The Big Blackfoot River Fisheries and Restoration Investigations for 2006 and 2007 Montana Fish, \ Wildlife and Parks The Big Blackfoot River Fisheries and Restoration Investigations for 2006 and 2007 by Ron Pierce, Craig Podner, Michael Davidson, Ladd Knotek and John Thabes Montana Fish, Wildlife and Parks 3201 Spurgin Road Missoula, MT 59804 June 2008 TABLE OF CONTENTS INTRODUCTION 5 EXECUTIVE SUMMARY 7 Bull Trout Recovery 11 Westslope Cutthroat Trout Recovery 15 STUDY AREA 19 PROCEDURES Fish Population Estimators 23 Water Temperature 25 Stream Habitat Surveys 25 Whirling Disease Sentinel Cage Exposures 26 WSCT Genetic Testing 26 Working with Private Landowners 27 Natural Channel Design 28 RESULTS/DISCUSSION PART I: Blackfoot River Environment Blackfoot River Discharge 30 Blackfoot River and Tributary Temperatures 31 PART II: Blackfoot River Trout Populations Lower River (Johnsrud and Scotty Brown Bridge Sections) 35 Middle River (Wales Creek, Canyon and Poorman/Dalton Sections) 35 Upper River (Upper Blackfoot Mining Complex Area) 38 PART III: Restoration: tributary assessments Ashby Creek 44 Bear Creek 45 Blanchard Creek 46 Braziel Creek 47 Chamberlain Creek 49 Copper Creek 50 Cottonwood Creek 51 (Enders) Spring Creek 53 Frazier Creek 55 Gold Creek 56 Hoyt Creek 57 Jacobsen Spring Creek 59 Kleinschmidt Creek 61 Lincoln Spring Creek 62 McCabe Creek 63 PART III: Restoration: tributary assessments (cont.) Monture Creek 64 (Murphy) Spring Creek 65 Nevada Spring Creek 66 Pearson Creek 68 Poorman Creek 69 Rock Creek 70 Shanley Creek 71 Snowbank Creek 72 Tamarack Creek 73 Warren Creek 74 Wasson Creek 76 Willow Creek, Bear Gulch and Sauerkraut Creeks 78 PART IV: Fisheries and Aquatic Assessments in the Clearwater Basin 84 Selected Tributaries Blind Canyon Creek 86 Boles Creek 88 Deer Creek 90 East Fork Clearwater River 92 Marshall Creek 94 Morrell Creek 97 Trail Creek 100 West Fork Clearwater River 103 Clearwater River Sections Section I: Big Blackfoot River to Salmon Lake 106 Section II: Salmon Lake to Seeley Lake 109 Section III: Seeley Lake to Emily A Dam Ill Section IV: Emily A dam to Rainy Lake 114 Section V: Rainy Lake to Clearwater Lake 117 PART V: Whirling Disease Investigations 121 1 . Pilot assessment of the association between stonefly assemblages and the incidence and severity of whirling disease in tributaries of the Blackfoot River, Montana 123 2. Exploratory assessment of association between invertebrate "EPT" taxa and the incidence and severity of whirling disease in tributaries of the Blackfoot River, Montana 128 3 . Environmental conditions linked to Myxobolus cerebralis infection in basin-fed streams of the Blackfoot Watershed, Montana 134 4. Relationships of migratory (hybrid) rainbow trout spawning life histories to risk of Myxobolus cerebralis infection in the Blackfoot Basin 149 PART V: Whirling Disease Investigations (cont.) 5. Status review of Mountain Whitefish (Prosopium williamsoni) in the Blackfoot Basin: a pilot study help identify risk of whirling disease 163 RESULTS PART VI: Other Special Studies 1 . Fluvial westslope cutthroat trout movements and restoration relationships in the upper Blackfoot Basin, Montana 174 2. A stream restoration and native fish conservation prioritization strategy for the Blackfoot River Basin 188 RESULTS PART VII: Backcountry fisheries Investigations 199 Streams 202 Lakes 215 RECOMMENDATIONS 267 ACKNOWLEDGEMENTS 268 LITERATURE CITED 269 APPENDICES 271 Appendix A: Summary of catch and size statistics for Blackfoot tributaries, 2006-07. Appendix B: Summary of two-pass estimates for Blackfoot tributaries, 2006-07. Appendix C: Mark and recapture estimates for the Blackfoot River, 2006. Appendix D: Summary of stream discharge measurements for 2006-07. Appendix E: Restoration streams and table of activities through 2007. Appendix F: Potential restoration projects in the Blackfoot drainage through 2007. Appendix G: Restoration streams and cooperators through 2007. Appendix H: Summary of water temperature in the Blackfoot drainage, 2006-07. Appendix I: Summary of water chemistry readings for 2006 and 2007. Appendix J: Westslope cutthroat trout genetic sampling sites and results, 2006-07 Appendix K: Blackfoot Basin restoration prioritization scorecard through 2007. INTRODUCTION The Blackfoot River Basin is the site of a "wild trout" restoration and conservation initiative. This initiative began 20-years ago (1988-89) when fisheries- related assessments identified 1) the over-harvest of native trout, 2) stream degradation (at a basin-scale), and 3) toxic mine waste (in the headwaters) as primary threats to Blackfoot River fisheries (Peters and Spoon 1989, Peters 1990). These findings led to the initial adoption of protective angling regulations in 1990 followed by the implementation of pilot-level restoration projects. Early project successes led to the development of a private lands restoration methodology for the Blackfoot River and the expansion of tributary restoration from the mid-1990s to the present. While the guiding philosophy of wild trout conservation provides the foundation for this endeavor, the cooperation of many resource agencies, conservation groups and private landowners (i.e. Blackfoot Cooperators - see below) form the social and technical network necessary to fund and implement the initiative. This initiative provides a more specific framework for the recovery of dwindling stocks of imperiled native fish when integrated with targeted harvest regulations, site-specific restoration and landscape protection measures often undertaken in remote but critical areas of the watershed. Fisheries restoration in the Blackfoot is an iterative process in which the scope and scale of restoration expands as information and stakeholder cooperation is generated. This iterative process usually begins with fisheries assessments, which often lead to restoration measures of individual tributary stocks, and so involves methods such as enhancing flows in rearing areas, preventing juvenile fish loss to irrigation in migration corridors, reconstructing altered streams, fencing livestock from spawning areas, while expanding these types of actions to adjacent tributaries as human-induced limiting factors are identified and as opportunities allow. Including the backcountry and upper-river mining areas, Montana Fish, Wildlife and Parks (FWP) has now inventoried (or otherwise assessed) 182 streams, including all major tributary streams within the Blackfoot River basin during the last 20-years. Excluding the backcountry, we have identified human-induced fisheries impairments on a great majority (>90%) of low-elevation water bodies (Appendix F). With information derived from these and related investigations, and with the cooperation of many stakeholders, the Blackfoot Cooperators have now targeted -50 tributaries with >600 individual fisheries-related projects (Appendix E). Correcting environmental (riparian) damage over large tracts of mixed land ownership involves protection (e.g. conservation easements) and restoration of biologically important but fisheries-impaired streams. Improving habitat involves both passive (e.g. compatible grazing) and active (e.g. channel reconstruction) measures depending on the degree of degradation and a stream's recovery potential. The geographic focus of restoration has been lower-river tributaries and bull trout "core area" streams; however, fisheries restoration and conservation measures are now expanding to streams in the Lincoln Valley and Clearwater River basin and other peripheral areas of the Blackfoot Basin, such as the Nevada Creek and Potomac Valley. In addition to the scale and scope of restoration, stakeholder involvement in the fisheries initiative continues to expand among non-profit groups (NPG), agencies and landowners. The Big Blackfoot Chapter of Trout Unlimited (BBCTU) is a leading NPG with a full-time project manager dedicated to restoration oversight. The Blackfoot Challenge (BC) helps by coordinating Total Maximum Daily Load (TMDL) studies and fund-raising for water quality impaired streams and facilitates conservation easements and landowner educational programs. Likewise, the North Powell Conservation District (NPCD), The Nature Conservancy (TNC) and Five Valleys Land Trust (FVLT) are engaged in the development of certain fisheries-related conservation projects. TNC specifically has made a great contribution by protecting -90,000 acres of industrial forestland and streams within from development. The combined services of federal agencies - U.S. Fish and Wildlife Service - Partners for Fish and Wildlife (USFWS), Natural Resource Conservation Service (NRCS), U. S. Forest Service (USFS) and Bureau of Reclamation (BOR) provide a wide range of resource expertise, project funding and technical services. The Bonneville Power Administration (BPA), the Trout Unlimited Western Water Project and Department of Natural Resource Conservation (DNRC) are helping coordinate drought mitigation, instream flow and water-leasing projects. Northwestern Energy (NWE) - Milltown Mitigation Funds help cost-share FWP research, monitoring and reporting (i.e. this report and all field studies described). Private landowners, including Plum Creek Timber Company, private foundations and others also contribute significant resources to fisheries-related projects. Together this affiliation - the Blackfoot Cooperators, form the general support base of the Blackfoot River Fisheries Restoration Initiative. A general summary of their support by individual stream is located in Appendix G. This expansion continues to generate new fisheries initiatives and restoration opportunities. One recent initiative - the NRCS sponsored Expedited EQIP Bull Trout and Westslope Cutthroat Trout Conservation program directed $2,000,000 in federal (Farm Bill) resources to 13 "priority" native trout streams. A second initiative - the Native Fisheries Habitat Conservation Plan (HCP) - a cooperative venture related to the BC Blackfoot Community Project recently placed an ~8,000-acres easement dedicated to native fish conservation on former Plum Creek Timber Company lands (Results Part III). Other important fisheries projects involve the ongoing removal of Milltown at the mouth of the Blackfoot River and the impending clean-up of the Mike Horse Mine, a contaminated mining area that poses extreme ecological risk to the upper Blackfoot River. Despite the many fisheries improvement projects, adverse human pressure upon salmonid habitat (and native trout specifically) in the Blackfoot River Basin is widespread, and as described in this report, these pressures will continue to pose a daunting conservation challenge well into the future. 1992 1994 1996 1998 2000 2002 2004 2006 2008 EXECUTIVE SUMMARY The 2006-2007 reporting period ended with the eighth straight year of drought (Results Part I). During this time, the Blackfoot River Basin was subject to extreme wildfires, increasing water temperatures, and the Blackfoot River set a new (recent) record for warming (Figure 1). Through this 8-year period of warming, the flow regime of the Blackfoot Basin has expressed a pattern of early runoff, and a consistent lowering of the spring hydrograph and lower summer base-flows (Figure 2). During the last two years, we surveyed fish populations in seven established Blackfoot River monitoring sites, and added two new survey sites to the upper-most reach of the Blackfoot River (Results Part II). Lower Blackfoot River surveys identified relatively high trout population abundance (densities and biomass) in the lower river near Johnsrud but very low population abundance in the middle Blackfoot River below Nevada Creek (Figure 3, Results Part II, Appendix C). Low salmonid densities in the middle Blackfoot River relate to weak recruitment of juvenile fish to river populations and other factors. Limited recruitment stems from low number of natural tributaries upstream of Nevada Creek and adverse alterations of tributaries between the North Fork and Lincoln. The middle Blackfoot River has potential for greatly improved fisheries through restoration and with the cooperation of nearby landowners. Westslope cutthroat trout (WSCT) specifically Figure 1. Maximum annual water temperatures for the lower Blackfoot River downstream of Belmont Creek, 1994-2007. 6000 £ 4000 u 1898-1999 ■2000-2007 ^^^ <.^ ^' i^^ ^^ J>^ ^ ^ / vd^ .cP Figure 2. Blackfoot River mean hydrograph (1898-1999) and the 2000-2007 mean hydrograph (USGS Bonner guage). ^ 100 o o o (A (A (Q E o 20 -Johnsrud (rm 13.5) -Scotty Brown (rm 43.9) -Wales Creek (rm 63.0) 1989 1990 1991 1993 1996 1998 2000 2002 2004 2006 Year of survey Figure 3. Total trout biomass estimates (fish >6.0") for the three lower Blackfoot River sampling locations, 1989-2006. have potential for 120 improvement through targeted tributary restoration as identified in a radio- telemetry WSCT study involving spawning movements (Results Part VI). Based on past and recent fish population studies in the upper Blackfoot River, we again report on impacted fisheries in the area of the Mike Horse Mine where WSCT have expressed a declining trend in densities for >30 years (Results Part II). Two mining companies (Asarco and Atlantic Richfield Company) are identified as responsible for cleanup costs ($37,000,000) as announced by the Montana's Governor Brian Schweitzer and Attorney General Mike McGrath on April 25* 2008. With cleanup slated to begin, the removal of streamside contaminants and restoration of riparian habitats are both necessary for a meaningful level of WSCT improvement in the mine-impact area. With successful remediation and restoration project, the damaged area has potential to recover resident WSCT and perhaps increase downstream recruitment of fluvial WSCT to the middle Blackfoot River (Results Part II, VI). In addition to river assessments, we present the results of fisheries and habitat assessments for 29 tributaries undergoing restoration (Results Part III). The results of these restoration-related investigations identify great promise for improvement to wild trout fisheries in the presence of river warming and whirling disease once damaging land (riparian) activities and other human-related factors limiting populations are corrected. In addition to identifying many successes, these assessments shed light on the complexities of native fish recovery and the inherent challenges of ensuring successful restoration on (private and public) lands managed for agricultural purposes. Our assessments identify not only a great need to continue restoration actions, but also a fundamental need to continue to track restoration progress to help ensure projects meet their intended outcomes. In Results Part FV, we report on important tributaries of the Clearwater River basin as well as five reaches of mainstem Clearwater River. The Clearwater River drainage is the largest tributary to the Big Blackfoot River by drainage area. The Clearwater system is a unique drainage within the Blackfoot Watershed and it supports exceptional and diversified aquatic resources, including many native fish populations with unique life history traits. Because of the interconnected nature of stream, river and lake environments, species richness is high and adfluvial migratory life forms are common. The Clearwater Basin supports some of the only natural and currently viable adfluvial bull trout populations in the region. Fisheries emphasis and restoration accomplishments on streams in the greater Blackfoot Basin have generally not yet included the Clearwater system. The need for relevant fisheries information has recently become imminent as the rapid conversion of corporate timberlands to smaller residential properties has forced natural resource managers and conservation advocates to prioritize lands for protection and acquisition. With the recent completion of fisheries work in the Clearwater Basin, we are now able to integrate key findings into the broader Blackfoot River fisheries restoration prioritization strategy (Pierce et al, 2006). The new prioritization is located in Results Part VI (Appendix K). This inclusion of the Clearwater Basin into the broader prioritization strategy represents an initial attempt to direct conservation efforts to important native fish populations. With the introduction of the exotic parasite Myxobolus cerebralis, whirling disease has expanded in recent years. It is now firmly established at the lower elevations of the watershed where infection among salmonids, particularly rainbow trout, vary within and between streams (Results Part V). The escalation of the disease (severity and distribution) as measured by histological scores of rainbow trout exposed toM cerebralis corresponds with an increase in cranial deformities in Blackfoot River rainbow trout, as well as a recent decline in rainbow trout in the middle Blackfoot River downstream of Monture Creek, a highly infected spawning stream. In 2006-07, we explored certain biotic (benthic) relationships of the M cerebralis parasite within the Blackfoot Basin. We began this work by exploring relationships between the incidence and severity of disease (as measured on the MacConnell Baldwin scale from sentinel cage exposures) and stonefly and EPT (Ephemeroptera, Plecoptera and Trichoptera) assemblages (Results Part V). Significant negative correlations were demonstrated between measures of disease and several metric relationships, including EPT richness and pollution sensitive taxa richness. Our findings further identified cold stenothermic species only in streams where infections went undetected (Results Part V). To further assess ecological relationships of whirling disease, we then investigated physical relationships between a group of five-landscape and four reach- scale environmental conditions and the presence of infected fish for 13 basin-fed tributaries to the Blackfoot River. In this study, infections were present and severity of disease was high in meandering streams (e.g. Monture Creek) in broad valleys with gentle relief and warmer summer temperatures. Conversely, streams with higher gradients, lower levels of fine sediment within the substrate, low summer temperatures and stenothermic species supported little to no infection despite the near proximity of higher infection rates in adjacent waters (Results Part V). Basic applications of this research relate to the need to continue to implement a more sensitive level of streamside management in environments prone to infection, while maintaining cold-water environments where possible. In a companion study also geared towards restoration and management applications of sport fisheries, we identified the relative spawning use of variously infected streams by fluvial Blackfoot River rainbow trout using radio telemetry (Results Part V). This study confirmed dispersed spawning throughout colder, higher gradient tributaries of the lower Blackfoot Basin in areas where infections remain low. However, a majority of rainbow trout from the middle Blackfoot River spawned in lower Monture Creek (a warmer lower-gradient stream) in waters highly infected by M cerebralis. Concentrated spawning within a single, localized and highly infected stream raises concern for an additive level of recruitment loss to trout populations in the middle Blackfoot River. This reach of river has long been identified with recruitment problems brought on by drought, low winter survival of juvenile fish, and human-related habitat impairments in spawning tributaries. To offset potential RBT losses in disease prone waters of the middle Blackfoot Basin, stakeholders should 1) better manage riparian areas for channel stability, increased shade and erosion reduction, 2) promote native fish recovery and migratory life histories, and 3) restore (or enhance) habitats favoring salmonid life stages less affected by the pathogen. As our ability to predict whirling disease in the Blackfoot Valley improves, we are now beginning to examine the influence of M cerebralis on mountain whitefish (Results Part V), a species now expressing the clinical signs of whirling disease within western Montana (and the Blackfoot River), but whose susceptibility remains in question. This evaluation began with a status review of the species in the Blackfoot Basin (Results Part V). This review identifies the middle Blackfoot River as supporting relatively high densities of juvenile whitefish; however this distribution also overlaps with the presence ofM cerebralis and high severity of disease in other species (e.g. rainbow trout). Recent sampling of mountain whitefish in the upper Blackfoot River has identified low juvenile densities where high infections are now prevalent. A pilot-level mountain whitefish telemetry study planned for 2008 will attempt to identify whitefish spawning sites, similar to the rainbow trout telemetry study, and in so doing help identify disease risk at the local scale. These investigations should coincide with controlled lab studies examining the susceptibility of whitefish to the parasite, which hopefully will lead to sentinel exposures of age-0 whitefish within the wild. In Results Part VII, we also report on stream and lake assessments in the "backcountry" of the Blackfoot Basin. This work identifies native species in high abundance throughout the upper Monture basin, but non-native Oncorhynchus hybrids throughout the upper North Fork upstream of the North Fork Falls in relatively low abundance. Hybrids above the North Fork Falls are self-sustaining and their presence is traced to historical stocking in (at least) two high mountain lakes. Our surveys suggest non-natives in the upper North Fork are poorly suited to the backcountry environment. Because of very low population densities, these fisheries provide only minimal ecological value. Furthermore, this backcountry source of non-native hybrids poses risks of hybridization to native WSCT in downstream waters. Options to convert these non- natives to a native species (i.e. WSCT) capable of thriving in the backcountry environment should be further explored. In summary, this report consolidates the results of recent FWP Blackfoot River fisheries restoration and related investigations. Our objectives are to: 1) summarize the status of Blackfoot River wild trout and their environments; 2) summarize fisheries- related monitoring in tributaries undergoing restoration; 3) report on a whirling disease investigations, 4) present the results and management considerations stemming from both rainbow trout and westslope cutthroat trout telemetry studies; 5) present the current status of backcountry fisheries investigations; and 6) help guide future fisheries restoration and other conservation actions basin-wide and to specifically expand native fish protection measures to the Clearwater River basin. 10 Bull Trout Recovery Bull trout is a "species of special concern" in Montanan and a species listed as "threatened" under the Endangered Species Act. The Blackfoot Basin supports stream- resident and migratory (i.e. fluvial and adfluvial) bull trout. The recovery of migratory bull trout in the Blackfoot Basin fundamentally relies on restoration and protection of "core areas" i.e. critical waters and watershed that provide for the spawning, rearing and movement (Figure 4). Within these areas, migratory bull trout exhibit local adaptations that involve spawning in discrete areas, tributary use by early life-stages, large home ranges, extensive migrations at higher flows, and seasonal use of larger, more productive river (or lake) habitats. Migratory bull trout also require complex habitats, colder water, groundwater upwelling for spawning and lower sediment and more tributary access than currently exists in many areas of the Blackfoot Watershed. Stream resident bull trout require similar environments and complete their life-cycle in tributary streams. Adfluvial bull trout are rare in the upper Clark Fork Basin but occupy the Clearwater chain of lakes and migrate to tributaries for spawning and rearing. The life-histories of fluvial bull trout Clearwater above Rainy Lake ^North Fork Landers Fork opper fr^ci^^ Creek Figure 4. Bull trout "core areas" for the Blackfoot Basin (MBTRT 1996). have been extensively studied in the Blackfoot basin (Swanberg 1997, Schmetterling 2003, Pierce et al. 2005) and are the primary focus of current recovery actions. Until recently, there has been very little applied research (or restoration) directed to the recovery of adfluvial bull trout within the Clearwater Basin. In 2006-2007, FWP and a University of Montana graduate student captured, radio-tagged and tracked >50 adult adfluvial bull trout from the Clearwater chain of lakes 11 and Clearwater River. Study objectives were to identify migration patterns, spawning areas, conservation needs and the impacts of several upstream migration barriers on the main stem river between the Clearwater lakes. This study identified several migratory components of the Clearwater population and complex movements between and among the various lakes. Three spawning tributaries (Morrell Creek, West Fork Clearwater and East Fork Clearwater) were identified as the primary spawning and rearing habitats for adfluvial bull trout. Adult bull trout from Salmon and Seeley Lakes migrated either up (Salmon Lake) or downriver (Seeley Lake) before ascending Morrell Creek for spawning. Seeley Lake fish also expressed upstream movements and, combined with the downstream migrants from Lake Inez and Lake Alva, ascended the West Fork drainage. Bull trout from Rainy Lake, and to a lesser extent Lake Alva, migrated up the Clearwater River and spawned in the East Fork drainage. At this early stage of the study, primary concern include the dewatering and entrainment of migratory bull trout in Morrell Creek, fragmentation of key migratory corridors (mainly due to dams) on the mainstem Clearwater River and development pressures along existing subdivisions. Landscape- level development pressures on Plum Creek Timber Company holdings "S in many of the C important tributary o o drainages are ■o considered a SL primary long-term threat to bull trout in the Clearwater basin. Fisheries inventories were also completed on all perennial tributaries (>25) to the Clearwater River. In addition to streams identified in telemetry studies. Boles Creek, Deer Creek, Camp Creek and Trail Creek drainage were identified as supporting bull trout. Since 1990, many actions targeting the 140 120 100 80 60 40 20 D Monture D North Fork ■ Copper Creek M Hi t frti i J' ^ ^ ^ cf c^ # <# J" <# J> ^ ^ # ^J? # # ^i^* # Figure 5. Bull trout redd counts for index reaches in three primary fluvial bull trout spawning streams, 1989-2007. o o 45 40 35 30 :: 25 o 20 +j O 10 5 r -| Monture Creek "iforth Fork Copper Creek n r ■1 |-| _ j=L^LL^LL^ 1 1 1 1 '"' 1 -W-^- 1 1 1 1 '"' 1 1 1 o> Tt eo o p>i lo h-- 00 o> o> o o o o o> o> o> o o o o T- T- T- CM Cv| CM Cv| o> ^ eo o p>i lo h-- 00 o> o> o o o o o> o> o> o o o o T- T- T- Cv| Cv| CM Cv| 0> 0> CNI Tt lO h-- 00 o> o o o o o> o> o o o o T- T- CM CM CM O Figure 6. CPUE for juvenile bull trout near spawning sites of three primary spawning streams, 1989-2007. 12 recovery of fluvial bull trout in the Blackfoot Watershed were completed (Pierce et al 2005), and many others are ongoing. In 2006 and 2007, the Blackfoot Cooperators developed habitat restoration projects in five core areas, and these included: 1) enhancing instream flows, improving fish passage flows and riparian fencing on Cottonwood Creek; 2) channel reconstruction in Hoyt Creek, a tributary to Monture Creek; 3) flow enhancement on Murphy Spring Creek, 4) and the reconstruction of Jacobsen and Enders Spring Creeks, and continued habitat work on both Rock Creek and Kleinschmidt Creek (all are in the North Fork Blackfoot River basin). Bull trout recovery work in the development phases includes 1) improving access, ditch screening and instream flows in Snowbank Creek (a tributary of Copper Creek), and 2) fish screening and flow enhancement on Morrell Creek. Other recent important bull trout recovery actions include the restoration of fish passage at Milltown dam, and a native fish Habitat Conservation Plan (HCP) conservation easement on Sauerkraut Creek. Following the adoption of protective harvest regulations and the initiation of initial bull trout recovery actions, bull trout densities in the lower Blackfoot River increased during the 1990s, with an inclination towards large fish (Pierce et al 2004). However, with the onset of drought in 2000, all measures of fluvial bull trout abundance showed declines, with the exception of Copper Creek (Figure 2, Results Part III). Bull trout declines correspond with an 8-year trend of increasing water temperatures and low stream flows. During this period increasing water temperatures have reduced thermally suitable bull trout (summer) habitat in the lower reaches of several core areas with recent water temperatures >70°F (Figure 7, Appendix H). This warming includes areas such as lower Monture Creek (Results Part III), an area identified as refugia for fluvial bull trout during periods of river warming (Swanberg 1997). If global warming trends continue as predicted, a reduction in thermally suitable streams adjacent to seasonally unsuitable habitat (e.g. Blackfoot River) is expected to reduce populations to incrementally smaller patches of habitat over a range of spatial scales (Reiman et al. 2008). The cumulative effects of warming and other human-induced adverse influences to bull trout habitat clearly identifies the need to continue restoration in core area streams where local human- induced warming, dewatering and habitat degradation is identified (Appendix F) and can be corrected. Redd surveys in index reaches show notable declines in both Monture Creek and the North Fork with declined of 65% and 66% from recent highs (Figure 5). Conversely, total redd counts in Copper Creek in 2006-07 have increased sharply (75%)) above the long-term (1989-2005) mean (Results Part III). These recent increases appear to relate to increased stream productivity as the result of wildfire (Results Part III). Similar to redd counts, juvenile bull trout densities have declined near Monture Creek and the North Fork spawning area, whereas densities of juvenile bull trout in Copper Creek have recently increased (Figure 6). Bull trout declines were also detected in the lower Blackfoot River at both Johnsrud and Scotty Brown monitoring locations between 2000 and 2006. In the Scotty Brown section of the lower Blackfoot River, bull trout (>6.0") densities declined from 7.7 to 4.4 bull trout/1000' between 2000 and 2006 (Results Part II). Bull trout were present in the Canyon Section in low densities in 1999 but absent from our 2006 surveys. Likewise densities in the upper Blackfoot River appear very low (Appendix C). 13 70 65 ■ 55 50 45 Belmont Creek at Mouth: July I 25% -Min -Median -Max . 75% 1996 2001 2002 2003 2004 2005 2006 2007 65 55 Copper Creek at Sucker Creek Bridge: July 111 1996 1999* 2001 2003 2004 2006 2007 ■ 25% -Min -Median -Max • 75% 75 70 ^ 65 .55 50 — ^ 45 U^ 40 Cottonwood Creek at HWY 200: July I 25% -Min -Median -Max . 75% 1997 2001 2002 2003 2004 2005 2006 2007 75 70 65 60 .55 50 45 40 Gold Creek Near Mouth: July T -r -r I -r — .— -^- - — - I ~ -\- TJJ T — — -^ ■ 25% -Min -Median -Max » 75% 1999 2000 2001 2002 2003 2004 2005 2006 2007 Monture Creek at FAS Temperatures: July 75 70 65 60 .55 50 45 — ^ I 25% -Min -Median -Max > 75% 40 1994 1996 1999 2001 2002 2003 2005* 2006 2007 70 65 55 50 45 40 North ForkatOvando-Helmlvllle Road: July rrt U ■ 25% -Min -Median -Max • 75% 1994 1996 2000 2001 2002 2003 2004 2005 2006 2007 Figure 7. Summary of July water temperatures at monitoring sites in the lower reaches of six bull trout core areas. 14 Surveys of juvenile Bull trout in upper Cottonwood Creek indicate low, but stable juvenile densities (Results Part III). However, livestock degradation of riparian areas has damaged bull trout habitat in middle Cottonwood Creek. Recent fish population surveys have failed to detect bull trout in the middle reach of Cottonwood Creek. Despite a general decline in bull trout at a basin-scale, monitoring has detected bull trout expansion in certain streams where improved fish passage and beneficial habitat restoration work has been completed. Examples of this expansion include Murphy Spring Creek, Nevada Spring Creek and Snowbank Creek (Results Part III). Although bull trout are particularly sensitive to many threats, at this time whirling disease appears to be less of a concern for bull trout than for other salmonids. Compared with WSCT, rainbow trout and brook trout, bull trout exhibit a greater physiological resistance to whirling disease (Vincent 2002). In 2006, we continued to monitor whirling disease near bull trout spawning and rearing areas of Cottonwood Creek, Monture Creek and the North Fork, using sentinel fish exposures. The tests indicate that whirling disease is not yet present at these locations; however, the disease is present at various levels in lower reaches of these streams. With increased warming, whirling disease is expected to expand in the upstream direction (Results Part V). Westslope Cutthroat Trout Conservation WSCT is also a "species of special concern" that has declined over much of their historic range within the last century. In Montana, these declines are most pronounced east of the Continental Divide in the upper Missouri River drainage (Shepard et al. 2003). In the Blackfoot Watershed, WSCT occupy -90% of historical range. Genetically unaltered WSCT dominate the upper basin upstream of the North Fork (Figure 8); whereas, introgressed tributary stocks are also common in the lower basin (including Clearwater sub-basin), and these test at levels that generally exceed 90% "genetic purity" (Figure 8, Appendix J). Historical accounts suggest WSCT were once abundant in river systems of western Montana (Lewis 1805, Shepard et al. 2005), where populations expressed a range of migratory (fluvial and adfluvial) and stream-resident life history traits. WSCT in the Blackfoot Basin still posses these three life history traits with 1) fluvial fish in the Blackfoot River (and tributaries), 2) stream-resident WSCT within smaller segments of tributaries, and 3) adfluvial fish in Clearwater Basin chain-of-lakes and nearby streams. Because of their low densities, unique life histories and high recreational value, fluvial WSCT of the Blackfoot River have become a guiding conservation target within the Blackfoot Basin. WSCT of the Blackfoot River typically occupy a large home-range and spawn in tributaries where the young rear before they migrate to a large river to mature (Results Part VI). WSCT have become increasingly rare as a result of habitat loss and degradation, competition with non-native fishes, genetic introgression and fish passage barriers (Mclntyre and Reiman 1995, Shepard 2003), all of which are present primarily at the mid-to low elevations of the Blackfoot Basin. 15 genfef k test pegEtlti^ ■> H^tvitjisatJen not dstect«d O t90% WSCl marUers W5Cr (Life hJstotY ''^'^-^ 5tf«4in-pes!deiivf ^^« Absent Figure 8. Generalized WSCT life history traits and summary of genetic test results. As identified in many Blackfoot fisheries studies (Results Part III and V), human alterations to WSCT habitat are pervasive throughout the tributary system and adversely influence many spawning and rearing areas of both resident and migratory WSCT stocks. As reflected in nearby tributaries, population trends at long-term monitoring sites in the Blackfoot River identify fluvial WSCT as: 1) declining in the upper river; 2) at static and very low densities in the middle river; but 3) increasing in abundance in the lower river (Figure 9, Results Part II). WSCT declines in the upper river upstream of Lincoln correspond with the release of toxic mine waste and related a population collapse within and downstream of the Upper Blackfoot Mining Complex (Spence 1975, Peters 1990, Stratus 2007, Results Part II). Extremely low densities of WSCT in the middle-to upper river reflect habitat loss and limited spawning opportunity in a majority of adjacent streams (Pierce et al 2000, 2001, 2002, 2004, Results Part III). As identified in a recent telemetry study, WSCT use of tributaries of the middle Blackfoot River was not detected over a large contiguous area (-43% of the upper basin above the North Fork) despite WSCT populations in the headwaters of nearby tributaries. This loss of spawning access and opportunity stems from disrupted migration corridors and habitat loss (i.e. riparian degradation) in the lower reaches of tributaries. The expressed loss of recruitment to the 16 Blackfoot River is expressed in extremely low WSCT densities in two mid-river sampling sites (river mile 63.0 and 95.3, Figure 9). Conversely, WSCT in the restoration focus area of lower Blackfoot River have significantly increased. These increases stem from the cumulative influences of protective regulations, increased fish passage and habitat restoration targeted at primary spawning tributaries such as Chamberlain Creek and many other streams (Results Part III). 140 120 100 #71000' +/- 95% CI '91 '96 '00 '02 13.5 '91 '96 '00 '02 '06 '02 '06 63.0 43.9 :o6 95.3 '71'72'88'06 '72'73'88'06 '73'75'88'99'06 '7r88'99 107.2 119.6 124.3 128.5 Figure 9. WSCT densities at eight sampling locations on the Blackfoot River. The horizontal axis shows the year of survey and the river-mile mid-point of the survey. In concert with fluvial bull trout recovery, the focus of WSCT recovery is re- establishing the fluvial life-history form by: 1) reducing or eliminating "controllable" sources of anthropogenic mortality; 2) maintaining and restoring existing spawning and rearing habitats; 3) restoring damaged habitats; 4) improving connectivity from the Blackfoot River to fluvial spawning areas; while 5) maintaining genetically "pure" population isolates (e.g. Ashby Creek, Results Part III). Most of the current WSCT- related work (-40 streams) occurs in bull trout core areas or tributaries to the lower Blackfoot River (Pierce et al. 1997; 2001; 2002; 2004; 2006; Results Part III). The distribution of whirling disease generally conforms to low-elevations of the basin below many known WSCT spawning and rearing sites with some exceptions, including Chamberlain Creek, an important fluvial WSCT spawning stream in lower Blackfoot Watershed. Despite high infection rates in test fish (rainbow trout), continued population monitoring of WSCT in Chamberlain Creek identify stable densities (Results Part III and V). The recovery of WSCT within Chamberlain and Wasson Creeks provides model examples of WSCT recovery within a tributary with benefit the broader WSCT metapopulation of the Blackfoot River. 17 During 2006 and 2007, we completed fisheries inventories to the backcountry of the watershed, including tributaries in the upper Monture, upper North Fork and upper Cottonwood Basins (Results Part VII). Inventories identified a WSCT-dominated community in relatively high abundance throughout the upper Monture basin. Upstream of the Monture Falls, we found a population of WSCT that tested genetically pure (Results Part VII, Appendix J). Although inventories are not yet complete, to date we failed to identify WSCT in the upper North Fork upstream of the North Fork Falls. At this time, it is unclear whether WSCT were ever present to this headwater area of the North Fork; however, our findings of a single sculpin upstream of the North Fork Falls indicate some level of post-glacial native fish passage and presence. Oncorhynchus hybrids above the North Fork Falls are now self-sustaining and their presence is traced to historical stocking in Lower Twin and Parker Lakes. Downstream of the North Fork Falls, we found naturalized rainbow trout in Camp Lake and Lake Otatsy (Results Part V). Likewise, the Cottonwood Lakes surveys identified a rainbow population upstream of a genetically pure WSCT population. These rainbow-dominated backcountry environments all place nearby populations of WSCT at increased risk of introgression (Results Part V, Appendix J). STUDY AREA The Blackfoot River, located in west-central Montana, begins at the junction of Beartrap and Anaconda Creeks (within the Upper Blackfoot Mining Complex), and flows west 132 miles from the base of the Continental Divide to its confluence with the Clark Fork River at Bonner, MT (Figure 10). The Blackfoot River is one of twelve renowned "blue ribbon" trout rivers in Montana with a 1972 appropriated "Murphy" in-stream flow water right of 700 cfs at the USGS Bonner (#12340000) gauging station. The 50-year mean annual discharge is 1,554 (cfs) near the mouth (USGS 2008 provisional data). This river system drains a 2,320-mile^ watershed through a 3,700-mile stream network, of which 1,900 miles are perennial streams capable of supporting fishes. The physical geography of the watershed ranges from high-elevation glaciated alpine meadows, timbered forests at the mid-elevations, to prairie pothole topography on the valley floor. Glacial landforms, moraine and outwash, glacial lake sediments and erratic boulders cover the floor of the entire Blackfoot River valley and exert a controlling influence on the habitat features of the Blackfoot River and the lower reaches of most tributaries. The Blackfoot River is a free flowing river to its confluence with the Clark Fork River where Milltown dam, a run-of-the-river hydroelectric facility eliminated upstream fish passage from 1907 when the dam was constructed to its removal in 2008. In April 2008, the first uninhibited movements of fish from the Clark Fork to the Blackfoot River were documented. 1 1 iLU ^H Utuvmiwy ^ m^Klieria 1 1 PlL5n CreeK 1 IDFW 1 j DtKf pfwate Figure 10. Land ownership map of the Blackfoot River Watershed. 19 The Blackfoot River is also one of the most popular, scenic, physically diverse and biologically complex rivers in western Montana. Segments of the river system however support low densities of wild trout due to an array of natural conditions and human impairments. Densities of imperiled native trout (westslope cutthroat trout (WSCT) - Oncorhynchus clarki lewisi and bull trout - Salvelinus confluenttis) are particularly low. Natural limiting factors involve drought stressors, areas of high instream sediment loads, low instream productivity, naturally intermittent tributaries, summer warming and periods of severe icing of the lower mainstem river channel. Human impairments apply to mining-related contamination in the upper Blackfoot Basin, the loss of upstream fish passage at the mouth of the Blackfoot River, expansion of exotic organisms including whirling disease at the low elevations of the watershed, and pervasive human-induced perturbations to habitats on >90% of tributaries. The sum of natural conditions and human impairments produce an array of trout assemblages that vary regionally within the watershed and longitudinally among river and tributary reaches. Current land ownership in the Blackfoot watershed is approximately 42% National Forest, 25% private ownership, 19%) Plum Creek Timber Company, 7% State of Montana, and 6% Bureau of Land Management. In general, public lands and large tracts of Plum Creek Timber Company properties comprise large forested tracts in mountainous areas of the watershed, while private lands occupy the foothills and lower valley areas (Figure 10). Traditional land-use in the basin includes mining, timber harvest, agriculture and recreation activities, all of which have contributed to habitat degradation or fish population declines. Excluding the backcounty and the upper Blackfoot mining area, 158 inventoried streams or river reaches, 145 have been altered, degraded or otherwise identified as fisheries-impaired (Pierce et al. 2005, Appendix F). The majority of habitat degradation occurs on the valley floor and foothills of the Blackfoot watershed and largely on private agricultural ranchlands. However, problems also extend to commercial timber areas, mining districts, and state and federal public lands. Distribution patterns of most salmonids generally conform to the physical geography of the landscape, with species richness increasing longitudinally in the downstream direction (Figure 11). Species assemblages and densities offish can also vary greatly at the lower elevations of the watershed. Native species of the Blackfoot Watershed are bull trout (Salvelinus confluentus), westslope cutthroat trout (Oncorhynchus clarki lewisi), mountain whitefish (Prosopium williamsoni), pigmy whitefish (Prosopium coulteri), longnose sucker (Catostomus catostomus), largescale sucker (Catostomus macrocheilus), northern pikeminnow (Ptychocheilus oregonensis), peamouth (Mylocheilus caurinus), redside shiner (Richardsonius balteatus), longnose dace (Rhinichthys cataractae) slimy sculpin (Cottus cognatus) and mottled sculpin (Cottus bairdi). Non-native species of the Blackfoot Watershed include rainbow trout (Oncorhynchus mykiss), kokanee (O. nerka), Yellowstone cutthroat trout (O. clarki bouvieri), brown trout (Salmo trutta), brook trout (Salvelinus fontinalis), artic grayling (Thymallus arcticus), white sucker (Catostomus commersoni), fathead minnow (Pimephales pomelas), northern pike (Esox lucius), brook stickleback (Culaea inconstans), Pumpkinseed (Lepomis gibbosus), largemouth bass (Micropterus salmoides) and yellow perch (Perca flavescens). 20 Bull Trout Westslope Cutthroat Trout K ■ ■ r. Brown Trout X Brook Trout Mountain Whitefish Figure 11. Generalized distribution of six salmonids within the Blackfoot Basin. 21 Most salmonids (WSCT, bull trout, rainbow trout and brown trout) in the main stem river system exhibit fluvial life-history characteristics, whereas tributaries support both migratory and resident populations. Native fishes within the Clearwater basin also exhibit adfluvial life-histories. WSCT has a basin-wide distribution and is the most abundant species in the upper reaches of the tributary system. Bull trout distribution extends from the main stem Blackfoot River to headwaters of larger tributaries north of the Blackfoot River main stem, including the Clearwater River Basin. Rainbow trout distribution is limited to the Blackfoot River downstream of Nevada Creek and lower reaches of the lower river tributaries, with the exception of Nevada Creek upstream and downstream of Nevada Reservoir. Rainbow trout occupy -10% of the perennial streams in the Blackfoot watershed, with river populations reproducing primarily in the lower portions of larger south-flowing tributaries. The exception to this is the upper North Fork within the Scapegoat Wilderness where self-sustaining lake populations have expanded to nearby tributaries. Brown trout inhabit -15% of the perennial stream system with a distribution that extends from the Landers Fork down the length of the Blackfoot River and into the lower foothills of the tributary system. Brook trout are widely distributed in tributaries, but rare in the main stem Blackfoot River below the Landers Fork. 22 PROCEDURES Methods associated with Results Part II, III, IV and VII are identified below; those related to Special Projects are located in Results Part V and VI. Fish Population Estimators Fish were captured using a boat or backpack mounted electrofishing unit. In small streams, we used a battery powered (Smith/Root) backpack mounted DC electrofishing unit. The anode (positive electrode) was a hand-held wand equipped with a 1 -foot-diameter hoop; the cathode (negative electrode), a braided steel wire. On the Blackfoot River, we used an aluminum drift boat mounted with a Coffelt Model VVP-15 rectifier and 5,000 watt generator. The hull of the boat serves as the cathode and two fiberglass booms, each with four steel cable droppers, serve as anodes. We used direct current (DC) waveform with output less than 1000 watts, which is an established method to significantly reduce spinal injuries in fish associated with electrofishing (Fredenberg 1992). Juvenile trout including young-of-the year (YOY) were sampled in the tributaries from August to November. Extra effort was used to sample stream edges and around cover to enable comparisons of densities between years and sampling sections. Captured fish were anesthetized with either tricaine methanesulfonate (MS-222) or clove oil, weighed (g) and measured (mm) for total length (TL). For this report, we converted all weights and lengths to standard units. Fish population surveys relied on mark-and-recapture or multiple-pass depletion density estimators and a simple catch-per-unit-effort (CPUE) statistic for small stream surveys. For the Blackfoot River below Lincoln we used mark-and-recapture density estimator. Using this method, estimates are considered valid if recaptures > 4 fish. We used depletion estimator on the upper-most mainstem of the Blackfoot River (upstream of Lincoln) and for small stream density estimates. Age classes were based on length- frequency analyses. For all Blackfoot River population surveys using mark-and-recapture, we also estimated biomass and calculated condition factor using Fisheries Analysis Plus software (FA +). The formulas for these calculations are: N= (m+l)(c+l)-l r+1 Biomass Estimate = N (Wt) CF=(WtL/LL) 100,000 N= population point estimate m= the number of marked fish c= the number offish captured in the recapture sample r= the number of marked fish captured in the recapture sample CF= condition factor WtL= average weight of length group Ll= average length of length group 23 Standard deviations (SD) for the mark-and-recaputure surveys were calculated using the equation: SD = sqrt {((m+ 1) (c + l)(m- r) (c - r)) / ((r + iy(r +2))} The 95% confidence intervals (CI) were calculated using the equation: 1.96*S) For fish population estimates in small streams, we used a standard two-pass depletion estimator and standard equations for calculating variance (Leathe 1983). For this estimator: N= (nil ni-n2 P = rii-rii Til Where: A^= point estimate, ill = the number offish collected on the first pass «2 = number offish captured on the second pass P = probability of capture (>0.5 for n>50 or >0.6 for n<50 for valid estimates) And, SD = nin^Jni+riTl (ni-n2X And, the 95% confidence interval forN = 1.96 (SD). In those few cases where a three-pass estimator was necessary, we used a maximum likelihood estimator using the Lockwood and Schneider. (2000) formula: N=[n+\ ln-T+ 1] [kn-X-T+ 1 + (k- i) I kn - X+ 2 + (k- i)\ < 1.0 Where n is the smallest integer satisfying Equation. Probability of capture (p) and variance of A^ are then estimated by: P= T kN-X Variance of jV= N(N-T) Where, T^-N{N-T){{kpf l{\-p)-\ N = point estimate / = pass number, k = number of removals (passes), C, = number offish caught in i* sample, X= an intermediate statistic used below, T= total number offish caught in all passes. Standard error ofN= Square root of variance of A^. 24 95% Confidence intervals (CI) were calculated using jV+ 2(Standard error) For initial small stream population assessments, we commonly use a single pass catch-per-unit effort (CPUE) method as a simple index of relative abundance (Appendix A). From monitoring sections with both CPUE and depletion estimates, we also recently developed linear regressions to help predict densities from CPUE (Pierce et al 2004). These regressions confirm correlations between CPUE and density estimates for fish <4.0" (y=1.7236x-0.1513; R^=0.86; P=<0.001) and for fish >4.0" (y=1.3162x + 0.5495; R^=0.86; P=<0.001). Although these regressions demonstrate CPUE to be a simple predictor of population density, estimates derived from these equations do not have a confidence interval like the actual (depletion) population density estimate, and should be used with caution. For this report, we use either CPUE or actual depletion estimates for tributary assessments. CPUE refers to the number offish collected in a single electrofishing pass and is adjusted per 100' of stream (i.e. CPUE of 8 means 8 fish captured per 100' of sampled stream). Actual population estimates are referred to as density/100'. CPUE catch statistics are located in Appendix A. Depletion estimates are located in Appendix B. Mark-and-recapture, biomass and condition factors for the Blackfoot River are located in Appendix C. Stream Temperatures Water temperatures (° F) were recorded at either 48 or 72-minute intervals using Hobo temperature or tidbit data loggers. All raw data plot for each station and monthly summary statistics are located in Appendix H. For this report we also standardized many temperatures summaries using July (the identified period of peak warming) data and display median, quartile and minimum and maximum temperatures values consistent with other (TMDL) temperature summaries within the Blackfoot River Basin. Objectives of the temperature data collections are many, and they include: 1) continue long-term data collections at established monitoring sites; 2) profile temperatures over the length of the river; 3) identify and monitor thermal properties of tributaries entering the river; 4) identify thermal regimes favorable and unfavorable for trout; 5) monitor temperature triggers used in the Blackfoot Emergency Drought Plan; 6) monitor stream restoration projects; and 7) establish winter baseline and influence of upwelling in bull trout spawning area; 8) assess relationships of water temperature to movements of rainbow trout; and 9) compile data for future studies. Stream Habitat Surveys Basic habitat surveys methods for small stream focus on precision, repeatability and efficiency. We sampled a minimum of every third habitat unit and began in a randomly selected downstream habitat unit and proceeded upstream. When habitats were sub-sampled (50%, 33%) or 25%) intensity), we began at a selected pool and measured: 1) maximum pool depth and the downstream riffle crest depth to calculate the residual depth, 2) wetted width and bank-full width at the maximum pool depth and at the riffle crest, and 3) total pool length. Pool frequency was then calculated by measuring the survey distances using either 1:24,000 scale maps or aerial photos. A total census of large wood (> 4" DBH and >6') was performed for all habitat units throughout the entire 25 length of the survey on all streams. Wohlman pebble counts were conducted at a minimum of one representative riffle. Stream discharge was measured using a Marsh- McBirney flow meter at the start location. We also noted overhead canopy and under- story vegetation, stream bank stability, stream degradation and Rosgen channel-type. Whirling Disease Sentinel Cage Studies Whirling disease surveys involving sentinel fish exposures were undertaken throughout the Blackfoot Watershed in 2006 and 2007. Sentinel cage studies are controlled experiments used to detect levels of whirling disease. Cages consist of an 18 x 24" cylindrical screened container placed into a stream site, which allows stream water to flow through the cage. Each cage contained 50 uninfected rainbow trout or WSCT (35- 60 mm) supplied by a state fish hatchery. Timing of field exposure was based on anticipated mean daily temperatures in the 50's (F), which correlates with peak triactinomyxon (TAM) production, and corresponds to peak infection rates in fish (Vincent 2000), except in spring creeks (Kleinschmidt and Nevada Spring Creek) where recent research indicated peak infection occurred in late winter and early Spring (Anderson 2004). The exposure period for each live cage was standardized at 10 days. At the end of the 10-day exposure period, the trout were transferred to Pony, MT, where they were held for an additional 80 days at a constant 50 ° F temperature to insure the WD infection if present would reach its maximum intensity (Vincent 2000). At the end of the holding period, all surviving fish were sacrificed and sent to the Washington State University Animal Disease Diagnostic Laboratory at Pullman, WA. At the lab, the heads were histologically examined using the MacConnell-Baldwin histological grading scale, which ranks infection intensity from (absent) to 5 (severe) (Baldwin et al. 2000). The results of this histological rating were presented as mean grade infection. Mean grade infections above 2.7 are likely to result in population level declines (Vincent 2000). Each sentinel cage also had an accompanying thermograph to establish mean daily water temperatures during the exposure period. WSCT Genetic Investigations In 2006 and 2007, we tested Oncorhynchus genetic composition in WSCT habitat throughout the backcountry and Clearwater basin (Appendix J). Samples consisted of fin-clips taken from a minimum 25 individual fish when possible, unless hybridization was identified phynotypically in which case we relied on a small sample to confirm observations. Samples collected were immediately preserved in 95% ethyl alcohol at stream-side and taken to the University of Montana, Conservation Genetics Laboratory for analysis, where one of two methods of genetic analysis was used. The Paired Interspersed Nuclear DNA Element-PCR (PINE-PCR) method is used in 2006 to determine each fish's genetic characteristics at 21 regions of nuclear DNA. In 2007, a more recent methodology - the "indel" technique-was also employed (Ostberg and Rodriguez 2004: Appendix J). Both methods distinguish WSCT, from rainbow trout and Yellowstone cutthroat trout and can be used to determine whether a sample came from a suspected genetically pure population of one of these fishes or one in which hybridization between two or all three of them has occurred. With a sample size of 25 fish, the PINES- PCR method has a 95% chance of identifying as little as \% introgression; whereas, the newer indel method has a 99%) chance of detecting as little as \% introgression with the 26 25 fish sample. Working with Private Landowners Typically, each tributary restoration project involves multiple landowners, professional disciplines, fianding sources, and involvement of the watershed groups. Restoration has focused on addressing obvious impacts to fish populations such as migration barriers, stream de-watering, fish losses to irrigation canals, and degraded riparian areas. All projects are cooperative endeavors between private landowners and the restoration team, and occur throughout the drainage. Projects are facilitated at the local level by agency resource specialists in cooperation with two watershed groups (BBCTU and BC) or local government groups such as the North Powell Conservation District (NPCD) or state and federal agencies such as the Montana Fish, Wildlife and Parks (FWP) or U.S. Fish and Wildlife Service, Partners for Fish and Wildlife (USFWS). The non-profit (501(c)3 status of watershed groups provide a mechanism for generating tax-deductible private funds. FWP biologists identify priorities by performing fisheries studies, communicating biological findings, review proposed fisheries projects, provide funding support and monitor fisheries on completed projects. Federal (USFWS, USPS and NRCS) biologists and other agency specialists (BOR, DNRC) help develop and fund projects usually in conjunction with watershed groups (BBCTU) and landowners and FWP. Agency staff and project leaders generally enlist help from interagency personnel or consultants including range conservationists, hydrologists, engineers, and water right specialists as necessary. Watershed groups (NPG) help with fundraising, administration of budgets, bid solicitation, apply for permits, oversee consultants and contractors, assist with landowner contacts, coordinate volunteers, help resolve local conflicts and address other social issues. Project funding comes from many sources including landowner contributions, private donations, foundation grants, state and federal agencies. Project managers from the agencies and watershed groups jointly undertake fundraising. BBCTU generally obtains project permits on behalf of cooperating landowners. Project bids (consulting and construction) conform to State and Federal procurement policies. These policies included the development of a Blackfoot watershed qualified vendors lists (QVL) derived through a competitive process managed primarily through BBCTU. A minimal project cost triggers use of the QVL. The watershed groups solicit bids from the QVL for both consulting and contractor services. Bid-contracts are signed between the watershed group and the selected vendor upon bid acceptance. Depending on the specific project, landowners are responsible for certain costs, construction and project maintenance. Addressing the source of stream degradation usually requires developing riparian/upland management options sensitive to the requirements of fish and other riparian-dependent species. Written agreements (10-30 year period) with landowners to maintain projects are arranged with cooperators on each project. These agreements vary by funding source and may include agencies, the NPCD and/or the Fish and Habitat Committee of BBCTU. Landowner awareness of the habitat requirements offish and wildlife, and their full participation and commitment to project goals and objectives are crucial to the long-term success of the restoration initiative. We encourage landowners to participate fully in all phases of restoration from fish population 27 data collection and problem identification to project development and monitoring of completed projects. Although many restoration projects have been completed in the Blackfoot River watershed, this effort is still educational at a broad level and is far from complete. Natural Channel Design and Fish Habitat Restoration (from Brown, Decker, Pierce and Brant 2001) Habitat restoration relies on both passive and active methods. Passive methods rely on riparian management changes by addressing the source of fisheries impairments, which generally require incorporating grazing BMPs in degraded riparian areas, shrub plantings, enhancing instream flows and screening irrigation ditches. Active restoration involves entering the channel with machinery and reconstructing severely damaged channels or restoring habitat features (e.g. wood) if necessary for fisheries improvement. For channel reconstruction and habitat restoration in the Blackfoot River drainage, we rely on a natural channel design philosophy (NCDP). This philosophy requires a multidisciplinary approach to stream restoration along with an understanding of historical riparian land use. Project complexity and risk define a specific combination of design methods. Methods involve a geomorphic approach that fits the proper stream to the proper stream valley. The Rosgen stream classification provides the basis of this approach (Rosgen 1994; Rosgen 1996). NCDP quantifies channel shape, pattern, and gradient (Rosgen 1996). Riparian health, instream habitat, and fish population surveys, along with measurements of discharge, sediment, and bed and bank stability, permit the assessment and evaluation of existing and potential channel conditions as well as biological attributes of the project. The NCDP aims to restore natural channel stability, or dynamic equilibrium, and habitat to impaired streams. Streams in dynamic equilibrium are generally more biologically productive, and provide higher quality and more complex habitat than altered or unstable streams. Geomorphic indicators (bankfull channel), prediction analysis (reference reaches and dimensionless ratios), and method validation (regional curves) define naturally functioning channels, and provide the basis for natural channel design. At the reach level, stream geomorphology is quantified in both project and reference reaches. The reference reach should be naturally functioning, provide optimal fish habitat, and serve as a model for the design channel. "Bankfull" indicators and other geomorphic variables are measured in both reaches. Bankfull elevation, a geomorphic indicator signifying the point of incipient flooding, coincides with the stage above which the stream accesses its floodplain or flood-prone area (Rosgen 1996). By doing the work that creates the average morphologic channel characteristics, bankfull discharge forms and maintains the channel over time (Dunne and Leopold 1978). Channel pattern (plan view characteristics), dimension (channel size and shape), and profile (longitudinal elevations and gradients) are measured. Appropriate designs may include creating aquatic habitat, prescribing a revegetation plan, and constructing an appropriate floodplain. Synthesizing reference reach field data and incorporating regional stream information helps identify design channel parameters. Regional data and dimensionless ratios help predict channel attributes relative to the watershed area and bankfull characteristics. Watershed discharge, sediment entrainment, and bankfull channel cross 28 sections are then hydraulically modeled to validate bankfull discharge. Design dimensions are developed relative to bankfull discharge. Comparing design dimensions to dimensionless ratios and a reference reach database further validates the design. The final restoration design seeks to mimic a stream in dynamic equilibrium with its watershed, and to provide a diverse and complex channel capable of conveying flows, transporting sediment, and integrating essential habitat features related to fish population recovery goals. Vegetation colonization through mature shrub and sod mat transplanting, as well as other revegetation efforts; along with woody materials and rock provide immediate fish habitat and temporary bank stability. These structures allow for shrub colonization which, when established, provides long-term channel stability and habitat complexity. Proper land management is essential to the success of these methodologies. Most restoration projects necessarily incorporate compatible grazing strategies and other land management changes. 29 RESULTS PART I: BLACKFOOT RIVER ENVIRONMENT Blackfoot River Discharge: Provisional USGS data at the Bonner gauging station #12340000 During 2006 and 2007, the Blackfoot River watershed was subject to a 7**^ and 8* During this time, mean basin discharge was 1,480 cfs (in 6000 year of consecutive drought 2006) and 1,254 cfs (in 2007), compared to the long-term mean of 1,563 cfs (Figure 12, Table 1). Since the advent of current drought (in 2000), annual flows in the Blackfoot River at the USGS Bonner gauging station averaged 83% (range, 60 - 95%) of normal; minimum monthly flows ranged from 47-80% of normal; and late summer low flows (July and August) averaged 62-71% of normal. August) fell to 47 and 61 % of normal (Table 1). regime of the Blackfoot Basin has expressed a pattern of early runoff, a consistent lowering of "flushing" flows and below normal base-flows during the late summer. Based on wetted-riffle surveys, FWP identified a "minimum instream flow" value Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Figure 12. Blackfoot River monthly mean discharge at the USGS Bonner, MT gauging station: a comparison of the historical (1898-1999) mean and flows from 2000 through 2007. In 2007 mid-summer flows (July and Through this 8 -year drought, the flow Mean 1898-1999 554 599 781 2052 4867 4877 1844 831 670 653 649 606 2000 542 532 744 2770 3741 2779 1004 528 521 608 517 450 2001 445 421 560 959 2980 2595 996 584 441 472 520 477 2002 455 429 503 1980 4036 5814 1939 788 600 517 487 439 2003 441 553 881 2885 4081 3579 970 554 491 467 520 471 2004 458 438 932 2471 3402 2838 1181 635 670 650 625 598 2005 670 668 726 1626 3690 3178 1167 579 504 504 518 841 2006 530 460 604 3279 5217 3379 1019 558 497 497 1097 633 2007 542 559 1531 2025 4365 2545 861 511 461 483 503 672 Mean 2000-07 510 508 810 2249 3939 3338 1142 592 523 524 598 573 % of (1898-1999) normal 92 85 104 109 81 68 62 71 78 80 92 95 Table 1. Provisional mean monthly flow statistics for the Blackfoot River at the USGS Bonner gauge: the historical (1898-1999 mean and monthly flow statistics from 2000-2007). 30 for "blue-ribbon" lower Blackfoot River near Bonner, MT at -700 cfs (FWP historical files). Below 700 cfs, water recedes from riffles of the lower River at an accelerated rate (Figure 13), and this process reduces productivity and the ability of the river to sustain productive fisheries. Flow records identify the Blackfoot River fell below 700 cfs an average of 221 days (61%) (range = 189-259 or 52-71%) per year from 2000 through 2007, compared to a historic (1898-2007) mean of 192 days per year. Wetted Perimeter (FT> M» ' 700 QFS y Year 2000 number days below 700 cfs 236 200 ■ f 2001 236 J 2002 219 /^ 2003 259 c 2004 207 2005 202 2006 189 100 ■ 2007 221 1000 2000 3000 4000 Flow (cfs) Figure 13. Minimum instream value of 700 cfs for the Blackfoot River based on wetted-perimeter riffle surveys (FWP historical files). The table shows the number of days Blackfoot River flows at the Bonner guage fell below minimum (700 cfs) instream flows from 2000 through 2007. Blackfoot River and tributary temperatures Temperatures studies during 2006 and 2007 involved: 1) baseline and long-term data collections at established sites throughout the Blackfoot watershed; 2) assessments of restoration projects; 3) identifying thermal regimes (natural and anthropogenic) favorable and unfavorable for trout; 4) monitoring temperature triggers of the Blackfoot Emergency Drought Plan; 5) relating rainbow trout migrations and spawning to thermal properties of the river and tributary system; and 6) predictions of whirling disease. Summaries and integration of water temperature data are found throughout this report. All raw and summary data for all monitoring sites are also located in Appendix H. In sum, we collected 60 water temperature samples, including 48 individual samples at 25 tributaries sites (Appendix H, Results Part III), along with 12 samples at 6 31 sites in the Blackfoot River during 2006 and 2007 (Figure 14). These data sets show a wide range of seasonal, spatial and inter-annual summer temperatures for waters of the Blackfoot River. This data includes temperatures commonly >70 °F, which we define (in this report) above the optimal range of most salmonids and temperatures >65 °F, which are considered harmful to bull trout. A summary of all July water temperature data for five long-term monitoring sites of the Blackfoot River is shown on Figure 15. These plots identify a recent warming trend, as well as reach-related temperature differences such as warming of the Blackfoot River between the Cutoff and Raymond Bridge sections, as well as cooling influence of the North Fork (mile 54.1) between the Raymond Bridge and Scotty Brown sections of the Blackfoot River. H Tributary sites LJ Mainstcm sites Figure 14. Temperature data collection sites in the Blackfoot Watershed for 2006 and 2007. Names identify Blackfoot River monitoring sites and relate to July summary graphs in Figure 13. 32 Blackfoot River above Belmont Creek Blackfoot River at USGS Bonner Guage 1996* 1997* 1999* 2001 2002* 2003 2004* 2005* 2006 2007 1996* 1998* 2001 2002 2003 2004 2005 2006 2007 Blackfoot River at Scotty Brown Bridge 80 75 _70 U- 5 65 CO S.60 E 6.0") in the Johnsrud section was 69% rainbow trout (n=722), 17% brown trout («=180), 12%) WSCT (n=l29) and 2% bull trout (n=20). The total trout point estimate (fish >6.0") for the Johnsrud section increased from 130 to 187 fish/1000' an increase of 30%) between 2004 and 2006. The total trout biomass estimate for fish >6.0" in the Johnsrud Section in 2006 was 84.9 lbs/1000' compared to 77.6 lbs/1000' in 2004. Densities of native WSCT (> 6.0") increased from 14.0 to 23 fish/1000' (Figure 17). Because of small sample size and a low recapture rate, we were unable to generate a valid bull trout estimate. However, catch statistics suggest lower densities compared to past years. The 2004 point estimate for brown trout (> 6.0") showed an increase from 19 in 2004 to 22 fish/1000' in 2006 (Figure 17). The density estimate for rainbow trout (>6.0") indicates a notable increases from 90/1000' in 2004 to 140/1000' in 2006 (Figure 15). In 2006, we observed no northern pike in the Johnsrud section, compared with one in 2004, two in 2002, six in 2000, two in 1998, one in 1996, and none prior to 1996. Scotty Brown Bridge section: The 2006 percent trout composition in the Scotty Brown Bridge Section was 31% rainbow trout («=166), 33% WSCT («=176), 29% brown trout («=156), and 7 % bull trout («=40). Total trout densities (fish >6.0") increased from 50.0 to 62.0 fish/1000' between 2004 and 2006. The total trout biomass estimate for fish >6.0" in the Scotty Brown Section in 2006 was 52.2 lbs/1000' up slightly from 48.1 in 2004. Density estimates for rainbow trout (fish >6.0"), shown in Figure 17, identify an increase from 9.0 in 2004 to 23 fish/ 1000' in 2006. Likewise, brown trout (fish >6.0") density increased from 10 fish/ 1000' in 2004 to 16 fish/ 1000' in 2006 (Figure 17). Estimated bull trout densities (fish >6.0") increased from 2.0 to 4.0 fish/1000' between 2004 and 2006 (Figure 17). WSCT densities (fish >6.0") were static at 19 in 2006 compared to 18 fish/1000' in 2004 (Figure 17). Despite these modest increases, densities of all species remain below pre-drought (e.g. 2000) population levels (Figure 17). Middle Blackfoot River Survey Sections Wales Creek Section: This section, located between the North Fork Blackfoot River and Nevada Creek suffers from impaired water quality (high levels of fine sediment, summer water temperatures, and nutrient levels) and degraded tributaries, which limits juvenile trout production and recruitment to this reach of the Blackfoot River (Pierce et al. 2001; 2004, 2007). In May 2006, trout species composition (%) of total catch for fish >6.0") in the Wales Creek section was 87%) brown trout («=1 17), 1% rainbow trout («=9), 4 % WSCT («=5) and 2 % bull trout («=3). We estimated total trout density (fish > 6.0") for the Wales Creek section at 1 1 fish /lOOO' in 2006 compared to 9. 1 in 2004. Of the total trout estimate, the brown trout (fish > 6.0") point estimate was 8 fish/1000'. Similar to past years (2002 and 2004), we were unable to generate density estimates for rainbow trout, WSCT and bull trout due to the low population densities within the section. The total trout biomass estimate for fish >6.0" in the Wales Creek in 2006 was 11.7 lbs/1000' In 2006, we also sub-sampled for mountain whitefish in the Wales Creek section 35 Johnsrud Section Scotty Brown Section Rainbow Trout Rainbow Trout 1989 1990 1991 1993 1996 19 2000 2002 2004 2006 1990 1991 1993 1996 1998 2000 2002 2004 2006 Brown Trout Brown Trout 1989 1990 1991 1993 1996 1998 2000 2002 2004 2006 Westslope Cutthroat Trout 1989 1990 1991 1993 1996 1998 2000 2002 2004 2006 Westslope Cutthroat Trout 1989 1990 1991 1993 1996 1998 2000 2002 2004 2006 35 11 10 30 9 () 8 ^ 2t) SI / O) o -i ','0 6 b (/) g 1R F J? ■^ o t/i 1(1 m c 2 1 u 5 16 n 14 o 12 ^ s 10 + o 8 o o B 1989 1990 1991 1993 1996 1998 2000 2002 2004 2006 Bull trout 1989 1990 1991 1993 1996 1998 2000 2002 2004 2006 Figure 17. Blackfoot River fish population density and biomass estimates (fish >6.0") for the Johnsrud Section (left column) and Scotty Brown Section (right column), 1989 to 2006. 36 to help identify population status of the species within the Blackfoot River. These results are located in the mountain whitefish status review. Results Part V. Canyon section: Fish populations in the Canyon section were sampled in 2006, 1999, 1988 and 1971. In 2006, brown trout (n=3\) continued to dominate the salmonid community by comprising 80% of the sample, followed by 20% WSCT («=8). Bull trout were present in low numbers in 1999, but absent in the 2006 survey. The total trout estimate (fish >6.0") in the Canyon Section declined from 52 fish/1000' in 1999 to 15 fish/1000' in 2006. Similar to previous samples, a reliable estimate of age I brown trout was not attained in 2006. The density estimate of the Age II and older brown trout (fish > 8.5") identified large decline from 33 fish in 1999 to 7.2 fish/1000' in 2006 (Figure 18). Similar to the Wales Creek section, we surveyed mountain whitefish in the Canyon section for the first time. The results of that targeted survey are reported in the mountain whitefish status o o o o 50 45 40 35 30 25 20 15 10 5 L i'' '-.i 1 1 \ ' ' 1971 1988 1999 2006 Figure 18. Estimated densities of age-2 and older brown trout in the Canyon Section of the Blackfoot River, 1971, 1988, 1999 and 2006. review in Results Part V. 200 - 180 o 5s160 S 140 ^ 120 100 I 80 = 60 01 a Poorman/Dalton section: Fish populations were sampled in the Pooman/Dalton section (mile mid-point at 107.2) in 2006 for the first time since 1988. In 2006, brown trout (n=l92) continued to dominate the salmonid community followed by WSCT (n=l4), bull trout {n=2) and brook trout (n=l). The total trout estimate (age 1 and older) increased from 50 to 60 fish/1000' in 2006 (Figure 19) periods of protracted drought. 40 20 WSCT Brook trout trout Brown trout -3 At irt CM oo oo CO o o CM 1^ CM oo oo Oi CO o o CM 1^ CM at oo CO oo o a> o T- CM Figure 19. Estimated densities of age 1 and older WSCT (left), brook trout (middle) and brown trout (right) in the Poorman/Dalton section of the Blackfoot River 1971, 1972, 1988 and 2006. Both 1988 and 2006 samples were taken following 37 Upper Blackfoot River Surveys: Upper Blackfoot Mining Complex This section summarizes FWP fisheries investigations upstream, within and downstream of the Upper Blackfoot Mining Complex. This summary includes not only the most recent (2006) fisheries investigations, but also the long-term (>30 years) fisheries changes, some of which pre-date the release of toxic mine waste into the headwaters of the upper Blackfoot River (Spence 1975). For a full summary of mining- related aquatic resource damages in the upper Blackfoot River Basin, the reader is referred to the Stratus (2007) resource damage report to the State of Montana Natural Resource Damage Program, Department of Justice. Introduction Several hard-rock mines are located within Upper Blackfoot Mining Complex (UBMC), located about 15-miles east of Lincoln, of which the Mike Horse Mine is the largest. Beginning in 1898, the Mike Horse Mine produced lead, zinc, and copper, and in 1941 the Mike Horse tailings dam was constructed on lower Beartrap Creek using contaminated mine tailings. About 30-years after that and in anticipation of expanded mining, FWP established a baseline fisheries inventory in the UBMC in 1971-1973 (Spence 1975). In June 1975 during a heavy "rain on snow" flood event, the Mike Horse tailings dam washed-out and released an estimated 200,000 yd^ of contaminated tailings into Figure 20. Photo taken immediately downstream of the Mike Horse tailings dam following its collapse in 1975 (photo by Spence 1975). 