and Wa- Griffith 1994). piti creeks t becausbu , e more fish spawnen i d Cutthroa tTayloe trouth n i tr Fork basin appear Cache Creek, total estimate productioy dfr n there to conform to a resident form of life history (Lik- was about twice that of Wapiti Creek. d Grahaan s mne 1988; Riema d Appersoan n n 1989), because we captured very few (3%) large Discussion spawners (>25 characteristi) 0mm f fluviaco l pop- Spawner and Redd Distribution ulations (Shepard et al. 1984b; Rieman and Ap-

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 Previous studies have documented low-order person 1989). Ireland (1993) found that after tributaries as important spawning habitat for cut- spawning, most cutthroat trou Cachn i t e Creed kan throat trout (Rieman and Apperson 1989). We also Wapiti Creek moved relatively short distances found most redds (88%) in first- and second-order (<2,000 m) to summer and winter habitats in sec- tributaries. However, we found high variation in ond- and third-order tributaries. Historically, f theso e e us sites. Nearl l reddal y s (99%) were spawners following a fluvial life history were probably commo e basin th t negativ n bu i n, - in e observe uppee th n di r reache high-elevatioe th f so n teractions with nonnative , habitat disrup- subbasin f Cachso Wapitd ean i creeks even though tion, and angler exploitation in lower portions of these areas represen relativela t y small proportion the basin probably selected against this life history (15%basine th f )o . Substantially fewer redds were form (Liknes and Graham 1988). observe lower-elevation di n tributarie- ba e th n i s sin, although these areas contained some spawning Spawning Habitat suitabld gravelha d ean s water temperatures (for Abundance of potential spawning gravels was a example, Deadhorse and Little Wapiti creeks). key factor influencing redd density at the reach CUTTHROAT TROUT SPAWNING HABITAT 775

TABLE 5.—Mean proportion of fines smaller than 6.35 mm and 0.85 mm and the mean fredle index (F/) of egg pocket salmonin si d redds fro Rocke mth y Mountain region, USA.

Percent fines smaller than: Trout species Location Tim f samplineo g 6.35 mm 0.8m 5m F-t N Reference Cutthroat Montana Near emergence 41.6 17.9 2.0 11 This study Cutthroat Montana 27.4 5.9 13 Weave Fraled an r y ( 1993) Cutthroat Idaho Near emergence 24.4 6.5 7.6 13 R. Thurow and J. King, unpublished data Steelhead Idaho Near emergence 15.4 3.7 10.5 9 R. Thurow and J. King, unpublished data Brook Near spawning 12.1 6.4 31 Youn . (1989al t ge ) Brown Wyoming Near spawning 15.0 3.0 69 Grost etal. (1991)

scale (Figur . Simila3) e r relations between redd their significantly lower concentrations of fines densit availabilitd yan f spawninyo g gravel were (Chapman 1988; Young et al. 1989). The lack of note Copy b d e (1957 r Yellowstonfo ) e cutthroat differences in our egg pocket and non-egg-pocket trout and by Beard and Carline (1991) for brown sample havy sma e been influence inabilite th y db y trout. The lack of strong association between of hollow core sampler preservo t s e verticaeth l spawning gravel abundanc d severaan e l reach stratificatio f redno d samples (Everes . 1980al t e t ) characteristics (gradient, proportio f graveno d an l or by rapid infiltration of fines into redds soon after pea gravel) suggests that local hydrologic features construction Gross . (1991a , al t te ) observed. Time influenced spawning gravel availability. Although of sampling also may result in redd particle size distributioe th f spawninno g substrat s patchewa y differences (Grost et al. 1991). We sampled redds within the Taylor Fork drainage, there appeared ttime o th emergencef eo t a , whereas some othe- rau be adequate habita supporo t t t resident spawners thors sample pocketg deg s nea time rth spawnf eo - because redd superimposition was minimal. The ing (Table 5). Everest et al. (1987) and Lisle and distributio f reddno s apparentl t stronglno s ywa Lewis (1992) recommended sampling redds near influence levele th f y finsdo b e sedimen spawnn i t - e tim th f emergenco e documeno t e t actual incu- ing gravel, because we found no negative relations bation conditions. between redd density and various measures of redd We