John J. Fraley and Bradley B. Shepard, MontanaDepartment of Fish,Wildlife and Parks P.O.Box 67, Kalispell,Montana 59903

Life History,Ecology and PopulationStatus of MigratoryBull Trout (salvelinusconfluentus) in the FlatheadLake and RiverSystem, Montana

Abstract

Life history, ecology, and population trends ofmigratory bull trout (Sabelinus confluentus) were investigated in the Flathead Lake and River system of northwest Montana and southeast British Columbia. We conducted these studies to obtain informa- tion to manage the species in light of threats posed by timber harvest, hydropower development, and a proposed coal mine. We estimated that about half the adult bull trout in Flathead Lake enbarked on a spawning migration from May through July, swimming 88-250 km to reach tributaries of the North and Middle Forks of the Flathead River. Bull trout entered the tributaries when water temperatures dropped below l2oC, and spawned from late August through early October after water temperatures were below 9oC. They spawned in areas of tributaries with low gradient, loosely compacted gravel, groundwater influence, and cover. After spawning, females left the tributaries and returned to the lake sooner than males. Most spawners were six or seyen years old and they averaged 628 mm in length. Juveniles were found close to the substrate in streams with summer maximum temperatures less than l5oC. Juveniles migrated out of the tributaries to the river system from June through August, at age I (18%), lI(49yo),IlI(32%), and IV (l%). Population status was monitored through redd counts and estimates of juvenile abundance in natal tributaries. The population may be limited by quantity and quality of rearing and spawnrng habitat, and spawning escapement. Specific requirements for spawning and rearing habitat, and general sensitivity of each life stage, make the bull trout an excellent indicator of environmental disturbance.

Introduction Flathead River Basin Studies (Graham et al. 1980, Fraley et al. 1981, Shepard et al. 1982, The bull trout (Salttelinus confluentus) is the 1984b, Graham et al. 1982, Fraley and Graham largest species of fish native to the Flathead 1982, Graham and Fredenbergl9S2,Leathe and drainage,attaining a length of nearly one meter Graham 1982). We studied bull trout age and and a weight of l0 kg. The bull trout inhabiting growth both in the lake and in the river system, the inland waters of northwestern North America harvest by anglers, the adult spawning migration, is considered a separate speciesfrom the smaller, spawning site selectionand use, and the densities, coastal Dolly Varden (Saluelinus malma) habitat selection, and emigration of juveniles (Cavender 1978).The bull trout population in the growing in tributaries. Methods included tagging, Flathead system is largely migratory, growing to gillnetting, stream trapping and electrofishing, maturity in lakes and migrating through the river creel survey, otolith and scale analysis, redd system and into the tributaries to spawn. Juve- counts, and substrate analysis (Graham and niles live in tributary streams from one to four Predenberg 1982, Shepard and Graham l9B3). years before migrating to the lakes. In this paper we summarize our findings on Much information has been published con- the life history, ecology, and population status cerning the life history of coastal Dolly Varden of adfluvial bull trout in the Flathead Lake (e.g.,Blackett 1968,Armstrong and Morton 1969, and inlet river system and compare our information Armstrong and Morrow 1980,Balon 1980).Pub- to the results of other investigators. lished information on the bull trout is limited. McPhail and Murray (1979), Leggett (1969), and Allan (1980) studied various aspects of the life StudyArea history of bull trout in British Columbia and The Flathead Lake and River system is a head- Alberta. Gould (1987) described the character- water drainage of the Columbia River Basin (Fig- istics of larval bull trout. ure l). Flathead Lake is a large oligomesotrophic The Montana Department of Fish, Wildlife lake with a surface area of 476 km2 and a mean and Parks has studied the bull trout population depth of 32.5 m (Potter l97B). The upper 3 m in the Flatheaddrainage since 1953(Block 1955, of Flathead Lake is regulated by Kerr Dam, con- Hanzel 1976).More intensive work was under- structed on the outlet in 1938. The Flathead taken from 1979-1984during the EPA-sponsored River enters the north end of the lake. This study

Northwest Science,Vol. 63, No. 4, l9B9 133 UPPERFLATHEAD RIVER BASIN

CANAOA USA

STUOY A REA

+ I N HUNGRY HORSE I DAM O 16km

BIG FORK DAM FLATHEAD LAKE

Figure 1. The upper Flathead River Basin. The 28 tributaries shownwere used by spawningbull trout.