38 Beartrap Creek, the upper Blackfoot River and adjacent floodplains (see photo Figure 20). The breach of the Mike Horse tailings dam and associated mining wastes within the UBMC resulted in 1) the acute and chronic contamination of the upper Blackfoot River (Stratus 2007), 2) the collapse of fisheries (Spence 1975, Peters and Spoon 1989, Pierce et al 2000), and 3) the physical downstream migration of heavy metals and the biological uptake of toxins within the aquatic food web (Moore et al 1991). The upper-river location and toxic nature of existing contaminants continue to pose significant ecological risks to the future of the mainstem Blackfoot River (Stratus 2007). In anticipation of a meaningful cleanup, FWP completed a series of fish population surveys throughout the UBMC in 2006. Our purpose was to identify the spatial and temporal nature of fish population declines, identify the status of nearby populations and provide information to help guide settlement discussions and aid in restoration planning in a manner that helps recover fisheries within the impact area. Study Area The very headwaters of Blackfoot River begin within the UBMC at the junction of Beartrap and Anaconda Creeks (Figure 21). In 2006, FWP fisheries crews surveyed 11 new sites within the immediate and known impact area, and we resurveyed two historical monitoring sites downstream of UBMC that were established prior to the failure Mike Horse tailings dam (Spence 1975, Figure 20). Pass Oreek ^ iP«s Ploce Section Fl«$cher Section Shove &tJkh MikeJHo rse tailings dam ■^— Fah Population survey sffes Figure 21. Study area: Upper Blackfoot River and tributary fish populations survey sites completed in 2006. 39 30 20 10 Methods We performed single pass electrofishing CPUE surveys at all sites and depletion population estimate surveys at two downstream long-term monitoring (Flescher and Hogum) sections (see Procedures Section). We also established an undisturbed "control" to compare native fish densities, and to help identify the current level fisheries impairments in the upper Blackfoot River. The control, located on upper Cottonwood Creek (Blackfoot Clearwater Game Range) was surveyed at the same time as the upper river, using identical methods. The control contained a similar fish community, and similar valley and channel types as both the Hogum and Flescher sections. In addition to fish population sampling, we collected water chemistry readings (pH, TDS and conductivity) at all sites (Appendix I). 22.C 23.5 13.3 '■ T ^ I — -^ — J 0.2 T 1 1 1 ] • 1 ,<3^ # 30-year trend of decreasing WSCT densities in the upper Blackfoot River. The comparison of the Cottonwood Creek control to the 2006 densities in the Blackfoot River (Flescher and Hogum sections) is presented in (Figure 24). This comparison shows significantly lower densities in the impact verses the control sites. 3M-| T E r" 1 IDd a iM ^^ *-l QtnnnadCiih iKibBl 2 n^m D jc^jb^il^ "fc^n 5 Cotton^vuKJ Creek BacMbol g) Flesher BlackfeQt gs Hogum Figure 24. WSCT density (left, all ages) and bull trout density (right, all ages) in Cottonwood Creek (control site) compared to the Flescher and Hogum reaches of the upper Blackfoot River, September 2006. Discussion The UBMC has been identified with high concentrations of arsenic, cadmium, lead, copper manganese and zinc. These toxic substances are found in the surface water, floodplains and groundwater and within the biota (Stratus 2007). Fish population surveys in this and earlier studies show the area of mining impacts begin in the mining complex and include the headwaters of certain perennial tributaries (Paymaster, Beartrap and MikeHorse Creek) as well also the lower reaches of Anaconda Creek and extend down the Blackfoot River from the headwater to Flesher and Hogum sections. Unlike other streams where water chemistry (pH) was tested. Paymaster Creek tested as acidic with a pH as low as 4.0 and was identified as fishless. Relatively unimpacted headwater stream upstream of the mining area still contain predominantly WSCT populations. Although populations have been greatly reduced, exisiting headwater populations have potential for expansion in the dowsntream direction under suitable clean-up conditions. In addtion to stream-resident populations, a proper clean-up has potential to benefit migratory (fluvial) WSCT based on the location of spawning sites identified in a recent FWP telemetry study (Pierce et al 2007, Results Part VI)). 41 FWP has periodically monitored fish population in the UBMC since the eariy 1970s (Spence 1975, Peters and Spoon 1989, Pierce et al 2000). These surveys identify a substantial long-term decline in age 1 and older WSCT. In the Pop's Place sample site, near the marsh area in the UBMC, the estimated density of age 1 and older WSCT in 1971, before the Mike Horse tailings dam failure in 1975, was approximately 100 fish/1,000' (Spence 1975). In 1988, after the release of toxic Mike Horse tailings into the Blackfoot River, the WSCT population declined to an estimated 15 fish/1,000' (Peters and Spoon 1989). By 1999, no age 1 and older WSCT were sampled at the Pop's Place monitoring section (Pierce and Podner, 2000). In 2006, we attempted to access the Pops Place section, but were denied access to the survey site. Few young of the year (YOY) WSCT have been found at Pop's Place during any of the sampling events, including the 1971 sampling prior to the dam failure. Given the multitude of contaminant sources in the UBMC, it is likely that mine-related stress affected the age WSCT at least as early as 1971 (Stratus 2007). Populations of non- native adult brook trout, which typically are not as sensitive to metals toxicity as are the native cutthroat trout (Nehring and Goettl, 1974), did not decline substantially at Pop's Place between 1971 and 1999. Population declines in the Flescher Section are consistent with the upstream spatial and temporal patterns of declining WSCT densities in the downstream direction. Before the 1975 tailings dam breach, the estimated WSCT density at Flesher section was 69 age 1 and older fish/1,000'. Shortly after the dam breach (1975), WSCT density dropped to 30 fish/1,000' (Spence 1975). By 1988 the WSCT density declined further to an estimated 15/1,000' (Peters and Spoon 1989), and in 2006, the estimated WSCT density was less than 9/1,000 feet. By contrast, adult brook trout densities at Flesher were relatively stable from 1973 to 2006. In 1973, the WSCT densities at Flesher section were nearly three times higher than the non-native brook trout density. The decline in the WSCT population, coupled with little change in brook trout density, suggest a species- selective toxicity, which has led to a dominance of the non-native trout, albeit at low densities. Cottonwood Creek was selected as a background area to estimate baseline conditions for the upper Blackfoot River. Fish population data were collected in the upper Blackfoot and in Cottonwood Creek in September 2006 using the same field methods and in similar environments. The native trout populations in the Flesher and Hogum reaches of the Blackfoot River were substantially lower than the populations in Cottonwood Creek. The densities of WSCT (all ages) in both the Flesher and Hogum reaches of the upper Blackfoot River were over five times lower than WSCT densities in Cottonwood Creek control site. Bull trout are also reduced in the Blackfoot River compared to Cottonwood Creek. The density of bull trout (all ages) in Cottonwood Creek was >15/1,000'. By contrast, in the upper Blackfoot River, only one bull trout was found in the Flesher and Hogum reaches combined, a length of 6,500'. Brook trout were present at all locations; the density of brook trout (all ages) at Cottonwood Creek (22/1,000') was between the density in the upper Blackfoot at Flesher (52/1,000') and Hogum (5/1,000'). Whether through direct toxicological effects or sub-acute impacts such as avoidance, it is highly likely that the reduced native trout populations downstream of the UBMC are the result of releases of hazardous substances from upstream mine sites. 42 Fluvial WSCT are currently in very low abundance in the upper Blackfoot River. A recently completed telemetry study identified extensive spawning migrations of adult WSCT telemetered downstream of Lincoln to upstream spawning sites (see related study in Results Part VI, Pierce et al 2007). This study identified fluvial WSCT spawning in the upper Blackfoot River near the Hogum and Flescher survey sections in the area of mining-related declines. These results identify mine-related impacts not only resident WSCT at a local scale but also fluvial WSCT stocks over a much broader area. These findings suggest the recovery of fluvial WSCT stocks should be fully considered as primary (habitat) targets when developing specific restoration actions in the UBMC. 43 RESULTS PART III: RIVER RESTORATION TRIBUTARY ASSESSMENTS Results Part III summarizes recent (2006-07) inventory and monitoring activities for 29 restoration project streams. Nine previous FWP reports spanning the 1988-2005 fisheries investigations provide additional information to these and earlier restoration endeavors (Peters and Spoon 1989, Peters 1990; Pierce and Peters 1991; Pierce, Peters and Swanberg 1997; Pierce and Schmetterling 1999; Pierce and Podner 2000; Pierce, Podner and McFee, 2001; 2002: Pierce, Anderson and Podner 2004, Pierce and Podner 2006). All fish population survey results (catch and size statistics and density estimates) for these 29 streams are located in Appendices A and B. Summaries of all related data (discharge, temperature, water chemistry, and WSCT genetics) collected during the 2006- 07 period are located in Appendices D, H, I and J, respectively. The locations of individual study streams and fish population survey sites are located below in Figure 1. Ashby Cr: Figure 1. Location map with 29 restoration-related study streams and fish population survey sites undertaken in 2006 and 2007. Ashby Creek Restoration objectives: Protect the genetic purity of a WSCT population in the upper Ashby Creek watershed by using an existing wetland complex as a migration barrier, and improve WSCT habitat by creating a natural channel that provides complexity, increases riffle-pool habitat features and available spawning substrate and increase shade and small diameter wood recruitment to the stream channel. Improve and re-establish wetland functionality. 44 -.nd Project Summary Ashby Creek is a 2"" order tributary to Camas Creek in the Union Creek basin that supports a genetically pure WSCT population. The upper reaches of Ashby Creek originate in forested areas on Plum Creek and BLM properties before entering private ranch lands near mile 3.0. Historical and recent agricultural practices have significantly altered Ashby Creek. Alterations include diversion for irrigation and channel ditching to increase farmable acreage, livestock degraded stream banks, loss of woody plant communities, irrigation- related dewatering and the draining of downstream wetlands. Over the past several years a comprehensive restoration project plan was developed, and finally implemented in 2007. The project included 1) reconstruction of three miles of stream that had been historically ditched, 2) enhanced instream flows, 3) improved upstream fish passage and the installation of a fish screen at a diversion point, 4) riparian grazing changes, and 5) shrub plantings. This project also connected Ashby Creek to an 80-acre scrub-shrub wetland in a manner that is designed to inhibit the upstream movement offish (Figure 2). Fish Populations and other monitoring activities In 2005, we established pre-project fish population survey sites at an upstream reference at mile 4.0. In 2007 we also established two fish population survey sites in the new channel at mile 3.0 and a downstream survey site at mile 2.0. At this early stage of post-project monitoring, we have identified no fish within either of the two new (treatment area) population-monitoring sites (Appendix A). Bear Creek Restoration Objectives: Restore habitat degraded by historical activities in the channel, restore fish passage and thermal refugia, and improve recruitment of trout to the Blackfoot River. Figure 2. Ashby Creek stream channel restoration project area and fish population survey locations, 2007. 45 2nd_ Restoration rli t -ih- rfi Project Summary Bear Creek, a small order tributary to the lower Blackfoot River, flows six miles north to its mouth where it enters the Blackfoot River at rm 12.2 with a base-flow of 3-5 cfs. Bear Creek is one of the colder tributaries to the lower Blackfoot River. Located on industrial forest and agricultural lands. Bear Creek has a long history of adverse habitat changes, which include placement of undersized culverts, road drainage and siltation, irrigation, channelization of the stream, excessive riparian grazing and streamside timber harvest (Pierce et al. 1997; Pierce and Schmetterling 1999). At least one road crossing is still considered a barrier to movement. Prior to restoration activities, these fisheries impairments contributed to the loss of migration corridors and the simplification and degradation of salmonid habitat. Many of these impairments were corrected in the late 1990s, and these included: 1) upgrading culverts and addressing road drainage problems; 2) improving water control structures at irrigation diversions; 3) reconstructing 2,000' of channel; 4) enhancing habitat complexity on an additional 2,000' of stream; 5) shrub plantings; and 6) the development of compatible riparian grazing systems for one mile of stream. 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Figure 3. Densities of age 1 and older rainbow trout for Bear Creek at mile 1.1, 1998-2007. Fish Populations and other monitoring activities Bear Creek supports predominately rainbow trout and lower numbers of brown trout and brook trout in the lower stream, along WSCT in the upper basin and very low densities of juvenile bull trout. Following restoration actions in 1999, fisheries monitoring has identified Bear Creek as an increasingly important spawning and rearing tributary to the lower Blackfoot River sport fishery. In 2006 and 2007, we continued to monitor fisheries in a reconstructed section (mile 1.1) of Bear Creek (Figure 3, Appendix A and B). This monitoring has identified a general trend of increasing densities of juvenile rainbow trout. We also tested for whirling disease at mile 1.1 in 2006 and found no infection. ind Blanchard Creek Blanchard Creek is a small 2^" order tributary to the lower Clearwater River entering at mile 2.9. It suffers from a long history of adverse activities resulting in riparian and fish habitat degradation. These include changes to the hydrograph (12% above natural) related to extensive timber harvest (DNRC unpublished data), side-casting of road grade materials to the stream channel by Missoula County road maintenance crews, excessive livestock degradation to the riparian area, channel instability and chronic dewatering through irrigation. Lower Blanchard Creek was the site of a water 46 lease during the 1990s that provided minimal (3 cfs) instream flows. Enhanced base flows led to a large increase in the abundance of juvenile rainbow trout, which identified Channel measurements stream channel length ( ft ) 12,56eft Sinuosity 1.11 Total # Pools 157 Total # Riffles 157 # Pools Sampled 52 # Riffle Sampled 52 Pool Frequency 125/1 000ft Riffle Frequency IZ^IOOOft # Pools Measured with UNO 21(40%) # Riffle Measured with LWD 10(19%) Pool Length Wetted Pool Surface Area Maximum Pool Depth 29 + 15(9-77) 475 + 272(59-1367) 2.0 + 0.5(1.0-3.1) Riffle Length Wfetted Riffle Surface Area Maximum Riffle Depth 29 + 24(4-113) 512 + 441 ( 41- 2083) 0.8 + 0.2(0.4-1.2 ) Wetted Pool Width @ Max Depth 15 + 5(7-33) Wetted Riffle Wdth @ Max Depth 18 + 5(9-31) Wetted Pool Width @ Riflle Crest 18 + 5(6-36) Bankfiill Wdth @Max Riffle Depth 32 + 11(14-68) Bankfiill Width @ Max Pool Depth 25 + 7(10-49) Bankfiill Width @ RifHe Crest 31+11(14-69) Riflle Crest Depth 0.6 + 0.1(0.3-1) Residual Pool Depth 1.3 + 0.5(0.5-2.4) Mean Wetted Wdth of Pools 17 + 5(7-33) All measurements in standard (ft) units Table 1. Stream habitat survey results for the lower 2.3 miles of Blanchard Creek. Blanchard Creek as a productive rainbow-spawning stream when minimum instream flows were maintained (Pierce et al 2004). The water lease was recently terminated, and lower Blanchard Creek now is again chronically dewatered during the irrigation season. Lower Blanchard Creek is now the site of a high-density subdivision proposal. Fish Populations and other monitoring activities In 2006 and 2007, we completed a stream habitat survey from the mouth to the confluence of the North Fork (at mile 2.3). We also conducted fish population surveys at three headwater locations (mile 3.3, 5.6 and 9.4). Results of the habitat survey are located in Table 1. Rainbow trout were present only at mile 3.3 (CPUE = 4.3) and WSCT increased in the upstream direction (Appendix A). Braziel Creek Braziel Creek drains a small watershed (-4.5 miles^) located along the southeastern foothills of Hoodoo Mountain, just south-southeast of Helmville. This 2"''- order tributary is -3.9 miles in length, maintains a mean gradient of 4807mile, and generates an estimated base-flow of 0.5-1.0 cfs before entering Nevada Creek at mile 24.5 about 2.0 miles downstream of the Nevada Creek reservoir. The upper 2.1 miles of Braziel Creek flows through Bureau of Land Management land and the lower 1.8 miles of stream flows through private ranch lands (Figure 4). 47 Livestock Elevation grazing has heavily impacted the riparian under- story vegetation of alder and grasses on lower Braziel Creek. Below the ELM properties, the riparian over-story consists of sparsely mixed community of ponderosa pine and Douglas fir. Intense livestock hoof-shear has left the stream banks and channel heavily degraded, over- widened and shallow. Fish habitat consisted of small pocket pools created by large boulders, small woody debris jams, undercut banks and young overhanging conifers. Large woody debris recruitment to the stream channel is minimal. A culvert observed at the West Fork road crossing (mile 1.5) was perched approximately 10-12 inches, thereby creating a possible fish barrier. Our observations of the Braziel Creek riparian area on ELM properties identified a healthier riparian area and stable banks. 7000 6500 ■■ 6000 ■ ■ 5500 - ■ 5000 ■ ■ 4500" 4000 ■■ 3500 - ■ 3000 ■ - 2500 * Fisheries Survey Locations Hoodoo Mtn A Headwaters ______ -■ West Foi k Braziel Creels y.--"^ Un-iumied Tril)iiaiv / ^^ Sec 1 Sec 2/ ^^j,^ NeviKlii Creeh; Wests! ope Cutthroat Trout Perennial Stream 1 lirterinittent Piiv.ite Ranch Land 1 1 ■ ' 1 BLH Stream Mileaye Figure 4. Longitudinal profile for Braziel Creek. 14 12 10 Fish population monitoring and other monitoring activities Fish population survey sites were established at two sites (mile 0.7 and 1.4) on lower Braziel Creek in 2006. A comparison of the CPUE for age 1 and older WSCT between the sites identify noticeably lower densities at the lower site compared to the upper site (Figure 5). WSCT YOY were common at both sampling sites. Sculpins were abundant at mile 0.7, but absent from the mile 1.4 survey site. « 6 IVIile 0.7 IVIile1.4 Figure 5. CPUE for age 1 and older WSCT at two locations on Braziel Creek, 2006. Chamberlain Creek Restoration objectives: Improve access to spawning areas; improve rearing conditions for WSCT; improve recruitment of WSCT to the Blackfoot River; provide thermal refuge and rearing opportunities for fiuvial bull trout. 48 Project Summary Chamberlain Creek is a small Garnet Mountain tributary to the middle Blackfoot River, entering near rm 43.9 with a base-flow of 2-3 cfs. Prior to 1990, sections of lower Chamberlain Creek was dewatered and severely altered (grazing and channelization), leading to sharp declines in WSCT densities (Peters 1990). During the early 1990s, Chamberlain Creek was also one of the first comprehensive restoration projects within the Blackfoot Basin. Restoration emphasized road drainage repairs, riparian livestock management 30 changes, fish habitat restoration, _ 25 irrigation upgrades (consolidation of ditches, water conservation, elimination of fish entrainment fish installation diversion), conservation easements and improved stream fiows through water leasing. Restoration occurred throughout the drainage with emphasis in the lower mile of stream. o o o 20 15 and ladder on a (fl c Q 5 X Pro. ject od ^1^ ■ Pen T . rii m / H h hsA A / n Figure 7. Densities of age 1 and older WSCT in Chamberlain Creek at mile 0.1, 1989-2007. Fish Populations and other monitoring activities Chamberlain Creek is a WSCT-dominated stream over its entire length although lower reaches also support rainbow and brown trout in low abundance. Following restoration and WSCT recovery in lower Chamberlain Creek, radio telemetry identified Chamberlain Creek as a primary spawning stream for fiuvial WSCT from the Blackfoot River (Schmetterling 2001). In 2006-07, we continued to monitor fish population 75 70 65 60 .55 50 45 40 — — j = — " • 25% - — — -Min -Median - -Max • 75% . 1999 2000 2005* 2007 Figure 8. July water temperatures for Chamberlain Creek at stream mile 0.1, 1999, 2000, 2005 and 2007. 49 densities, water temperature and whirling disease in the restoration area near the mouth. Fish population surveys at mile 0.1 identify >10 years of stable densities of age 1 and older WSCT (Figure 7). Periodic water temperature monitoring suggest recent warming (Figure 8, Appendix H) In 2007, we continued to test Chamberlain Creek for whirling disease (using sentinel exposures^) at an established downstream monitoring site (mile 0.1), and at two upstream (miles 0.7 and 3.8) locations. The two lower sentinel cages were placed up-and downstream of artificial ponds to help assess whether these stream-connected ponds might contribute to a high severity of disease as previously identified (Pierce et al 2006). This test identified mean grade infection rates were slightly higher downstream of the ponds than upstream of the ponds (mean grade infection = 1.89 below verses 1.21 above). Interpreting these results however remains difficult as the mean grade infection at mile 0.1 was notably lower in 2007 (1.89) than 3.78 when last tested in 2005. Consistent with earlier exposures, whirling disease was not detected at mile 3.8. Copper Creek Copper Creek, the largest tributary to the lower Landers Fork entering at mile 3.6, is a critical spawning and rearing stream for genetically pure fluvial WSCT and fluvial bull trout in the upper Blackfoot River drainage. Copper Creek supports an entirely native fish community basin-wide, and provides the only major spawning migration of fluvial bull trout in the upper Blackfoot River basin. Copper Creek's consistent cold- water temperatures help moderate temperatures in the lower Landers Fork. During August 2003, the Snow/Talon wildfire on the Helena National Forest ran through the Copper Creek drainage. This high intensity, stand replacement fire burned significant portions of the basin including a fluvial bull trout spawning site approximately three weeks prior to spawning. 40 35 30 25 20 D Cutthroat ' m Bull Trout ' O 15 10 :te rk,f Fish Populations and other monitoring activities In 2006-07, we continued fish population and temperature monitoring in Copper Creek in the area of the Snow/Talon wildfire. This monitoring included bull trout redd counts, surveys of juvenile trout abundance. Following the wildfire, bull trout redd counts have shown a substantial increase in the total number of redds in Copper Creek (Executive Summary). Similar to redd counts, electrofishing surveys at a long-term monitoring site (mile 6.2) indicate a post-fire increase in the numbers of age 0+ native fish (Figure 6, Appendix A). Water temperature monitoring identifies a significant warming in Copper Creek, in the post-fire environment (Executive 1989 1998 1999 2002 2004 2005 2007 Figure 6. CPUE for age 0+ native trout at stream mile 6.2 in Copper Creek, 1989-2007. 50 Summary). We interpret the fisheries increases as a response to increased water temperature and sunlight, and a nitrogen influx from the burn area, all of which would contribute to enhanced secondary productivity as expressed by the native fish community. Cottonwood Creek Restoration objectives: improve degraded habitat; eliminate fish losses to ditches; and restore instream fiows and migration corridors for native fish. irrigation Project Summary Cottonwood Creek, a 3"^ order stream, flows ~16-miles south from the Cottonwood Lakes and enters the middle Blackfoot River at rm 43 with a base- flow of -15 cfs. Genetically pure WSCT and bull trout dominate the headwaters of Cottonwood Creek. Rainbow trout, brook trout and brown trout dominate middle to lower stream reaches. Cottonwood Creek is considered a "core area" and was designated as "proposed critical habitat" under the ESA for the recovery of bull trout. Cottonwood Creek has been the focus of irrigation-related flsheries improvements since 1996. In 2006-07, flsheries improvements were undertaken in both the middle and upper reaches of Cottonwood Creek. In upper Cottonwood Creek, the Blackfoot Cooperators replaced an undersized (and perched) culvert at mile 15.9. This project restored flsh passage to -1.0 mile of stream, while • Water Chemistry ■ Fish Population Survey Figure 9. Cottonwood Creek and adjoining spring creek complexes and flsh population survey locations. 36 34 32 _ 30 o 28 SS 26 S 24 -i 22 + 20 i!? >> 14 S 12 S 10 ° I 4 2 No fish D Cutthroat trout D Bull Trout ^- # rh Figure 10. Densities for age 1 and older native salmonids in Cottonwood Creek at mile 12.0, 1996- 2007. 51 correcting channel incision and erosion problems downstream of the culvert. Grazing and irrigation-related projects are now the focus of developing projects on public lands along middle Cottonwood Creek. These developing projects result from livestock-related degradation of riparian areas on the both the University of Montana (Bandy Experimental Ranch) and FWP (Blackfoot Clearwater Game Range - Dryer Ranch) properties. Additional fisheries impacts involve the unauthorized use of an FWP diversion during the native fish migration period. This use has compromised native fish migrations to upstream spawning areas in recent years. In an attempt to address these concerns, both U of M and FWP have assessed their properties and are identifying steps to make necessary changes. These assessments include fish population surveys and spawning site surveys (this report, Pierce et al 2006), as well as a series of riparian health assessments conducted by the U of M Forestry School in 2007. On the FWP properties, corrective measures include 1) the exclusion of livestock from immediate stream banks and degraded spring creek complexes, 2) upgrades at three stream crossing, 3) off-stream water developments, 3) the removal of a diversion, and 4) dedication of irrigation water rights to instream flows. On the Bandy Ranch, irrigation pump sites are being modified and attempts to correct riparian grazing problems are being pursued. Corrective measures, if successful, are expected to improve riparian health along the middle of Cottonwood Creek, while enhancing native fish conditions in middle and upper reaches of Cottonwood Creek. Fish Populations and other monitoring activities In 2006 and 2007, we continued to monitor fish populations in upper Cottonwood Creek (mile 12.0) where enhanced flow, irrigation ditch screening and diversion upgrades were made. Prior to 1997 this 30 25 20 8 ^ 10 £ 15 o 1 ■ Cutthroat D Brown □ Brook i^^ M section was completely dewatered during late summer and fall by irrigation. We also resurveyed middle Cottonwood Creek (mile 7.5) in 2007 as well as three nearby spring creeks on FWP lands (Figure 9). The middle Cottonwood Creek and nearby spring creek monitoring sites were originally established in 1991 prior to the current level of riparian degradation (Pierce et al. 1997). At the upper Cottonwood Creek monitoring site (mile 12.0), age 1 and older WSCT have remained stable in recent years, following an initial increase in the late 1990s. Bull trout densities have remained static at low densities (Figure 10). By contrast, fisheries at the middle Cottonwood Creek (mile 7.5) monitoring site show community-level decline in the area of livestock over-use (Figure 11). Likewise, sampling on three adjoining spring creeks (entering middle Cottonwood Creek as miles 6.4, 6.7 and 7.5) show declines in species richness and abundance. In 1991, 1991 1997 2007 Figure 11. CPUE for salmonids in Cottonwood Creek at mile 7.5 in 1991, 1997 and 2007. 52 brown trout were identified throughout the spring creeks, and WSCT were present in two of three spring creeks. However, in replicate 2007 surveys, all trout species were identified in reduced densities in areas where riparian conditions have deteriorated. Brown trout were found in low densities in only one spring creek, and WSCT were absent from all spring creek samples. Brook trout have also expressed declines compared to 1991 (Pierce et al 1997, Appendix A). Water temperature monitoring in lower Cottonwood Creek shows continued warming since 2001 (see Figure 7 in Executive Summary). Sentinel exposures near the mouth (mile 1.1) from 2006 and 2007 show a continuous high severity of whirling disease (mean grades of 3.96 and 4.25, respectively). Enders Spring Creek Restoration objectives: Restore the spring creek to natural "C4-type" spawning channel, reduce water temperatures to level suitable for bull trout, reduce instream sediment levels, enhance habitat quality utilizing in-stream structures, vegetation and provide suitable substrate for spawning. Project Summary Enders Spring Creek is a heavily degraded T'-order spring creek that enters the North Fork of the Blackfoot River at mile 6.3. Stream discharge was measured at 6.5 cfs on May 30, 2007 (Appendix D). Like other spring creek tributaries to the North Fork, Enders Spring Creek has a long-history of adverse human-related changes to salmonid habitat. These stem from past agricultural activities and include the loss of sinuosity, channel widening and • Temperature Sensor ■ Fish Population Survey Figure 12. Enders Spring Creek stream channel restoration project area and fish population survey location, 2007. heavy sediment loading in pools and glides. Enders Spring Creek is the last major spring creek to the North Fork that requires active restoration. Restoration will include two-miles of complete channel reconstruction similar to the Jacobsen Spring Creek project (see Jacobsen Spring Creek), and this work is slated for 2008. Like all other spring creeks restoration projects on the North Fork, ensuring compatible grazing strategies will be critical to the future success of this project. 53 Fish Populations and other monitoring activities In 2006 and channel measurements 2007 in advance of channel reconstmction, we conducted a series of baseline studies that included fish population surveys at mile 0.5, and a habitat (channel) survey that also included water temperature, water chemistry, substrate and discharge measurements. Fish population survey recorded primarily brook trout, brown trout, mountain whitefish at low densities and very low Pre-restoration stream Channel Length Sinuosity Channel Gradient Width / Depth Ratio Total # Pools # Pools Sampled Pool Frequency # Pools Measured with LWD Pool Length Wetted Pool Surface Area Maximum Pool Depth Wetted Pool Width @ Max Pool Depth Wetted Width @ Riffle Crest Bankfull Width @ Max Pool Depth Bankfull Width @ Riffle Crest Riffle Crest Depth Residual Pool Depth 10,600 1.1 0.004 23-41 45 45 4.2 pools /1 000ft 34 (76%) 57 + 41 (9-161 ) 1358+1162(74-5252) 2.2 + 0.8(1-4) 23 + 8(10-40) 21+7(7-38) 26 + 10(11-68) 26 + 10(8-50) 0.5 + 0.2(0.1 -0.9) 2 + 0.8(0.4-4) All measuements in standard (ft) units Table 2. Summary of pre-project habitat survey results for Enders Spring Creek. densities of bull trout (Appendix A). The habitat survey measured low sinuosity, high W/D ratios and bankfull widths ranging from 8-68' (Table 2). Pre-project (2007) water temperature recorded maximum summer temperature of 60. IT near the mouth (Appendix H). Wolman pebble count at mile 0.1 identified a gravel- dominated (D75 = 68mm, D50= 31mm and D25=9mm) substrate in a representative riffle, heavy loading of fine sediment was noted in pools and glides. Frazier Creek In 2006, FWP explored fisheries restoration opportunities in Frazier Creek. Frazier Creek, a small Although a 3'^''-order Figure 13. Stage discharge staff and water temperature sensor locations in the Frazier Creek drainage, 2007. 54 0.2 0.4 0.6 0.8 staff elevation (ft) Figure 14. Stage discharge curve developed for Frazier Creek at mile 0.2 in Mav 2007. basin with 3.8-miles of perennial stream, enters the middle Blackfoot River at river-mile 59.4 from the Garnet Mountains. Frazier Creek is one of several small Garnet Mountain tributaries to the middle Blackfoot River that currently support no known fluvial WSCT use due to instream reservoirs as well as other fisheries impairments (Results Part V, Appendix F). The North Fork of Frazier Creek is -2.0 miles in length and enters Frazier Creek at mile 1.0 with an estimated base-flow of 0.3 cfs. Both the mainstem and North Fork provide water for intensive irrigation via two large instream reservoirs, one smaller (empty) reservoir on the mainstem and three irrigation diversions (Figure 13). In fish-bearing water above the middle reservoir, Frazier Creek is fragmented at two locations by the upper reservoir and a partial fish passage barrier (perched culvert) at mile 0.3 on the lower reaches of the North Fork. The stream is segmented into sections of 2.6-miles above the upper reservoir, 0.7- miles above the middle reservoir and 1.7-miles above a culvert (fish passage) barrier on the North Fork. In addition to fragmented habitat, the stream channel and riparian vegetation on the lower 0.3-miles of Frazier Creek suffers from past livestock grazing. The mainstem riparian understory vegetation above the middle and upper reservoirs and along the North Fork drainage show signs of heavy degradation from livestock grazing; however, the over-story is relatively dense providing shade to the stream channel. 1.2 Date Staff reading Discharge 31-May-07 0.54 * 0.08 31-May-07 0.83 * 1.39 31-May-07 0.89 *2.01 31-May-07 1.47 * 7.33 6-Jun-07 0.55 ** 0.11 8-Jun-07 0.8 ** 1.31 12-Jun-07 0.81 ** 1.38 18-Jun-07 0.68 ** 0.63 2-JUI-07 0.55 ** 0.11 18-Jul-07 0.55 ** 0.11 8-Aug-07 0.48 ** 0.0 12-Sep-07 0.5 ** 0.01 15-Sep-07 0.57 ** 0.18 27-Sep-07 0.56 ** 0.14 22-Oct-07 0.65 ** 0.49 * Discharge used to build rating curve Disciiarge calculated from staff reading Table 3. Staff readings and discharge measurement on lower Frazier Creek, May through October 2007. 55 Flow and temperature monitoring We investigated stream flow and temperature relationships in Frazier Creek during the summer of 2007. On May 31, 2007, we placed staff plate near the mouth (mile 0.2) and regulated reservoir outflows in order to develop a stage-discharge rating curve that would served to monitor stream flows during the irrigation season (Figure 14). Eleven staff readings were then taken on lower Frazier Creek and these recorded a discharge range 1.4 cfs in June to no-flow in September (Table 3). During the flow evaluation, the majority of the water from upper Frazier Creek was stored in reservoirs and for irrigation use. The outlet flow to Frazier Creek was primarily from reservoir seepage. Temperature sensors were placed at five locations (Figure 13, Appendix H). Maximum temperatures increased from 68°F to 78°F between the upper site (mile 1.2) located upstream of the upper Frazier Creek reservoir and the next downstream site (mile 0.4) downstream the middle reservoir. Sensors recorded some moderation in temperatures near the mouth (mile 0.1) with maximum temperature at ~73°F. Fish populations Frazier Creek supports a genetically "pure" disjunct population of stream-resident WSCT with no other fish species (Pierce et al 2006). Both Frazier Creek reservoirs provide holding areas for WSCT with reproduction and rearing occurring in the isolated reaches above each reservoir (Pierce et al 2000). Because of the fragmented headwaters and the loss of fisheries in the lower reaches, Frazier Creek is lacking sport fishery value to the Blackfoot River (Pierce et al 2005). Reestablishing complete upstream WSCT connectivity from the Blackfoot River is currently in confiict with irrigation practices and a concern given the potential for invasive (e.g. hybridizing) species, particularly under existing environmental (e.g. degraded habitat) conditions. However this stream-resident WSCT population has conservation value and potential for improvement by eliminating headwater fragmentation, ditch screening and implementing alternative riparian grazing strategies Gold Creek Restoration objectives: Restore pool habitat and morphological complexity; restore thermal refugia for Blackfoot River native fish species. Summary Gold Creek is the largest tributary to the lower Blackfoot River, entering at mile 13.5. The majority (90%) of the Gold Creek watershed is industrial forest. Past harvest of riparian conifers combined with the actual removal of large wood from the channel reduced habitat complexity on the lower three-miles of Gold Creek. The result of this fish habitat simplification was low densities of age 1 and older fish. In 1996, we installed 66 habitat structures made of native material (rock and wood) constructing 61 new pools in the 3-mile section (Schmetterling and Pierce 1999). Prior to restoration work (1996), we established a baseline fish population survey section (mile 1.9) in the treated area for future monitoring. 56 25 20 15 10 D Bull trout ■ Cutthroat B Brown D Rainbow ra * hM J? ] J ,^ ,^ ,^ # * Figure 15. CPUE for age 1+ fish in Gold Creek at mile 1.9, 1996 - 2006. Fish Populations and other monitoring activities Gold Creek is a major spawning tributary to the lower Blackfoot River for bull trout, WSCT, rainbow trout, and brown trout. Resident brook trout also inhabit the drainage. Gold Creek's mainstem and confluence area provides thermal refugia for Blackfoot River bull trout during periods of river warming. In 2006 and 2007, we continued to monitor 1) post- restoration fish population monitoring in the project area (mile 1.9), 2) bull trout redds, and 3) water temperatures and whirling disease near the mouth. Fish population surveys show a 10-year upward trend in CPUE for fish >4.0" within the restoration project area (Figure 15). However, these samples identify consistently low bull trout densities, and in 2006 we failed to detect bull trout in the monitoring sites. Our bull trout redd counts identified only one redd in 2007 and none in 2006 compared to seven in 2005. Water temperature recordings at mile 1.9 show a nine- year warming trend (Executive Summary, Appendix H). Whirling disease tests in 2006 and 2007 near the mouth of Gold Creek have recorded no infection. Hoyt Creek Restoration objectives: Project objectives were to restore Hoyt Creek to its historic floodplain elevation for channel stability, wetland values, and irrigation efficiency improvement; maximize cooling influence on water temperatures, reduce sediment production in the downstream direction, maximize the quality of undercut bank habitat through riparian vegetation root mass and develop and implement a management plan. Project Summary grazing Project Are; ■ Fish Population Survey • Temperature Sensors Ovando, IVIT Figure 16. Hoyt Creek stream restoration project area, temperature sensor and fish population monitoring locations, 2006-2007. Hoyt Creek, a small r* order spring-fed tributary to Dick Creek, originates from alluvial aquifers located just 57 ■ Cutthroat a Brook D Brown □ Rainbow D Mountain wliitefisii □ Long nose sucl75°F, which exceed the suitable range necessary for salmonids (Figure 18). Summary statistics for all water temperature monitoring are located in Appendix H. 80 75 70 65 .60 55 50 45 I •- I 1 ^" 1 ^" 1^ — T — -» — ' 2001 2005 2007 Figure 18. July water temperatures for Hoyt Creek at mile 1.3 for 2001, 2005 and 2007. 58 Jacobsen Spring Creek Restoration objectives: Maximize secondary instream productivity; maximize quality of shoreline rearing areas; restore spawning site potential by reducing levels of fine sediment in riffles to a level suitable for spawnmg; summer temperatures for bull trout provide high reduce water suitable (<60°F); quality • Temperature Sensor ■ Fish Population Survey pools with high level of complex cover; maximize use of existing channel belt- width and existing shoreline areas. Project Summary Jacobsen Spring Creek forms from two spring creeks that merge at mile 0.7 and generate a base-flow of 4-7cfs near the mouth (Figure 19). This small spring creek system enters the North Fork of the Blackfoot River at mile 4.7. According to landowner accounts, Jacobsen Spring Creek historically supported both bull trout and WSCT. Jacobsen Spring Creek was severely degraded due to historic grazing and timber Channel measurements Pre-restoration Post-restoration % change Figure 19. Jacobsen Spring Creek stream channel restoration area and fish population survey location, 2007. stream Channel Length 3150 3800 20.6 Sinuosity 1.2 1.4 16.7 Total # Pools 19 58 205 # Sampled Pools 10 29 190 Pool Frequency 6.0 /1000ft 15.3/ 1000ft 155 # Pools Measured with LWD 9 ( 90%) 28 ( 97%) 211 Pool Length 37 + 21 (14-79) 21 +6(13-34) 43 Wetted Pool Surface Area 858 + 626(224-1859) 208 + 52(112-299) -76 Maximum Pool Depth 1.7 + 0.7(0.9-3.3) 3 + 0.4(2-3.4) 76 Wetted Pool Width @ Max Depth 20 + 10(9-44) 11+2(7-14) -45 Wetted Width @ Riffle Crest 24 + 12(8-47) 9 + 2(7-14) -62 Bankfull Width @ Max Pool Depth 21 +10(9-44) 12 + 2(9-19) -43 Bankfull Width @ Riffle Crest 24 + 12(8-47) 11+2(8-16) -54 Riffle Crest Depth 0.6 + 0.2(0.4-0.9) 1 +0.2(0.5-1.5) 67 Residual Pool Depth 1.1 +0.7(0.3-3) 1.7 + 0.3 (0.8-2.2) 55 All measurementsin standard (ft) units Table 4. Pre-and post restoration channel measurements for the lower 0.7 miles of Jacobsen Spring Creek. 59 harvest practices, the consequences of which include an over-widened stream channel, low sinuosity, elevated water temperatures and excessive sediment loading (Pierce et al 2006). However, early habitat investigations identified the spring creek as possessing the basic habitat components necessary for improved fisheries such as stable groundwater inflows, gravel substrate and a relatively dense riparian spruce forest that has potential to provide shade, complexity, and wood to the stream channel. Between 2005 and 2007, the entire 17,220' of Jacobsen Spring Creek (both channels) was reconstructed. The project emphasized a deep and narrow channel with higher sinuosity, the inclusion of backwater and shoreline rearing areas, gravel in pool tail-outs, and the placement of instream wood and sod mats on the stream banks to facilitate recovery. The project also included shrub plantings and the adoption of livestock management changes consistent with project objectives. Figure 20. Pre (2004) and post-project (2006) max\min daily waters temperatures for Jacobsen Spring Creek near mouth, summer 2004 and 2006. n Fish Populations and other monitoring activities In 2006-07, we returned to at our pre-project baseline monitoring locations and completed 1) a post-project habitat survey downstream of mile 0.7, 2) a fish population survey site (mile 0.6) established in 2005, and 3) water temperature monitoring site near the mouth. A comparison of the pre- and post habitat survey results are presented in Table 4. Among the changes to the physical channel, our survey results show a 48% decrease in the wetted- width of the channel and a 76% increase in pool depth. Water temperature changes include a 10°F reduction in maximum summer temperatures between the pre-project (2004) and post- project (2006) (Figure 20, Appendix H). 18 16 14 12 10 8 6 4 2 i ■ Brown trout D Brook D Rainbow DWSCT ■ MWF 2005 2006 Year 2007 Figure 21. CPUE for salmonids in Jacobsen Spring Creek at stream mile 0.6, 2005-2007. 60 At this early stage of post-project fisheries monitoring, fisheries have expressed no appreciable change (Figure 21); however, rainbow trout spawning (redds) were identified in Jacobsen Spring Creek in the spring of 2007 and rainbow alevins were present in constructed backwater areas during the 2007 surveys. Mountain whitefish, absent from previous surveys, were also identified in the new channel in 2007. With summer temperatures now cooler than lower North Fork, Jacobsen spring creek should attract an increased level of bull trout use in the future. Kleinschmidt Creek Restoration objectives: Reduce whirling disease infection levels; restore stream channel morphology for all life stages of trout; increase recruitment of trout to the Blackfoot River; and restore thermal refugia and rearing areas for North Fork Blackfoot River bull trout. Project Summary Kleinschmidt Creek, a spring creek tributary with a base flow of ~9 cfs, joins with Rock Creek at mile 0. 1 before entering the North Fork of the Blackfoot River at mile 6.2. Kleinschmidt Creek has a long history of stream degradation involving livestock over-use and channel alterations related to instream rock dams, undersized culverts and highway channelization (Pierce 1991). Restoration of Kleinschmidt Creek began in 1991, and expanded substantially in 2001 with 6,250' of stream reconstructed to a longer (8,494'), narrower, deeper and more sinuous channel. The work has reduced water maximum water temperatures from a high of -70° F to <60° F (Pierce et al 2006). In 2006 restoration continued with -600' of channel reconstruction and riparian grazing changes in the upper-most perennial section of stream. Summaries of pre-and post-project fisheries and related , assessments (water temperatures, discharge, substrates, channel morphometries and whirling disease) are described in Pierce et al. 1997; 2002; 2004; and 2006. o o Fish Populations and other monitoring During the 2006 and 2007, we resurveyed at two locations (mile 0.5 and 0.8) of lower Kleinschmidt Creek established in 1998 prior to channel reconstruction. These sites were established to assess 0) Q ,5*\5* ,5* ,5* ,5* ,5* Kfc hs »y *V ^ ^ ^ ^ ^ ^ Mile 0.8 w/ wood Figure 22. Estimated densities of age 1 and older brown trout in two sections of Kleinschmidt Creek, 1998-2007. 61 restoration techniques involving the placement of large instream wood into E4-type channels. We placed no instream wood in the reconstructed channel at mile 0.5, whereas the rest of the channel, including the mile 0.8 survey site, included instream wood placements. Both sites show higher densities of age 1 and older brown trout compared to the pre-project periods; however the section with wood has continuously recorded higher brown trout densities (Figure 22). WSCT and bull trout were not detected in the two monitoring section between 1998 and 2003; however both native species were consistently identified in very low densities in the section with wood (mile 0.8) in recent years. Lincoln Spring Creek Restoration objectives: Improve overall habitat conditions, improve spawning and rearing habitat for salmonids, eliminate fish passage barriers, and improve water quality conditions. Project Summary Lincoln Spring Creek is a large spring creek tributary to Keep Cool Creek, which enters the Blackfoot River at mile 105.2. This T^-order, low-gradient spring creek is -6.3 miles in length (Figure 23) and originates from an alluvial aquifer under the Lincoln Valley and generates variable base-flow that seasonally rises and falls with the aquifer. The stream flows west through private ranchland and the town of Lincoln before entering Keep Cool Creek at mile 0.6. It is primarily a gravel based stream with a surrounding spruce riparian over-story. Fisheries-related impairments include irrigation practices, heavy livestock grazing and residential impacts and undersized culverts. The activities have suppressed riparian vegetation and contribute to an over- widened and shallow stream channel, fine sediment loading and generally A^l^ Elevation 4600- A^a- simplified habitat. The Blackfoot Cooperators by BBCTU currently reconstructing upper -8,000 Lincoln Spring Creek (mile 2.9 to 4.6). Specific restoration fish 4S40- led are the of 4520- 4500- 4480- ' A^^- 4440 it F|tiHrln'95 » «0 <0 N .oooooooo oooooooo Figure 24. Densities of age 1 and older fish in McCabe Creek at mile 2.2, 1999-2007. 63 to increasing stream flows, reducing water temperatures in Dick Creek, eliminating WSCT losses to ditches, and restoring habitat complexity to a damaged stream channel. Fish Populations McCabe Creek is a WSCT dominated stream, with brook trout present in lower stream reaches. Due to cool summer temperatures, McCabe Creek likely supported bull trout historically. In 1999, prior to restoration, we established a fish population survey section in a degraded section of stream (mile 2.2), an area of low habitat complexity and chronic low flows. In 2006, we continued to monitor fisheries at mile 2.2 (Figure 24). Both WSCT and brook trout (age 1 and older) have increased in the project area compared to the pre- project (1999) condition. Less encouraging is an increase in brook trout at the monitoring site. Monture Creek Restoration objectives: Restore habitat for spawning and rearing bull trout and WSCT; improve recruitment of bull trout and WSCT to the Blackfoot River; improve staging areas and thermal refugia for fluvial bull trout. Project Summary Monture Creek, a large tributary to the middle Blackfoot River, is a primary spawning and rearing tributary for fluvial bull trout and fluvial WSCT (Swanberg 1997, Schmetterling 2001). Monture Creek also serves as thermal refugia for fluvial bull trout during periods of Blackfoot River warming. Reproduction of WSCT and bull trout occurs primarily in the mid-to-upper basin. Fluvial rainbow trout inhabit and reproduce the lower portions of the drainage (Results Part V). Brook trout are found in the lower basin downstream of the intermittent reach at mile 14 (Results Part VII). In addition to monitoring in the lower Monture Basin, in 2006-07 we also inventoried tributary fisheries and lakes in the backcountry of the upper Monture Creek basin and several tributaries (Results Part VII). Riparian areas in the mid-to-lower reaches of Monture Creek have a long history of riparian timber harvest and improper grazing practices, with resulting adverse impacts to riparian habitats (Fitzgerald 1997). All lower tributaries of Monture Creek from Dunham Creek downstream likewise were identified as fisheries-impaired (Appendix F). Many identified problems were corrected through a decade of cooperative restoration activities (Pierce et al. 1997; Pierce et al. 2001), which contributed to improving the health of Monture Creek. Despite improvement, excessive livestock access continues in certain riparian areas of lower Monture Creek. Fish Populations and other monitoring Monitoring for 2006 and 2007 period included: 1) bull trout redd counts; 2) assessments of juvenile bull trout abundance at one long-term monitoring station; 3) water temperature monitoring; 4) continued whirling disease studies; and 5) radio telemetry study involving fluvial rainbow trout from the Blackfoot River. Bull trout redd counts have been upward trending since restrictive angling regulations in 1990, but also show a sharp recent decline {see Figure 5 Executive 64 Summary). This downturn is consistent with other drought-related bull trout declines in the lower Blackfoot Basin. Likewise, assessments of juvenile bull trout abundance at a long-term monitoring station revealed increases through the 1990s, but also a recent decline proportional to declining redds (see Figure 6 Executive Summary). Results from a rainbow trout telemetry study clearly identify lower Monture Creek as the primary spawning tributary for the middle Blackfoot River upstream of Clearwater River. Spawning occurred primarily in lower Monture Creek, but extended upstream as far as lower Dunham Creek (Results Part V). Lower Monture Creek first tested positive for whirling disease in 2000. Since then, whirling disease in the primary rainbow spawning areas have increased to a severe level (mean grade infection > 4.0 on the MacConnell Baldwin scale. Results Part V). Conversely, whirling disease tests near native fish spawning areas located on the National Forest (mile 12.6) have failed to detect the disease in both 2006 and 2007. Water temperature monitoring in 2006-2007 was completed at two sites (miles 2.0 and 12.9). This data shows significant warming in the downstream direction (Appendix H) and a long-term (> 10-year) trend of increasing temperatures in lower Monture Creek (see Figure 7 Executive Summary). Murphy Spring Creek Restoration objectives: Restore habitat conditions suitable to WSCT and juvenile bull trout; prevent irrigation ditch losses; maintain minimum instream flows and provide rearing and recruitment for fluvial bull trout and cutthroat trout to the North Fork. Project Summary Murphy Spring Creek, a small WSCT dominated tributary, originates on the north side of Ovando Mountain and flows six miles south and enters the North Fork at mile 9.9. Murphy Spring Creek has a history of irrigation impacts and fish passage problems (Pierce et al. 2006). 22 20 o o O Irrigation problems involve chronic dewatering and entrainment of WSCT to the Murphy ditch at mile 1.8. Fish passage problems involved an undersized culvert at mile 0.5 and the defunct condition of the Murphy diversion. The culvert reduced the upstream movement of juvenile bull trout from the North Fork, while the diversion reduced downstream movement of WSCT from the headwaters to the North Fork through dewatering and entrainment. ■ WSCT n Bull Trout □ Brook Trout 2001 2005 2006 2007 Figure 25. CPUE for salmonids in Murphy Spring Creek at mile 0.6, 2001-2007. 65 The Murphy Spring Creek restoration project began in 1998 with the installation of a new diversion fitted with a Denil fish ladder. In 2000, we replaced the culvert with a larger baffled culvert designed to allow the upstream movement of YOY bull trout. In 2004-05, the Blackfoot Cooperators expanded restoration actions by developing an instream flow agreement that granted habitat maintenance flows as well as a 2.2 cfs minimal instream flow in Murphy Spring Creek. In 2006, a Coanda fish screen was placed at a diversion as a measure to eliminated losses of WSCT. Fish population and other monitoring activities Fish population surveys conducted in 2006 and 2007 show a modest increase in numbers for all salmonids. Prior to 2001, bull trout were absent from this location. Brook trout densities have also increased at this site (Figure 25). Nevada Spring Creek Restoration objectives: Restore habitat suitable for cold-water trout; improve downstream water quality, and reduce thermal stress in Nevada Creek and the Blackfoot River. Project Summary Nevada Spring Creek, a tributary of lower Nevada Creek, originates from an artesian spring and flows through agricultural lands to its junction with Nevada Creek at mile 6.2. The spring source produces between six and nine cfs. Nevada Spring Creek is j oined near the source by Wasson Creek, a small, basin-fed tributary that brings and additional base flow of approximately two cfs during the non- irrigation season. Water temperatures at the artesian source are a constant year- around 44°F. Restoration of Nevada Spring Creek has been ongoing for several years. A habitat restoration project for the entire 4.2 miles of Nevada Blackfoot River HWY141 HWY200 Nevada Spring Creek Channel Restorat Wasson Creek :t Area Wasson Creek Nevada Creek stream mile 6.2 I Fish Population Survey Temperature Sensor Figure 26. Nevada Spring Creek stream channel restoration project area and fish population survey and temperature sensor locations, 2007. Spring Creek was completed between 2001 and 2004. The project entailed the complete reconstruction of Nevada Spring Creek, riparian grazing changes, instream flow enhancement, wetland restoration and shrub plantings. Prior to restoration, summer water temperatures in the lower portion of Nevada Spring Creek exceeded >75°F due in part an over-widened channel (Pierce et al. 2002). This warming and agricultural runoff 66 from adjacent lands contributed to water quality degradation, and created unsuitable habitat conditions for coldwater salmonids in the lower portion of Nevada Spring Creek (Pierce et al. 2002). A complete before and after summary of channel measurements is located in a previous monitoring report (Pierce et al. 2006). Fish populations and other monitoring activities Prior to channel restoration, Nevada Spring Creek supported low densities of brown trout in upper non-game (redside northern and o o o 16 14 12 10 ^< 6 (fl o 4 Q □ Brown Trout r+i D V\festslope Cutthroat Trout -A -| r -| Y _rf rt-i r*i nrii f in ir^ ir^ 1 — ,- i> Mile 1.1 / / / i' •^ / / / Mile 3.5 Figure 27 Densities for age 1 and older salmonids at two locations on Nevada Spring Creek, 2000-07. reaches and species shiners, pikeminnow. largescale sucker) in lower reaches (Pierce et al 2002). WSCT were historically abundant in Nevada Spring Creek based on accounts by a long-term landowner (Frank Potts, personal communication). In 2006 and 2007, we continued post-project fish population monitoring at two sites (mile 3.5 (upper site near the source) and 1 . 1 (lower site)), and water temperatures and whirling disease monitoring near the mouth. Near the spring source, densities of age 1 and older brown trout have recently declined; however, WSCT densities show a large recent increase (Figure 27). The brown trout decline appears to relate to a reduction in juvenile recruitment. By contrast, the sharp increase in WSCT densities coincides with upstream restoration and the screening of fish from two upstream irrigation ditches in Wasson Creek (see Wasson Creek section). 80 75 — 70 9> 65 reconstruction ^ 60 55 50 45 1994* 2000 2001 2003 2004 2005 2006 2007* Figure 28. July water temperatures for Nevada Spring Creek near the mouth, 1994, and 2000-2007. 67 Water temperature monitoring near the mouth shows recent increases in water temperature from 2004 through 2007 with temperatures now approaching pre-project levels (Figure 28). These increases began one year after channel reconstruction and result from loss of cooler spring water to off-channel wetland cells. In 2007, the warming exceeded >75°F. Options to correct to this problem are now being examined. Whirling disease monitoring in 2006 found a mean 1.97 grade infection compared to 2.2 of 2005 (Results Part IV). Pearson Creek Restoration objectives: Improve status of WSCT population and increase recruitment of fluvial WSCT to the Blackfoot River. Project Summary Pearson Creek is a small 2 flow of one cfs. Pearson Creek has a history of nd order tributary to Chamberlain Creek with a base- channel adverse riparian (grazing harvest) lower channel. 40 35 30 25 20 3* 15 Q 10 Project -| l+l \ r " r*! \ h ^ — rh [Tl [^ .^ Mile 0.5 ,5> / r^ ^ ^ ,> 4? ,5» ,5> ,5> ,5> ,S> Mile 1.1 / •V / alterations and irrigation and land management and timber practices in its two-miles of Beginning in 1994, Pearson Creek has been the focus of a holistic restoration project involving channel reconstruction and instream habitat work, instream flow enhancement (water leasing), conservation easements and riparian grazing changes. Additional riparian grazing improvements completed in 2006 included riparian corridor fencing for the lower two miles of stream, off-stream water developments and armoring a road crossing. Fish Populations Pearson Creek is a fluvial WSCT spawning stream connected to the Chamberlain Creek WSCT population. In 2006 and 2007, we continued fish population monitoring at two sites in lower Pearson Creek. The upstream site (mile 1.1) was established in 1999 prior to instream restoration activities. Following an initial increase between 1999 and 2000, age 1 and older WSCT have remained static at higher densities. In 2005, we established the downstream site (mile 0.5) following the degradation of stream banks by cattle. Fish population sampling results for both sites are summarized in Figure 29. Figure 29. Densities of age 1 and older WSCT in Pearson Creek at miles 0.5 and 1.1, 1999-2007. 68 Poorman Creek Restoration objectives: Improve riparian habitat conditions and enhance instream flows; restore migration corridors; improve recruitment of native fish to the Blackfoot River. Project Summary Poorman Creek, one of the larger tributaries from the Garnet Mountains and it enters the Blackfoot River at river mile 108. Poorman Creek has been identified with hardrock and placer mining, irrigation dewatering, fish losses to ditches, channel instability, excessive riparian grazing pressure, subdivision impacts and multiple undersized culverts. Beginning in 2002 and continuing through the present, a comprehensive 90 80 7(^ Catch / 100' SO 50 40 30 20 10 0' Brpwn Tmuf D wecT n er«* TrPHf I . M + l"-l" !■ y i " i — I *"* ! , i " iL i " i 2001 I 2003 I 200S | 2007 2002 2004 2000 Mil« 13 -f 2001 I 2003 I 200S | 200T 2002 2004 200& Mil« 1.C restoration projects was implemented on lower Poorman Creek. Restoration projects involved instream fiow enhancement and ditch screening through the fiood-to-sprinkler irrigation conversion, culvert to bridge replacements and riparian grazing changes (corridor fencing, off-stream water) and shrub plantings. Lower Poorman Creek is now entering the passive recovery phase. The recovery of riparian plant communities and improved channel stability now hinges on the continuation of compatible grazing practices, a process expected to take several years. Several upstream culverts were also recently replaced with structures that allow fish passage on the Stemple Pass County Road through the combined assistance of the Blackfoot Cooperators. Fish Populations and other monitoring activities Poorman Creek supports genetically pure WSCT, brown trout and brook trout, and is one of only two known Garnet Mountain stream that still supports bull trout reproduction. Native fish densities increase in the upstream direction while non-native fish occupy lower Poorman Creek. In 2006-07, we repeated fish population surveys at two sites (mile 1.3 and 1.5) in lower Poorman Creek (Figure 30). In 2001, these sites were established up-and downstream of active irrigation diversion and prior to flow enhancement and passive restoration actions. Recent survey results suggest an initial favorable population response Figure 30. CPUE for fish in Poorman Creek at two locations, 2001-2007. 69 for brown trout and WSCT (mostly age fish) up-and downstream of the irrigation conversion project area. Water temperature and whirling disease monitoring was conducted in 2007 at mile 2.2. Water temperature statistical results are found in Appendix H. Whirling disease identified a sharp increase in the severity of whirling disease with a mean grade infection of 4.69 compared to 0.78 in 2004. Rock Creek Restoration objectives: Restore migration corridors for native fish; restore natural stream morphology to improve spawning and rearing conditions for all fish using the system. Project Summary Rock Creek, a basin-fed stream over most of its length, receives significant groundwater inflows downstream of mile 1.6. Rock Creek is the largest tributary to the lower North Fork of the Blackfoot River, but has been degraded over most of its 8.2-mile length due to a wide range of past channel alterations and riparian management activities (Pierce 1990; Pierce et al. 1997, 2006). Rock Creek has also been the focus of continued restoration since 1990. Restoration actions involved working with 13 separate landowners on grazing improvements, instream flow enhancement, and channel reconstruction and revegetation. "Active" restoration is now completed over the entire length of Rock Creek and its primary tributaries, the South Fork of Rock Creek, Salmon Creek and Dry Creek. From this time forward, project success hinges on the ability of all cooperators to managing instream flows and livestock in riparian area, while allowing the passive re- colonization of woody riparian plants. Recovery of riparian areas, including plant communities, will take many years. Fish Populations and other monitoring activities Rock Creek supports a mixed salmonid community. Rock creek provides spawning of brown trout and rainbow trout in lower reaches, a resident brook trout population, limited bull trout rearing and a migration corridor for fluvial WSCT to headwater areas. In 2006 and 2007, we continued to monitor fish populations in lower Rock Creek (mile 1.6) where the stream was reconstructed in 1999. We also resurveyed fish populations at three upstream sites (miles 3.9, 6.4 and 7.5) established in 1994 or 1996. Figure 31. Densities of age 1 and older brown trout in Rock Creek at mile 1.6, 2001-2007. 70 85 80 75 r7o 65 We also monitored water temperatures at a site established upstream of the gaining reach (mile 1.7) to identify whether restoration actions have influenced water temperature conditions across Kleinschmidt Flat. Following a period of increase, fish population surveys in lower Rock Creek (mile 1.6) show a stable brown trout-dominated community with no significant changes in densities in the last few years (Figure 3 1). At this site, a bull trout was recorded in 2006 for the first time since 2001. Surveys at the three upstream monitoring sites recorded low densities of WSCT and brown trout at all three survey locations. Brook trout were found at all sampling locations. In 2007, we identified moderate densities of age-0 Oncorhynchiis (presumed WSCT) in moderated densities at the mile 6.4 sampling location. Summaries of catch and size statistics and density estimates are located in Appendix A andB. July 2007 water temperatures at mile 1.7 show continued high water temperature problems (>80°F) on Kleinschmidt Flat (Figure 32). These temperatures clearly demonstrate the need for the continued (passive) recovery of woody plants (and increased shade) along the channel. Shanley Creek Restoration objectives: Restore habitat for all fish species; restore migration corridors for native fish; reduce loss offish to irrigation ditches; maintain instream flows. Project Summary Shanley Creek is the primary tributary to Cottonwood Creek. At a total length of 10.3 miles, Shanley Creek is a small 2"'' order stream with an estimated base-flow of 3-4 cfs. Since 1994, Shanley Creek has been the focus of several riparian improvement projects with emphasis on correcting riparian grazing problems and screening an irrigation canal to reduce fish losses. In 2006, we observed excessive livestock grazing on University of Montana (U of M) properties and established grazing plans were no longer followed on neighboring private lands. An assessment of U of M riparian areas by the Forestry School identified riparian health as "at risk" due to cattle-related impacts. 60 55 50 45 J -r 1 1 — — 1 1 (■ ^ ^ 2007 Figure 32. July water temperatures for Rock Creek at mile 1.7, 1999 and 2007. 71 22 20 18 16 14 12 10 8 8 4 2 171 Westslope cutthroat trout I I Brown trout I I Brook trout I I Xl r+1 Fish Populations and other monitoring activities In 2006, we longitudinally surveyed fish population at four sites (Figure 33). These surveys included a new upstream site (mile 2.0) upstream of the Bandy diversion in a segment of stream with less livestock disturbance and without the influence of irrigation. These surveys show twice the fish in the upstream sites compare to the lower sites and a sharp decrease in WSCT near the mouth. We also tested lower Shanley Creek for whirling disease for first time in 2007. This test identified a severe mean grade infection of 4.9. Snowbank Creek Restoration objectives: Restore migration corridor for native fish; enhance instream flows; eliminate loss of bull trout and WSCT to irrigation ditch; improve recruitment of native fish to Blackfoot River. Mile 0.2 Mile 1.4 Mile 1.8 Mile 2.0 Figure 33. Densities for age 1 and older salmonids sampled at four locations on Shanley Creek, 2006. D Cutthroat trout ■ Bull Trout Introduction Snowbank Creek is a 1^* order tributary flowing 4.4 miles through the Helena National Forest and enters Copper Creek at mile 5.9. Snowbank Creek was identified as fisheries impaired in 2003 during an assessment of a defunct diversion at mile 0.4. The Snowbank diversion was constructed in 1962 to divert water to create a put-and-take fishery at Snowbank Lake (FWP historical files). We identified fisheries impairments in lower Snowbank Creek to include: 1) native fish entrainment from a diversion to Snowbank Lake; 2) fish passage problems at the diversion and a culvert near the mouth; 3) dewatering 35 30 25 b ° 20 I 15 O 10 Mile 0.1 below culvert l^ 0- im &^^ J ///// Mile 0.4 below diversion >Y/// Mile 0.41 above diversion Figure 34. CPUE for native fish at three locations on Snowbank Creek, 2003 - 2007. 72 below the diversion; and 4) the lack of a legitimate water right that would allow the legal use of Snowbank Creek water for Snowbank Lake (Pierce et al 2004, 2006). Because of the water right problem, the diversion to Snowbank Lake was closed in 2005. In 2007, the USPS obtained a water right that allows the filling of Snowbank Lake. This right provides for restored fish passage, fish screening at the diversion and a minimal instream flow of 4 cfs in lower Snowbank Creek during base-flow periods. Fish Populations and other monitoring activities In August 2006 and 2007, we continued fish population surveys near the mouth (mile 0.1) and up-and downstream of the diversion (located at mile 0.4) at monitoring sites established in 2003 (Figure 34, Appendix A). Our surveys identify both WSCT and bull trout densities have increased sharply with enhanced stream flows. Juvenile bull trout absent from original surveys have now recolonized lower Snowbank Creek. Sampling above the diversion also recorded higher YOY densities for both native species, in addition to finding an adult bull trout that negotiated its way passed the diversion (Appendix B). The presence of adult bull trout during the pre-spawning period and YOY above the diversion suggests bull trout reproduction may be occurring. Tamarack Creek Description Tamarack Creek is a heavily altered small T^-order stream (1.5 miles in length) located on the southeast slopes of Lockwood Point, which historically entered the Blackfoot River at river mile 7.7. At some point. Tamarack Creek was historically rechanneled and diverted to an artificial lake (Lockwood Lake). The original channel is now seasonally dewatered and flows under Highway 200 through a perched culvert that no longer provides upstream fish passage (Figure 35). Below the diversion, the riparian vegetation consists of canary grass, noxious weeds and snowberry. Fish ____ B(>(^jm\ habitat consists of overhanging canary grass and manmade pools. Above the valley floor, the riparian vegetation consists of a Douglas fir and ponderosa pine canopy with an under- story of rocky mountain maple, willow, snowberry, grasses and forbs. Fish habitat in the upper section is composed of plunge pools formed by large boulders and LWD, some undercut banks and dense under-story vegetation. Extensive logging and road building has occurred on the higher FW» 4nH wm iWii v)m ^IDB 4onp iiim Vlt*-: aw !flW -4 PHUiiHl4iih-Ln4l^ ^4 ^.^^ ■|« •/■■■■■ ■. . -^^ ,^-^ - 4-<>— Hl^T imtii^tCimnmTmt IVvHril^tn 1 1 ' 1 Qjt 1.jl <,!(i l.FJ Figure 35. Longitudinal profile for Tamarack Creek. 73 slopes, and landowner accounts report high turbidity during periods of heavy precipitation. Fish populations and other monitoring In 2007, we performed a fish population survey in Tamarack Creek (mile 0.1) for the first time. This survey recorded low densities (CPUE = 6.2) of resident WSCT. Although no other fish species were observed in the stream sample pumpkinseed sunfish were observed in Lockwood Lake. Stream discharge was measured at 0.38 cfs at stream mile 0.1. A WSCT genetic sample (n=27) was taken at the monitoring site, the results of which are pending. Warren Creek Restoration Objectives: Restore riparian vegetation and stream habitat for all life stages of trout; improve spawning and rearing conditions; increase recruitment of trout to the middle Blackfoot River; moderate whirling disease. Project Summary Warren Creek, a small tributary to the middle Blackfoot River, originates on Ovando Mountain and flows 12 miles southwest through knob-and-kettle topography to its confluence with the Blackfoot River at rm 50 (Figure 36). Warren Creek water is used for irrigated hay production and livestock watering. Irrigation causes the middle section of Warren Creek to dewater, although the lower section gains inflow from springs and maintains perennial base-flows of 3-5 cfs. Some riparian areas in mid-to- lower Warren Creek were cleared, heavily grazed, dredged and historically straightened in some cases with dynamite (Don McNally, personal communication). These actions all contribute to extensive degradation of salmonid habitat over the length of Warren Creek. Since 1995, Warren Creek has been the focus of extensive restoration actions. The actions involve removal of several streamside corrals, implementation of grazing plans, shrub planting, several miles of channel reconstruction, instream flow enhancement near the mouth, wetland restoration and the enrollment of private landowners in conservation easement programs. In 2006 a re-entry into a channel reconstruction project (between mile 5.1 and 6.8) was needed to correct channel incision problem in a newly constructed segment of Warren Creek. Figure 36. Warren Creek project areas and monitoring sites, 2007. 74 60 D Rainbow SWSCT ■ Brook D Brown Fish Populations and other monitoring activities In 2006 and 2007, FWP continued to monitor fish populations at five locations (miles 8.2, 6.7, 3.6, 2.1 and 1.1), all in areas of previous restoration actions (Figure 37). The mile 8.2 monitoring site was established in 1995 to monitor fish population response to a riparian grazing project. Since grazing exclusions were put into efi^ect, the stream has OT-CMCO^lOCDh- oooooooo oooooooo Mile 1.1 OT-CMCO^lOCDh- oooooooo oooooooo Mile 2.1 OT-CNfCQ'^lOCDh- CT^lOtOh- OOOOOOOO ooooo OOOOOOOO ooooo J C>JC>J75°F) in exceeding levels considered stressful to salmonids ¥ 2000 2001 2002 2003 2004 2007 Figure 38. July water temperatures for lower Warren Creek (mile 1.1), 2000-2007. 75 (Figure 38). These results identify a need to establish woody species and/or identify irrigation-related options to cool the stream. Wasson Creek The goal of the Wasson Creek project is to "ensure that Wasson Creek will be a significant source of WSCT recruitment to Nevada Spring Creek, Nevada Creek and the Blackfoot River, and provide sufficient forage production for economic sustainability to ranchlands, while demonstrating a successful collaborative effort'' Restoration Objectives: Restore channel maintenance and minimal instream flows; restore migration corridors in lower Wasson Creek in order to provide recruitment of WSCT to Nevada Spring Creek; restore channel conditions to support spawning and rearing conditions in lower Wasson Creek; prevent fish losses to irrigation ditches; prevent the introduction of unwanted fish into the drainage. Project Summary Wasson Creek is a small 2"''-order basin-fed tributary to Nevada Spring Creek. Wasson Creek begins on the Helena National Forest and enters private ranchland near mile 4.0. Wasson Creek joins Nevada Spring Creek -100' below the (artesian) spring source, bringing a base-flow of ~2 cfs during the non-irrigation season. Wasson Creek has a long history of fisheries-related impairments that include fish passage barriers throughout the system, irrigation dewatering and entrainment Blackfoot River Nevada Spring Creek Channel Restoration Wasson Creek : Area Wasson Creek Nevada Creek stream mile 6.2 ■ Fish Population Survey • Temperature Sensor Figure 39. Wasson Creek stream restoration project area and fish population survey and temperature sensor locations, 2007. o o 30 25 20 15 B O 10 ^ ■ Brown Trout HWSCT B n CO ■^ lO «D o o o o o o o o CN CN CN CN Mile 0.1 CO ^ o o o o o o o o CN CN CN CN CN Mile 2.8 CO O O ■^ lO CD t~- O O O O O O O O CN CN CN CN CN Mile 3.0 Figure 40. CPUE for WSCT and brown trout at three monitoring sites in Wasson Creek, 2003-2007. 76 offish to ditches, excessive livestock damage to streambanks, channel straightening and water quality impairments from agricultural runoff Fisheries elements of the project include: 1) grazing management over the length of the project; 2) irrigation changes to accommodate instream flows (low flows and channel maintenance) and fish passage; and 3) channel reconstruction and floodplain containment in the lower (Pierce et al. 2006). The final element to the project was the installation of two fish screens at two irrigation diversions in the spring of 2007. With the exception of grazing changes on upstream properties, Wasson Creek is now entering the recovery phase of the project. Fish Populations and other monitoring activities To assess the restoration project, we repeated fish population sampling at three locations (mile 0.1, 2.8 and 3.0) and continued water temperature monitoring near the mouth (mile 0. 1) through 2007. Our surveys identify the Wasson Creek WSCT population as now expanding in size (densities and distribution) through the lower project area (Figure 40). The 2007 sampling specifically identified a noticeable increase in WSCT downstream of mile 2.8 following the recent installation offish screens. Water temperature monitoring at the mouth show the stream has been cooling since 2004 (Figure 41, Appendix H). This cooling appears to result from cumulative restoration measures including the early recovery of streamside plants, increased flows, and the passive narrowing of the channel in response to stream-side grazing changes. Whirling disease monitoring testing in 2007 near mile 1.4 found no infection. 2003 2004 2005 2006 2007 Figure 41. July temperatures for Wasson Creek near the mouth with Nevada Spring Creek, 2003-2007. 77 Willow Creek, Bear Gulch and Sauerkraut Creeks: A contiguous native fish Habitat Conservation Plan (HC) in the upper Blackfoot Basin. Introduction Within the upper Blackfoot Basin near Lincoln, Willow Creek, Bear Gulch (a tributary to Willow Creek) Sauerkraut Creek, and two miles of the Blackfoot River all fall within a contiguous area of private land located in foothills the Garnet Mountains south of Lincoln. This area adjoins the Helena National Forest and small parcels of State land. In April 2008, -8,000 acres the private land in this contiguous area was recently placed under a Native Fish Habitat Conservation Plan (HCP) easement (Figure 42). The easement, the first of its kind within the Blackfoot Basin, is an outcome of the Blackfoot Challenge Community Project effort to secure the future of some 89,000 acres formerly owned by Plum Creek Real Estate Investment Trust (REIT). The specific purpose of this easement is to enhance the recovery of WSCT and bull trout in concert with the existing Habitat Conservation Plan appurtenant to the lands formerly owned by Plum Creek REIT. In so doing, the proposed conservation easement would conserve fish and wildlife habitat by preventing subdivision, development, and other forms of habitat loss, and perpetuate the ranching and logging lifestyle of the private landowners on the land under easement. The streams encumbered and other nearby streams support genetically pure WSCT, low densities of bull trout in certain areas, as well as non-native fish (e.g. brook trout and brown trout) in lower reaches of most streams. On the western portion the project area is Sauerkraut Creek, a small 2"''-order tributary that enters the Blackfoot River at river-mile 102. 