believe that the erosive geology and asso- substrate composition. ciated sediment loading in Taylor Fork contributed Studie f spawninso g habitat selectio salmoy nb - hige th ho t percentage f fineso reddn i s s compared nids have focuse microhabitae th n do t featuref o s with other sites in the region. The proportion of redds (e.g., Thurow and King 1994). The high vari- small fine studr sou (<0.8yn i redd) 5mm s (17.9%) ation we observed in redd density across the basin averaged two to six times higher than in redds and even among nearby study reaches suggests that elsewhere. These small fine readile ar s y removed bioti abiotid can c factors operatin t largega r scales during redd excavation (Youn . 1989al t ge ; Grost e alsar o importan describinn i t g spawning habitat . 1991)al t t apparentle ye , y reaccumulat higo et h suitability (Beard and Carline 1991; Bozek and levels before emergence, attesting to the high fine-

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 Rahel 1991) r datOu a. suggest that temperature, sediment load in spawning reaches. High propor- fish density, and availability of spawning gravels tions of fines within the undisturbed Wapiti Creek interacte influenco dt e redd density - acrosen e sth subbasin further illustrate the highly erosive nature tire basin. of the siltstone geology of the basin. Proportion f smalso l fine reddn i s s were highly Redd Characteristics variable, accountin significana r gfo t proportiof no Taylor Fork redds contained highese somth f eo t variation (R 0.75= 2 0.001 = F, e ,P th . Smalln )i - proportion f fino s e sediments observeg eg n i d scale differences in sediment transport and infil- pockets of salmonid redds in the Rocky Mountain tration, particularly in streams with high loads of region (Table 5). Several factors may account for fine sediment, commonly cause high variation ni differences among sites, including sampling meth- smaller particle sizes among redds (Lisle and Lew- , geologyod sedimend an , t transport. Other inves- is 1992). Everes . (1987al Chapmat d te )an n (1988) tigators, using freeze core samplers, have docu- emphasized that predictin effecte gth f fino s e sed- mente e uniquth d e qualitie g pocketeg d f o san s iments on salmonid embryo survival requires ac- 776 MAGEE ET AL.

curate samplin pocketg eg greatef d go san r atten- respectively) shown to adversely affect salmonid tio o measurint n g temporal change substratn i s e emergence success (McNei d Ahnelan l l 1964; conditions during incubation findingr Ou . s suggest Shepar . 1984aal t de ; Cederhol Reid man d 1987; that accurate prediction f effecto s f fineo s s also Reise Whitd an r e 1988). require knowledg f spatiao e l variabilit f redo y d Previous studies have demonstrated that reduced substrate conditions and knowledge of the influ- emergence success from high sedimentation can enc f basio e n geolog n substratyo e size compo- juvenilw resullo n i t e densitieadulw - lo re td an s sition (Everes t ale t . 1987; Lisl Lewid ean s 1992). cruitment (Cederholm and Reid 1987; Scrivener r datOu a further suggest that land disturbance and Brownlee 1989). Despite high fine-sediment exacerbated the high levels of fine sediments in levels and low predicted emergence success of fry Taylor Fork redds. Previous reports have noted that from Taylor Fork redds, there was little supporting small finemose th s te (<0.8sensitivar ) 5mm - ein evidenc recruitmentw lo r efo . Juvenil aduld an e t dicatoe leve th f disturbanc o f l o r watershedn i e s densitie pooln i s f Cacho s e Creek (43.5 fish/100 (McNei d Ahnelan l l 1964; Cederhol d Reiman d m2) and Wapiti Creek (32.5/100 m2) are notably 1987; Young et al. 1991; see also Scrivener and higher than bote averagth h e (9.2/100 md 2an ) Brownlee 1989). The high-disturbance Cache maximum (26.1/100 m2) density in pools reported Creek subbasin had significantly greater propor- for seven other westslope cutthroat trout popula- tions of small fines in redds than the relatively tion Idahn si Montand oan a (Shepar . 