134 Fraley and Shepard was conducted in the upper Flathead Basin which age of maturity was reported for bull trout in includes Flathead Lake and the river systemup- Arrow Lakes, British Columbia (McPhail and stream from Flathead Lake. Murray 1979). The South, Middle and North forks drain The diet of bull trout in the lake consisted areasof approximately equal size in portions of almost exclusively of fish. Whitefish species and the Great Bear and Bob Marshall wildernesses, yellow perch (PercaJlaoescans) were the most im- Glacier National Park and the Flathead National portant food items, followed by kokanee(Onco- Forest. The upper North Fork drains southern rhynchus nerka) and,nongame species(Table l). British Columbia. The South Fork is regulated Small bull trout have been found to feed inci- by Hungry Horse Dam, located 8 km above its dentally on Mysis in Flathead Lake. Mysis relicta mouth. The Swan River enters Flathead Lake was discoveredin Flathead Lake in l98l and den- near the mouth of the Flathead River. Bull trout sities increased dramatically through 1986. coexist with 23 other species of fish in the Kokanee were the major food item for bull trout Flathead Lake and River svstem (Leathe and in Pend Oreille Lake, Idaho (Jeppsonand Platts Graham 1982). 1959),while whitefish were the major food in Up- Most bull trout that spawn in the North and per Priest Lake (Bjornn l96l). Middle Fork drainages mature in Flathead Lake, The annual growth increment for bull trout but fish maturing in large lakes of Glacier Na- in Flathead Lake, based on analysis of scales, tional Park may spawn in some tributaries. There ranged from 60-132mm (Table 2). Back calcula- are a few populations of bull trout in tributaries tions of length at annulus formation were made of the North Fork that spend their entire lives from 1,813 scale samples.Aging was checked in the streams. with otoliths from 451 of the fish. Agreement of Bull trout originally used the tributaries of aging between otoliths and scalesranged from all forks of the Flathead and the Swan rivers. The 100 percent for fish zero to three years of age, construction of Bigfork Dam in 1902 blocked bull to 52 percent for older, mature fish. Growth of trout migrations into the Swan River. Limited lake-residentfish was relatively constant after age numbers of bull trout move downstream from the IV. Growth rates of bull trout in Flathead Lake Swan drainage via a marginal fish ladder, as were similar to those reported for Priest and Up- evidenced by tag returns. Hungry Horse Dam, per Priest Lakes, Idaho (Bjornn l96l). a 164.6-mstructure which was closed in 1951, Not all mature bull trout spawnedannually. blocked all movements of bull trout into the Adult-size fish were relatively less abundant in South Fork drainage and probably resulted in a the lake during the summer and fall, as compared substantial reduction of the population in to the spring. It appearedthat 38 to 69 percent Flathead Lake. (average57To)left the lake eachspring and sum- The 28 tributaries used by spawning bull mer to spawn.The frequency of successiveyear trout in the North and Middle Fork drainages spawning varied by age and sex (Leathe and (Figure l) are characterized by gravel-rubble Graham 1982).Alternate year spawning has been substrate, low flows of 0.057-1.70m3isec, and reported for inland Dolly Varden char (Arm- maximum summer water temperaturesless than strong and Morrow 1980). 150c. Upstream Migration Resultsand Discussion Bull trout maturing in Flathead Lake began their LifeHistory spawning migration into the river system during April and moved slowly upstream, arriving in the Lake Residence North and Middle forks during late June and Ju- Bull trout populations residing in Flathead Lake ly. They traveled more than 250 km to spawn in were found to include recently arrived juveniles some North Fork tributaries in British Colum- from the Flathead River system,subadult fish less bia. The shortest distance traveled from Flathead than about 450 mm in length, and mature fish Lake was 88 km to the mouth of Canyon Creek five to six years or more in age. Most bull trout in the North Fork drainage. Observationsand tag in Flathead Lake matured at age VL A similar returns from 1974-1982indicated that adult bull

Life History and Ecologyof Migratory Bull Trout r35 TABLE l. Composition by number, weight, and frequency of occurrence and calculated index of relative importance (lRI, George and Hadley 1979) for major food items in the stomachs of 95 bull trout collected between November and January, 1979, 1980 and l98l in Flathead Lake.