1 . Stream gradients range from 530'/mile between stream mile 7.0 and 6.0 to 807mile in the lower mile of stream. The perennial mainstem Sauerkraut Creek is seven miles in length. WSCT occupy the entire stream as well as (at least) one unnamed tributary (in section 32) to lower Sauerkraut Creek. The headwaters of Sauerkraut Creek are located on the Helena National Forest and the lower three miles of land is located on private lands, most of which (-2.25 miles) falls within the HCP area (Figure 42). On the eastern portion of the project area is Willow Creek, a 2"'' order tributary to the upper Blackfoot River, entering at river-mile 102.5. Stream gradients below the confluence of the West Fork range from 2007mile between stream mile 6.0 and 5.0 and decrease to 5-207mile between stream mile 1.0 and the mouth. Land ownerships on the lower 6.0 miles of stream are mixed private lands, which include HCP parcels as well as non-encumbered private land. The primary tributary to lower Willow Creek is Bear Gulch, a small perennial stream that is currently biologically disjunct near its confluence with Willow Creek. The majority of the Bear Creek basin falls within the HCP boundary. Baseline Assessments In 2007, FWP conducted fish population and related surveys (water temperature and discharge) in Willow Creekm, Sauerkraut Creek and Bear Gulch. In addition to FWP surveys, channel morphology (e.g. sediment, cross sections, reference and departure conditions), riparian health and grazing practices were also assessed (Watershed Consulting 2007). These surveys were intended to serve as an easement baseline, help 78 IT'-!! ^T-i 1hm:>WrFi ^Wm >^ 1^1'-: ; __2 K-^4 III hjf Bear Gulch. SausrKrsut Cr^eh — — I P ■ I I u . Q9Q samiJllng S4t«*. 9 Fl< PopiJHtian T WdlEi- Icmp. Figure 42. Location map showing checker-board landownership and the 2007 fish population, water temperature and stream discharge survey sites. identify restoration actions and provide a basis for long-term monitoring. Fish population surveys were conducted longitudinally at three to five locations per stream (Figure 42). Upstream sites were selected in reference reaches and lower reach survey sites were taken in areas identified as fisheries impaired. Specific fisheries survey objectives were to 1) determine the distribution and densities of species with emphasis on trout; 2) help assess the downstream influences of anthropogenic actions; 3) identify relationships of tributary fish communities to those of the Blackfoot River; and 3) help identify restoration objectives and management needs. We also completed water temperature recordings at four locations, including lower Sauerkraut Creek and three sites on Willow Creek and instantaneous discharge longitudinally at ten locations on the three study streams between 7/23-26/2007 (Figure 42). We timed the flow measurements to coincide with the lower Blackfoot River flows receding to "minimal flow" thresholds of -700 cfs at the Bonner USGS gauging station. We assumed that tributaries reflected general basin (lower Blackfoot River) conditions, and that our study streams, where not dewatered by irrigation and were likewise near minimal instream flow values. We compared these measured flow values with minimum instream flow values generated from the FWP Blackfoot Basin "Montana Method" minimum instream flow model to approximate minimum instream (and departure) flow values for both Willow Creek and Sauerkraut Creek. FWP Survey Results Figure 44 show the relative abundance (CPUE) for all salmonids sampled in the 79 Table 5. Summary of 12 stream discharge measurements, July 2007. Date Stream miie Discharge (tfe) Samplinq location Location Legal Description Lat Long Bear Gulch jmami 1.2 0.03 T13N,R9W,S3B N46.91521 W1 12.72527 Bear Gulch imami 0.7 0.01 T14N,R9W,S34D N46.92245 W1 12.71989 Sauerkraut Cr ir&nmi 3 0.93 T13N,R9W,S8A N46.90111 W1 12.75749 Sauerkraut Cr ir&fMM 2.7 0.51 TT13N,R9W,S5D N46.90537 W1 12.75443 Sauerkraut Cr ir&fMM 0.15 1.7B T14N,R9W,S29C N46.93233 W1 12.76775 Willow Creek 7/23/2007 5.7 1.2B T13N,R9W,S3C N46.90171 W1 12.72139 Willow Creek 7/24/2007 5.2 1.41 T13N,R9W,S3A N46.90910 W1 12.71300 Willow Creek (west channel) 7/25/2007 4.7 0.72 T14N,R9W,S34D N46.91595 W1 12.71220 Willow Creek (east channel) 7/25/2007 4.7 1.78 T14N,R9W,S34D N46.91635 W1 12.71 190 Irrifiation Ditch 7/24/2007 4.3 0.51 T14N,R9W,S34A N46.92299 W1 12.71372 Irrifiation Ditch 7/24/2007 4.1 0.32 T14N,R9W,S34A N46.92464 W1 12.71288 Willow Creek 7/24/2007 4.1 0.57 T14N,R9W,S34A N46.92444 W1 12.71313 Willow Creek 7/24/2007 1.7 0.06 T14N,R9W,S28A N46.93544 W1 12.73946 project area. Related catch and size statistics and population density estimates are located in Appendices A and B. The Sauerkraut Creek surveys identify WSCT at all sampling sites and very low densities of bull trout at both upper and lower sampling sites. Densities of WSCT show a significant decline in the downstream direction between mile 3.2 and 2.7. Salmonid abundances then increased significantly in the downstream direction between miles 2.7 and 0.2. Brook trout and brown trout increase in the downstream direction. The Willow Creek surveys identified WSCT in the headwater areas of the Willow Creek basin (upstream of 4.7); however, our surveys identified the loss of the all salmonids between stream mile 4.7 and 3.6. This absence of salmonids extends to the Willow Creek confluence with the Blackfoot River. Plots of all raw water temperature data are located in Appendix H. Plots of maximum daily water temperatures for Willow Creek are located in Figure 43. Measurements of stream discharge (including primary ditches on Willow Creek) are summarized in Table 5. Sauerkraut showed surprisingly high temperatures of >73° F. Water temperatures collected on Willow Creek show large increases (10°F) between monitoring stations at mile 3.7 and 1.6 and maximum temperatures >77°F. 90 lower (mile 1 .6) middle (mile 3.7) upper (mile 5.4) 40 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ R S (^ s R s R R R R s R R 1 1 g ft 1 IV IV i 1 g Figure 43. Summary of maximum daily water temperatures (°F) for three monitoring sites on Willow Creek, Summer 2007. 80 Discussion The restoration needs of native fish vary within and among the three study streams, although populations in all three streams reflect common sets of human- related limiting factors that influence both the distribution and abundance of fish. As expected, densities of WSCT were higher in the upper reaches of all three streams where habitat conditions are at (or near) natural reference conditions. Moving in the downstream direction, adverse (human-related) changes to the WSCT habitats result in significant population decreases in reaches of all three-study streams. Bull trout were found only in Sauerkraut Creek in very low densities. These bull trout are believed to be migratory fish from the Blackfoot River using Sauerkraut Creek for rearing purposes. Limited bull trout rearing within small Garnet Mountain is consistently identified in small non- spawning streams where suitable native fish (cold and clean) habitats and biological connection to the Blackfoot River persist. Recently, radio- telemetry studies identified spawning migrations of Blackfoot River WSCT in Sauerkraut Creek. These findings of migratory use indicate Sauerkraut Creek still provides for the life-history Figure 44. CPUE for salmonids in Willow Creek (top). Bear Gulch (middle) and Sauerkraut Creek (bottom). re o CPUE for salmonids at five sampling locations on Willow Creek 16 14 12 10 8 6 4 2 o o 14 12 10 8 13 ^ u 4 25 20 §15 510 DWSCT n Brook trout no salmonids Mile 1.7 Mile 3.6 Mile 4.7 Mile 5.2 Mile 5.7 CPUE for WSCT in Bear Gulch at four sampling locations 1 1 1 Mile 0.1 Mile 0.6 Mile 0.7 Mile 1.2 CPUE forsalmonids at three sampling locaitons on Sauerkraut Creek ■ Bulltrout DWSCT D Brook Trout D Brown trout Mile 0.2 Mile 2.7 Mile 3.2 needs of migratory Blackfoot River native fish. WSCT are identified as distributed throughout the length of Sauerkraut Creek although densities decreased significantly in a highly disturbed (mining area) segment of stream immediately downstream of the HNF boundary. Our surveys identify the greatest restoration needs for Sauerkraut Creek within a localized area. This stream segment has a history of placer mining, diversions and excessive grazing. These activities have altered riparian vegetation, damaged channels reduced flow and limit native fisheries. Channel reconstruction, grazing management and other restoration actions should be considered in this area. In the downstream direction, excessive riparian grazing has likewise impaired native fish habitats. Restoration actions in the downstream direction should (at a minimum) include passive restoration action (grazing changes) and improvements to at least one road crossing. We recorded water temperatures higher than expected with a maximum of 73 °F near the mouth of Sauerkraut Creek. Record high temperatures basin-wide in 2007, reduction of riparian vegetation and recent increases in local beaver activity near the mouth all likely contribute to this warming. Measured flows up and downstream of the impacted sites indicate minimum base flow values range from -1.0 cfs at mile 3.0 and from -1.8 cfs to -2.7 cfs near the mouth. This range minimum instream flows near the mouth represent measured flows (1.8 cfs at mile 0.15) and modeled flows (2.7 cfs at the mouth). The lower-most western (unnamed) tributary to Sauerkraut Creek also supports WSCT. Restoration needs have not been assessed for this stream. Compared to Sauerkraut Creek, the influences of riparian degradation are more pronounced in Willow Creek. Bull trout were absent from the Willow Creek surveys, and WSCT densities decrease sharply upon entering more intensively managed pasturelands. Our surveys identified a sharp decreasing trend in WSCT densities in the downstream direction beginning at mile 5.7. Between stream-mile 4.7 and 3.6, salmonid populations were absent, and this absence of all salmonids continues to the lower Willow Creek confluence area. This absence includes more tolerant salmonids such as brook and brown trout usually found in lower reaches of adjacent tributaries. These species were present in 1999. Fish population surveys identified a disjunct population of WSCT in Bear Gulch with no other fish species present. WSCT were absent near the Bear Gulch confluence with Willow Creek Stream surveys identify damaged (incised and over-widened) channels, dewatering, fish passage barriers (culverts and diversions), impacts to vegetation and elevated water temperatures all within the lower five miles Willow Creek. In lower Willow Creek, maximum water temperature increase 10°F between mile 3.7 and 1.6 where temperatures exceeded 78''F. These findings of unsuitable native fish habitat confirm comprehensive restoration is needed for the lower five miles stream if some level of WSCT recovery to occur. Stream flow measurements suggest minimum instream flow values range from -2.3 cfs at mile 4.7 (based on measured flows) to a minimum flow -2.8 at the mouth (based on the Montana Method model). Willow Creek is reported as supporting a population of western pearlshell mussels. WPM is a long-lived, highly sensitive species with great conservation value. We made no observations of an existing mussel population during our fieldwork. A more comprehensive survey of fresh water mussels should be considered. 82 Similar to Willow Creek, densities of WSCT in Bear Gulch decrease significantly in the downstream direction. Similar to Willow Creek, we found no WSCT upon entering primary pasturelands. Assessments indicate dewatering and channel alterations (widening and degraded stream banks) near the mouth contribute to isolation and decreasing population trends in the downstream direction. Similar to other channels evaluated, grazing management changes are necessary for fisheries improvements in Bear Gulch. 83 Fisheries and Aquatic Assessments in the Clearwater River Watershed In terms of drainage area, the Clearwater River is the largest tributary system to the Big Blackfoot River. This complex system is comprised of an interconnected series of lakes and river sections, with many smaller coldwater tributary streams that enter throughout. The main stem flows south from its headwaters near Ptarmigan Mountain and the Swan River divide to the confluence with the Big Blackfoot River west of Ovando. Tributary basins stretch from the Swan Mountain divide and Bob Marshall Wilderness border on the east to the upper Jocko River divide (Flathead Indian Reservation) on the west. ff^r^\ Clearwater Lake Raiijv Lake Clearwater Drainage • Town /\/ Highway A/ Stream ^1 Lake I I Watershed N w 20 MIfr* -^- Figure 1. Map of the Clearwater River watershed and Clearwater chain of lakes. The Clearwater System is a unique drainage within the Blackfoot Watershed and Clark Fork Basin for a number of reasons. The drainage still supports exceptional and diversified aquatic resources, including many native fish populations with unique life history characteristics. Because of the interconnected stream and lake environments, species richness is high and adfluvial migratory life forms are common. The Clearwater Drainage supports some of the only natural and currently viable adfluvial bull trout populations in the region. These resources are largely intact due to the predominant public land (USFS and DNRC) and the Plum Creek Timber (PC) Company land base in the watershed that has (thus far) 84 precluded widespread subdivision and human development. Despite high aquatic resource value, this system has also been unique in the Blackfoot watershed because of the absence of data describing population trends, habitat use and key liming factors for native fish and other aquatic species. Fisheries emphasis and restoration accomplishments on streams in the greater Blackfoot and Clark Fork basins have generally not yet included the Clearwater system. The need for information has recently become imminent as the rapid conversion of corporate timberlands to smaller residential properties has forced natural resource managers and conservation advocates to prioritize lands for protection and acquisition. The need for updated and accurate aquatic resource information prompted a recent large- scale effort to collect baseline data for the Clearwater watershed. Since 2005 this effort has focused on the tributary and river systems, and emphasized native trout species (bull trout and WSCT). Assessments that are underway include: 1) Basin-wide stream sampling (electrofishing) to determine fish species distribution and relative abundance in all fish-bearing reaches and amphibian presence/absence. Sampling and analysis of Oncorhynchus spp. genetic composition to identify where genetically non-introgressed WSCT populations persist. Adfiuvial bull trout telemetry involving all known viable lake populations. 4) Bull trout redd counts in known spawning tributaries for migratory populations. 5) Assessment of main stem Clearwater River barriers to upstream fish passage (graduate student project in cooperation with the USFS and the University of Montana). Assessment of road crossings on private lands to identify barriers to upstream fish passage and limiting factors for natural stream processes. Assessment of stream protection practices on private lands; specifically riparian clearing and compliance with Montana stream protection laws. Monitoring of temperature regimes in major tributaries and main stem reaches. Collection of wetted perimeter and flow measurements on numerous tributaries. These efforts supplement ongoing lake monitoring by FWP and tributary habitat assessment and fish passage work by the USFS. Since all of these activities are ongoing, results will be reported in subsequent documents. However, the following sections describe basic information for key river sections and tributaries that support bull trout and migratory WSCT populations. These streams (in alphabetical order) are highlighted to emphasize their importance within the basin. Waters supporting bull trout and migratory WSCT are prioritized because these are Species of Concern in Montana and federally listed as Threatened under the Endangered Species Act (bull trout). These streams likely also represent the most intact and critical aquatic habitats for protection in the Clearwater System as native trout are an excellent indicator of habitat quality due to their sensitivity to human- caused perturbations and complex life-histories. 85 Summaries for Selective Clearwater Drainage Streams Blind Canyon Creek c 6 - Sample Sites ^ >i y\'"———_ ^-'^^ S 1 uj 4 ■ ^^^-^-^^^ WCT, EBT, DV Epheiiieml Stieinii USFSI PC 1 USPS 1 PCTC 1 USPS 1 2 3 4 5 6 7 Siren ni Mileage Blind Canyon Creek is a second- order tributary of Trail Creek located two miles northeast of the town of Seeley Lake in the Morrell Creek watershed. The main stem flows generally west, then south, from its headwaters near Devine Peak and Morrell Mountain. It has no major tributaries. Blind Canyon Creek is an important stream for native salmonids in the Clearwater drainage, with limited use (inferred) by migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) reflect relatively high habitat quality, abundant WSCT, and the presence of stream-resident bull trout. Boles Creek likely supports migratory WSCT and possibly adfluvial bull trout population, although this has yet to be verified. Figure 2. Longitudinal profile for Blind Canyon Creek. Land Ownership and Habitat Conditions Lower and middle Blind Canyon Creek flows through a "checkerboard" of public (USPS) land and industrial forest (Plum Creek Timber Company-PC) land (Figure 2). The upper drainage (~ 5 mi^) consists entirely of roadless USFS holdings. Blind Canyon Creek enters Trail Creek at approximately river mile 3.9. Stream gradients range from 700 ft/mile (~ 13.3 %) near its headwaters to about 116 ft/mile (~ 2.2 %) near its mouth at Trail Creek. The majority of the upper Blind Canyon Creek watershed consists of functional, non- degraded aquatic, riparian and terrestrial habitats. The upper drainage is primarily intact forest, with large roadless tracts. Riparian areas, water quality and channel morphology are largely intact, providing shade, instream habitat complexity, consistent recruitment of woody debris and adequate stream buffers. However, the lower 4.2 miles of the Blind Canyon Creek watershed is approximately half industrial forest and has higher road densities and recent logging, with some examples of poor road drainage that lead to increased water temperatures, increased sediment levels, and decreased habitat quality. Overall, riparian corridors are intact, helping to mitigate land management-related impacts. The impact of road crossings and the road matrix are unknown for Blind Canyon Creek. There are at least two crossings on the main stem channel and one is a bridge that likely does not prevent upstream fish passage or significantly impair natural stream integrity. There is also an undersized culvert on a small, ephemeral tributary (section 20) that has been determined to be a partial barrier to fish movement. The density and location of roads in certain sections is also a concern, particularly where the road encroaches on the stream and riparian corridor in middle reaches. 86 Temperature Regime Little data is available which describes the water temperature regime in Blind Canyon Creek. Water temperatures collected in conjunction with other investigations (i.e., electrofishing surveys) in late June and early July indicate a range of 8-11. 5°C. It is unlikely that temperatures exceed optimal levels for native trout (~ 15°C) later in the summer. Native salmonids, including bull trout and WSCT, require cold water in spawning, staging and rearing habitats. Future monitoring of water temperatures is planned for Blind Canyon Creek. Fish Populations Since 1995, three locations on Blind Canyon Creek have been sampled using backpack electrofishing to determine fish species composition and Oncorhynchiis spp. genetic contribution (see Figure 3). These locations were at stream miles 1.6, 3.3, and 4.5. These samples indicated that Blind Canyon Creek is dominated by native species, primarily WSCT. Bull trout and brook trout were also documented at all three sites. Most of the WSCT are mildly genetically introgressed, having hybridized with rainbow trout. Surveys and genetic analyses conducted in 2006 and 2007 suggest that the average genetic contribution of WSCT is 99.7 % for the population. Sculpins have been documented in the lower end of Blind Canyon Creek. Proportion of Catch (Combined data, 1995-2007) n 9. n fi n A n 9 m H ■ ^^ Immmii <^^ v^/y/M II: |: |: m.'iM I) ^ \mmd ■ DV ElEBT ■ RBT HRBxCT qwct 1.6 3.3 Location (stream mile) 4.5 Figure 3. Proportion of catch for salmonid species at three sites on Blind Canyon Creek, 1995-2007. Amphibian Community Tailed frogs have been observed at the lower two sample sites on Blind Canyon Creek. Failure to detect Columbia spotted frogs does not necessarily indicate absence. 87 Sample Sites EBT. DV, WCT 1 EBT, WCT 1 Peieiiiiial Stieiini 1 PCTC 1 USPS 1 PC 1 USPS 1 4 5 6 7 Stream Mleaji^ Figure 4. Longitudinal profile for Boles Creek. Boles Creek Boles Creek is a second- order tributary of Placid Creek located approximately six miles southwest of the town of Seeley Lake. The main stem flows generally east, then north, from its headwaters near Gold Creek Peak and Elk Meadow. Boles Creek enters Placid Creek just upstream of Placid Lake. It's most major tributary is an unnamed stream that enters from the south and originates at Spook Lake. Boles Creek is an important stream for native salmonids in the Clearwater drainage, with apparent limited use by migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) refiect relatively abundant stream-resident bull trout and WSCT populations, and high habitat quality. Boles Creek may also support modest adfiuvial bull trout and WSCT populations, although this has yet to be verified and warrants further investigation. These migratory populations likely reside in Placid Lake and use Boles Creek for spawning and rearing. Land Ownership and Habitat Conditions Boles Creek flows through a "checkerboard" of public (USPS) land and industrial forest (PC) land (Figure 4). It enters Placid Creek at approximately river mile 0.2, just upstream of Placid Lake. Stream gradients range from 300 ft/mile (~ 5.7 %) near stream mile 3 to about 33 ft/mile (~ 0.6 %) near the headwaters at Elk Meadow. No major dewatered reaches have been observed and human water use is limited. The Boles Creek watershed consists of marginally functional, semi-degraded aquatic, riparian and terrestrial habitats. The entire drainage is a checkerboard of primarily intact forest and managed industrial forest with high road densities and logging. Where Boles Creek flows through USPS lands, riparian areas, water quality and channel morphology are largely intact, providing shade, instream habitat complexity, consistent recruitment of woody debris and adequate stream buffers. However, portions of the road system lie directly adjacent to the main stem and encroach on the stream corridor in many locations. Where Boles Creek flows through industrial forest, high road densities are evident. Riparian encroachment and road drainage lead to increased water temperatures, increased sediment levels, and decreased habitat quality. Recent wildfires in upper and middle portions of the drainage (2003, 2007) have also had a significant impact on the stream. These will undoubtedly contribute to increased water temperature and sediment levels in the short term. Wildfires also expedited timber management (salvage) activities in 2007-2008. Road locations, maintenance and stream crossings may be limiting factors for native fish populations, but need to be more thoroughly assessed. Boles Creek has a total of four crossings that cross the main stem channel. Two of these crossings are bridges that likely 88 have little impact on fish movement or natural stream integrity. The others are open-bottom arch and circular culverts which do not likely impede fish movement at most flow levels. Temperature Regime The water temperatures in Boles Creek have not been rigorously monitored. Water temperatures collected in conjunction with other investigations have been marginal, ranging fi-om 12° C to 16.5° C in late June and early July (2006). It is likely that temperatures often exceed optimal levels for native fish later in the summer. Native salmonids, particularly bull trout, require cold water in spawning, staging and rearing habitats. Non-native salmonids, such as brook trout and brown trout, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures above 15° C (-18° C for WSCT). It is likely that maximum water temperatures are a limiting factor for native trout in Boles Creek. More comprehensive data will be collected to evaluate water temperature in 2008. Fish Populations 1 0.8 0.6 0.4 0.2 Proportion of Catch (Combined data, 1995-2006) m 0.1 0.9 2.5 Location (stream mile) 3.5 7.0 ■ DV ElEBT HLL ■ DVxEBT H RBxCT QWCT Figure 5. Proportion of catch for trout species at five locations on Boles Creek in 1995-2006. Since 1995, five locations on Boles Creek have been sampled by backpack electrofishing to determine fish species distribution and Oncorhynchus spp. genetic composition (see Figure 5). These locations were at stream miles 0.1, 0.9, 2.5, 3.5, and 7.0. These samples have indicated that Boles Creek is dominated by non-native species, primarily brook trout. Native fish populations appear to be genetically introgressed. WSCT have been hybridized with rainbow trout and genetic analyses (2006, FWP, unpublished data) suggest a hybrid swarm with > 98% WSCT genetic contribution. Although the population has not been tested, morphological characteristics indicate that bull trout x brook trout hybrid individuals are common. Sculpins, longnose suckers, and a brown trout have also been observed in lower stream reaches. Amphibian Community Anecdotal observations suggest that tailed frogs are common at the majority of sites on Boles Creek. Lack of data on other species does not confirm absence. 89 Deer Creek o 6- 1 £5- 1 i E 4- Siimple Sites ^y Westslo|)e Cutthroat Trout, Brook Trout Perennial Stieam USPS Plum Creek Timl)er Company | USPS C 1 2 3 4 5 6 7 3 9 10 11 Stieam Mileage Figure 6. Longitudinal profile for Deer Creek. Deer Creek is a third-order tributary of the Clearwater River system located approximately three miles northwest of the town of Seeley Lake. This stream fiows directly into Seeley Lake at the northwest corner of the lake. Deer Creek flows generally east and drains the area between the Marshall Creek and Placid Creek watersheds. It has one major tributary (Fawn Creek), which enters from the south at stream mile 3.4 and two smaller tributaries. These include Sheep Creek, which enters from the south (mile 6) and an un- named tributary that enters from the north (mile 8). Deer Creek is an important stream for native salmonids in the Clearwater drainage, particularly for WSCT. Bull trout have also been detected in low densities, but population viability is questionable. Because the stream flows directly into Seeley Lake, use by adfiuvial trout is likely. However, this has not been verified. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see prioritization section) reflect relatively high habitat quality, relatively abundant WSCT populations, and high potential for use by migratory life forms. Land Ownership and Habitat Conditions Most of Deer Creek flows through industrial forest (PC) land (Figure 6). The uppermost ~1 mile lies on USFS lands. Deer Creek enters the Clearwater River at river mile 24.8 (Seeley Lake). Stream gradients range from 635 ft/mile (~ 12.0 %) near the stream's headwaters to about 49 ft/mile (~ 0.9 %) near its confiuence with Seeley Lake. No major dewatered reaches have been observed and human water use is currently minimal. The majority of the Deer Creek watershed consists of large tracts of undeveloped timber land with high road densities and extensive timber management. Riparian areas have been compromised by past logging in some areas and the stream is likely affected by impacts associated with the extensive road matrix (e.g., increased sediment levels and water temperatures, altered hydrology). Only the extreme upper watershed (~ 1 mi^) is comprised of intact, roadless forest. Middle and lower portions of the drainage still maintain high stream habitat quality and generally intact stream corridors that are capable of recovery where impacts have been observed. Road maintenance and stream crossings on Plum Creek Timber Company roads need to be thoroughly evaluated. There are currently five stream crossings on the main stem of Deer Creek. These crossings are all bridges that likely have little impact on fish movement or natural stream integrity. There are also two crossings of Fawn Creek. Both are pipe-arch culverts that are suspected to be partial barriers to fish movement. Additionally, there is one crossing of Sheep Creek. It is also a pipe-arch culvert that is suspected to be a partial barrier to fish movement. 90 Temperature Regime The water temperature regime in Deer Creek has not been rigorously monitored. Anecdotal water temperatures collected in conjunction with other investigations have been marginal, ranging from 9° C to 18° C in mid-summer (July and August) 2006. It is likely that temperatures often exceed optimal levels for native fish throughout the summer. Native salmonids, particularly bull trout, require cold water in spawning, staging and rearing habitats. Non-native salmonids, such as brook trout and brown trout, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures >15° Celsius (>18° C for WSCT). It is likely that maximum water temperatures in Deer Creek are a limiting factor for native trout populations. Fish Populations Since 1995, five locations on Deer Creek have been sampled by backpack electrofishing to determine fish species distribution and Oncorhynchiis spp. genetic composition (see Figure 7). These locations were at stream miles 0.8, 2.6, 4.4, 5.6, and 7.7. These samples indicate that Deer Creek is dominated by WSCT (native) and brook trout (introduced). However, the population of WSCT has been slightly hybridized with rainbow trout. Bull trout were observed in very low densities in 1995 and 2000, but have not been detected in Deer Creek since. The upper end of Deer Creek has been sampled in 2000 and 2006 at stream mile 7.7, and no fish were documented either time. A very high-gradient area near stream mile 7.4 is likely a total fish barrier. Lower reaches of Deer Creek have a higher proportion of non-native species (primarily brook trout) relative to middle and upper reaches. Brown trout and long nose dace have been documented in Deer Creek in very low numbers, and sculpins are present throughout the fish-bearing portion of the drainage. Proportion of Catch (Combined data, 1995-2007) 0.8 0.6 0.4 0.2 m 11 No Fish hDV hebt hll hRBxCT nwcT 0.8 2.6 4.4 5.6 Location (stream mile) 7.7 Figure 7. Proportion of catch for trout species at five locations on Deer Creek in 1995-2007. Surveys and genetic analyses conducted by FWP in 2006 suggest that the Deer Creek WSCT population is a hybrid swarm with a 99.5% WSCT genetic contribution. However, a limited sample also suggests that the population of WSCT found in Sheep Creek may be genetically non-introgressed. This is significant because most WSCT populations in the 91 Clearwater River drainage have been mildly hybridized with introduced rainbow trout, Yellowstone WSCT or both. Amphibian Community Tailed frogs have been observed at nearly every sample site on Deer Creek. Columbia spotted frogs have also been observed in multiple sites on Deer Creek. Occurrence of both species seems to be widespread in the Deer Creek drainage. East Fork Clearwater River EH an SamnlB S^w IJ5FS 1 5l| ^.-l^l^HiJ^ Figure 8. Longitudinal profile for East Fork Clearwater River. The East Fork of the Clearwater River (hereafter, "East Fork") is a second-order tributary to the Clearwater River, located approximately 13 miles north- northwest of the town of Seeley Lake. It flows generally west from its headwaters near Sunday Mountain. It has one major tributary, an un-named stream that enters from the north at stream mile 1.2. The East Fork is a very important stream for native salmonids in the Clearwater drainage, particularly migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) reflect relatively abundant adfluvial and stream-resident bull trout and WSCT populations and high habitat quality. Native salmonid populations in the East Fork are likely not as abundant as those in the West Fork Clearwater or Morrell Creek, but this stream is a very important stronghold for these species in the watershed. The East Fork supports one of the four known adfluvial spawning populations in the Clearwater Drainage. This stream is particularly important for the viability and persistence of the Rainy Lake bull trout population, as it is currently the only spawning stream accessible via upstream and downstream movement of adults and sub-adults. Currently, a dam on the main stem Clearwater River (at stream mile 38.6) below Rainy Lake acts a complete barrier to upstream fish passage. Recent radio telemetry investigations have demonstrated that adfluvial bull trout emigrate downstream from Rainy Lake. When this occurs, fish are unable to return to Rainy Lake or the primary spawning tributary (East Fork). Land Ownership and Habitat Conditions The East Fork watershed lies entirely on USES lands (Figure 8). It enters the Clearwater River (outlet of Clearwater Lake) at approximately river mile 40.4, upstream of Rainy Lake. Stream gradients range from 557 ft/mile (~ 10.5 %) near the headwaters to about 166 ft/mile (-3.1 %) near stream mile 2. The majority of the East Fork watershed consists of functional, non-degraded aquatic, riparian and terrestrial habitats. The entire drainage is primarily intact forest, with modest 92 road densities and limited past logging. Riparian areas, water quality and channel morphology are largely intact, providing shade, instream habitat complexity, consistent recruitment of woody debris and adequate stream buffers along the entire main stem. Although the lower mile of the East Fork does have a road running very near it that may contribute elevated excess sediment, the USFS has recently completed road BMP upgrades and road crossing improvements in the watershed that have reduced any recognized impacts. The water temperature regime in the East Fork represents one of the lowest anywhere in the Clearwater River or Blackfoot River drainages. Native salmonids, particularly bull trout, require cold water in spawning, staging and rearing habitats. Non-native species, such as brook trout, are generally more tolerant of warmer water temperatures. Bull trout become stressed when subjected to maximum daily temperatures > 15° C (> -18° C for WSCT). Nearly all other larger streams (capable of supporting bull trout) in the Clearwater River drainage exceed this temperature during summer months. Because the East Fork typically does not exceed 15° C (see Figure 9), native fish and other species are able use this tributary as a thermal refuge, as well as for spawning and other life stages. 0) a. Upper East Fork Clearwater River Daily Min., Max., and Avg. (2007) ^ — OOIDCNOJIDCNOJCD CN CO r^ r^ Fi 15 65 00 00 00 ID CN O) (D CO 0)0)0) Figure 9. Temperature data from the Upper East fork Clearwater River in 2007. Fish Populations Since 1995, six locations have been sampled using backpack electrofishing to determine fish species composition and Oncorhynchus genetic composition. These locations were at stream miles 1.1, 2.5, 3.2, and 4.9, as well as two sites in the un-named tributary that enters the East Fork at stream mile 1 . 1 from the north. These investigations have revealed that the stream is dominated by native species (Figure 10). Brook trout have been found in the main stem of the Clearwater River, just 1.5 miles upstream of the mouth of the East Fork, and may eventually move into the East Fork. Brook trout were not observed in Rainy Lake or any of the drainage upstream of the lake until the late 1990s when they were illegally introduced into Clearwater Lake. Sculpins have also been observed in the upper end of the unnamed tributary. 93 1 n 0.8 0.6 ^ 0.4 ^ 0.2 Proportion of Catch p- HDV QWCT H RBxCT iw IN i ^ 1.1 2.5 Location (stream m 3.2 4.9 T-0.3 T-2.4 ile) Tributary Figure 10. Proportion of catch for salmonids at six locations on the East Fork of the Clearwater River in 1995-2007. Although the East Fork of the Clearwater River is strongly dominated by native species, the population of WSCT has been hybridized with rainbow trout and Yellowstone WSCT. Surveys and genetic analyses conducted in 2006 and 2007 suggest that the average genetic contribution of WSCT is 97% for the population. Amphibian Community Although not rigorously investigated, it appears that the amphibian community of the East Fork of the Clearwater River is limited. Tailed frogs have been documented in the lower end of the unnamed tributary. No other amphibians were observed during limited stream surveys, but Columbia spotted frogs and other species may be present. Marshall Creek _ 5 1 S (1111 pie Sites - " Lake l^hirsiiaii EBT. WCT 1 DV, EBT, RBT, WCT Peieiinitii Stieciiii Plum CieekTiiiil)er Coni|)iiiiy USPS Marshall Creek is a second order tributary of the West Fork of the Clearwater River (hereafter. West Fork) located 6-10 miles northwest of the town of Seeley Lake. It flows generally east and drains the divide between the Clearwater River and Jocko River watersheds, near Sunset Peak. It has no major tributaries, although it does flow through Lake Marshall. Marshall Creek essentially consists of two reaches: 1) the upper reach from headwaters to Lake Marshall and 2) the outlet of Lake Marshall to the confluence with the West Fork. 3 4 5 Sti einn Mileage Figure 11. Longitudinal profile for Marshall Creek. 94 Marshall Creek is an important stream for native salmonids in the Clearwater drainage, particularly for migratory bull trout. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see prioritization section) reflect relatively high aquatic habitat quality and the presence of bull trout in Lake Marshall and in stream reaches upstream of the lake. Limited data is available for the lower reach between the lake outlet and West Fork. Marshall Creek supports a small, disjunct adfluvial bull trout population and represents one of four known adfluvial spawning populations in the Clearwater Drainage. Upper Marshall Creek appears to be the only spawning stream utilized by adfluvial bull trout from Lake Marshall (based on the distribution of juvenile fish). Very high water temperatures (typical of lake outlets) in Marshall Creek downstream of Lake Marshall likely discourage use by bull trout from June- September and likely makes this reach unsuitable as year-round juvenile rearing habitat. Recent radio-telemetry investigations have demonstrated that adfluvial bull trout in the West Fork of the Clearwater River enter lower Marshall Creek in the spring, when water temperatures were low, and returned to the West Fork after a short time. Upper Marshall Creek also supports an abundant migratory rainbow trout population that was previously introduced into the lake. This population has hybridized with and overwhelmed the native WSCT population in the upper drainage. Land Ownership and Habitat Conditions Most of Marshall Creek flows through industrial forest (PC) land (Figure 11), although the uppermost ~2 miles lie on USFS lands. Marshall Creek enters the West Fork of the Clearwater River at river mile 4.9. Stream gradients range from 445 ft/mile (~ 8.4 %) near stream mile 5 to about 5 ft/mile (<1%) just above Lake Marshall. The majority of the Marshall Creek watershed consists of large tracts of undeveloped land with high road densities and logging (primarily industrial forest). Riparian areas have been compromised by past logging in some areas, and the stream is likely affected by poor road drainage. These factors likely contribute to increased sediment levels and water temperatures, and decreased habitat quality. However, the headwaters are comprised of intact, roadless forest and functional riparian and aquatic habitats. Middle and lower portions of the drainage still maintain high stream habitat quality and generally intact stream corridors that are capable of recovery where impacts have been observed. Two stream crossings have been built in the Marshall Creek drainage and both cross the main stem channel. These crossings are both bridges that likely have little impact on fish movement or natural stream integrity. The water temperature regime in Marshall Creek downstream of Lake Marshall is above the optimal range of temperatures for native salmonids. Although data from 2007 (Figure 12) were biased by unusually low water and warm temperatures, this section of stream was not nearly as resilient as other coldwater tributaries in the drainage under these conditions. However, the water temperatures in Marshall Creek upstream of Lake Marshall have not been monitored. Water temperatures collected in conjunction with electrofishing investigations indicate a range of 8.5°- 12° C in early to mid- September. It is unlikely that temperatures often exceed optimal levels for native salmonids later in the summer, but future monitoring will be conducted to confirm this. Native salmonids, particularly bull trout and WSCT, require cold water in spawning, staging and rearing habitats. Non-native species, such as brook trout and rainbow trout, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures > 15° C (>18° C 95 for WSCT), which occurred regularly for approximately twelve weeks in 2007 downstream of the lake. In this system, native salmonids may use lake stratification as a thermal refuge during warm summer months. o 3 oi a. E 01 Marshall Creek Below Lake - 2007 Daily Min., Max., and Avg. 00 55 CM en CD CO 1^ 00 00 £N £N S2 o CO Figure 12. Temperature data from Marshall Creek downstream of Lake Marshall. Fish Populations Since 1995, four locations on Marshall Creek have been sampled using backpack electrofishing to determine fish species composition and Oncorhynchiis spp. genetic contribution (see Figure 13). These locations were at stream miles 1.8, 3.6, 4.5, and 5.0. These samples have indicated that Marshall Creek is dominated by non-native species, primarily brook trout and rainbow trout. Proportion of Catch (Combined data, 1995-2007) 0.5 I ^ 1.8 3.6 4.5 Location (stream mile) 5.0 HDV BDVxEBT SEBT ■ RBT nwcT Figure 13. Proportion of catch for salmonids at four locations on Marshall Creek in 1995-2007. The bull trout population inhabiting the lake and upper stream reaches currently exists at very low densities based on electrofishing and lake gill net surveys. The population of WSCT has been hybridized with rainbow trout. Surveys and genetic analyses conducted in 2006 and 2007 suggest that the average genetic contribution of WSCT is less than 5% for the population upstream of the lake. Since 1971, WSCT are the only fish species that have been 96 stocked in the Marshall Creek drainage, all in Lake Marshall. Prior to this (post 1942), rainbow trout and unspecified WSCT were stocked in Lake Marshall periodically. In addition, the only known stocking of Lake Dinah (at the headwaters of Marshall Creek) was with rainbow trout in 1941. Although the bull trout population has not been genetically tested, multiple bull trout x brook trout hybrid individuals have been observed in upper Marshall Creek. Long nose dace and mountain whitefish have also been found in the lower end of Marshall Creek. Species captured in Lake Marshall gill net surveys include mountain whitefish and longnose sucker, in addition to the salmonids species already mentioned. Amphibian Community Tailed frogs have been documented at the majority of sites on Marshall Creek. Columbia spotted frogs and other species have not been observed, but may be present. Morrell Creek ffrw I" 8 a ID 11 12 13 -14 15 18 17 ia ID U Morrell Creek is a third order tributary of the Clearwater River located near the town of Seeley Lake. The main stem flows generally southward from its headwaters near Matt Mountain and Sunday Mountain. Major tributaries, most of which form the Trail Creek system and Drew Creek, converge and enter Morrell Creek at stream mile 0.5. These streams are described in separate sections of this report. Morrell Creek is a very important stream for native salmonids in the Clearwater drainage, particularly migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) reflect relatively abundant adfluvial and stream-resident bull trout and WSCT populations and high habitat quality. Morrell Creek supports one of the two largest adfluvial bull trout populations in the upper Clark Fork Basin and represents one of four known adfluvial spawning populations in the Clearwater Drainage. Morrell Creek is particularly important for the viability and persistence of the Seeley Lake and Salmon Lake bull trout populations, as it is currently the only spawning stream readily accessible to these fish. Morrell Creek also supports an abundant migratory WSCT population and a genetically non-introgressed stream- resident population in upper reaches. Figure 14. Longitudinal profile for Morrell Creek. Land Ownership and Habitat Conditions Most of Morrell Creek flows through public (USPS and DNRC) land and industrial forest (one section of PC) land (Figure 14), but the lower three miles flow through highly developed private lands, including the Double Arrow Golf Course and Subdivision. The stream enters the Clearwater River at approximately river mile 17.3, between Seeley Lake and Salmon Lake. Stream gradients range from 690 ft/mile (-13 %) near the headwaters to about 40 ft/mile (~ 0.8 %) near the mouth. 97 The majority of the Morrell Creek watershed consists of functional, non-degraded aquatic, riparian and terrestrial habitats. The upper drainage is primarily intact forest, with large roadless tracts and some sections with modest road densities and logging (primarily industrial forest). Riparian areas, water quality and channel morphology are largely intact, providing shade, instream habitat complexity, consistent recruitment of woody debris and adequate stream buffers along middle and upper reaches. However, the lower 2.5 miles of Morrell Creek are plagued with detrimental land-use practices, including stream encroachment, removal of riparian vegetation, poor road drainage, and excessive water diversion, which contributes to stream dewatering and fish entrainment losses in irrigation ditches. These problems lead to increased water temperatures, increased sediment levels, and decreased habitat quality. A ~ 0.25-0.5 mile stretch of stream that dewaters in low water years in the fall (stream mile 1.0) prevents upstream and downstream movement of fish, which is particularly problematic for migrating adult bull trout attempting to reach spawning areas. A natural dewatered section is also present at stream mile 9.5-10.5. This section typically loses surface fiows in August and remains dry until late fall or spring runoff Road maintenance and stream crossings on USPS and county roads could be improved, but are not considered a limiting factor for fish in this drainage. Road crossings on Morrell Creek are nearly all bridges and are not considered impairments to upstream fish passage or natural stream integrity. Temperature Regime Water temperatures in Morrell Creek are cold relative to other tributaries in the Clearwater River drainage (see Figures 15-17). Native salmonids, particularly bull trout and WSCT populations, require cold water in spawning, staging and rearing habitats. Non-native salmonids, such as brook trout and brown trout, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures above 15° C (-18° C for WSCT). Nearly all third order or larger streams in the Clearwater River drainage frequently exceed this temperature. Abundant native fish are an indication of relatively intact habitat quality and water quality (see figures 2, 3, 4); native fish are able to use Morrell Creek during all stages of their lives. Upper Morrell Creek 2007 Daily MIn., Max., and Avg. Figure 15. Temperature data from upper Morrell Creek in 2007. 98 Lower Morrell Creek 2007 Daily Min., Max., and Avg. o 3 a! a. E 0) Figure 16. Temperature data from lower Morrell Creek in 2007. Lower Morrell Creek 2006 Daily Min., Max., and Avg. CD CM en CD CD C! S S ^ £i! in CD CD CD ■*-!-(X3-5r-!-(X3ir5T- CX3 CX3 CX3 CX3 in CM C35 CD C3^ C3^ C3^ T- CM CM Figure 17. Temperature data from lower Morrell Creek in 2006. Fish Populations Since 1995, thirteen locations on Morrell Creek have been sampled for fish species composition and Oncorhynchus spp. genetic analysis (see Figure 18). These locations were at stream miles 0.4, 1.6, 2.6, 4.4, 4.5, 6.1, 7.8, 10.2, 10.5, 13.4, and 14.2, as well as two sites in the unnamed tributary that enters Morrell Creek from the north, just downstream of Morrell Falls, at (tributary) stream miles 0.8 and 1.3. These samples have indicated that the majority of Morrell Creek is dominated by native species. However, as is often the case with degraded habitats, the lower end of Morrell Creek (~3 miles) is dominated by non-native species (brown trout and brook trout) and WSCT x rainbow trout hybrids. Hybrids have been detected throughout lower portions of the Morrell Creek system, but upper reaches (above stream mile 9.5) appear to be non-introgressed. Although not rigorously investigated, sculpins are present in Morrell Creek from stream mile 4.5 to Morrell Falls and possibly in the lower 4.5 miles as well. Morrell Falls is at stream mile 13.8, and is a barrier to fish movement. Surveys conducted in 2007 (stream mile 14.2) revealed no fish above Morrell Falls in Grizzly Basin. 99 Proportion of Catch (Combined data, 1995-2007) 1 -, 0.8 b-M BDV HEBT bll ■ RBT H RBxCT QWCT 0.4 1.6 2.6 4.4 4.5 6.1 7.8 10.2 10.5 13.4 14.2 0.8 1.3 Location (stream mile) (Tributary) Figure 18. Proportion of catch for salmonids at thirteen locations on Morrell Creek in 1995-2007. Adult bull trout implanted with radio transmitters in 2006 in Seeley Lake and Morrell Creek provided important information on migration timing, habitat use, spawning behavior, and limiting factors. These findings will be summarized in the Clearwater bull trout telemetry project report. Surveys and genetic analyses conducted in 2007 suggest that the population of WSCT found in upper Morrell Creek (above the intermittent stretch from stream mile -9.5 to -10.5) is genetically non-introgressed. This is significant because most WSCT populations in the Clearwater River drainage (including lower Morrell Creek and the Trail Creek drainage) have been mildly hybridized with introduced Yellowstone WSCT, rainbow trout, or both. WSCT make up > 99% (genetic contribution) of Oncorhynchus populations in middle and lower reaches of Morrell Creek. Amphibian Community Tailed frogs have been observed from stream mile 4.5 to above Morrell Falls, possibly in the lower 4.5 miles also. Columbia spotted frogs have also been observed in Morrell Creek. However, locations are anecdotal and distribution appears to be limited. Trail Creek Trail Creek is a third- order tributary of Morrell Creek, located just east of the town of Seeley Lake. It flows generally south and drains the west side of the Pyramid Pass area between and 7 - o 6 - Siiniple Sites / Jl5- ^^^Y^^^^:^-— ^^^X 1 4- LU 3 - 2 - C -^^^-""^ EBT, WCT, LL | EBT, WCT Peieniiifil Stiefim Piiviite IPCTC USFSIPCTCIUSFSIPCI USPS 123456789 10 Sti'eiiiiiMileii<|e Figu re 19. Longitudinal proflle for Trail Creek. 100 Pyramid Peak and Devine Peak. The Trail Creek drainage includes three major tributaries, Blind Canyon Creek, Swamp Creek and Mountain Creek, which all enter from the east between stream mile 2.0 and 3.9. Blind Canyon Creek is described in a separate section of this report, while other tributaries will be included in subsequent reports. Trail Creek is an important stream for native salmonids in the Clearwater drainage, particularly migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see prioritization section) reflect relatively high habitat quality, relatively abundant migratory and stream resident WSCT populations (upper reaches), and high potential for use by migratory and stream resident bull trout. Land Ownership and Habitat Conditions Most of Trail Creek flows through public (USPS) land and industrial forest (three sections of PC land), but the lower two miles flow through highly developed private lands, including the Double Arrow Golf Course and Subdivision (Figure 19). Trail Creek enters Morrell Creek at approximately stream mile 0.8, on the Double Arrow Golf Course. Stream gradients range from 1055 ft/mile (~ 20 %) near the headwaters to about 50 ft/mile (~ 0.9 %) near stream mile 3. The upper and middle portions of Trail Creek watershed consist of generally functional aquatic, riparian and terrestrial habitats. These reaches lie within "checkerboard" ownership of primarily intact USPS holdings and managed industrial forest with higher road densities and logging. Only the extreme upper watershed (~ 4 mi^) is comprised of intact, roadless forest and fully functioning riparian and aquatic habitats. Where Trail Creek flows through other USPS lands, riparian areas, water quality and channel morphology are largely intact, providing shade, instream habitat complexity, consistent recruitment of woody debris and adequate stream buffers. On industrial forest properties, habitat quality varies with management intensity and road characteristics. Riparian corridors have been maintained (consistent with Montana SMZ laws), but removal of some riparian vegetation and road drainage issues likely contribute to increased sediment levels and decreased habitat quality. Middle portions of the drainage still maintain relatively high stream habitat quality and stream corridors are capable of recovery where impacts have been observed and addressed. However, the lower two miles of Trail Creek are plagued with detrimental land- use practices, including stream encroachment, removal of riparian vegetation, and poor road drainage. As the stream passes through the Double Arrow Subdivision and Golf Course, riparian clearing and human encroachment are common. Road maintenance and stream crossings on USPS, PC, and Missoula County roads could be improved, and may be a limiting factor for fish in this drainage. Trail Creek has at least six crossings that cross the main stem channel. Only one of these crossings has been surveyed, and it has been determined to be a partial barrier for fish passage. More comprehensive surveys of the crossings of Trail Creek will be conducted in the near future. Temperature Regime The water temperature regime of Trail Creek is cold relative to other tributaries in the Clearwater River drainage (see Pigure 20). Native salmonids, particularly bull trout and WSCT populations, require cold water in spawning, staging and rearing habitats. Non-native salmonids, such as brook trout and brown trout, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures 101 above 15° C (> -18° C for WSCT). Nearly all third-order or larger streams in the Clearwater River drainage frequently exceed this temperature. Trail Creek -2007 a. E o Figure 20. Temperature data from Trail Creek in 2007 at stream mile -3.8. Fish Populations Since 1995, seven locations on Trail Creek have been sampled for fish species composition and Oncorhynchus spp. genetic analysis (see Figure 21). These locations were at stream miles 0.5, 1.2, 1.8, 2.9, 3.8, 5.5, and 7.8. These samples indicate that most of Trail Creek is strongly dominated by non-native brook trout. However, despite relatively low densities of native fish species in Trail Creek, the population of WSCT found in upper reaches appears to be non-introgressed. Bull trout were found in Trail Creek surveys conducted in 1995 and 2002. Since then, the only evidence of bull trout in Trail Creek was a bull trout x brook trout hybrid observed in 2006. Brown trout have also been found in lower Trail Creek, in 2002 and 2006. Although not rigorously investigated, sculpins appear to be present in Morrell Creek from stream mile 2 to the top upper extent offish-bearing reaches. They likely also inhabit the lower 2 miles, but this has not been verified. Proportion of Catch (Combined data, 1995-2007) 1 0.8 0.6 0.4 0.2 ■ ■,!„,■ -,J,n, ■ „, ■ n, ■ n llDV ^EBT HLL HRBxCT DWCT 0.5 1.2 1.8 2.9 3.8 Location (stream mile) 5.5 7.8 Figure 21. Proportion of catch for salmonid species at seven sites on Trail Creek, 1995-2007. 102 Bull trout and other migratory species have been inhibited by an irrigation diversion dam near the mouth for at least two decades. In 2001, a Denil fish ladder was installed on the structure to provide upstream fish passage. Fish entrainment in the ditch and stream dewatering has also been issues at this location. Surveys and genetic analyses conducted in 2006 and 2007 suggest that the population of WSCT found in upper Trail Creek (above stream mile 6) may be genetically non- introgressed. In middle and lower portions of the stream, the WSCT population is hybridized with rainbow trout, but the WSCT genetic contribution is > 98%. Most WSCT populations in the Clearwater River drainage (including lower Trail Creek and lower Morrell Creek) have been mildly hybridized with introduced Yellowstone WSCT, rainbow trout, or both. Amphibian Community Although not rigorously investigated, tailed frogs have been observed in Trail Creek above stream mile 3.8. Columbia spotted frogs have not been noted in Trail Creek. Failure to detect Columbia spotted frogs does not necessarily indicate absence. West Fork Clearwater River The West Fork of the Clearwater River (hereafter "West Fork") is a third-order tributary of the Clearwater River located approximately six miles northwest of the town of Seeley Lake. It fiows generally southeast and drains the divide between the Clearwater River and Swan River watersheds, near Sunset Peak. There is only one major tributary, Marshall Creek, which enters the West Fork at stream mile 4.9. Marshall Creek is described in a separate section of this report. The West Fork is a very important stream for native salmonids in the Clearwater drainage, particularly migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see prioritization section) reflect relatively abundant adfluvial and stream-resident bull trout and WSCT populations, and high habitat quality. The West Fork supports one of the largest adfluvial bull trout populations in the upper Clark Fork Basin and represents one of four known adfluvial spawning populations in the Clearwater Drainage. The West Fork is particularly important for the viability and persistence of the Lake Inez and Lake Alva bull trout populations, as it is apparently the only suitable spawning stream accessible to these fish. Currently, a dam on the main stem Clearwater River (stream mile 29.3) between Seeley Lake and the mouth of the West Fork completely blocks upstream fish passage from downstream areas, including Seeley Lake and Salmon Lake. Recent radio- telemetry investigations have demonstrated that a large number of adfluvial bull trout from / Sample Sites •■" *-4- Diverse 1 EBT. WCT. DV |WCT. DV| No Fi5h Perennial Stream USFSl Plum Creek Timber Co. 1 USPS 7 8 9 10 11 STieiHii Mileage 12 13 14 15 ie Figure 22. Longitudinal profile for the West Fork Clearwater River. 103 Seeley Lake attempt to migrate into the West Fork of the Clearwater River, but are impeded by the dam. Other native fish populations, such as WSCT and mountain whitefish, are likely also impacted. Interim (manual) passage of bull trout was begun in 2007 and a permanent fish passage solution is being developed. The West Fork also supports an abundant migratory WSCT population and a genetically non-introgressed stream-resident population in upper reaches. Land Ownership and Habitat Conditions Most of the West Fork fiows through industrial forest (PC) land (Figure 22). The uppermost ~4 miles lies on USFS holdings and the lower -1.5 miles fiows through privately owned and USFS lands. The West Fork enters the Clearwater River between Seeley Lake and Lake Inez. Stream gradients range from 470 ft/mile (~ 9 %) near stream mile 12 to about 50 ft/mile (~ 0.9 %) near the mouth. The majority of the West Fork watershed consists of large tracts of undeveloped land with high road densities and logging (primarily industrial forest). Riparian areas have been compromised by past logging in some areas, and the stream is likely affected by poor road drainage. These factors likely contribute to increased sediment levels and water temperatures, and decreased habitat quality. However, the upper watershed (~ 7 mi^) is comprised of intact, roadless forest and functional riparian and aquatic habitats. Middle and lower portions of the drainage still maintain high stream habitat quality and generally intact stream corridors that are capable of recovery where impacts have been observed. Three stream crossings have been built in the West Fork drainage and all cross the main stem channel. These crossings are all bridges that likely have little impact on fish movement or natural stream integrity. The water temperature regime in the West Fork is above the optimal range of temperatures for native salmonids. Although data from 2007 (Figures 23 and 24) were biased by unusually low water and warm temperatures, this stream was not nearly as resilient as other coldwater tributaries in the drainage under these conditions. Native salmonids, particularly bull trout and WSCT, require cold water in spawning, staging and rearing habitats. Non-native species, such as brook trout and brown trout, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures > 15° C (-18° C for WSCT), which occurred regularly for approximately six weeks in 2007. The detrimental effects of these extreme temperatures likely contributed to high rates of adult bull trout mortality observed in 2007. Upper West Fork Clearwater River Daily Min., Max., and Avg. (2007) S! 3 a. E 0) CM 00 55 CM CM CD 05 CD CO 1^ £2 00 00 CM CM o m CM Figure 23. Temperature data from upper West Fork of the Clearwater River. 104 Lower West Fork Clearwater River Daily IVIin., Max., and Avg. (2007) 20 2 - />v /\ A^yr-O^ Avv'Wv A /A//^A.A^^^Sa--CH /^ y\ aaa//V/v-^ ^-wt^^^^x^::^^ . />i i/^/y/ ^"^^^^^tP^ /^ <^ .V/7 W^^-^ V ^--^ K^l^i^.. ^^ i/^:;^/^'-^^:^^^^'^^^:/^^ ^^^'v^'^%ip%3^. ^,--^v;^^7C-^«!=^-' N^-^^"^"'^;^ 1 1 ■ ■ ■ 1 ■ ■■■■■■■■■■■■■■■■■■■■■I 1 1 ■ " ■ 1 !■'■■■'■■■■'■■■■■■■■■■■ 1 Figure 24. Temperature data from lower West Fork Clearwater River. Fish Populations Since 1995, nine locations have been sampled to determine fish species composition and Oncorhynchus spp. genetic composition (Figure 25). These locations were at stream miles 2.4, 3.5, 5.3, 7.1, 9.6, 11.1, 11.2, 11.6, and 12.7. These investigations have revealed that, while the upper end of the stream is dominated by native species, non-native brook trout strongly dominate the lower ~8 miles of the West Fork. Sculpins are common throughout the drainage, and both long-nosed dace and mountain whitefish are present in the lower ~4 miles. A very high-gradient section near stream mile 12 appears to be a migration barrier. Surveys conducted in 2007 at stream mile 12.7 revealed no fish above the high gradient section. Proportion of Catch (Combined Data, 1995-2007) 1 0.8 0.6 0.4 0.2 — F=1 I JIL 11 DV 0EBT BLL QWCT EJMWF 2.4 3.5 5.3 7.1 9.6 11.1 Location (stream mile) 11.2 11.6 12.7 Figure 25. Proportion of catch for salmonids at nine locations on the West Fork of the Clearwater River in 1995-2007. Surveys and genetic analyses conducted in 2006 and 2007 suggest that the population of WSCT found in the upper-most fish-bearing reach (~ mile 11) of the West Fork is genetically non-introgressed. WSCT (> 99% genetic contribution) in middle and lower portions of the stream (below mile 11) have been hybridized with rainbow trout and 105 Yellowstone WSCT. Most WSCT populations in the Clearwater River drainage have been mildly hybridized with one or both of these introduced Oncorhynchus species. Amphibian Community Tailed frogs are present throughout the fish-bearing waters of the West Fork. Columbia spotted frogs have also been observed in the lower half of the drainage. Failure to detect them in upper reaches does not confirm their absence. Clearwater River Main Stem Sections 3.9 5 i 3.8 3.7 :^// Soiiii|>le Sites ==-= 7' Rainhow, Brown Peienniiil Stream Piiviite 1 State 1 Piivate 1 4 5 6 Stream MIeage 10 Figure 26. Longitudinal profile for Clearwater River Section I. Clearwater River Section I (Big Blackfoot River to Salmon Lake). The Clearwater River is the largest tributary drainage (fourth- order) of the Big Blackfoot River. The main stem Clearwater system is comprised of an interconnected series of lakes and river sections, with many smaller tributary streams that enter throughout. The main stem flows south from its headwaters near Ptarmigan Mountain and the Swan River divide to the confluence with the Big Blackfoot River west of Ovando. Its major tributaries include the East Fork of the Clearwater River, West Fork of the Clearwater River, Camp Creek, Morrell Creek, and the Placid Creek system (Owl Creek). Clearwater River Section I is the reach from the Salmon Lake outlet to the mouth. This reach encompasses Black "Lake", Blanchard "Lake" and Elbow "Lake", which are actually just wider, lower gradient portions of the river. The only major tributary stream in this section is Blanchard Creek, which enters from the west at stream mile 2.8. Two smaller tributaries. Lost Horse (mile 5) and Lost Prairie (mile 6) Creeks, also enter from the west. Characteristics of all of these drainages will be summarized in a future report. Clearwater River Section I is an important stream for salmonids in the Clearwater drainage, particularly migratory life history forms and sport fish. This reach is seasonally inhabited by salmonids and is suspected to be an important migratory corridor for fish moving between the Blackfoot River and upper Clearwater system. The extent of use of this corridor has not been determined for bull trout, WSCT or other native fish species, and definitely warrants further investigation. Anecdotal reports and field data highlight the seasonal abundance of non-native salmonids such as brown trout and the year-long use by high densities of introduced northern pike, particularly in slower backwater areas. Connectivity between the Clearwater and Blackfoot Rivers may be an important component of larger scale watershed native fish restoration efforts. Currently, hand-made rock dams and a large irrigation diversion may restrict upstream fish passage at low flows. 106 Land Ownership and Habitat Conditions The Clearwater River Section I flows through a mix of publicly owned (State of Montana-DNRC) lands and privately owned lands (Figure 26). A large portion of this reach borders the Blackfoot-Clearwater Wildlife Management Area. Most of the privately owned parcels along this river were agricultural (grazing and timber), but are experiencing rapid subdivision and residential development. A significant portion of the Clearwater River waterfront on DNRC holdings is also leased for private residential use, particularly along Elbow Lake. Stream gradients are very low, ranging from 62 ft/mile (0.11%) below Salmon Lake to near 0.0% near the wide, wetland dominated portions referred to as "lakes". Most of the Clearwater River Section I flows through moderately impacted riparian and terrestrial habitats. Channel alterations, removal of riparian vegetation, and water diversion exacerbate the effects of similar practices upstream. High levels of recreational use and annual manipulations to the shoreline and channel alter the natural integrity of the river corridor. For instance, riparian clearing, establishment of lawns adjacent to cabins, construction of artificial dams to enhance backwater areas for recreation, etc. are common. There is also a large irrigation diversion dam near the mouth they may partially impede upstream fish passage. Water Temperature Regime The water temperature regime in the Clearwater River Section I is much above the optimal range of temperatures for salmonids (see Figure 27). Native salmonids require cold water in which to live and reproduce. Non-native species, such as rainbow trout, brook trout and brown trout, are generally capable of succeeding in warmer waters, but the surface discharge from Salmon Lake make this river reach excessively warm in July and August. Because of these high maximum water temperatures, use by salmonids is presumably seasonal. Fish movement is likely extensive and frequent in this reach. This highlights the importance of connectivity for aquatic organisms as they migrate or seek thermal refuge in lakes or near colder tributaries. 30 Clearwater River Section 1 -2007 Daily Min., Max., and Avg. 4> Ay^ feX^ J" 4" ^^^^ J^ <3\> ^\- ^\- ^- 4 ^^ 4^ <# ^o^ jy Figure 27. Temperature data from Clearwater River section I (stream mile 3.2), 2007. 107 Fish Populations In 1995, four locations on the Clearwater River Section I were sampled to determine fish species composition (see Figure 28). These locations were at stream miles 0.1, 6.1, 9.0, and 9.6. These samples indicated that rainbow trout and brown trout were the dominant salmonid species during early summer. Overall, very little information on fish movement, composition or abundance is available for this reach. Efforts to collect these data will continue. 1 0.8 0.6 0.4 0.2 Proportion of Catch 1995 ii^L r"^^_ m ELL ■ RBT HRBxCT DWCT EilMWF 0.1 6.1 9.0 Location (stream mile) 9.6 Figure 28. Proportion of catch for salmonids at four locations on the Clearwater River Section I in 1995. SamniB Sitw Clearwater River Section II (Salmon Lake to Seeley Lake) Clearwater River Section II is the reach between the Seeley Lake outlet and the inlet of Salmon Lake. The actual outlet of Seeley Lake is often misrepresented, since the lake's outlet arm extends for more than 1.5 miles past the main lake body. The actual lake outlet (gradient break) occurs at Riverview Drive (T16N, R15W, section 3), just south of Seeley Lake. Clearwater River section II has two major tributaries: Owl Creek (outlet of Placid Lake) at stream mile 15.8 and Morrell Creek at stream mile 17.3. The Morrell Creek drainage is extremely important for native trout spawning and rearing, and is as Plvna T n^ J Bi flitf*.-Ciinlii*MTiflin ?tb^ I ^ f^o _ I -^iHlB'U^Fa U 15 16 IT ie 19 a 21 22 a Figure 29. Longitudinal profile for Clearwater River Section II. 108 described in an accompanying report section. Owl Creek is an intermittent stream reach due to Placid Lake management (dam at outlet) that connects the Placid Creek drainage with the Clearwater River. This stream does not appear to be utilized heavily by native fish, but does support a substantial kokanee migration fi-om Salmon Lake. Owl Creek and other streams in the Placid Creek drainage will be described in a subsequent report (Boles Creek is described above). The Clearwater River Section II is a vital stream segment for native salmonids in the Clearwater drainage, particularly migratory bull trout. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) refiect high use by migrating adfluvial bull trout, relatively abundant migratory WSCT, and its location linking adfluvial trout populations with spawning and nursery habitats. Numerous adfiuvial bull trout redds were also found in this section in 2006 and 2007. Recent bull trout telemetry studies demonstrated that the Clearwater River Section II is vital as a migration corridor between Salmon Lake and Morrell Creek, and between Seeley Lake and Morrell Creek. Morrell Creek supports one of the two largest adfiuvial bull trout populations in the upper Clark Fork Basin and represents one of four known adfiuvial spawning populations in the Clearwater Drainage based on redd count data and juvenile distribution. Morrell Creek is particularly important for the viability and persistence of the Seeley Lake and Salmon Lake bull trout populations, as it is currently the only spawning stream readily accessible to these fish. Adfiuvial westslope cutthroat and other fish species likely also use this river section extensively as they move between lakes and migrate at various life stages. Land Ownership and Habitat Conditions The Clearwater River Section II is approximately an equal mix of privately owned and publicly owned lands (State of Montana-DNRC). However, portions of the DNRC land is leased to private parties as residential waterfront property. Stream gradients range from 30 ft/mile (~ 0.6%) near Seeley Lake to about 6 ft/mile (~ 0. 1 %) near stream mile 19. The majority of this Clearwater River section consists of slow moving stream habitat with associated wetlands. The riparian corridor is largely intact, although there are some lessees and landowners that have recently cleared and manipulated the stream corridor. Stream encroachment, removal of riparian vegetation, and illegal water diversion are common in this section. In low water years (e.g., 2006-2007) this reach of stream was completely dewatered in areas as illegal pumping, unauthorized dams and upstream water usage exacerbated already minimal instream fiows and elevated water temperatures. In late summer and early fall, stream dewatering from Seeley Lake to Morrell Creek inhibits upstream and downstream movement offish, which is particularly problematic for migrating adult bull trout attempting to reach spawning areas (based on 2006-2007 telemetry data). Water Temperature Regime 109 Clearwater River Below Seeley Lake - DV Spawning Area 2007 Daily MIn., Max., and Avg. 3 oi a. E 01 # ^# ^<^^ ^^ ^^ ^<^ ^<^ *^ 4^^^ / # 4>^ <^^ 4^^ # c# ^o^ ,c^<^,c^^^ Figure 30. Temperature data from Clearwater River Section II (stream mile 19.2), 2007. The water temperature regime in the Clearwater River Section II is much above the optimal range of temperatures for native trout populations and other salmonids from the end of June through early September in low water years (Figures 30 & 31). Because this reach naturally receives surface discharge from Seeley Lake, fish populations have adapted over time and use this river corridor seasonally. For example, adult bull trout inhabit and move through this section primarily in June and early September to reach Morrell Creek when water temperatures are favorable. During warmer summer periods, native trout are most abundant in the cooler waters of tributaries and at depth in lakes (due to lake thermal stratification). Fish movement is likely extensive and frequent in this reach. This highlights the importance of connectivity for aquatic organisms as they migrate or seek thermal refuge in lakes or near colder tributaries. Clearwater River - Seeley Lake Outlet 2006 Dally Min., Max., and Avg. Figure 31. Temperature data from Clearwater River Section II (stream mile 22.9), 2006. Fish Populations There is likely high variability in the way various fish species use this reach of the Clearwater River. As a result, species composition and relative abundance likely changes seasonally. As mentioned previously, this reach is a primary migratory corridor between 110 Morrell Creek and Seeley/Salmon Lakes for migratory native trout. It also allows movement of many other species between lakes and throughout the lower Clearwater system. Sampling in 1995 at two locations (river mile 17.6 and 20.9) in late August indicated that brown trout and brook trout were common, along with documented mountain whitefish, sculpin and northern pikeminnow. Non-indigenous warm water fish introduced into Seeley Lake and Salmon Lake (e.g., yellow perch, northern pike, and largemouth bass) are likely also found in this section sporadically during warmer temperature periods. Expanded sampling efforts are planned for this reach during different temperature regimes to document fish community characteristics and the seasonal change in species composition. Oncorhynchus spp. genetic testing has not been completed in this reach due to the history of rainbow trout stocking and current WSCT stocking in lakes. It is very unlikely that indigenous, non-introgressed WSCT make up a significant component of the fish community. ilB 4.[Q < zf.' affi Einllif J &*-, ^.impla 5»M faLE2SHL£!lii~_ P^h^ 25 36 37 2S 2Si 3 Clearwater River Section III (Seeley Lake to Emily-A Dam) Clearwater River Section III is the reach between the inlet to Seeley Lake and the constructed fish barrier (Emily-A Dam) located just downstream of the mouth of the West Fork Clearwater River and Lake Inez. The fish barrier was constructed by FWP in 1964 to prevent upstream colonization by introduced fishes. Since construction, all introduced species found in the lower system have also been introduced upstream of the barrier. The only species not found upstream are northern pikeminnow and peamouth (both native fish). Restoring upstream fish passage at the dam is a restoration priority in the Clearwater Drainage as the dam continues to impede native fish. Clearwater River Section III has no major tributaries, although Benedict Creek enters just below the dam. Fisheries and aquatic information for this stream will be summarized in a subsequent report. The Clearwater River Section II is a vital stream segment for native salmonids in the Clearwater drainage, particularly migratory bull trout. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) reflect high use by migrating adfluvial bull trout, relatively abundant migratory WSCT, and its location linking adfluvial trout populations with spawning and nursery habitats in the West Fork Clearwater River (hereafter. West Fork). Recent bull trout telemetry studies demonstrated that the Clearwater River Section III is vital as a migration corridor between Seeley Lake and the West Fork Clearwater River. The West Fork supports one of the two largest adfluvial bull trout populations in the upper Clark Fork Basin and represents one of four known adfluvial spawning populations in the Clearwater Drainage based on redd count data and juvenile distribution. The West Fork is Figure 32. Longitudinal profile for Clearwater River Section III. Ill particularly important for the viability and persistence of the Seeley Lake, Lake Inez, and Lake Alva bull trout populations. It is not known if emigrants from the West Fork also inhabit Salmon Lake. Adfluvial westslope cutthroat and other fish species likely also use this river section extensively as they attempt to move between lakes and migrate at various life stages. Land Ownership and Habitat Conditions The Clearwater River Section III flows through public (USPS) and small private lands parallel to State Highway 83 (Figure 32). Stream gradients are very low, averaging just 6 ft/mile (0.01%) in this reach. The majority of the Clearwater River Section III consists of functional, slightly degraded aquatic, riparian and terrestrial habitats. Along this main stem, there are limited examples of stream encroachment and removal of riparian vegetation. Most of the reach contains low gradient stream habitat with associated wetlands, debris jams and abundant beaver activity. The Emily-A dam, at the upstream boundary of the section, poses the greatest human disturbance and threat to river function and stream connectivity. Water Temperature Regime 25 1 Clearwater River - Seeley Lake Inlet 2006 Daily Min., Max., and Avg. # 4^ ^^^4^"^^ ^^^^>^V^^^ Figure 33. Temperature data from Clearwater River Section III upstream of Seeley Lake (stream mile 28.6), 2006. 112 Clearwater River at Bnily-A Dam - 2007 Daily Min., Max., and Avg. 01 a. E 30 25 20 15 10 t& (9 # 4" ^^ ^^ # ^N<^ ^ ^ kO^ X<3 .C^" Figure 34. Temperature data from Clearwater River Section III at Emily-A Dam (stream mile 29.3), 2007. The water temperature regime in the Clearwater River Section III is much above the optimal range of temperatures for salmonids during summer months, primarily due to surface discharge from Lake Inez (see Figures 33 and 34). As a result, fish species composition in the reach changes seasonally. This was observed through repeated sampling at Emily-A Dam in 2007. Bull trout and WSCT become stressed when subjected to maximum daily temperatures above 15-18° C, which typically occurs in late June through August. During these months, salmonids must seek refuge in colder tributaries or lakes to escape lethal temperature extremes. Recent investigations have shown that adult adfluvial bull trout will use the Clearwater River Section III as a migration route, even when temperatures are above optimal. However, these behaviors may be unnatural as salmonid migrations are interrupted by the dam and fish remain in the section as they try to ascend past the obstruction. Road Crossings The Clearwater River Section III has two road crossings. They are both bridges that do not inhibit upstream fish movement or natural stream function. Fish Populations 113 1 _, Proportion of Catch 1995 ■ DV 0EBT HLL ■ RBT BRBxCT DWCT n A 0.2 - n 28.5 29.3 Location (stream mile) Figure 35. Proportion of catch for salmonids at two locations on Clearwater River Section III in 1995. In 1995, two locations on the Clearwater River Section III were sampled at stream miles 28.5 and 29.3 in July to determine fish species composition (see Figure 35). These samples indicated that brown trout and brook trout comprised a large proportion of the salmonid composition in this section. Given the warm water temperatures in July and higher tolerance of introduced trout species, this is not unexpected. Multiple species of native, non- salmonid fish were also documented including northern pikeminnow, redside shiner, peamouth, longnose sucker, etc. There is likely high variability in the way various fish species use this reach of the Clearwater River. As a result, species composition and relative abundance changes seasonally based on sampling in 2007 at Emily-A Dam. As mentioned previously, this reach would be a primary migratory corridor between the West Fork and Seeley/Salmon Lakes for migratory native trout. It would also allow movement of many other species between lakes and throughout the lower Clearwater system. This reach supports abundant spawning by kokanee migrating from Seeley Lake. Non-indigenous warm water fish introduced into Seeley Lake (e.g., yellow perch, northern pike, and largemouth bass) are likely also found in this section sporadically during warmer temperature periods. Expanded sampling efforts are planned for this reach during different temperature regimes to document fish community characteristics and the seasonal change in species composition. Clearwater River Section IV (Emily-A Dam to Rainy Lake outlet) 114 Clearwater River Section IV is the reach between the Emily-A Dam and the Rainy Lake fish barrier just downstream of the Rainy Lake outlet. It encompasses both Lake Alva and Lake Inez. This river reach includes the mouths of the West Fork Clearwater River and Camp Creek, as well as several smaller tributaries including Colt, Uhler, and Richmond Creeks. The West Fork is described above, but other tributary attributes will be summarized in subsequent reports. Clearwater River Section IV is a very important stream for native salmonids in the Clearwater drainage, particularly as a migration corridor for migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see Stream Prioritization section) reflect high use by migrating adfluvial bull trout, relatively abundant fluvial WSCT populations, and high habitat quality. This main stem reach is particularly important as a migration route for migratory native salmonids, as well as other sport fish populations. Clearwater River Section IV is an important migration corridor between Lake Inez, Lake Alva, and the West Fork of the Clearwater River, a key adfluvial bull trout spawning stream as documented through radio telemetry studies in 2007-2008. The West Fork of the Clearwater River supports one of the two largest adfluvial bull trout populations in the upper Clark Fork Basin and represents one of four known adfluvial spawning f 412 JOS A.(H- 3.3E Emll}f J tti / / /. LjkiiMj UifLfilm e.illTr^Mlgr. BrMtTlMLWCr^BT PBIt— Ml^lltJ- I WCT pm^jjH I ijsf^ I fiw-m USfS 29 3D 31 33 3:^ 33 37 3S 33 Figure 36. Longitudinal profile for Clearwater River Section IV. populations in the Clearwater Drainage. The West Fork of the Clearwater River is particularly important for the viability and persistence of the Lake Inez and Lake Alva bull trout populations, as it is currently the only spawning stream readily accessible to these fish. Recent studies have shown that some adult adfluvial bull trout from Rainy Lake also move downstream into this section of river. The Rainy Lake fish barrier prevents adult and juvenile fish from returning to Rainy Lake. Investigations are ongoing in order to determine the most prudent solution to this problem as the dam also prevents upstream colonization by several undesirable nonnative fish including northern pike and brook stickleback. A seasonal fish obstruction is also present at the outlet of Lake Inez as the lake is impounded each summer to enhance recreation. Installation of a fish ladder is planned for this site in 2008. Land Ownership and Habitat Conditions The Clearwater River Section IV flows almost entirely through publicly owned (USFS) lands and a small amount of privately owned lands. Most of the privately owned land along this section of river is adjacent to Lake Inez or just downstream. Stream gradients are very low, ranging from 33 ft/mile (0.06%) near the Rainy Lake fish barrier to near ft/mile (0.0%) adjacent to the lakes. 115 Nearly all of Clearwater River Section IV flows through functional, non-degraded riparian and terrestrial habitats. The channel is generally a low gradient corridor with associated wetlands and abundant beaver activity where wider floodplains exist. Water Temperature Regime The water temperature regime in this section is well above the optimal range of temperatures for native salmonids during July and August (see Figure 37). Native salmonids, particularly bull trout, require cold water in which to live and reproduce. Non-native species, such as brook trout, brown trout and other non-salmonids, are generally more tolerant of warmer temperatures. Bull trout become stressed when subjected to maximum daily temperatures > 15° C (-18° C for WSCT), which occurs for a significant portion of most summers in this reach. During these months, native salmonids must seek refuge in colder tributaries in order to escape lethal temperature extremes. Recent investigations have shown that adfluvial bull trout will use the Clearwater River Section IV as a migration route, even when temperatures are above optimal. However, these fish attempt to minimize the time they spend in the warm river and spend the majority of their time in deep water in Lake Alva or Lake Inez. a. 30 25 20 15 10 Clearwater River Below Lake Alva - 2007 Daily Min., Max., and Avg. Thermograph Malfunction r^^ — OOLO — OOLOOMCnCDCO T^ T- Osl 2? LO LO (D (D (D T^ Osl Osl 15 fi fi 00 r^^ — r^^T-ooLOOMO) ^ ^ Osl CO 65 65 00 00 T^ Osl Osl en en en Figure 37. Temperature data from the Clearwater River Section 4 downstream of Lake Alva (stream mile 35.4), 2007. Road Crossings Clearwater River Section IV has two road crossings. Both crossings are bridges that do not impede upstream fish passage or natural stream function. Fish Populations 116 Proportion of Catch 35.4 Loci 3.8 3.6 Rainy Lake Bull, Cutthroat Trout Cutthroat, Brook Trout Perennial Stream USPS 38 39 40 41 42 Stresm MIe 43 44 44.6 Figure 38. Proportion of catch fo Section IV in 1995. Figure 39. Longitudinal profile for Clearwater River Section V. In July 1995, three locations at stream miles 35.4, 35.5, and 38.7 on Clearwater River Section IV were sampled to determine fish species composition (see Figure 38). These samples indicated that brook trout comprised a large majority of the fish in the portion of river below Lake Alva, while WSCT comprised a large majority of fish above Lake Inez. It is likely that a brook trout stronghold that exists in Uhler Creek, which enters the Clearwater River Section IV at stream mile 35.0 just downstream of Lake Alva, is contributing significantly to the brook trout population there. The few WSCT detected were not tested for genetic purity as hybridization is likely common (due to past rainbow trout stocking) and many of the non-introgressed fish were likely of hatchery origin (stocked in lakes). Presence of sculpins was not noted in the 1995 sampling, but long-nosed dace were common. The age of available data for the Clearwater River Section IV and the variable seasonal species composition in this reach necessitates additional sampling, which is planned for 2008. Amphibian Community In 1995 sampling, no amphibians were noted. However, Columbia spotted frogs have been observed in these sections of the Clearwater River. More rigorous investigations are warranted. Clearwater River Section V (Rainy Lake to Clearwater Lake) The upper Clearwater River is a third-order that extends from the outlet of Clearwater Lake to the Outlet of Rainy Lake, where a man-made dam prevents upstream fish passage. This reach generally flows south from its headwaters near Ptarmigan Mountain and the Swan 117 River divide. Tributaries of this reach (described separately) include the East Fork of the Clearwater River (hereafter "East Fork") and Bertha Creek. This section of the Clearwater River is a very important reach for native salmonids in the Clearwater drainage, particularly migratory life history forms. High biological ranking and prioritization of this watershed relative to other Clearwater Drainage streams (see stream prioritization section) reflect high use by migrating adfluvial bull trout, relatively abundant adfluvial WSCT populations, and high habitat quality. The lower portion of the Clearwater River Section V is an important migration corridor between Rainy Lake and the East Fork, which is documented bull trout spawning habitat. The East Fork supports one of four known adfluvial spawning populations in the Clearwater Drainage and likely also supports a stream-resident population. The East Fork is particularly important for the viability and persistence of the Rainy Lake bull trout population, as it is likely the only spawning reach accessible to these fish. Recent radio-telemetry studies have shown that adult adfluvial bull trout from Rainy Lake move upstream into this section of river via the Clearwater River. These studies have also shown that adult adfluvial bull trout from Lake Alva (and likely from Lake Inez) occasionally attempt to migrate into the East Fork of the Clearwater River, but are hindered by the Rainy Lake fish barrier, immediately downstream of this section. Land Ownership and Habitat Conditions The Clearwater River Section V flows entirely through publicly owned (USES) lands (Figure 39). Stream gradients are very low, ranging from 126 ft/mile (2.4%) near Clearwater Lake to 80 ft/mile (1.5%) near Rainy Lake. The Clearwater River Section V consists of functional, non-degraded aquatic, riparian and terrestrial habitats. The drainage is intact forest, with modest road densities and almost no logging. Riparian areas, water quality and channel morphology are largely intact, providing shade, instream habitat complexity, consistent recruitment of woody debris and adequate stream buffers along its length. Water Temperature Regime The water temperature regime in the portion of the Clearwater River Section V that is below the mouth of the East Fork of the Clearwater River is lower than average for the Clearwater River drainage, likely remaining suitable year-round for salmonid species (see Figure 40 & 41). However, upstream of the confluence with the East Fork, temperatures are frequently above optimal for native salmonids in July and August as this is the outlet of Clearwater Lake. Bull trout become stressed when subjected to maximum daily temperatures >15° C (-18° C for WSCT), which occurs for a significant portion of most summers. During these months, native salmonids must seek refuge in colder tributaries or lake environments in order to escape temperature extremes. Nonnative trout (i.e., brook trout in this reach) are more tolerant of warmer water temperatures. 118 Clearwater River Above Rainy Lake - 2007 Daily Min., Max., and Avg. -- CN OM 00 in CN O) (D CO T- CN r^ r^ r^ r^ Figure 40. Temperature data from the Clearwater River section V, downstream of the East Fork of the Clearwater River, 2007. Clearwater River Below Clearwater Lake outlet - 2007 Daily Min., Max., Avg. -- CM CM (D (D (D r^ r^ r^ r^ ^ ,— on ^ ,— oo ID rsi rsi oi rsi 00 00 00 O) O) a> Figure 41. Temperature data from the Clearwater River section V, upstream of the East Fork of the Clearwater River, 2007. Road Crossings The Clearwater River Section V has two road crossings. Both of these are bridges and likely do not impeded upstream fish movement or natural channel function. Fish Populations 119 •I Proportion of Catch 0.8 - 0.6 - 0.4 - 0.2 ^ - - - HDV 0EBT HRBxCT DWCT — - H= — ii ^ ■i! ■M ■i ■i P r r ■ 38.8 39.3 39.8 39.9 40.5 40.7 42.3 43.2 Location (stream mile) Figure 42. Proportion of catch for salmonids at eight locations on Clearwater River Section V, 1995-2007. Since 1995, eight locations on the Clearwater River Section V have been sampled to determine fish species composition and Oncorhynchus genetic composition (see figure 4). These locations were at stream miles 38.8, 39.3, 39.8, 39.9, 40.5, 40.7, 42.3, and 43.2. These samples indicated WSCT and rainbow-westslope cutthroat hybrid trout dominate the section. The area near Rainy Lake appears to have the highest proportion of hybrid fish, likely as a result of past rainbow trout stocking. Bull trout were present downstream of and very near the mouth of the East Fork of the Clearwater River. It is likely that bull trout only use this habitat seasonally as high water temperatures may be limiting during summer. Brook trout have been found in the upper-most sites. This is the result of an illegally introduced population in Clearwater Lake. No sculpin have been noted in this section of the Clearwater River. Amphibian Community Sampling conducted in 2006 indicated the presence of tailed frogs and Colombia spotted frogs in this section of river. 120 Whirling Disease RESULTS PART V: Special Studies Results Part V is a whirling disease special study section with five related studies, all of which variously rely on sentinel exposures of fish (rainbow trout) in waters of the Blackfoot Basin. We begin with a brief introduction to the pathogen (worm and fish hosts) and summarize the current status of whirling disease within the Blackfoot Basin. Individual studies then begin with a pilot-level investigations focusing on biotic relationships of the pathogen to "indicator" aquatic benthic organisms (e.g. stoneflies, caddis and mayflies). Our studies then expand to environmental predictors of disease as measured by both biotic (benthic) and physical characteristics of basin-fed tributaries of the Blackfoot River. Lastly, we examine potential effects (and implications) of whirling disease of two susceptible species (rainbow trout and mountain whitefish) that inhabit the Blackfoot River. Figure 1. Generalized distribution of M. cerebralis infected waters in the Blackfoot Watershed as identified by sentinel cage exposures of RBT. Introduction Whirling disease, caused by the exotic Mean grade category Infection Level Description myxosporean parasite Myxoboliis cerebralis, was first detected in the Blackfoot River in 1995 near Ovando. Since then, the disease has increased in both distribution severity (Figure 1). The disease is now present throughout the entire mainstem Blackfoot River and continues to expand in the lower reaches of some tributaries. Despite this expansion, sentinel exposures 0.0-2.0 2.01-2.74 2.75-3.7 3.71-5.0 Low Medium High Very l-ligli and Table 1. Mean grade category descriptions (Baldwin et al. 2000). Com m on Name Susceptibility Rainbow Trout 3 Westslope Cuttiiroat 2 Brool< Trout 2 Bull Trout 1 Brown Trout 1 Mountain Whitefish 2S undertaken in 2006 and 2007 also suggest a recent reduction in severity of disease in the lower Blackfoot River (Table 3). The low-elevation distribution of the disease currently overlaps with the distribution of many salmonids. Myxobolus cerebralis has a complex, two- host life-cycle involving a salmonid and the aquatic oligochaete worm, Tiibifex tiibifex. There are also two spore forms of the parasite; a fragile triactinomyxon (TAM) that is released by the worm and infects young trout and a hardy myxospore later released by infected fish and ingested by the worm host, where the myxospore is then converted back to the TAM stage. The development and severity of whirling disease in exposed salmonids is dependent on many factors involving: 1) the fish host (species, strain, age, size); 2) the worm host; 3) the environment 121 Table 2. Susceptibility to whirling disease among species of salmonids in the Blackfoot River. Scale of to 3 or S: = resistant; 1= partial resistance; 2 = susceptible; 3 = highly susceptible; S = susceptibility is unclear (conflicting reports). Adapted from MacConnell and Vincent (2002). (water quality parameters, water temperature, flow rates); and 4) the overlap of contact with both spore types (overlap of TAM with susceptible fry species and myxospore being encountered by the worm). These variables all have potential to influence infection and severity among salmonids across the Blackfoot Basin. As an indirect measure of TAM abundance and disease severity, sentinel cages were first deployed in the Blackfoot Watershed in 1998. Sentinel cage monitoring has continued through 2007 at established Blackfoot River sites and throughout tributaries in order to assess disease expansion. A mean grade infection is determined from histology results from sentinel fish exposed in each cage to determine disease severity at individual locations (Table 1). Concurrent with the recent escalation of the disease was an increase in the clinical signs (cranial and skeletal deformities) through 2004, and modest population declines in rainbow trout in the middle Blackfoot River downstream of Monture Creek, a highly infected rainbow trout spawning stream. Previous studies have classified salmonids based on susceptibility to the disease, which varies considerably by species (Table 2). All salmonids in the Blackfoot Watershed (WSCT, bull trout, rainbow trout, brown trout, brook trout and mountain whitefish) can be infected by the parasite. Rainbow trout are reported to be the most susceptible and brown trout and bull trout more resistant. The susceptibility of mountain whitefish is unclear but remains a concern and a focus of investigations in the Blackfoot Basin. Waterbody IVIean Grade Infection Blackfoot River 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Blackfoot River-Below Gold Cr 0.22 nd 2.44 nd 0.59 2.42 2.2 2.06 nd Blackfoot River-Below Elk Cr nd nd 2.3 nd 1.59 nd 2.3 nd 0.64 0.22 Blackfoot River-above Clearwater 1.1 0.22 3.11 nd 2.79 3.16 3.41 2.96 2.03 1.33 Blackfoot River-Below North Fork 0.25 nd nd nd nd nd 2.64 2.86 0.79 nd Blackfoot River-below Nevada Cr 0.84 nd 0.9 2.12 3.93 3.28 0.1 0.31 Blackfoot River-Below Lincoln 0.6 nd 2.44 nd nd 3.89 2.25 nd Blackfoot River-Headwaters nd nd nd 0.02 0.32 nd 0.07 Basin-fed Streams Johnson Creek nd nd nd nd nd nd nd nd West Twin Creek nd nd nd nd nd nd nd East Twin Creek nd nd nd nd nd nd nd nd Bear Creek nd nd nd nd nd nd nd na nd Union Creek nd nd nd nd nd nd nd nd nd Gold Creek nd 0.12 nd nd Belmont Creek nd nd nd 0.19 0.38 1.55 2.48 0.3 3.44 Elk Creek nd nd 2.84 4.32 4.82 nd nd Clearwater River nd nd nd nd nd nd nd nd nd CottonwoodCreek 3.66 4.52 nd nd 4.5 nd nd 3.78 3.96 4.25 Chamberlain Creek 0.16 2.71 3.88 nd 2.63 nd 4.33 3.78 nd 1.89 Monture Creek 1.76 nd 3.22 nd nd 4.81 4.57 4.26 Warren Creek 0.21 2.1 1.72 nd nd nd nd 0.0 nd nd North Fork Blackfoot River nd nd 0.78 nd nd 0.27 nd nd Arrastra Creek nd nd nd nd nd 0.34 1.23 0.02 0.14 nd Beaver Creek nd nd nd nd nd nd 0.45 0.85 0.3 Poorman Creek nd nd nd nd nd nd 0.78 ND nd 4.69 Landers Fork nd nd nd nd nd nd 0.14 Upper Willow Creek nd nd nd nd nd nd nd nd Wasson Creek nd nd nd nd nd nd nd nd Spring Creeks Jacobsen Spring Creek nd nd nd nd nd nd 0.13 nd nd nd Rock Creek nd 2.3 3.9 nd 3.38 nd nd nd nd Kleinschmidt Creek 2.83 3.56 4.52 3.77 nd 4.9 4.7 nd nd nd Nevada Spring Creek nd nd nd nd nd 3.66 2.22 1.94 nd Grentier Spring Creek nd nd nd nd nd nd 0.06 1 nd nd Lincoln Spring Creek nd nd nd nd nd nd 5 4.7 nd nd Table 3. Summary of histological results summarized as mean grade infections from sentinel cages placed in the Blackfoot River (top), the confiuence areas of basin-fed tributaries (middle) and spring creeks (lower) for 1998-2007. Pilot assessment of the association between stonefly assemblages and the incidence and severity of whirling disease in tributaries of the Blackfoot River, Montana Wease Bollman, Ron Pierce, and Lisa Eby Introduction This brief study was intended to investigate whether stonefly assemblages could be useful bellwethers of the presence or severity of whirling disease in tributary streams to the Blackfoot River. At least one previous study (Bollman 1998) demonstrated associations between stonefly richness and certain observational measures related to reach-scale habitat integrity; these measures included streambank stability, condition of the riparian zone, and stream channel morphological elements. Loss of streambank stability, riparian zone integrity, and natural channel morphology may contribute to instream conditions favoring the presence of Tubifex tubifex, the oligochaete intermediate host for the whirling disease organism (Myxobolus cerebralis). Thus, we hypothesize that metrics describing stonefly assemblages may be useful in predicting the presence and severity of whirling disease. Methods Benthic invertebrates were sampled on July 26-27, 2006 from single riffles in each of 13 tributary streams of the Blackfoot River. A D-frame net with 1000 micron mesh was used. Substrates were disturbed by kicking along transects; sampling effort was timed and distance approximated by stepping off Table 1 lists sampling sites, the time expended for each sample, and the approximate distance over which substrates were disturbed. Samples were preserved in 95% ethanol at streamside, and delivered to Rhithron Associates in Missoula for sorting and identification of organisms. In the laboratory, samples were sorted under dissecting stereoscopes, using lOx - 30x magnification. A random selection of 500 organisms was taken from each sample; stoneflies collected in these subsamples were separated from the remaining organisms and preserved. All stoneflies remaining in each sample were then collected, and these were preserved separately. Stoneflies from both the random subsample and the total sample collection were identified using published keys; specimens were identified to the lowest taxonomic level possible considering the maturity of the animals and the availability of appropriate keys. Generally, at least genus level was achieved; in many cases, species could be determined. Early- instar capniids were left at family level. Two samples yielded a total of 6 extremely immature specimens. These were identified to family level, but were not considered in the subsequent analysis since it was not possible to determine whether they represented unique taxa or were early instars of taxa already included in the taxa lists. No further analysis of the 500 organism random subsamples was performed, other than the inclusion of stoneflies from those subsamples in the present exploration. All sample fractions were preserved and retained at Rhithron for possible further analysis. Physical and chemical data as well as data related to incidence (percent of reaches with infection >3 in 2005) and severity (mean MacConnell-Baldwin scale value in 2005) of whirling disease in each stream were collected and compiled by Montana Fish Wildlife and Parks personnel. Correlation matrices (Spearman rank R) were constructed using these data and stonefly data, and these matrices were examined for suggestive associations. In all, 18 metric expressions summarizing the stonefly data were analyzed for correlation with the 2 whirling disease measures. Data from Bear Creek was deleted from the data set, since the status of whirling disease in that stream in 2005 was not known at the time of this study. 123 Results Nineteen stonefly taxa in 7 families were present in the 13 samples. A total of 1348 stoneflies were identified. Figures 1 and 2 illustrate correlation between stonefly taxa richness and 2 measures of whirling disease incidence and severity. Correlations were not significant, but Chamberlain Creek clearly presents as an outlier in these analyses. When Chamberlain Creek was removed from the dataset, correlation between stonefly taxa richness and measures of whirling disease severity and incidence were significant (R=-0.711 143, p < 0.05 and R=-0. 736352,/? < 0.05). Three functional feeding groups were represented in the stonefly collection from the 13 samples: shredders were represented by 8 taxa, predators by 10 taxa, and collectors by 1 tax on. There was a significant association of predatory taxa richness with infection severity (R= - 0.616833,/' < 0.05), but not with infection incidence. This relationship is illustrated in Figure 3. Other significant correlative relationships were demonstrated between whirling disease severity and sensitive stonefly taxa richness and abundance, and between sensitive taxa richness and richness within the family Chloroperlidae and whirling disease incidence. However, neither sensitive taxa nor Chloroperlid taxa were well-distributed among these sites. No other significant associations could be demonstrated between either measure of whirling disease and measures of richness, relative abundance, or absolute abundance of various stonefly families, functional groups, or tolerance characteristics. Several taxa were collected only in streams with no incidence (i.e. 0% of reaches with >3 on the MacConnell-Baldwin scale) of whirling disease. These were the nemourids Visoka cataractae (collected from 2 sites) and Zapada oregonensis (one site), the perlid Calineuria californica (2 sites), the taeniopterygid Taeniopteryx sp. (one site), and perlodids Isoperla sp. (one site), Kogotus sp. (2 sites), and Megarcys sp. (4 sites). It should be noted that Kogotus sp. was collected from Arrasta Creek, which had a low mean severity rating (0.02). The low severity rating despite 0% of reaches severely infected suggests that whirling disease probably is present though not widespread in Arrasta Creek. Two taxa were collected only from streams with whirling disease infection: Pteronarcella badia (2 sites) and Claassenia sabulosa (2 sites). 124 Tables and Figures Sampling date Waterbody Time expended (min: sees) Distance Description of effort 7/27/2006 Gold Creek 6:20 36 feet Single transect (riffle) 7/27/2006 Bear Creek 6:20 36 feet Single diagonal transect (riffle) 7/27/2006 W. Twin Creek 6:15 36 feet Triple diagonal transect (riffle) 7/27/2006 E. Twin Creek 6:20 36 feet Triple diagonal transect (riffle) 7/26/2006 Cottonwood Creek 5:00 36 feet Single diagonal transect riffle 7/26/2006 Monture Creek 5:30 36 feet Single transect (riffle) 7/26/2006 Chamberlain Creek 8:10 36 feet Triple transect (riffle) 7/26/2006 Arrastra Creek 6:40 36 feet Triple diagonal transect (riffle) 7/27/2006 Belmont Creek 6:04 28 feet Single transect (riffle) 7/26/2006 Blanchard Creek 7:15 27 feet Double transect (riffle) 7/26/2006 Landers Fork 6:20 36 feet Single transect (riffle) 7/27/2006 Johnson Creek 6:00 36 feet Double diagonal transect (riffle) 7/26/2006 Elk Creek 6:30 36 feet Triple diagonal transect (riffle) Table 1. Summary of sampling events: July 2005. 10 -1 12 3 4 2005 Mea nseverity (MacCon nell-Baldwin scale) Figure 1. Association between stonefly taxa richness and severity of whirling disease {R=-0A6,p>0.05). 125 10 (!> 7 ^ 3lanchard Johnson Ov- EAST TWIN CREEK Landers Fork WestTwin Arrasta GOLD CREEK -20 CHAU BERLAIN CREEK fl- BELUONT CREEK JLONTURE CREEK ELK CREEK COTTONWOOD CREEK "^ V 20 40 60 80 100 %ofreach es with mean infectio n severity >3 (ft/bcC onnell-Baldwin sea le) in 2005 120 Figure 2. Association between stonefly taxa richness and incidence of whirling disease. (R= 0.51,/? > 0.05). m T3 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 BLANCHASD CREEK — — LANDERS FORK East Twin Gold MONTURE CREEK ---- -o- ---- BELIIONT CREEJ COTTONW OOD CREEK ELK CREEK 12 3 4 2005 Mean severity (MacConnell-Baldwin scale) Figure 3. Association between predatory stonefly taxa richness and severity of whirling disease. (R=-0.51,/» > 0.05). Chamberlain Creek was not included in this analysis. 126 Discussion Although significant trends could be demonstrated, none of the correlative relationships explored in this study gave results definitive enough to support the hypothesis that characteristics of stonefly assemblages can predict the incidence or severity of whirling disease as measured here. In each analysis, there is considerable overlap of results between infected and uninfected streams. It seemed justifiable to delete Chamberlain Creek from the correlation analyses, since there are some unique conditions in that watershed immediately upstream of the sampling location. As noted, this site presents as an outlier; despite high incidence of infection in this stream, the sample collected here yielded the highest stonefly taxa richness of any sample in this study. Historic reconstruction that reclaimed an altered channel may have resulted in ideal habitat conditions for stoneflies. Reconstructed reaches were located immediately upstream of the sampling location as well as two artificial upstream ponds that drain into Chamberlain Creek in the immediate upstream area. These ponds are thought to harbor t. tubifex "hotspof conditions such as warmer water effluent and organic sediments used to line the ponds. Since there were some taxa that may have been confined to uninfected streams, further investigation of potential indicator taxa in the family Perlodidae may be promising. Further, the taxa that were identified in this study as potential indicators will be relatively easy to identify streamside, if care is taken to sample at appropriate times of year. 127 Exploratory assessment of association between invertebrate "EPT" taxa and the incidence and severity of whirling disease in tributaries of the Blackfoot River Wease Bollman, Ron Pierce, and Lisa Eby Indroduction In an earlier pilot study, we investigated whether patterns in stonefly (Plecoptera) assemblages could suggest the presence of whirling disease in tributary streams to the Blackfoot River. In this study, we include 2 other aquatic insect orders, mayflies (Ephemeroptera) and caddisflies (Trichoptera). These 3 groups comprise the so-called "EPT" orders of aquatic insects, which are often considered to be general indicators of clean water and undisturbed habitat conditions. Methods Sampling methods and laboratory processing and identification protocols are described in the previous paper (Bollman et al. 2006), to which the reader is referred. Random subsamples of 500 organisms from all aquatic invertebrate groups were taken from whole samples, and these were identified. Physical and chemical data, as well as data related to the incidence (percent of reaches with infection >3 in 2005) and severity (mean McConnell-Baldwin scale value in 2005) of whirling disease in each stream were made available by Montana Fish Wildlife and Parks. Correlation matrices (Spearman rank R) were constructed using these data and the invertebrate data, and the matrices were examined for suggestive associations. In all, 22 metric expressions (listed in Table 1) summarizing the invertebrate data were analyzed for correlation with the 2 whirling disease measures. Ephemeroptera and Trichoptera taxa from both the random subsamples and the total sample collections were removed and identified, and these data were combined with the Plecoptera data generated for the earlier study. The relative abundance of each EPT taxon was examined for correlative association with whirling disease measures. As before. Bear Creek was not included in this analysis, since the status of whirling disease in that stream in 2005 was not known. The EPT fractions of the whole samples were further analyzed with an ordination study (nonmetric multidimensional scaling: McCune and Grace 2002). For this analysis, the Sorenson (Bray-Curtis) distance measure was used. The resulting plot was examined to see if groupings of invertebrate assemblages distinguished infected sites from uninfected sites. Bear Creek was included in this analysis. Results Sixty-seven EPT taxa were identified in the 13 samples. A total of 10,644 EPT individuals were present. Significant correlation could be demonstrated between both measures of whirling disease and 4 metric expressions summarizing the invertebrate assemblages. Table 2 summarizes the correlation coefficients of these relationships. Figures 1-4 graph the results of each metric individually against the percentage of stream reaches with infection severity greater than 3, as measured on the MacConnell-Baldwin scale. Several taxa were collected only in streams with no incidence (i.e. 0% of reaches with >3 on the MacConnell-Baldwin scale) of whirling disease. A few taxa were only collected from streams with whirling disease infection. These data are summarized in Tables 3 and 4. Taxa that were collected at a single site were not included in the tables. Figure 5 illustrates the ordination plot of the aquatic invertebrate assemblages collected at the 13 sites. Final stress for this analysis was 7.02, indicating a good fit of the ordination model to the data. 128 Table 1. Metrics tested for association with incidence and severity of whirling disease. Measures of habitus, physiology, or life history Air Breather Richness Burrower Richness dinger Richness Cold Stenotherm Richness Hemoglobin Bearer Richness Semivoltine Richness Swimmer Richness Univoltine Richness Tolerance measures Hilsenhoff Biotic Index Metals Tolerance Index Pollution Sensitive Richness Sediment Sensitive Richness Sediment Tolerant Richness Functional measures Filterer Richness Predator Richness Taxonomic composition measures Baetidae/Ephemeroptera E Richness EPT Richness Hydropsychidae/Trichoptera P Richness T Richness Taxa Richness Table 2. Spearman rank order correlation coefficients (R) for associations between 2 measures of whirling disease and 4 metric expressions derived from invertebrate taxonomic data from 13 samples. Correlations are significant at p < .05000. 2005 mean infection % reaches > grade 3 infected 2005 mean infection 1.000000 0.913293 % reaches > grade 3 infected 0.913293 1.000000 Cold Stenothei-m Richness -0.673981 -0.710569 EPT Richness -0.600712 -0.608616 Metals Tolerance Index 0.744218 0.842184 Pollution Sensitive Richness -0.691541 -0.745014 129 Figure 1. Association between cold stenotherm taxa richness and incidence of whirling disease (R = -0.71, p < 0.05). Dotted lines indicate 95% confidence intervals. 14 10 c .c o E 8 E ta ;□ o o EAST TWIN CREEK WfeSTTWiN CREEK ^-::P: JOHNSON CREEK o ""BTJWCHAftO CREEK GOLD CREEK ■■■ *■ ■■■■ - -.BELMONT CREEK C H AylB ER L^ NT R EEK- . ,. J. CtHKiHBHEBD Cf Wi. !P' 60. 60 % > grade 3 infected 100 120 Figure 2. Association between EPT richness and incidence of whirling disease (R = -0.61, p < 0.05). Dotted lines indicate 95% confidence intervals. 36, 34 S 32 ID C ^ 30 Q. Lu 28 BLANCHARD CREEK 26- 24- EASTTWm CREEK . o :-v, ^LjSWOlBRCHemK o ARRASTA CREEK ' SifBtfffMr WtopraK ^:-- f ao-^ m IVlONTU RE CREEK Q- ; i CH*IBERLA1N ; t>- JH -COTTONWifSOD CREEK .■■^? 40 60 % > grade 3 infected 80 100 120 130 Figure 3. Association between the metals tolerance index and incidence of whirling disease (R = 0.84,/? < 0.05). Dotted lines indicate 95% confidence intervals. 3.8 3.6 3-4 3.2 3.0 2.8 2.6 2.4 2:2 2.0 iyS- 1.6 1.4 1.2 -20 L'.M [) EFS FO F K f-.P P.-STO. CREEK JO H HO(/il C I! EEK o jClUiltd'EFlLa1il.CLH£EIS:.. 20 40 60 80 % > grade 3 infected 100 120 Figure 4. Association between pollution sensitive taxa richness and the incidence of whirling disease (R = -0.74, p < 0.05). Dotted lines indicate 95% confidence intervals. 14 12'- ii EAST TWIN CREEK o WBSfimirftKIH^HK -20 20 40 60 •%>igj=ade 3 infected 80 100 120 131 Table 3. Taxa collected only from streams without documented whirling disease infection. The number of sites where the taxon was collected is indicated in parentheses. Taxa collected at only a single site are not included in the table. Ephemeroptera Drunella spinifera (2) Caudatella edmundsi (2) Epeorus deceptivus (2) Epeoriis longimanus (5) Rhithrogena sp. (4) Ironodes sp. (2) Baetis alius (2) Baetisflavistriga (2) Plecoptera Visoka cataractae (2) Calineuria califomica (2) Megarcys signata (4) Trichoptera Parapsyche elsis (4) Agraylea sp. (2) Lepidostoma unicolor (4) Rhyacophila alberta (3) Table 4. Taxa collected only from streams with documented whirling disease infection. The number of sites where the taxon was collected is indicated in parentheses. Taxa collected at only a single site are not included in the table. Ephemeroptera Timpanoga hecuba (3) Plecoptera Pteronarcella badia (2) Claassenia sabulosa (2) Trichoptera Brachycentrus occidentalis (3) Hydropsyche oslari (2) Figure 5. Ordination (nonmetric multidimensional scaling) of aquatic invertebrate assemblages from 13 stream sites in the Blackfoot drainage. v-a.:. MOHTUpE_ ELK o o ; BLiiNCHAR' o' i ' jCH/*^BER E O BEAR . la: COTTONW o : ..BELWaWT- o ■ LflNDEPS O : WTWIM, o. i JOHNSON : ARRASTA ETWIN o '■■■-■--t- Discussion The significant association of EPT taxa richness with whirling disease incidence and severity supports the finding reported by McGinnis and Kerans (undated report) for western Montana drainages. These investigators determined that the risk of whirling disease in watersheds increased as EPT richness diminished {r^ = 0.35, p = 0.07). Neither E richness, P 132 richness, nor T richness, when considered singly, produced significant correlation with the incidence of whirling disease, suggesting that different groups may have been advantaged in different environs. Although the association of EPT richness with whirling disease incidence was strong in our analysis, both Monture Creek and Belmont Creek supported relatively large numbers of EPT taxa despite the presence of whirling disease; EPT richness in samples collected in these drainages equaled or exceeded EPT richness in 3 drainages (Arrastra Creek, Johnson Creek, and West Twin Creek) that apparently did not harbor the infection. These findings suggest that it may be important to consider the presence or absence of individual taxa instead of metric summations of taxa richness. Since both mean water temperature and July-August maximum water temperature were significantly associated with the incidence of whirling disease (R = 0.621 p < 0.05 and R = 0.799 p < 0.05 respectively), it was not surprising that the number of cold stenotherm taxa present in samples was also significantly correlated with whirling disease incidence. Pollution sensitive taxa are often cold stenothermic; the significant relationship between that metric and the whirling disease measures is predictable given the overlap between those 2 groups of aquatic invertebrates. However, similar to EPT richness, neither cold stenotherm richness nor pollution sensitive richness were perfect predictors of the presence of whirling disease. These correlations suggest that the incidence and severity of whirling disease may be related to water temperature. Those taxa that were collected only in infected streams (Table 4) are generally tolerant of warmer thermal conditions, while those collected only in uninfected streams (Table 3) are generally cold stenotherms. A thermal association with whirling disease is also evident in the ordination. While no strong clustering of sites is apparent, the ordination suggests a sorting of assemblages with respect to temperature tolerances. There is a temperature gradient evident from the lower left comer of the plot to the upper right corner when July- August maximum temperature is considered. The plot suggests that sites with higher July-August maximum temperature tend to be infected with whirling disease. Other than Arrastra Creek, which demonstrated a low mean severity and 0% severely infected reaches, all infected sites plot in the upper right of the ordination space. The significant association between whirling disease incidence and the metals tolerance index was an unexpected result. Among the group of sites studied, the incidence of whirling disease tended to be higher in streams where the invertebrate assemblage was more tolerant to metals contamination, when tolerance was measured by this index. Curiously, sites in watersheds with a high percentage of reaches (48.5% and greater) with whirling disease incidence have higher forest cover as a percentage of watershed area than uninfected watersheds (R = 0.853 p < 0.05). Forested area ranges from 73.4% to 87.3%) in drainages without demonstrated whirling disease infection, and from 88.9%) to 95.1%) in infected drainages. References Bollman, W., R. Pierce and L. Eby. 2006. A pilot assessment of the association between stonefiy assemblages and the incidence and severity of whirling disease in tributaries of the Blackfoot River, Montana. Unpublished report to Montana Fish Wildlife and Parks. Ron Pierce, project manager. McGinnis, S. and B.L. Kerans. Undated report. A preliminary assessment of land use and aquatic invertebrates as indicators of whirling disease risk in Montana. MSU. 133 Environmental conditions linked to Myxobolus cerebralis infection in basin-fed streams of the Blackfoot Watershed, Montana Ron Pierce Montana Fish, Wildlife and Parks Lisa Eby University of Montana Wease Bollman Rhithron Associates Dick Vincent Montana, Fish Wildlife and Parks Abstract The exotic parasite Myxobolus cerebralis, the infectious agent that ultimately causes whirling disease in salmonids, has successfully invaded many river systems throughout the interior West. Given the patchy distribution and variable effects of whirling disease, it is important to identify the physical and biological characteristics of watersheds that influence infection. In the Blackfoot Basin of western Montana, we investigated relationships between a group of five landscape-and four reach-scale environmental conditions and the presence of infection in hatchery rainbow trout {Onchorynchus mykiss) within basin-fed tributaries of the Blackfoot River in order to determine variables correlated with infection. Study results found that rainbow trout developed little to no infection in streams with higher gradients, lower levels of fines within the substrate and low summer temperatures despite the near proximity to higher infection rates in rainbow trout in adjacent waters. Infections were present in streams with summertime maximum water temperatures over 19°C; a logistic regression model including maximum water temperatures, fine sediment (<0.84mm) and channel gradient explained the presence of rainbow trout infection within our thirteen study streams. In our study area, rainbow trout infections were present and infection rates were high in meandering streams in broad valleys with gentle relief and warmer summer temperatures. We also examined the relationship between invertebrate taxa and our ability to detect infection. Species richness of stenohaline species was the single best indicator of infection; a logistic regression with stenohaline species, richness of Ephemeroptera, Plecoptera and Trichopera (EPT) species, and sediment sensitive species richness described was the best biotic model for describing the variation of infection in our dataset. In western Montana, environments with infected fish are often prone to the additive effects of habitat alterations that may increase sediment or temperature regimes. Such changes, whether natural or anthropogenic, may increase disease stressors in wild trout populations. Key words : Blackfoot River Basin, Myxobolus cerebralis, infection, environmental characteristics, water temperature, macroinvertebrate assemblage, habitat 134 Introduction Whirling disease is a chronic disease caused by the invasive myxosporean parasite Myxobolus cerebralis (Hoffman 1990) that was introduced into North America in the 1950s (Bartholomew and Reno 2002). Myxobolus cerebralis has a complex, two-host life cycle involving the aquatic oligochaete worm Tubifex tubifex and most members of the salmonidae, the youngest of which (age-0 fry) have been shown to be the most susceptible to infection (Ryce 2004). High mortality and recruitment collapse has occurred in certain infected rainbow trout {Oncorhynchus mykiss) populations in Montana (MacConnell and Vincent 2002) and Colorado (Nehring and Walker 1996). First detected in Montana in 1994 in the Madison River, this disease has spread to salmonid-dominated river systems throughout western Montana; however, infection rates are geographically highly variable among and within watersheds despite the ubiquitous presence of the salmonid host. This high variation of infection across regions, within drainages and within streams has been observed throughout the West including Colorado, Idaho, Utah, California and Montana (Nehring and Walker 1996, Modin 1998, Hiner and Moffitt 2001, Sandell et al. 2001, de la Hoz Franco and Budy 2004, Krueger et al. 2006). The extent of contact between vulnerable fry and the release of the infective triactinomyxon (TAM) stage of the parasite determine the degree of exposure for young fish and, ultimately, the magnitude of population-level effects. Thus, watershed characteristics and aspects of tributaries that influence T. tubifex availability and spore production can alter exposure rates. We can identify watershed characteristics governing whirling disease by examining distribution patterns of infection within the broader landscape. Environmental factors play a critical role in determining the result of host and parasite interactions (MacConnell and Vincent 2002). Temperature influences the growth, reproduction and survival of T. tubifex (de la Hoz Franco and Budy 2004) as well as spore and infective TAM production (El-Matbouli et al. 1999). Water velocity may also influence TAM survival and concentrations (Kerans and Zale 2002, MacConnell and Vincent 2002). Substrate size and nutrients influence the distribution and abundance of T. tubifex (Sauter and Gude 1996, Arndt et al. 2002). Fewer studies examine how these factors interact in a field setting to help explain the distribution and prevalence in the environment (Hiner and Moffitt 2002, De la Hoz Franco and Budy 2004, Krueger et al. 2006). Linking these potentially limiting factors to patterns on the landscape in a variety of different system types and regions is useful as we attempt to separate the effects of correlated variables in the field, understand different limiting factors across different types of systems, and provide generalizations of vulnerability to disease. Response variables associated with infection (e.g., severity, spore production) increase with average water temperature (and variation in water temperature) in Idaho and Utah studies (Hiner and Moffitt 2002, de la Hoz Franco and Budy 2004), but decrease with water temperature in the tailwater section of Madison River MT (Krueger et al. 2006). Given that the temperature range observed in the Madison River study is similar to the other studies (range in study 10.1-13.6°C), this effect could be due to the inverse correlation among fine sediment and temperature in the dataset or other factors related to flow alterations. Beyond temperature, the other variables that have a positive relationship with disease severity include water velocity (de la Hoz Franco and Budy 2004), fine sediments (Krueger et al. 2006), and density of oligochaetes and chironomids (Hiner and Moffitt 2002). The distribution of M cerebralis infected rainbow trout in basin-fed environments seems to adhere to a fairly predictable geographic pattern of increasing infection rates in the downstream direction (Smith 1998, Sandell et al. 2001, Hubert et al. 2002). Although the longitudinal relationships of infection has been described in some areas (Sandell et al. 2001, De la Hoz Franco and Budy 2004), the variables linked to infection (and severity) vary by geographic region and the specific physical characteristics influencing infection have not been quantitatively evaluated in western Montana tributaries. Improving our understanding to better predict infection in the field will require expanding our understanding of the environmental 135 mechanisms that result in spatial overlap of fry with high concentrations of triactinomyxon (Downing et al. 2002, Kerans and Zale 2002). This would allow fisheries managers to better predict the species and streams in which high or low severity of whirling disease might be expected. If fisheries managers could link disease potential on the landscape (based on geomorphic and physiochemical predictors), they could forecast which species, life histories or stream-types are most vulnerable to whirling disease and better determine broad-scale (versus localized) population effects. In addition, depending on the factors determining exposure potential, managers may be able to offset disease effects in a particular tributary through habitat restoration or other management techniques. Whirling disease, first identified in the Blackfoot Basin in 1995, is currently present throughout the mainstem Blackfoot River and lower reaches of many tributaries (Pierce et al 2006). Although not quantified, there tends to be a general inverse relationship between channel elevation, channel slope and infection rates, which appears to result from a lack of suitable habitat (slow water and fine substrate) to support T. tubifex (Smith 1998). Biotic predictors (e.g., sentinel exposures, T. tubifex or TAMs) are time consuming, expensive, and require specialized identification training and equipment. The ability to identify high disease risk areas potentially before severe infection levels occur is critical for practitioners to focus research and management effort. This research investigates the M cerebralis infection of rainbow trout with abiotic attributes of streams within a heterogeneous area of the Blackfoot Basin. In addition, riffle invertebrate assemblages were correlated with infection, as these data are routinely collected in environmental or water quality assessments. Our goals of this project are both to develop a model that identifies abiotic environmental conditions that explain the presence of infection, and to examine the relationship of the occurrence of infection with invertebrates to determine whether they may be a good indicator of infected environments. Study area - The Blackfoot River, a 5* order tributary (Strahler 1957) of the upper Columbia River, lies in west-central Montana and flows west 211 km from the Continental Divide to its confluence with the Clark Fork River in Bonner, Montana. The River drains a 3,728 km^ watershed through 3,040 km of perennial streams and generates a mean annual discharge of 45.2 m''/s (United States Geological Survey 2006). The geography of the watershed is a physically diverse, geo-structurally controlled glacial landscape with alpine and subalpine mountains at the upper elevations, montane forests at the mid-elevations and semi-arid glacial pothole and outwash topography on the valley floor. Many tributaries of the Blackfoot River begin in high cirque basins, flow through alluvial valleys with meandering streams in broad valleys with gentle relief, while others flow through confined steeper channels of non-glacial origin before entering the Blackfoot River. Lands in the upper Blackfoot Basin are mostly public (65%) headwater areas with about 35 percent privately held lands consisting primarily of timbered foothills and agricultural bottomlands. The Blackfoot River is a renowned trout river in Montana and contains diverse self- sustaining wild trout populations, most of which reproduce in tributaries (Pierce et al 2007, 2006). Salmonids of the Blackfoot watershed include brook trout (Salve limis fontinalis), brown trout (Salmo triitta), bull trout {S. confluentus), mountain whitefish (Prosopium williamsoni), rainbow trout (Oncorhynchus mykiss), and westslope cutthroat trout (O. clarkii lewisi) all of which possess some level of whirling disease susceptibility (MacConnell and Vincent 2002). 136 D A99«49ni«nt sit«s I'rJDhnssonCr 21 West Twin Cr 31 East Twin Cr 4] Buac Cr 5) Oatd Cr 6) B*lmonl Cr 71 Blanched Cr 8) Bh Cr 9\ ChamtHflaJn Cr 101 CotfonwHtd Cr 11>MonlureCr 1?}ArrastrF] Cr K\ Landars forts I I It M »l3 on the MacConnell-Baldwin scale. As an index to severity, mean lesion scores of >2.75 have been associated with significant levels of mortality in wild rainbow trout populations (Vincent 2002). Analyses - Scatterplots of the response variables (index of mean infection or percent of fish with infection grade >3) indicted non-linear relationships with many of our predictor variables. Typically infections were either not-detected or infection rates were relatively high (majority > grade 3) with very few intermediate values. In addition, streams with no or very low (<0.05 mean score) or relatively high histological scores (>3 severity) have remained relatively consistent from year to year (Pierce et al. 2006). Given these considerations we decided to use infection presence and non-detect in our analyses. We used two approaches to link environmental conditions to infection. First, we used classification and regression trees (CART, Venables and Ripley 1997) to examine whether the groups (non-detect verses infected) reflect differences in any of the environmental predictor variables. Classification and regression trees attempt to partition a data set by recursively explaining subsets of the data using either continuous or categorical variables (Breiman et al. 1984). Because of the small data set (and only 5 streams with infection present), in Splus we set the minimum node size to be 3 but reduced the number of potential splits of the dataset ("pruned the tree") to prevent overfitting the data. In addition to CART analyses, we used logistic regression to identify the environmental variables that best explained the variation in non-detect/presence data. Logistic regressions were run with a backward elimination method (likelihood ratio) and performed in SPSS version 11. Results Our first set of analyses examined which landscape-scale parameters best explained the presence of infection in tributaries. We included percent valley slope, percent forest cover, sinuosity, channel type, and stream order. Valley type and valley slope were not both used in this analysis because of the inherent relatedness of the factors (Table 1). Next we used the same analytical techniques to examine which tributary characteristics best explained whether streams were vulnerable to infection. To remove correlated predictive variables, we first examined the relationships among the sediment measures. There were significant correlations among all of the McNeil coring measures, including the Fredle and geometric mean measures, as well as many of the Wolman pebble count measure (e.g., D50). To reduce correlated predictor variables and given the known relationship of T. tubifex with fine sediment, we used the percentage of substrate less than 0.84 mm to describe substrate composition in our analyses. This measure of substrate composition significantly correlated with several stream measures, including entrenchment and sinuosity. Conductivity was significantly correlated with pH, Total Dissolved Solids, entrenchment, and reach slope. In addition, average temperature, maximum temperature, and reach slope were significantly correlated. Given the known importance of temperature to the biology of T. tubifex and spore production, we included temperature (versus slope) in the analyses. To examine what stream reach characteristics described the presence of infection, we excluded correlated predictor variables and included the following variables: (1) maximum summertime temperature, (2) substrate less than 0.84mm, (3) width/depth ratio, and (4) conductivity (Table 2). 139 Mean Percent > Percent Percent Stream lesion grade 3 Presence/ Sinuosity Channel watershed valley Stream Valley score infected non-detect type forested slope order type Johnson 1.1 B4 75.8 8.6 2 II E.Twin 1.1 C4 73.4 7.7 2 II W.Twin 1.3 C4 92.6 5 3 II Bear 1.6 C4b 86.8 5.7 2 VI Gold 1.4 C3 90 1.4 3 V Belmont 2.48 48.5 1 1.0 B4 95.1 1.3 3 V Elk 4.82 95.3 1 1.8 E4 83.1 0.4 3 VIII Blanchard 1.3 C4 88.9 3 2 VI Cottonwood 3.78 100 1 1.1 C3 77.4 0.7 3 VIII Monture 4.81 96.9 1 1.5 C3 74.6 0.7 4 VIII Arrastra 0.02 1.3 C4 86.4 1.2 2 V Chamberlain 3.78 78 1 1.1 C4 95 2.1 2 VI Landers Fork 1.2 C4 87.3 0.9 4 V Table 1 . Variables used in the landscape scale analyses. Valley slope and valley type are correlated and were not included in the same analysis. Stream Presence/ Width/Depth % substrate Conductivity Mean Temp °C Max Temp °C non-detect ratio <0.084mm (July/Aug) (July/Aug) Johnson 9.5 7.9 39 10.3 13.7 E. Twin 11.5 5.9 18 11.8 14.8 W. Twm 10.8 5.3 9 11.8 15.6 Bear 14.4 5.9 82 11.2 17.5 Gold 20.3 11 162 13.0 18.7 Belmont 1 34.2 9.2 258 13.2 18.3 Elk 1 11.2 29.1 230 13.9 20.6 Blanchard 33.3 6.3 95 14.9 18.3 Cottonwood 1 17.6 11.3 220 11.8 20.9 Monture 1 48.0 12.1 135 14.7 20.6 Arrastra 34.1 9.3 159 10.1 14.4 Chamberlain 1 19.2 6.6 67 13.9 19.4 Landers Fork 27.3 6.5 204 11.5 17.1 Table 2. Variables used in the reach scale presence/non-detect analyses. 140 0) o c 0) M Q. Q 5 Valley Type VIM -•"M- 2.0 4.0 6.0 Valley Slope 8.0 Figure 2. Plot of the whirling disease detection associated with valley slope and valley type. Finally, we used similar analytical techniques to examine how much variation in whirling disease presence could be described by the invertebrate community indices. There are many potential indices to describe the invertebrate community; we focused on those that were linked to ecological mechanisms and we excluded highly 1 n(** ♦ ) ♦ correlated indices. We included cold stenotherm richness, EPT richness, sediment sensitive richness, and sediment tolerant richness in analyses to investigate how well invertebrate community indices describe variation in whirling disease. When we examined the landscape predictors to the presence of infection (Table 1), the classification and regression tree analysis predicted infection presence based on percent valley slope (if valley slope is <0.8 present). In this Blackfoot River Basin dataset, if the valley slope is less that 0.8 then it is also a Valley type 8 (Rosgen classification). These valley types are characterized by wide, gentle valley slopes with well- developed floodplain adjacent to river or glacial terraces typically containing alluvial valley fill (Rosgen 2006). This analysis misclassified two tributaries, where there was a predicted non- detect when in fact infection was present. The misclassified tributaries were Chamberlain Creek (valley type 6, valley slope 2.1) and Belmont Creek (valley type 5, valley slope 1.3); although these two streams had higher slopes they did have warm summertime temperatures (max temp > 18°C). In this data set, valley type is significantly correlated with valley slope, channel slope, temperature, and percent of substrate less than 0.84mm, pH, conductivity and Total Dissolved Solids. Results from the logistic regression were consistent with the CART analysis that valley slope was the only parameter not excluded from the final model (valley slope p=0.022, 67% misclassification. Figure 2). In our analyses examining which reach level variables were linked to infection, the CART model predicted infection in streams with maximum summertime temperatures above 19.02°C and the only misclassified tributary was Belmont Creek (maximum temp <19.02°C but infection present). Two different logistic regression models correctly classified all of the streams with two predictor variables: maximum temperature and conductivity or maximum temperature and width/depth ratio. The best individual predictor of disease presence was maximum summertime temperature (Figure 3). Not surprisingly, there was a significant correlation between our landscape level (valley slope) and several reach level variables (maximum temperature Pearson Correlation R=-0.74, p=0.006; fine sediment <0.8mm R=0.585, p=0.046; and Total Dissolved Solids R=-0.806, p=0.002). 10.0 141 1 01 0.8 .2 0.6 Q I 0.4 5 0.2 1 0.8 0) U) (Q Si 0.6 □ S) I 0.4 s 0.2 ♦ ♦ 1 o 0.8 u (0 0) .12 0.6 □ £ 0.4 !E 5 0.2 12 14 16 18 Max Summer Temp ♦ ♦ ♦♦ 20 22 10 20 30 40 Width/Depth Ratio 50 60 .m 0.6 a B 0.4 - !c S 0.2 »«» ♦ ♦ I ♦ ♦ ♦ ♦ ♦ ♦ -♦-♦t^ 10 15 20 % substrate < 0.84nim 25 30 50 100 150 200 Conductivity 250 300 Figure 3. Scatterplots of the variables that were significant predictors of infection presence at the reach scale from the logistic regression analysis. A whirling disease score of 1.0 indicates detection and indicates infection was not detected in the stream. Sixty-seven EPT taxa were identified in the samples and a total of 10,644 EPT individuals were present. Several taxa were collected in more than one stream and exclusively in streams with no infection, including Drimella spinifera (2 sites), Caudatella edmimdsi (2 sites), Epeoriis longimamis (5 sites), Epeorus deceptiviis (2 sites), Rhithrogena sp. (4 sites), Ironodes sp. (2 sites) Baetis alius (2 sites), Baetis flavistriga (2 sites), Visoka cataractae (2 sites), Calineiiria californica (2), Megarcys signata (4 sites), Parapsyche elsis (4 sites), Agraylea sp. (2 sites), Lepidostoma imicolor (4 sites), Rhyacophila alberta (3 sites). The CART model predicted the presence of infection in streams with cold stenotherm richness of less than 6 species. This model misclassified 3 of the 13 streams, with Belmont incorrectly classified as a non- detect stream and Gold and Blanchard incorrectly classified as streams with infection. The logistic regression model indicated that we could correctly classify 12 of our 13 streams with two predictor variables: EPT species richness and cold stenotherm species richness. In this analysis, Belmont Creek was the only stream incorrectly classified (Figure 4). Tributary Cold Stenotherm Richness Sediment Sensitive Richness EPT Richness Sediment Tolerant Richness Arrasta 7 3 30 Bear 7 2 29 Belmont 7 4 30 Blanchard 5 4 39 2 Chamberlain 5 3 26 1 Cottonwood 1 4 21 East Twin 12 4 34 Elk Gold Johnson Landers Fork M onture West Twin 10 26 33 29 33 32 29 Table 3. Indices developed from the invertebrate data and used in analyses 142 CD CO CD CD CO 1 - ♦ ♦ ♦ ♦ 1 0.8 0.8 0.6 0.6 0.4 0.4 0.2 n ▲▲ ▲▲ 0.2 u 15 \ ^^ •• 1 25 35 1 U 45 EPT Richness ♦ ♦ — ♦ ♦ — ♦-♦ — ♦ ♦ 5 10 15 Cold Stenotherm Richness Figure 4. Scatterplots of the invertebrate variables that were significant predictors of infection presence from the logistic regression analysis. A score of 1.0 indicates infection present and indicates infection was not detected or has a mean histological score of <0.05. Discussion Our study failed to detect infection of M. cerebralis in rainbow trout in colder high gradient streams but identified high infection rates in lower-gradient broader valley environments. These findings are consistent with down-valley morphological transitions of stream valleys and channel -types (Rosgen 1996) and a central tenet of stream ecology that abiotic conditions and biotic communities change predictably along longitudinal (upstream- downstream) gradients (Vannote et al. 1980). In western Montana watersheds, channel features often transition in a predictable downstream path from steep cold streams to broader alluvial valleys. However due to basin heterogeneity, stream environments conducive to infection such as those in the Blackfoot Basin also vary substantially at a reach and sub-basin scales. In contrast to the lower Blackfoot Basin, the middle Blackfoot Basin includes many tributary environments suitable to T. tubifex, high TAM exposure and higher infection rates. In our analysis of landscape variables, environments conducive to infection in fish were found in broad valleys with gentle, down-valley elevation relief (Valley type 8, Rosgen 1996). In the down-valley continuum of Rosgen channel-types, streams within these broad gentle valleys occupy are C and E-type channels, which are slightly entrenched. In our study area, high infection rates were prevalent in meandering streams in broad valleys with gentle relief and warmer summer temperatures. Alluvial floodplains are the most predominant landforms, which can produce relatively high sediment supply. Soils are developed over alluvium, thus, streams in this valley are susceptible to accelerated bank erosion. In our study area, these streams are also prone to anthropogenic activities that elevate sediment and water temperature levels (Pierce et al 2006). By contrast, mountain streams of the lower Blackfoot Basin support lower temperatures, lower in-channel sediment levels within spawning areas, and low infection rates. These streams occupy steep forested valleys, narrow floodplains with moderate side-slopes formed primarily of colluvium. At the reach scale, we had a strong positive correlation with infection and water temperature, similar to other studies (Hiner and Moffitt 2002, de la Hoz Franco and Budy 2004). This was expected given the entire life cycle of the parasite in both the fish and worm host is temperature dependant and natural outbreaks occur during temperatures optimal to the parasite (MacConnell and Vincent 2002). Summertime temperatures identify conditions favorably for spore TAM production and release, and thus the potential of a stream to support a high infection rates. Temperature also influence fish growth, development and immune response, all of which may influence the degree to which fish are susceptible to the development of the disease 143 (MacConnell and Vincent 2002). Maximum annual water temperature also provides a simple and direct method of comparison to the other basins (De la Hoz Franco and Budy 2004). Similar to other studies, we found a relationship with conductivity (in addition to temperature), the only water quality parameter correlated to disease severity in Oregon (Sandell et al. 2001). Although Sandell et al. (2001) provide a potential mechanistic explanation associated with conductivity influencing TAM recognition of living tissue, in our study conductivity is significantly correlated with Total Dissolved Solids, entrenchment, slope, and geometric mean sediment size. Therefore, high conductivity is indicative of stream reaches with low gradient channels, more fine sediments, greater entrenchment and higher Total Dissolved Solids. We did not find that stream size using bankfull area or stream-order as potential surrogates for discharge explained a significant amount of the variation in the occurrence of infection. But in a similar study, both water temperature and discharge were positively correlated with infection (de la Hoz Franco and Budy 2004). In the Blackfoot Basin, the variable geology infiuences the formation of large, cold-water bodies with very low incidence of infection. The results from analyses of our invertebrate indicators corroborate results from the analyses of abiotic factors. Cold stenotherm species richness classified infection fairly well in the CART analysis and EPT richness and cold stenotherm species richness predicting the presence of infection in the regression analyses. Developments of high severity have been shown in areas of natural and man-made impoundments (Hiner and Moffit 2002), both of which are present in our study area and likely infiuence infection. Impoundments and can affect community structure of macroinvertebrates trap fine sediment and organic matter and increase warming and thus create optimal conditions for T. tubifex. These conditions can lead to the production and release of TAMs from infected worm populations. Both Belmont and Chamberlain Creeks have ponds (either man-made or beaver) upstream of the sentinel cage sites. Throughout this study Belmont was an outlier, a cold stream with whirling disease. The Belmont basin consists of 92% private industrial forest with 6.4-miles of logging roads per mile^, which once generated an estimated 200 tons/year of fine sediments from roads alone (Sugden, 1994). This combination of elevated sediment trapped in areas of increased recent beaver activity is likely creating this outlier "hot spof condition for infection. Management Implications - As we begin to better understand the environmental factors that influence infection, we can begin to predict areas that we expect to be naturally prone to having severe impacts of the disease. Reach scale variables imply that within certain watersheds we might manage the potential impacts of disease by protecting and restoring habitat in tributary spawning and rearing areas to minimize factors that favor habitat of T. tubifex (e.g. sediment) or otherwise increase whirling disease (e.g., temperature). In our study area, infections were identified in meandering streams in broad valleys with gentle relief and warmer summer temperatures. In western Montana, environments of this type are often prone to excessive grazing and other land-use activities that potentially elevate water temperatures and instream sediment levels. Zendt and Bergerson (2000) found the highest relative abundance of T. tubifex in areas where riparian zones were heavily disturbed. Our study further suggests that anthropogenic warming or increases in sediment supply may increase infection (and severity) and shift disease distribution (and effects) in the upstream direction. For example, Monture Creek has been identified as the primary rainbow trout spawning tributary to the Blackfoot River. Water temperature monitoring beginning in 1993 has shown a >1°C increase in maximum annual temperatures. Relative influences of regional trends versus anthropogenic influences driving this temperature change are unclear as confounding factors such as habitat degradation of riparian areas is occurring within the watershed, which is likely influencing several environmental conditions conducive infection rates. If the exposure potential shifts upstream, so will the likely impacts of the disease on various species that 144 typically spawn higher in the watershed. While there is little we can do within the Blackfoot Basin to affect climate, land-management related water temperature changes may be able to reduce disease severity. If we can predict which streams are likely to be sites with high disease severity because of habitat degradation, we can develop and/or implement existing stream restoration techniques to correct human-related factors in areas that may contribute to severity. Examining the potential for restoration to prevent and/or reverse trends in disease severity is needed. In addition, a priori predictions may allow managers to change fishing regulations to decrease other sources of mortality that would allow these populations to better handle the stress of extra juvenile mortality resulting from whirling disease. Hopefully, this information will help us maintain self-sustaining wild salmonid populations in parasite positive streams. Conclusions Determining the physical variables correlated with infection is critical to understanding the spatial characteristics of the whirling disease impacts on fish populations. This study improves our ability to better predict the aquatic environments prone to infection. Understanding the environmental and spatial relationships within the watershed allow more concise interpretation of the effects on susceptible species, when considered within a context of overlapping life-histories for vulnerable species. The role of anthropogenic habitat degradation must also be considered in terms of combined effects of multiple threats including disease. Based on our study, increased water temperature appears to be coupled with infection. Habitat degradation often increases temperature, in which cases we can generally predict a higher infection rates. Acknowledgments We would like to acknowledge the USPS for sifting samples, NRCS and Blackfoot Challenge for funding the McNeil core samples. Northwestern Energy for funding field technicians Ryen Aasheim Anderson, Mike Davidson and Craig Podner, and the Whirling Disease Foundation for contributing funds to help complete this project. We would also like to thank Eileen Ryce for her review of this manuscript. 145 References Anderson, R.A. 2004. Occurrence and seasonal dynamics of the whirling disease parasite, Myxobolus cerebralis, in Montana spring creeks. Masters Thesis Montana State University Bozeman, MT. Arndt, R.E., E.J. Wagner, Q. Cannon, and M. Smith. 2002. Triactinomyxon production as related to rearing substrate and diel light cycle. Pages 87-91 in J.L. Bartholomew and J.C. Wilson, editors. Whirling disease: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda, Maryland. Baldwin, T. J., E. R. Vincent, R. M. Silflow, D. Stanek. 2000. Myxobolus cerebralis infection in rainbow trout (Oncorhynchus mykiss) and brown trout (Salmo trutta) exposed under natural stream conditions. Journal of Veterinary Diagnostic Investigations 12:312-321. Bartholomew, J.L. and P.W. Reno. 2002. The history and dissemination of whirling disease. Pages 3-24. in J.L. Bartholomew and J.C. Wilson, editors. Whirling disease: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda, Maryland. Breiman, L., J.H. Friedman, R.A. Olshen, and C.J. Stone. 1984. Classification and regression trees. Chapman and Hall, New York. De la Hoz, E, and P. Budy. 2004. Linking environmental heterogeneity to the distribution of prevalence oi Myxobolus cerebralis: a comparison across sites in a northern Utah watershed. Transactions of the American Fisheries Society. 133:1176-1189. Downing, D.C., T.E. McMahon, B.L. Kearns, and E.R. Vincent. 2002. Relation of spawning and rearing life history of rainbow trout and susceptibility to Myxobolus cerebralis infection in the Madison River, Montana. Journal of Aquatic Animal Health 14: 191-203. El-Matbouli, M., T.S. McDowell, D.B. Antonia, K.B. Andree, and R.P. Hendrick. 1999. Effect of water temperature on the development, release, and survival of triactinomyxon stage of Myxobolus cerebralis in its oligochaete host. International Journal for Parasitology 29:627- 641. Hiner, M., and CM. Moffitt. 2001. Variation in Myxobolus cerebralis infections in field- exposed cutthroat and rainbow trout in Idaho. Journal of Aquatic Animal Health 13:124- 132. Hiner, M. and CM. Moffitt. 2002. Modeling Myxobolus cerebralis infections in trout: associations with habitat variables. Pages 167-179. in J.L. Bartholomew and J.C. Wilson, editors. Whirling disease: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda, Maryland. Hoffman, G.L. 1990. Myxobolus cerebralis, a worldwide cause of salmonid whirling disease. Journal of Aquatic Animal Health 2:30-37. Hubert, W. A. and six coauthors. 2002. Whirling disease among Snake River cutthroat trout in two spring streams in Wyoming. Whirling disease: reviews and current topics. American Fisheries Society Symposium 29:181-193. Kearns, B.L., and A.V Zale. 2002. The ecology oi Myxobolus cerebralis. Pages 145-166. in J.L. Bartholomew and J.C. Wilson, editors. Whirling disease: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda, Maryland. Krueger, R.C, B.L. Kearns, E.R. Vincent, and C Rasmussen. 2006. Risk of Myxobolus cerebralis infection to rainbow trout in the Madison River, Montana, USA. Ecological Applications 16:770-783. Landolt, P, and M, Sartori. 1997. Ephemeroptera and Plecoptera: biology-ecology-sytematics. MacConnell, E. and E. R. Vincent 2002. Review: the effects oi Myxobolus cerebralis on the salmonid host. Pages 95-108 in J. L. Bartholomew and J. C Wilson, editors. Whirling disease: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda Maryland. 146 Merritt, R. W., K. W. Cummins and M. B. Berg, eds. 2008. An Introduction to the Aquatic Insects of North America, Fourth Edition. Kendall/Hunt Publishing Company. Dubuque, Iowa, xi + 1158 p. Modin, J. 1998. Whirling disease in California: A review of its history, distribution, and impacts, 1965-1997. Journal of Aquatic Animal Health 10:132-142. McNeil, W.J. and W.H Ahnell. 1964. Success of pink salmon spawning relative to size of spawning bed material. U.S. Fish and Wildlife Service Special Scientific Report Fisheries 469. MWDTF (Montana Whirling Disease Task Force). 1996. Final report and action recommendations. Montana Whirling Disease Task Force, Helena, MT. Nehring, B., and P. G. Walker. 1996. Whirling disease in the wild: the new reality in the intermountain west. Fisheries 21: 28-30. Pierce., and R. Aasheim, and C. Podner 2005. An integrated stream restoration and native fish conservation strategy for the Big Blackfoot River Basin, Montana Fish, Wildlife and Parks, Missoula, MT. Pierce, R., and C. Podner. 