1984bal t de ; undisturbed Wapiti Creek subbasin. Cederholm Ireland 1993). Beard and Carline (1991) found ev- and Reid (1987) found that road surface erosion idence for recruitment limitation of a resident was the major contributor to higher proportions of brown trout population as reflected by low redd small fines in redds in disturbed subbasins of the density, high ratio f maturo s e female o reddst s , Clearwater River, quantift no d di ye .W high rate f redo s d superimposition embryw lo , o the source of small fines in our study; however, survival, and low juvenile and adult density. We road surface t responsiblerosiono s - in wa n r fo e predicte embryw dlo o surviva r studyou n i l; how- creased sedimentatio f Cacho n e Creek because ever we observed high numbers of redds, low redd most surveyed spawning areas were located up- superimposition, low ratio of females to redds, and stream from roads. Streambank Cachn i s e Creek high juvenil aduld ean t density. were degraded (Snyder et al. 1978) and we spec- Several factors may buffer the effects of high ulate that increased erosion resulting from live- sedimentatio f reddsno . Hig mortality hfr coms yi - stock disturbanc primara e b y y ema sourc - el f eo mon in resident salmonids, but low fry densities evated small fines; cattl commoe ear streamn no - may resul n decreasei t d competition, increased banks and within the stream channel during in- growth, and compensatory survival (McFadden cubation (Magee 1993). 1969). High fry survival is suggested for Taylor Fork cutthroat trout populatioA . n estimat- ju f eo Is Fine Sediment Limiting Recruitment? venil aduld ean t cutthroat trou Cachn i t e Creen ki High levels of fine sediment in Taylor Fork redds fall 1994 (our unpublished data; N = 3,726, 95% le vero predictiondt w ylo f overalso l embryo sur- confidence interval = 2,894-4,588) was 50% of vival (mean, 8.5%) and numbers of fry produced e predicteth d number producey fr f o s Cachn di e per redd (mean, 30). Our estimates of emergence Creek in 1992 (Table 3; N = 7,354). Typical annual Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 success were mad y usinb e g proportion f fino s e mortality rate r cutthroasfo t trou estimatee ar t t da sediment less tha ndiametern i 6.3 m 5m , basen do d 30-50an y r olde%fr fo r rfo age-classe% 95 s the experiments of Weaver and Fraley (1993) with (Rieman and Apperson 1989). artificial egg pockets planted in stream substrate. Variatio substratn i n e quality among redds (F/ Actual emergence succes havy ma se been even = 0.48-5.14) also could lead to underestimates of lower than predicted due to high proportions of fry recruitment. With a large number of redds, the small fines (<0.85 mm). Small fines are more det- small proportion with high substrate quality yma rimental to survival of incubating eggs (Reiser and produce enoug compensato t y emer-hfr w lo r efo White 1988) and intrude deeper into spawning gence success in remaining redds (Lisle and Lewis gravels (Beschta and Jackson 1979) than coarser 1992). Reductions of fines during redd building fines. The proportions of both small (<0.85 mm; may counterac effecte th t f higo s h sediment loads mean, 17.9%) and coarse (<6.36 mm; mean, (Everest et al. 1987; Chapman 1988), but this is 44.3%) fine Taylon si r For considerable kar y above probably not an important factor for Taylor Fork the level r thesfo s e particle sizes 30%d (10an , cutthroat trout becaus smale th f leo siz spawnf eo - CUTTHROAT TROUT SPAWNING HABITAT 777 e verth yd higan s h er proportio f finess o ni t I . ing reache highed sha r redd densities than others. possible thapopulatioe th t evolves nha d othe- re r However, knowledg basif eo n scale variables (e.g., productive adaptation compensato st nate th -r efo temperature, fish species compositio densityd nan ) urally high sedimen e tbasi loath n i d (e.g., dif- was also important. For example, extrapolation of ference sizeg eg ,n i sfecundity depthg eg , ; Holtby dat n redo a d density obtained fro ma subse f o t and Healey 1986) r example Berghn Fo . de n eva , Taylor Fork reaches without more general knowl- and Gross (1989) reported that small salmonid edge of redd distribution across the entire basin eggs survive better than large eggs in highly sedi- would resul higa n i ht degre f erroreo . High vari- mented substrates. Small, resident cutthroat trout atio fisn ni h populatio habitad nan t characteristics in Taylor Fork therefor havy ma ee adaptee th o dt across watersheds illustrate importance sth - us f eo inherently high fines and have better emergence ing multiple scale defino t s e suitable habitad an t success than the values we applied from Weaver to design appropriate monitoring programs (Fris- and Fraley (1993). Additional measurementf o s . 