Wet Index of Relative Item Number (%) weight-g. (%) Importance (lRI)

Pygmy whitefish (2.4) 37.0 (4.0) J-Z Lake whitefish l (0.s) 104.1 (r r.2) 4.3 Mountain whitefish I (0.5) 24.3 (2.6) 4.4 Unidentified whitefish l1 (s.3) 28t.2 (30.3) 15.0 Total whitefish l8 (8.7) 446.6 (48.l ) 23.5

Kokanee 2 (1.0) 82.8 (8.e) 4.0 Unidentified trout/salmon 2 (l.0) t3.2 (1.4) 1.5 Total trout/salmon I (l.e) 96.0 (10.3) 5.r

Sculpin 3 (1.s) 7.6 (0.8) l.B Redside shiner (2.4) 15.0 (1.6) 2.Q Peamouth I (0.5) 3.6 (0.4) 0.7 Sucker 2 (1.0) 74.4 (8.0) Yellow perch 83 (40.3) 105.1 (11.3) 24.6 Total nongame 94 (4s.6) 205.7 (22.r) 31.0 Unidentified fish 90 (43.7) t8l.l (le.s) 4t.4

TABLE 2. Back calculated lengths at annulus formation of bull trout in the upper Flathead Basin (n in parentheses). Calcula- tions were made based on methods in Hesse (1977).

Total length (mm) at annulus

Drainage III IV VI VII VIII IX

Aduhs and Juoeniles Upper Flathead(1968-81) 66 121 196 292 385 475 566 657 73r (l,Br3) (l,s3B) (r,16r) (927) (66e) (34e) (t2e) (32) (4) FlatheadLake (1968-81) 68 r29 204 29r 384 472 566 658 73r (93r) (931) (e28) (8s3) (603) (2er) (r02) (28) (4) North Fork of the 73 lr7 165 301 440 538 574 FlatheadRiver drainage (s33) (306) (60) (12) (B) (7) (3) (197s-81)

Middle Fork of the 52 100 165 297 399 488 567 o)) FlatheadRiver drainage (34e) (300) (t72) (61) (s7) (s0) (24) (4) (r980-81)

Juteniles Only North Fork drainage lta 155 228 (s2s) (2eB) (s2) (4)

Coal Creek /J 124 202 323 (l4s) (62) (23) (14) Red MeadowCreek 65 ll3 168 360 (l4s) (l l3) (2e) (7) Trail Creek ll9 158 228 (473) (264) (46) (4) Whale Creek 56 98 t39 (s2) (34) (6)

136 Fraley and Shepard trout remained at the mouths of spawning Bull trout spawners selected areas in the tributaries for two to four weeks during which stream channel characterized by gravel time feeding was thought to be limited. substrates, low compaction and low gradient Based on observations at stream trapping (Table 4). Groundwater influence and proximity sites,adult bull trout entered tributary streams to cover also were important factors influencing at night from July through September; the ma- spawning site selection. These relatively specific jority entered in August. Becausemost bull trout requirements resulted in a restricted distribution moved through the traps in pairs, we believe bull of spawning in the Flathead drainage. Bull trout trout formed pairs near the mouth of the spawn- from Flathead Lake spawned in only 28 percent ing tributary. Bull trout which entered the spawn- of the 750 km of available stream habitat accord- ing tributaries were generally not in final spawn- ing to basin-wide surveys from 1980-1982. ing condition, but held in the tributaries for up to a month or more in deeper holes or near log TABLE4. Meanmeasurements of physical habitat variables or debris cover before spawning. Similar pre- in 34 stream reaches where no redds were located, 29 reaches where redd frequency averaged 1.2 spawning behavior and spawning timing was redds/km (low), and 3l reaches where redd fre- reported for bull trout in Mackenzie Creek quency averaged 6.9 redds/km (high). (McPhail and Murray 1979)and John Creek (Leg- gett 1969) in British Columbia. Redd frequency categories Most bull trout spawners in the North and Middle Forks were six or seven years of age Parameter Low High (Table 3) whereas most spawners in the Swan Stream order 3.0 3.1 3.6 systemwere five or six years old (Leathe and Enk l98s). D-90 (cm; the size of material 51 37 33 larger than 90% of the substrate) TABLE 3. Age of bull trout spawnersin the Flatheadsystem. Gradient (percent) 3.2 r.8 L5 Boulder (percent of substrate) 16 12 I0 Percent by Age Gravel-Cobble 54 62 62 Stream 678 (combined percent of substrate) NorthFork Flathead River 24 39 34 l9sa = 4l) High quality pool fl (percent of stream) Middle Fork Flathead River 10 48 35 l0 Overhang cover (percent) l98l (N = 3I) Total cover (percent) Swan River 33 35 23 9

Most bull trout spawned during September and TABLE5. Averagetotal lengths of adultbull trout spawners early October in the Flathead River system, as in the Flathead drainase. did adfluvial bull trout in Idaho (Heimer 1965) and British Columbia (McPhail and Murray Average Number 1979).Initiation ofspawning in the Flathead ap- Stream Year Length (mm) of Fish peared to be related largely to water temperature, North Fork t979 638 36 although photoperiod and streamflow probably r977 645 32 also played a part. Spawning began when water 1953 ol / lo) temperatures dropped below 9-l0oC. McPhail Middle Fork 1980 618 and Murray (1979) reported that 9oC was the 1957 622 87 threshold temperature for the initiation of spawn- Both Forks 1975 628 46 ing in Mackenzie Creek, British Columbia.