2006. The Big Blackfoot River Restoration Report for 2004 and 2005. Montana Fish, Wildlife and Parks. Missoula, MT. Pierce, R., C. Podner, M. Davidson, and R. Vincent. In Review. Relationships of migratory (hybrid) rainbow trout spawning life histories to risk of Myxobolus cerebralis infection in the Blackfoot River Basin, Montana. In submission to Transactions of the American Fisheries Society. Rosgen, D. 1996. Applied Fluvial Geomorphology. Wildlands Hydrology, Pagosa Springs Colorado. Ryce, E. K. N, A. V. Zale and E. MacConnell. 2004. Effects offish age and parasite dose on the development if whirling disease in rainbow trout. Diseases of Aquatic Organisms Vol. 59(3):225-233. Sandell, T. A., H. V. Lorz, D. G. Stevens, and J. L. Bartholomew. 2001. Dynamics of Myxobolus cerebralis in the Lostine River, Oregon: implications for resident and anadromous salmonids. Journal of Aquatic Animal Health 13:142-150. Sauter, G., and H. Gude. 1996. Influence of grain size on the distribution of tubificid oligochaete species. Hydrobiologia 333:97-101. Shepard, B. B., and P. J. Graham. 1982. Monitoring spawning gravel bed material used by bull trout on the glacier view district Flathead National Forest. Montana Fish Wildlife and Parks Completion Report. Shepard, B., S. Leathe, T. Weaver, and M. Enk. 1984. Monitoring levels of fine sediments within tributaries to Flathead Lake and impacts of fine sediment to bull trout recruitment. Wild Trout III. Smith, L. 1998. Study on the distribution and abundance of Tubifex tubifex within Cottonwood Creek in the Blackfoot drainage. Masters Thesis, University of Montana, Missoula, Montana. Strahler, A.N. 1957. Quantitative analysis of watershed geomorphology. Transactions of the American Geophysical Union 38:913-920. Sugden, B. 1996. Belmont Creek watershed analysis. Plum Creek Timber Company. Clearwater Unit, Missoula, MT. Thorp, J. H. and A. P. Covich, eds. 2001. Ecology and classification of North American freshwater invertebrates (Second Edition). Academic Press. San Diego, California, xiii + 1056 p. USGS. 2006. Gauging station 1234000 provisional unpublished data. Vannote, R. L., GW. Minshall, K.W. Cummins, JH. R. Sedell, and C.E. Cushing. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences.; 37: 130-137. 147 Venables, W.N. and B.D. Ripley. 1997. Modem applied statistics with S-PLUS (2"'' edition). Springer, New York. Vincent, E. R. 1996. Whirling disease and wild trout: the Montana experience. Fisheries 21 (6):32-33. Vincent, E.R. 2000. Whirling disease report 1997-98. Montana, Fish, Wildlife and Parks. Project 3860. Helena, Montana. Vincent, E. R. 2002. Relative susceptibility of various salmonids to whirling disease with emphasis on rainbow and cutthroat trout. Whirling disease: reviews and current topics. American Fisheries Society Symposium 29:109-115. Wolman, M. 1954. A method of sampling coarse river-bed material. Transactions of the American Geophysical Union 35:951-956. Zendt, J. S. and E. P. Bergersen. 2000. Distribution and abundance of the aquatic oligochaete host Tiibifex tubifex for the salmonid whirling disease parasite Myxobolus cerebralis in the upper Colorado River basin. North American Journal of Fisheries Management 20:502-5 12. 148 Relationships of migratory (hybrid) rainbow trout spawning life histories to risk of Myxobolus cerebralis infection in the Blackfoot River Basin, Montana Ron Pierce, Craig Podner, Michael Davidson and Richard Vincent, Montana Fish, Wildlife and Parks Abstract - The middle Blackfoot River Basin in western Montana is the site of a low-elevation whirling disease epizootic among rainbow trout {Oncorhynchus mykiss hybrids) caused by a recent invasion of the exotic parasite Myxobolus cerebralis. To assess exposure of Blackfoot River rainbow trout to the parasite, we investigated the spawning life histories of adult rainbow trout with respect to the distribution and severity of disease in spawning and rearing areas in two distinct reaches of the Blackfoot River. Radio-telemetry confirmed Blackfoot River rainbow trout migrate from wintering sites within the Blackfoot River to spawning tributaries. Over 90% of telemetered rainbow trout in the middle Blackfoot River spawned in a low-gradient, infected stream where fry emerged in early July during the vulnerable, highly infectious period. By contrast, spawning of lower Blackfoot River rainbow trout was dispersed among smaller, colder, higher gradient tributaries, most of which fell below disease detection levels. For fluvial rainbow trout risk of exposure varies at a sub-basin scale and relates to the geographical arrangement and properties of tributaries, including the longitudinal relationship of disease to spawning and early rearing areas. Prior to the invasion ofM cerebralis, the middle Blackfoot River was identified with recruitment-limitations caused by winter mortality and anthropogenic activities. Management implications suggest that riparian restoration and habitat enhancement with emphasis on migratory native fish within and upstream of infected waters may buffer effects of the disease. Key words: Blackfoot River, rainbow trout, whirling disease, migration, movement patterns, tributary, and population risk Introduction Whirling disease (WD hereafter), a parasitic infection caused by the myxosporean Myxobolus cerebralis, has been associated with significant declines in wild rainbow trout (RBT hereafter) populations in certain streams in the western United States (Nehring and Walker 1996; Vincent 1996). WD was first detected in Montana in 1994 within the renowned Madison River following large and unexplained declines in RBT abundance. Soon thereafter, WD was described as one of the single greatest threats to wild trout (MWDTF 1996). Yet with time and the expansion of WD, it appears the infiuences of WD on interior populations of RBT are highly variable among watersheds (Nehring and Walker 1996, Modin 1998, Sandell et al. 2001). M. cerebralis has a complex, two-host life cycle involving the aquatic oligochaete worm Tubifex tubifex, and most salmonids, which include trout, whitefish and salmon. Susceptibility to disease depends on species (MacConnell and Vincent 2002), fish age and size (Ryce et al. 2005), and parasite dose at time of exposure (Vincent 2002). Young trout, particularly RBT, are most vulnerable when infected at less than nine weeks of age (Ryce et al. 2004). Coincidence between this vulnerable period and the release of the infective triactinomyxon (TAM) stage of the parasite largely determines the degree of exposure for young fish and, ultimately, the magnitude of population-level effects. High mortality and recruitment collapse can occur in highly exposed populations (Nehring and Walker 1996, but ^ee Sandell etal. 2001). Environmental conditions play an important role in the distribution of infection and level of severity within and among streams of the Blackfoot Basin. In Cottonwood Creek, a tributary to the middle Blackfoot River, Smith (1998) initially identified a longitudinal distribution with tubifex worms and WD absent from upper glacial valleys but abundant tubifex worms and a high severity of disease present in lower-valley stream reaches. More recently. 149 water temperature, channel gradient and fine sediment were primary environmental predictors of WD presence within basin-fed streams of the Blackfoot Basin like Cottonwood Creek. Although WD has resulted in very large population declines of RBT in certain Montana (Madison River - Vincent 1996, Baldwin et al. 1998) and Colorado Rivers (Nehring and Walker 1996), population effects are regionally variable. Previous studies of RBT vulnerability to WD in Montana have focused on the tailwater fishery of the Madison River where trout spawn in side-channels (Downing et al. 1999, Krueger et al 2006). In western Montana, significant declines in RBT in Rock Creek, a large tributary to the upper Clark Fork River near Missoula, followed the introduction of WD (Montana Fish, Wildlife and Park, unpublished data). Likewise, RBT in lower Cottonwood Creek have declined 50% from pre-WD estimates (Peters 1990, Smith 1998, Pierce et al. 2006). Both Rock Creek and lower Cottonwood Creek have experienced community-level shifts toward brown trout (Montana Fish, Wildlife and Parks, unpublished data), a species with partial WD resistant. These observations suggest some risk of population, or community-level, changes in the middle Blackfoot River where RBT declines are now being detected near infected RBT spawning tributaries. Predicting WD effects on RBT populations in western Montana requires assessing the juxtaposition of streams with high vulnerability to infection and the location of spawning and rearing sites. □ S«nlirtel cagA * Wal«r tamfiaraujr« 1) Johnson Cr 2) West Twin Cr 3) Eft9t Twin Cr 4)OoldCr 5) Belmom Cr e) Bh Cr 7\ CotlonMood Cr \ 8J Montuf* Cr 9) Dunh^EnCr 10) Churnbciialii Cr 11) North Fork 14 21 2Shfn Figure 1. Study area: Blackfoot River Basin with sentinel cage, water temperature and Blackfoot River (USGS) discharge monitoring sites. Previous studies identify Blackfoot River RBT reproduction within tributaries (Peters and Spoon 1989, FWP unpublished data); however, the relative importance of tributary stocks has not been evaluated, nor has the influence and spatial extent of possible WD effects upon fluvial RBT of the Blackfoot River. To investigate these questions, we assessed the overlap of 150 fluvial RBT spawning sites with M cerebralis infection in 10 spawning streams. Our study objectives were to: 1) identify the spawning life-histories of fluvial adult RBT of the Blackfoot River; 2) identify the relative use of spawning tributaries by fluvial stocks of the Blackfoot River; and 3) identify disease severity in spawning streams using sentinel exposures of age-0 RBT. Our purpose is to assess disease risk for migratory RBT stocks for two reaches of the Blackfoot River, to gain a better understanding of fluvial RBT and to identify management measures that could buffer possible RBT declines within rivers of western Montana. Study Area The Blackfoot River, a 5* order tributary (Strahler 1957) of the upper Columbia River, lies in west-central Montana and flows west 212 km from the Continental Divide to its confluence with the Clark Fork River in Bonner, Montana (Figure 1). The River drains a 5,998-km2 heterogeneous watershed through 3,038 km of perennial streams and generates a mean annual discharge of 44.8m''s (United States Geological Survey 2006). The Blackfoot River flows freely to its confluence with the Clark Fork River where Milltown dam, a run-of- the-river hydroelectric facility, has blocked upstream fish passage to the Blackfoot River since 1907. The physical geography of the watershed is geo-structurally controlled and regionally variable with subalpine forests dominating the high mountains, montane woodlands at the mid- elevations and semi-arid glacial (pothole and outwash) topography on the valley floor. Primary tributaries of the upper Blackfoot River (upstream of the Clearwater River) flow through a broad upper valley and alluvial bottomlands. Downstream of the Clearwater River, mountains constrict the Blackfoot River to a narrow canyon. With some exceptions, tributaries enter the lower Blackfoot River through a mountainous area with conflned channels, steeper gradients and colder summer temperatures. WD was flrst detected in the middle Blackfoot Watershed in 1995 in lower Cottonwood Creek. Since then, the disease has increased in distribution and severity. WD now infects the entire mainstem Blackfoot River and lower reaches of many tributaries (Pierce et al. 2006). Tributaries to the Blackfoot River provide spawning and rearing for migratory RBT, as well as other WD susceptible salmonids including bull trout {Salvelinus confluentus), westslope cutthroat trout (O. clarki lewisi), mountain whiteflsh (Prosopium williamsoni), brown trout (Salmo trutta) and brook trout {Salvelinus fontinalis) (Pierce et al. 2006). Montana rivers are managed for a diversity of self-sustaining wild trout populations. Within the Blackfoot Basin, wild RBT are present at low-elevations, and population densities increase in the down-river direction (Pierce et al. 2006). Although RBT occupy only about 15% of the Blackfoot Basin, they comprise ~ 70% of the trout community in the lower Blackfoot River. Below the mouth of the North Fork, RBT contribute to a high-value recreational flshery for the mainstem Blackfoot River supporting an estimated 26,817 anglers in 2005 (Montana Statewide Angling Pressure Estimates 2005). For this study, we telemetered fluvial RBT from the Blackfoot River and examined life history and disease relationships up-and downstream of the mouth of the Clearwater River. The lower Clearwater River flows through a series of natural lakes causing high summer water temperatures (>27''C), and seems to support very little, if any, RBT reproduction (Peters 1990, Pierce et al. 2002). This river demarcates the mid-point of rainbow trout distribution within the Blackfoot River RBT distribution, and separates the Blackfoot Basin into two general sub- basins based on physical differences of tributaries. The lower reach of the Blackfoot River (Rl hereafter) extends from Clearwater River confluence 55.8 km downstream to the Blackfoot River confluence with the Clark Fork River. Except for the upper-most tributary to Rl (Elk Creek), RBT spawning tributaries originate in a mountainous region and tend toward smaller (second and third-order), higher gradient streams with colder summer temperatures. Conversely, Elk Creek, a low-gradient stream within an agricultural valley, supports elevated 151 summer temperatures and high instream levels of fine sediment (Pierce et al. 2006). The Blackfoot River between the confluence of the Clearwater River and North Fork Blackfoot River (R2 hereafter) has RBT spawning tributaries that are fewer but generally larger (third and fourth-order), flow within wider channels, have broader fioodplains with lower gradients and support warmer summer temperatures. An exception is the North Fork, a stream of wilderness origin that is larger and colder than all other R2 tributaries (Pierce et al. 2006). Methods Radio-telemetry - We assessed migration patterns, relative use of tributaries, timing of migration events and location of RBT spawning using radio-telemetry. Twenty-five RBT were captured in the lower Blackfoot River, phenotypically identified as RBT and implanted with continuous (12 hour on/off) Lotek'^'^ radio transmitters on 8 March 2004 («=4), between 28 February - 8 March 2005 («=10) and 7-22 March 2006 {n=\ 1) and tracked to spawning areas within tributaries. These fish ranged from 34.0 to 49.0 cm in total length (mean, 41.4) and from 408 to 1,270 g in weight (mean, 680). We selected larger "plump" female fish (based on absence of a kype) to increase the likelihood that telemetered fish were sexually mature, and to more accurately identify the timing and location of spawning events. Visual identification was later verified for the 21 of the 25 fish collected in 2005-06 through genetic analysis of fin clips using 17 fragments of nuclear DNA at the University of Montana, Trout and Wild Salmon Genetics Laboratory (Boecklen and Howard 1997). Transmitters were evenly distributed among fish in the lower 35.4 km of Rl («=12); whereas telemetered fish were captured only in a 6.4-km section in R2 (n=l3) due to shelf ice and limited river access. Fish were captured prior to spawning migrations (by electro-fishing) in suspected wintering pools. Individually coded transmitters weighed 7.7 g, had an estimated life of 450 days, emitted an individual coded signal, did not exceed 2 percent of fish weight (Winters 1997), and were implanted following standard surgical methods (Swanberg 1999). Technicians located telemetered fish on foot using a hand held three-element Yagi antenna or by truck using an omni-directional whip antenna. We located fish weekly prior to migrations, 2-3 times per week during migrations and spawning, once per week following spawning and generally once per month thereafter. We recorded upstream movements by river kilometer. We assumed fish spawned if they ascended a stream with suitable spawning habitats during the spring spawning period, and the upper-most location was the assumed spawning site. We estimated spawning dates as the median date between two contacts for a given event (i.e. spawning or migration) (Swanberg 1997). Peak spawning among spawners was identified as the median spawning date. We assumed the reach influenced by whirling disease extended from wintering locations to spawning sites. Water temperature and flows - Water temperatures and flows were measured in the Blackfoot River to assess their influences on RBT migrations. Thermographs (Onset^'^) were placed (2005-06) at rkm 12.7 at the U.S. Geological Survey gauging station (guage number: 12345000). We used both mean daily discharge and temperature to examine potential relationships with RBT movements. Onset thermographs were placed in lower Gold (2005-06) and Monture creeks (2004-06) where mean daily temperatures were calculated to identify relationships of tributary movements and spawning. To predict the timing of fry emergence for Gold and Monture Creeks, we calculated the incubation period using a 350°C degree-day span (Piper 1982), beginning at the estimated spawning date for each individual fish that spawned in Gold and Monture Creeks, and emergence was estimated at three weeks post- hatch. All thermographs recorded at 48-minute intervals. 152 Analyses of life histories - To test the potential influence of introgression on movement patterns, we compared the start date of migration and the total pre-spawning migration distance between hybrids, and "pure" RBT using Mann-Whitney rank sum tests. For the total group, we used linear regressions to assess potential associations between the start date of migration and distance to spawning sites; the total duration (days) and total distance (km) of migrations ; and the date spawners returned to the River and the total migration period (days). We used Kruskal-Wallis (ANOVA) on ranks to assess tributary size (i.e. stream-order) and the date of entry to a spawning stream and days spent within a tributary. For reach-stratified spawners, we used Mann Whitney rank sum tests to analyze the start of migration, dates RBT entered tributaries and the upstream distance to spawning sites upon entering a tributary, estimated spawning dates and dates RBT exited tributaries. All tests were evaluated at the a = 0.05 level of significance. WD infection and severity - We conducted sentinel exposures of 50 hatchery RBT fry (age-0 cohorts) at known RBT spawning sites in 10 streams to identify WD severity in individual streams and the spatial variation of M cerebralis among tributaries (Figure 1). These fish were exposed at 98-103 days post-hatch at mean length of 36mm in 2005 and 45mm in 2006. Exposures were completed in July within 9 weeks of the estimate post-hatch period for wild fish. This timing coincides with high RBT susceptibility (Ryce et al. 2004), estimated emergence of wild RBT fry and the corresponding peak TAM production period within rivers of western Montana (Vincent 2000) including the Blackfoot Basin (Fish, Wildlife and Parks, unpublished data). Flow (M^S) Temperature (C) Figure 2. Blackfoot River: Mean daily flow (left axis - gray line) and mean daily water temperatures (right axis - dark line) during rainbow trout spawning migration. The total migration period is shown by the arrowed horizontal lines. The vertical arrows show median dates spawners entered and exited tributaries. 153 The exposure period for each live cage was standardized at 10 days. At the end of that time, fry were transferred to Pony, MT, where they were held for an additional 80 days at a constant 10 ° C to ensure that WD, if present, would reach maximum intensity (Vincent 2000). At the end of the holding period, all surviving fish were sacrificed and sent to the Washington State University Animal Disease Diagnostic Laboratory at Pullman, WA. At the lab, fish heads were examined histologically and scored using the MacConnell-Baldwin grading scale, which ranks whirling disease from (absent) to 5 (severe) (Baldwin et al. 2000). Sentinel exposures were considered severe if a majority (%) of exposed RBT had histological (lesion) scores of >3 on the MacConnell-Baldwin scale. Lesion scores >3 are determined by severe cartilage damage and a dispersed inflammatory response that occurs in infected fish (Baldwin et al. 1998). Results Migratory life histories and spawning - For 25 telemetered RBT, we made a total of 1,594 contacts with an average of 64 contacts (range: 12-129) per fish. All 25 RBT were successfully tracked to spawning tributaries from Temperature (C) March 2004 to December 2006 (Table 1). Fourteen of twenty fish that underwent genetic analysis tested as post-Fi RBT hybrids with westslope cutthroat trout having a predominant rainbow trout genetic contribution; the remaining six tested as genetically unaltered rainbow trout (Leary 2005, 2006). Four migrants captured in 2004 that later entered Monture Creek («=3) and the North Fork {n=\) were untested. There were no significant differences between hybrid and "pure" RBT for either start (date) of migration (Mann Whitney, P=0.78) or the total pre-spawning distance moved (Mann Whitney, P=0.56). River temperatures and flows incrementally increased during the (2004- 2006) RBT pre-spawning migrations. In these years, migrations began between 19 March and 15 April on the rising limb of the hydrograph as mean daily temperatures approached 5°C (Figure 2). With the onset of 3/1 3/31 4/30 5/30 6/29 7/29 3/1 3/31 4/30 5/30 6/29 7/29 Figure 3. Water temperatures for Gold (top) and Monture Creeks in 2005 (gray) and 2006 (black). Duration within tributaries (2004-06) and the estimated emergence periods for 18 spawners are shown by arrowed left and right horizontal lines, respectively. The median spawning date for the tributary is shown by vertical arrows. 154 migration, twenty-four RBT moved up-river and one moved down-river. In nine days telemetered RBT traveled a median of 6.8 rkm to their respective spawning tributary. RBT from Rl moved a (median) distance of 10.0 rkm (range 0.5 - 56.8) compared to 6.6 rkm (range 2.7-21.4) for R2. For the total group, there was no relationship between the date migrations began and the total distance to spawning sites (linear regression, R^=0.008, P=0.89). However, RBT with longer pre-spawning distance (start locations and spawning sites) underwent migrations of longer duration (linear regression, R^=0.20, P=0.04), and RBT returned to the river later than fish exhibiting movements of shorter duration (linear regression, R^=0.36, P=0.003). Spawners spent an average of 17 days (range, 3-63) in tributaries and ascended a median of 3.0 km (range, 0.2-19.8) to their spawning grounds where they held for an average of six (range 1-14) days before returning to the Blackfoot River. We observed that R2 fish migrated significantly farther up tributaries (median, 7.1 versus 1.0 km) to spawning sites than Rl fish (Mann Whitney, P=0.005). Based on the distance between winter pools and spawning sites, fish moved a (median) distance of 12.1 rkm for the total group, and a median of 10.6 (range 1.1 - 63.2) rkm forRl fish compared to 12.6 (range 6.0 -27.5) forR2 fish. Migration events began slightly earlier and ended later for R2 fish although these differences were not statistically significant. RBT from the R2 began their migrations eight days earlier (median, 9 April versus 17 April; Mann Whitney, P=0.17), entered tributaries nine days earlier (median, 17 April versus 26 April; Mann Whitney, P=0.10) and spawned six days earlier (median, 28 April versus 4 May; Mann Whitney, P =0.40). However, the duration of tributary use was five days longer for R2 fish (median, 17 days versus 12 days), and fish exited \( 11 ^\ y 19 R*ach 1 R*ach 2 Ststt of mJgratkptt « u \ 1^ Si 2Si(m Figure 4. Start of migrations (open symbols) and upstream -most location (closed symbols) of spawning rainbow trout. tributaries six days later (median, 15 May versus 9 May; Mann Whitney, P=0.24) than Rl fish. 155 RBT spawned in six tributaries ranging from I"'' to 4* order with the Monture Creek watershed and Gold Creek supporting the highest proportion of spawners (n=l2 or 48%) and (n=5 or 20%) respectfully (Table 1). Fish from Rl spawned in four tributaries: Gold Creek (n=5), Belmont Creek («=4), East Twin («=2) and Monture Creek («=1); whereas R2 fish spawned in Monture (n=lO) and its tributary Dunham Creek (n=2), and only one spawning outside of the Monture Creek Basin, within the North Fork. Based on stream-order, spawners entered larger tributaries earlier than smaller tributaries (ANOVA, P=0.02). However, there was no significant difference with stream-order and time spent in tributaries (ANOVA, P=0.20), During the period of pre-spawning river migration (19 March to 15 April), mean water temperatures in the Blackfoot River were higher in 2005 (5.6°C) than 2006 (4.9°C). Thirteen RBT entered Monture Creek and five entered Gold Creek at mean water temperatures of 5.6 (range 3.6-8. 1°C), and RBT spawned at mean temperatures of 5.2 (range 3.4-8.0 °C) in these drainages. After spawning, all fish with active radios (n=24) exited the tributaries. Three of 24 (12%)) post-spawners (fish: 2, 9 and 25) moved downstream of Milltown Dam into the Clark Fork River during peak flow (May and July), including two spawners from Gold Creek and one that moved downriver >74-km after spawning in Monture Creek. However, the majority (n=\S or 76%) either returned to (n=9), or were within 1.6 km (n=9) of their original start locations; three (12%) moved downriver a mean of 14.0 km (range, 4.3-23.7) from their starting locations. Reach and fish# Start of river migration Pre-spawning river migration Tibutary spawning End of Migration Total km to Estimated Date Date returned to River km at rl grade 3) and identified only the North Fork with a low severity. The percent of Blackfoot River fish with high severity (> grade 3) was 43% in Rl compared with 66% in R2. Discussion A similar study east of the Continental Divide in Montana investigated RBT spawning life history and risk to juvenile survival within an infected "tailwater" section of the Madison River (Downing et al. 1999). By contrast, our study, undertaken west of the Continental Divide within a headwater basin of the upper Clark Fork drainage, examined fluvial life history within a "free-flowing" river system. Common to both areas are predictable migratory strategies involving pre-spawning migrants holding within wintering areas prior to upriver movement; the fidelity of most post-spawners to their initial tagging location; upstream migrations of similar distances (mean, 14.5 versus 18.7 km) to spawning grounds; and fry emergence by early July during the vulnerable, highly infectious period (Downing et al. 1999, FWP unpublished data). Life history differences between the Madison and Blackfoot sites involve primarily mainstem spawning within the Madison River compared to tributary spawning within the Blackfoot Basin. Although WD has been shown to vary within the mainstem of the Madison River (Downing et al. 1999, Krueger et al 2006), our study identifies large differences in histological scores among spawning tributaries. Like the Madison River, TAM production in the Blackfoot River express a seasonal peak in June and July, followed by declines by September (Downing et al. 1999, Montana Fish, Wildlife and Parks, unpublished data). This seasonal pattern of high TAM release overlaps with the emergence of wild RBT fry at the critical early life stage. In Blackfoot tributaries prone to high TAM production and high incidence of infection, the epizootic has been both rapid and severe. At category 3.0 severity, granulomatous lesions can be large and severely impact bone, causing distortion and breakage, which leaves the fish weak, less able to compete for food and habitat and ultimately increases chances of mortality. Mean lesion scores of >2.75 have been associated with significant levels of mortality in wild rainbow trout populations (Vincent 2002). Infections in Elk Creek increased from non-detectable to a mean lesion score of 2.8 within a single year (2002 to 2003) before increasing to 4.8 identified in this study. Likewise, infections in Monture Creek increased from non-detectable to a mean lesion score of 3.2 between 1999 and 2002 (Pierce et al 2006) before increasing to 4.8 in this study. M. Cerebralis infections to high severity in the middle Blackfoot River coincide with a temporal trend (1998-2004) of increased cranial deformities (a clinical sign of WD infection) and recent declines in RBT abundance in the Blackfoot River downstream of the Monture Creek confluence (Pierce et al 2006). Similar to spatial variability of infected waters within a Utah watershed (Hoz Franco and Budy 2004), infections within the Blackfoot Basin vary geographically depending on the 157 physical properties and arrangement of tributaries. For basin-fed tributaries within the Blackfoot Basin, conditions correlated with infection include wide alluvial valleys with warm water during summer, lower stream gradients and higher levels of fine sediments (Pierce et al. In review); all are conditions that favor habitat for T. tubifex or production of TAMs (Amdt et al. 2002, El-Matbouli et al. 1999). For Rl fish, spawning was dispersed in lower reaches of three morphologically similar (cold, high-gradient) tributaries to the lower Blackfoot Basin. Within this group of similar streams are three additional tributaries (Johnson, East Twin and Bear Creeks), all of which support known (Peters and Spoon 1989, FWP unpublished data) but limited RBT spawning (based on our telemetry findings) and low to no measurable infection. From Belmont Creek (rkm 35.4) downriver, this concentrated group of relatively "clean" tributaries enters the Blackfoot River at a mean interval of one stream per 5.8 km of river. Of 11 RBT tracked from wintering pools to tributaries within this area, ten expressed unidirectional (upstream) migration over a 9.2 km median distance to spawning areas including 8.7 km of the lower Blackfoot River. Unlike the upper reach, these movement patterns identify several overlapping spawning stocks, which cohabit this reach of the Blackfoot River. Sentinel exposures within this group of Rl tributaries consistently test at low levels (< grade 3 or non-detect) of severity compared to R2 tributaries where sentinel exposures consistently rank at high (> grade 3) severity (Pierce et al 2006). The RBT densities in the lower Blackfoot River remain stable (Pierce et al. 2006) despite an apparent annual loss of -15-20% of lower River RBT spawners over Milltown Dam identified in this study. In contrast to this Rl river area, the 51.5-km reach of the Blackfoot River between Belmont Creek and the North Fork (R2) contain fewer (five) RBT spawning streams - one per 10.3 km of river although most (four) enter within a 17.8-km section of Blackfoot River between Cottonwood Creek (rkm 69.2) and the North Fork (rm 86.9). Consequently, RBT recruitment sources within the 33.8-km section of the Blackfoot River between Belmont and Cottonwood Creek are limited. Elk Creek enters this reach but this stream is water quality (sediment and temperature) impaired (Blackfoot Challenge 2006), supports high severity of WD, and has experienced RBT declines in recent years (Pierce et al 2004). Of the five RBT spawning streams upstream of Belmont Creek, only the North Fork supports a low severity of WD, yet it supports limited RBT reproduction {this study) and recruits relatively fewer age-0 RBT to the Blackfoot River than downstream tributaries (Peters and Spoon 1989). Age-0 RBT abundance has been longitudinally evaluated in all RBT spawning streams during the early rearing mid-summer period (Peters and Spoon 1989, Peters 1990, Pierce et al. 2004, 2006). Juvenile inventories identify relatively high abundance of age-0 RBT within and downstream of all central spawning grounds and concentrated densities extend to the Blackfoot River below the mouths of all spawning tributaries identified in this study (Peters and Spoon 1989), a pattern of rearing consistent with the Madison River (Downing et al. 2002). For the Blackfoot Basin, this pattern of limited early dispersal suggests a higher risk of disease exposure throughout the lower reaches of most R2 tributary and mainstem rearing areas, but conversely low risk in Rl tributaries (except Elk Creek) and those fry dispersing to the mainstem Blackfoot River during the summer period when WD severity ranked high. Prior to the invasion of M cerebralis, Peters and Spoon (1989) identified Monture Creek as a primary source of RBT recruitment, but considered the middle Blackfoot River as recruitment limited. Our study confirmed this spawning relationship with >90% of telemetered R2 fish spawning within the lower Monture Basin with a central (median) spawning location of rkm 6.9 (range 0.3-19.8). Although the 2005 Monture Creek sentinel exposure identified a severe (97% > grade 3) exposure, the cage was located downstream (rkm 3.2) of the central spawning site. To clarify disease severity within the central spawning area, we further examined M. cerebralis exposures at rkm 7.4 and upstream of identified RBT spawning areas 158 (rkm 20.8) with additional sentinel exposures in 2006. Exposure results confirmed the high severity at the central spawning areas (95% > grade 3), but detected no upstream infection. The combined 2005-06 exposure results confirm risk of severe exposure within and downstream of primary Monture Creek spawning areas, yet the upstream attenuation to no infection suggests an upper segment of the Monture RBT spawning site remains at a low level of risk. The discrepancy between river migration distances in Rl an R2 (6.6 verses 10.0) raises concerns of disease-related recruitment losses in R2. In addition to a reduced level of river use, high lesion scores at the primary RBT spawning site (Monture Creek) indicate potential for a synergistic reduction in R2 recruits, including fish dispersing to downstream waters where trout populations are currently limited by the low number and poor quality of existing spawning streams (i.e. upstream of Belmont Creek). Oncorhynchus resistance to pathogens such as whirling disease can take many forms such as inherent life history strategies that help avoid exposure of M cerebralis at early life stages, or physiological resistance such as an innate immune response that limit the pathogen from infecting the host (MacConnell and Vincent 2002). Similar to nearby stocks within the Clark Fork Basin, a majority offish identified as RBT in the Blackfoot Basin were found to be mildly introgressed with westslope cutthroat trout. Physiological resistance of RBT/ westslope cutthroat trout hybrids to WD is untested, but it is possible that Fi hybrids may have an intermediate level of resistance between the low resistance of non-hybridized RBT and the "moderate" resistance of non-hybridized westslope cutthroat trout (MacConnell and Vincent 2002, Hendrick et al 1999). Within the Blackfoot watershed, the longitudinal distribution from pure westslope cutthroat trout predominant in the upper Blackfoot Basin to a more RBT- dominated community downstream of the North Fork suggests an inter-specific reduction in WD susceptibility among the Oncorhynchus community, particularly when further considered within a context of migratory life histories and environmental factors that influence infection (and severity) along a longitudinal continuum (Smith 1998, see prediction paper) Management ImpUcations - Management implications vary by river reach and involve the potential for an additive loss of recruitment to the middle Blackfoot River, and the need to offset this loss by correcting anthropogenic degradation of spawning and rearing streams. The middle Blackfoot River (upstream of the Belmont Creek) was previously identified with trout recruitment problems brought on by drought and winter mortality, limited spawning areas and degradation of existing spawning and rearing areas caused by agricultural and other land uses. For the middle Blackfoot River, this study and other tributary assessments suggest abundant restoration opportunities even in tributaries that host high levels of disease. Based on community-level changes in Rock Creek, brown trout clearly have potential for expansion under environments prone to WD. This naturally more resistant species has also shown significant population increases in highly infected spring creeks within the middle Blackfoot Basin once limiting factors related to physical habitat were corrected (Pierce et al. 2006). Like brown trout, native westslope cutthroat trout and bull trout could thrive within certain infected environments. While both westslope cutthroat trout and bull trout possess partial resistance to WD (MacConnell and Vincent 2002), both species also possess life history strategies that help avoid exposure of M cerebralis at early life stages by spawning in headwaters of the Blackfoot Basin (including Monture Creek) where contact with M. cerebralis at critical stages (age-0) is reduced. Young cutthroat trout and bull trout migrate to infected waters at more disease-resistant (age-1 and older) stages. Both species migrate extensively within the Blackfoot Basin (Swanberg 1997, Schmetterling 2001, Pierce et al. 2007), including infected sections of the Blackfoot River prone to limited RBT recruitment (Pierce et al 2006). 159 Even moderate levels of WD resistance for certain native species can temper population effects within waters that support severe WD. One example of this is Chamberlain Creek, a tributary supporting primarily westslope cutthroat trout. Following remediation of dewatering, ditch entrainment, riparian grazing and channel alterations, westslope cutthroat trout densities in lower Chamberlain Creek increased from two to 80 fish/lOOm by 1994 and remained stable thereafter (Pierce et al. 1997, 2006). After this recovery, telemetered adult fluvial cutthroat from the Blackfoot River identified Chamberlain Creek as an important westslope cutthroat trout spawning stream to the lower Blackfoot River (Schmetterling 2001). Densities of westslope cutthroat trout have remained stable in lower Chamberlain Creek despite a high severity of WD (range of mean lesion scores, 2.7 to 4.3) between 1999 and 2005. Population trends for fluvial westslope cutthroat trout in the lower and middle Blackfoot River have been stable despite being epizootic among rainbow trout. Conclusions - Although future population (and community) effects are difficult to predict, our study clearly indicates disease risks to Blackfoot River RBT vary from the tributary to sub-basin scale. Our study suggests the middle Blackfoot River is at a higher risk of RBT recruitment loss through WD, perhaps at levels sufficient to affect angling success. Some highly infected valley-floor streams in the middle Blackfoot Valley seem predisposed to high WD because of their low gradient, high water temperatures and high sediment levels and synergistic effects of heavy grazing and other disturbances. By contrast, higher gradient mountain streams are less prone to infection. To offset potential RBT losses in disease prone waters of the middle Blackfoot Basin, stakeholders must 1) better manage riparian areas for channel stability, increased shade and erosion reduction, 2) promote native fish recovery and migratory life histories, and 3) restore (or enhance) habitats favoring salmonid life stages less affected by the WD pathogen. Acknowledgements Northwestern Energy, the Big Blackfoot Chapter of Trout Unlimited and U. S. Fish and Wildlife Service-Partners for Fish and Wildlife provided partial funding for this project. We also extend thanks to the landowners including the Two Creeks, Knob and Kettle, Paws Up and Heart-bar-Heart Ranches for allowing access to their lands. Pat Saffel and Robb Leary with Montana Fish, Wildlife and Parks and Ryen Aasheim with the Big Blackfoot Chapter of Trout Unlimited for their assistance with the project. Lisa Eby, Eileen Ryce and Pat Byorth reviewed and improved the quality of the manuscript. Literature Cited Anderson, R., A. 2004. Occurrence and seasonal dynamics of the whirling disease parasite, Myxobolus cerebralis, in Montana spring creeks. Master of Science thesis, Montana State University, Bozeman. Arndt, R.E., E. J. Wagner, Q. Cannon, and M. Smith. 2002. Triactinomyxon production as related to rearing substrate and diel light cycle. Pages 87-91 in J.L. Bartholomew and J.C. Wilson, editors. Whirling disease: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda, Maryland. Baldwin, T. J., E. R. Vincent, R. M. Silflow, D. Stanek. 2000. Myxobolus cerebralis infection in RBT (Oncorhynchus mykiss) and brown trout (Salmo trutta) exposed under natural stream conditions. Journal of Veterinary Diagnostic Investigations 12:312-321. Baldwin, T. J., J. E. Peterson, G. C. McGree, K.D. Staigmiller, E. S. Motteram, C. C. Downs and D. R. Stanek. 1998. Distribution oi Myxobolus cerebralis in salmonid fishes of Montana. Journal of Aquatic Animal Health 10:361-371. Blackfoot Challenge. 2005. A basin-wide restoration action plan for the Blackfoot Watershed. Boecklen, W. J. and Howard, D. J. 1997. Genetic analysis of hybrid swarms: numbers of markers and power of resolution. Ecology 78 (8) pp. 261 1-2616. 160 Downing, D.C., T. E. McMahon, K.L. Kerans and E.R. Vincent. 2002. Relation of spawning and rearing of rainbow trout and susceptibility to Myxobolus cerebralis infection in the Madison River, Montana. Journal of Aquatic Animal Health 14:191-203. El-Matbouli, M., T.S. McDowell, D.B. Antonia, K.B. Andree, and R.P. Hendrick. 1999. Effect of water temperature on the development, release, and survival of triactinomyxon stage of Myxobolus cerebralis in its oligochaete host. International Journal for Parasitology 29 : 627-64 1 . Goldberg, T. L., E. C. Grant, K. R. Inendino, T. W. Kassler, J. E. Claussen, D. P. Phillip. 2005. Increased infection disease susceptibility resulting from outbreeding depression. Conservation Biology 19: 455-462. Hendrick, R. P., M. El-Matbouli, M. A. Adkinson, and E MacConnell. 1999. Susceptibility of selected inland salmonids to experimentally induced infections with Myxobolus Cerebralis, the causative agent of whirling disease. Journal of Aquatic Animal Health 11:330-376. Leary R. 2005, 2006. Rainbow trout genetics lab reports. Wild Trout and Salmon Genetics Laboratory, Division of Biological Sciences, University of Montana, Missoula. Kerans, B.L. and A.V. Zale. 2002. The ecology oi Myxobolus cerebralis. Whirling disease: reviews and current topics. American Fisheries Society Symposium 29: 145-166. Krueger, R.C., B.L. Keams, E.R.Vincent and C. Rasumussen. 2006. Risk of Myxobolus cerebralis infection to rainbow trout in the Madison River, Montana, USA. Ecological Applications, 16:770-783. MacConnell, E. and E. R. Vincent 2002. Review: the effects o^ Myxobolus cerebralis on the salmonid host. Pages 95-108 in J. L. Bartholomew and J. C. Wilson, editors. WD: reviews and current topics. American Fisheries Society, Symposium 29, Bethesda Maryland. Modin, J. 1998. Whirling disease in California: A review of its history, distribution, and impacts, 1965-1997. Journal of Aquatic Animal Health 10:132-142. Montana Fish, Wildlife and Parks. 2006. Statewide angler pressure surveys for 2005. MWDTF (Montana Whirling Disease Task Force). 1996. Final report and action recommendations. Montana WD Task Force, Helena, MT. Nehring, B., and P. G. Walker. 1996. Whirling disease in the wild: the new reality in the intermountain west. Fisheries 2\ (6): 28-30. Peters, D.J. and R. Spoon. 1989. Preliminary inventory of the Big Blackfoot River. Montana Department of Fish, Wildlife and Parks, Missoula, Montana. Peters, D. 1990. Inventory of fishery resources in the Blackfoot River and major tributaries to the Blackfoot River. Montana Department of Fish, Wildlife and Parks, Missoula, Montana. Pierce R. and C. Podner. 2000. Blackfoot River fisheries inventory, monitoring and restoration report. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce R., C. Podner and J. McFee. 2002. Blackfoot River fisheries Restoration progress report for 2001. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce, R., R. Anderson and C. Podner. 2004. The Big Blackfoot River Restoration Progress Report for 2002 and 2003. Montana Fish Wildlife and Parks, Missoula Montana. Pierce, R., R. Aasheim and C. Podner. 2005. An integrated stream restoration and native fish conservation strategy for the Big Blackfoot River basin. Montana Fish Wildlife and Parks, Missoula, Montana. Pierce, R., R. Aasheim and C. Podner. 2007. Fluvial westslope cutthroat trout movements and restoration relationships in the upper Blackfoot Basin, Montana. Intermountain Journal of Sciences Vo\. 13(2). Piper, R. G 1982. Fish Hatchery Management. USDI, United States Fish and Wildlife Service. Report 329-150. 161 Ryce, E. K. N, A. V. Zale and E. MacConnell. 2004. Effects offish age and parasite dose on the development if whirling disease in rainbow trout. Diseases of Aquatic Organisms Vol. 59(3):225-233. Ryce, E. K.N, A. V. Zale, E. MacConnell and M. Nelson. 2005. Effects offish age versus size on the development of whirling disease in rainbow trout. Diseases of Aquatic Organisms, Vol. 63 (1): 69-76. Sandell, T. A., H. V. Lorz, D. G. Stevens, and J. L. Bartholomew. 2001. Dynamics of Myxobolus cerebralis in the Lostine River, Oregon: implications for resident and anadromous salmonids. Journal of Aquatic Animal Health 13: 142-150. Schmetterling, D. A. 2001. Seasonal movements of fluvial westslope cutthroat trout in the Blackfoot River drainage, Montana. North American Journal of Fisheries Management 21: 507-520. Shepard, B. B., B.E. May and W. Urie. 2003. Status of westslope cutthroat trout (Onchorynchus clarki lewisi) in the United States: 2002. A report to the Westslope Cutthroat trout Interagency Conservation Team. Smith, L. 1998. Study on the distribution and abundance of Tubifex tubifex within Cottonwood Creek in the Blackfoot drainage. Masters Thesis, University of Montana, Missoula, Montana. Strahler, A.N. 1957. Quantitative analysis of watershed geomorphology. Transactions, American Geophysical Union 38:913-920. Swanberg, T. R. 1997. Movements of and habitat use by fluvial bull trout in the Blackfoot River. Transactions of the American Fisheries Society 126: 735-746. Swanberg, T. R., D. A. Schmetterling, and D. H. McEvoy. 1999. Comparison of surgical staples and silk sutures for closing incisions in RBT. North American Journal of Fisheries Management 19:215-218. USGS 2006. Gauging station 1234000 provisional unpublished data. Vincent, E. R. 1996. Whirling disease and wild trout: the Montana experience. Fisheries 21 (6):32-33. Vincent, E. R. 2000. Whirling disease report 1997-98. Montana Fish, Wildlife and Parks. Project 3860. Helena, Montana. Vincent, E. R. 2002. Relative susceptibility of various salmonids to WD with emphasis on rainbow and cutthroat trout. Whirling Disease: reviews and current topics. American Fisheries Society Symposium 29:109-115. Winter, J. D. 1996. Advances in underwater biotelemetry. Pages 555-590 m B.R. Murphy and D. W. Willis, editors. Fisheries Techniques, 2"'' edition. American Fisheries Society, Bethesda, Maryland. Zendt, J. S. and E. P. Bergersen. 2000. Distribution and abundance of the aquatic oligochaete host Tubifex tubifex for the salmonid WD parasite Myxobolus cerebralis in the upper Colorado River basin. North American Journal of Fisheries Management 20:502-5 12. 162 Status review of Mountain Whitefish {Prosopium williamsoni) in the Blackfoot Basin: A pilot study to help identify risk of whirling disease A report to the Whirling Disease Foundation Introduction Mountain whitefish - Prosopium williamsoni - (MWF) is a salmonid endemic to the Pacific Northwest of both the U.S and Canada. Native to western Montana, they are found primarily in cold, medium-to large rivers and in some lakes and reservoirs, and their distribution extends east and west of the Continental Divide. West of the Divide, they range throughout the upper Clark Fork and Flathead Basins. East of the Divide, their range extends throughout the headwaters of both the upper Missouri and Yellowstone Basins. Despite their generally ubiquitous presence in the river systems of western Montana, the life histories and population status of MWF have not been fully documented, nor has the vulnerability of MWF to whirling disease been fully investigated. To help document the status of MWF and to begin to identify relationships of MWF to whirling disease in the Blackfoot Basin, Montana Fish, Wildlife and Parks (FWP) compiled all available historic fish population survey information within the Blackfoot Basin into a (GIS) database and began (in 2006) more targeted surveys of MWF within the Blackfoot River. This status review and related fieldwork are running concurrent with plans for controlled laboratory exposures of MWF fry to Myxobolus cerebralis followed by histological examination of infected fish. If successful, laboratory tests will help identify the age and size of susceptible fry and develop measures of disease severity. The testing of MWF using sentinel exposures in the field is expected in the near future. Only one laboratory test has focused on the susceptibility of MWF to whirling disease (MacConnell et al. 2000). These researchers found when exposed within seven weeks of life to a high dose of TAMs, MWF experienced direct and rapid mortality. Other MWF exposed at lower doses survived but developed the clinical signs of whirling disease (blacktail, whirling behavior and skeletal (caudal) deformities). This study concluded that exposed MWF that were susceptible to infection by M. cerebralis, could develop whirling disease, and could serve as host for developing of M. cerebralis myxospores. This study observed that caudal lesions were prevalent in infected whitefish, and that these closely resembled lesions found in wild juvenile mountain whitefish collected from the Madison River in 1999. Certain aspects of the study were inconclusive because of an unrelated level of high MWF mortality during testing. In addition to early lab results, field-based research and anecdotal reports likewise indicate MWF may have a high prevalence of M cerebralis infection and could suffer population-level impacts. Whirling disease has been detected in MWF in the Salt River of Wyoming (Gelwicks and Zafft 2000). Barry Nehring of Colorado Division of Wildlife reported a 70% to 80% prevalence ofM cerebralis infection among wild MWF of the Roaring Fork River, Colorado. In Mission Creek, Montana, biologists from the Confederated Salish and Kootenai Tribes recently reported clinical signs of whirling and caudal deformities in MWF (Craig Barfoot, personal communication). Likewise, caudal deformities in juvenile MWF were recently detected within infected waters of the middle Blackfoot River of Montana. One local example of possible population declines within the Blackfoot Basin appears to be the recent loss of MWF from Hoyt Creek, a small spring creek tributary to Monture Creek. In 1992 prior to the introduction of whirling 163 disease to waters of the Blackfoot Basin, juvenile MWF were identified as common in lower Hoyt Creek; however in 2006 following the local escalation of whirling disease, MWF were absent from the same Hoyt Creek sampling location. Infected spring creeks like Hoyt Creek have been shown to support continuously high TAM production during the early MWF rearing period (i.e., from winter through early summer; Anderson 2004, R. Pierce, unpublished data). Infected basin-fed tributaries however show variable infection levels during the early summer depending on the environmental properties (e.g. water temperature) of individual streams (Pierce et al., in review). MWF Status summary MWF Distribution and WD overlap Understanding potential MWF disease relationships requires understanding the distribution and basic life history of MWF with emphasis on the vulnerable juvenile life-stages. Fish population surveys conducted within the Blackfoot Basin between 1989- 2006 identified the presence of MWF from the confluence of the Blackfoot River upstream -125 river miles and present at the lower elevations with -25 of the larger tributaries (Figure 1, Table 1). This distribution identified MWF mostly in the larger streams of basin-fed origin as well as the lower reaches of connected tributaries including several smaller spring creeks like Hoyt Creek, all of which are located in streams within the lower-valleys of the Blackfoot Basin. This distribution overlaps closely with high infection rates based on sentinel exposures. Our review of the historic MWF information identifies at a basin scale primary YOY rearing areas within the middle Blackfoot Basin from Elk Creek to Arrastra Creek and within the lower reaches of nearby tributaries (Figure 2). This distribution pattern Figure 1 and 2. Fish populations survey sites (1989-2007) where the presence of MWF is documented (top) and YOY abundance classes (bottom). 164 overlaps closely with the known distribution of whiriing disease including a large degree of spatial overlap with high severity of disease with rainbow trout. Basic life history - MWF are long-lived and possess some life history variation that often involves movement between habitats at multiple life stages. Spawning migration and spawning areas are highly variable between regions (Northcote and Ennis 1994). Although not well documented, spawning migrations often range from 10-30km (Northcote and Ennis 1994), however, spawning migrations >60 km have been identified (Davies and Thompson 1976). Migratory fish seem to undergo a complex sequence of seasonal movements beginning with passive dispersal of fry, followed by late summer movements to deeper water and higher velocity feeding habitats and autumn migrations to downriver over-wintering habitat. Spawning migrations of river populations are often in an upstream direction, although downstream spawning migrations from summer foraging areas to spawning locations in lower reaches of larger tributaries or into main- stem of rivers have also been documented (McPhail and Troffe 1998). MWF are long- lived and usually reach sexually maturity by the age of six. Fecundity is a function of female body size, thus larger females produce more eggs than smaller females, ranging from 1,400 to 24,000 eggs in Montana females (Brown 1952). MWF seem to use a wide range of habitats for spawning, and no spawning site preparation (redd construction) occurs by females (McPhail and Troffe 1998). Instead, MWF often spawn in (small) groups and eggs are broadcast over the substrate in riffles or rapids in late fall or early winter. Egg collection in western Montana by FWP hatchery personnel in the fall of 2007 indicate November as a primary spawning period. Spawning occur at temperatures below 6*'C, incubation requires 250-280 (°C) temperature units and emergent MWF fry were recently detected on March 15, 2008 in the upper Madison (Dick Vincent, FWP unpublished data). According to Northcote and Ennis (1994), throughout their life MWF progressively move to faster and deeper waters as body size increases. Fry emergence occurs in spring at which time sac fry seek out side-channels or protected backwaters along stream margins (Brown 1952). Fry leave these habitats by early summer and passively disperse downstream to protected areas where fish school, before further dispersing to deeper sections of stream during summer. Consistent with this movement pattern, in summer 2006, FWP and Dr. Lisa Eby undertook a targeted YOY survey in a known spawning area in Rattlesnake Creek (a tributary of the Clark Fork River near Missoula) where they detected very low densities of YOY. However, YOY were observed nearby in relatively high abundance in riffles of a much larger river (the lower Blackfoot River), suggesting a run-off-related out-migration of YOY although high densities of YOY have been identified in the lower reaches of tributaries during summer as well (Figure 2). This general pattern of early downstream dispersal is consistent with trapping studies in tributaries to the Flathead River where YOY out-movements were identified during the runoff period (Craig Barfoot, personal communication). Older fish prefer pools but are often associated with runs (riffle breaks) and riffles for foraging areas slightly upstream of pools or in deeper depressions or quiet areas associated with the downstream side of woody debris. MWF survey in the Blackfoot River: Wales and Canyon Creek sections Based on past electro-fishing observations, MWF are identified as common throughout the mainstem Blackfoot River downstream of Lincoln (rm 108); however, as a non-target species, quantification of mainstem populations has not been a priority in the past. To assess MWF sampling techniques and develop a monitoring baseline for the Blackfoot River, MWF recently (in 2006) targeted two population surveys in the 165 Blackfoot River under differing flow conditions. One survey was completed in May during the peak of runoff in the middle Blackfoot River (Wales Creek Section), and the other was completed in the upper Blackfoot River (Canyon Section) in September during base-flow conditions. Both surveys used mark-and-recapture and identical drift boat boom-mounted electrofishing methods. These surveys emphasized estimates of population densities and size structure metrics (length-weight and age-and-growth). Survey Results - The spring survey in the Wales Creek section generated a very low capture efficiency and produced an unreliable estimate of population density. Conversely, the fall survey in the Canyon Section resulted in much higher capture efficiency and a more reliable density estimate. A comparison of these density estimates and related statistics for MWF over two years of age (>8.0") for both sections is located in Table 2. Table 2. Comparison of mark-and-recapture survey results for MWF (>8.0") in two sections of the Blackfoot River. The Wales Creek is a spring estimate and the Canyon Section is a fall estimate. River-mile Date Section SizeCiass Efficiency Estim/1000' Stream mid-point Sampied Length (ft) Species (in) Marl(ed Captured Recaptured (R/C) Totai Estim t Ci tCi l/ales Creek Section 63 20-lay-02 7603 IWF >aO 74 77 4 005 1169 + 923 154 + 119 CanyonSection 95,3 2tSep-06 5422 IWF >8,0 177 121 23 0,19 904 + 324 167 + 59 Weight-length and age-size class assessment - Length frequency and weight- length scatter grams and condition factor plots for both sites are presented in Figures 3, 4 and 5, respectfully. The data indicates the upper Blackfoot River supports a "top-heavy" population, particularly in the upper river (Canyon Section). Condition factor (Wr) measurements (Anderson and Neuman 1996) showed a higher mean condition of 107 in the Wales Creek section compared to a mean of 96 in the Canyon Section. Scales from 36 MWF were also collected from the two sections on the Blackfoot River during sampling. Aging the scales from the Wales Creek section show little or no growth had occurred since winter annulus was formed; therefore the outer edge was considered the final annulus. Because the Canyon section scales were collected in September the ages were stated with a plus (+), although little growth probably will occur after the September collection data. The fall age size groups are probably the same because growth after late September would not be significant; therefore 0+ mountain whitefish in the fall would be the same size as a yearling (age 1 fish) in the spring (Table 3). 166 Table 3. Estimated size-groups for each age classes found in the Wales Creek and Canyon section of the Blackfoot River, 2006. Wales Creek Section Canyon Section Age Class Size Group (est.) Age Class Size Group (est.) 1 4 - 4.9 inches 0+ 4 - 4.9 inches 2 5 - 7.9 inches 1 + 5 - 7.9 inches 3 8 - 10.6 inches 2+ 8 - 10.6 inches 4 10.7 - 11.8 inches 3 + 10.7 - 11.8 inches 5 + > 1 1 .9 inches 4+ > 1 1 .9 inches 240 220 200 180 160 140 120 100 80 60 40 20 # of Fish D Wales Cr Sec ■ Canyon Sec Using the age structure comparisons for both survey sections, there is a noticeable lack of younger MWF in the Canyon section compared to the Wales Creek section (Figure 6). For the Canyon Section, all the year classes between 2005-2002 are either low or missing, however larger numbers of 2001 and older whitefish were collected. With exception of the low numbers of 2006- year class (YOY) the year distribution , , ^'^^^^ Figure 6. Comparisons of age structure between the Wales section looks much ^ , . ,^ . . . ^^^^ , Creek section and Canyon section using size groups, 2006. Refer to Table 3 for ages. class in the Creek 4-4.9 5-7.9 8-10.6 10.7-11.8 Size-age groups (inches) >11.9 167 Percent of catch Percent of catch 25 20 15 10 0^ N=184 m^»— 45 40 35 30 25 20 15 10 5 N=274 Hi \-a- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Size class (inches) 123456789 101112131415 1617 Size class (inches) Figure 3. Length-frequency histograms for MWF in the Wales Creek (left) and Canyon Creeks (right) sections of the Blackfoot River, 2006 Lenth-weight relationship for MWF in the Wales Creel< section 500 / *»Jf _ 400 - 3 \^* J£* Weig ir M^^ 100 ^rW ^^,4^*'^ n ^ •M^t**^ 100 200 300 400 Length (mm) Wieght-legth relationship for MWF in the Canyon Section 800 700 ^ 600 ■2 500 1,400 ■ 80 _ra ^ 60 40 20 100 150 200 250 300 350 400 450 Length (mm) 200 250 300 350 Length (mm) 400 Figure 5. Relative weight for MWF in the Wales Creek (left) and Canyon Section (right) of the Blackfoot River. 450 168 Discussion Although not always appreciated by the common angler, the ecological importance of MWF is high, particularly for large salmonids like bull trout or other predatory game fish. If whirling disease were to reduce MWF populations, this could potentially impact not only the fish community, but also the overall food web including a myriad of terrestrial predators (and scavengers) that also rely on MWF as a key forage species Although this review improves our understanding of MWF, still little is known about the local MWF life histories or the vulnerability of fry to whirling disease. What is known is that many streams in the middle Blackfoot Basin, identified as supporting high densities of juvenile MWF in the past, are now highly infected and the clinical signs of whirling are now being detected in MWF in certain western Montana waters, including the middle Blackfoot River. The recent absence of MWF from Hoyt Creek elevates disease concerns for MWF. The spatial overlap of MWF andM cerebralis is a specific concern in the middle Blackfoot Basin where the high densities of YOY overlap with a high severity of disease in other species (e.g. rainbow trout). Until susceptibility is better identified and exposures of MWF are undertaken in the wild, it is difficult to interpret local population changes in places like Hoyt Creek or examine the extremely weak juvenile MWF numbers in highly infected waters of the upper Blackfoot River. In the case of MWF in the upper Blackfoot River, it is interesting to further consider the "top-heavy" population within a context of potential disease implications. Although other factors (e.g. movement) could explain the very low abundance of juvenile fish (age through age 3) in the Canyon section, weak juvenile year classes are consistent with recent increases in whirling disease to high levels of severity (i.e. prevalence of high severity >3 for rainbow trout on MacConnell-Baldwin scale) between 2003-2005. Unlike other susceptible salmonids such as rainbow trout, MWF is by comparison a long-lived fish with individuals approaching 30-years of age (Northcote and Ennis 1994). This longevity is important given the potential population effects resulting from whirling disease in waters like the Blackfoot River would not show up in adult populations for many years. In summary, assessing potential disease effects on MWF in the wild requires an understanding of the movement and habitat use with emphasis on spawning sites, early rearing areas and related movement patterns. Once disease susceptibility is better identified, and if spawning and rearing sites of MWF can be better identified, the known temporal-spatial conditions associated with high (or low) whirling disease infection (and severity) in the Blackfoot Basin can be applied to MWF. To aid in our understanding of local MWF life history, a pilot-level telemetry emphasizing movements and locations of spawning sites is planned for the Blackfoot River in 2008. If the timing and location of spawning can be identified, existing winter temperature information and environmental conditions conducive to infection could be used to identify incubation and hatching and better assess exposure risk of M Cerebralis to fry at a more refined spatial scale. With a better understanding of MWF life history, habitats and susceptibility, the effects of whirling disease on MWF populations could be examined using sentinel cages as well as the continuation of population densities surveys and winter water temperature monitoring in suspected spawning and early rearing areas. This information will help in evaluating overlap between M cerebralis and vulnerable fry. 169 Recommended future work • Complete lab exposures of MWF fry and identify the age and size and other factors related to susceptibility as follows: 1) expose MWF fry at TAM concentrations at 50, 100, 500, 1000, 2000 and 5000 TAMS/fish at three months of age; and 2) expose four month old fry to 1000, 2000 and 5000 TAMS/fish. There will be a control group of MWF for each of the two exposure experiments. The objective will be to determine at what TAM exposure intensity significant mortality will occur and then each of the exposure groups will have histology work at WADDL. Attempt total spore counts to determine how many myxospores potentially are added to the streams. • Attempt sentinel exposures in the wild in areas where vulnerability and fry overlap. • Repeat MWF sampling at pre-disease population survey sites in tributaries in order to detect possible disease-related MWF population changes. • Identify adult spawning and early rearing life histories of MWF within the Blackfoot River Basin and various tributaries in order to determine specific MWF streams at risk. • Continue to monitor the MWF population in the upper Blackfoot River Canyon Creek section. • Identify a funding source to complete juvenile life-history work and develop a more refined study through a U of M graduate study Literature Cited Anderson, R. B. 2004. Occurrence and seasonal dynamics of the whirling disease parasite, Myxobolus cerebralis, in Montana spring creeks. Master of Science Thesis, Montana State University. Anderson, R. O., and R. M. Neuman. 1996. Length, weight and associated structural indices, pages 447-482 in B. R. Murphy and D. W. Willis editors. Fisheries techniques, 2"'' edition. American Fisheries Society. Bethesda, Maryland. Barfoot, Craig. Fisheries biologist with the Confederated Salish and Kootenai Tribe. Personal communication. Brown, C. J. P. 1952. Spawning habits and early development of the mountain whitefish, Prosopim williamsoni, in Montana. Copeia. 2: 109-1 13. Davies, R. W. and G. W. Thompson 1976. Movements of mountain whitefish (Prosopium williamsoni) in the Sheep River watershed. Alberta. Journal of Fisheriers Research Board Canada. 33:2395-2401. Gelwicks, K. and D. Zafft. 2000. Effects of Myxobolus cerebralis on salmonids in the mainstem Salt River, Wyoming. Whirling Disease Symposium: Solutions to Whirling Disease: Putting the Pieces Together. Pp 95-97. Coeur d' Alene, Idaho. MacConnell, E., A.V. Zale and M. Quinn. 2000. Susceptibility of mountain whitefish, Prosopium williamsoni, to Myxobolus cerebralis. USFWS Fish Health Laboratory, unpublished report. MacConnell, E. and E. R. Vincent. 2002. Review: The Effects oi Myxobolus cerebralis on the Salmonid Host. Pp. 95-108 in J.L. Bartholomew and J.C. Wilson, editors. Whirling Disease: Reviews and Current Topics. American Fisheries Society, Symposium 29. Bethesda, Maryland. 170 McPhail, J. D. and P. M. Troff. 1998. The mountain whitefish (Prosopium williamsoni): a potential indicator species for the Fraser System. Environmental Conservation Branch Aquatic and Atmospheric Sciences Divisions, Vancouver BC DOE FRAP 1998-16. Northcote, T, G. and G. L. Ennis. 1994. Mountain whitefish biology and habitat use in relation to compensation and improvement possibilities. Review in Fisheries Science. 2(4):347-371. Pierce R., L. Eby, W. Bollman and D. Vincent. In review. Prediction of whirling disease in basin-fed streams of the Blackfoot Watershed, Montana. Submitted to Transactions of the American Fisheries Society. Ill Table 1 : Catch and size statistics for MWF in the Blackfoot Basin. Stream River Mile Section Date Sampled Length (ft) MWF abundance Total Number Captured Number Captured 1st Pass MWF(<4.0") Captured 1st Pass Range of Lengths (in) Mean Length (in} Total CPUE (#/100') YOYCPUE {#/100') 15-Sept-£ 23-Aug-89 26-Aug-96 440 Present 18 12 12 2.9-3.5 3.2 27 27 15-Sep-99 460 Present 32 32 32 3.0-3.7 3.7 7.1 7.1 Beaver Creek 0.2 24-Aug-89 477 Present 3 3 10.2-12.2 11.4 0.6 0.0 Belmont Creek 0.1 25-Jul-89 9-Aug-01 365 576 Present Present 2 2 2 2 2 2 3.3-3.6 3.3-3.6 3.5 3.5 0.5 0.3 0.5 0.3 Blackfoot River-Johnrud Scotty Brown Raymond Bridge Wales Creek Section H2-0 ditch Pocha Ditch (trap) 5/19-7/15/05 Canyon reach Poorman/Dalton Section Hefner Ditch Hogum Section 13.5 30-May-06 17680 Common 43.9 25-May-06 20064 Common 59.5 26-Aug-99 5745 Common 64 24-May-06 7603 Common 190 84 4.0-15.4 10.4 83.7 24-JUI-95 31-Aug-04 525 1000 Common Common 7 12 7 12 10 2.7-3.2 2.7-3.7 2.9 3.3 1.3 1.2 0.0 1.0 86.5 19-May-05 Present 31 31 3 3.7-7.7 46 95.3 20-Sep-06 5422 Common 277 177 3.9-16.3 12.8 107.2 21-Sep-06 6800 Common 4 4 12.6-15.5 14.1 0.1 0.0 114 25-Jul-OO 8-Aug-01 570 1340 Common Common 10 4 10 4 10 4 2.7 2.6-2.8 2.7 2.7 1.8 0.3 1.8 0.3 11-Sep-O 3.5-15.3 Blanchard Creek Clearwater River 15-Sept-94 350 Present 11 9 14-Sep-95 420 Present 7 12-Aug-97 550 Present 23-Sep-98 425 Present 19-Sep-02 310 Present 1 3.1-3.8 3.3-43 3.1 3.7-43 3.9 3.8 3.1 41 8-Aug-95 29-Aug-95 10496 10.4 17-Jul-95 492 Present 23-Jun-06 66 Present 11-Jul-06 66 Present 2-Sep-04 555 Present 29-Aug-OO 465 Present 16-Sep-02 450 Present 1-Oct-03 465 Present 9.4 8.6-8.7 7.3-7.9 2.6 17 0.2 3.0 6.1 2.6 1.4 19-Sep-02 310 Present 1 1 0.3 Chamberlain Creek 0.1 22-Sept-89 200 Present 17-Sep-98 430 Present 1 1 41 41 0.2 0.0 Clearwater Ditch 0.1 2-Sep-03 22-Sep-05 4224 567 Present Present 2 2 1 3.7-43 40 0.0 0.3 0.0 0.0 0.0 Copper Creek 7.4 Cottonwood Creek 2 30 1 2 30 1 2 30 3.6-3.7 3.1-3.8 3.7 3.4 0.4 67 0.2 0.4 67 0.0 47 28-Jul-92 240 Present 1 1 2.7 2.7 0.4 0.0 5.0 7-JUI-92 225 Present 3 3 15.2-15.6 15.5 1.3 0.0 0.1 6-May-92 6-Sep-01 420 360 Present Abundant 4 4 7.9-11.0 9.4 1.0 0.0 6-May-92 7.5-7.9 3-Oct-91 198 Present 5-Sep-OO 375 Present 22-Sep-03 430 Present 1 4 42 4 24 3.7 3.2-43 3.1-48 3.7 3.6 3.9 0.5 1.1 0.0 1.1 5.6 1.1 5-Sep-OO 354 Present 3 2 2 3.6-3.9 3.8 0.6 0.6 3.0 3-Oct-91 108 Present 2 2 4.5-4.7 46 1.9 0.0 0.2 17-Aug-OO 16-Aug-01 490 510 Present Present 2 2 2 1 1 3.7-7.2 7.1-7.7 5.4 7.4 0.4 0.2 0.2 0.0 1.9 10-Aug-98 21-Aug-OO 400 387 Present Present 1 1 7.6 41 7.6 41 0.0 0.0 2.6 6-Aug-96 569 Present 1 1 11.6 11.6 0.2 0.0 Hogum Creek 0.1 10-Aug-95 108 Present 1 1 2.6 2.6 0.9 0.0 0.4 28-JUI-99 405 Present 1 1 5.1 5.1 0.2 0.0 Hoyt Creek 0.2 8-Sep-92 200 Present 28 20 17 3.1-5.0 3.6 10.0 8.5 Landers Fork 0.1 13-Sept-89 781 Present 1 1 1 3.5 3.5 0.1 0.1 Marshall Creek 2 29-Jun-95 443 Present 1 1 8.7 8.7 0.2 0.0 37 29-Jun-95 394 Present 1 1 7.3 7.3 0.3 0.0 172 Table 1 (cont.): Catch and size statistics for MWF in the Blackfoot Basin. stream River Mile Date Sampled Section Length (ft) MWF abundance Total Number Captured Number Captured 1sl Pass MWF (<4.0") Captured 1st Pass Range of Lengths (in) Mean Length (in) Total CPUE (#/100') YOY CPUE (#/100') Monture Creek 0.4 9-Aug-89 21-Aug-02 480 446 Present Present 2 1 2 1 10.2-10.5 2.4-3.2 10.3 2.8 0.3 0.2 0.0 0.2 2.2 16-Aug-OO 204 Present 1 1 3.7 3.7 0.5 0.5 5.4 16-Aug-OO 18-Aug-05 456 460 Present Present 1 1 2.5 2.5 0.2 0.2 8.6 14-Aug-02 680 Present 12.9 25-Sep-68 400 Present 2 2 2.4-2.7 2.5 0.4 0.4 13.9 25-Sep-68 400 Present 1 1 2.7 2.7 0.2 0.2 Nevada Creek 0.3 14-Sep-OO 465 Present 12 12 5.9 5.9 2.6 0.0 0.7 1-N0V-89 650 Present 2 2 4.0-4.2 4.1 0.3 0.0 5.1 29-Sep-05 6336 Present 60 60 3 3.3-8.8 4.7 0.9 0.0 27 5-Jul-OO 600 Present 29.0 12-Apr-90 9-Aug-94 27-Sep-OO 400 430 522 Present Present Abundant 2 1 2 11.6-12.8 8.6 12.2 8.6 0.5 0.0 0.0 0.0 Nevada Spring Cr. 0.8 21-Sep-04 500 Present 1 5.2 5.2 0.0 0.0 1.1 15-Sep-05 500 Present 1 4.1 4.1 0.2 0.0 2.8 26-Sep-OO 18-Sep-01 450 450 Present Present 1 1 4 4.4 4.0 4.4 0.2 0.2 0.0 0.0 North Fork snorkel survey 1.2 19-Sep-85 12150 Present 17 17 0.1 Blackfoot River 2.6 10-Aug-89 10-Sep-98 22-Aug-02 590 770 660 Present Present Present 4 1 1 6.4-11.7 3.5 9.1 3.5 0.7 0.1 0.0 0.1 NF snorkel survey 4.0 17-Sep-85 29-Aug-98 20430 20430 Present Abundant 305 305 1.5 7.6 29-Aug-02 850 Present 3 3 3 3.1-3.5 3.3 0.4 0.4 7.9 16-Aug-89 15-Aug-OO 735 672 Present Present 6 1 6 1 1 1 3.^10.7 2.6 8.2 2.6 0.8 0.1 0.1 0.1 Weaver ditch at road xing 10.4 23-Sep-94 28-Aug-96 13-Aug-02 300 375 450 Present Present Present 2 4 12 2 4 12 1 12 3.1-4.0 1.8-2.4 3.6 2.1 0.7 1.1 2.7 0.3 0.0 2.7 Weaver ditch at road xing 22-Sep-94 210 Present 38 38 12 3.1-4.5 4.1 18.1 5.7 Rangitch Ditch at N.F. mile 11.6 11.6 23-Aug-05 300 Present 1 1 1 2.8 2.8 0.3 0.3 Rowland Fish camp 12 15-Aug-89 757 Present 1 1 1 2.7 2.7 0.2 0.2 NF snorkel survey 15.5 19-Aug-85 18480 Present 77 77 0.4 Lund Ditch at N.F 15.5 24-Aug-05 310 Present 8 8 8 2.3-2.9 2.6 2.6 2.6 Owl Creek 1.2 23-Aug-90 500 Present 1 1 1 3.2 3.2 0.2 0.2 4.2 19-Jul-95 50 Present 7 7 10.4-12.2 11.7 14.0 0.0 Rock Creek 0.0 2-Aug-94 385 Present 3 3 3 2.0-3.5 2.6 0.8 0.8 Wales Creek 0.1 8-Aug-OO 6-Oct-03 396 391 Present Present 30 3 30 2 30 2 3.1 3.0-3.7 3.1 3.2 7.6 0.5 7.6 0.5 Warren Creek 0.1 11-0ct-91 11-Sep-OO 186 294 Present Present 58 6 47 4 22 3 3.2-4.5 3.0-4.1 3.9 3.6 25.3 1.4 11.8 1.0 0.4 11-0ct-91 180 Present 13 10 1 3.8-4.7 4.3 5.6 0.6 1.1 11-Sep-02 8-Sep-04 576 345 Present Present 1 1 1 3.2 3.7 3.2 3.7 2.1 12-Sep-OO 333 Present 3 2 2 3.8-4.1 4.0 0.6 0.6 West Fork Clearwater River 2.3 22-Aug-06 492 Present 2 2 1 3.5-4.0 3.7 0.4 0.2 3.3 23-Aug-06 492 Present 2 2 1 3.3-4.2 3.7 0.4 0.2 173 RESULTS PART VI: Other special studies Fluvial westslope cutthroat trout movements and restoration relationships in the upper Blackfoot Basin, Montana Ronald W. Pierce, Ryen B. Aasheim, Craig S. Podner, Montana Fish, Wildlife and Parks, Missoula, Montana 59804 ABSTRACT We telemetered fluvial westslope cutthroat trout ( Oncorhynchus clarki lewisi, WSCTj in order to relate migratory life history traits to restoration opportunities in the upper Blackfoot Basin (upstream of the North Fork confluence) of Montana. Telemetry confirmed life-history similarities to fish of the lower basin but also identified higher fidelity to spawning areas and mainstem pools as well as movements through intermittent channels to headwater spawning areas. Anthropogenic influences limit fluvial WSCT abundance and their ability of reproduce and place sensitive areas of the Blackfoot River environment at increased risk. Road crossings, riparian grazing and irrigation practices, primarily in tributaries of the Garnet Mountains, adversely influence fluvial WSCT from the tributary to sub-basin scales. Localized life history characteristics demonstrated in the upper Blackfoot River environment confirm the value of fisheries investigations at reach and regional scales. Understanding local life history strategies is vital when planning fluvial native fish recovery in watersheds of geo-spatial and anthropogenic variability. Telemetry results indicate that WSCT conservation and recovery in the upper Blackfoot basin will rely on restoration of tributaries, protection of intermittent channels, changes in grazing and timber harvest practices on alluvial stream channels and careful management of private ponds (to avoid hybridization). These assessments identified a fundamental need to work with private landowners for fluvial WSCT recovery at a metapopulation scale to be effective. Key words: upper Blackfoot River, telemetry, movement, fluvial WSCT, tributary restoration, private land. Pierce, R., R. Aasheim and C. Podner. 2007. Fluvial westslope cutthroat trout movements and restoration relationships in the upper Blackfoot Basin, Montana. Intermountain Journal of Sciences Volume 13 (2). 