1986al sel t e l ; Hanki Reeved nan s 1988; Bozek egg deposition, fecundity, and emergence success Rahed an l 1991). are needed to evaluate compensatory mechanisms. r resultOu s suppor theore th t y that resident sal- Acknowledgments monid populations are typically not limited by re- This study was supported by funds provided by duced spawning success and that recruitment is the U.S. Forest Service Intermountain Research probably limited by available rearing habitat, ex- Station. Jack Mclntyre was instrumental in its cept when seedin s veri g(Everesw lo y t ale t . fundin d developmentan g e thanW . k Brucy Ma e 1987). However, the high levels of fine sediments for facilitating substrate analyses e IrelandSu ; , in Taylor Fork redds, among the highest reported Mike Jones Jennifed an , r Staple r fielfo s d assis- literature inth r salmonidefo s (Tabl ; Chapmae5 n tance Robd an ; b Lear r providinyfo - g ge dat n ao 1988; Scrivener and Brownlee 1989; Beard and netic status. Review Alay b s n Barta, Mason Bry- Carline 1991), suggest that fine sediments may be , Brad Shepard Weaverm To ,Whiteb Bo ,d an , approachin levee gth l that would limit recruitment. anonymous reviewers significantly improved the The loss of larger, more fecund fluvial cutthroat manuscript. trout together with degradation of remaining hab- itats may threaten the persistence of cutthroat trout References basine inth . Populations with reduced geneti- cdi versity and those in degraded habitats will be less Alt, D., and D. W. Hyndman. 1986. Roadside geology resilient and at increased risk of local f Montanao . Mountain Press. Missoula, Montana. Beard, T. D., and R. F. Carline. 1991. Influence of from both stochasti deterministid an c c processes spawnin othed gan r stream habitat feature span o s - (Riema Mclntyrd nan e 1993). tial variability of wild brown trout. Transactions of A potential confounding r factostuds ou n ywa i r the American Fisheries Society 120:711-722. the genetic introgression of Taylor Fork cutthroat Beschta, R. L., and W. L. Jackson. 1979. The intrusion trout. Littl knows ei n abou phenotypiw ho t c traits of fine sediments into a stable gravel bed. Journal influencee ar levee th f introgressio o y l db n (Leary Fisheriee oth f s Research Boar Canadf do a 36:204- 210. . 1995)al t e . Comparative studie f habitao s e us t Bjornncoauthorsx si d . CT ,an , . 1977. Transporf o t survivad an f cutthroao l t trout having varying lev- granitic sedimen streamn - effects i t it in d n o san s Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 f introgressioo s el e neededar n , given that pure sects and fish. University of Idaho, Forest, Wildlife, populations represent only a small proportion of and Range Experiment Station, Bulletin 17, Mos- remaining populations (Liknes and Graham 1988: cow. Lear. 1995)al t ye . . HubertA . BozekW . d A. M 1992an ,, . Segregatiof no resident trou streamn i t predictes sa threy db e hab- Spatial Variation itat dimensions. Canadian Journa f Zoologo l : y70 866-890. Our results concur with studies illustrating that . RahelJ Bozek . F . d A.. M ,an ,1991 . Assessing habitat different spatial scale importane sar charactern i t - requirement f youno s g River cutthroat izing fish habitat requirement assessind san g land trout by use of macrohabitat and microhabitat anal- use effects (Frissell et al. 1986; Bozek and Rahel yses. Transactions of the American Fisheries So- 1991). Bozek and Rahel (1991) noted that many ciety 120:571-581. Cederholm, C. J., and L. M. Reid. 1987. Impact of forest fish habitat studies are conducted at the micro- managemen n coho t o (Oncorhynchus ki- habita r reaco t h spatial scalesr studyou n I ., reach sutch) population Clearwatee th f o s r River, Wash- measurements helped explai certaiy nwh n spawn- ington: a project summary. Pages 373-398 in E. 778 MAGEE ET AL.