Life History and Ecologyof Migratory Bull Trout r37 remained near the redd an averageof two weeks upper FlatheadRiver Basin.Juveniles were pres- after spawning.Bull trout redds in the Flathead ent in many reachesthat were not used by adult drainage averaged2.0 m long x 1.0 m wide, and spawners; they apparently swam upstream to sometimesoverlapped. Block (1955)observed one these sectionsto grow. Distribution also was in- male spawn with three females in succession;the fluenced by water temperature as juvenile bull size of the redd expanded each time. McPhail and trout were rarely observedin streamswith sum- Murray (1979),Leggett (1969),and Block (1955) mer maximum temperatures exceeding l5oC. provided detailed descriptions of spawning Oliver (1979),Allan (1980)and Pratt (1984)also behavior and spawning site activities. After reported that bull trout distribution was affected spawning, the spent adults moved out of the by temperature. tributaries and downstreamto the lake, possibly Young-of-the-year bull trout were generally feeding on mountain whitefish (Prosopium found in side channel areas and along the stream uilliamsoni) during the journey. margins in Flathead tributaries. Blackett (1968) Fecundity varied with fish size, averaging reported a similar habitat preference for juvenile 5,482 eggs per female for a sample of 32 adults Dolly Varden char in southeastAlaskan streams. averaging 645 mm. One female bull trout McPhail and Murray (1979)found young-of-the- weighed l5 pounds and contained 12,000eggs. year bull trout in areas of low velocity near Bull trout in Arrow Lakes, British Columbia, were stream edges. smaller and contained fewer than 2,000 eggs per Densitiesof bull trout juveniles in Flathead female (McPhail and Murray 1979).The sex ratio tributaries were greatestin pools, and lower but of bull trout spawnersaveraged 1.4 females per generally similar in runs, riffles and pocketwater male in Trail Creek in the North Fork drainage, habitat. Juvenile bull trout were found closely and 1.37females per male in the Swan drainage. associatedwith stream substrate. Pratt (1984) studied microhabitat preferencesin the Flathead Incubation and Emergence drainage and reported that juvenile bull trout After depositionby early October,bull trout em- (less than 100 mm) usually remained near the bryos incubated in the redd for several months stream bottom, closeto streambedmaterials and before hatching in January. The alevinsthen re- submerged fine debris. Juveniles 100 mm or mained in the gravel and absorbedthe yolk sac, longer also remained near cover, including larger with the first fry appearing in electrofishing instream debris.As the juvenile bull trout grew, samplesin mid April. Emergence occurred ap- they becameless associated with the streambed. proximately 200 days after egg deposition.New- During stream residence,juvenile bull trout ly emerged fry averaged 23-28mm and more than were opportunistic feeders, mainly ingesting doubled their length during the first summer of aquatic invertebrates (especially Diptera and growth (see Table 2). Ephemeroptera)in similar percentagesas they Weaver and White (1985) found that incuba- were available in the stream (Fraley et al. l98l). tion time was dependent on temperature. Bull Bull trout larger than I l0 mm also ate small trout trout eggs required l13 days (340 temperature and sculpin. units) to 50 percent hatch in Coal Creek, a tribu- Snorkeling estimates of juvenile bull trout tary of the North Fork of the Flathead River. The densitiesin Flathead drainage tributaries aver- fry emerged (635 from the gravel223 days temp- aged 1.5 fish/100 m2 of stream surface area erature units) after egg deposition. Intergravel (range: 0.1-7.1).Juvenile bull trout are difficult temperatures during (Octo- the incubation period to observe becauseof their close associationwith ber-March) in Coal Creek ranged from 1.2-5.4oC. the stream bottom, so these numbers are prob- Survival to emergencein Coal Creek averaged53 ably underestimates. Electrofishing estimates percent. McPhail (1979) and Murray reported the high as 15.5 fish/100 m2 in certain best survival of bull trout embryos at 2-4"C. :lrl*"*.* Most juvenile bull trout in the Flathead drain- fuuenile Occurrence and Emigration age remained in the tributaries for one to three Juvenile bull trout were present in about half of years before emigrating to the river system.Of the stream reachessurveyed during studies in the 246 juvenile bull trout captured in downstream