174 INTRODUCTION Concern over the declines in both abundance and distribution of westslope cutthroat trout (WSCT) throughout the subspecies range have prompted fisheries managers to attempt to identify the mechanisms responsible for declines and develop effective conservation and recovery programs (Behnke 1992, Shepard et al. 1997, 2003, Pierce et al. 2005). Historical accounts suggest WSCT were once abundant in river systems of western Montana (Lewis 1805, Behnke 1992, Shepard et al. 2005), where populations expressed a range of migratory (fluvial and adfluvial) and stream-resident life history traits (Behnke 2002, Shepard et al. 2003). Fluvial WSCT often occupy large home ranges, spawn in tributaries where the young rear for up to three years, migrate to a large river to mature and then return as adults to their natal tributaries to spawn (Schmetterling 2001, Behnke 2002). Fluvial WSCT have become increasingly rare as a result of habitat loss and degradation, competition with non-native fishes, genetic introgression and fish passage barriers (Mclntyre and Reiman 1995, Shepard 2003), all of which are common in the Blackfoot watershed (Pierce et al. 2005). Radio telemetry has recently been used to elucidate migratory life history traits of native trout species in the lower Blackfoot Basin (i.e. from the North Fork downstream; Swanberg 1997, Schmetterling 2001), such as extensive spawning migrations (>80 km) to natal tributaries by WSCT (Schmetterling 2001, 2003). Telemetered native trout have also helped to identify specific population recovery and protection actions at critical sites; validate restoration assumptions; and monitor fluvial use of completed restoration projects (Swanberg 1997, Schmetterling 2001, Pierce et al. 2004). Two examples of these applications include Dunham Creek and Chamberlain Creek, both recently restored tributaries to the lower Blackfoot River. Dunham Creek involved a bull trout (Salvelmus confJuentus) tagged in the lower Blackfoot River, tracked to an unknown and severely altered (channelized) spawning site, and then entrained in an irrigation ditch during the out-migration (Swanberg 1997). This information, generated during the formative years of bull trout recovery planning, led to restoration of the channelized site and screening of the Dunham ditch (Pierce et al 2002), and contributed to the designation of Dunham Creek as proposed critical habitat for bull trout under the Endangered Species Act (USFWS 2002). The second example is Chamberlain Creek, a tributary to the lower Blackfoot river where, after chronic issues such as dewatering, entrainment, grazing and channel alterations were remediated (Pierce et al 1997), telemetered WSCT indicate that fluvial adults were beginning to use the tributary for spawning in greater numbers (Schmetterling 2001). And higher numbers (densities) of WSCT continue to persist in this stream, years after the restoration efforts (Pierce et al. 2006). The results from these and other telemetry-based investigations have been integrated into monitoring and restoration planning allowing these activities to be targeted more efficiently. However, these applications have focused primarily on the lower Blackfoot basin and other sub-basins within the Blackfoot watershed (Clearwater River Basin and upper Blackfoot River Basin) have not been emphasized. Because of the successful interface between understanding life history traits through applied research and restoration planning and implementation in the lower Blackfoot Basin, we investigated fluvial adult WSCT movements and related our findings to anthropogenic impairments in upper Blackfoot basin where WSCT are present (Pierce et al. 2004). We hypothesized the physical and human environment of the upper Blackfoot basin would locally influence WSCT movement patterns, and areas with low densities of fluvial WSCT therein would reflect human disturbance of aquatic habitat. Study objectives were to 1) describe movement patterns of fluvial WSCT in 175 the upper Blackfoot Basin following Schmetterling (2001), and 2) discuss restoration implications by comparing known upper basin impairments (Pierce et al 2004) with movement of adult WSCT as well as spawning, summering and wintering needs in the upper Blackfoot Basin. The purpose of this study is to characterize seasonal movements over a sub-basin scale so that specific recovery actions can be directed at important, but anthropogenically impaired habitat and movement corridors with the goal of conserving and restoring the fluvial WSCT life history in the upper Blackfoot Basin. STUDY AREA The Blackfoot River, a 5**^ order tributary (Strahler 1957) of the upper Columbia River, lies in west-central Montana and flows west 211 km from the Continental Divide to its confluence with the Clark Fork River at Bonner, Montana (Figure 1). The Blackfoot River drains a 3,728 km^ watershed through 3,040 km of perennial streams, and discharges a mean annual flow of 45.2 m''/s (United States Geological Survey 2004). Higher elevation, glaciated mountains to the north and a lower relief, nonglaciated landscape to the south define the physical geography of the Blackfoot watershed. Northern tributary streams begin in high cirque basins and flow through alluviated glacial valleys, where sections of stream are often seasonally intermittent. The Garnet Mountains to the south of the Blackfoot River produce small streams that are naturally perennial to the Blackfoot River although most are anthropogenically degraded or dewatered during the irrigation season. Lands in the upper Blackfoot Basin are mostly public (65%) headwater areas, with the private lands consisting primarily of timbered foothills and agricultural bottomland. Lower Ba&kn Upper Basin CabJn Cr LflQflnd A How jndl&mpiralurB mo«^hafln(|-Glt< /^ IrriErmrtlefTi; channelx ClarKRM'li.RIVM' WIKwCr DlamorKl Cr Pooriman Cr Figure 1 . Study area: upper Blackfoot River Basin with water temperature and flow monitoring station and intermittent stream channels. The regional (natural and human-induced) variability of the basin is further 176 expressed within the valley of the Blackfoot River. The upper Blackfoot River occupies a lower gradient, alluvial channel with long segments without tributary input, and those tributaries that are present are often seasonally intermittent or degraded in lower reaches often as a result of agricultural activities. The upper river supports low instream (secondary) productivity and water quality impairment from non-point agricultural sources increases between Nevada Creek and the North Fork Blackfoot River (Ingman et al. 1990). At the junction of the North Fork, the divide between the upper and lower basins, the lower Blackfoot River receives a large influx of colder water, which reduces summer water temperature, improves water quality and approximately doubles the base flow of the lower Blackfoot River (Ingman et al. 1990, Pierce et al 2006, United States Geological Survey 2006). Contained by glacial boulders and bedrock, the lower river channel is steeper, geomorphically stable and bedrock controlled. The lower Blackfoot River has higher secondary productivity (Ingman et al. 1990) and much higher densities of WSCT than the upper Blackfoot River (Pierce et al. 2004). The density of adult WSCT in the upper mainstem Blackfoot River near Nevada Creek are as low as 4/km compared to 58/km in the lower Blackfoot River near Chamberlain Creek and few, if any, fluvial WSCT from the lower Blackfoot River migrate to the upper Blackfoot basin upstream of the North Fork confluence (Schmetterling 2001, 2003, Pierce et al. 2006). Unlike the lower Blackfoot basin and despite no isolating mechanism, the upper Blackfoot Basin is absent of fluvial rainbow trout (O. mykiss) reproduction with the exception of Wales Creek (Shepard et al. 2003, Pierce et al. 2005). Here, WSCT occupy about 90 percent of headwater tributaries although population abundances usually decrease in the downstream direction due to tributary alterations (Pierce et al. 2004). The loss of spawning areas has been identified as a major reason for the decline and low abundance of WSCT within the upper Blackfoot River. Correcting anthropogenic impairments in the upper Blackfoot Basin is increasingly a restoration focus (Blackfoot Challenge 2005), but prior to this study no attempt has been made to identify problems specifically affecting fluvial WSCT. Within the upper Blackfoot Basin, the first 88 km of upper mainstem Blackfoot River above the confiuence of the North Fork Blackfoot River is naturally stratified into three (hereafter upper, middle and lower) reaches, among which anthropogenic impairments are spatially variable (Pierce et al 2004). The upper reach extends 33.4 river kilometers (rkm) from Poorman Creek (rkm 174.2) to Arrastra Creek (rkm 140.8) and is a densely wooded C4 alluvial channel -type (Rosgen 1996). This reach begins at the downstream end of an intermittent section of the mainstem where groundwater and spring creek inflows reenter the mainstem Blackfoot River. The middle reach, also a C4 channel -type, extends 32.5 km from Arrastra Creek downstream to Nevada Creek (rkm 108.3). This reach is less wooded and the channel loses slope, becomes highly sinuous, prone to bank erosion and deposition of fine sediment. Riparian livestock grazing is more common in downstream areas (Marler 1997, Confluence Consulting 2003) and the lower section of this reach is increasingly dewatered during the irrigation season (Pierce et al 2005). Other than at reach boundaries no tributaries enter the middle reach. The lower reach extends 22.3 km from Nevada Creek, a water quality (nitrate, phosphate, total suspended solids and temperature) impaired tributary, to the mouth of the North Fork (rkm 86) (Ingman et al. 1990, Pierce et al. 2006). Below Nevada Creek, the Blackfoot River transitions from a low gradient alluvial (C4) channel to a more confined, higher gradient geologically controlled (B3 and F3) channel (Rosgen 1996). Several small but degraded and dewatered tributaries enter this reach from the Garnet Mountains (Pierce et al. 2005). 177 METHODS Radio telemetry WSCT were captured in the upper Blackfoot River, phenotypically identified, implanted with continuous radio Lotek^M transmitters (between 13 March - 18 April 2002 and 18 March - 13 April 2003) and tracked fish through one fiall spawning migration cycle. Visual identification was later verified through genetic analysis of fin clips using 17 fragments of nuclear DNA at the University of Montana, Trout and Wild Salmon Genetics Laboratory (Boecklen and Howard 1997). Transmitters were evenly distributed (10-11 per reach) within each of the three study reaches. Fish were captured prior to spring run off, presumably prior to spawning migrations (by angling or electro- fishing) in suspected wintering pools. Individually coded transmitters weighed 7.7 g, had an estimated life of 450 days, did not exceed 2 percent of fish weight (Winters 1997) were implanted following standard surgical methods (Swanberg 1997, Schmetterling2001). Fish were located from the ground, using either an omni-directional whip antenna mounted on a truck or a hand held three-element Yagi antenna when walking. When ground tracking failed to locate a fish, we relied on fixed wing aircraft flying approximately 100-200 meters above the river, equipped with a three-element Yagi antenna attached to the wing strut. Similar to Schmetterling (2001), fish were located at least three times per week immediately prior to and during spring migrations and spawning, once-per week while holding in tributaries or the Blackfoot River following spawning, and once per month thereafter. For each ground-based relocation within a habitat unit, we triangulated the fish's location to within an estimated 5 m and recorded with the location using GPS. Within tributaries and the Blackfoot River, locations were expressed as the distance upstream from the mouth in river kilometers. Following Schmetterling (2001), fish were assumed to have spawned if they ascended a stream (or river reach) with suitable spawning habitats during a spring spawning period, and the upper-most location was the assumed spawning site. Because of high flows and poor instream visibility, we were unable to visually validate spawning at most assumed spawning areas. We therefore relied on the presence of juvenile (age and I) WSCT within <2 km of all identified spawning areas (FWP unpublished data) to support spawning site assumptions. The mean date between two contacts surrounding an event, such as a migration start or spawning date was used to estimate the date of an event (Schmetterling 2001). We considered relocations from November through April to represent winter habitat use, while a spring spawning-migration period was delineated from May through 14 July and summer habitat use from 15 July through October. Blackfoot River daily discharge data were obtained from U.S. Geological Survey (USGS) gauging station (No. 12335100) located in the middle reach at rkm 115.5 to examine potential relationships between discharge and fish movement. We also placed thermographs (Onset^'^) at the USGS guage to evaluate the effect of maximum daily water temperature on the onset of migration and spawning. We used the FWP "dewatered stream lisf to identify naturally intermittent reaches (Pierce et al. 2005), and we compared the basin area above intermittent channels between the lower and upper Blackfoot subbasins. 178 Because of small sample size, all first-year WSCT spawners fi'om 2002 and 2003 were grouped by reach and reach differences were then tested by the dates migrations began and dates WSCT entered tributaries using a Kruskal-Wallis one way analysis of variance (ANOVA) on ranks. To explore between-year (2002 and 2003) differences influencing the onset of movement and spawning, we compared daily water temperatures for the May through 14 July spawning migration period using paired t- test. Mann-Whitney rank sum tests were then used to test between-year differences in the dates migrations began and the date first year WSCT spawners entered tributaries. Potential associations between date migrations began and total pre-spawning distance moved, and spawning tributary size (drainage area) and number of days WSCT spent in each of these tributaries was assessed with linear regressions. Second-year (repeat) spawners were tracked in 2003 but not included in our analyses because of the limited transmitter life during the second migration/spawning period. All results were tested at the alpha 0.05 level of significance. RESULTS Over the course of this study we tagged and tracked 31 WSCT to spawning sites, and those fish with active transmitters were then tracked to summering and wintering areas. These 31 fish were located each an average of 39 times (range: 17-88) between the March 2002 and December 2004 study period. Four spawners tagged in 2002 were tracked as repeat spawners in 2003 and these fish were used to identify spawning site fidelity. Twenty-nine (94%) of the 31 fish tested genetically pure WSCT. Two fish (6%) contained all WSCT genetic markers plus two of seven rainbow trout genetic markers and were classified as post-Fl generation hybrids (Martin 2004). Because of their visual WSCT features the low level of hybridization we included these fish in our analyses. Overall, twenty-eight (90%) fish migrated to tributaries, while three migrated to spawning sites in the upper main stem Blackfoot River during the two-year study (Figure 2, Table 1). Figure 2. Capture locations (open symbols) and assumed spawning sites (closed symbols) of telemetered WSCT for 2002 (left) and 2003 (right). Numbers refer to individuals in Table 1. During the migration and spawning periods, river temperatures were similar between 2002 and 2003 (P = 0.29), and WSCT migrations began on the rising limb of the hydrograph as temperatures approached 4 °C (Figure 3). Twenty-two WSCT 179 migrated upstream, nine moved downstream and one repeat spawner (fish # 8) moved upstream (in Temperature (C) 2002) and downstream (in 2003) 1 before ascending spawnmg streams. The period of migration in the Blackfoot River averaged 16 days and fish moved an average of 21 km in the Blackfoot River before reaching spawning tributaries or main stem spawning sites (Table 1). Tributary spawners entered spawning streams at mean water temperatures of 6-7 °C and migrated another 8 km to spawning sites. 60 50 40 30 20 10 Discharge (cms) Figure 3. Relationships of water temperature (top) and discharge (bottom) to dates WSCT began migrations (range and median) in 2002 Among (grey) and 2003 (black). The range is shown by the horizontal bar and the three median migration start date by vertical arrows, reaches, the start of spawning migrations incrementally increased in the upstream direction from 29 April in the lower reach, to 1 May (middle reach) to 4 May in the upper reach, however differences were not significant (ANOVA, P = 0.89). Between years, WSCT began their spawning migrations 17 days later (13 April versus 26 March) in 2002 (range: 54 days) than in 2003 (range: 61 days). Although slight annual variation was detected (P = 0.085) differences were not significant. Likewise, the starting dates of WSCT migrations were not associated with the distance moved (R^ = 0.08 P = 0.24). Overall WSCT spawning occurred in nine tributaries varying from T* to 4* order (see Table 1 and Figure 2 for locations). Arrastra Creek and Willow Creek supported the highest proportion of telemetered spawners (9 or 29% and 5 or 16%) respectfully, and each of these tributaries also had at least one 2002 repeat spawner return in 2003. WSCT entered tributaries from mid-April through mid- June (mean date: May 16). There were no significant differences in the date WSCT entered 180 spawning tributaries either among reaches (ANOVA, P = 0.42) or between years (P = 0.17). WSCT spent an average 51 days in tributaries (range 4-402) and spent significantly different amounts of time in the seven different spawning tributaries (R^ = 0.36, P = 0.002), staying the longest in the largest tributary, the North Fork. The majority of WSCT tagged in the lower river reach (6 of 11 or 55%) migrated downriver to the lower reach boundary before ascending the North Fork for spawning (n=3) or two tributaries to the upper North Fork (Dry Fork (n=2) and Cabin Creek (n=l)). Three other lower reach fish entered Wales Creek (n=3), a tributary adjacent to the lower reach; while two ascended the middle river reach to spawn in Arrastra Creek (located at the middle-upper reach boundary). Most (9 of 10 or 90%) WSCT tagged in the middle river moved upriver to either Arrastra Creek (n=6). Sauerkraut Creek (a tributary to the upper river reach, n = 1), or through the upper reach to Willow Creek (n = 2). Only one middle reach fish migrated downriver before ascending the North Fork. Similar to middle reach fish, most (9 of 10) WSCT originally in the upper river reach migrated upriver, however unlike the concentrated spawning of most middle reach fish, spawning of upper reach WSCT was dispersed Table 1 . Summary of capture locations, spawning movements sites and dates, time spent in tributaries and fate of post-spawning WSCT, 2002 and 2003; PM = post spawning mortality. Year and Fisli River capture Prespawning Prespawning distance (km) " Spawning Use of intermittent Spawning Days in reach no. location (rl''Pi9'*ll*|ih9 V)ttill^kfT4^Mk Hwf f lNfl fc PUMUBli'&IIAlIll CIIIH4.1A hUbMU PitUH {>illHl El ^::? |m(.imkMllvi[]V Figure 8. Longitudinal profile for Hayden Creek. Fish Population and other monitoring activities We surveyed fisheries at stream-mile 0.1 and identified low numbers of bull trout and WSCT (CPUE =1.8 and 0.8, respectively). Sculpins were present, but no amphibians were observed. Water Chemistry readings identified a pH of 7.62, very low eii-^gtion y 1 w» w conductivity of 44uS and very low TDS reading TDS of 21ppm. Un-named Creek Creek Pass near Hahn Description We sampled an unnamed tributary that enters Monture Creek at stream-mile 27.2 and drains a small basin (1.8 mile^) near Hahn Creek Pass. It lies entirely on the Lolo National Forest adjacent to the southern boundary of the Bob Marshall Wilderness. Approximately 2.2 stream-miles in length, only the Slrt^m mllcjgt' Figure 9. Longitudinal profile for Un-named Creek. 205 lower 0.7-mile of stream is perennial. Stream gradient is 325' in the lower mile, but increases significantly to 1200' / mile in the upper reaches (Figure 9). Base flow discharge was estimated at 0.5-l.Ocfs. The riparian under-story vegetation is primarily alder, rocky mountain maple and young conifers above a ground-cover of mixed with horsetail, forbs and various grasses beneath a canopy of lodgepole pine. Stream channel substrate is primarily boulder, bedrock and cobble with gravel and detritus. Large woody debris recruitment to the stream channel is moderate creating small plunge pools and cover for fish habitat. Fish population and other monitoring activities In 2006, we conducted a fish population survey at stream-mile 0.5. Sampling recorded low numbers of WSCT at a CPUE of 0.9 fish and these fish averaged 7 inches in length. No fish were found upstream of a bedrock nick point observed ~ 450' upstream of the survey section. No other fish species or amphibians were sampled or observed. Water chemistry data was collected recording: pH of 8.0, low conductivity of 48uS and low TDS of 24ppm. Wedge Creek Eifr^anon )t HKW jHh ir FllllUbl 'int^LiU J 7.5-- T-- 6.5- 6-- 4J5-- J- FuiHli A Description Wedge Creek is a high-gradient stream (mean gradient = 1,030' /mile), 2.1-mile length and drains a small 1.9 mile^ basin on the southern slopes of Fenn Mountain. It lies within the Lolo National Forest just south of the Bob Marshall Wilderness. Wedge Creek is a l*** order tributary stream that enters Monture Creek at stream-mile 20.3 with an estimated base-flow of 0.5-l.Ocfs. Wedge Creek is classified as a "Rosgen Al" channel-type, and it is characterized by high-gradient stream channels with cascading step pools created by bedrock and boulders. The riparian over-story vegetation is dense and predominately a Douglas fir, larch and lodgepole pine forest above a thin under-story of rocky mountain maple, young conifers, snowberry, ferns and grasses. 3J5- n»iK»T*rimt( Hp Flih Pt'Pllllhll S#P'1I l-th-M^ilaiidl Fai-iai siht^m r/a«a^ Figure 10. Longitudinal profile for Wedge Creek. Fish population and other monitoring activities In 2006, we conducted a fish population survey at stream mile 0.1 on Wedge Creek. No fish or amphibians were sampled or observed. Water chemistry reading identified a pH of 8.43, conductivity of 135uS and low TDS of 67ppm. Yellowjacket Creek Description Yellowjacket Creek is a small 1**' order stream that flows west 1.9 stream-miles to its confluence with Monture Creek at stream-mile 18.4. It drains ~ 0.9 mile^ of Lolo National Forest 206 land with an average stream gradient of 1,205' / mile and generates an estimated base-flow of 0.2 - 0.5 cfs. The riparian under-story is relatively dense, primarily rocky mountain maple and alder, ferns and beargrass beneath an over- story of lodgepole pine and Douglas fir mixed with Englemann spruce. Large woody debris recruitment to the stream channel is moderate. Stream channel substrate is predominately cobble and boulders with gravel. The stream becomes boulder and bedrock dominated as gradient increases in the upstream direction. Elfrvai1on)i:14HX>4ttt 7.5 6.5- 5.J5 5- 4i t£- it ht'intovSinrvL'Ktfttn tivnHw Wp^iivCrtffe WiaBl*#< ■Ciinlii*MHfliii FHhaiy P41Alllll.tl SVA.'lIll 1-ii* HiUlgiLiI Ffli-IM '4 -giT-***!! ri«*»ff* Figure 11. Creek. Longitudinal profile for Yellowjacket Fish populations and other monitoring activities We conducted a fish population survey section at stream-mile 0.1 where we found only YOY WSCT at a CPUE of 10. No adult fish, other species or amphibians were observed. Water chemistry readings identified a pH of 8.23, conductivity of 1 13uS and TDS of 57ppm. North Fork of the Blackfoot River unnn Kxa iaaa SDDD 4DD[( 3WK> IIhvbJLcc] i" 'tlHFtI SH%f L-K I'ldirJdMi Description The North Fork of the Blackfoot River is the largest tributary to the Blackfoot River. Beginning on the Continental Divide, the headwaters of the North Fork drain a glaciated region of the Scapegoat Wilderness. The North Fork flows west and southwest a total of 41.5-miles. At stream-mile 22, the North Fork exits the Wilderness and then enters Kleinschmidt Flats, a large glacial outwash plain, near stream-mile 12.0. The North Fork enters the middle Blackfoot River at river-mile 54. Below the North Fork Falls (at mile 26.7) the lower North Fork variably supports fluvial bull trout and WSCT, brown trout, rainbow trout and very low densities of brook trout depending on the specific stream reach. In 2006-07, we conducted a series of fish population surveys above the North Fork Falls. These included three sites (stream-miles 27.2, 33.3 and 36) on the mainstem North Fork, two sites on the East Fork of the North Fork and seven smaller tributaries (Broadus, Cooney, Dobrota and Theodore, Pony, Scotty and Sourdough Creeks). We also surveyed Canyon Creek - a tributary to Dry Fork, which is located downstream of the North Fork Falls (Figure 12). 2D0O 3 4 i«h HWfMtftlV- L?JitP^pCr'«t: Description The East Fork of the North Fork is a 3™ order stream that originates -1.5 stream-miles upstream of Parker Lake. From Parker Lake, the East Fork flows northwest -13 miles to its confluence with the North Fork at stream-mile 27. The East Fork ^^ Eli-vrttion drains eight tributary streams over a -67 mile^ basin, which includes the Helena and Lolo National Forest portions of the Scapegoat Wilderness (Figure 14). In 2006, we established two fish population surveys sections (stream miles 7.0 and 1 1.7) on the East Fork. The lower survey site falls within the 1988 Canyon Creek burn area, and it contained significant amounts of LWD within the channel. A dense forest of lodgepole pine has become established in the riparian zone along with a corridor of 55*0" •>m)-- idV)-- AOK HialCi'iiB. '£«IHlJ«llfll Cl-IOL Tl4«lDailllll.llAlJ lltHtfe Cvi ^\it HaIaii^ lliA4*ill4Mmiyn»ih ukUttm^f^ta iii^wiitim i 4 ^iiinnh'.qi «J Figure 16. Longitudinal profile for Cooney Creek. Fish populations and other monitoring activities A fisheries survey at stream-mile 0.4 recorded low numbers of rainbow trout (CPUE = 0.2), and no other fish species were present. Water chemistry readings identified moderate conductivity of 160uS, low TDS of 80ppm and a water temperature of 56.3°F during the survey. Dobrota Creek Description Dobrota Creek is a 1*** order tributary to the North Fork, located on Lolo National Forest and within the Scapegoat Wilderness. Dobrota Creek drains a small basin (-6.1 mile^) on the southern slopes of Scapegoat Mountain. Dobrota Creek flows in a southerly direction for about 4.0 miles to its confluence with the North Fork at stream-mile 35.9 near the Carmichael Guard Station. The lower 2.0 miles of Dobrota Creek has an average gradient of 192' / mile, compared to l,434'/mile in the upper 2.0 miles of stream SljIHl P Figure 17 . Longitudinal profile for Dobrota Creek. 210 (Figure 17). The Canyon Creek fire burned this area intensely in 1988, resulting in a stand-replacement of lodgepole pine forest. High rates of erosion are occurring in areas against steep hillsides where plant re-growth is slow. This erosion process, however, is recruiting high amounts of LWD from standing snags. The riparian plant communities are composed of willow, alder, young lodgepole pine and a robust mixture of grasses and forbs that contributes to bank stability. Dobrota Creek classifies as a "Rosgen B3" with cobble-dominated substrate along with gravel, boulders mix with large areas of bedrock. Because of the lack of overhanging vegetation, fish habitat is primarily localized to LWD plunge and scour pools and boulder pocket water. Fish populations and other monitoring activities A survey of fish populations at stream-mile 0.1 found low numbers of rainbow trout (CPUE = 2.7) and no other fish species were found. Genetic samples were analyzed and confirmed introgression of RBT with WSCT and Yellowstone cutthroat trout. Water chemistry recorded conductivity at 148uS, TDS of 74ppm and a water temperature of Sl.Q^F during the survey. 8500 8000 - ■ 7500 - ■ 7000- ■ 0500 - ■ 0000-- 5500- ■ ■k Fislieries Siiivey Location G Gonetic Samples Galusha Peak Lost Pony Creek Description Lost Pony Creek is a T* order perennial tributary to the middle reaches of the East Fork. Located in Helena National Forest and Scapegoat Wilderness, Lost Pony Creek drains a small basin (-3.6 mile^) on the southern slopes of Galusha Peak. Lost Pony - Elevation Creek flows in a southerly direction 3.8 miles to its confluence with the East Fork at stream-mile 6. 1 . Stream gradients range from 208' / mile near the mouth to 750' / mile in the upper reaches. At stream-mile 0.85 the outlet stream from West Twin Lake enters with ~0.2-0.4cfs. Below this junction Lost Pony Creek has an estimated base-flow of 0.5cfs. The stream banks are stable as a result of very dense riparian shoreline vegetation composed of 5000-- 4500- . Triljutaiy from West Twin Lake East Fork of North Fork Blackfool Rivei Rainbow Trout Dominatet itiliiiailiibiiilFuAii iTpiipiTi VJlJiJiiiii Scotty Creek Description Scotty Creek is a 2 order tributary flowing south -4.8 miles through Helena National Forest land to its confluence with the East Fork at stream-mile 9.3. The small watershed (-4.4 mile^) drains a small cirque lake and the slopes of Olson and Pyramid Peaks that lie along Red Ridge within the Scapegoat Wilderness Area. Stream gradients range from 710' / mile near the headwaters to 2207 mile near the mouth (Figure 19). Riparian vegetation is very dense and consists of willows and grasses within a regenerating lodgepole pine forest. The forest was severely burned during the 1988 Canyon Creek fire and is now recruiting high concentrations of LWD to the stream channel. Fish Populations and other monitoring activities A fish population survey at stream-mile 0.2 identified hybrid rainbow trout at low densities (CPUE = 0.4). One "cutthroat trout" was identified (CPUE = 0.2) in the field and numerous western toads were observed. Genetic analyses identify rainbow trout hybridized with Yellowstone cutthroat trout and minor genetic contribution of WSCT. Sourdough Creek flit J it Figure 19. Longitudinal profile for Scotty Creek. Description Sourdough Creek, a 2"'' order tributary, drains a small (-5.1 mile^) basin on the western slopes of Red Mountain, as well as a series of small cirque lakes before flowing north -3.4 miles and joining the East Fork at stream-mile 9.4. Sourdough Creek lies entirely in Helen National Forest and Scapegoat Wilderness. Stream gradients average 185' / mile in the lower 2.0 miles of stream, and then increase to 1,060' / mile between stream-mile 2.0 and 3.0 before decreasing in the very headwaters (Figure 20). The riparian under-story vegetation is a dense shrub community (willow, alder and red-osier dogwood) and various forbs and grasses. Only the lower 0.1-mile of Sourdough Creek was affected by the 1988 SH*' Eli^vrilen SHU" sm CkinLJlHV Eaisi Fuk id iHMili Fuk BtacsfMi FtuiiaM Tiori tmitian Pft-tlP^H Ml't.'^ 1 \ J ^ i.i f|it4_Mllvi3V Figure 20. Longitudinal profile for Sourdough Creek. 212 Canyon Creek fire; thus, the drainage supports a mature lodgepole pine and subalpine fir forest. Fish habitat consists of overhanging vegetation, under-cut banks and large boulder substrates. Larger woody debris recruitment in the majority of the stream channel is low compared to the nearby bum area. Fish Populations and other monitoring activities A fish population survey conducted at stream-mile 0.45 recorded very low numbers of rainbow trout (CPUE = 0.5) and fish averaged 7.2 inches in length. No other fish species or amphibians were observed. Only three genetic samples were collected and these fish were identified as rainbow trout introgressed with Yellowstone cutthroat trout and WSCT. Theodore Creek Description A small 1**' order perennial Ei..i,B« tributary, Theodore Creek flows northerly -2.4 miles through Lolo National Forest (Scapegoat Wilderness) before entering the upper North Fork at stream-mile 33.6. This high-gradient stream (mean gradient 5307mile) drains as small basin (1.6 mile^) on the north- eastern slopes of Galusha Peak and generates an estimate base-flow of 0.5-l.Ocfs (Figure 21). The 1988 Canyon Creek fire burned the riparian plant community along Theodore Creek, which now consists of dense communities of young Englemann spruce, black Cottonwood and lodgepole pine, along with willow, forbs and grasses at the stream margin. LWD recruitment to the channel is high and over-hanging shrubs contribute extensively to instream habitat features. The survey location on lower Theodore Creek falls into a "Rosgen C4" type-channel with a predominately gravel substrate. Fish population and other monitoring activities A fish population survey at stream-mile 0.2 found no fish. Spotted frogs were observed. Water chemistry measurements recorded conductivity at 166uS, TDS at 83ppm and a water temperature at 56.8*'F during the survey. Canyon Creek Description Canyon Creek is a tributary to the Dry Fork of the North Fork. The Dry Fork is a large glaciated basin that enters the North Fork downstream of the North Fork Falls. The headwaters of Canyon Creek begin in a marsh within a cirque basin upstream of Canyon Lake. The upper Canyon Creek basin is proposed wilderness and the lower basin falls within the Scapegoat Wilderness area of the Lolo National Forest. The Canyon Lake outlet stream joins with Conger Creek, a small tributary stream draining the slopes of Omar Mountain and Canyon Point, and f^tmi ndtogt Figure 21. Longitudinal profile for Theodore Creek. 213 H44- SHi' HK-r nw- tm- ' — -I^F^ dH-lto^ '■■•'- ■•■■■■' ■'-■■ T' • ^ 4 ■HtJiH t^^mTi** I I T t.ktbiVmlf^HI tl. dt tj M together they form a 2" -order stream that enters the Dry Fork of the North Fork Blackfoot River near rm 5.0. Approximately 5.6-miles Pi,iu« in length, stream gradient decreases from 165' / mile at the headwaters to 40' / mile in the middle reaches before increasing to 184' / mile near the mouth (Figure 22). Canyon Creek flows northerly and contributes an estimated base-flow of ~5 to lOcfs to Dry Creek. The 1988 Canyon Creek forest fire burned the lower basin. A strong regeneration of lodgepole pine and high concentrations of LWD recruitment to the stream channel is now occurring. The middle to upper reaches of the drainage were unaffected by the fire, the stream banks are stable supporting dense over-story populations of lodgepole pine, Douglas fir, Englemann spruce above an under-story of willows, alders, rocky mountain maple, forbs, shrubs and various grasses. Large woody debris recruitment to the stream channel is moderate occurring at its natural pace. Stream channel classification at the survey location is predominately a "Rosgen B3" type channel with a substrate of cobble / gravel mixed with boulders. Log scour pools, overhanging vegetation, undercut banks and boulder pocket pools are primary habitat features. Fish Population and other monitoring We conducted a fish population survey on Canyon Creek at sm 1.5 in 2007. Fish sampling recorded a WSCT CPUE of 8.1. We failed to detect bull trout or other fish species. Bull trout were present in Canyon Lake in past surveys. Genetic analysis showed this population to be pure WSCT. Frog tadpoles were present but no adults were observed. Water chemistry readings were: conductivity at 1 18uS, TDS at 60ppm and a water temperature at 57. tV during the survey. Figure 22. Longitudinal profile for Canyon Creek. 214 Lake surveys in the Blackfoot Basin Introduction and study area During the field season of 2006 and 2007, fisheries crews surveyed both the high mountain lakes in the "backcountry" areas (i.e. roadless and wilderness areas) as well as several low- elevation lakes located on the floor of the Blackfoot valley (Figure 23). The backcountry lakes were last surveyed more than 20 years ago and many of these were historically stocked with rainbow trout and "undifferentiated" cutthroat trout. Our survey objectives were to describe physical and biological attributes of lakes and evaluate accessibility and relative levels of recreational use. Surveys included fish population assessments, amphibian searches, bathymetric mapping, water chemistry measurements and a description of recreation sites and trail networks. We surveyed a total of 15 backcountry lakes, including three in the upper Monture Basin, seven in the North Fork Basin and five in the Landers Fork Basin. In addition to the backcountry lakes, we also surveyed Nevada Reservoir and eight additional lower-elevation natural lakes that possess some level of public access. All surveyed lakes are also included in Figure 23. Additional lake surveys are planned for the summer of 2008. s Figure 23. Lake survey sites: The lakes identified in black are the backcountry lakes and lakes identified in brown are lower-elevation lakes. Procedures Backcountry lakes - Using pack stock, fisheries crews established a series of remote base camps near all known fish-bearing and a few fishless backcountry lakes. Descriptive information (e.g., elevation, surface area, specific location, etc.) was approximated from USGS topographic maps and existing GIS data. Fish sampling was conducted using overnight sets of sinking experimental gill nets. We used standardized net dimensions and mesh size (125'x 4'; 5 panels; 0.75", 1.00", 1.25", 1.5", and 2.0" bar) specified for alpine lake sampling in Montana. Nets were set for a single sampling period (minimum 10 hrs) usually beginning between 18:00 and 20:00. Nets were typically anchored to a log or rock on the shoreline (small mesh end) near a point or prominent feature with 215 gradual depth contour. We used an inflatable boat to stretch and set the remainder of the net (maximum depth rarely exceeded 30 ft). Small lakes (< 10 acres) were sampled for one netting period. On larger lakes and in instances where fish abundance was obviously low, we set two nets concurrently at widely spaced locations. For comparative purposes, all gill net catch results were standardized by species as number of fish per net-hr. Nets were located at previous sampling locations where possible. Fish caught in gill nets were sacrificed and processed on shore. We weighed and measured each individual, assessed sex and maturity, and recorded a qualitative description of stomach contents. Scales were removed from a sub-sample for subsequent age and growth analyses. In most instances where Oncorhynchus Clarkii lewisi were suspected, we preserved 25 fin clips in individual vials filled with 95% ethanol. Amphibian surveys were conducted during lake- perimeter surveys. All amphibians were identified to species and life stage. Total observed abundance of each species and life stage was approximated for each lake. Basic water chemistry measurements were collected at shoreline and mid-lake locations using a hand-held electronic meter. Measurements included surface water temperature, pH, conductivity (uS/cm), and total dissolved solids (TDS; ppm). Water clarity was also measured from an inflatable boat with a Secchi disk between 10:00 and 17:00 while wearing polarized sunglasses. The Secchi depth reported was the mean of two replicates by independent observers. In some cases, the maximum observable Secchi depth was estimated because it exceeded the maximum depth of the lake. Lake bathymetric maps were created using field location and depth measurements at a series of points that characterized each lake. Although the total number of points varied based on lake size and depth variability, the protocol always included locations along the entire lake perimeter and at least five transects across the water body. At each transect point, latitude and longitude measurements were collected from an infiatable boat using a hand held GPS unit. Water depth was measured with a hand held electronic depth-finder. The total number of points collected per lake generally ranged from 150-300. Field data were transferred to spreadsheet files in the office and shipped to the FWP Information Services Unit. Once formatted, point data were processed by TIN (triangulated integrated network) mapping software to produce bathymetric maps with 2-10 ft contours. The program also calculated surface area and lake volume. Features of interest such as trails, inlet streams, outlet streams and campsite/fire ring locations were later plotted on each map. Results Summary results of all lake surveys are located in Table 2. More detailed individual descriptions of all surveyed lakes are identified below. 216 Table 2. Summary of lake survey information collected in 2005-2007. Lake Survey Date Location Traiihead Miies in Lake Features Recreation Fisheries Water Chemistry Morph. Elev. Acres Max Depth Camp Sites Use #of Fish Fish/Hr Genetics Size Range (inches) Secchi depth IDS PH Cond. Braziel Lake Oct-06 Helena NF,T17N R11WS28/33. N/A No Access Isolated basin 5141' 7.8 18.5 None Heavy, local resident, cattle None 1 gill net, 24 Hr N/A N/A 2 101 8.36 203 Browns Lake 5-17- 2006, 10/2/2006 Powell County, T14N R11W S16C,17D,20A/B /C/D,21B/C,29B N/A Kettle Basin 4292' 530 27.4' 10, developed Heavy 270 RBI (2004) Set times were not recorded N/A 4.2-24.2 AVG = 12.3 22.6 166 8.5 334 Camp Lake Jul-05 LoloNF,T17N R11WS32. North Fork Blackfoot River 8 Glacial Valley 6161' 15.5 13.9' 1, Hardly Discernable Light 16RBT 1 gill net, 14.5 Hr = 1.1 RBT 5.9-13.8 Avg = 9.0 13.9 49 84 97 Canyon Lake Jul-05 LoloNF,T17N R11WS28/33. North Fork Blackfoot River 8.5 Glacial Valley 5741' 11.2 6.8' 1, Marginal Light 18WCT 1 gill net, 18.5 Hr = 0.97 WSCT 6.5-13.5 Avg = 10.2 6.8 83 8.81 170 Coopers Lake May-06 LoloNF,T15N R10W S6B/C,12A. N/A Glacial Valley 4491' 200 70' 3, developed Heavy WSCT EBT 0.072 0.028 N/A 9.5-16.9 10-14.6 24.6 66 8.23 144 Heart Lake Jul-05 Helena NF,16N R8W S17C/18D/19A Indian Meadows 4.2 Glacial Valley 6424' 28.3 55.8' 4, Well Established Heavy 17WCT 2 gill nets 26.8 Hr = 0.63 Planted WSCT 13.6-18.5 Avg = 16.1 37.5 108 8.71 216 Lake Otatsy Jul-05 LoloNF,T17N R11WS32/ T16N R11WS18 North Fork Blackfoot River 7 Glacial Valley 6069' 19.1 32.6' 2, Well Established 1, Marginal Moderate 21 RBT 1 gill net, 14Hr = 1.5 RBT 6.1-11.5 Avg = 9.7 21.5 44 8.13 89 Lower Copper Lake Jun-06 Helena NF,T15N R9WS10C,9D. Option 1 : Stonewall Mt Option 2: end of Copper Cr Rd Option 1:3 Option 2:1 Glacial valley 6870' 6.1 20.3' None Light 2WCT 1 gill net, 19.2 Hr = 0.10 WSCT 13.3-14.8 AVG = 14.0 202 14 7.77 27 Lower Twin Lake Jul-05 Helena NF,T16N R9W S6. Option 1 : Meadow Creek Option 2: Indian Meadows Option 1:13.75 Option 2:12 Glacial valley 5900' 6.6 11.6' None Light 25 Trout 1 gill net, 7Hr = 3.66 Hybrids YCTx WSCTx RBT 5.7-23.6 Avg = 12.7 6.6 112 87 226 Maddie Lake Jul-06 Powell County, T15N R10W S30C 2 Isolated basin 4560' 5.6 33.7' None Light None, Redside shiners abundant 1 gill net, 21 Hr = 0.0 N/A N/A 13.5 132 8.37 265 Meadow Lake Jun-05 Helena NF,T16N R9S18. Option 1 : Meadow Creek Option 2: Indian Meadows Option 1:10 Option 2:11.7 Glacial valley 5800' 44 15' 1,Well Established Medium 2 RBT 1 gill net, 16Hr = 0.13 RBT 8.1-14.5 Avg = 10.3 4.4 91 8.03 181 Middle Cottonwood Lake Jul-06 LotoNF,T16N R14WS3D/10A N/A Glacial Valley 4835' 10.3 30.5 One, primitive at boat launch Medium 10 RBT 1 gill net 20.75 Hr = 048 WSCT 8 RBT 7.0-14.6 AVG = 10.1 21 111 8.78 224 Monture Lake #1 Jun-05 LoloNF,T18N R12WS17. Monture Creek 14.5 Glacial Cirque 7217' 5.5 48.7' A few places to pitch camp on eastern shore Very Light 2WCT 1 gill net 17Hr = 0.12 WSCT 7.6 and 9.1 23.5 3 6.84 5 Monture Lake #2 Jun-05 LotoNF,T18N R12WS17. Monture Creek 15.25 Glacial Cirque 7709' 6.9 18.4' None None None 1 gill net 17.5 Hr = Ofish N/A N/A 9.3 1 7.55 5 Monture Lake #3 Jun-05 LoloNF,T18N R12WS18. Monture Creek 16.75 Glacial Cirque 7641' 4.5 18.4' None None None 1 gill net 17.5 Hr = N/A N/A 12 2 6.65 5 Nevada Reservoir May-06 Powell County, T12N R10W S13A/B/D,14A,18 C,19B. N/A Man Made 4615' 350 65.0' One, primitive at boat launch Moderate/ Heavy WSCT RBT YP Set times were not recorded N/A 5.0-14.0 7.2-16.6 5.2-10.1 5.6 89 8.18 179 Parker Lake Jul-05 Helena NF,T16N R9WS9 Indian Meadows 8.5 Glacial Valley 6000' 18.9 6.2' 2, one at the base of each peninsula Moderate 54YCT 1 gill net, 18Hr = 4.17 Hybrids YCTx WSCTx RBT 5.9-14.9 Avg = 10.6 6.2 142 8.36 291 Two Point Lake Jul-05 Helena NF,T16N R9WS10 Indian Meadows 8.5 Glacial Valley 6187' 9.5 10.5 None Very Light None 1 gill net 19Hr = N/A N/A 10.5 129 9 258 Upper Copper Lake Jun-06 Lewis and Clark County, 11 5N R9W S8D,9C,16B Option 1 : Stonewall Mt Option 2: end of Copper Cr Rd Option 1:3.75 Option 2:2 Glacial Cirque 77.06- 11.1 36.6' None Low 37WCT 1 gill net, 19.75 Hr = 1.87 WSCT 6.1-154 AVG = 1 1 .3 36.6 8 8.19 18 Upper Cottonwood Lake Jul-06 LoloNF,T16N R14W S3C/D. Glacial Valley 4867' 3.8 53.2' None Low 9 RBT 1 gill net, 20.5 Hr = 0.44 WSCT 8 RBT 8.6-17.0 AVG = 13.3 5.2 117 8.96 236 Upper Twin Lake Jun-05 Helena NF,T16N ROW S5/8. Option 1 : Meadow Creek Option 2: Indian Meadows Option 1:13 Option 2:10.75 Glacial valley 5969' 6.3 10.4' 1, open area on Virest shore Light None 1 gill net 13.25 Hr = N/A N/A 104 150 8.57 300 Upsata Lake Apr/May- 07 Powell County, T15N R13W S2,3,10 N/A Isolated basin 4130' 90.6 43.0' 4 at FAS Moderate/ Heavy YP LMB NP Set times were not recorded N/A N/A 11.8 197 8.26 394 Webb Lake Jul-05 Helena NF,T16N R9WS14 Indian Meadows 6.5 Glacial Valley 6079' 6 5.5' Numerous places to pitch camp at Webb Lake Guard Station Heavy 5WCT 2 gill nets 32.5 Hr = 0.15 WSCT ™th some RBT markers 6.5-14.5 Avg = 8.5 5.5 109 9.25 216 217 Braziel Lake Nevada Creek US fflghway 141 , "V Braziel Lake BLM section Description: In the upper Braziel Creek drainage, Braziel Lake is a small (7.4 acres) kettle lake (elevation of 5,141') and is located 3.2 miles due east of Hoodoo Mountain and 1.2 miles southwest of Nevada Lake (Reservoir) in the eastern Garnet Mountain Range (see above map). Private ranch surrounds the lake except for the very west- southwest side of the lake (section 22), which is Bureau of Land Management land (BLM). Location: T12N, RlOW, Sections 14, 15, 22 and 23; Latitude N47.78848°, Longitude Wl 12.83057°; Nearest Town: Helmville. Pubic Access: The very comer of section 22 of BLM land touches Braziel Lake (see above map), which appears to limit public access to this single site. "Comer-crossing" at section lines is not recommended. Recreationists are advised to obtain permission from the adjacent landowners to reduce conflict and ensure compliance with applicable access laws and rules. Camp Sites and Use: Because the majority of lake's perimeter is privately owned land, there are no camping areas on Braziel Lake. Angling Opportunity: Our 2006 gill net survey found no fish in Braziel Lake. However, the lake was recently planted by adjacent private landowner. Many areas of the immediate shoreline are marshy with cattails, sedges and rushes. Stocking History: The lake has been periodically stocked by private landowners, most recently in 2007 Angling Pressure: Unknown. Other Nearby Lakes: Nevada Lake (Reservoir) is only other lake in the immediate vicinity of Braziel Lake. Nevada Reservoir can be access from Highway 141 approximately 10.6 miles southeast of Helmville or 20.2 miles northwest of Avon. 218 Braziel Lake: Biological & Physical Information Date Sampled: 10/18-19/2006 Water Code: None Game Fish Present: WSCT Other Fish Species Present: None Size Captured: NA Trout Condition (Wr): NA SampHng Methods: Sinking Gill Net Gill Net Catch Rate: NA Natural Recruitment: NA Amphibians Observed: None Management Objectives: Explore opportunities to improve public access to the lake. Consider stocking program that provides for limited public recreation. Currently Stocked: Yes (by private landowner) Last Stocked: 2007 (private landowners) Species: Westslope Cutthroat Trout Recommended Stocking Frequency: NA Water Chemistry: pH: 8.36 TDS: 101 ppm Conductivity: 203 uS Secchi Depth: 2.0 ft Brazjel Lake ' Sbtfeih^ GIN mtt LMoihan 219 Browns Lake Description: Browns Lake is a large (549.9-acre) "pothole" lake located in the "knob and kettle" topography of Kleinschmidt Flats near the middle Blackfoot River watershed. The lake sits at an elevation of 4,294' and is bordered primarily by private ranchlands except for about 80-acres of state land on the northeast comer of the lake. Portions of the eastern shoreline are owned by USFW for waterfowl production. Ward Creek enters at the northeast comer and exits at the southwest comer of the Browns Lake with estimated flows ranging from 0.5-5cfs. Location: T14N, RllW, Sections 16,17,20,21 & 29; Latitude N46.9523°, Longitude W113.0110°; Nearest Town: Ovando. Public Access: Access to Browns Lake can be reached from Ovando or Helmville by using the Ovando-Helmville Road then tuming east for 4 miles on the Browns Lake Road. Access can be obtained from US Highway 200 by tuming south for 3 miles on the Browns Lake Road. Browns Lake is signed from Ovando and Highway 200. Camp Sites and Use: Browns Lalce has an established campground under lease and maintained by Montana Fish, Wildlife and Parks. This "fee" area has 15 camp sites with fire rings, two boat ramps and toilets. There are numerous other undeveloped (primitive) camp sites on the eastem shore of the lake along Browns Lake Road. Angling Opportunity: Browns Lake is a productive "put-and-take" rainbow trout fishery. The lake also supports an abundant number of longnose suckers. Primary means of fishing is done from a boat or through the ice. Shoreline topography lends well to shoreline angling from most of the lake's perimeter. Stocking History: From 1953 through 2006, annual stocking has occurred with primarily rainbow trout augmented with periodic stockings of westslope cutthroat trout in 1959, 1989, 1994 and 1998 and Coho salmon between 1953- 55, and 1968-69. Currently Browns Lake is stocked annually with -50,000 rainbow trout. Angling Pressure: From March 2005 to Febmary 2006, estimated angling pressure for Browns Lake supported 10,078 angler-days. Other Nearby Lakes: Kleinschmidt Lake (a fishless lake surrounded by private land), approximately one mile to the northwest is the only other lake in the immediate vicinity of Browns Lake. 220 Browns Lake: Biological & Physical Information Date Sampled: 9/9-10/2007, 5/10, 16, 17 & 10/2/2006, 9/8-1 1/2004 Water Code: 04-6210 Game Fish Present: Rainbow trout Other Fish Present: Longnose sucker, redside shiner Size Captured: Mean 14.8 inches (range 6.2 - 22.8 inches) Trout Condition Factor (Wr): Mean 113 + 16.1 (range 42 - 157) Amphibians Observed: Spotted frogs (adults & larvae), painted turtles Samphng Methods: Floating and Sinking Gill Nets Gill Net Catch Rate: 0.9/trout/net/hr Natural Recruitment: NA Management Objectives: Continue existing annual stocking fishery and investigate alternative strains and timing of rainbow trout plants. Currently Stocked: Yes Last Stocked: Stocked annually Species: RBT Recommended Frequency: Annually Water Chemistry: pH: 8.50 TDS:166ppm Conductivity: 334 uS Secchi Depth: 22.6 ft Browns Lake Rainbow Trout Size Distribution 2007 a 0.25 re O •S 0.20 c O t 0.15 o Q. e 0.10 Q. 0.05 0.00 Stocking History Year Species # Stocked 2006-1932 periodically RBT 2,779,083 1998 -1959 periodically WSCT 72,856 1969-1940 periodically Coho 327,590 1937 Kokanee 63,200 1931, 1928 Chinook 73,000 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Total Length (inches) 221 Browns Lake LLI 0:113010+469583 Area= 549.9 acres Volume = 6023,9 acne/feet Contour Interval = 5 feet MaK. Deplti = 27.4 ^— PkHiciiiy Oill Ntfi LiH^ulkHib s striking Qill Met Lctcatlons 111 Camp Lake X,. Deconimissiolned Ti-ail / ' ' ■ I, I 4, ^ Canyon'-L ake; '- ■ ''^^ ^W-^-fe^^ ;.^uvi! :! /'iivt' ' '^ I I .5r^ ^ 1 e im IU3 iSo Sod ] Iniedi -^ 235 Lower Twin Lake Lower Twin Lake ''^., ■> ' it R 1 r h T Upper Twin Lake /" "' ^ "^ . ■^. i Description: Lower Twin Lake is a small (8.6 acres) glacial lake located -2.4 miles north-northeast of Bugle Mountain and 2.7 miles west-southwest of Pyramid Peak within the Scapegoat Wilderness. The lake lies on the Helena National Forest (Lincoln Ranger District) at 5,900' in East Fork of the North Fork Blackfoot River drainage. Lower Twin Lake's outlet feeds Lost Pony Creek. Location: T16N, R9W, Sections 6; Latitude N47. 17288°, Longitude Wl 12.78843°; Nearest Town: Lincoln. Public Access: There are two options to access Lower Twin Lake. Option 1: Meadow Creek Trailhead, near the northeast comer of the Kleinschmidt Flats east of Ovando take USFS Trail #483 (Meadow Creek Trail) 12 miles to its intersection with USFS Trail #481 (Main Line Pack Trail) then travel 0.25 miles northwest to the Twin Lakes Trail #425 and follow it 1.5 miles passed Upper Twin Lake to Lower Twin Lake. Option 2: hidian Meadows Trailhead, east of Lincoln in the Landers Fork and Copper Creek drainage, take the USFS Trail #481 (Main Line Pack Trail) northwest about 10.5 miles to the USFS Trail #425 (Twin Lakes Trail) and follow it another 1.5 miles to the lake. Both trails leading to Lower Twin Lake have a moderate degree of difficultly. Camp Sites and Use: No campsites were observed on Lower Twin Lake, recreating is essential in this wilderness area. "Leave no trace" camping and Angling Opportunity: Lower Twin Lake currently supports self-sustaining hybridized population of westslope cutthroat, Yellowstone cutthroat and rainbow trout. Areas of the shoreline topography are lined with sedges, dense brush and lodgepole pine; however, there are some areas that lend well to shoreline angling. Stocking History: Fish planting records show Lower Twin Lake was planted once in 1950 and twice in 1952 with undesignated cutthroat trout. Genetics analyses confirm rainbow trout were also introduced at some point. Angling Pressure: Very light Other Nearby Lakes: Upper Twin Lake is in the very near vicinity of Lower Twin Lake, 0.9 miles southeast along Twins Lakes Trail (USFS Trail #425) on the way into Lower Twin Lake. If you accessed Lower Twin Lake using the Main Line Pack Trail (USFS Trail #481) from hidian Meadows Trailhead, you can back-track 2.9 miles southeast to Parker Lake. Meadow Lake is 4.0 miles south-southwest of Lower Twin Lake. To access Meadow Lake will require you to back-track to the Main Line Pack Trail (USFS Trail #481) before continuing on Meadow Creek Trail (USFS Trail # 483). 236 Lower Twin Lake: Biological & Physical Information Date Sampled: 6/21/2005 Water Code: 04-6900 Sampling Methods: Sinking Gill Net Gill Net Catch Rate: 3.6trout/net/hr Natural Recruitment: Present Game Fish Present: Hybridized Cutthroat-Rainbow Trout Other Fish Present: None Size Captured: Mean 12.3 inches (range 5.7 - 23.6 inches) Trout Condition (Wr): Mean 82 + 21.78 (range 25 - 1 1 1) Genetics: All fish possessed genetic markers for Yellowstone and westslope cutthroat trout and rainbow trout. Genetic analyses identify two somewhat reproductive ly isolated populations of hybrid rainbow trout and hybrid Yellowstone cutthroat trout Amphibians Observed: Spotted frogs (adults), tadpoles and eggs present, western toad (adult) Management Objectives: Identify opportunities to convert to WSCT Currently Stocked: No Last Stocked: 1952 Species: CT Recommended Frequency: NA Water Chemistry: pH: 8.70 TDS:112ppm Conductivity: 226 uS Secchi Depth: NA Stocking History Lower Twin Lake Size Distribution 2005 ' 1 1 1 1 1 MM lllll 1 1 1 Year Species # Stocked 1952 CT 3,120 1952 CT 6,864 1950 CT 3,600 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Total Length (inches) Lower Twin Lake Water Code: 046900 LLID: 1127894471730 Survey Date: 6/21. '2005 Surface Area: 8,6 acres Volume: 29.5 acre-feet Max. Depth: ll.Sfeet Contour Interval: 2 feet 237 Maddie Lake i^ ■y ^ \ Maddie Lake \ rA Dry Creek Road y^' i I ■jyT" Description: Maddie Lake is a small (5.7 acres) kettle lake located -4.1 miles north-northeast of Marcum Mountain. The lake is part of The Nature Conservancy land holdings, located about 1.1 miles east of Kleinschmidt Flats off the Dry Creek Road. It lies at 4,560' in the Rock Creek drainage, a tributary to the North Fork Blackfoot River. Location: T15N, RlOW, Sections 30; Latitude N47.02429°, Longitude Wl 13.91551°; Nearest Town: Ovando. Public Access: Approximately 8.3 miles east of Ovando along Highway 200, follow the Dry Creek Road north 6.5 miles to a Nature Conservancy gate, walk in area only. The lake is 1.0 mile from the gate and requires bush- whacking from a second Nature Conservancy gate off the main road into the lalce. The lake is on Nature Conservancy land. Camp Sites and Use: No camping areas were observed around Maddie Lake. Because of its size, relative remote location and lack of sport fishing opportunities, the lake experiences light use. Angling Opportunity: The 2006 gill net survey found no game fish within Maddie Lake although red-side shiners are present in abundance. The immediate shoreline is marshy with floating vegetation mats and overgrown with alder and red-osier dogwood. The surrounding topography is a mixture of aspen and coniferous forest with an alder, willow and dogwood under-story. Stocking History: NA Angling Pressure: NA Other Nearby Lakes: There are a series of small kettle lakes in the immediate vicinity of Maddie Lake. Most are grown in with sedges and rushes. Tupper Lake lies 0.3 miles to the southeast and lies entirely on private land with no access. Coppers Lake is located -7.0 road miles north of Maddie Lake on the Dry Creek Road. 238 Maddie Lake: Biological & Physical Information Date Sampled: 6/7-8/2006 Game Fish Present: None Other Fish Present: Redside shiners Size Captured: NA Trout Condition Factor (Wr): NA Water Code: 04-6950 Samphng Methods: Sinking Gill Net Gill Net Catch Rate: NA Natural Recruitment: NA Amphibians Observed: Painted turtles Management Objectives: Consider future WSCT stocking if public access is secured. Currently Stocked: No Last Stocked: None known Species: NA Recommended Frequency: NA Water Chemistry: pH: 8.37 TDS:132ppm Conductivity: 265 uS Secchi Depth: 13.5 ft Maddie Lake LLIO 1139150470243 Area = 5.7 acres; \ft3tume= f{}2 6 acna^eat ContioL* IrJerval - B feet MiK. D«piri = 33.? f»9[ -f ■ SJnrk^ng CHU tfet tocBHon a Aa ^ ^30 i^ f<^..ia4ii,\yv\rx^ Description: One of a cluster of three un-named lakes, Monture Lake #3 is a small (6.0 acre) high elevation (7,641 ft) glacial cirque lake located on the Lolo National Forest (Seeley Lake Ranger District) on the south-southwestern slope of Moser Mountain, just south of the Bob Marshall Wilderness. The outlet stream, together with that from Monture Lake #1, forms the headwaters of Middle Fork Monture Creek in the upper Monture Creek drainage. Monture Creek is a major tributary to the middle reach of the Blackfoot River. Location: T18N, R12W, Sections 18; Latitude N47. 31356°, Longitude W113. 17903°; Nearest Town: Ovando. Access: Access to Monture Lake #3 can be obtained by traveling north 8.0 miles from Highway 200 at Ovando on USFS Road 89 to the Monture Creek Campground/Trailhead. From the trailhead, travel along USFS Trail #25 which changes to USFS Trail #27 (Monture-Hahn Pack Trail) at trail mile 1.6 near Falls Creek. Continue north along USFS Trail #27 for 12.6 miles and 1.1 mile above the Monture Creek Falls to USFS Trail #37 IW. Follow USFS Trail #37 IW northeast up the Middle Fork of Monture Creek about 2.5 miles to the first tributary stream flowing down from the northwest. Follow this small tributary upstream 2.0 miles to the lake. There is no established trail leading to the lake. An alternate route is to continue following USFS Trail #37 IW to Monture Lake #1, a distance of 3.7 miles from the USFS Trail #27. From Monture Lake #1, an ascent up and over a finger- ridge of Moser Mountain that separates the two cirque lake basins is required. Depending on the route followed, the average distance from trailhead to Monture Lake #3 is approximately 17.3 miles. Trails #25 and #27 are categorized as mainline (primary) trails and are well maintained with moderate ascents and difficultly. Trail #37 IW is a primitive trail with areas of steep ascents and more difficult travel. Camp Sites and Use: No camping areas were observed around Monture Lake #3. Angling Opportunity: Our 2005 gill net survey identified Monture Lake #3 as fishless. The shoreline on the southern side of the lake is relatively flat, brushy and timbered with sub-alpine firs. The remaining shoreline surrounding the lake is rugged, steep, rocky talus slopes and lightly timbered. Stocking History: No history offish plants. Angling Pressure: NA Other Nearby Lakes: There are two other cirque lakes in the vicinity of Monture Lake #3. Monture Lake #1 is 425' lower in elevation than lake #3 (7,217ft), and it is found about 1.4 miles east on the western slope of Foolhen Mountain and is accessed from the USFS Trail 37 IW that leads up to the lake. Monture Lake #2 is 1.3 miles northeast of Monture Lake #3 on the southeastern slope of Moser Mountain (elevation of 7,700') and access to it requires an ascent from Monture Creek Lake # 1 . There are no established trails between the three lakes and only 248 the one primitive trail leads to Monture Lake #1. Travel between the lakes is difficult and requires good physical health. Monture Lake #3: Biological & Physical Information Date Sampled: 6/28/2005 Water Code: 04-6987 Game Fish Present: None Other Fish Present: None Size Captured: NA Trout Condition Factor (Wr): NA Amphibians Observed: Spotted Frogs (adults, eggs), Long-toed Salamander Samphng Methods: Sinking Gill Net Gill Net Catch Rate: NA Natural Recruitment: NA Management Objectives: No change - manage as fishless. Currently Stocked: No Last Stocked: NA Species: NA Recommended Frequency: NA Water Chemistry: pH: 6.65 TDS: 2 ppm Conductivity: 5 uS Secchi Depth: 12.0 ft Monture Lake #3 Water Code; 046987 LLID: 1131783473140 Survey Dote: 6'28j2005 Surface Area: 6,0 acres Volume: 37,2 acre-feet Max. Depth: 18.4 feet Contour Interval: 5 feet ^— Siiikiiifj Gill Net Location 249 Nevada Lake Reservoir ^^^3WyiV\ .\, \ \ ^\—,' ^^^Sfe^ ir /^ J ^R ^ 5^r f^^ ^^' 1^ 1 y \ \^2? 12^^^J ■) Jilapjl \)Pt^° ^?oix 470 940 1,410 1,8&0 ^. -• . A 'i .^- "' ^ - ^^ V. -4 Description: Upper Twin Lake is a small (6.3 acres) glacial lake located -2.6 miles east-northeast of Bugle Mountain and 2.5 miles south-southwest of Pyramid Peak within the Scapegoat Wilderness. The lake lies on the Helena National Forest (Lincoln Ranger District) at 5,967' in the East Fork of the North Fork Blackfoot River drainage. Location: T16N, R9W, Sections 5 & 8; Latitude N47. 16432°, Longitude Wl 12.77245°; Nearest Town: Lincoln. Public Access: There are two options to access to Upper Twin Lake: Option 1: From Meadow Creek Trailhead, east of Ovando, take USFS Trail #483 (Meadow Creek Trail) 12 miles to its intersection with USFS Trail #481 (Main Line Pack Trail) then travel -0.25 miles northwest to the Twin Lakes Trail #425 and follow it 0.6 miles to the lake. Option 2: From the Indian Meadows Trailhead, east of Lincoln in the Landers Fork and Copper Creek drainage, take the USFS Trail #481 (Main Line Pack Trail) northwest 10.5 miles to the USFS Trail #425 (Twin Lakes Trail) and follow it 0.6 miles to the lake. Both trails leading to Upper Twin Lake have a moderate degree of difficultly. Camp Sites and Use: One primitive campsite was observed at Upper Twin Lake on the northern end of the lake near the small inlet stream. "Leave no trace" camping and recreating is essential in this wilderness area. Angling Opportunity: Gill net surveys identified Upper Twin Lake as Ashless. Stocking History: Fish planting records show that Upper Twin Lake was stocked four times, beginning in 1943 with unspecified cutthroat trout. In 1969 and 1971, Yellowstone cutthroat were also planted. The most recent stocking occurred 1988 with westslope cutthroat trout. Angling Pressure: NA Other Nearby Lakes: In the vicinity of Upper Twin Lake is Lower Twin Lake located -0.9 miles northwest along the Twin Lalces Trail #425. If you accessed upper Twin Lake via the Main Line Pack Trail (USFS trail #481) from the Indian Meadows Trailhead, you can retrace it back approximately 2.0 miles southeast to Parker Lake. If you accessed Upper Twin Lake area using the Meadow Creek Trailhead (USFS Trail #483), Meadow Lake is -2.9 miles southwest of Upper Twin Lake on the return trip to the trailhead. 261 Upper Twin Lake: Biological & Physical Information Date Sampled: 6/20/2005 Water Code: 04-7530 Game Fish Present: None Other Fish Present: None Size Captured: NA Trout Condition Factor (Wr): NA Samphng Methods: Sinking Gill Net Gill Net Catch Rate: NA Natural Recruitment: NA Amphibians Observed: Spotted frogs (adults) Management Objectives: The lake can not sustain natural reproduction and appears marginal from the perspective of winter survival. Recommend the lake remain Ashless. Currently Stocked: No Last Stocked: 1988 Species: WSCT Recommended Frequency: NA Water Chemistry: pH: 8.57 TDS:150ppm Conductivity: 300 uS Secchi Depth: 10.4 ft Stocking History Year Species # Stocked 1988 WSCT 3,990 1971 YCT 1,035 1969 YCT 990 1943 CT 4,000 Upper Twin Lake WSlSr C&dc: 047630 LLID^ 1127724471642 Survey Data: St2f>S2eOS Surface Ar-ea: 7,0 acres Volume: 25.7 acre-feet Max. Deptti: 10.4 feet Cwitcuf (nlerval: iteei — Sinking Gill Net Location 262 Upsata Lake Upsata Lake Upsata Lake Road Woodworth Road 1^ Highway 200 i Description: Upsata Lake is a 90.6 acre glacial "pothole" lake (elevation of 4,130') located in the "knob-kettle" topography of the middle Blackfoot River watershed. Upsata Lake is 6.1 miles south of Dunham Point and -11.2 miles east-northeast of Ovando. Location: T15N, R13W, Sections 2, 3 & 10; Latitude N47.07795°, Longitude Wl 13.22162°; Nearest Town: Ovando. Public Access: Upsata Lake lies along Upsata Lake Road, which is easily accessed from the Woodworth Road. From Highway 200, take Woodworth Road northwest 3.3 miles, then turn right (east) on Upsata Lake Road for 0.9 miles to the Upsata Lake Fishing Access Site. Camp Sites and Use: Four campsites and a boat launch are located at the Upsata Lake Fishing Access Site. These camp sites receive moderate to heavy use. Angling Opportunity: Upsata Lake supports an abundant population of "stunted" yellow perch and low numbers of largemouth bass and northern pike. Upsata Lake is surrounded by private property with the exception of the Fishing Access Site at the southwest end of the lake. Stocking History: From 1959 to 1992, Upsata Lake was on a yearly stocking schedule of rainbow trout. Starting in the early 1990's, largemouth bass were also stocked annually to curb rising populations of yellow perch that were illegally introduced. Upsata Lake is now periodically supplemented with adult largemouth bass. Angling Pressure: From March 2005 to February 2006 angling pressure estimated at 770 angler-days per year. Other Nearby Lakes: The Cottonwood Lakes lie along USFS Road 477 (Cottonwood Lakes Road) to the northwest of Upsata Lake approximately 12.0 miles. Browns Lake is southeast of Upsata Lake 18.2 miles and can be accessed from Highway 200, east of Ovando, or from the Ovando-Helmville Road, south of Ovando. 263 Upsata Lake: Biological & Physical Information Date Sampled: 8/7/2006, 4/24/2007 and 5/2/2007 Game Fish Present: Yellow Perch, Largemouth Bass and Northern Pike Other Fish Present: Redside shiners Size Captured: NA Trout Condition (Wr): NA Amphibians Observed: Spotted Frogs (adults) and Painted Turtles Water Code: 04-7560 Samphng Methods: NA Gill Net Catch Rate: NA Natural Recruitment: Present Management Objectives: Status quo Currently Stocked: No Last Stocked: 1992 Species: LMB Recommended Frequency: NA Water Chemistry: pH: 8.26 TDS:197ppm Conductivity: 394 uS Secchi Depth: 11.8 ft Stocking History Histogram: No Data Year Species # Stocked 1988-1991 LMB 3,613 1959-1992 RBT 371,921 Lake Upsata LLID: 11 £630446763 J Ar«a= gO-6a5fes VMijme= 1,185 dcraffest ContaLir IntervBJ = 5 feet Max.. DesPi ^ 43 tee: 2» 4M iw am 264 Webb Lake _^, Webb Lake Trail #481 Description: Webb Lake is a shallow (24.7 acres) glacial valley trough lake located ~ 2.8 miles northeast of Red Mountain at an elevation of 6,079' . The lake lies within the Scapegoat Wilderness, Helena National Forest (Lincoln Ranger District). Its outlet flows into Ringeye Creek, a tributary to the Landers Fork of the Blackfoot River. Location: T16N, R9W, Sections 14; Latitude N47. 14522°, Longitude Wl 12.69466°; Nearest Town: Lincoln. Public Access: Access to Webb Lake can be obtained by taking Highway 200 east from Lincoln 6.2 miles to the Copper Creek Road. Follow the Copper Creek Road 10.2 miles to the Indian Meadows Trailhead. From the trailhead follow USFS Trail #481 (Main Line Pack Trail) 6.5 miles to Webb Lake. The trail is in very good condition with moderate difficultly. Camp Sites and Use: Because of its relatively short distance within the Scapegoat Wilderness boundary on a well maintained trail system, Webb Lake receives relatively heavy recreational use. Camping opportunities exist in the vicinity of the inlet (near USFS Webb Lake Guard Station) located at the northwest end of the lalce. Pack stock is not allowed within 200ft of the shore except on the trail. "Leave no trace" camping and recreating is essential in this wilderness area. Angling Opportunity: Webb Lake supports a small hybrid cutthroat trout population through natural reproduction. Many areas of the shoreline topography are brushy with sedges and rushes extending out into the lake. The adjacent topography is heavily timbered with moderately steep slopes; however, there are areas that lend well to shoreline angling. Stocking History: Webb Lake was stocked with undifferentiated cutthroat trout between 1940 and 1952. Angling Pressure: Moderate Other Nearby Lakes: In the vicinity of Webb Lake are three other lakes. Approximately 4.8 miles from the Indian Meadows Trailhead, Heart Lake lies along a short spur trail (USFS Trail #424) that branches off the Main Line Pack Trail (USFS Trail #481) at trail mile 3.9 then reconnects back into the Main Line Pack Trail at mile 4.8, 2.1 trail miles to the southeast of Webb Lake. Continuing northwest along USFS Trail #481 west of Webb Lake, but 0.6 miles before Parker Lake, Two Point Lake is 0.5 miles north of the Main Line Pack Trail along USFS Trail #479. Parker Lake lies 2.3 miles west of Webb Lake along USFS Trail #481, or 7.0 miles from Indian Meadows Trailhead. Travel is moderately difficult to these other lakes. 265 Webb Lake: Biological & Physical Information Date Sampled: 7/19/2005 Water Code: 04-7590 Game Fish Present: Hybridized Westslope Cutthroat Trout Other Fish Present: None Size Captured: Mean 8.5 inches (range 6.4 - 14.5 inches) Trout Condition Factor (Wr): Mean 90 + 14.6 (range 67 -106) Genetics: Westslope cutthroat trout hybridized with rainbow trout. SampHng Methods: Sinking Gill Net Gill Net Catch Rate: 0.15troufnet/hr Natural Recruitment: Present (limited) Amphibians Observed: Spotted Frogs Management Objectives: Recommend no change Currently Stocked: No Last Stocked: 1952 Species: CT Recommended Frequency: NA Water Chemistry: pH: 9.25 TDS:109ppm Conductivity: 216 uS Secchi Depth: 5.4 ft Stocking History Webb Lake Cutthroat Trout Size Distribution 2005 0.05 0.00 Year Species # Stocked 1940-1952 CT 34,485 8 9 10 11 12 Total Length |inches) Inlet Webb Lake Water Co de:Q 47590 LLID; 1126949471452 Survev Date: 7/1 9..'20O5 Surface Area; fi.7 acres Volume; 14,8 acre-feet Max. Deptli: 5.4 feet Contour Interval: 2 feet 266 RECOMMENDATIONS - Identify a sustainable fisheries technician-funding source in order to continue the current FWP fisheries restoration program at the level outlined in this report. The need stems from the loss of the Milltown Mitigation Funds in 2009. - Encourage watershed groups and resource agencies that promote and develop fisheries restoration and grazing plans to implement a grazing monitoring plan to better ensure fisheries restoration projects are successful. - Expand on the ground restoration to the Clearwater River Basin with support provided through watershed groups including the Blackfoot Challenge, Big Blackfoot Chapter of Trout Unlimited, Clearwater Resource Council as well as other supporting agencies and organizations. - Complete restoration projects in all bull trout "core areas" and current restoration streams. Expand restoration to the upper Blackfoot and Clearwater Basin with emphasis placed on native fish priority streams. - Continue to monitor the spread and impacts of whirling disease and the results of restoration on infection rates. Examine the susceptibility of whirling disease on mountain whitefish. Incorporate pertinent results into the restoration program. - Increase landscape protection on critical fish and wildlife habitat in cooperation with the Montana Land Reliance, Nature Conservancy, US Fish and Wildlife Service, Montana Fish, Wildlife and Parks, Blackfoot Challenge and Plum Creek Timber Company and extend protective measures to critical waters in the Clearwater River Basin. - Continue fish populations monitoring at the Johnsrud and Scotty Brown Bridge section of the Blackfoot River, and major tributary restoration projects as funding allows. - Examine methods of converting rainbow trout above the North Fork Falls to WSCT. - Increase FWP enforcement efforts in bull trout spawning and staging areas. Initiate additional bull trout regulation protection measures as necessary - Complete the cleanup of the Mike Horse mine in a manner that allows the recolonization of WSCT, and develop a post-project monitoring program that measures metal uptake in aquatic biota. 267 ACKNOWLEDGEMENTS The 2006-07 fisheries fieldwork was made possible through a Northwestern Energy (Milltown Mitigation) grant, for which we especially thank Sam Milodragovich and the Milltown Technical Advisory Committee. We extend our thanks to the Backcountry Horsemen (Randy Kappas and Kirk Sybrant) and the U.S. Forest Service (Helena and Lolo National Forests) for assisting with the backcountry work. The whirling disease work was made possible with the help of Dr. Lisa Eby (University of Montana), Wease Bollman (Rhrithron Associates), Dick Vincent (FWP) and David Kumlien (the Whirling Disease Foundation). The Big Blackfoot Chapter of Trout Unlimited helped fund the writing of this report. Both Ryen Aasheim (BBCTU) and Linneae Schroeer-Smith (FWP) helped improve the quality of this report. We extend a special thanks to all the landowners that cooperated with data collections, restoration projects and allowed us access to streams during the 2006 and 2007 field seasons. 268 LITERATURE CITED Anderson, R. A. 2004. Occurrence and seasonal dynamics of the whirling disease parasite, Myxobolus cerehralis, in Montana spring creeks. Master of Science thesis, Montana State University, Bozeman. Baldwin, T. J., E. R. Vincent, R. M. Silflow, D. Stanek. 2000. Myxobolus cerebralis infection in RBT {Oncorhynchus mykiss) and brown trout (Salmo trutta) exposed under natural stream conditions. Journal of Veterinary Diagnostic Investigations 12:312-321. Brown, C, G. Decker, R. Pierce and T. Brant. 2001. Applying natural channel design philosophy to the restoration of instating native fish habitat. Practical approaches to conserving inland native fish. Western Division AFS Symposium, Missoula, Montana. Dunne, T., and L. B. Leopold. 1978. Water in Environmental Planning. Freeman and Co. New York. Fish Analysis Plus (FA+) Montana Fish, Wildlife and Parks fisheries software. Fitzgerald G. 1997. Analysis and inventory of riparian vegetation along Nevada and Monture Creeks using ADAR imagery. MS. Thesis University of Montana, Missoula. Fredenberg, F. 1992. Evaluation of electrofishing-induced spinal injuries resulting from field electrofishing surveys in Montana. Montana Department of Fish, Wildlife and Parks, Bozeman, Montana. Leathe, S. 1983. Inter-office memo of the two-pass depletion estimator. Montana Fish, Wildlife and Parks, Helena. Lewis, M. 1805. Lewis and Clark Expedition journal entry 13 June. Lockwood R.N. and J. C. Schneider. 2000. Stream fish population estimates by mark-and-recapture and depletion methods. Chapter 7 in Schneider, James C. (ed.) 2000. Manual of fisheries survey methods II: with periodic updates. Michigan Department of Natural Resources, Fisheries Special Report 25, Ann Arbor. Mclntyre, J. D., and B. E. Reiman. 1995. Westslope WSCT. Pages 1-15 in M. K. Young, editor. Conservation assessment for inland WSCT. U. S. Forest Service General Technical Report. RM-256. Nehring, R. B., and J. P. Goettl Jr. 1974. Acute toxicity of a zinc-polluted stream to four species of salmonids. Bulletin of Environmental Contamination and Toxicology 12(4):464-469. Peters, D. J. and R. Spoon. 1989. Preliminary inventory of the Big Blackfoot River. Montana Department of Fish, Wildlife and Parks, Missoula, Montana. Peters, D. J. 1990. Inventory of fishery resources in the Blackfoot River and major tributaries of the Blackfoot River. Montana Fish Wildlife and Parks, Missoula, Montana. Pierce R. and D. J. Peters. 1991. Aquatic investigation in the middle Blackfoot River, Nevada Creek and Nevada Spring Creek corridor. Montana Department of Fish, Wildlife and Parks, Bozeman, Montana. Pierce, R. 1991. A stream habitat and fisheries analysis for six tributaries to the Blackfoot River. Montana Department of Fish, Wildlife and Parks, Missoula, Montana. Pierce, R., D. Peters and T. Swanberg. 1997. Blackfoot River restoration progress report. Montana Fish Wildlife and Parks, Missoula, Montana. Pierce, R., and D.A. Schmetterling. 1999. Blackfoot River restoration project progress report, 1997-1998. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce, R. and C. Podner. 2000. Blackfoot River fisheries inventory, monitoring and restoration report. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce, R. C. Podner and J. McFee. 2001. Blackfoot River fisheries inventory, monitoring and restoration report. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce, R., C. Podner and J. McFee. 2002. The Blackfoot River fisheries inventory, restoration and monitoring progress report for 2001. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce, R., R. Anderson and C. Podner. 2004. The Big Blackfoot River Restoration Progress Report for 2002 and 2003. Montana Fish Wildlife and Parks, Missoula Montana. Pierce, R., R. Aasheim and C. Podner. 2005. An integrated stream restoration and native fish conservation strategy for the Big Blackfoot River basin. Montana Fish Wildlife and Parks, Missoula, Montana. Pierce, R and C. Podner. 2006. The Big Blackfoot River Fisheries Restoration Report for 2004 and 2005. Montana Fish, Wildlife and Parks, Missoula, Montana. Pierce, R., R. Aasheim and C. Podner. 2007. Fluvial westslope cutthroat trout movements and restoration relationships in the upper Blackfoot Basin, Montana. Intermountain Journal of Sciences Vol. 13(2). 269 Reiman. B., E, and D. Isaak, S. Adams, D. Horan, D. Nagel, C. Luce and D. Myers. 2007. Anticipated climate warming effects on bull trout habitats and populations across the interior Columbia River basin. Transactions of the American Fisheries Society, 13: 1552-565. Ricker, W. E. 1975. Computation and interpretation of biological statistics offish populations. Bulletin of the Fisheries Research Board of Canada, Bulletin 191. Ottawa, Canada. Rosgen, D. 1994. A classification of natural rivers. Catena 22: 169-199. Rosgen, D. 1996. Applied Fluvial Geomorphology. Wildland Hydrology, Pagosa Springs Colorado. Schmetterling, D.A. 2003. Reconnecting a fragmented river: movements of westslope cutthroat trout and bull trout after transport upstream of Milltown Dam, Montana. North American Journal of Fisheries Management 23:721-731. Schmetterling, D. A. 2001. Seasonal movements of fluvial westslope cutthroat trout in the Blackfoot River drainage, Montana. North American Journal of Fisheries Management 2 1 : 507-520. Shepard, B. B., B. E May and W. Urie. 2003. Status of westslope cutthroat trout (Onchorhunchs clarki lewisi) in the United States: 2002. A report to the Westslope Cutthroat Interagency Conservation Team. Shepard, B. B., B. E May and W. Urie. 2005. Status and conservation of westslope cutthroat trout within the Western United States. North American Journal of Fisheries Management 25: 1426-1440. Spence, L. E. 1975. Upper Blackfoot River study: A premining inventory of aquatic and wildlife resources. Montana Department of Fish and Game & the Anaconda Mining Company. 86 pp + appendices. Stratus Consulting. 2007. Preliminary evaluation of injuries and damages: Upper Blackfoot Mining Complex, Lewis and Clark County, Montana, A report to the Natural Resources Damage Claim, Montana Department of Justice, Helena, MT. Swanberg, T. R. 1997. Movement of and Habitat Use by Fluvial Bull Trout in the Blackfoot River, Montana. Transactions of the American Fisheries Society 126:735-746. Vincent, E. R. 2000. Whirling disease report 1997-98. Montana Fish, Wildlife and Parks. Project 3860. Helena, Montana. Vincent, E. R. 2002. Relative susceptibility of various salmonids to WD with emphasis on rainbow and cutthroat trout. WD: reviews and current topics. American Fisheries Society Symposium 29:109-115. 