. GundyT Sald an o, editors. Streamside manage- ural stream: a simulation approach. Canadian Jour- ment: forestry and fishery interactions. University nal of Fisheries and Aquatic Sciences 49:2337- of Washington, Colleg f Foreseo t Resources, Con- 2344. tributio , Seattlen57 . Magee, J. P. 1993. A basin approach to characterizing Chapman . 1988W . D ,. Critical revie variablef wo s used spawning and fry rearing habitats for westslope cut- defino t e effect finef so reddn si largf so e salmonids. throat trou sediment-rica n ti h basin, Montana. Mas- Transactions of the American Fisheries Society 117: ter's thesis. Montana State University, Bozeman. 1-21. McFadden, J. T 1969. Dynamics and regulation of sal- . Cope1957B e . choicO ,Th . f spawnino e g sitey b s monid populations in streams. Pages 313-329 in T. cutthroat trout. Proceeding Utae th hf o sAcadem y G. Northcote, editor. Symposium on salmon and of Sciences, Arts Letterd an , s 34:73-79. trou n streamsi t . Universit f Britiso y h Columbia, . McLemoreE . . WardEverestF C . , J H. .d . 1980F an , . Vancouver. An improved tri-tube cryogenic gravel sampler. McNeil, W. J., and W. H. Ahnell. 1964. Success of pink U.S. Forest Service Research Note PNW-350. salmon spawning relativ sizo e t f spawnineo d gbe Everest fiv. H.d F , an e, coauthors. 1987. Fine sediment materials. U.S. Fish and Wildlife Service Special and salmonid production—a paradox. Pages 98-142 Scientific Report Fisheries 469. . Cundy T . Sal inE d oan , editors. Streamside man- Overton, C. K., S. P. Wollrab, B. C. Roberts, and M. A. agement: forestr fisherd an y y interactions. Univer- Radko n pressI . . R1/R4 (Northern/Intermountain sit f Washingtono y , Colleg f Foreso e t Resources, Region) fish and fish habitat standard inventory pro- Contributio , Seattlen57 . cedures. U.S. Forest Service General Technica- Re l Frissell, C. A., W. J. Liss, C. E. Warren, and M. D. port INT. Hurley. 1986. A hierarchical framework for stream Plans, W. S., W. F. Megahan, and G. W. Minshall. 1983. habitat classification: viewing streams in a water- Methods for evaluating stream, riparian, and biotic shed context. Environmental Management 10:199- conditions. U.S. Forest Service General Technical 214. Report INT-138. . WescheA . Grost Hubert. A T . . T.d R W ,., an , 1991. Platts, W. S., R. J. Torquemada, M. L. McHenry, and C. Description of brown trout redds in a mountain K. Graham. 1989. Changes in salmon spawning stream. Transactions of the American Fisheries So- and rearing habitat from increased delivery of fine ciety 120:582-588. sediment to the South Fork Salmon River, Idaho. Reeves. H . G d . Hankinan 1988 , G. . EstimatinD , g total Transactions of the American Fisheries Society 118: fish abundanc d totaan e l habitat are n smali a l 274-283. streams base n visuao d l estimation methods- Ca . Reiser, D. W, and R. G. White. 1988. Effects of two nadian Journal of Fisheries and Aquatic Sciences sediment size-classes on survival of steelhead and 45:834-844. chinook salmon eggs. North American Journaf o l HealeyC. Holtby M. .B. L. ,and 1986, . Selectionfor Fisheries Management 8:432-437. adult siz n femali e e coho salmon (Oncorhynchus Rieman, B. E., and K. A. Apperson. 1989. Status and kisutch). Canadian Journal of Fisheries and Aquatic analysis of salmonid fisheries: westslope cutthroat Sciences 43:1946-1959. trout synopsi analysid san f fisherso y information. Ireland . 1993C . S , . Seasonal distributio habitad nan t Idaho Department of Fish and Game, Federal Aid f westslopo e us e cutthroat trousediment-rica n i t h in Sport Fish Restoration, Project F-73-R-11, Boise, basin in Montana. Master's thesis. Montana State Idaho. University, Bozeman. Rieman, B. E., and J. D. Mclntyre. 1993. Demographic Leary, R. F, F. W. Allendorf, and G. K. Sage. 1995. and habitat requirements for conservation of bull Hybridization and introgression between introduced trout. U.S. Forest Service General Technical Report and native fish. Pages 91-101 in H. L. Schramm, INT-302. Jr. and R. G. Piper, editors. Uses and effects of Scrivener, J. C., and M. J. Brownlee. 1989. Effects of

Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013 cultured fishes in aquatic ecosystems. American forest harvesting on spawning gravel and incubation Fisheries Society, Symposium 15, Bethesda, Mary- surviva f chuo l m (Oncorhynchus keta)d cohan o land. salmo . kisutch)n(O Carnation i n Creek, Britis- hCo Liknes . 1984A presene . G , Th . t statu distributiod san n lumbia. Canadian Journa f Fisherieo l Aquatid san c of the westslope cutthroat trout (Salmo clarki lewisi) Sciences 46:681-696. Continentae easwesd th f tan to l Divid Montanan ei . Shepard, B. B., S. A. Leathe, T. M. Weaver, and M. D. Montana Department of Fish, Wildlife, and Parks, Enk. 1984a. Monitoring level f fino s e sediment Helena. within tributaries to , and impacts of Liknes, G. A., and P. J. Graham. 1988. Westslope cut- fine sedimen buln o t l trout recruitment. Pages 146- throat trout in Montana: life history, status, and 156 in E Richardson and F. H. Hamre, editors. Wild management. Pages 53-60 in R. E. Gresswell, ed- trout III. , Vienna, Virginia. itor. Status and management of interior stocks of Shepard, B. B., K. L. Pratt, and P. J. Graham. 1984b. cutthroat trout. American Fisheries Society, Sym- Life histories of westslope cutthroat trout and bull posiu , mBethesda4 , Maryland. trout in the upper basin, Montana. Lisle. LewisJ d . E.T ,an , . 1992. Effect f sedimeno s t Montana Department of Fish, Wildlife, and Parks, transport on survival of salmonid embryos in a nat- Kalispell. CUTTHROAT TROUT SPAWNING HABITAT 779

. GriffithS . J d .an Smith, 1994W . .R , Surviva f raino l - trou froy fr tm varying substrate compositiona n i s trouw bo t during their first wintee Henryth n i r s natural stream channel. North American Journaf o l Snake Forth f keo River, Idaho. Transactione th f so Fisheries Management 13:817-822. American Fisheries Society 123:747-756. Wolman, G. M. 1954. A method of sampling coarse Snyder, G., J. Black, P. Swetik, and H. F. Haupt. 1978. river-bed material. Transactions, American Geo- Modeling forest water quality of the upper Gallatin physical Union 35:951-956. and Madison watersheds, using the benchmark . Wesche sys A . Hubert . A - T . d W . Young, an , 1989K. . .M , tem. U.S. Forest Service, Intermountain Forest and Substrate alteratio spawniny b n g brook troua n i t Range Experiment Station, Forestry Sciences Lab- southeastern Wyoming stream. Transactions of the oratory, Moscow, Idaho. American Fisheries Society 118:379-385. Thurow, R. F, and J. G. King. 1994. Attributes of Yel- . WescheYoung A . Hubert . A T . . K. d M ,W , . an , 1991. lowstone cutthroat trout redds in a tributary of the Selection of measures of substrate composition to , Idaho. Transaction Americae th f o s n estimate surviva emergenco t l f salmonideo td o san Fisheries Society 123:37-50. detect changes in stream substrates. North American . GrossR . M . d 1989Berghevan an n, de R . . NaturaE , l Journa f Fisherieo l s Management 11:339-346. selection resulting from female breeding competi- Zar, J. H. 1984. Biostatistical analysis, 2nd edition. a Pacifitio n i n c salmon (coho: Oncorhynchus- ki Prentice-Hall, Englewood Cliffs Jerseyw Ne , . sutch). Evolution 43:125-140. Weaver, T. M., and J. J. Fraley. 1993. A method to Received March 31, 1995 measure emergence success of westslope cutthroat Accepted April 15, 1996 Downloaded by [Montana State University Bozeman] at 11:50 01 July 2013