138 Fralev and Shenard migrant traps placed in three tributaries to the which successfullyspawned in the river-tributary North and Middle forks, about half (49%) were system.We converted the redd counts to approx- age II, a third (32%) age III, and l8 percent age imate fish numbers by making the following I (Table 6). Only I percent of the emigrantswere assumptions: l) 75 percent of all redds were age IV. The ages of emigrating juveniles were located, and 2) an average of 3.2 spawners similar in Idaho and British Columbia (Bjornn entered the tributary for each completed redd. I961, Oliver 1979, McPhail and Murray 1979). From partial trapping results on several tribu- The average lengths at annulus formation ofAge taries in 1977-1981,we estimated a spawner:redd I, II, and III juvenile bull trout in tributaries of ratio of 3.2:1. In 1953, 55 bull trout entered the North Fork Flathead were 73, ll7 and 155 Trail Creek and constructed l8 redds for a mm, respectively(Table 2). spawner:reddratio of 3.2:l (Block 1955).During 1954, 160 bull trout constructed 48 redds in Trail TABLE6. Percentand number of ageI, II, III andIV bull Creek, yielding a ratio of 3.3:1. Based on these trout emigrating from tributary streams. assumptions,we calculated that an average of 3,450 bull trout successfully spawned annually Years of Age Classes in the Flathead drainage during our period of migration record. Location sampling IV Bull trout spawned in 28 tributaries to the Red Meadow Cr. 1,973,79 676180 North and Middle forks (seeFigure l), but only (3) (42) (r0) (0) a small percentageof the stream reacheswere Trail Creek 1977, 79 3443193 used for spawning. Important spawning tribu- (4r) (s2) (23) (4) taries in the North Fork were Howell, Trail, Geifer Creek l98l 037630 Whale, Big and Coal creeks. Major spawning (o) (26) (4s) (o) tributaries in the Middle Fork were Morrison- Lodgepole,Granite, Ole, Trail and Dolly Varden All Sites ('o) 1849321 The portion of the drainage in Canada (number) (44) (r20) (78) (4) creeks. supported 23-31 percent (mean 29%) of the spawning in the North Fork drainage during the 1980-82period. Howell Creek supported l3-19 Emigration of juveniles from the tributaries percent(mean l6%) of all North Fork spawning. into the Flathead River systemtook place large- ly from June through August (Table 7), similar Monitoring of bull trout spawning at selected to the emigration period reported for the sites indicated that escapementwas highest in Wigwam drainage, British Columbia (Oliver 1982 (Table 8). These sites are consideredrep- 1979).After juvenile bull trout entered the river resentativeof the drainage, and comprised 32, system they appeared to move rapidly down- 30, 31, and 43 percent of the total drainage-wide stream into the main stem Flathead River, arriv- counts in 1980, 1981, 1982, and 1986, respective- ing below the South Fork during August and ly. Monitoring areas reflected drainage-wide September. Although juvenile bull trout were trends. captured by electrofishing in the main stem Juvenile bull trout densitieshave been used throughout the year, their numbers peaked dur- as an index of population status.Juvenile bull ing the fall months (McMullin and Graham l98l). trout populations in sectionsof Coal and Mor- Snorkel observationsindicated that some juve- rison creekshave been monitored for a six-year niles lived along the shallow margins of the Mid- period (Table 9). Numbers of juvenile bull trout dle and North forks. Residencein the lower Flat- in these sectionswere highest in 1987 for both head River before entry into Flathead Lake has streams.Continued population estimatesin these not been well documented. streams will provide valuable baselineinforma- tion for future monitoring. Trends in SoawnerAbundance Sampling for bull trout in Flathead Lake in- Drainage-widecounts of bull trout reddsin 1980 dicated that the population had been relatively (568),l98l (7l4), 1982(l I 38),and 1986(81 4) were stable through 1981. Average catches of bull usedto indexthe numberof adfluvialbull trout trout in sinking nets were 1.2 to 2.1 fish per net

Life Historyand Ecologyof MigratoryBull Trout 139 TABLE T Number of stream trapping days, number of juvenile bull rrour passeddownstream through traps, and number of trapped juvenile bull trout per trap day by month from North Fork tributaries during 1976 to 1980 and Middle Fork tributaries durinq l98l.