270 APPENDICES Appendix A: Summary of catch and size statistics for Blackfoot tributaries, 2006-07. Appendix B: Summary of two-pass estimates for Blackfoot tributaries, 2006-07. Appendix C: Mark and recapture estimates for the Blackfoot River, 2006. Appendix D: Summary of stream discharge measurements for 2006-07. Appendix E: Restoration streams and table of activities through 2007. Appendix F: Potential restoration projects in the Blackfoot drainage through 2007. Appendix G: Restoration streams and cooperators through 2007. Appendix H: Summary of water temperature in the Blackfoot drainage, 2006-07. Appendix I: Summary of water chemistry readings for 2006 and 2007. Appendix J: Westslope cutthroat trout genetic sampling sites and results, 2006-07 Appendix K: Blackfoot Basin restoration prioritization scorecard though 2007 271 Appendix A : Catch and size statistics for tributaries to the Blackfoot River excluding the Clearwater Basin, 2006-2007. stream River Date Mile Location (T,R,S) Sampled Section Length (ft) Species Number YOY{<4.0") YOY CPUE Total Number Captured 1st Captured 1st Range of Mean CPUE(#/100') (#/100') in 1st Captured Pass Pass Lengths (in) Length (in) in 1st Pass Pass Anaconda Creek 0.1 15N,6W,27B 25-JUI-06 330 CT EB 4 1 4 1 5.2 5.5 4.9-5.8 5.5 1.2 0.3 0.0 0.0 0.5 15N,6W,22C 25-JUI-06 346 CT 46 46 26 3.8 2.0-8.8 13.3 7.5 2 13N,16W,26A 9-Aug-07 390 No fish found Ashby Creel< 13N,16W,35B 31-JUI-06 300 CT EB Sculpins 75 1 present 59 1 Spotted frogs 34 present 1.5-8.4 5.1 3.1 5.1 9-Aug-07 351 No fish found Observed Spotted frogs 12N,16W,3A 31-JUI-06 300 CT EB Sculpins 79 2 present 62 2 Spotted frogs 30 present 1.4- 2.0- 7.5 2.1 4.1 2.1 9-Aug-07 300 CT EB Sculpins 122 27 present 97 22 Spotted frogs 49 21 present 1.4- 2.0- 7.6 5.6 4 2.6 19.7 0.3 20.7 0.7 32.3 7.3 11.3 0.0 10.0 0.0 16.3 7.0 Bear Creek lower river trib. 13N,16W,18B; 2-Aug-06 13N,16W,7C 374 RB CT LL EB Sculpins 241 2 15 16 abundant 180 2 13 15 141 5 7 1.0- 4.3- 1.9- 2.2- 7.5 6.0 9.8 7.5 2.5 5.1 5.2 3.8 i-Aug-07 374 RB LL EB Sculpins 254 59 14 abundant 201 46 14 118 32 4 1.5-9.0 2.3-12.0 2.4-7.0 3.2 3.5 4.6 48.1 0.5 3.5 4.0 53.7 12.3 3.7 37.7 0.0 1.3 1.9 31.6 8.6 1.1 Bear Gulch 0.1 13N,9W,34A 3-Aug-06 250 No fish Spotted frogs present 0.6 13N,9W,34A 3-Aug-06 250 CT Spotted frogs 19 present 19 9 3.0-6.0 4.1 7.6 3.6 0.7 13N,9W,34AB 20-JUI-06 450 CT 36 36 27 1.3-6.4 3.1 8.0 6.0 1.2 13N,9W,3B 20-JUI-06 480 CT 85 85 60 2.4-7.1 3.6 17.7 12.5 Beartrap Cr 0.2 1.2 15N,6W,27B 15N,6W,27C 25-JUI-06 25-JUI-06 400 325 No fish No fish Blackfoot River (above Hogum Cr Rd) (above Flesher Pass Rd) (above Pass Cr) (above Shave Cr) 119.6 14N,7W,5D 11-Sep-06 4000 DV CT LL EB MWF Sculpins 1 64 22 18 18 abuundant 1 48 15 13 15 Western toads 28 9 5 2 present 15.3 1.8-16.9 2.6-18.3 2.7-11.8 3.5-15.3 15.3 4.5 6.8 5.4 12.2 124.3 15N,7W,35B&26C 5-Sep-06 2457 DV CT LL EB LNS Sculpins 1 93 1 117 4 abuundant 1 75 1 91 3 Spotted frogs 63 64 1 common 12.4 1.9-9.9 2.9 2.1 -10.3 3.2-6.5 12.4 2.9 2.9 4.0 5.2 130.5 15N,6W,20A 26-Jul-06 725 EB Sculpins 19 present 19 Spotted frogs present 4.2 LNS 1.9-7.0 present 1318 15N,6W.21D 26-Jul-06 540 CT Spotted frogs 1 present 2.0 0.0 1.2 0.4 0.3 0.4 0.0 3.1 0.0 3.7 0.1 2.6 0,2 0.0 0.7 0.2 0.1 0.1 0.0 2.6 0.0 2.6 0.0 0.8 0.2 271 Appendix A : Catch and size statistics for tributaries to the Blacl(foot River excluding the Clearwater Basin, 2006-2007 (cont'd). stream Chamberlain Creek River Date Section Totai Number Miie Location (T,R,S) Sampied Length (ft) Species Captured Number YOY(<4.0") YOYCPUE Captured 1st Captured 1st Range of IVIean CPUE(#;iOO') (#;iOO') in 1st Pass Pass Lengths (in) Length (in) in 1st Pass Pass Bianchard Creei< 3.3 15N,15W,36AB 15-Jun-06 492 CT* 3 3 1 3.7-7.3 5.2 0.6 0.2 RB* 21 21 7 3.5-8.2 5.0 4.3 1.4 5.6 15N,15W,34B Spotted frogs present 15-Jun-06 492 CT 27 27 19 2.7-7.4 4.1 5.5 3.9 Spotted frogs & Western toads present 9.4 15N,16W,25D 19-Jun-06 410 CT 39 39 26 2.2-7.2 3.7 9.5 6.3 N.F.Bianchard Creei< 0.15 15N,14W,31B 15-Jun-06 492 RB* 4 4 3 2.4-4.3 3.4 0.8 0.6 EB 13 13 11 1.3-4.9 2.2 2.6 2.2 SCUL present Spotted frogs present 2 15N,15W,26A 19-Jun-06 492 CT* 10 10 5 3.2-5.5 4.1 2.0 1.0 RB* 10 10 7 2.7-6.0 3.9 2.0 1.4 EB 11 11 4 3.3-7.0 4.7 2.2 0.8 SCUL present 6.3 15N,15W,17A 19-Jun-06 492 CT 7 7 3 1.9-6.1 4.1 1.4 0.6 EB 7 7 1 3.9-6.0 4.8 1.4 0.2 Braziel Creeic 0.7 12N,10W,15A 19-Oct-06 309 CT 57 57 50 1.6-6.1 2.4 18.4 16.2 YOY based on <3.0" Sculpins abundant YOY based on <3.0" 1.4 12N,10W,15D 19-Oct-06 325 CT No sculpins 130 observed 130 Spotted frogs 91 present 1.4-7.2 2.6 42.1 29.4 Broadus Creek 0.1 17N,10W,2A 12-Jul-07 280 RB No sculpins 4 observed 4 5.6-8.7 6.8 1.4 0.0 Burnt Cabin Creek 0.2 17N,12W,8D 24-Aug-06 303 CT 3 3 3 2.7-3.0 2.8 1.0 1.0 Canyon Creek 1.5 17N,11W,14C 14-Jul-07 393 CT No sculpins 32 observed 32 Spotted frogs present 4.2-9.1 6.4 8.1 0.0 0.1 15N,13W,32A 12-Sep-06 300 10-Sep-07 300 CT 163 144 111 2.0-7.5 3.2 48.0 37.0 LL 21 21 13 2.8-5.9 4.1 7.0 4.3 Sculpins abundant LNS present Spotted frogs present CT 234 184 144 1.7-8.5 3.0 61.3 48.0 LL 36 27 20 2.9-6.5 4.0 9.0 6.7 MWF 1 1 1 3.7 3.7 0.3 0.3 Sculpins abundant RSS & LNS present Spotted frogs present Cold Brook Creek 1.1 15N,15W,28B 15-Jun-06 410 CT 11 11 8 2.3-5.8 3.6 2.7 2.0 trib to Bianchard Cr Cooney Creek 0.4 17N,10W,1A 12-Jul-07 639 RB 1 1 7.9 7.9 0.2 0.0 No Sculpins observed Sample may include rainbow trout / cutthroat trout tiybrids 272 Appendix A : Catch and size statistics for tributaries to the Blacl4.0 34 25 15 1 0.56 0.96 60.8 + 19.4 26 + 0.4 20.3 + 6.5 8.7 + 0.1 EB >4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 All <4.0 34 15 0.56 60.8 + 19.4 20.3 + 6.5 >4.0 26 1 0.96 27 + 0.4 9.0 + 0.1 12N,16W,3A 31-JUI-06 300 9-Aug-07 300 CT <4.0 >4.0 30 32 8 9 0.73 0.72 40.8 + 6.0 44.5 + 6.8 13.6 + 2.0 14.8 + 2.3 EB <4.0 2 1.00 2.0 + 0.0 0.7 + 0.0 All <4.0 32 8 0.75 42.7 + 5.5 14.2 + 1.8 >4.0 32 9 0.72 44.5 + 6.8 14.8 + 2.3 CT <4.0 >4.0 49 48 17 8 0.65 0.83 75.0 + 13 57.6 + 3.5 25.0 + 43 19.2 + 1.2 EB <4.0 21 4 0.81 25.9 + 2.8 8.6 + 0.9 All <4.0 70 21 0.70 100 + 11.4 33.3 + 3.8 >4.0 49 9 0.82 60 + 4.1 20 + 1.4 BearCreek 1.1 13N,16W,18B,7C 2-Aug-06 374 RB YOY (<2.5) 113 42 0.63 179.9 + 23.0 48.1 +6.1 Age1+(>2.5) 67 19 0.72 93.5 + 10.0 25.0 + 2.7 LL YOY (<3.0) 4 1 0.75 5.3 + 1.9 1.4 + 0.5 Aqe1+(>3.0) 9 1 0.89 10.1 +0.9 2.7 + 0.2 EB YOY (<40) 7 1 0.86 8.2 + 1.1 2.2 + 0.3 Age 1+ (>4.0) 8 1.00 8.0 + 0.0 2.1 +0.0 8-Aug-07 374 All YOY 11 2 0.82 13.4 + 1.9 3.6 + 0.5 Age 1+ 84 20 076 110.3 + 8.2 29.5 + 2.2 RB YOY (<3.0) 101 32 0.68 147.8 + 15.3 39.5 + 41 Age1+(>3.0) 100 21 0.79 126.6 + 7.3 33.8 + 1.9 LL YOY (<40) 32 12 0.63 51.2 + 12.5 13.7 + 3.3 Age 1+ (>4.0) 14 1 0.93 15.08 + 0.6 40 + 0.2 EB YOY (<40) 4 1.00 40 + 0.0 1.1 +0.0 Age1+(>4.0) 10 1.00 10.0 + 0.0 2.7 + 0.0 All YOY 137 44 0.68 201.8 + 18.4 54.0 + 49 Age1 + 124 22 0.82 150.8 + 6.2 40.3 + 1.7 DV >4.0 1 1.00 1.0 + 0.0 0.0 + 0.0 CT <4.0 28 12 0.57 49 + 16.3 1.2 + 0.4 >40 20 4 0.80 25 + 3.0 0.6 + 0.1 LL <4.0 9 5 0.44 20.5 + 20.6 0.5 +0.5 >4.0 6 2 0.67 9.0 + 4.2 0.2 + 0.1 EB <4.0 5 4 0.20 25 + 11.7.6 0.6 + 2.9 >40 8 1 0.88 9.14 + 1.0 0.2 + 0.0 MWF <40 2 1.00 2.0 + 0.0 0.1 +0.0 >40 13 3 0.77 16.9 + 3.1 0.4 + 0.1 Blackfoot River 119.6 14N,7W,5D 11-Sep-06 4000 (above Hogum Cr Rd) All <4.0 44 21 0.52 84.2 + 27.6 2.1+0.7 >40 47 10 0.79 59.7 + 5.1 1.5 + 0.1 Blackfoot River 124.3 15N,7W,35B,26C 5-Sep-06 2457 (above Flesher Pass Rd) CT <4.0 63 13 0.79 79.4 + 5.6 3.2 + 0.2 >40 12 5 0.58 20.6 + 9.9 0.8 + 0.4 LL >4.0 1 1.00 1.0 + 0.0 0.0 + 0.0 EB <4.0 64 17 0.73 87.1 +8.7 3.5 + 0.4 >40 27 9 0.67 4.05 + 8.8 1.6 + 0.4 LNS <4.0 1 1.00 1.0 + 0.0 0.0 + 0.0 >40 2 1 0.50 40 + 6.8 0.2 + 0.3 All <4.0 128 30 0.77 167.2 + 9.9 6.8 + 0.4 >40 42 15 0.64 65.3 + 12.8 2.7 + 0.5 281 Appendix B: Two-pass depletion estimates for tributaries to tlie Blacl4.0 33 5 0.85 38.9 + 2.5 13 + 0.8 LL <4.0 13 1.00 13 + 0.0 4.3 + 0.0 >4.0 8 1.00 8.0 + 0.0 2.7 + 0.0 All <4.0 124 14 0.89 139.8 + 3.3 46.6+1.1 >4.0 41 5 0.88 46.7 + 2.1 15.6 + 0.7 10-Sep-07 300 CT <4.0 144 41 0.72 201.3+14.8 67.1 + 4.9 >4.0 40 9 0.78 51.6 + 5.1 17.2 + 1.7 LL <4.0 20 7 0.65 30.8 + 8.4 10.3 + 2.8 >4.0 7 2 0.71 9.8 + 3.3 3.3+1.1 MWF <4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 All <4.0 165 48 0.71 232.7+16.6 77.6 + 5.5 >4.0 47 11 0.77 61.4 + 6.0 20.5 + 2.0 Cottonwood Creek 12.0 16N,14W,24D 19-Sep-06 515 CT YOY(<3.0) 59 16 0.73 81+8.7 15.7+1.7 Age1+(>3.0) 34 11 0^68 50.3 + 9.3 9.8+1.8 DV" YOY (<4.0) 4 1.00 4.0 + 0.0 0.8 + 0.0 Age1+(>4.0) 4 1.00 4.0 + 0.0 0.8 + 0.0 EB" <4.0 8 1.00 8.0 + 0.0 1.6 + 0.0 >4.0 2 1 0.50 4.0 + 6.8 0.8+1.3 All <4.0 12 1.00 12 + 0.0 2.3 + 0.0 >£0 6 1 083 7.2+1.2 1.4 + 0.2 18-Sep-07 515 CT YOY(<3.0) 63 18 0.71 88.2 + 9.9 17.1 + 1.9 Age1+(>3.0) 45 18 060 75+17.3 14.6 + 3.4 DV YOY (<4.0) Age1+(>4.0) 3 1 1.00 3.0 + 0.0 0.6 + 0.0 EB <4.0 >4.0 4 1 1 1 0.75 0.00 5.3+1.9 1.0 + 0.4 All <4.0 >4.0 4 4 2 1 0.50 0.75 8.0 + 9.6 5.3+1.9 1.6+1.9 1.0 + 0.4 DV >4.0 1 1 0.00 Enders Spring Creek 0.5 14N,11W,31C 24-Aug-06 300 (trib to NFBLKFT) EB <4.0 53 1.00 53.0 + 0.0 17.7 + 0.0 >4.0 5 2 0.60 8.3 + 5.8 2.8+1.9 15-Aug-07 300 EB <4.0 >4.0 59 19 26 6 0.56 0.68 105.5 + 25.5 27.8 + 6.6 35.2 + 8.5 9.3 + 2.2 LL >4.0 2 1.00 2.0 + 0.0 0.7 + 0.0 MWF <4.0 2 1.00 2.0 + 0.0 0.7 + 0.0 RB* <4.0 >4.0 32 38 6 22 0.81 0.42 39.4 + 3.4 90.2 + 49.6 9.8 + 0.9 22.6+12.4 CT >4.0 2 1.00 2.0 + 0.0 0.5 + 0.0 LL <4.0 >4.0 8 29 2 6 0.75 0.79 10.7 + 2.8 36.6 + 3.8 2.7 + 0.7 9.1 + 1.0 Gold Creek 1.9 14N,16W,30D 14-Sep-06 400 All <4.0 40 8 0.80 50 + 4.2 12.5+1.1 >4.0 69 28 0.59 116.1+22.2 29.0 + 5.5 * Sample may include rainbow trout / cutthroat trout hybrids ** Sample may include bull trout / brook trout hybrids 282 Appendix B: Two-pass depletion estimates for tributaries to the Blackfoot River excluding Clearwater Basin, 2006-2007 (cont'd). River Date Section Size Class 2nd 3rd Probability of Total Estimate ± Stream IVIile Location (T,R,S) Sampled Length (ft) Species (inches) 1st Pass Pass Pass Capture CI Estim/100' ±CI Hoyt Creek New post restoration section 2007 0.2 15N, 12W, 19B 12-Sep-06 300 EB <4.0 >4.0 1 2 1 1.00 0.50 1.0 + 0.0 4.0 + 6.8 0.3+0.0 1.3+ 2.3 1.2 15N,12W,19C 12-Sep-06 300 EB <4.0 1 1.00 1.0+_0.0 0.3 + 0.0 >4.0 5 1.00 5.0 + 0.0 1.7 + 0.0 LL >4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 2.7 15N,12W,29C 4-Sep-07 300 EB <4.0 >4.0 1 1 1.00 1.00 1.0 + 0.0 1.0 + 0.0 0.3 + 0.0 0.3 + 0.0 4.3 15N,12W,28C 12-Sep-06 300 EB <4.0 9 4 0.56 16.2+10.2 5.4 + 3.4 >40 2 100 2.0 + 0.0 0.7 + 0.0 4-Sep-07 300 EB <4.0 4 1.00 4.0 + 0.0 1.3 + 0.0 >4.0 5 1 0.80 6.3+1.5 2.1 + 0.5 Jacobson Spring Creek 0.6 14N,12W,1CD 24-Aug-06 525 13-Aug-07 525 LL <4.0 2 1.00 2.0 + 0.0 0.4 + 0.0 >4.0 8 1 0.88 9.1 + 1.0 1.7 + 0.2 EB <4.0 23 1.00 23.0 + 0.0 4.4 + 0.0 >4.0 13 4 0.69 18.8 + 5.2 3.6+ 1.0 All <4.0 25 1.00 25.0 + 0.0 4.8 + 0.0 >4.0 21 5 0.76 27.6 + 4.1 5.3 + 0.8 LL <4.0 6 2 0.67 9.0 + 4.2 1.7 + 0.8 >4.0 11 2 0.82 13.4+ 1.9 2.6 + 0.4 RB <4.0 >4.0 1 1 1 1.00 0.00 1.0 + 0.0 0.2 + 0.0 EB <4.0 44 8 0.82 53.8 + 3.8 10.2 + 0.7 >4.0 15 2 0.87 17.3+ 1.4 3.3 + 0.3 All <4.0 51 10 0.80 63.4 + 4.6 12.1 + 0.9 >4.0 27 5 0.81 33.1 + 3.1 6.3 + 0.6 MWF <40 6 2 0.67 9.0 + 4.2 1.7 + O.i >4.0 1 Kleinschmidt Creek 0.5 14N,11W,6D,5C 16-Aug-06 500 Post 2001 reconstruction E channel "woodless" section LL [<4.3] 93 33 0.65 144.2+ 18.8 28.8 + 3.8 [>4.31 27 12 0.56 48.6+ 17.6 9.7 + 3.5 EB [<4.3] [>4.31 1 1 1 1.00 0.00 1.0 + 0.0 0.2 + 0.0 All [<4.3] 94 33 [>4.31 28 13 22-Aug-07 500 0.8 14N,11W,5C 16-Aug-06 500 Post-2001 reconstruction based on 3 pass est E channel "woody"section based on 3 pass est_ 0.65 144.9+ 18.4 29.0 + 3.7 0.54 52.3+20.3 10.5 + 4.1 LL [<4.3] [>4.31 52 27 14 13 0.73 0.52 71.2 + 8.0 52.1 + 22.2 14.2+1.6 10.4 + 4.4 EB [<4.31 2 1 0.50 4.0 + 6.8 0.8+ 1.4 All [<4.3] 54 15 0.72 74.8 + 8.7 15.0+1.7 [>431 27 13 052 52.1 + 22.2 10.4 + 4.4 22-Aug-07 500 LL [<4.3] 120 40 0.67 180 + 18.6 36+ 3.7 [>4.31 32 20 7 0.54 64 + 7.0 12.8 + 3.1 EB [<4.3] [>4.31 3 3 3 5 0.00 All [<4.3] 123 43 0.65 189.1 + 20.9 37.8 + 4.2 [>4.31 35 25 7 0.53 73.2 + 7.9 14.6 + 3.5 CT [<4.31 1 1.00 1.0 + 0.0 0.2 + 0.0 DV [>4.31 1 LL [<4.3] 78 14 0.82 95.06 + 5.0 19.0+ 1.0 [>4.31 49 16 0.67 72.8+ 11.4 14.6 + 2.3 EB [<4.3[ 8 2 0.75 10.7 + 2.8 2.1 + 0.6 [>4.31 4 2 0.50 8.0 + 9.6 1.6+ 1.9 All [<4.3] 87 16 0.82 106.6 + 5.5 21.3+1.1 [>431 53 19 064 82.6+14.5 16.5 + 2.9283 Appendix B: Two-pass depletion estimates for tributaries to the Biacl(foot River excluding Clearwater Basin, 2006-2007 (cont'd). stream River Date Section Size Class IVIile Location (T,R,S) Sampled Length (ft) Species (inches) 1st Pass 2nd Pass 3rd Pass Probability of Capture Total Estimate + CI Estim/100' + CI Lincoln Spring Creek Pre-restoration 3.8 14N,9W,13D 21-Aug-07 321 LL >4.0 13 5 0.62 21.1 +8.4 6.6 + 2.6 EB <4.0 >4.0 117 20 43 8 0.63 0.60 185.0 +22.8 33.3 + 11.5 57.6 + 7.1 10.4 + 3.6 All <4.0 117 43 0.63 185.0 +22.8 57.6 + 7.1 >4.0 33 13 0.61 54.5 + 14.3 17.0 + 4.4 McCabe Creek 2.2 15N,12W,5C 20-Aug-07 340 CT YOY ( <3.0) Age1+(>3.0) 97 54 7 10 EB YOY ( <3.0) Age1+(>3.0) 23 22 All YOY ( <3.0) Age1+(>3.0) 120 76 10 12 0.93 0.81 104.5+1.7 66.3 + 4.4 0.87 0.91 26.5 + 1.7 24.2 + 1.1 0.92 0.84 130.9 + 2.2 90.3 + 4.1 31 +0.5 19.5 + 1.3 7.8 + 0.5 7.1 +0.3 38.5 + 0.7 26.5 + 1.2 Nevada Spring Creek Old stream mile was 0.8 Post restoration stream mile ( ) 0.8(1.1) 13N,11W,10C 13-Sep-06 500 CT >4.0 LL >4.0 17 RB >4.0 NPM <4.0 >4.0 3 27 RSS <4.0 >4.0 15 7 15 1 Sucker <4.0 All <4.0 >4.0 19 57 15 11 12-Sep-07 500 CT >4.0 12 LL >4.0 Old stream mile was 3.0 Post restoration stream mile ( ) 3(3.5) 13N,11W,11D 13-Sep-06 470 CT >4.0 LL <4.0 >4.0 5 26 RSS <4.0 >4.0 12-Sep-07 470 CT >4.0 24 LL <4.0 >4.0 21 All >4.0 45 1.00 5.0 + 0.0 0.71 24.1 +5.4 1.00 1.0 + 0.0 1.00 0.81 3.0 + 0.0 33.1 +3.1 0.00 0.86 8.2 + 1.1 1.00 1.0 + 0.0 0.21 0.81 90.3 + 203.6 70.6 + 4.8 0.92 13.1+0.7 0.67 4.5 + 2.9 0.00 0.60 0.77 8.3 + 5.8 33.8 + 4.3 1.00 1.0 + 0.0 0.92 26.2 + 1.0 0.81 25.9 + 2.8 0.87 51.9 + 2.5 1.0 + 0.0 4.8 + 1.1 0.2 + 0.0 0.6 + 0.0 6.6 + 0.6 1.6 + 0.2 0.2 + 0.0 18.1 +40.7 14.1 +1.0 2.6+0.1 0.9 + 0.6 1.8 + 1.2 7.2 + 0.9 0.2 + 0.0 5.6 + 0.2 5.5 + 0.6 11.0 + 0.5 Pearson Creek 0.5 15N,13W,33D 18-Sep-06 300 CTYOY age 1 + <3.2 >3.2 11-Sep-07 300 CTYOY age 1 + <3.2 >3.2 10 1.1 14N,13W,3B 18-Sep-06 405 11-Sep-07 405 EB <4.0 >4.0 0.63 12.8 + 6.2 0.80 12.5 + 2.1 1.00 1.00 2.0 + 0.0 2.0 + 0.0 4.3 + 2.1 4.2 + 0.7 CTYOY <3.2 6 3 0.50 12 + 11.8 3.0 + 2.9 age1 + >3.2 68 19 0.72 94.4 + 9.8 23.3 + 2.4 EB <4.0 1 1.00 1.0 + 0.0 0.2 + 0.0 CTYOY <3.2 4 1.00 4.0 + 0.0 1.0 + 0.0 age1 + >3.2 79 10 0.87 90.5 + 3.1 22.3 + 0.8 0.5 + 0.0 0.5 + 0.0 ■ 284 Appendix B: Two-pass depletion estimates for tributaries to the Blacl4.0 65 5 16 2 16-Aug-07 <4.0 >4.0 242 6 65 1.5 14N,9W,36A 10-Aug-06 270 16-Aug-07 270 <4.0 >4.0 266 31 46 7 0.75 0.60 86.2: 8.3 4 :7.6 5.8 0.73 1.00 330.9: 6.0 4 : 17.2 0.0 0.83 0.77 321.6: 40.04- 17.6 + 1.6 1.7 + 1.2 LL <4.0 >4.0 172 6 48 0.72 1.00 238.6 + 15.6 6.0 4 0.0 48.7 4 3.2 1.2 4 0.0 CT <4.0 70 17 0.76 92.5 + 7.7 18.9 4 1.6 67.5 4 3.5 1.2 4 0.0 LL <4.0 71 14 0.80 88.4 4 5.5 30.5 4 1.9 >4.0 8 4 0.50 16.0 4 13.6 5.5 4 4.7 EB <4.0 3 1.00 3.0 4 0.0 1.0 4 0.0 >4.0 1 1.00 1.0 4 0.0 0.3 4 0.0 Ail <4.0 74 14 0.81 91.3 4 5.3 31.5 4 1.8 >4.0 9 4 0.56 16.2 4 10.2 5.6 4 3.5 CT <4.0 76 13 0.83 91.7 4 4.6 34.0 4 1.7 LL <4.0 190 33 0.83 230.0 4 7.4 85.2 4 2.8 >4.0 31 7 0.77 40.04 4 4.6 14.8 4 1.7 119.1 4 3.2 14.8 4 1.7 Roci< Creei< 1.6 14N,11W,5A 6-Sep-06 510 DV <4.0 >4.0 1 1.00 1.0 4 0.0 0.2 4 0.0 LL <4.0 24 7 0.71 33.9 4 6.3 6.6 4 1.2 >4.0 16 5 0.69 23.3 4 5.9 4.6 4 1.2 EB <4.0 35 3 0.91 38.3 4 1.2 7.5 4 0.2 >4.0 6 2 0.67 9.0 4 4.2 1.8 4 0.8 15-Aug-07 <4.0 >4.0 59 23 10 7 <4.0 >4.0 <4.0 >4.0 31 19 16 5 0.83 0.70 71 44.0 33.1 4 6.8 0.71 0.79 43.7 4 7.1 24.1 4 3.2 0.50 0.60 32.0 4 19.2 8.3 4 5.8 13.9 4 0.8 6.5 4 1.3 8.6 4 1.4 4.7 4 0.6 6.3 4 3.8 1.6 4 1.1 <4.0 >4.0 47 24 19 6 0.60 0.75 78.9 4 18.1 32.0 4 4.8 15.5 4 3.6 6.3 4 0.9 15N,11W,35B 13-Aug-07 <4.0 >4.0 1 15 0.00 0.93 <4.0 >4.0 1 23 0.00 0.91 <4.0 >4.0 15N,11W,24D 13-Aug-07 <4.0 >4.0 4 31 0.75 0.81 5.3 4 1.9 38.4 4 3.5 1.0 40.4 7.3 4 0.7 <4.0 >4.0 15 35 15N,10W,9B 15-Aug-07 <4.0 >4.0 0.80 0.83 18.8 4 2.6 42.2 4 3.1 0.89 0.67 10.1 4 0.9 4.5 4 2.9 3.6 4 0.5 8.0 4 0.6 2.3 4 0.2 1.0 40.7 <4.0 >4.0 65 35 22 6 0.66 0.83 98.3 4 14.1 42.2 4 3.1 22.6 4 3.3 9.7 4 0.7 LND Sucl4.0 <4.0 >4.0 74 39 5 1 23 7 0.69 0.82 1.00 1.00 107.4 4 12.6 47.5 4 3.5 5.0 4 0.0 1.0 4 0.0 24.7 4 2.9 10.9 4 0.8 1.1 4 0.0 0.2 4 0.0 285 Appendix B: Two-pass depletion estimates for tributaries to the BlaclTo i 100 1.0 + 0.0 0.1 +0.0 CT <4.0 >4.0 29 36 15 3 0.48 0.92 60.1 +28.9 39.3+1.2 8.0 + 3.8 5.2 + 0.2 LL <4.0 >4.0 40 9 5 3 0.88 0.67 45.7 + 2.1 13.5 + 5.1 6.1+0.3 1.8 + 0.7 EB <4.0 >4.0 39 58 25 12 0.36 0.79 108.6 + 78.0 73.1 + 5.4 14.4+10.3 9.7 + 0.7 All <4.0 >4.0 108 104 45 18 0.58 0.83 185.1+29.7 125.8 + 5.5 24.5 + 3.9 16.7 + 0.7 2.7 13N,9W,4D 26-JUI-07 630 CT <4.0 11 1.00 11.0 + 0.0 1.7 + 0.0 EB <4.0 >4.0 5 2 1 0.80 1.00 6.3+1.5 2.0 + 0.0 1.0 + 0.2 0.3 + 0.0 All <4.0 >4.0 16 2 1 0.94 1.00 17.1+0.6 2.0 + 0.0 2.7 + 0.1 0.3 + 0.0 3.2 13N,9W,8A 2-Aug-07 360 DV >4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 CT <4.0 >4.0 54 22 12 2 0.78 0.91 69.4 + 5.8 24.2 + 1.1 19.3 + 1.6 6.7 + 0.3 EB <4.0 >4.0 1 3 1.00 1.00 1.0 + 0.0 3.0 + 0.0 0.3 + 0.0 0.8 + 0.0 All <4.0 >4.0 55 26 12 2 0.78 0.92 70.4 + 5.7 28.2 + 0.9 19.5+1.6 7.8 + 0.3 0.2 15N,13W,9B 14-Aug-06 360 CT Age1+(>4.0) 1 1.00 1.0 + 0.0 0.3 + 0.0 LL YOY (<3.0) Age1+(>3.0) 3 25 1 3 0.67 0.88 4.5 + 2.9 28.4+1.6 1.3 + 0.8 7.9 + 0.4 EB YOY (<4.0) Age1+(>4.0) 5 4 2 0.60 1.00 8.3 + 5.8 4.0 + 0.0 2.3+1.6 1.1+0.0 All YOY Age1 + 8 30 3 3 0.63 0.90 12.8 + 6.2 33.3+1.4 3.6+1.7 9.3 + 0.4 1.4 15N,13W,3B 14-Aug-06 261 CT YOY (<3.0) Age1+(>3.0) 1 17 5 1.00 0.71 1.0 + 0.0 24.1 + 5.4 0.4 + 0.0 9.2 + 2.1 EB YOY (<3.0) Age1+(>3.0) 6 9 1 0.83 1.00 7.2 + 1.2 9.0 + 0.0 2.8 + 0.5 3.4 + 0.0 All YOY (<3.0) Age1+(>3.0) 7 26 1 5 0.86 0.81 8.2+1.1 32.2 + 3.2 3.1+0.4 12.3 + 1.2 1.6 15N,13W,3B 6-Sep-06 466 CT YOY (<3.0) Age1+(>3.0) 13 30 3 7 0.77 0.77 16.9 + 3.1 39.1 + 4.7 3.6 + 0.7 8.4+1.0 EB YOY (<3.0) Age1+(>3.0) 7 6 3 0.57 1.00 12.2 + 8.1 6.0 + 0.0 2.6+1.7 1.3 + 0.0 All YOY ( < 3.0) Age1+(>3.0) 20 36 6 7 0.70 0.81 28.6 + 6.1 44.7 + 3.9 6.1 + 1.3 9.6 + 0.8 2 15N,13W,3A 14-Aug-06 285 CT YOY (<3.0) Age1+(>3.0) 12 32 2 11 0.83 0.66 14.4+1.8 48.8+10.3 5.1+0.6 17.1+3.6 EB YOY (<3.0) Age1+(>3.0) 14 14 3 3 0.79 0.79 17.8 + 2.8 17.8 + 2.8 6.3 + 1.0 6.3+1.0 Shanley Creek All YOY (< 3.0) 26 5 0.81 32.2 + 3.2 11.3+1.1 Age1+(>3.0) 46 14 070 66.1 +9.5 23.2 + 3.4 286 Appendix B: Two-pass depletion estimates for tributaries to the Blackfoot River excluding Clearwater Basin, 2006-2007 (cont'd). River Date Section Size Class 2nd 3rd Probability of Total Estimate ± Stream IVIile Location (T,R,S) Sampled Length (ft) Species (inches) 1st Pass Pass Pass Capture CI Estim/100' ± CI Snowbank Creek oi 15N,8W,9A 15-Aug-06 396 7-Aug-07 300 DV <4.0 3 2 0.33 9.0 + 26.3 2.3 + 6.6 >4.0 4 1 0.75 5.3 + 1.9 1.3 + 0.5 CT <4.0 17 2 0.88 19.3 + 1.3 4.9 + 0.3 >4.0 14 2 0.86 16.3 + 1.5 4.1 +0.4 All <4.0 20 4 0.80 25 + 3.0 6.3 + 0.8 >4.0 18 3 0.83 21.6 + 2.2 5.5 + 0.5 DV <4.0 20 3 0.85 23.5 + 2.0 7.8 + 0.7 >4.0 26 5 0.81 32.2 + 3.2 10.7 + 1.1 CT <4.0 24 2 0.92 26.2 + 1.0 8.7 + 0.3 >4.0 47 14 0.70 66.9 + 9.2 22.3 + 3.1 All <4.0 44 5 0.89 49.6 + 2.0 16.5 + 0.7 >4.0 73 19 0.74 98.7 + 8.9 32.9 + 3.0 below diversion 0.4 15N,8W,9A 15-Aug-06 450 7-Aug-07 450 DV <4.0 >4.0 8 4 0.50 16 + 13.6 3.6 + 3.0 CT <4.0 21 7 0.67 31.5 + 7.8 7.0 + 1.7 >4.0 37 7 0.81 45.6 + 3.7 10.1 +0.8 All <4.0 21 7 0.67 31.5 + 7.8 7.0 + 1.7 >4.0 45 11 0.76 59.6 + 6.3 13.2 + 1.4 DV <4.0 11 2 0.82 13.4 + 1.9 3.0 + 0.4 >4.0 17 6 0.65 26.3 + 7.9 5.8 + 1.8 CT <4.0 12 3 0.75 16.0 + 3.4 3.6 + 0.7 >4.0 62 18 0.71 87.4 + 10.1 19.4 + 2.2 All <4.0 23 5 0.78 29.4 + 3.7 6.5 + 0.8 >4.0 79 24 0.70 113.5 + 12.5 25.2 + 2.8 above diversion 0.41 15N,8W,9A 15-Aug-06 500 DV <4.0 >4.0 1 1 0.00 CT <4.0 101 27 0.73 137.8 + 11 27.6 + 2.2 >40 30 8 073 40.9 + 6.0 8.2 + 1.2 All <4.0 101 27 0.73 137.8 + 11 27.6 + 2.2 >4.0 31 9 0.71 43.7 + 7.1 8.7 + 1.4 7-Aug-07 500 DV <4.0 22 4 0.82 26.9 + 2.7 5.4 + 0.5 >4.0 2 1.00 2.0 + 0.0 0.4 + 0.0 CT <4.0 73 24 0.67 108.8 + 14.1 21.8 + 2.8 >4.0 86 22 0.74 115.6 + 9.4 23.1 +1.9 All <4.0 95 28 0.71 134.7 + 12.9 26.9 + 2.6 >4.0 88 22 0.75 117.3 + 9.1 23.5 + 1.8 Spring Creek, 0.6 15N,11W,21B 14-Sep-06 385 trib to NFBLKFT (Murphy's Spring Cr) 6-Sep-07 385 DV <4.0 2 1.00 2.0 + 0.0 0.5 + 0.0 >4.0 2 1.00 2.0 + 0.0 0.5 + 0.0 CT <4.0 61 26 0.57 106.3 + 23.7 27.6 + 6.1 >4.0 15 2 0.87 17.3 + 1.4 4.5 + 0.4 EB <4.0 11 5 0.55 20.2 + 12 5.2 + 3.1 >4.0 4 2 0.50 8.0 + 9.6 2.1 +2.5 All <4.0 74 31 0.58 127.3 + 24.9 33.1 +6.5 >4.0 21 4 0.81 25.9 + 2.8 6.7 + 0.7 DV >4.0 2 1.00 2.0 + 0.0 0.5 + 0.0 CT <4.0 56 16 0.71 78.4 + 9.3 20.4 + 2.4 >4.0 17 2 0.88 19.3 + 1.3 5.0 + 0.3 EB <4.0 19 5 0.74 25.8 + 4.7 6.7 + 1.2 >4.0 6 1.00 6.0 + 0.0 1.6 + 0.0 All <4.0 75 21 0.72 104.2 + 10.4 27.1+2.7 >4.0 25 2 0.92 27.2 + 1.0 7.1+0.3 287 Appendix B: Two-pass depletion estimates for tributaries to the Blackfoot River excluding Clearwater Basin, 2006-2007 (cont'd). stream River Date Section IVIile Location (T,R,S) Sampled Length (ft) Species Size Class (inches) 2nd 1st Pass Pass 3rd Pass Probability of Total Estimate ± Capture CI Estim/100'±CI Warren Creek 15N,12W,31C 13-Sep-06 345 LL <4.0 >4.0 2 14 5-Sep-07 345 LL <4.0 >4.0 9 16 2.1 15N,12W,31A 13-Sep-06 345 8.2 15N,12W,25C 5-Sep-07 330 EB <4.0 >4.0 65 65 16 6 One. <4.0 1.00 0.93 2.0 i 15.8- 0.0 ■0.6 0.44 0.94 20.3 j 17.1 ■ 20.6 ■0.6 0.75 0.91 86.2 + 7.6 71.6-H.9 1.00 1.0-1-0.0 0.6 + 0.0 4.4 + 0.2 5.9 + 6.0 4.9 + 0.2 LL <4.0 >4.0 4 1 0.75 5.3 + 1.9 1.5 + 0.6 CT <4.0 >4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 EB <4.0 >4.0 2 1.00 2.0 + 0.0 0.6 + 0.0 All >4.0 7 1 0.86 8.2 + 1.1 2.4 + 0.3 5-Sep-07 345 EB >4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 13-Sep-06 468 EB <4.0 >4.0 7 58 1 4 0.86 0.93 8.2 + 1.1 62.3 + 1.2 1.7 + 0.2 13.3 + 0.3 10-Sep-07 468 EB <4.0 28 6 0.79 35.6 + 4.0 7.6 + 0.8 >4.0 92 18 0.80 114.4 + 6.2 24.4 + 1.3 6.7 15N,12W,35B 13-Sep-06 386 EB <4.0 1 1.00 1.0 + 0.0 0.3 + 0.0 >4.0 4 1.00 4.0 + 0.0 1.0 + 0.0 5-Sep-07 385 No salomn lids present 26.1 +2.3 21.7 + 0.6 0.3 + 0.0 Wasson Creek 0.1 13N,11W,11D 17-Aug-06 300 CT <4.0 >4.0 LL <4.0 >4.0 1 11 LNS >4.0 >4.0 All <4.0 >4.0 1 19 14-Aug-07 300 CT >4.0 LL <4.0 >4.0 ALL <4.0 >4.0 1 24 11 LNS >4.0 RSS >4.0 Old stream mile was 0.6 Post restoration stream mile ( ) Old stream mile was 2.4 Post restoration stream mile ( ) Old stream mile was 2.6 Post restoration stream mile ( ) 0.6(1) 13N,11W,13B 29-Aug-06 300 Based on 3 pass est CT <4.0 >4.0 13 0.67 9.0 + 4.2 0.00 0.82 13.4 + 1.1 1.00 2.0 + 0.0 0.00 0.79 24.1 +3.2 44.1 +25.6 1.00 1.00 1.0: 3.0- 0.0 0.0 1.00 0.54 1.0 + 0.0 44.3 + 18.1 1.00 2.0 + 0.0 0.69 21.1 +0.5 LNS >4.0 1.00 14-Aug-07 300 CT <4.0 >4.0 1.00 0.33 1.0 + 0.0 18.0 + 37.2 RSS <4.0 2.4(2.! 13N,10W,7C 17-Aug-06 312 CT <4.0 >4.0 4 15 0.75 0.80 5.3 + 1.9 18.7 + 2.6 14-Aug-07 312 CT <4.0 >4.0 13 32 0.77 0.88 16.9 + 3.1 36.6 + 1.9 2.6 (3) 13N,10W,7C 17-Aug-06 300 CT <4.0 >4.0 25 39 0.92 0.87 27.2 + 1.0 44.7 + 2.2 14-Aug-07 300 CT <4.0 >4.0 61 24 0.85 0.75 71.6 + 3.3 32.0 + 4.8 3.0 + 1.4 4.5 + 0.6 0.7 + 0.0 8.0 + 1.1 14.7 + 8.5 0.3 + 0.0 1.0 + 0.0 0.3 + 0.0 14.8 + 6.0 0.7 + 0.0 7.0 + 0.3 0.3 + 0.0 6.0 + 12.4 1.7 + 0.6 6.0 + 0.8 5.4 + 1.0 11.7 + 0.6 9.1 j 14.9- 0.3 ■0.7 23.9 + 1.1 10.7 + 1.6 288 Appendix B: Two-pass depletion estimates for tributaries to the Blacl40 7 100 7.0 + 0.0 14 + 0.0 3.6 14N,9W,34A 25-Jul-07 371 No fish found 4.7 13N,9W,34D 25-Jul-07 560 CT <4.0 6 1 0.83 7.2 + 1.2 1.3 + 0.2 >4.0 11 1 0.91 12.1 +0.7 2.2 + 0.1 EB <4.0 9 2 0.78 116 + 2.4 2.1 +0.4 >4.0 12 1.00 12.0 + 0.0 2.1 +0.0 All <4.0 15 3 0.80 18.8 + 2.6 3.3 + 0.5 >4.0 23 1 0.96 24.1+0.5 43 + 0.1 5.2 13N,9W,3A 24-Jul-07 300 CT <4.0 4 2 0.50 8.0 + 9.6 2.7 + 3.2 >4.0 22 1.00 22.0 + 0.0 7.3 + 0.0 EB <4.0 9 5 0.44 20.3 + 20.6 6.8 + 6.9 >4.0 15 100 15.0 + 0.0 5.0 + 0.0 All <4.0 13 7 0.46 28.2 + 22.2 9.4 + 7.4 >4.0 37 100 37.0 + 0.0 12.3 + 0.0 5.7 13N,9W,3C 23-Jul-07 395 CT <4.0 28 8 0.71 39.2 + 6.6 9.9 + 17 >4.0 32 5 0.84 37.9 + 2.6 9.6 + 0.7 EB <4.0 15 5 0.67 22.5 + 6.6 5.7 + 17 >4.0 20 2 0.90 22.2 + 11 5.6 + 0.3 All <4.0 43 13 0.70 616 + 9.1 15.6 + 2.3 >4.0 52 7 0.87 60.1+2.7 15.2 + 0.7 * Sample may Include rainbow trout / cutttiroat trout fiybrlds ** Sample may Inolude bull trout / brook trout tiybrlds *** Genetics testing pending CT = Cutttiroat trout DV = Bull trout (Dolly Varden) LL = Brown trout (Locti Leven) RB = Rainbow trout EB = Eastern brook trout MWF = Monutain wtilteflsti LNS = Longnose sucker LSS = Largescale sucker LND = Longnose dace RSS = Redslde stilner ONC = Oncorfiynctius (undifferentiated) 289 Appendix C: Mark and recapture and biomass estimates for the Biackfoot River, 2006. River Mile IVIid- point Date Sampled Section Length (ft) Species Size Class (inches) (R/C) Total Estimate : 95% CI Total Biomass (lb/section) Estimate/1000' ± 95% CI Biomass (lb/1000') Condition Factor /1 000' Biackfoot River, Johnsrud Section 13.5 30-May-06 17680 5 - 9.9 10 -11.9 >12 355 45 77 319 46 28 4 70 14 0.14 0.14 0.20 2422.8 - 265.8 + 368.2 + ■ 590.1 183.4 144.4 381.43 130.73 363.80 137.04 + 33.4 15 + 10.4 20.8 + 8.2 21.57 7.39 20.58 0.19 3300 + 534 1500.58 186.6 + 30.20 84.87 40.60 36.77 33.75 RB> 6 424 359 61 0.17 2466.7 + 512.4 874.48 139.5 + 29 49.46 38.36 LL 6-11.9 > 12 78 44 54 29 13 12 0.24 0.41 309.4+ 123.01 102.9 + 34 74.55 274.27 17.5 + 6.96 5.8 + 1.95 4.22 15.51 41.65 39.74 LL > 6 122 83 25 0.30 396.4 + 110.6 161.64 22.42 + 6.26 9.14 36.01 CT 6- 11.9 > 12 63 19 47 11 6 5 0.13 0.45 437.9 + 265.3 39.0 + 17.5 143.47 36.40 24.8 + 15 2.21 +1.0 8.12 2.06 43.09 34.91 CT > 6 82 58 11 0.19 407.1 + 183.1 181.16 23.03 + 10.36 10.25 41.50 DV > 6** 13 8 1 0.13 38.92 Biackfoot River, Scotty Brown Bridge 43.9 25-May-06 20064 4 - 11 10.9 ■ 13.9 14 38 22 30 47 24 33 0.13 0.17 0.27 266.4 -1 114 + 104.4 155.1 72.8 - 43.1 59.80 84.17 162.87 13.3 + 7.7 5.68 + 3.6 5.2 + 2.1 2.98 4.19 8.12 0.18 452.3 + 157.9 360.70 22.54 + 7.9 17.98 6-11.9 > 12 46 39 55 41 12 13 0.22 0.32 201.5 119 + - 79 40 62.80 175.93 10.4 + 3.9 5.93 + 2.0 3.13 8.77 319.9 + 86.5 15.9 + 4.3 11.9 • 12 44 46 67 45 10 16 0.15 0.36 277.2 + 126.2 - 125.3 - 37.3 95.45 139.33 13.8 + 6.2 6.29 + 1.86 4.76 6.94 379.9 + 103.2 89.00 + 49.2 4.44 + 2.5 1246.6 + 211.4 62.1 + 10.5 40.59 39.01 34.24 37.88 38.92 36.01 38.21 37.29 Wales Creek Section 24-May-06 31635 6-11.9 > 12 20 49 18 46 6 10 0.33 0.22 56 + 25.63 212.6 + 93.4 23.22 299.56 1.77 ■ 6.72 ■ 0.81 2.95 0.73 9.47 266.7 + 92.5 8.43 + 2.92 337.3 +121 10.7 + 3.8 24-May-06 7603 1220 + 914 160 + 120 38.08 34.21 Canyon Section 20-Sep-06 5422 10.7 + 5.5 LL Age 1 + LL Age 2+ 4.5 • 8.5 " 22 15 14 9 0.29 0.33 68.0 ; 39.0 - 39.9 23.5 59.12 48.92 12.5 + 7.4 7.2 + 4.3 10.90 4.90 177 121 24 867.6 + 276 36.71 37.29 LL > 6 " 21 13 3 0.23 76 + 52 68.00 14 + 9.5 12.54 36.49 > 6 25 19 5 0.26 85.7 + 47.1 79.33 15.8 + 8.7 14.63 36.51 6 - 11.9 ** > 12 28 149 19 102 3 21 0.16 0.21 144 + 105.6 701.3 + 235.1 70.87 563.31 26.6 + 19.5 129.3 + 43.4 13.07 103.89 36 34.78 Poorman - Dalton Section 107.2 21-Sep-06 DV > 6 ** 2 CT Age 1 + > 3.0 ** 12 5 1 0.20 LL Age 1 + >4.5 123 116 39 0.34 361.7 + 74.1 192.2 53.2 + 10.9 28.3 38.5 EB Age 1 + 4.5 - 11.9 ** 1 0.00 All Age 1 + 135 122 40 0.33 407.0 + 84.2 216.5 59.9+ 12.4 31.8 39.6 All 133 114 39 0.34 384.3 + 79. i 56.5 +11.7 These estimates did not meet the minimal number of recaptures for a valid estimate and should be used with caution. CT = Cutthroat trout DV = Bull trout (Dolly Varden) LL = Bro\A/n trout (Loch Leven) RB = Rainbow trout EB = Eastern brook trout MWF = Mountain whitefish 290 Appendix D: Summary of stream discharge measurements for 2006 and 2007. stream name Legal Description Stream Mile Date Discharge (cfs) Lat Long Location / Comments Bear Gulch T14N,R9W,34D 0.7 23-Jul-07 0.011 N46.92245 W1 12.71 989 ~150ft upstream of Dalton Mtn Rd. Bear Gulch T13N,R9W,3B 1.2 24-Jul-07 0.026 N46.91521 W1 12.72527 Blanchard Creek T14N,R14W,5B 1 25-Sep-06 1.503 N47.00349 W1 13.40356 Upstream of Ricfiards diversion Cottonwood Lake T16N,R14W,10A 26-Apr-07 0.026 N47. 16371 W1 13.34865 Stream connecting upper and middle Cottonwood Lal C» C> C» C> o' o o o o A r V Si r uT <>r r^ nT ^ ^' ^ <^ <^ <^ «f cT A A XS JO '^' <<> ^ ^ ^<^ X O? A^ o.^ P^ V- V- ^ fr $r $>' r. ^ p* ^ '^^ r.>^ .->^ ^^^ i?-^-' i^^" ^^^^ ^^^'^ cT.cT .p^ ..p^ .p >' ' o'y 'b'^ ^^ N*" ^' N^' T^" Month Max Temp Min Temp Ave Temp StDev Temp Var Temp June 68.61 61.09 65.01 2.32 5.38 July 77.1 59.38 67.51 3.98 15.85 August 72.48 52.66 63.91 4.25 18.07 September 66.57 43.74 54.86 5.27 27.73 October 57.67 37.83 46.01 4.72 22.31 Blackfoot River @ Scotty Brown Bridge (Mile -46.1) -2006 Month Max Temp Min Temp Ave Temp StDev Temp Var Temp April 49.06 39.17 43.19 2.28 5.18 May 54.23 40.74 46.56 2.73 7.47 June 63.16 48.02 54.25 3.72 13.87 July 68 55.27 61.72 2.76 7.62 August 65.31 50.61 59.3 3.04 9.24 September 61.02 44.89 52.85 3.69 13.61 October 53.19 47.5 50.42 1.44 2.07 310 Blackfoot River @ USGS Gage Station (l\/lile - 7.9) - 2006 Month Max Temp Min Temp Ave Temp StDev Temp Var Temp March 41.95 32.95 38.56 2.09 4.38 April 50.25 38.81 43.31 2.59 6.71 May 53.35 41.95 48.63 2.76 7.62 June 67.09 50.76 56.88 3.89 15.12 July 69.26 56.99 63.35 2.48 6.13 August 63.86 53.35 59.26 2.13 4.52 September 59.61 46.10 53.30 3.32 11.01 October 54.41 36.22 46.02 3.75 14.08 November 46.06 33.93 38.21 2.61 6.83 December 43.82 31.9 3576 2.01 4.04 311 Copper Creek @ Sucker Creek Bridge (Mile - 1.1) - 2006 9> 9> ?> 9> ^ ^ ^ ^ S> c> c> c> c> 9> 9> ?* 9> ?> ^ ^ ^ ^ .•$> .^ -•$> -^ .^ -^ ii „ -^ ^' ^ *^ ,^ / / / / .^ / / / V Month Max Temp Min Temp Ave Temp StDev Temp Var Temp June 55.97 45.38 48.91 3.56 12.66 July 58.04 43.92 49.54 3.98 15.84 August 57.35 45.38 50.10 3.17 10.04 September 55.28 43.92 48.74 2.49 6.23 October 51.08 43.92 47.51 1.93 3.74 313 Landers Fork@ Hwy 200 (Mile - 1.1) -2006 <§* <5> ^ »s> ^ ^ ^ ^ Q Q Q Q Q Q ' ^ ^ ^ Ch J^ qT \>r ^? pT ,T ^ ^ T <§» <§» <§» r5? ,5? ^ iV' A' 'V'b -Jb -Q> - ^V' ' t^ ^^ .^ ^^ ^^ o^ i'' i^ i^ i^ ^S J^ J^ J^ ^ " i? <^ ^ 'T / i.^ / / '?' ^ ^ ^ ■^ Month Max Temp Min Temp Ave Temp StDev Temp Var Temp June 73.15 65.59 69.38 2.33 5.43 July 80.12 60.8 70.42 4.04 16.31 August 75.22 52.49 64.71 4.71 22.17 September 66.96 43.92 54.36 5.17 26.75 October 56.66 47.53 51.11 2.35 5.54 Nevada Creek below Nevada Spring Creek (Mile -4.5) -2006 (0 0) Q. E 0) 'V^ 5§» ^5§? ^^ ^5? ^s? sS* S? ^"^ ^ S? S? S? ^ ^ ^^ ^^ ^ ^^ ' ^' K^' <^ <^' ^' ^^ ^^ ^^ ^ ,Si ,$> ^ ^ ^ ^ ^ ^ ^ J> J> J> J^ .S J^ ^' '^ K> Ky '^ <6^ K^^ ^^^ rf ^ ^ ^ Month Max Temp Min Temp Ave Temp StDev Temp VarTemp June 69.00 58.70 63.67 3.37 11.34 July 71.10 53.20 63.38 3.75 14.08 August 63.50 46.80 56.16 3.47 12.01 September 58.70 41.70 50.65 3.85 14.86 Nevada Spring Creek @ lower bridge (Mile -1.1) -2006 (0 0) Q. E (1) ♦ ^^ ^^ ;>^ i^ i^ i^ i«> ^^ ^ ^ ^ ^ ^ -i ^ ^ ^ < K ^ ^ ^^ # # Month Max Temp Min Temp Ave Temp StDev Temp Var Temp June 69.70 56.00 60.85 3.39 11.52 July 67.70 49.00 60.47 3.60 12.99 August 61.50 45.40 54.16 3.85 14.80 September 56.70 41.70 49.53 3.60 12.98 316 Nevada Spring Creek @ upper fenceline (Mile -3.5) -2006 to Jo to to to Jb to to to to Jo to to J=> to' ''3' ci' V 5?* 5r ,5r 5r 5r J^ J> J> MorillT— Max Temp Iviin Temp Ave Temp STOev'Temp VarTemp June 51.80 46.80 48.52 1.71 2.94 July 53.20 45.40 48.76 2.11 4.47 August 55.30 45.40 48.71 2.33 5.45 September 54.60 42.50 47.57 2.79 7.77 North Forl< Blacl' Month Max Temp Min Temp Avg Temp StDev Temp Var Temp April 53.78 35.5 43.8 3.9 15.25 May 58.26 38.36 48.46 3.88 15.04 June 64.28 44.84 53.44 4.06 16.5 July 69.56 51.83 61.31 3.94 15.49 August 66.03 47.07 55.65 4.08 16.61 September 62.84 42.59 53.74 4.73 22.4 319 Blackfoot River above Belmont Creek (Mile - 21 .8) - 2007 ^ ^ ^ ^ ^ ^ ^ ^ .^ 9i\9i^ 9i C< ^ A (b^_<^_Q>\c> ^'' ^'^''o,/a>^^'.>''^ ^^^ ^^** ^^'^ /^s.^"^' s*' ^' «b^ v 9> 9>^ P' 9y 6^ 9> 9> 9>^ 9>^ 9>^ P 9> P 6^ 6 Month Max Temp Min Temp Avg Temp StPev Temp Var Temp April 56.02 38.99 48.02 4.14 17.14 May 58.82 41.42 51.43 3.66 13.42 June 67.45 47.11 57.66 4.75 22.59 July 76.46 61.1 69.31 3.28 10.74 August 73.68 54.07 63.07 3.97 15.8 September 68.9 48.49 58.79 4.72 22.23 Blackfoot River @ Scotty Brown Bridge (Mile -46.1) 2007 J" J" / / J" / / / s^^ / J J J" Month Max Temp Min Temp Avg Temp StDev Temp Var Temp April 53.8 37.9 46.22 3.55 12.58 May 57.69 39.87 48.65 3.46 11.95 June 66.57 45.46 55.46 4.77 22.74 July 72.18 56.03 65.15 3.61 13.02 August 67.74 52.4 59.62 3.16 9.97 September 63.39 50.18 58.11 3.11 9.7 322 Blackfoot River @ USGS Gage Station (Mile - 7.9) - 2007 /" /^ <,^ • ^' / ^^^' #^ / of J" i^ <^^ '^ n'V ^' f^ ^ T?>' N* N^' Month Max Temp Min Temp Avg Temp StDev Temp Var Temp January 43.54 31.9 36.38 3.71 13.76 February 43.54 32.48 36.85 3.38 11.43 March 44.94 33.64 39.47 2.53 6.42 April 52.64 39.61 45.01 3.18 10.13 May 56.64 42.42 56.71 2.85 8.14 June 67.52 48.27 57.9 4.1 16.77 July 72.86 60.6 67.53 2.62 6.85 August 68.68 54.13 61.24 2.94 8.63 September 65.19 52.74 59.58 3.09 9.54 Chamberlain Creek @ mouth (Mile - 0.1) - 2007 IS a E |2 75 70 65 60 55 50 45 40 ^ ^ \ ' j^:;.:^.o=i|-t^^^^=^::^ 4i^^^4J-^^'^ -s> .-^^->-] ; L^^FTrT^ Tt^^ -s -■ ■ ■ 1 1 l~ 1 1 1 1 1 1 1 1 1 O^ .9^ o^ .o^ 9^ ^ .' A n'b «0 ^ ^ ^ ^ v>^' / y^-^'* .>•*.>•* y v-^* ^-^^ ^^* v-^* 'o^ ^^ v" m" ■v 9, ^ a ^ ^y ^y -y y ^v r' ^ Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 56.7 41.7 49.58 3.34 11.16 July 63.5 47.6 56.39 3.52 12.39 August 61.5 43.2 51.53 3.77 14.23 September 57.4 39.5 49.78 3.97 15.77 324 Copper Creek @ Sucker Creek Bridge (Mile - 1.1) -2007 65 60 55 50 45 y^ 40 J w VvV, \ V ^\ = .^ ^v \' ^\ v\: ^^ ^5- \ sV vvS vs NV V 0^ o^ (^ ^^ .^>' .P^ v^ .p^ ,^^ u.^ J. ^ Jr

^'^' <^'''^' <0^ V V Month Max Temp Min Temp Avg Temp StDev Temp Var Temp July 68.83 46.74 56.54 5.46 29.79 August 67.09 43.96 54.06 5.41 29.25 September 66.22 39.47 52.79 6.06 36.68 Enders Spring Creek (Mile - 0.1 ) - 2007 S2 .0) 65 60 55 50 45 40 35 1 ^^ % s\- \^ ^\ \' \^' w\ ^» ^sv^ K. p^ 5^ <^ 5^^ 5^^ J^ J^ J^ ^ J' J' 5^ J^ J^ J^ J^ ^^ Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 57.35 39.48 46.49 4.29 18.4 July 60.11 43.19 51.64 4.25 18.06 August 57.35 43.19 49.86 3.41 11.61 September 55.28 41.72 48.59 3.05 9.28 326 Frazier Creek above Upper Frazier Creek Reservoir (Mile -1.2) -2007 0) a E 0) ^^^ ^^-^^ V>^ /" /" ,/' / ,/ J^ J" J" J" C,^ J^ T?' -b Month Max Temp Min Temp Avg Temp StPev Temp Var Temp May 51.79 38.72 43.52 2.93 8.57 June 60.8 42.46 50.15 3.99 15.96 July 68.33 49.67 60.04 3.79 14.33 August 64.91 44.65 54.77 4.27 18.25 September 60.8 39.48 50.62 5.19 26.94 Frazier Creek @ mouth (Mile - 0.1 ) - 2007 |2 #^' #^' ^^^ /" /" y^>^ J" i^"^ s^^ i^"^ i^"" c,^ c^"^ T?' 'S Month Max Temp Min Temp Avg Temp StDev Temp Var Temp May 61.48 46.82 52.68 3.28 10.78 June 69.71 50.38 58.77 4.24 17.95 July 73.84 57.35 66.54 3.53 12.46 August 69.71 52.49 60.17 3.54 12.55 September 63.54 48.96 56.88 3.57 12.75 327 Frazier Creek below Middle Frazier Creek Reservoir (Mile -0.4)- 2007 S2 0) .0) .^ ^ ^<3J^ ^ ^ .^' ^ ^ ^ ^ ^ ^ ^ J!> J^ J^ ^t?P! ^ ^^^ ^^^ b^" b^" b^ *,>^\.>^ ^^ v>^ v^^ ^^ ^iT / / / / / / J" J" J" J" / Month Max Temp Min Temp Avg Temp StDev Temp Var Temp May 54.58 46.82 50.21 1.93 3.71 June 58.04 46.18 52.71 2.48 6.13 July 62.17 49.67 56.57 2.72 7.41 August 61.48 47.53 55.16 3.15 9.93 September 59.42 43.92 51.91 4.1 16.82 328 Frazier Creek "North Fork" (Mile -0.1) -2007 6' 6' 6' p* S> 6^ P' vP' P' P P 9> P^ ' ^ ^ ^r r^-" ^-^ r Ok'^ V A ^' V^ oF N*" '^ «b'' nV' r^ ^ '^ r^' ^''n^'"'^'' b." n'^' rQT Month April May June July August September Max Temp 51.96 58.09 67.01 71.14 66.42 62.37 Min Temp 35.11 36.54 44.71 50.84 46.94 42.46 Avg Temp 43.2 46.55 54.23 62.72 56.31 53.88 StPev Temp 3.62 4.21 5.04 4.7 4.28 4.57 Var Temp 13.14 17.73 25.44 22.12 18.33 20.87 329 Hoyt Creek (Mile - 1 .3) - 2007 3 0) ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ N*' X ^r rv^ 0>> ^> ^^ ^^ ^y ^^ ^^ .^ .5b .5b^ ^ ' 'b' ^Q' Month Max Temp Min Temp Avg Temp StPev Temp Var Temp June 76.62 44.65 60.18 7.43 55.23 July 78.71 51.08 66.86 6.3 39 August 74.53 46.1 59.34 6.26 39.22 September 69.71 42.46 56.91 6.34 40.26 Hoyt Creek (Mile - 4.3) - 2007 Si 3 Q. E 65 60 55 50 45 40 \s\ S\' \\\ w \ \\' \s \0s s \ ^1 \\ \"^^ \ ^ .^ ^ .^ ^ ^ ^ ^ ^ ^ ^ -v*' so''' y v>^' v^' s»^' v^' ^»*' ^->« v^*' j^ ^^ j>' cf' ? N"^ kO-^ kV' M^ A"^ J^ Jf- J>^ kS- aS- nf J> ^ ^- ^. nV n> ■^' Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 60.11 50.38 54.44 3.08 9.5 July 62.85 47.53 55.44 3.92 15.34 August 61.48 45.38 53.59 3.63 13.15 September 59.42 42.46 52.51 4 16.04 330 Landers Fork @ Hwy 200 (Mile - 1.0) - 2007 IS 0) a E o ^ ^ ^ ^.^.^.^.^ ^ ^ ^ ^ ^ ^ ^ ^ N^"n?"n?' rj^ n? ^' ' N*^' n^' 'b'^' ' o?>'^o^'^ ^ * N^ n? n? ^^ \^ ^^ A^ nV „:^^ <6^ .^ \^ t>' Month Max Temp Min Temp r Avg Temp StDev Temp Var temp June 63.67 46.49 54.81 4.16 17.33 July 71.02 52.91 62.67 3.91 15.3 August 68.93 48.16 58.34 4.51 20.36 September 66 45.93 56.61 4.61 21.29 331 Monture Creek @ USPS Bridge (Mile - 13.1) - 2007 60 £ 55 9) S 50 0) Q. E I- 45 40 \^ _ ~s I 5$^ ^ ^ .^ .^ ^ ^ ^ ^ ^ ^ «?>• v->*'' v*''' S^' #' ^•^' V^' ^•^' ,.»*' ^ ^ ^ cS^' c^"^ ^ •P <>■ '(>■ 4' nO' Month Max Temp Min Temp Avg Temp StPev Temp Var Temp June 55 42.41 48.82 2.88 8.3 July 55.84 46.06 50.3 2.5 6.25 August 53.88 44.1 47.73 2.2 4.85 September 51.09 42.7 46.21 1.96 3.85 Nevada Creek below Nevada Spring Creek (Mile -4.5)- 2007 80 75 £ 70 B 65 0) 60 Q. a) 55 50 45 4t ^ ft 5k l -s' ^q^ v^% ^:::V ^ ^ ^ .^ ^ ^ ^ ^ ^^' so^' b^' b^' b^' J^ J^ J^ J^ > V^ y y v^ o>.^ ^^ Jl^ ^ <^' $>- >J>^ >^ >^ >^ ^^ ^ ^ ^ ^ ^ ^ ^ ^^'^ ^^^ V^>>^ V^>>^ V'^^ J^ Month Max Temp Min Temp Avg Temp StPev Temp Var Temp June 67 46.8 56.83 5.09 25.95 July 75.3 53.9 64.91 4.33 18.78 August 71.1 50.4 59.5 4.17 17.37 September 65.6 47.6 56.99 4.26 18.16 North Fork Blackfoot River @ Ovando-Helmville Rd (Mile -2.6) -2007 J Month January February March April May June July August September Max Temp 42.42 45.76 48.8 50.47 52.69 60.4 64.78 62.2 61.07 Min Temp 31.65 33.37 33.96 36.24 38.22 42.69 48.52 46.59 45.2 Avg Temp 36.99 39.02 40.58 42.37 44.81 50.15 55.84 53.02 51.94 StPev Temp 2.42 2.18 2.83 3.07 2.75 3.79 4.12 3.86 3.92 Var Temp 5.85 4.75 7.99 9.44 7.61 14.34 16.98 14.89 15.35 333 3 0) North Fork Blackfoot River @ USPS Bridge (Mile - 17.5) 2007 60 ^ 55 50 45 40 V N N' •>\^s .\ V\s' sS \[ v^ \ ^ ^ ^ .^ ^ ^ ^ ^ ^^ ^^ ^^ ^^ ^^ ^ . >'*^ >'*'' >'*^ ^>^ o.^'" «.>^ b.>'' o>^ V=^ J?>^ J?>^ J?>^ V^^ ^"^ ^^ N* r^ r§> * ^ N* '^ '^ *>^ ^^ ^^ ^ N- TT v q>" t^'^ ffp'' V ,$> Month June July August September Max Temp 58.7 63.5 61.5 61.5 Min Temp 44.7 49 46.8 43.9 Avg Temp 50.39 55.36 53.37 52.23 StDev Temp 3.22 3.82 3.6 3.86 Var Temp 10.34 14.63 12.98 14.87 334 Rock Creek @ County X-ing Rd (Mile - 1 .6) - 2007 85 80 ^ 75 u. ^ 70 ■i 65 o a E 60 55 50 45 40 ! S* ^S^^ P ^ ^s^s"^^ S k ^_ --s ^-^ ^ - ■ \^^ ^ ^ ^ ^ ^ O^ O^ <^ <^ O^ Q^ O^ - _ ,><^' ^^' >j>' >5>' >5>' ^^' J^ J^ Ji J^ J( J( ^ Month Max Temp Min Temp Avg Temp StPev Temp Var Temp June 71.4 43.8 58.55 6.49 42.06 July 82.96 53.02 67.77 6.08 36.97 August 78.8 49.38 60.33 5.95 35.39 September 75.98 42.95 58.96 7.78 60.6 Sauerkraut Creek (Mile 0.5) - 2007 75 70 £ 65 SJ 60 1 55 I 50 I- 45 40 35 i s. .'\''\y^^^" J ^- ^. t ^^ ,^ 1 1 1 1 1 1 1 1 1 1 1 ^ ^ ^ .^ .^ ^ ^ ^ ^ ^ ^ ^ .O^' ssf" J>' b^' b^"^' b^' b^' ^>>« ^>>« ^C.^ J^ J^ Ji'j^ ^ 'V t? Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 66.61 43.47 53.66 5.59 31.22 July 73.44 49.04 60.78 6.46 41.7 August 71.63 44.59 57.01 6.46 41.7 September 68.36 39.81 53.1 6.62 43.84 335 Shanley Creek @ Woodworth Rd (Mile - 0.4) - 2007 I a E .0) 65 60 55 50 45 40 m V ^ V 1 'l|i -M M'||~- I'lM s \ 1 35 / / N^^ N^^ X^^"^ / X^^"^ C b^ /^ o>^ o.>^ ..>^ ^>^ cs>^ V-^ V>' ^ ^ ^ ^ 'b' 9^.9?'''^ ^ N^' Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 57.27 40.85 49.49 3.85 14.83 July 59.53 44.48 52.74 3.75 14.09 August 58.39 44.2 51.4 3.11 9.69 September 57.27 40.28 50.4 3.8 14.48 Upper Willow Creek (Mile - 0.7) - 2007 75 70 £ 65 3 60 ^ 55 E « 50 45 40 SNj S^\ '^.. ^v\ ~0~ s sy \ N N. \^ \v, Vv vO' ^ ^ .^^' v>' ^ .^' ^ v>' ^ ^ ^ ^ J^' ^ .4^ .s<5> ^ V- V^^^ c,^"^ cf ^ Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 64.9 46.1 55 4.84 23.42 July 72.5 54.6 63.78 3.85 14.86 August 69 47.6 57.28 4.47 20.01 September 62.8 43.2 53.44 4.56 20.83 336 Warren Creek @ lower bridge (Mile -1.1)- 2007 80 75 £ 70 3 65 a E .0) 1 ,"~ s ^^'"fr^^'iih ^s ^ 1 1 60 55 50 45 ^ ^ <^ <^ <$^ ^ N^^' N^^' b^' b^' -^ -^"^ v>> vO v>> v> 'V ^' K^' <^' <§>' V^' ^ ^ ^ ^ ^ y J^ J^ J^ J^ c,^ cf ^ I' 1 1 1 1 1 1 1 1 1 1 1 ^ ^ .^.^ .^ .^ ^ ^ ^ ^ ^ ^ N^' 'V- Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 64.41 45.28 54.68 4.03 16.25 July 66.17 50.02 57.57 3.39 11.5 August 59.82 47.23 53.98 2.73 7.46 September 61.53 43.6 53.76 4.53 20.6 337 Wasson Creek @ H wy 1 41 (Mile - 1 .3) - 2007 70 65 £ 60 3 55 ^ 50 E ^ 45 40 35 iklN^lMLL ' [^ L^'^'^ "^If'-^'l^^ s v>' ^ v>' ^ ^ VN7 VJ VN? V> v>' ^ ^' ' ^ ^ >.« ^ n>' ^ ^' ^ ^ N' ^^ . cf ^ N^' Month Max Temp Min Temp Avg Temp StDev Temp Var Temp July 67 51.1 59.82 3.9 15.18 August 63.5 44.7 53.78 3.95 15.62 September 60.1 40.2 56. .86 4.57 20.93 West Twin Creels (Mile - 0.1) - 2007 70 65 60 2 55 0) Q. E 0) 50 45 40 5i^ -s V O \ v^ ^^s ^\. fi^ ^ ^ .^ ^ ^ ^ ^ ^ ^ ^ N*' v>^' v>^' >^' # V^"" >>^' # ^^^' ^^^' ^^^' ^^^' O.^' ' '¥ ' n? 'V>' N^' Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 57.06 42.83 50.9 3.2 10.25 July 64.49 48.4 58.28 3.53 12.46 August 62.47 47.29 54.78 3.1 9.6 September 60.47 43.67 53.57 3.97 15.76 338 Willow Creek @ Dalton Mtn Rd (Mile - 1.6) - 2007 80 75 iT 70 65 0) S 60 0) I" 55 0) I- 50 45 40 ^' \' \ ^\ ^^ \'. \ s. ^\ -\ v^' ^ ^ ^ ^ .^ .^ .^ ^ ^ 5^ ^ ^ ^ ^ ^ ^y 1^ Month Max Temp Min Temp Avg Temp StPev Temp Var Temp June 72.46 46.1 59.05 6.46 41.75 July 78.01 54.58 67.12 5.78 33.38 August 74.53 48.96 59.72 5.65 31.89 September 69.71 43.19 56.64 6.13 37.6 Willow Creek (Mile - 3.7) - 2007 I 0) a E 70 65 60 55 50 45 40 35 i\ SsV -N \^ ^v^-^vV \ \ v\\ \\ N ..-.-- 5^-^ ^ ^ ^.^ .^ ^ A A .^ ^^ ^^ > ^ s:^\^\<^' !J^' b^' !f^ !}^' 's^' ^ ^ ^ ^ ^ ^ ^^'' ^<^ > cs> 0,> < 'b^ .V" „C>^ .<6^ .y ^y Jr o.y cX K^ J> ^f ^ N^" T?" <^ 'V n? n'^" N^" 'V '^'^ Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 60.8 43.19 51.96 4.54 20.62 July 68.33 50.38 60.04 3.98 15.1 August 64.91 46.1 54.69 4.07 16.59 September 60.8 40.23 52.06 4.89 23.93 339 Willow Creek (Mile - 5.4) - 2007 0) 3 70 65 60 S 55 0) Q. E 0) 50 45 40 N -^ ^ N N">.S; 5^ ^ ^ 5^ .5^ .5^. .^.5^ 5^ 5^ 5^ ^ ^ ^ ^ ^ \^' <^ <\^ Month Max Temp Min Temp Avg Temp StDev Temp Var Temp June 59.42 40.97 49.34 4.27 6.26 July 65.59 47.53 57.1 3.96 15.68 August 62.17 46.1 53.22 3.53 12.49 September 57.35 40.97 50.65 4.17 17.36 340 Appendix 1 : Summary of water chemistry readings for 2006. stream name Date River IVIile PH Conductivity (uS) TDS (ppm) Tem P°F Lat Long TRS Anaconda Cr@ mile 0.1 25-Jul-06 0.1 8.12 160 80 50.2 N47.03485 W1 12.35775 15N,6W,27B Anaconda Cr@ mile 0.5 25-Jul-06 0.5 8.14 161 80 49.8 N47.03524 W1 12.351 25 15N,6W,22C Beartrap Cr @ mile 0.2 25-Jul-06 0.2 8.27 296 147 58.6 N47.03397 W1 12.35729 15N,6W,27B BeartrapCr@ mile 1.2 25-Jul-06 1.2 8.22 183 90 62.6 N47.02128 W1 12.34695 15N,6W,27C Blacl0.05) among the fish in the sample. Thus, when this reach of Donovan Creek was sampled it appears to have possessed a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (97%) westslope cutthroat trout genetic contribution. Kendall Creek 3283 Individuals were collected from two locations in Kendall Creek: site 1 (N=15) and site 2 (N=12). There was no evidence of genetic differences between the fish collected from the two sites so they were combined into one for further analysis. PINE fragments usually characteristic of rainbow trout were detected at all six diagnostic loci analyzed in the sample that distinguish rainbow from westslope cutthroat trout. The rainbow trout fragments were not randomly distributed (Poisson distribution, P<0.001) among the fish in the sample. In contrast, they were detected in only two fish. One of these (#27) possessed PINE fragments characteristic of only rainbow trout suggesting it to be a non-hybridized rainbow trout. Another fish (#1) possessed PINE fragments characteristic of rainbow trout at all six diagnostic loci analyzed for this fish and possessed PINE fragments characteristic of westslope cutthroat trout at all seven diagnostic loci analyzed for this fish. This individual, therefore, appears to be a first generation hybrid between rainbow and westslope cutthroat trout. The remaining 25 fish in the sample possessed PINE fragments characteristic of only westslope cutthroat trout suggesting them to be non-hybridized westslope cutthroat trout. When Kendall Creek was sampled, therefore, it appears to have contained a mixture of non-hybridized westslope cutthroat trout, non-hybridized rainbow trout, and first generation hybrids between these fishes. Tura Creek 3284 Fish were collected from two locations in Tura Creek: site 1 (N=10) and site 2 (N=5). Unfortunately, only poor quality DNA was obtainable from the fish collected from site 1 so no data are available from this collection. In the site 2 sample, PINE fragments characteristic of only westslope cutthroat trout were detected. With the sample size of five, we have only a 43 % chance of detecting as little as a one percent rainbow trout and a 33% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Thus, we can not reasonably exclude the possibility that the Tura Creek population may be slightly hybridized with rainbow trout, westslope cutthroat trout, or both of these fishes. Although the status of this population is uncertain, conservatively it should be considered to be non-hybridized westslope cutthroat trout unless further data indicate otherwise. 349 Straight Creek above falls 3286 PINE fragments characteristic of only westslope cutthroat trout were detected in the sample. A previous PINE analysis offish collected from Straight Creek above the falls (sample # 1961, N=l 8) also detected fragments characteristic of only westslope cutthroat trout. With the combined sample size of 26, we have a 96% chance of detecting as little as a one percent rainbow trout and an 88% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Straight Creek above the falls, therefore, very likely contains a non-hybridized westslope cutthroat trout population. Crow Creek near mouth 3287 PINE fragments characteristic of only westslope cutthroat trout were detected in the sample. With the sample size of 25, we have a 95% chance of detecting as little as a one percent rainbow trout and an 87% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Crow Creek, therefore, very likely contains a non- hybridized westslope cutthroat trout population. Maryann Creek 3288 PINE fragments characteristic of only westslope cutthroat trout were detected in the sample. With the sample size of 25, we have a 95% chance of detecting as little as a one percent rainbow trout and an 87% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Maryann Creek, therefore, very likely contains a non- hybridized westslope cutthroat trout population. Lookout Creek 3289 PINE fragments characteristic of only westslope cutthroat trout were detected in the sample. With the sample size of 25, we have a 95% chance of detecting as little as a one percent rainbow trout and an 87% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Lookout Creek, therefore, very likely contains a non- hybridized westslope cutthroat trout population. Rock Creek 3290 PINE fragments characteristic of only westslope cutthroat trout were detected in the sample. With the sample size of 25, we have a 95% chance of detecting as little as a one percent rainbow trout and an 87% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Rock Creek, therefore, very likely contains a non- hybridized westslope cutthroat trout population. Robb Leary Ben Wright 350 TABLE 1 Diagnostic PINE markers for westslope cutthroat, Yellowstone cutthroat, and rainbow trout. X indicates the fragment is present in the particular taxon. Markers Yellowstone Westslope Rainbow Hpa1 5VHpa1 3' 232 X 153 X 72 X X 70 X 69 X X 66 X Fok1 5'/Tc1 369 X 366 X X 230 X 159 X 138 X 110 X Hpa1 5V33.6+2 395 X 388 X X 266 X 248 X 148 X X 351 Montana Conservation Genetics Laboratory Division of Biological Sciences * University of Montana * Missoula, MT 59812 (406)243-5503/6749 Fax (406)243-4184 July 11,2006 Ron Pierce Genetics Contact, Region 2 Mt. Dept. of Fish, Wildlife, and Parks 3201 Spurgin Road Missoula, MT 59801 Ron: The paired interspersed nuclear DNA elements (PINE) technique has been used to analyze DNA from the following trout samples: Sununaiy of results. Sample # Individuals Water Name/Location/Collection Date/ Collector a N b c # markers Species ID d Power (%) % WCT 3299 10 Blackfoot River 20 R7Y4 RBTXWCT 10 3/22/2006 Ron Pierce RBT? ^Number offish successfully analyzed. If combined with a previous sample, the number in parentheses indicates the combined sample size 'TSTumber of markers analyzed that are diagnostic for the non- native species (R=rainbow trout, W=westslope cutthroat trout, Y= Yellowstone cutthroat trout). ■^Codes: WCT = westslope cutthroat trout {Oncorhynchus clarki lewisi); RBT = rainbow trout (O. mykiss); YCT = Yellowstone cutthroat trout (O. clarki bouvieri). Only one species code is listed when the entire sample possessed alleles from that species only. However, it must be noted that we cannot definitively rule out the possibility that some or all of the individuals are hybrids. We may not have detected any non-native alleles at the loci examined because of sampling error (see Power %). Species codes separated by "x" indicate hybridization between those species. ''Number corresponds to the percent chance we have to detect 1% hybridization given the number of individuals successfully analyzed and the number of diagnostic markers used. For example, 25 individuals are required to yield a 95% chance to detect as little as 1% hybridization with rainbow or an 87% chance to detect as little as 1% hybridization with Yellowstone cutthroat trout into what once was a westslope cutthroat trout population. Not reported when hybridization is detected. Indicates the genetic contribution of the hybridizing taxa in the order listed under c to the sample assuming Hardy-Weinburg proportions. This number is reported if the sample appears to have come from a hybrid swarm. That is, a random mating population in which species markers are randomly distributed among individuals. Indicates number of individuals with genetic characteristics corresponding to the species code column when the sample can be analyzed on the individual level. This occurs when marker alleles are not randomly distributed among individuals and hybridization appears to be recent and/or if the sample appears to consist of a mixture of populations and hybrids and non- hybrids can be reliably distinguished. Methods and Data Analysis The PINE technique uses short synthetically made segments of DNA called primers, in pairs, to search for relatively small segments of organismal DNA flanked by particular, often viral, DNA inserts. During the polymerase chain reaction (PCR), the primers bind to the ends of the inserts and 352 many copies of the organismal DNA between the primers are made. While the DNA from some organisms may have two appropriately spaced inserts to which the primers can attach, the DNA from other organisms may have only one or none of the appropriately spaced inserts in particular regions. During PCR we will fail to copy DNA in the latter two cases. Thus, the PINE technique coupled with PCR is used to search for evidence of genetic variation based on the presence or absence of particular DNA fragments. The fragments are labeled by the primers used to produce them and their length in terms of the number of nucleotides in the fragment. The fragments are made using dye labeled nucleotides and after PCR are separated from each other via electrophoresis in polyacrylamide gels. Smaller fragments move through the gels at a faster rate than larger fragments. The use of dye labeled nucleotides allows one to visualize the position of the fragments in the gels after electrophoresis using a spectrophotometer and the size of the fragments is determined by comparison to the position of synthetic fragments of known size that were also migrated into the gel. When DNA from westslope cutthroat trout, Oncorhynchus clarki lewisi, and rainbow trout, O. mykiss, is compared with PINE analysis and three different pairs of primers seven fragments are usually characteristic of westslope cutthroat trout and seven fragments are usually characteristic of rainbow trout (Table 1). Likewise, when DNA from westslope and Yellowstone cutthroat trout, O. c. bouvieri, is compared using the same procedure two fragments are usually characteristic of westslope cutthroat trout and four fragments are usually characteristic of Yellowstone cutthroat trout (Table 1). Fragments produced from the DNA of one taxon and not another are commonly termed diagnostic or marker loci because they can be used to help determine whether a sample came from a non- hybridized population of one of the taxa or a population in which hybridization between them has or is occurring. Individuals from a non-hybridized population will possess fragments characteristic of only that taxon. In contrast, since half the DNA of first generation hybrids comes from each of the parental taxa the DNA from such individuals will yield all the fragments characteristic of the two parental taxa. In later generation hybrids, the amount and particular regions of DNA acquired from the parental taxa will vary among individuals. Thus, DNA from later generation hybrid individuals will yield only a subset of the parental fragments and the particular subset will vary among individuals. In a sample from a random mating hybrid swarm, that is a population in which the genetic material (i.e. fragments) of the parental taxa is randomly distributed among individuals such that essentially all of them are of hybrid origin, the frequency of the fragment producing allele from the non-native taxon is expected to be nearly equal among the diagnostic loci since their presence can all be traced to a common origin or origins. Thus, if a sample contains substantial variation at only a single marker locus where the presence of the fragment is usually characteristic of a non-native taxon and lacks such fragments at all other markers this is probably not indicative of hybridization. Rather, it much more likely represents the existence of genetic variation for the presence or absence of the fragment within this particular population of the native taxon. An important aspect of PINE marker loci is that individuals homozygous for the presence allele (pp) or heterozygous (pa) will both yield the fragment. That is,p is dominant to a. Thus, in order to estimate the genetic contribution of the native taxon to a hybrid swarm we concentrate on the marker loci at which the/? allele is characteristic of the non-native taxon. Furthermore, we must assume that genotypic distributions in the population reasonably conform to expected random mating proportions. Under this assumption the frequency of the native a allele is approximately the square root of the frequency of individuals in the population lacking the fragment (aa). The frequency of the non-native allele then is one minus this value. We focus on the/? alleles characteristic of the non- native taxon because with low levels of hybridization it is the presence of these alleles that are likely to provide evidence of hybridization. With low levels of hybridization, it is likely all individuals in 353 the sample will genotypically hepp or pa where the/? allele is characteristic of the native taxon. Thus, like in non-hybridized populations all individuals in the sample will yield the fragment providing no evidence of hybridization. Failure to detect evidence of hybridization in a sample does not necessarily mean the population is non-hybridized because there is always the possibility that we would not detect evidence of hybridization because of sampling error. In order to assess the likelihood the population is non- hybridized, we determine the chances of not detecting as little as a one percent genetic contribution of a non-native taxon to a hybrid swarm. This is simply 0.99 ^^^ where N is the number offish in the sample and X is the number of marker loci where the/? allele is characteristic of the non-native taxon. In samples showing evidence of hybridization, that is; fragments characteristic of a non-native taxon were detected at two or more marker loci, we used two approaches to determine if the population appeared to be a hybrid swarm. First, contingency table chi-square analysis was used to test for heterogeneity of allele frequencies among the marker loci. Next, we computed a hybrid index for each individual in the sample. Each diagnostic locus at which an individual possessed a PINE fragment characteristic of the non-native taxon was given a value of one. Each diagnostic locus at which an individual did not possess a PINE fragment characteristic of the non-native taxon was given a value of zero. These values summed over all diagnostic loci represent an individual's hybrid index. The observed distribution of hybrid index scores was then statistically compared to the expected random binomial distribution based on the estimated native and non-native genetic contributions to the sample. If the allele frequencies were statistically homogeneous among the diagnostic loci and the observed distribution of hybrid indices statistically conformed to the expected random binomial distribution, then the sample was considered to have come from a hybrid swarm. Heterogeneity of allele frequencies among marker loci can arise in very old hybrid swarms as the frequencies over time diverge from each other due to genetic drift. In this case, however, the non- native fragments will still be randomly distributed among individuals. Thus, samples with these characteristics were also considered to have come from hybrid swarms. There are two likely reasons why a non-random distribution of non-native fragments may be observed among individuals in a sample. It may contain individuals from genetically divergent populations with different amounts of hybridization or hybridization may have only recently occurred in the population. Based on PINE data alone, these two situations will generally be difficult to distinguish from each other. Regardless of the explanation, when the non-native fragments are not randomly distributed among individuals in a sample estimating a mean level of hybridization has little, if any, biological meaning and, therefore, is often not estimated. Results and Discussion: Blackfoot River 3299 All of the fish in the sample contained PINE fragments usually characteristic of rainbow trout at six or seven of the diagnostic loci analyzed that distinguish rainbow from westslope cutthroat trout. PINE fragments usually characteristic of westslope cutthroat trout were also detected at two of the seven diagnostic loci analyzed that distinguish westslope cutthroat from rainbow trout. The frequencies of the westslope cutthroat trout PINE fragments were statistically heterogeneous {X 6= 32.925, P<0.001) among the diagnostic loci and, they were not randomly distributed (Poisson distribution, X j= 288.478, P<0.001) among the fish in the sample. In contrast, significantly fewer individuals possessed no PINE fragments characteristic of westslope 354 cutthroat trout and significantly more possessed PESTE fragments characteristic of westslope cutthroat trout at one or two loci than expected by chance (Fig. 1). Overall, these results suggest that the sample contained individuals from two or more genetically different populations. On an individual basis, ten fish in the sample (#'s 1751, 1752, 1754, 1755, 1757, 1858, 1862,1863, 1967, and 1969) possessed PINE fragments characteristic of both rainbow and westslope cutthroat trout definitely indicating that they were post first generation hybrids between these fishes with a predominant rainbow trout genetic contribution. The remaining ten fish in the sample possessed PESTE fragments characteristic of only rainbow trout suggesting that they may be non-hybridized rainbow trout. We cannot exclude the possibility, however, that some or all of these latter individuals may be slightly hybridized with westslope cutthroat trout but, evidence of this was not detected because of sampling error. Robb Leary John Powell 355 Table 1: Diagnostic PESTE markers for westslope cutthroat, Yellowstone cutthroat, and rainbow trout. X indicates the fragment is present in the particular taxon. Markers Yellowstone Westslope Rainbow Hpal 5'/Hpal 3' 232 X 153 110.5 72 X 70 69 X 66 Fokl 5'/Tcl 369 366 X 230 159 X 138 X 110 Hpal 5733.6+2 395 388 X 266 248 X 148 X 356 25 20 (0 t 15 o 0) I 10 Blackfoot Telemetry Hybrid Index ■ Observed n Expected 11 2 3 4 5 Hybrid Index Figure 1. Number of diagnostic loci at which fish collected from the Blackfoot River and implanted with radio telemeters possessed PINE fragments characteristic of westslope cutthroat trout (hybrid index). Observed is the observed number of fish and expected is the expected number offish if the westslope cutthroat trout PINE fragments were randomly distributed among the individuals in the sample. 357 Montana Conservation Genetics Laboratory Division of Biological Sciences * University of Montana * Missoula, MT 59812 (406)243-5503/6749 Fax (406)243-4184 April 21, 2007 Ladd Knotek Genetics Contact, Region 2 Montana Fish, Wildlife, & Parks 3201 Spurgin Road Missoula, Montana 59801 Ladd: In order to determine if there is evidence of hybridization, we used a combination of insertion/deletion events (indel loci) and microsatellite loci to analyze DNA extracted from fin clips taken from trout sampled from the following populations: Suimnaiy of results. Sample # Water Name/Location/Collection Date/ Individuals Collector 3351 Flat Creek 17N24W12 47.247 114.842 6/21/2002 Ladd Knotek 3352 Falls Creek 16N08W35 7/31/2002 Laura Bums 3353 Ringeye Creek 16N09W13 7/31/2002 Laura Bums 3354 Randolph Creek 24 20N31W30 2 47.449 115.584 7/13/2006 Ladd Knotek 3355 Upper Copper Lake 15N09W09 6/27/2006 Ron Pierce 3356 Rainy Creek 23 2 a be d e N # Markers Taxa ID Power (%) % 26 R13Y8 WCT 25 R13Y8 WCTXYCT 25 R13Y8 WCT 26 R13Y8 WCT WCTXRBT 25 R13Y8 WCT 25 R13Y8 WCT 19N32W12 R99Y98 99.2X0.8 R99Y98 R99Y98 WCT X RBT 358 8/16/2002 Ladd Knotek 3357 East Fork Twin Creek 19N29W18 8/22/2002 Ladd Knotek 26 R13Y8 WCT R99Y98 Sample # Water Name/Location/Collectlon Date/ Individuals Collector 3380 14 East Fork Big Creek 18N30W09 a be d e N # Markers Taxa ID Power (%) % 16 R6Y4 WCT WCTXRBT 3375 12 8/15/2002 Ladd Knotek Middle Fork Big Creek 18N30W13 16 R6Y4 WCT WCT X RBT 3359 3360 3361 3362 3363 3364 3365 1.8 3366 3367 8/15/2002 Ladd Knotek Dominion Creek 19N31W19 7/13/2006 Ladd Knotek East Fork North Fork Blackfoot River 16N09W07 47.164 112.795 7/1 1/2006 Ron Pieree Sourdough Creek 16N09W17 47.147 112.756 7/12/2006 Ron Pieree Scotty Creek 16N09W08 47.155 112.757 7/12/2006 Ron Pieree Lost Pony Creek 16N10w6l 47.173 112.796 7/11/2006 Ron Pieree Rye Creek 03N20W25 8/10/2006 Chris Clancy East Fork Clearwater River 19N15W32NE1/4 9/12/2002 Ladd Knotek CleaiTvater River (above Rainy Lake) 19N16W01NW1/4 7/19/2006 Ladd Knotek Bertha Creek 19N16W28NW1/4 7/19/2006 Ladd Knotek 12 R13Y8 WCTXRBT 5 R 1 3Y8 WCT X YCT X RBT 3 R 1 3Y8 WCT X YCT X RBT 5 R 1 3Y8 WCT X YCT X RBT 5 R 1 3Y8 RBT X YCT X WCT 28 R13Y8 WCT? R99Y99 26 R 1 3Y8 WCT X YCT X RBT 25 R13Y8 WCTXRBT 8 R13Y8 WCT? R88Y72 83.9X10.0X6.1 96.7 X 1.5 X 99.4X0.6 359 3368 Rattlesnake Creek 28 R13Y8 WCTXYCTXRBT 46.999 113.84 8/9/2006 Ladd Knotek Sample # Water Name/Locatioii/Collection Date/ Individuals Collector 3369 Morrell Creek (upper) 17N15W01NE1/4 9/11/2002 Ladd Knotek 3370 MorreU Creek (middle) 17N15W24NE1/4 8/8/2006 Ladd Knotek 3381 MoiTell Creek (lower) 17N15W25SW1/4 8/8/2006 Ladd Knotek 3371+3382 Savenac Creek 19N29W10+20N29W26 8/28/2002+7/31/06 Ladd Knotek a be d e N # Markers Taxa ID Power (%) % 3373 11 3374 3383 3384 3385 3376 3377 3378 Monture Creek below falls (lower) 17N12W32 47.18 113.16 8/22/2006 Ron Pierce Monture Creek below falls (upper) 47.236 113.156 8/21/2006 Ron Pierce Monture Creek above falls 18N13W25 47.28 113.20 8/23/06 Ron Pierce East Fork Monture Creek 17N12W08 47.245 113.158 8/21/2006 Ron Pierce Middle Fork Monture Creek 18N12W31 47.277 113.181 8/23/2006 Ron Pierce Windlass Gulch 7/27/2006 Dave Schmetterling Scotchman Gulch 8/7/2006 Dave Schmetterling Sluice Gulch 8/2/2006 Dave Schmetterling 7 R13Y8 WCT? 9 R13Y8 WCTXRBT 9 R13Y8 WCTXRBT 27 R13Y8 WCTXRBT 13 R13Y8 WCT WCT X RET 12 R13Y8 WCT? 25 ROYS WCT 16 R13Y8 WCT? 16 R13Y8 WCT? 24 R13Y8 WCTXRBT 25 R13Y8 WCTXRBT 25 R13Y8 WCT R84Y68 99.6X0.4 R96Y86 R99Y98 R99Y92 R99Y92 R99Y98 360 TSIumber offish successfully analyzed. If combined with a previous sample, the number in parentheses indicates the combined sample size. 'Number of diagnostic loci analyzed for the non-native taxa (R=rainbow trout Oncorhynchus mykiss, W=westslope cutthroat trout O. clarki lewisii, Y= Yellowstone cutthroat trout O. c. bouvieri). "Codes: WCT = westslope cutthroat trout; RBT = rainbow trout; YCT = Yellowstone cutthroat trout . Only one taxon code is listed when the entire sample possessed alleles from that taxon only. It must be noted, however, that we cannot definitively rule out the possibility that some or all of the individuals are hybrids. We may not have detected any non-native alleles at the loci examined because of sampling error (see Power %). Taxa codes separated by "x" indicate hybridization between those taxa. ''Number corresponds to the percent chance we have to detect 1% hybridization given the number of individuals successfully analyzed and the number of diagnostic markers used. For example, with 25 individuals we have better than a 99 % chance to detect as little as 1% hybridization with rainbow trout or a 98% chance to detect as little as 1% hybridization with Yellowstone cutthroat trout in a hybrid swarm (a random mating population in which taxa markers are randomly distributed among individuals such that essentially all of them in the population are of hybrid origin) that once was a westslope cutthroat trout population. Likewise, with 25 individuals we have better than a 99% chance to detect as little as a 1% rainbow trout genetic contribution in a hybrid swarm that once was a Yellowstone cutthroat trout population. Not reported when hybridization is detected. Taxa as in b. Indicates the genetic contribution of the hybridizing taxa in the order listed under c. This number is usually reported only if the sample appears to have come from a hybrid swarm. 'indicates number of individuals with genetic characteristics corresponding to the taxa ID code column when the sample can be analyzed at the individual level. This occurs when marker alleles are not randomly distributed among individuals and hybrids and non-hybrids can be reliably distinguished. Methods and Data Analysis The indel technique (Ostberg and Rodriguez 2004) uses short synthetically made segments of DNA called primers, in pairs, to detect areas of DNA in trout that have undergone insertion or deletion (indel) events. During the polymerase chain reaction (PCR), the primers bind to specific areas of the organismal DNA and many copies of the DNA between the primers are made using dye labeled nucleotides. The indel events have resulted in length differences (alleles) in the region of DNA copied between the primers that characterize different trout taxa. These differences have been found to be useful for analysis of hybridization (e.g. Ostberg et al. 2004; Ostberg and Rodriguez 2006). After PCR, the alleles are separated from each other using capillary electrophoresis and visualized using an applied Biosystems 3130x1 genetic analyzer. The alleles are labeled by the primers used to produce them and the number of nucleotides in the sequence. After electrophoresis, the alleles detected in an individual are determined by comparison to synthetic fragments of DNA of known length and alleles from previously analyzed individuals. Microsatellite loci are segments of DNA in which small nucleotide sequences (usually two to five nucleotides) are consecutively repeated numerous times. By using PCR amplification methods similar to those used for indel loci, specific microsatellite loci can be analyzed for differences in the number of repeat units. These differences result in size differences among alleles which are detected using the procedure used to detect indel alleles. We obtained data from seven indel loci and seven microsatellite loci. At 13 of these loci, westslope cutthroat trout, Oncorhynchus clarki lewisii, and rainbow trout, O. mykiss, rarely, if ever, share alleles in common (Table 1). This situation also pertains to a comparison of westslope and Yellowstone cutthroat trout, O. c. bouvieri, at eight loci and Yellowstone cutthroat and rainbow trout at 14 loci (Table 1). Finally, seven loci usually distinguish all three taxa from each other (Table 1). Loci at which taxa rarely, if ever, share alleles in common are often termed diagnostic or marker loci because the alleles detected at them can be used to help determine if a sample came from a non- hybridized population or a population in which hybridization between two or more taxa has or is 361 occurring. Individuals from a non-hybridized population will possess alleles at all diagnostic loci analyzed characteristic of only that taxon. In contrast, since half the DNA from first generation hybrids (Fi) comes from each of the parental taxa Fi individuals will possess alleles characteristic of both the hybridizing taxa at all diagnostic loci analyzed. In later generation hybrids (post Fi), the amount and particular regions of DNA acquired from the parental taxa will vary among individuals. Thus, the particular alleles detected in post Fi hybrids will be highly variable at the diagnostic loci analyzed within and among individuals. An important aspect of both indel and microsatellite alleles is that they demonstrate a codominant mode of inheritance. That is, all genotypes are readily distinguishable from each other. Thus, at diagnostic loci the genotype of individuals in a sample can directly be determined. From these data, the proportion of alleles from different taxa in the population sampled can be directly estimated at each diagnostic locus analyzed. These values averaged over all diagnostic loci yields an estimate of the proportion of alleles in the population that can be attributed to one or more taxa (proportion of admixture). When evidence of hybridization is detected, the first issue to address is whether or not the sample appears to have come from a hybrid swarm. That is, a random mating population in which the alleles of the hybridizing taxa are randomly distributed among individuals such that essentially all of them are of hybrid origin. A common attribute of hybrid swarms is that allele frequencies at diagnostic loci are usually similar among them because their presence can all be traced to a common origin or origins. Thus, one criterion we used for the assessment of whether or not a sample appeared to have come from a hybrid swarm was whether or not the allele frequencies among diagnostic loci reasonably conformed to homogeneity using contingency table chi-square. In order to determine whether or not alleles at the diagnostic loci were randomly distributed among the fish in a sample showing evidence of hybridization, we calculated a hybrid index for each fish in the sample. The hybrid index for an individual was calculated as follows. At each diagnostic locus, an allele characteristic of the native taxon was given a value of zero and an allele characteristic of the non-native taxon a value of one. Thus, at a single diagnostic locus the hybrid index for an individual could have a value of zero (only native alleles present), one (both native and non-native alleles present), or two (only non-native alleles present). These values summed over all diagnostic loci analyzed yields an individual's hybrid index. Considering westslope cutthroat and rainbow trout, therefore, non-hybridized westslope cutthroat trout would have a hybrid index of zero, non- hybridized rainbow trout a hybrid index of 26, Fi hybrids a hybrid index of 13, and post Fi hybrids could have values ranging from zero to 26. The distribution of hybrid indices among fish in a sample was statistically compared to the expected random binomial distribution based on the proportion of admixture detected estimated from the allele frequencies at the diagnostic loci. If the allele frequencies appeared to be statistically homogeneous among diagnostic loci and the observed distribution of hybrid indices reasonably conformed to the expected random distribution, then the sample was considered to have come from a hybrid swarm. In very old hybrid swarms, allele frequencies at diagnostic loci can randomly diverge from homogeneity over time because of genetic drift. In this case, however, the observed distribution of hybrid indices is still expected to reasonably conform to the expected random distribution. Thus, if the allele frequencies were statistically heterogeneous among the diagnostic loci in a sample, but the observed distribution of hybrid indices reasonably conformed to the expected random distribution the sample was also considered to have come from a hybrid swarm. The strongest evidence that a sample showing evidence of hybridization did not come from a hybrid swarm is failure of the observed distribution of hybrid indices to reasonably conform to the expected 362 random distribution. The most likely reasons for this are that the population has only recently become hybridized or the sample contains individuals from two or more populations with different proportions of admixture. At times, the distribution of genotypes at diagnostic loci and the observed distribution of hybrid indices can provide insight into which of these two factors appears mainly responsible for the non-random distribution of the alleles from the hybridizing taxa among individuals in the population. At other times, the distribution of genotypes at diagnostic loci and the observed distribution of hybrid indices may provide little or no insight into the cause of the non- random distribution of alleles among individuals. The latter situation is expected to be fairly common as the two factors usually responsible for the non-random distribution of alleles are not necessarily mutually exclusive. Regardless of the cause, when alleles at the diagnostic loci do not appear to be randomly distributed among individuals in a sample, estimating the proportion of admixture has little if any biological meaning and, therefore, is generally not estimated and reported. Failure to detect evidence of hybridization in a sample does not necessarily mean the population is non-hybridized because there is always the possibility that we would not detect evidence of hybridization because of sampling error. When no evidence of hybridization was detected in a sample, we assessed the likelihood the population is non-hybridized by determining the chances of not detecting as little as a one percent genetic contribution of a non-native taxon to a hybrid swarm. This is simply 0.99 where N is the number of fish in the sample and X is the number of diagnostic loci analyzed. Results and Discussion: Flat Creek 3351 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Flat Creek (Table 2). With the sample size of 26, we have better than a 99% chance of detecting as little as a one percent rainbow trout and better than a 98% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. The Flat Creek population, therefore, strongly appears to be non-hybridized westslope cutthroat trout. Falls Creek 3352 Alleles characteristic of both westslope and Yellowstone cutthroat trout were detected at two of the eight diagnostic loci between these fishes that were analyzed in the sample from Falls Creek (Table 3). The allele frequencies were statistically homogeneous among the diagnostic loci (X^7= 10.201; P>0.10) and the Yellowstone cutthroat trout alleles appeared to be randomly distributed (X^;=0. 180; P>0.50) among the individuals in the sample. Thus, unlike a previous allozyme analysis (sample #492, collected 7/1/91, N=10) that provided no evidence of hybridization in the Falls Creek population this sample clearly indicates the population to be a hybrid swarm between westslope and Yellowstone cutthroat trout with a predominant (0.992) westslope cutthroat trout genetic contribution. We feel the discrepancy between the two samples more likely reflects sampling error rather than the situation in which the Falls Creek population has only recently become hybridized because with only ten fish in the first sample there is about a 15% chance we would not detect a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm. Ringeye Creek 3353 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Ringeye Creek (Table 2). With a sample size of 25, we have better than a 99% chance of detecting as little as a one percent rainbow trout and a 98% chance of detecting as little as a one percent 363 Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. The Ringeye Creek sample, therefore, strongly appears to have come from a non-hybridized westslope cutthroat trout population. Randolph Creek 3354 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at five of the 13 diagnostic loci between these fishes that were analyzed in the Randolph Creek sample (Table 4). Although the allele frequencies are statistically homogeneous (X^i2=ll.454: P>0. 