June July August September October

North Fork tributaries (1976.1980) Trap days 42 443 424 264 t3l Number of fish 709 340 il6 o Fish/trapday 1.00 t.60 0.80 0.44 0.04 Middle Fork tributaries(1981) Trap days 43 62 t4 Number of fish 60 28 l9 B Fish/trapday 1.40 0.38 0.26 0.57

TABLE 8. Bull trout redd counts for selected areas of tributaries chosen for monitoring in the Flathead Drainage.

t979 1980 t98l t982 l9B3 I 984 l9B5 1986 1987 l 988

North Fork: Big t0 20 IB 4l 22 9 9 l2 22 l9 Coal 3B 34 23 60 /J 6t 40 IJ 48 52 Whale 35 45 9B 2tl l4l 133 94 90 t43 136 Trail 3la 7B 94 56 2,t 25 69 64 62

Total North Fork tt7 r30 2r7 406 292 ZJJ t68b 184 277 269 Middle Fork: Morrison 254 75 32^ 86 o1 38 99 52 49 50 Granite t4 34 I Aa 3l 47 24 37 34 32 Lodgepole et l4 l8 23 23 23 20 42 2l r9 Ole l9 l9 )l 35 26 30 36 45 59

Total Middle Fork 7l t42 B3 t94 156 134 I 73D t67 149 160 Total Drainage MonitoringAreas r88 272 300 600 448 369 341 35r 426 429 aCounts may be undereslimateddue to incompletesurvey. "High flowsmay haveobliterated some of the redds.

TABLE 9. Juvenilebull trout densitiesin sectionsof a North Fork tributary(Coal Creek) and a Middle Fork triburary(Molrson Creek)from 1980-1985.

Population Estimate Date (Number/I50 m section) 95% Confidence Interval

Coal Creek 08105182 130 +36 (at Deadhorse Bridge) 03t23t83 99 +33

08/31/84 89 -zl 08t26t85 t67 +66 0Bn2t86 t49 +45 09t0|/87 179 +55 Morrison Creek 09t23t80 9l +48 09t0UB2 93 +q 08/lB/83 62 +ll 09t25t85 93 +27 08t27t86 114 08t25t87 l38 +10 08/30/88 t26

t40 Fralev and Shenard in 1967-1970,2.2to 2.9 fish per net in 1980-81 related to the percent fine material ( < 6.35 mm) (Leathe and Graham 1982).Average length ofbull in the gravels(Weaver and White 1985).Survival trout sampled in Flathead Lake increasedby 24 to emergenceranged from nearly 50 percent in mm from 1967 to 1980. A larger percentageof substrateswhich contained l0 percent fines, to the fish were greater than 500 mm in the 1980-81 zero percent in mixtures which contained 50 per- sampling period. The percentageof trophy fish cent fines. Juvenile bull trout could be affected (greater than 634 mm) was similar in both sam- by streambed changes because of their close pling periods. Migrating spawners,captured in association with the substrate. Shepard et al. the river system,were similar in size from 1953 (1984a) found a significant relationship (r2 = through l98l (see Table 5). 0.40, P < .01) between substrate score (a measure of unimbedded instream rock cover) and Sensitivityto EnvironmentalDisturbance juvenile bull trout densitiesin tributaries of the Swan River. All bull trout life stages are sensitive to en- vironmental disturbances.The population in the As our studies of bull trout in the Flathead River system we hope to Flathead system is threatened by severalmajor continue, define more precisely factors forms of resource development. The proposed the which negatively affect the population. is not Cabin Creek coal mine in the North Fork drain- It clear whether the tributaries juvenile age in British Columbia received preliminary ap- are at carrying capacity for bull trout, nor whether juvenile limited proval by the Canadian government and was re- densities are by spawnerescapement levels. The answerto these ferred by the U.S. and Canadiangovernments to questionswill require monitoring of the escape- an International Joint Commission for review ment levels resulting juvenile (Flathead International Study Board 1988).This and densitiesin the tributaries over a longer period of time. McPhail mining activity could harm bull trout spawning and Murray (1979)suggested that limitations in and rearing habitat in the upper North Fork and "ecological juvenile rearing habitat may form an in Howell Creek, the major spawning tributary greatly population in the Canadian portion of the drainage. The ma- bottleneck," affecting overall levels of bull trout. jor concernsare increasedsedimentation, altera- tion of flow and water quality degradation Bull trout in the Flathead River system are (Biological ResourcesCommittee 1987).In ad- dependent on habitat quality and management dition, timber harvest and road construction in of the interconnected river, lake, and tributaries. both the North and Middle Fork drainages are Cumulative losses of spawning and rearing potential threats to bull trout spawning and rear- habitat would reduce the bull trout population ing habitat. in Flathead Lake. Increased fishing pressure is often associated with resource development. Because of the re- Acknowledgments stricted distribution of bull trout spawning in the We thank Steve Leathe, Laney Hanzel, Tom basin and the limited size of the known annual Weaver, Karen Pratt, Don Read and Steve escapement(3,000-4,000 individuals), harvest of McMullin for their research efforts on bull trout fish by anglers could reduce the population. Any in the basin. Pat Graham directed portions ofthe increasein harvest by anglers in a particular area studies. Many technicians contributed directly to or subbasincould result in a loss of recruitment the study. The USDE Environmental Protection from that site, in turn reducing the overall Agency, USDA Forest Service,USDI Bureau of population in Flathead Lake. Reclamation, and the USDE Bonneville Power The Iong overwinter incubation and develop- Administration supplied funds and cooperation. ment phase for bull trout embryos and alevins Mark Gaub prepared the illustrations; Dennice (223 days in Coal Creek)leaves them particular- Hamman and Frances Roe typed the manuscript. ly vulnerable to increasesin fine sedimentsand Harold Mundie, Loren Bahls, Alan Martin, Jim degradation of water quality. In laboratory ex- Vashro, Fred Holm, Karen Pratt, Tom Weaver, periments, survival was shown to be inversely and Mark Gaub reviewed the manuscript.

Life Historyand Ecologyof MigratoryBull Trout l4l Literature Clted Graham, P. J., B. B. Shepard, and J. J. Fraley. 1982. Use of stream habitat classifications to identify bull trout Allan, J. H. 1980.Life historynotes on the Dolly Vardenchar spawning areas in streams. In W. B- Armantrout, (ed.) (Saluelinus malma) in the upper ClearwaterRiver, Acquisition and utilization of aquatic habitat inven- Alberta.Alberta Energyand Natural Resources,Fish tory information. Proceedings of a symposium held and Wildlife Division, Red Deer, Alberta, Canada. October 28-30, 1981, Portland, Oregon. Pp. 186-198. Armstrong,R. H., and J. E. Morrow. 1980.The Dolly Varden Hanzel, D. A. 1976. The seasonal,area and deoth distribu- Char Saluelinusmalma. In E. K. Balon,(ed.) CAars: lion of cutthroat trout and Dolly Varden in Flathead salmonid fishesof the genusScluelinus. W. Junk. The Lake. Job Performance Report, Project No. F-33-F-10, Hague,The Netherlands.Pp. 99-140. Job I-a. Montana Department of Fish, Wildlife and Armstrong,R. H., andW. M. Morton.1969. Revised annorated Parks, Kalispell, Montana. 3 p. bibliography on the Dolly Varden char. Research Heimer, J. T. 1965. A supplemental Dolly Varden spawning ReportNo. 7.Alaska Department of Fishand Game, area. University of Idaho, Moscow, Idaho. M.S. Thesis. Juneau,Alaska. Hesse, L. 1977. FIRE I, a conputer program for the com- Balon,E. K. 1980.Comparative ontogeny of chars,In E. K. putation of fishery statistics. Nebraska Tech. Ser. No. Balon, (ed.) CAars: Salmonid fishes of the senus l, Nebraska Game and Parks Commission. Proj. No. Saluelinus.W. Junk. The Hague,The Netheriands. F-10-R.Pp.60. Pp.703.720. Jeppson, P. W., and W. S. Platts. 1959. Ecology and conrrol BiologicalResources Committee. 1987. Predicted impacts of of the Columbia River squawfish in northern Idaho the proposedCabin Creekcoal mine on the aquatic Lakes. Trans. Amer. Fish. Soc., 88:197.203. and riparian resourcesof the FlatheadRiver Basin, Leathe, S. A., and M. D. Enk. 1985. Cumulative imoacts of British Columbiaand Montana. Reporr to the Flat- microhydro developmenton the fisheriesin the S*an head River InternationalStudy Board, Washington, drainage, Montana. Montana Department of Fish, D.C. 352.p. Wildlife and Parks, Kalispell, Montana. ll4 p. Bjornn, T. C. 1961.Harvest, age structure,and growth of Leathe, S. A., and P. J. Graham. 1982. Flathead Lake fish game fish populationsfrom Priest to Upper Priest food habits study-Final report. Montana Department Lakes.Trans. Amer. Fish Soc. 90:27-31. of Fish, Wildiife and Parks, Kalispell, Montana. 137 p. Blackett,R. F. 1968.Spawning behavior, fecundity, and ear- Leggett, J. W. 1969. The reproductive biology of the Dolly ly life history of anadromousDolly Varden (So/oeljnzs Varden char Sahelinus malma (Walbarm). Univer- malma) in southeasternAlaska. J. Fish Res. Board sity of Victoria, Victoria, British Columbia, Canada. Can. 30:543-548. M.S. Thesis. Block,D. G. 1955.Trout migration and spawningstudies on McMullin, S. L., and P. J. Graham. 1981. The impact of the North Fork drainageof the FlatheadRiver. Mon- Hungry Horse Dam on lhe kokanee fishery of the Flathead tana StateUniversity, Missoula, Montana. M.S. Thesis. River. Montana Department of Fish, Wildlife and Parks, Kalispell, Cavender,T. M. 1978.Taxonomy and distributionof the bull Montana. 98 p. McPhail, D., trout (Saloelinus confluentus), from the American J. and C. B. Murray. 1979. The early life-history and ecology (Saluelinus Northwest.Calif. Fish Game 64:139-174. of Dolly Varden malma) in the upper Arrow Lakes. Department FlatheadRiver International Study Board. 1988. Board report of Zoology and Institute of Animal Resources, to the InternationalJoint Commission,Washington, University of British Columbia, D.C. Vancouver, British Columbia. ll3 p. Oliver, G. 1979. A final report on the present fisheries Fraley,J., and P. Craham. 1982.Physical habitat, geologic use of the Wigwam River with an emphasis on the migra- bedrocktypes and trout densitiesin tributariesof the tion, life history and spawning behavior of Dolly FlatheadRiver drainage,Montana. 1n N. B. Arman. Varden char, Sab elinu s malma alb aum). Fisheries trout, (ed.) Acquisition and utilization of aquatic $ investigation in tributaries of the Canadian portion habitat inventoryinformation. Proceedings of a sym- of Libby Resenoir, British Columbia Fish and Wildlife posiumheld October28-30, 1981, Portland, Oregon. Branch, Victoria, British Columbia, Canada. Pp. l78-185. Potter, D. S. 1978. The zooplankton of Flathead Lake: An Fraley,J., D. Read, and P. Graham. 1981.Flathead River historical review with suggestions for continuing lake fishery study: April I981. Montana Department of resource management. University of Montana, Fish,Wildlife and Parks,Kalispell, Montana. 132p. Missoula, Montana. Ph.D. Dissertation. George,E. L., and W. F. Hadley.1979. Food and habitatpar- Pratt, K. L. 1984. Habitat selection and species interactions titioning betweenrockbass (Ambloplites rupestris)and of juvenile westslope cutthroat trout (Salmo clarki smallmouthbass (Micropterus dolomieul young-of- Lewisi) and bull trout (Sabelinus confluentus) in the the-year.Trans. Amer. Fish. Soc. 108:253-26I. upper Flathead River Basin. University of Idaho, Gould,W. R. 1987.Features in the earlydevelopment of bull Moscow, Idaho. M.S. Thesis. (Salaelinus troul conJluentus). Northw. Sci. 6l:264268. Shepard, B. B. and P. J. Graham. 1983. Fish resource monitor- Graham, P. J., and W. Fredenberg.1982. Flathead Lake ing program for the upper Flathead Basin. Montana Fisherman Census.Montana Department of Fish, Department of Fish, Wildlife and Parks, Kalispell, Wildlife and Parks, Kalispell,Montana. 44 p. Montana.6l pp. Graham,P. J., D. Read, D. Leathe,J. Miller, and K. Pratt. Shepard, B. 8., J. J. Fraley, T. M. Weaver, and P. Graham. 1980.Flathead River Basin fishery study. Montana 1982. Flathead River fisheries study. Montana Deparr Department of Fish, Wildlife and Parks, Kalispell, ment of Fish, Wildlife and Parks, Kalispell, Montana. Montana.ll8 o. 86 p.

142 Fraley and Shepard Shepard, B., S. Leathe, T. Weaver, and M. Enk. l984a. Weaver, T. M., and R. G. White. 1985. Coal Creek fisheries Monitoring levels of fine sediments within tributaries monitoring study number III. Quarterly progressre- to Flathead Lake, and impacts of fine sediment on port to United States Department of Agriculture, bull trout recruitment. Paper presented at the Wild Forest Service, Montana State Cooperative Fisheries Trout III Symposium. Yellowstone National Park' Research Unit, Bozeman, Montana. 94 p. Wyoming. ll p. Shepard, B., K. Pratt, and P. Graham. I9B4b. Life histories of westslopecutthroat and bull trout in the upper Flathead River Basin, Montana. Montana Department of Fish, Wildlife and Parks, Kalispell, Montana. 85 p.

Receiaed 14 March 1988 Acceptedfor publication 16 February 1989

Life History and Ecology of MigratoryBull Trout 143