10) among the diagnostic loci, the rainbow trout alleles do not appear to be randomly distributed (X^5=21.820; P<0.001) among the fish in the sample. Rather, there are significantly more fish with a hybrid index of zero or greater than one than expected by chance (Figure 1). Furthermore, the hybrid indices in the sample divide the fish into discrete categories with individuals having a hybrid index of zero or greater than one. The Randolph Creek sample, therefore, appears to have contained a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout with a predominant westslope cutthroat trout genetic contribution. Upper Copper Lake 3355 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Upper Copper Lake (Table 2). With a sample size of 25 , we have better than a 99% chance of detecting as little as a one percent rainbow trout and a 98% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. The Upper Copper Lake population, therefore, very likely is non-hybridized westslope cutthroat trout. Rainy Creek 3356 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at seven of the 13 diagnostic loci between these fishes that were analyzed in the sample from Rainy Creek (Table 4). Although the allele frequencies are statistically homogeneous (X^;2=6.048; P>0.50) among the diagnostic loci in the sample, the rainbow trout alleles are not randomly distributed (X^2=59.548; P<0.001) among the fish in the sample. In contrast, there are significantly more individuals with a hybrid index of zero or greater than two and significantly fewer with a hybrid index of one than expected by chance (Figure 2). Furthermore, the hybrid indices divide the fish in the sample into two distinct groups one of which appears to be non-hybridized westslope cutthroat trout and the other definitely hybrids between westslope cutthroat and rainbow trout. This sample, therefore, appears to have been a mixture of non-hybridized westslope cutthroat trout and hybrids. At OmylOOl *, we detected a single copy of the 266 allele in the sample which is usually characteristic of Yellowstone cutthroat trout (Table 1). This could indicate a small amount of hybridization with Yellowstone cutthroat trout or it could simply be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of Yellowstone cutthroat trout. Because of the very low frequency oi OmylOOl *266, we cannot reasonably begin to distinguish between these possibilities but, there is a reasonable chance this allele is simply westslope cutthroat trout genetic variation as we have previously detected it in some non-hybridized westslope cutthroat trout populations (Table 1). Regardless of whether or not the presence of OmylOOl *266 represents evidence of hybridization with Yellowstone cutthroat trout, the sample clearly contained hybrids between westslope cutthroat and rainbow trout and Rainy Creek should be considered to contain a mixture of non-hybridized westslope cutthroat trout and hybrids. 364 East Fork Twin Creek 3357 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from the East Fork of Twin Creek (Table 2). With a sample size of 26 , we have better than a 99% chance of detecting as little as a one percent rainbow trout and better than a 98% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. The East Fork of Twin Creek sample, therefore, very likely came from a non-hybridized westslope cutthroat trout population. Big Creek Drainage Samples were collected from the Big Creek drainage from two locations in the East Fork and two locations in the Middle Fork. Thus, the first issue to address is whether there is any evidence of genetic differences between the two samples collected from the same stream. If so, then it would not be appropriate to combine the samples from the same stream into one for further analysis. In the two East Fork Big Creek samples, evidence of genetic variation was detected at all 14 loci analyzed (data not presented). Contingency table chi-square analysis indicates that the allele frequencies are statistically heterogeneous (P<0.05) between the samples at two of these loci. This could indicate that genetic differences exist between the samples or these significant differences could represent chance departures from homogeneity due to the number of comparisons performed. In order to distinguish between these possibilities, we compared the chi-square statistic at the two loci showing significant differences to the modified level of significance proposed by Rice (1989). These differences are not significant at the modified level suggesting they most likely represent chance departures from homogeneity. Thus, there is no compelling evidence of genetic differences between the two East Fork Big Creek samples. The two samples, therefore, were combined into one for subsequent analysis. Evidence of genetic variation was detected at 13 of the 14 loci analyzed in the two Middle Fork Big Creek samples (data not presented). The allele frequencies were statistically heterogeneous at three of these loci. These differences, however, are not significant at the modified level suggesting they most likely represent chance departures from homogeneity rather than true genetic differences between the samples. Thus, the two Middle Fork Big Creek samples were combined into one for further analysis. When the East and Middle Fork Big Creek samples are compared, the allele frequencies are statistically heterogeneous between them at one of the 14 loci analyzed. This difference remains significant at the modified level indicating that genetic differences exist between the fish in the East and Middle Fork of Big Creek. These samples, therefore, were treated separately in subsequent analyses. East Fork Big Creek 3380 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at all 13 of the diagnostic loci between these fishes that were analyzed in the East Fork Big Creek sample (Table 4). Although the allele frequencies are statistically homogeneous (X^;2=7.515; P>0.50) among the diagnostic loci, the rainbow trout alleles are not randomly distributed (X^7=379.188; P<0.001) among the fish in the sample. In contrast, there were significantly more individuals in the sample with a hybrid index of zero or greater than seven and significantly fewer individuals with a hybrid index of one through six than expected by chance (Figure 3). The hybrid indices also divide the individuals into two distinct categories one of which appears to be non-hybridized westslope cutthroat trout and the other hybrids between westslope cutthroat and rainbow trout. The East Fork Big Creek sample, therefore, appears to be a mixture of non-hybridized westslope cutthroat trout and hybrids. 365 Middle Fork Big Creek 3375 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at 1 1 of the 13 diagnostic loci between these fishes that were analyzed in the sample from Middle Fork Big Creek (Table 4). Although the allele frequencies are statistically homogeneous (X^;2=14.864; P>0. 10) among the diagnostic loci, the rainbow trout alleles are not randomly distributed (X^5 = 135.111; P<0.001) among the fish in the sample. Rather, there were significantly more individuals in the sample with a hybrid index of zero or greater than four and significantly fewer individuals with a hybrid index of one through four than expected by chance (Figure 4). The hybrid indices also divide the individuals into two distinct categories one of which appears to be non-hybridized westslope cutthroat trout and the other hybrids between westslope cutthroat and rainbow trout. The Middle Fork Big Creek sample, therefore, appears to be a mixture of non-hybridized westslope cutthroat trout and hybrids. Dominion Creek 3359 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at 1 1 of the 13 diagnostic loci between these fishes that were analyzed in the sample from Dominion Creek (Table 4). The allele frequencies are statistically homogeneous {X^ i2=l 369; P>0.50) among the diagnostic loci but, the rainbow trout alleles are not randomly distributed (Z^^ =46. 8 11; P<0.001) among the fish in the sample. The hybrid indices divide the individuals into two distinct categories one of which is composed of individuals with values of zero or one and the other with values of five or more (Figure 5). This distribution suggests the sample mainly contained individuals from a hybrid swarm between westslope cutthroat and rainbow trout with a predominant westslope cutthroat trout genetic contribution (the former group) and a couple of migrants into Dominion Creek from a hybridized population that has a substantial rainbow trout genetic contribution (the latter group). This population, therefore, should simply be considered to be hybridized between westslope cutthroat and rainbow trout. The above results somewhat contradict those obtained from a previous PINE analysis of fish collected from Dominion Creek (sample #3080; col. 8/16/02; N=13 and 14). The sample collected from above a culvert showed no evidence of hybridization but, the sample collected from below the culvert indicated it came from a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.98) westslope cutthroat trout genetic contribution. The present sample was also collected from above the culvert but, it clearly shows evidence of hybridization. The simplest explanation for the discrepancy between the two samples collected from above the culvert is that evidence of hybridization was not detected in the first sample because of sampling error. With the 14 fish in the sample and six diagnostic PINE loci, we had about a 20% chance of not detecting a one percent rainbow trout genetic contribution to a hybrid swarm with westslope cutthroat trout. East Fork North Fork Blackfoot River 3360 Alleles characteristic of both rainbow and Yellowstone cutthroat trout were detected at ten of the 14 diagnostic loci between these fishes that were analyzed in the sample from the East Fork of the North Fork Blackfoot River (Table 5). The allele frequencies were statistically homogeneous {X^i 3=15396; P>0. 10) among the diagnostic loci but, the Yellowstone cutthroat trout alleles were not randomly distributed (X ^9= 18.940; P<0.05) among the fish in the sample. All of the fish in the sample, however, were definitely of hybrid origin between rainbow and Yellowstone cutthroat trout (Figure 6). At Omml037-1 *, a single copy of the 147 allele was detected in the sample. This allele is usually characteristic of westslope cutthroat trout and its presence, therefore, suggests that at least some of the fish in the East Fork of the North Fork Blackfoot River may have a minor westslope cutthroat trout genetic contribution. Thus, considering all the data the East Fork of the North Fork Blackfoot River should be considered to contain hybridized fish with a predominant rainbow trout genetic contribution, a moderate Yellowstone cutthroat trout genetic contribution, and a minor contribution from westslope 366 cutthroat trout. These results are highly concordant with those obtained from a previous allozyme analysis offish collected from the East Fork of the North Fork Blackfoot River (sample #1203; col. 8/1/96; N=9). The previous results also indicated the population to contain hybrids among rainbow, Yellowstone cutthroat, and westslope cutthroat trout with a predominant rainbow trout genetic contribution. Sourdough Creek 3361 Alleles characteristic of both rainbow and Yellowstone cutthroat trout were detected at 13 of the 14 diagnostic loci between these fishes that were analyzed in the sample from Sourdough Creek (Table 5). The allele frequencies were statistically homogeneous (X^;5= 10.628; P>0.50) among the diagnostic loci but, the Yellowstone cutthroat trout alleles were not randomly distributed (X^;2=346.963; P<0.001) among the fish in the sample. All of the fish in the sample, however, were definitely of hybrid origin between rainbow and Yellowstone cutthroat trout (Figure 7). At Omy0004*, two copies of the 77 allele were detected in the sample. This allele is usually characteristic of westslope cutthroat trout and its presence, therefore, suggests that at least some of the fish in Sourdough Creek may have a minor westslope cutthroat trout genetic contribution. Thus, considering all the data Sourdough Creek should be considered to contain hybridized fish with a substantial rainbow and Yellowstone cutthroat trout genetic contribution and a minor contribution from westslope cutthroat trout. Scotty Creek 3362 Alleles characteristic of both rainbow and Yellowstone cutthroat trout were detected at all 14 diagnostic loci between these fishes that were analyzed in the sample from Scotty Creek (Table 5). Although the allele frequencies were statistically homogeneous (X^;5=10.202; P>0.50) among the diagnostic loci, the Yellowstone cutthroat trout alleles were not randomly distributed (X^;2=346.963; P<0.001) among the fish in the sample. All of the fish in the sample except one, however, were definitely of hybrid origin between rainbow and Yellowstone cutthroat trout (Figure 8). The exception was one individual that may have been a non-hybridized rainbow trout (Figure 8). The conclusion that this individual was a non- hybridized rainbow trout is tentative because the small sample size precludes a reliable assessment of this likelihood. At OkilO*, a single copy of the 145 allele was detected in the sample. A single copy of the 77 allele was also detected at Omy0004*. These alleles are usually characteristic of westslope cutthroat trout and their presence indicates that at least some of the fish in Scotty Creek may have a minor westslope cutthroat trout genetic contribution. Thus, considering all the data Scotty Creek should be considered to contain hybridized fish with a substantial rainbow and Yellowstone cutthroat trout genetic contribution and a minor contribution from westslope cutthroat trout. Lost Pony Creek 3363 Alleles characteristic of both rainbow and Yellowstone cutthroat trout were detected at eight of the 14 diagnostic loci between these fishes that were analyzed in the sample from Lost Pony Creek (Table 5). The allele frequencies were statistically heterogeneous (X^;5=36.252; P<0.001) among the diagnostic loci but, the Yellowstone cutthroat trout alleles appeared to be randomly distributed (X^7=3.228; P>0.50 ) among the fish in the sample. At Ssa408* and Omml037-1 *, alleles characteristic of westslope cutthroat trout were detected in the 367 sample (Table 4). Although the allele frequencies were statistically heterogeneous (X^;2=77.509; P<0.001) among the diagnostic loci, the westslope cutthroat trout alleles appeared to be randomly distributed (X^3=2.486; P>0. 10) among the fish in the sample. Considering all the data, therefore. Lost Pony Creek appears to contain a hybrid swarm among rainbow, Yellowstone cutthroat, and westslope cutthroat trout with a predominant (0.839) rainbow trout genetic contribution. Rye Creek 3364 With the exception of Ssa408*, alleles characteristic of only westslope cutthroat trout were detected in the Rye Creek sample. At Ssa408*, a single copy of the 282 allele was detected. This allele is usually characteristic of rainbow trout. Its presence, therefore, could indicate a small amount of hybridization with rainbow trout or it could simply be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of rainbow trout. The low frequency of the allele precludes us from reliably distinguishing between these possibilities but, the failure of two previous allozyme analyses (sample #130; col. 9/12/84; N=26 and sample #898; col. 4/27/94; N=10) to detect evidence of hybridization lends some support to the latter interpretation. The conclusion that Ssa408*282 simply represents westslope cutthroat trout genetic variation, however, is tentative. Thus, we conclude the status of the Rye Creek population is presently uncertain. With this uncertainty we suggest the conservative approach is to consider the Rye Creek population to be non-hybridized westslope cutthroat trout unless further data indicate otherwise. East Fork Clearwater River 3365 Samples were collected from two locations in the lower East Fork Clearwater River in 2002 and two locations from the upper river in 2006. The first issue to address, therefore, is whether there is any evidence of genetic differences between the two 2002 and the two 2006 samples. Contingency table chi- square analysis indicated the allele frequencies at the loci showing evidence of genetic variation (data not presented) were statistically homogeneous (P>0.05) between the two 2002 and two 2006 samples. Thus, there was no evidence of genetic differences between the samples collected within a year so they were combined for subsequent analysis. The next issue to address is whether there is any evidence of genetic differences between the fish collected from the lower and upper East Fork Clearwater River. At the 11 loci showing evidence of genetic variation between the two samples (data not presented), the allele frequencies were statistically heterogeneous at one locus. This difference, however, is not significant at the modified level suggesting it most likely represents a chance departure from homogeneity rather than the existence of genetic differences between the samples. Thus, there is no compelling evidence of genetic differences between the samples so they were combined into a single East Fork Clearwater River sample for subsequent analysis. Alleles characteristic of both westslope cutthroat and rainbow trout were detected at five of the thirteen diagnostic loci between these fishes that were analyzed in the East Fork Clearwater River sample (Table 4). The allele frequencies were statistically heterogeneous (X^;2=26.866; P<0.01) among the diagnostic loci but, the rainbow trout alleles appeared to be randomly distributed (X^5=2.830; P>0. 10) among the fish in the sample. Alleles characteristic of both westslope and Yellowstone cutthroat trout were detected at three of the eight diagnostic loci between these fishes that were analyzed in the sample (Table 3). The allele frequencies were statistically homogeneous (X^7=12.892; P>0.05) among the diagnostic loci in the sample and the Yellowstone cutthroat trout alleles appeared to be randomly distributed (X^2=0-998; P>0. 10) among the fish in the sample. Considering all the data, therefore, the East Fork Clearwater River appears to contain a hybrid swarm among westslope cutthroat, rainbow, and Yellowstone cutthroat trout with a predominant 368 (0.967) westslope cutthroat trout genetic contribution. This sample also contained five bull trout. Clearwater River (above Rainy Lake) 3366 Samples were collected fi^om three areas in the Clearwater River above Rainy Lake. The allele frequencies were statistically homogeneous among the samples at all six loci showing evidence of genetic variation (data not presented). Thus, there was no evidence of genetic differences among the samples and they were combined into one for further analysis. Alleles characteristic of both westslope cutthroat and rainbow trout were detected at three of the 13 diagnostic loci between these fishes that were analyzed in the sample (Table 4). The allele frequencies were statistically homogeneous (X^]2=15A79: P>0. 10) among the diagnostic loci and the rainbow trout alleles appeared to be randomly distributed (X^;=0.397; P>0.50) among the fish in the sample. The Clearwater River above Rainy Lake, therefore, contains a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.994) westslope cutthroat trout genetic contribution. At Omml037-1 *, a single copy of the 127 allele was detected in the sample. This allele is usually characteristic of Yellowstone cutthroat trout. Its presence, therefore, could indicate a small amount of hybridization with Yellowstone cutthroat trout or it could simply be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of Yellowstone cutthroat trout. Because of the low frequency oiOmmI037-l *127, we cannot reasonably distinguish between these possibihties. Regardless of whether or not the presence of Omml037-1 *127 indicates hybridization with Yellowstone cutthroat trout, the population is clearly a hybrid swarm between westslope cutthroat and rainbow trout and should simply be considered to be hybridized. Bertha Creek 3367 Alleles characteristic of only westslope cutthroat trout were detected in the sample from Bertha Creek (Table 2). With the sample size of eight, we have only an 88% chance of detecting as little as a one percent rainbow trout and a 72% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. Although no evidence of hybridization was detected, we, therefore, cannot reasonably exclude the possibility that the Bertha Creek population may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both and that evidence of this was not detected because of sampling error. With this uncertainty, we suggest the conservative approach is to consider the Bertha Creek population to be non-hybridized westslope cutthroat trout unless further data indicate otherwise. Rattlesnake Creek 3368 A single copy of an allele characteristic of Yellowstone cutthroat trout was detected at two loci in the sample from Rattlesnake Creek (Table 3). Although the allele frequencies are statistically homogeneous (X^7=6.027; P>0.50) among the diagnostic loci, the Yellowstone cutthroat trout alleles were not randomly distributed (X^;=12.556; P<0.001) among the individuals in the sample. In contrast, they were detected in only one fish providing clear evidence of hybridization with Yellowstone cutthroat trout but, also indicating the sample did not come from a hybrid swarm between westslope and Yellowstone cutthroat trout. 369 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at nine of the 13 diagnostic loci between these fishes that were analyzed in the sample (Table 4). The allele frequencies were statistically homogeneous (X^;2=7.571; P>0.50) among the diagnostic loci but, the rainbow trout alleles do not appear to be randomly distributed (X^2=l 1-3 17; P<0.01) among the fish in the sample. Rather, there are significantly more individuals with a hybrid index of zero or greater than two and significantly fewer individuals with a hybrid index of one than expected by chance (Figure 9). This sample, therefore, does not appear to have come from a hybrid swarm between westslope cutthroat and rainbow trout and may contain some non-hybridized westslope cutthroat trout. Since the hybrid indices do not clearly separate the fish definitely of hybrid origin and those possibly being non-hybridized into distinct groups, from a management perspective Rattlesnake Creek should simply be considered to contain hybrids among westslope cutthroat, Yellowstone cutthroat, and rainbow trout. The above results are quite consistent with those obtained from previous allozyme (sample #150; col. 10/4/85; N=32 and sample # 188; col. 10/3/86; N=30) and PINE analyses (sample # 2271; col. 7/31/02; N=24 and sample #'s 3090 N=10, 3091 N=16 and 3092 N=16 all col. 5/1/04) offish collected from Rattlesnake Creek. All the samples provided clear evidence of hybridization with rainbow trout and the first allozyme analysis also clearly provided evidence of hybridization with Yellowstone cutthroat trout. Morrell Creek drainage Samples were collected from three areas of Morrell Creek. The allele frequencies were statistically heterogeneous (P<0.001; data not presented) among the samples at one of the seven loci at which evidence of genetic variation was detected and, this difference remains significant at the modified level. Since genetic differences exist among the samples, they were treated separately in subsequent analyses. Morrell Creek (upper) 3369 Alleles characteristic of only westslope cutthroat trout were detected in the sample from upper Morrell Creek (Table 2). With the sample size of seven, we have only an 84% chance of detecting as little as a one percent rainbow trout and a 68% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. Although no evidence of hybridization was detected, we, therefore, cannot reasonably exclude the possibility that the upper Morrell Creek population may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both and that evidence of this was not detected because of sampling error. With this uncertainty, we suggest the conservative approach is to consider the upper Morrell Creek population to be non-hybridized westslope cutthroat trout unless further data indicate otherwise. Morrell Creek (middle) 3370 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at three of the 13 diagnostic loci between these fishes that were analyzed in the sample from middle Morrell Creek (Table 4). The allele frequencies were statistically homogeneous (X^i2=l0.\63: P>0.50) among the diagnostic loci but, the rainbow trout alleles do not appear to be randomly distributed (X^2 =0.25.382; P<0.001) among the fish in the sample. In contrast, they were detected in only one fish (Figure 10). Furthermore, the distribution of hybrid indices clearly divides the fish in the sample into a group that appears to be non- hybridized westslope cutthroat trout and the one individual definitely of hybrid origin. At the time of sampling, therefore, middle Morrell Creek appears to have contained a mixture of non-hybridized westslope cutthroat trout and a relatively small percentage of hybrids between westslope cutthroat and rainbow trout. Morrell Creek (lower) 3381 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at only one of the 13 370 diagnostic loci between these fishes that were analyzed in the sample (Table 4). This could indicate a small amount of hybridization with rainbow trout or it could simply be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of rainbow trout. In this situation we tend to favor the former interpretation because fish definitely of hybrid origin between westslope cutthroat and rainbow trout have been detected further up the drainage. Assuming Occ38*150 represents hybridization with rainbow trout, the allele frequencies were statistically homogeneous {X^ i2=\\. 591; P>0.10) among the diagnostic loci and the rainbow trout alleles appeared to be randomly distributed (X^;=0.073; P>0.50) among the fish in the sample. Lower Morrell Creek, therefore, appears to contain a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.996) westslope cutthroat trout genetic contribution. Savenac Creek 3371 and 3382 Samples were collected from Savenac Creek from two areas in 2002 and two different areas in 2006. The allele frequencies were statistically heterogeneous (P<0.05) between the two 2002 samples at two of the six loci showing evidence of genetic variation (data not presented). These differences, however, are not significant at the modified level suggesting they more likely represent chance departures from homogeneity due to the number of comparisons performed rather than the existence of genetic differences between the samples. These samples, therefore, were combined into one for further analysis. The allele frequencies were statistically homogeneous ( P>0.05; data not presented) between the two 2006 samples at all seven loci showing evidence of genetic variation. These two samples, therefore, were also combined into one for subsequent analysis. Finally, the allele frequencies were statistically homogeneous (P>0.05; data not presented) between the 2002 and 2006 samples at all seven loci showing evidence of genetic variation. Thus, all the samples from Savenac Creek were combined into one for further analysis. Alleles characteristic of both westslope cutthroat and rainbow trout were detected at two of the 13 diagnostic loci between these fishes that were analyzed in the Savenac Creek sample (Table 4). The allele frequencies were statistically homogeneous (X^;2=16.887; P>0.10) among the diagnostic loci but, the rainbow trout alleles do not appear to be randomly distributed (X^;=6.661; P<0.01; Figure 1 1) among the fish in the sample. At the time of sampling, therefore, Savenac Creek appears to have contained a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. The hybrid indices, however, do not separate what may be non-hybridized westslope cutthroat trout and hybrids into discrete categories. Thus, reliably identifying non-hybridized westslope cutthroat trout in Savenac Creek on an individual basis will be problematic and from a management perspective the creek should simply be considered to contain hybrids between westslope cutthroat and rainbow trout. Monture Creek drainage Samples were collected from five areas in the Monture Creek drainage: two areas below the falls, one above the falls, one in East Fork Monture Creek, and one in Middle Fork Monture Creek. The allele frequencies were statistically heterogeneous (P<0.01) among the five samples at three of the 12 loci showing evidence of genetic variation (data not presented). These differences remain significant at the modified level indicating that genetic differences exist among the samples. Most of the above genetic divergence appeared to be due to the sample collected from the lower reaches of Monture Creek below the falls. Thus, we excluded this sample and re-analyzed the data using only the remaining four samples. The allele frequencies were statistically heterogeneous (P<0.001) among these four samples at two of the four loci showing evidence of genetic variation. These differences remain 371 significant at the modified level indicating that genetic differences exist among these samples. All five samples, therefore, were treated separately in subsequent analyses. Monture Creek below falls (lower) 3373 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at eight of the 13 diagnostic loci between these fishes that were analyzed in the sample collected from the lower reach of Monture Creek below the falls (Table 4). The allele frequencies were statistically homogeneous (X^]2=9.356: P>0.50) among the diagnostic loci but, the rainbow trout alleles do not appear to be randomly distributed (X^4=219.797; P<0.001) among the fish in the sample. In contrast, eleven fish in the sample appear to be non-hybridized westslope cutthroat trout and two were definitely hybrids between westslope cutthroat and rainbow trout (Figure 12). Since the hybrid indices clearly separate the non- hybridized and hybridized fish into discrete classes, this portion of Monture Creek should be considered to contain a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. The above results are highly concordant with those obtained from a previous PINE analysis offish collected from lower Monture Creek (sample #2891; col. 9/15/03; N=27). This sample contained 23 fish appearing to be non-hybridized westslope cutthroat trout and four hybrids between westslope cutthroat and rainbow trout. Monture Creek below falls (upper) 3374 Alleles characteristic of only westslope cutthroat trout were detected in the sample from the upper reach of Monture Creek below the falls (Table 2). With the sample size of 12, we have a 96% chance of detecting as little as a one percent rainbow trout but, only an 86% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. We, therefore, cannot reasonably exclude the possibility that the fish in Monture Creek in the upper reach below the falls may be slightly hybridized with Yellowstone cutthroat trout and that evidence of this was not detected because of sampling error. With this uncertainty, we suggest the conservative approach is to consider the fish in the upper reach of Monture Creek below the falls to be non-hybridized westslope cutthroat trout unless further data indicate otherwise. Monture Creek above falls 3383 Alleles characteristic of only westslope cutthroat trout were detected in the sample from Monture Creek collected above the falls (Table 2). With the sample size of 25, we have a 99% chance of detecting as little as a one percent rainbow trout and a 98% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. Monture Creek above the falls, therefore, very likely contains a non-hybridized westslope cutthroat trout population. East Fork Monture Creek 3384 Alleles characteristic of only westslope cutthroat trout were detected in the sample from the East Fork Monture Creek (Table 2). With the sample size of 16, we have a 99% chance of detecting as little as a one percent rainbow trout but, only a 92% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. We, therefore, cannot reasonably exclude the possibility that the fish in East Fork Monture Creek may be slightly hybridized with Yellowstone cutthroat trout and that evidence of this was not detected because of 372 sampling error. With this uncertainty, we suggest the conservative approach is to consider the fish in East Fork Monture Creek to be non-hybridized westslope cutthroat trout unless fiirther data indicate otherwise. Middle Fork Monture Creek 3385 Alleles characteristic of only westslope cutthroat trout were detected in the sample fi^om the Middle Fork Monture Creek (Table 2). With the sample size of 16, we have a 99% chance of detecting as little as a one percent rainbow trout but, only a 92% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. Thus, we cannot reasonably exclude the possibility that the fish in Middle Fork Monture Creek may be slightly hybridized with Yellowstone cutthroat trout but, evidence of this was not detected because of sampling error. With this uncertainty, we suggest the conservative approach is to consider the fish in Middle Fork Monture Creek to be non-hybridized westslope cutthroat trout unless further data indicate otherwise. Windlass Gulch 3376 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at all 13 diagnostic loci between these fishes that were analyzed in the sample collected from Windlass Gulch (Table 4). Although the allele frequencies were statistically homogeneous (X^;2=4.682; P>0.50) among the diagnostic loci, the rainbow trout alleles do not appear to be randomly distributed (J^5= 152.294; P<0.001) among the fish in the sample. Rather, the sample appears to have been a mixture of non- hybridized westslope cutthroat trout, possibly one non-hybridized rainbow trout, and hybrids between westslope cutthroat and rainbow trout (Figure 13). Because the value of the hybrid indices is highly variable among the fish definitely of hybrid origin and some fish have relatively low values, this makes reliably identifying the non-hybridized westslope cutthroat trout on an individual basis problematic. From a management perspective, therefore. Windlass Gulch should simply be considered to contain hybrids between westslope cutthroat and rainbow trout. At OmylOOl *, a single copy of the 212 allele was detected in the sample. This allele is usually characteristic of Yellowstone cutthroat trout. Its presence, therefore, could indicate a small amount of hybridization with Yellowstone cutthroat trout or it could simply be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of Yellowstone cutthroat trout. Because of the low frequency oi OmylOOl *212, we cannot reasonably distinguish between these possibilities. Regardless of whether or not the presence of OmylOOl *212 indicates hybridization with Yellowstone cutthroat trout, the population clearly contains hybrids between westslope cutthroat and rainbow trout and should simply be considered to be hybridized. Scotchman Gulch 3377 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at 1 1 of the 13 diagnostic loci between these fishes that were analyzed in the sample collected from Scotchman Gulch (Table 4). Although the allele frequencies were statistically homogeneous (X^i2=l 1-736; P>0.10) among the diagnostic loci, the rainbow trout alleles do not appear to be randomly distributed (X^4= 146.963; P<0.001) among the fish in the sample. In contrast, the sample appears to have been a mixture of non- hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout (Figure 14). Because the hybrid indices do not divide the non-hybridized westslope cutthroat trout and hybrids into discrete categories, we cannot at the individual level reliably identify non-hybridized westslope cutthroat trout. From a management perspective, therefore, Scotchman Gulch should simply be 373 considered to contain hybrids between westslope cutthroat and rainbow trout. Sluice Gulch 3378 Alleles characteristic of only westslope cutthroat trout were detected in the sample collected from Sluice Gulch (Table 2). With the sample size of 25, we have a 99% chance of detecting as little as a one percent rainbow trout and a 98% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm with westslope cutthroat trout. Sluice Gulch, therefore, very likely contains a non-hybridized westslope cutthroat trout population. Robb Leary John Powell Literature Cited Ostberg, C. O., and R. J. Rodriguez. 2004. Bi-parentally inherited species-specific markers identify hybridization between rainbow trout and cutthroat trout subspecies. Molecular Ecology 4:26-29. Ostberg, C. O., and R. J. Rodriguez. 2006. Hybridization and cytonuclear associations among native westslope cutthroat trout, introduced rainbow trout, and their hybrids within the Stehekin River drainage. North Cascades National Park. Tranactions of the American Fisheries Society 135:924-942. Ostberg, C. O., S. L. Slatton, and R. J. Rodriquez. 2004. Spatial partitioning and asymmetric hybridization among sympatric coastal steelhead {Oncorhynchus mykiss irideus), coastal cutthroat trout {O. clarki clarki) and interspecific hybrids. Molecular Ecology 13:2773-2788. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. 374 Table 1 Alleles at the diagnostic indel and microsatellite loci that usually differentiate among westslope cutthroat, Yellowstone cutthroat, and rainbow trout. Alleles in bold are occassionally shared between or among taxa. Taxa and alleles Locus Westslope Yellowstone Rainbow Indels Occ34 225 225 215 Occ35 230 230 200 Occ36 325 325 275 324 275 285 Occ37 270 270 260 Occ38 175 175 150 Occ42 190 190 160 160 Om55 220 180 199 221 200 Microsatellites Ssa408 183 199 170 195 174 226 178 282 182 183 186 190 194 198 202 206 210 214 218 222 226 230 234 238 246 250 254 262 282 375 Table 1 -continued Taxa and alleles Locus Westslope Yellowstone Rainbow Microsatellites Oki10 93 141 99 99 149 101 97 153 103 101 157 105 103 161 109 105 165 113 109 169 117 113 173 121 117 125 121 129 125 133 129 137 133 153 137 141 145 149 153 Omm1037-1 127 127 159 131 163 135 167 139 171 143 175 147 179 151 183 155 187 191 195 199 203 Omm1037-2 104 106 100 106 102 376 Table 1 -continued Taxa and alleles Locus Westslope Yellowstone Rainbow Microsatellites Omm1050 226 235 238 227 240 230 244 231 246 234 250 235 254 236 256 258 260 262 266 269 270 271 272 274 276 278 280 281 282 284 285 286 289 291 292 293 296 300 302 304 306 308 310 312 322 324 325 326 328 330 335 338 340 365 377 Table 1 -continued Taxa and alleles Locus Westslope Yellowstone Rainbow Microsatellites Omy0004 77 173 99 183 178 101 181 103 183 105 189 109 191 113 193 117 195 121 197 125 199 129 239 131 241 133 245 135 137 139 141 145 149 151 153 157 159 OmylOOl 228 212 159 232 216 174 236 220 176 240 224 178 242 242 182 244 258 184 248 262 186 252 266 190 254 270 192 256 274 194 258 278 196 260 282 198 262 286 200 264 290 202 266 294 204 268 298 206 270 306 208 272 310 210 276 318 214 280 218 284 222 226 230 378 Table 2 Allele frequencies at the loci showing evidence of genetic variation in samples from what appear to be non-hybridized westslope cutthroat trout collected from Bertha Creek, East Fork Monture Creek, East Fork Twin Creek, Flat Creek, Middle Fork Monture Creek, Monture Creek above the falls, Monture Creek below the falls, upper Morrell Creek, Ringeye Creek, Rye Creek, Sluice Gulch, and Upper Copper Lake. Sample and allele free luencies E.F. M.F. Monture Monture Locus Alleles Bertha Monture E. F. Twii 1 Flat Monture (above) (below) Occ36* 324 0.188 0.423 0.192 325 0.812 1.000 0.577 0.808 1.000 1.000 1.000 Om55* 220 221 1.000 1.000 0.904 0.096 0.673 0.327 1.000 1.000 1.000 Ssa408* 195 282 1.000 1.000 1.000 1.000 1.000 1.000 1.000 OkiW* 101 0.250 105 0.343 0.300 0.125 0.281 0.220 0.208 109 0.438 0.479 0.156 0.120 0.083 113 0.313 0.375 0.140 0.104 0.469 0.360 0.458 117 0.063 0.040 0.031 0.120 0.042 121 125 0.180 0.042 129 0.094 0.240 0.250 0.031 0.140 0.125 133 0.125 0.100 0.031 0.040 0.083 137 141 Omm1037-r 139 147 0.438 0.281 0.440 0.846 0.688 0.740 0.458 151 0.562 0.719 0.560 0.154 0.312 0.260 0.542 0mm 1050* 226 230 0.094 0.020 0.060 234 1.000 1.000 0.840 0.096 0.906 0.920 1.000 236 0.160 0.904 379 Table 2-continued Sample and allele frequencies E.F. M. F. Monture (above) Monture (below) Locus Alleles Bertha Monture E. F. Twin Flat Monture Omy1001* 228 0.188 0.160 232 0.260 0.077 0.020 236 0.438 0.420 0.403 0.094 0.042 240 0.125 0.406 0.058 0.188 0.260 0.250 244 0.140 0.080 248 0.020 0.019 252 0.038 0.031 0.208 256 0.125 0.385 0.042 260 0.063 0.094 0.094 0.140 0.042 264 0.019 0.094 0.100 0.083 268 0.063 0.281 0.406 0.220 0.333 272 0.188 0.063 0.160 276 0.031 0.031 0.020 280 380 Table 2-continued Locus Alleles Sample and allele frequencies Morrell Ringeye Rye Sluice Upper Copper Occ36* 324 325 Om55* 220 221 Ssa408* 195 282 Oki10* 101 105 109 113 117 121 125 129 133 137 141 Omm1037-1* 139 147 151 Omm1050* 226 230 234 236 0.143 0.054 0.857 1.000 0.946 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.982 0.018 1.000 1.000 0.500 0.188 0.220 0.208 0.148 0.083 0.071 0.021 0.074 0.720 0.188 0.357 0.167 0.259 0.479 0.071 0.021 0.292 0.204 0.111 0.060 0.104 0.021 0.042 0.130 0.125 0.063 0.074 0.500 0.375 0.021 0.500 0.020 0.458 0.500 0.604 0.500 0.980 0.542 0.042 0.040 0.292 1.000 0.958 0.946 0.054 0.860 0.100 0.708 381 Table 2-continued Sample and allele frequrncies Locus Alleles Morrell Ringeye Upper Rye Sluice Copper OmylOOr 228 232 236 240 244 248 252 256 260 264 268 272 276 280 0.019 0.071 0.271 0.093 0.180 0.063 0.214 0.146 0.185 0.080 0.071 0.375 0.019 0.333 0.143 0.296 0.160 0.021 0.037 0.093 0.300 0.220 0.286 0.056 0.071 0.042 0.146 0.093 0.040 0.020 0.021 0.458 0.143 0.056 0.037 0.019 0.021 0.104 382 Table 3 Allele frequencies at the diagnostic loci between westslope and Yellowstone cutthroat trout in samples showing evidence of hybridization between these fishes collected from the East Fork Clearwater River, Falls Creek, and Rattlesnake Creek. Alleles in bold are characteristic of Yellowstone cutthroat trout. Means are reported only if the sample appears to have come from a hybrid swarm. Allele frequencies and means may not sum to one in some samples because they also contained a rainbow trout genetic contribution (see Table 4). Locus Alleles Sample and allele frequencies E. F. Clearwater Falls Rattlesnake Om55* 220 0.738 221 0.167 Ssa408* 195 0.929 199 0.071 Oki10* 97 0.024 101 0.024 105 0.119 109 0.357 113 0.357 117 0.048 125 129 0.071 157 Omm1037-r 139 147 0.429 151 0.571 0mm 1037-2* 104 0.952 106 0.024 Omm1050* 226 230 0.143 234 0.786 235 236 0.071 my 0004* 77 0.976 181 0.024 191 1.000 1.000 0.020 0.360 0.120 0.020 0.120 0.340 0.020 0.380 0.020 0.600 1.000 0.100 0.860 0.040 1.000 0.964 0.018 0.946 0.018 0.250 0.196 0.375 0.143 0.036 0.321 0.642 1.000 0.018 0.036 0.892 0.018 0.964 0.018 383 Table 3-continued Alleles Sample and allele frequencies Locus E. F. Clearwater Falls Rattlesnake OmylOOr 228 0.310 232 0.071 0.200 236 0.024 0.240 0.036 240 0.167 0.340 0.250 244 0.161 248 0.048 0.040 252 0.024 256 0.167 0.143 260 0.020 0.054 264 0.071 0.160 0.089 268 0.119 0.161 272 0.018 276 0.071 Mean Westslope 0.967 0.992 Mean Yellowstone 0.015 0.008 384 Table 4 Allele frequencies at the diagnostic loci between westslope cutthroat and rainbow trout in samples showing evidence of hybridization between these fishes collected from the Clearwater River (Clear.), Dominion Creek, East Fork Big Creek, East Fork Clearwater River (E. F. Clear.), East Fork North Fork Blackfoot River (E. F. Black), Lost Pony Creek, Middle Fork Big Creek, Monture Creek below the falls, lower Morrell Creek, middle Morrell Creek, Rainy Creek, Randolph Creek, Rattlesnake Creek (Rattle.), Savenac Creek, Scothman Gulch (Scotch.), Scotty Creek, Sourdough Creek (Sour.), and Windlass Gulch. Alleles in bold are characteristic of rainbow trout. Means are provided only if the sample appears to have come from a hybrid swarm. Allele frequencies and means in some samples may not sum to one because they also contained a genetic contribution from Yellowstone cutthroat trout (see Tables 3 and 5). Locus Alleles Sample and allele frequencies Clear. Dominion E. F. Big E. F. Clear E. F. Black. Lost Pony M. F. Big OccSr Occ35* Occ36* Occ37* Occ38* Occ42* Om55* 215 0.041 0.156 0.024 0.900 0.900 0.094 225 1.000 0.959 0.844 0.976 0.906 200 0.083 0.064 0.900 0.900 230 1.000 0.917 0.936 1.000 1.000 275 0.020 0.041 0.094 0.024 1.000 1.000 0.125 324 0.120 0.041 0.064 0.095 0.063 325 0.860 0.918 0.844 0.881 0.813 260 1.000 0.083 0.064 1.000 1.000 0.031 270 0.917 0.936 1.000 0.969 150 0.020 0.041 0.064 0.071 0.900 1.000 0.063 175 0.980 0.959 0.936 0.929 0.937 160 0.064 0.800 1.000 0.063 190 1.000 1.000 0.936 1.000 0.937 199 0.064 0.700 0.400 200 0.083 0.095 0.100 0.600 220 0.860 0.542 0.656 0.738 0.781 221 0.140 0.375 0.281 0.167 0.219 385 Table 4-continued Sample and allele freq uencies E. F. E. F. E. F. Lost M. F. Locus Alleles Clear. Dominion Big Clear Black. Pony Big Ssa408* 178 182 186 194 0.031 0.100 0.400 0.200 0.400 0.100 195 1.000 0.959 0.875 0.929 0.100 0.969 198 0.064 210 0.100 214 0.200 218 0.041 246 0.031 254 0.031 Omm1037-1* 127 135 0.020 0.083 139 0.420 0.542 0.438 0.429 0.600 0.219 143 0.031 0.100 147 0.020 0.031 0.100 151 0.540 0.292 0.375 0.571 0.719 155 0.063 0.031 159 0.083 0.200 0.100 0.031 171 0.100 183 0.063 0.100 187 0.100 195 0.200 0.200 0mm 1037-2* 100 0.041 0.156 0.024 0.900 0.800 0.094 104 1.000 0.959 0.844 0.952 0.906 386 Table 4-continued Sample and allele freq uencies E. F. E. F. E. F. Lost M. F. Locus Alleles Clear. Dominion Big Clear Black. Pony Big Omm1050* 230 0.200 0.143 234 0.800 0.833 0.750 0.786 0.781 236 0.041 0.125 0.071 0.063 238 0.041 0.031 260 0.062 270 0.100 272 0.100 0.100 0.031 276 0.100 278 0.100 280 0.041 0.063 0.100 296 0.200 0.200 300 0.041 0.031 0.200 308 0.100 0.100 335 0.031 0.031 Omy0004* 77 131 137 139 141 151 159 1.000 1.000 0.844 0.063 0.094 0.976 0.200 0.600 0.100 0.100 0.300 0.400 0.100 0.969 0.031 387 Table 4-continued Sample and allele freq uencies E. F. E. F. E. F. Lost M. F. Locus Alleles Clear. Dominion Big Clear Black. Pony Big OmylOOr 172 178 186 190 194 200 202 214 218 222 0.040 0.041 0.031 0.031 0.143 0.125 0.125 0.250 0.200 0.100 0.100 0.300 0.033 0.033 228 0.360 0.031 0.167 0.100 232 0.040 0.292 0.406 0.071 0.033 236 0.060 0.583 0.188 0.024 0.133 240 0.100 0.083 0.094 0.167 0.533 244 0.040 0.063 0.033 248 0.048 252 0.040 0.024 0.033 256 0.040 0.063 0.167 258 0.031 260 0.020 262 0.033 264 0.120 0.071 0.033 268 0.060 0.063 0.119 270 0.020 272 0.020 Mean Westslope 0.994 0.967 0.061 Mean Rainbow 0.006 0.018 0.839 388 Table 4-continued Locus Alleles Monture Sample and allele frequencies lower Morrell middle Morrell Rainy Randolph Rattle. Savenac OccSr Occ35* Occ36* Occ37* Occ38* Occ42* Om55* Ssa408* 215 0.039 225 0.961 1.000 1.000 1.000 1.000 1.000 1.000 200 0.020 0.019 0.018 230 1.000 1.000 1.000 0.980 0.989 0.982 1.000 275 0.055 324 0.423 0.111 0.222 0.160 0.058 0.179 0.093 325 0.577 0.889 0.722 0.840 0.942 0.821 0.907 260 0.039 0.055 0.018 0.018 270 0.961 1.000 0.945 1.000 1.000 0.982 0.982 150 0.077 0.055 0.055 0.020 0.019 175 0.923 0.945 0.945 0.980 0.989 1.000 160 0.039 0.020 0.018 190 0.961 1.000 1.000 0.980 1.000 0.982 1.000 199 0.020 200 0.039 0.058 0.018 220 0.961 1.000 1.000 0.700 0.558 0.964 0.574 221 0.280 0.404 0.018 0.426 182 0.018 195 1.000 1.000 1.000 1.000 1.000 0.946 1.000 389 Table 4-continued Locus Alleles Sample and allele frequencies lower middle Monture Morrell Morrell Rainy Randolph Rattle. Savenac Omm1037-1* 139 0.231 143 147 151 0.538 155 0.192 159 0.039 179 183 0.500 0.500 0.556 0.700 0.780 0.020 0.020 0.055 0.389 0.260 0.200 0.020 0.321 0.642 0.018 0.018 0.759 0.241 Omml 037-2* Omm1050* 104 1.000 1.000 1.000 1.000 1.000 1.000 226 230 0.055 234 0.923 0.945 236 238 240 0.077 272 300 335 0.019 1.000 0.018 1.000 0.018 0.020 0.058 0.036 0.018 0.760 0.731 0.892 0.889 0.220 0.212 0.018 0.056 0.018 0.018 0.018 390 Table 4-continued Sample and allele frequencies lower middle Locus Alleles Monture Morrell Morrell Rainy Randolph Rattle. Savenac Omy0004* 77 137 141 0.923 0.039 0.039 1.000 1.000 0.980 0.020 0.989 0.019 0.964 0.018 1.000 OmylOOr 178 182 228 0.020 0.100 0.096 0.018 0.204 232 0.039 0.360 0.269 0.204 236 0.333 0.556 0.320 0.365 0.036 0.370 240 0.111 0.060 0.096 0.250 0.093 244 0.077 0.111 0.161 248 0.077 0.055 0.020 0.038 0.037 252 0.039 0.038 0.018 254 0.080 256 0.192 0.167 0.055 0.143 258 0.056 260 0.077 0.055 0.111 0.020 0.019 0.054 262 0.018 264 0.231 0.019 0.089 266 0.020 268 0.154 0.287 0.111 0.161 272 0.115 0.055 0.058 0.018 276 0.071 Mean Westslope 0.996 Mean Rainbow 0.004 391 Table 4-continued Sample and allele frequencies Locus Alleles Scotch. Scotty Sour. Windlass OccSr Occ35* Occ36* Occ37* Occ38* Occ42* Om55* Ssa408* 215 0.020 225 0.980 200 0.020 230 0.980 275 0.040 285 324 325 0.960 260 0.040 270 0.960 150 0.040 175 0.960 160 0.020 190 0.980 199 200 0.020 220 0.980 221 170 182 186 190 195 1.000 198 210 214 222 0.800 0.800 0.700 0.100 0.700 0.800 0.800 0.200 0.600 0.100 0.100 0.100 0.200 1.000 0.500 0.833 0.667 0.500 0.667 0.333 0.167 0.354 0.646 0.354 0.646 0.229 0.083 0.688 0.292 0.708 0.313 0.688 0.292 0.708 0.229 0.104 0.604 0.063 0.063 0.104 0.042 0.708 0.063 0.021 0.167 392 Table 4-continued Sample and allele frequencies Locus Alleles Scotch. Scotty Sour. Windlass Omm1037-1* 139 0.260 151 0.660 159 0.040 163 0.020 167 0.020 171 183 199 Omm1 037-2* 100 104 1.000 Omm1050* 230 0.120 234 0.820 236 238 240 256 273 0.020 280 284 285 0.020 286 296 304 308 0.020 312 325 341 348 0.400 0.100 0.700 0.200 0.100 0.100 0.100 0.500 0.167 0.833 0.167 0.167 0.167 0.341 0.432 0.068 0.091 0.045 0.023 0.227 0.773 0.045 0.477 0.159 0.023 0.045 0.023 0.023 0.023 0.023 0.068 0.023 0.045 0.023 393 Table 4-continued Sample and allele frequencies Locus Alleles Scotch. Scotty Sour. Windlass Omy0004* OmylOOr 77 0.980 131 135 137 139 145 153 0.020 157 163 176 0.020 182 186 190 192 198 200 0.020 202 0.020 212 214 218 226 232 236 0.100 240 0.060 244 0.420 248 0.200 256 0.080 260 0.080 264 0.100 0.100 0.100 0.300 0.100 0.333 0.333 0.300 0.100 0.100 0.167 0.167 0.167 0.750 0.023 0.023 0.023 0.045 0.068 0.068 0.023 0.045 0.045 0.091 0.023 0.023 0.045 0.136 0.068 0.068 0.182 0.114 0.136 Mean Westslope Mean Rainbow 394 Table 5 Allele frequencies at the diagnostic loci between rainbow and Yellowstone cutthroat trout in samples showing evidence of hybridization between these fishes collected from East Fork North Fork Blackfoot River (E. F. Black.), Lost Pony Creek, Scotty Creek, and Sourdough Creek (Sour.) Alleles in bold are characteristic of Yellowstone cutthroat trout. Mean rainbow and Yellowstone cutthroat trout allele frequencies are reported only when a sample appears to have come from a hybrid swarm. In some samples, means and allele frequencies may not sum to one because they also contained a westslope cutthroat trout genetic contribution (see Table 4). Locus Alleles E. F. Black. Sample and allele frequencies Lost Pony Scotty Sour. Occ3r Occ35* Occ36* Occ37* Occ38* Occ42* Om55* Ssa408* 215 0.900 225 0.100 200 0.901 230 275 1.000 285 325 260 1.000 270 150 0.900 175 0.100 160 0.800 190 0.200 180 0.200 199 0.700 200 0.100 170 182 0.100 186 0.400 190 194 198 199 0.300 210 214 0.200 222 0.900 0.100 0.900 0.100 1.000 1.000 1.000 1.000 0.400 0.600 0.200 0.400 0.100 0.100 0.100 0.800 0.200 0.800 0.200 0.700 0.100 0.200 0.700 0.300 0.800 0.200 0.800 0.200 0.200 0.200 0.600 0.100 0.100 0.100 0.200 0.500 1.000 0.500 0.500 0.833 0.167 0.667 0.333 0.500 0.500 0.667 0.333 0.667 0.333 0.167 0.667 0.167 395 Table 5- continued Sample and allele frequencies Locus Alleles E. F. Black. Lost Pony Scotty Sour. Oki10* Omm1037-r Omm1 037-2* 0mm 1050* Omy0004* 99 0.200 101 0.100 109 0.600 117 125 137 153 157 161 0.100 127 0.200 159 0.200 171 0.100 183 0.100 187 0.100 195 0.200 100 0.900 106 0.100 235 0.300 240 256 270 0.100 272 0.100 276 278 0.100 280 0.100 296 0.200 300 308 0.100 312 131 0.200 135 137 0.600 141 0.100 153 159 0.100 189 191 195 197 239 0.100 0.100 0.300 0.100 0.100 0.100 0.100 0.200 0.800 0.200 0.300 0.100 0.100 0.200 0.200 0.100 0.300 0.400 0.100 0.200 0.300 0.100 0.100 0.300 0.100 0.500 0.400 0.100 0.700 0.300 0.500 0.200 0.100 0.100 0.100 0.100 0.100 0.300 0.100 0.100 0.100 0.100 0.333 0.167 0.500 0.333 0.500 0.167 0.833 0.167 0.500 0.167 0.167 0.167 0.333 0.167 0.167 396 Table 5- continued Locus Alleles E. F. Black. Sample and allele frequencies Lost Pony Scotty Sour. Omy1001* 172 178 0.125 182 190 192 198 200 0.125 202 0.250 214 270 0.250 274 290 0.125 294 298 0.125 318 Mean Rainbow Mean Yellowstone 0.200 0.100 0.100 0.300 0.200 0.100 0.839 0.100 0.300 0.100 0.100 0.200 0.100 0.100 0.100 0.167 0.167 0.167 0.167 0.167 0.167 397 Randolph Creek 30 n ■ Observed 25 n Expected 1 20 1 Number of 5 1-1 n L I U 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 O q. ^ ^fe ^^ r^ r^ r^tx ^ Hybrid Index Figure 1.- Observed and expected random distribution of hybrid index values between westslope cutthroat and rainbow trout in a sample showing evidence of hybridization between these fishes collected from Randolph Creek. Note the observed distribution does not statistically conform (P<0.05) with the expected random distribution. 398 Rainy Creek 25 20 t 15 o I 10 ■ Observed D Expected n^ Ix h^ ^^ r^ r^ qtx r^ r^ Hybrid Index Figure 6.- Observed and expected random distribution of hybrid index values between rainbow and Yellowstone cutthroat trout in a sample showing evidence of hybridization between these fishes collected from East Fork North Fork Blackfoot River. Note the observed distribution does not statistically conform (P<0.05) with the expected random distribution. A hybrid index of zero is characteristic of rainbow trout and a hybrid index of 28 is characteristic of Yellowstone cutthroat trout. 403 1.2 1 I 0-8 o .- 0.6 0) n I 0.4 0.2 Sourdough Creek I ■ Observed D Expected O n. tx - 0.6 0) n I 0.4 0.2 Scotty Creek ■ Observed D Expected Hybrid Index Figure 8.- Observed and expected random distribution of hybrid index values between rainbow and Yellowstone cutthroat trout in a sample showing evidence of hybridization between these fishes collected from Scotty Creek. Note the observed distribution does not statistically conform (P<0.05) with the expected random distribution. A hybrid index of zero is characteristic of rainbow trout and a hybrid index of 28 is characteristic of Yellowstone cutthroat trout. 405 Rattlesnake Creek 30 25 ^ tn ?n L M- o ^ 1h ^fe ^^ rf:) r^ qtx ^ Hybrid Index Figure 14.- Observed and expected random distribution of hybrid index values between westslope cutthroat and rainbow trout in a sample showing evidence of hybridization between these fishes collected from Scotchman Gulch . Note the observed distribution does not statistically conform (P<0.05) with the expected random distribution. 411 Montana Conservation Genetics Laboratory Division of Biological Sciences * University of Montana * Missoula, MT 59812 (406)243-5503/6749 Fax (406)243-4184 March 7, 2008 Ladd Knotek Montana Fish, Wildlife, & Parks 3201 Spurgin Road Missoula, Montana 59801 Ladd: In order to determine if there is evidence of hybridization, we used a combination of insertion/deletion events (indel loci) and microsatellite loci to analyze DNA extracted from fin clips taken from trout sampled from the following populations: SummaiT of results. Sample # Water Name/Location/Collectlon Date/ Individuals Collector # Markers Taxa ID Power (%) % 3499 Devils Creek 17N25W14 8/1/2001 Ladd Knotek R13Y8 WCTXRBT W96.2XR3.8 3500 Enstache Creek 17N25W12 8/1/2001 Ladd Knotek 10 R13Y8 WCTXRBT W93.4XR6.6 3501 Ninemile Creek 17N24W18 8/1/2001 Ladd Knotek R13Y8 WCTXRBT 3502 Bertha Creek 19N16W28 7/24/2007 Ladd Knotek 18(26) R13Y8 WCTXRBT W99.1XR0.9 3503 Blind Canyon Creek 17N14W16 47.208 113.42 6/26/2007 Ladd Knotek 24 R13Y8 WCTXRBT W99.7XR0.3 3504 Boles Creek 16N16W31NE1/4 26 R13Y8 WCTXRBT W98.4XR1.6 412 7/6/2006 Ladd Knotek f Sample # Individuals Water Name/Location/C Collector 3505 Broadus Creek 47.258 112.83 7/12/2007 Ron Pierce 3506 Butler Creek 16N22W20 9/3/2002 Ladd Knotek 3507 Camp Creek 18N15W29 47.333 113.53 6/19/2007 Ladd Knotek 3508 Canyon Creek 47.220 112.97 7/14/2007 Ron Pierce 3509 Colt Creek 47.323 113.60 7/31/2007 Ladd Knotek 3510 Cooney Creek 47.258*112.81 7/12/2007 Ron Pierce 3511 Deborta Creek 47.269 112.81 7/13/2007 Ron Pierce 3512 FindeU Creek 17N05W32 7/24/2007 Ladd Knotek 3513 Lower Holloman Creek 11N18W17 8/8/2007 Ladd Knotek 3514 Upper Holloman Creek 11N18W16 8/8/2007 Ladd Knotek 3515 7 1 Lower Inez Creek 17N15W32 6/20/2006 Ladd Knotek N # markers Taxa ID R13Y8 WCTXRBT d e Power (%) % WCT W6.7XR93.3 30 R12Y8 WCT? 12 R13Y8 WCTXRBT R99Y98 25 25 R13Y8 WCT R99Y98 R13Y8 WCTXRBT W99.4XR0.6 R13Y8 WCT? R23Y15 R 1 3Y8 WCT X YCT X RET 22 R12Y8 R13Y8 WCT? WCT? R99Y97 R90Y76 10 R13Y8 WCTXRBT R13Y8 WCT?? WCT X RBI 413 Sample # Water Name/Location/Collectlon Date/ Individuals Collector a be d e N # Markers Taxa ID Power (%) % 3516 Middle Inez Creek 18N15W19 47.307 113.55 6/20/2006 Ladd Knotek 5 R13Y8 WCTXRBT W98.5XR1 .5 3517 Upper Inez Creek 13 R13Y8 WCT? R97Y88 18N15W08 47.333 113.53 6/20/2006 Ladd Knotek 3518 Montana Creek 12N25W12 8/9/2007 Ladd Knotek 8 R13Y8 WCT? R88Y72 3519 Lower Murphy Creek W97.4XY1 .7XR0.3 17N15W05 7/25/2007 Ladd Knotek 11 R13Y8 WCTXYCTXRBT 3520 Upper Murphy Creek 17N15W05 7/25/2007 Ladd Knotek 11 R13Y8 WCT? R94Y83 3521 North Fork Blackfoot River (above Deboi-ta) 47.268 7/13/2007 Ron Pierce 3522 North Fork Blackfoot River (below South) 47.197 112.88 7/12/2007 Ron Pierce R 1 3Y8 WCT X YCT X RBT R13Y8 WCTXYCTXRBT 3523 North Fork Blackfoot River (below Theodore) 47.248 112.84 7/12/2007 Ron Pierce R 1 3Y8 WCT X YCT X RBT 3524 West Fork Packer Creek 19N31W03 47.441 115.52 8/17/2006 Ladd Knotek R13Y8 WCTXRBT W98.9XR1.1 3525 Lower Packer Creek 19N31W11 47.425 115.52 8/17/2006 Ladd Knotek R13Y8 WCTXRBT 414 Sample # Water Name/Locatton/Collection Date/ Individuals Collector # Markers Taxa ID Power (%) % 3526 Upper Packer Creek 20N31W26 47.448 115.50 8/17/2006 Ladd Knotek 14 R13Y8 WCT? R97Y89 3527 Plant Creek 11N18W05 8/8/2007 Ladd Knotek 20 R13Y8 WCTXRBT W97.7XR2.3 3528 Lower Richmond Creek 18N16W12 47.327 113.58 6/13/2007 Ladd Knotek 12 R 1 3Y8 WCT X YCT X RBT? W99.4XY0.5XR0.3 3529 Upper Richmond Creek 18N15W07 47.338 113.55 6/13/2007 Ladd Knotek 11 R13Y8 WCT? R94Y83 3530 Lower Trail Creek 17N15W31 47.166 113.44 7/31/2007 Ladd Knotek 10 R13Y8 WCTXRBT W98.1XR1.9 3531 Upper Trail Creek 17N14W19 47.244 113.42 7/31/2007 Ladd Knotek 19 R13Y8 WCT R99Y95 3532 Irish Creek 12N25W21 8/10/2007 Ladd Knotek R13Y8 WCT? R84Y68 3533 Cache Creek 12N25W33 8/10/2007 Ladd Knotek 16 R13Y8 WCTXRBT W99.8XR0.2 3534 Miller Creek 11N18W18 8/8/2007 Ladd Knotek 10 R13Y8 WCT? R93Y80 3536 East Fork MorreU Creek 47.301 113.47 7/26/2007 Ladd Knotek 21 R13Y8 WCT R99Y97 Sample # Individuals Water Name/Location/Collection Date/ # Markers Taxa ID Power (%) % Collector 415 3537 West Fork Clearwater River 18N16W19 7/19/2007 Ladd Knotek 3538 16 2 Vaughn Creek 47.097 113.51 6/21/2006 Ladd Knotek 3539 7 1 White Creek 12N24W30NE1/4 8/9/2007 Ladd Knotek 3540 7 1 South Fork White Creek 12N24W30NW1/4 8/9/2007 Ladd Knotek 3541 8 1 North Fork Fish Creek 14N26W36 7/27/2004 Ladd Knotek 3542 East Fork Indian Creek 13N26W35 7/30/2004 Ladd Knotek 3543 West Fork Indian Creek 13N26W36 7/30/2004 Ladd Knotek 3544 Marshall Creek 47.272 113.68 9/17/2007 Ladd Knotek 3545 Lower Mill Creek 11N21W03 8/4/2003 Ladd Knotek 3546 Upper Mill Creek 11N21W09 8/4/2003 Ladd Knotek 3547 Lower Mormon Creek 11N20W17 7/11/2003 Ladd Knotek 17 18 R13Y8 WCT R99Y94 15 15 14 R13Y8 WCT WCT X RBT R13Y8 WCT WCT X RBT R13Y8 WCT WCT X RBT R13Y8 WCT WCT X RBT R13Y8 R13Y8 R13Y8 R13Y8 R13Y8 WCT? WCT? R13Y8 YCTXRBT WCT? WCT? WCT? R79Y62 R73Y55 Y0.2XR99.8 R79Y62 R98Y91 R97Y89 Sample # Water Name/Location/Collection Date/ # Markers Taxa ID Power (%) % 416 Individuals Collector 3548 Upper Moi-mon Creek 11N21W13 7/1 1/2003 Ladd Knotek R13Y8 WCT? R84Y68 3549 9 Middle Thompson Creek 13N25W14 8/29/2007 Ladd Knotek 10 R13Y8 WCT WCT X RBT TSIumber offish successfully analyzed. If combined with a previous sample, the number in parentheses indicates the combined sample size. 'TSIumber of diagnostic loci analyzed for the non-native taxa (R=rainbow trout Oncorhynchus mykiss, W=westslope cutthroat trout O. clarki lewisii, Y= Yellowstone cutthroat trout O. c. bouvieri). "Codes: WCT = westslope cutthroat trout; RBT = rainbow trout; YCT = Yellowstone cutthroat trout . Only one taxon code is listed when the entire sample possessed alleles from that taxon only. It must be noted, however, that we cannot definitely rule out the possibility that some or all of the individuals are hybrids. We may not have detected any non-native alleles at the loci examined because of sampling error (see Power %). Taxa codes separated by "x" indicate hybridization between those taxa. Number corresponds to the percent chance we have to detect 1% hybridization given the number of individuals successfully analyzed and the number of diagnostic markers used. For example, with 25 individuals we have better than a 99 % chance to detect as little as 1% hybridization with rainbow trout or a 98% chance to detect as little as 1% hybridization with Yellowstone cutthroat trout in a hybrid swarm (a random mating population in which taxa markers are randomly distributed among individuals such that essentially all of them in the population are of hybrid origin) that once was a westslope cutthroat trout population. Likewise, with 25 individuals we have better than a 99% chance to detect as little as a 1% rainbow trout genetic contribution in a hybrid swarm that once was a Yellowstone cutthroat trout population. Not reported when hybridization is detected. Taxa as in b. Indicates the genetic contribution of the hybridizing taxa in the order listed under c. This number is usually reported only if the sample appears to have come from a hybrid swarm. 'indicates number of individuals with genetic characteristics corresponding to the taxa ID code column when the sample can be analyzed at the individual level. This occurs when marker alleles are not randomly distributed among individuals and hybrids and non-hybrids can be reliably distinguished. Methods and Data Analysis The indel technique (Ostberg and Rodriguez 2004) uses short synthetically made segments of DNA called primers, in pairs, to detect areas of DNA in trout that have undergone insertion or deletion (indel) events. During the polymerase chain reaction (PCR), the primers bind to specific areas of the organismal DNA and many copies of the DNA between the primers are made using dye labeled nucleotides. The indel events have resulted in length differences (alleles) in the region of DNA copied between the primers that characterize different trout taxa. These differences have been found to be useful for analysis of hybridization (e.g. Ostberg et al. 2004; Ostberg and Rodriguez 2006). After PCR, the alleles are separated from each other using capillary electrophoresis and visualized using an applied Biosystems 3130x1 genetic analyzer. The alleles are labeled by the primers used to produce them and the number of nucleotides in the sequence. After electrophoresis, the alleles detected in an individual are determined by comparison to synthetic fragments of DNA of known length and alleles from previously analyzed individuals. Microsatellite loci are segments of DNA in which small nucleotide sequences (usually two to five nucleotides) are consecutively repeated numerous times. By using PCR amplification methods similar to those used for indel loci, specific microsatellite loci can be analyzed for differences in the number of repeat units. These differences result in size differences among alleles which are detected using the procedure used to detect indel alleles. We obtained data from seven indel loci and seven microsatellite loci. At 13 of these loci, westslope cutthroat trout, Oncorhynchus clarki lewisii, and rainbow trout, O. mykiss, rarely share alleles in 417 common (Table 1). This situation also pertains to a comparison of westslope and Yellowstone cutthroat trout, O. c. bouvieri, at eight loci and Yellowstone cutthroat and rainbow trout at 14 loci (Table 1). Finally, seven loci usually distinguish all three taxa from each other (Table 1). Loci at which taxa rarely share alleles in common are often termed diagnostic or marker loci because the alleles detected at them can be used to help determine if a sample came from a non-hybridized population or a population in which hybridization between two or more taxa has or is occurring. Individuals from a non-hybridized population will possess alleles at all diagnostic loci analyzed characteristic of only that taxon. In contrast, since half the DNA from first generation hybrids (Fi) comes from each of the parental taxa Fi individuals will possess alleles characteristic of both the hybridizing taxa at all diagnostic loci analyzed. In later generation hybrids (post Fi), the amount and particular regions of DNA acquired from the parental taxa will vary among individuals. Thus, the particular alleles detected in post Fi hybrids will be highly variable at the diagnostic loci analyzed within and among individuals. An important aspect of both indel and microsatellite alleles is that they demonstrate a codominant mode of inheritance. That is, all genotypes are readily distinguishable from each other. Thus, at diagnostic loci the genotype of individuals in a sample can directly be determined. From these data, the proportion of alleles from different taxa in the population sampled can be directly estimated at each diagnostic locus analyzed. These values averaged over all diagnostic loci yields an estimate of the proportion of alleles in the population that can be attributed to one or more taxa (proportion of admixture). When evidence of hybridization is detected, the next issue to address is whether or not the sample appears to have come from a hybrid swarm. That is, a random mating population in which the alleles of the hybridizing taxa are randomly distributed among individuals such that essentially all of them are of hybrid origin. A common attribute of hybrid swarms is that allele frequencies at diagnostic loci are usually similar among them because their presence can all be traced to a common origin or origins. Thus, one criterion we used for the assessment of whether or not a sample appeared to have come from a hybrid swarm was whether or not the allele frequencies among diagnostic loci reasonably conformed to homogeneity using contingency table chi-square. In order to determine whether or not alleles at the diagnostic loci were randomly distributed among the fish in a sample showing evidence of hybridization, we calculated a hybrid index for each fish in the sample. The hybrid index for an individual was calculated as follows. At each diagnostic locus, an allele characteristic of the native taxon was given a value of zero and an allele characteristic of the non-native taxon a value of one. Thus, at a single diagnostic locus the hybrid index for an individual could have a value of zero (only native alleles present), one (both native and non-native alleles present), or two (only non-native alleles present). These values summed over all diagnostic loci analyzed yields an individual's hybrid index. Considering westslope cutthroat and rainbow trout, therefore, non-hybridized westslope cutthroat trout would have a hybrid index of zero, non- hybridized rainbow trout a hybrid index of 26, Fi hybrids a hybrid index of 13, and post Fi hybrids could have values ranging from zero to 26. The distribution of hybrid indices among the fish in a sample was statistically compared to the expected random binomial distribution based on the proportion of admixture detected estimated from the allele frequencies at the diagnostic loci. If the allele frequencies appeared to be statistically homogeneous among the diagnostic loci and the observed distribution of hybrid indices reasonably conformed to the expected random distribution, then the sample was considered to have come from a hybrid swarm. In very old hybrid swarms, allele frequencies at diagnostic loci can randomly diverge from homogeneity over time because of genetic drift. In this case, however, the observed distribution of 418 hybrid indices is still expected to reasonably conform to the expected random distribution. Thus, if the allele frequencies were statistically heterogeneous among the diagnostic loci in a sample, but the observed distribution of hybrid indices reasonably conformed to the expected random distribution the sample was also considered to have come from a hybrid swarm. The strongest evidence that a sample showing evidence of hybridization did not come from a hybrid swarm is failure of the observed distribution of hybrid indices to reasonably conform to the expected random distribution. The most likely reasons for this are that the population has only recently become hybridized or the sample contains individuals from two or more populations with different proportions of admixture. At times, the distribution of genotypes at diagnostic loci and the observed distribution of hybrid indices can provide insight into which of these two factors appears mainly responsible for the nonrandom distribution of the alleles from the hybridizing taxa among individuals in the population. At other times, the distribution of genotypes at diagnostic loci and the observed distribution of hybrid indices may provide little or no insight into the cause of the nonrandom distribution of alleles among individuals. The latter situation is expected to be fairly common as the two factors usually responsible for the nonrandom distribution of alleles are not necessarily mutually exclusive. Regardless of the cause, when alleles at the diagnostic loci do not appear to be randomly distributed among individuals in a sample, estimating the proportion of admixture has little if any biological meaning and, therefore, is generally not estimated and reported. Failure to detect evidence of hybridization in a sample does not necessarily mean the population is non-hybridized because there is always the possibility that we would not detect evidence of hybridization because of sampling error. When no evidence of hybridization was detected in a sample, we assessed the likelihood the population is non-hybridized by determining the chances of not detecting as little as a one percent genetic contribution of a non-native taxon to a hybrid swarm. This is simply 0.99 where N is the number of fish in the sample and X is the number of diagnostic loci analyzed. Results and Discussion: Devils Creek 3499 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at five of the 13 diagnostic loci between these species that were analyzed in the sample from Devils Creek (Table 2). The allele frequencies were statistically homogeneous {X ^2=1 5.232; P> 0.10) among the diagnostic loci and the rainbow trout alleles appeared to be randomly distributed {X ^-=0.886; P>0.50) among the fish in the sample. This sample, therefore, appears to have come from a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.96) westslope cutthroat trout genetic contribution. Eustache Creek 3500 At eight of the 13 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed in the sample from Eustache Creek, alleles characteristic of both fishes were detected (Table 2). Although the allele frequencies were statistically heterogeneous {X i2=25.908; P<0.05) among the diagnostic loci in the sample, the rainbow trout alleles appeared to be randomly distributed among the fish {X 2=5.518; P>0.05). Thus, Eustache Creek appears to contain a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.93) westslope cutthroat trout genetic contribution. Ninemile Creek 3501 419 Because of poor DNA quality, we were not able to obtain the genotype of all individuals at all the loci analyzed in the Ninemile Creek sample. At nine of the 1 3 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, however, alleles characteristic of both fishes were detected in the sample (Table 3). The allele frequencies were statistically homogeneous {X i2=l4.296; P>0.10) among the diagnostic loci. Because of missing genotypes we were not able to obtain hybrid indices for all the fish and statistically determine whether or not the sample appeared to have come from a hybrid swarm. Of the seven fish in the sample, however, five were definitely of hybrid origin suggesting that the majority or all of the fish in this reach of Ninemile Creek were hybrids. Thus, from a management perspective this section of Ninemile Creek should simply be considered to contain hybrids between westslope cutthroat and rainbow trout. Bertha Creek (combined) 3502 This reach of Bertha Creek was previously sampled July 19, 2006 (sample #3367; N=8). Between the two samples, evidence of genetic variation was detected at five loci. The allele frequencies were statistically homogeneous between the samples at all these loci. Thus, there was no evidence of genetic differences between the samples so they were combined into one for further analysis. In the combined sample, alleles characteristic of both westslope cutthroat and rainbow trout were detected at two of the 13 diagnostic loci between these species that were analyzed (Table 2). The allele frequencies were statistically homogeneous {X 72=10.627; P> 0.50) among the diagnostic loci and the rainbow trout alleles were randomly distributed {X ;=0.078; P>0.50) among the fish in the sample. This sample, therefore, appears to have come from a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (>0.99) westslope cutthroat trout genetic contribution. Furthermore, these results suggest the population has been hybridized for many generations suggesting that the failure to detect evidence of hybridization in the first sample was simply due to sampling error. With the first sample size of eight and the estimated proportion of rainbow trout alleles in the population, we had a 1 5% chance of detecting no evidence of hybridization. Blind Canyon Creek 3503 Fish were collected from three reaches of Blind Canyon Creek. Among these samples, evidence of genetic variation was detected at seven loci. The allele frequencies were statistically heterogeneous among the samples only at OmylOOl * {X 20=^ 1 .604; P< 0.05). This could indicate that genetic differences exist among the samples or it could simply be a chance departure from homogeneity due to the number of comparisons performed. In order to distinguish between these possibilities, we compared the chi-square statistic at OmylOOl* to the modified level of significance proposed by Rice (1 989). At the modified level, this difference is not significant. Thus, it most likely represents a chance departure from homogeneity and there is no compelling evidence of genetic differences among the samples. The samples, therefore, were combined into a single Bertha Creek sample for further analysis. At two of the 1 3 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the combined sample (Table 2). The allele frequencies were statistically homogeneous {X ^2=1 1-319; P>0.10) among the diagnostic loci in the sample and the rainbow trout alleles appeared to be randomly distributed among the fish {X j=0. 101; P>0. 50). Thus, Blind Canyon Creek appears to contain a hybrid 420 swarm between westslope cutthroat and rainbow trout with a predominant (>0.99) westslope cutthroat trout genetic contribution. Boles Creek 3504 Fish were sampled from three areas of Boles Creek. Evidence of genetic variation was detected at nine loci among the samples. At all of these loci, the allele frequencies were statistically homogeneous among the samples. Thus, they were combined into one for further analysis. Alleles characteristic of both westslope cutthroat and rainbow trout were detected at six of the 13 diagnostic loci between these species that were analyzed in the combined sample (Table 2). Although the allele frequencies were statistically heterogeneous {X 1 2=37. 904; P<0.001) among the diagnostic loci, the rainbow trout alleles appeared to be randomly distributed {X 2=3.379; P>0. 10) among the fish in the sample. This sample, therefore, appears to have come from a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.98) westslope cutthroat trout genetic contribution. Broadus Creek 3505 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at seven of the 1 3 diagnostic loci between these species that were analyzed in the sample from Broadus Creek (Table 2). The allele frequencies were statistically homogeneous {X 1 2=7. 448; P>0.05) among the diagnostic loci and the rainbow trout alleles appeared to be randomly distributed {X 3=2.879; P>0. 10) among the fish in the sample. This sample, therefore, appears to have come from a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.93) rainbow trout genetic contribution. Butler Creek 3506 Samples were collected from above and below the dam in Butler Creek. Between the two samples, evidence of genetic variation was detected at five loci. The allele frequencies were statistically homogeneous between the samples at all these loci. The two samples, therefore, were combined into a single Butler Creek sample for subsequent analysis. With the exception of Ssa408*, alleles characteristic of only westslope cutthroat trout were detected in the combined sample (Table 5). At Ssa408*, one fish collected from below the dam possessed a single copy of the 282* allele. This allele is usually characteristic of rainbow trout (Table 1). Its presence in the sample, therefore, could indicate a small amount of hybridization with rainbow trout or it could be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of rainbow trout. In this situation, we can not distinguish between these possibilities. With this uncertainty and a lack of conclusive evidence for hybridization, the conservative approach would be to consider Butler Creek as containing a non-hybridized westslope cutthroat trout population unless future data indicate otherwise. Camp Creek 3507 At seven of the 13 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the Camp Creek sample (Table 3). 421 Although the allele frequencies were statistically homogeneous (X 72=10.101; P>0.50) among the diagnostic loci in the sample, the rainbow trout alleles were not randomly distributed among the fish (X 2=9.013; P<0.05). In contrast, significantly more fish had a hybrid index of three and significantly fewer fish a hybrid index of one than expected by chance (Figure 1). This sample, therefore, probably contained a mixture of hybridized and non-hybridized westslope cutthroat trout. The potentially non-hybridized westslope cutthroat trout and hybrids, however, do not fall into distinct groups. Thus, it will not be possible on an individual basis to reliably separate the non-hybridized fish from hybrids. From a management perspective, therefore. Camp Creek should simply be considered to contain hybrids between westslope cutthroat and rainbow trout with a predominant westslope cutthroat trout genetic component. Canyon Creek 3508 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Canyon Creek (Table 5). With the sample size of 25, we have better than a 99% chance of detecting as little as a one percent rainbow trout and better than a 98% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Canyon Creek, therefore, very likely contains a non-hybridized westslope cutthroat trout population. Colt Creek 3509 Samples were collected from three reaches of Colt Creek. Among these samples, evidence of genetic variation was detected at seven loci. The allele frequencies were statistically homogeneous among the samples at all these loci. Thus, the samples were combined into a single Colt Creek sample for subsequent analysis. Alleles characteristic of both westslope cutthroat and rainbow trout were detected at two of the 13 diagnostic loci between these species that were analyzed in the combined sample (Table 2). Although the allele frequencies were statistically heterogeneous (X ]2=28A63; P<0.01) among the diagnostic loci, the rainbow trout alleles were randomly distributed {X ;=0.389; P>0.50) among the fish in the sample. Fish definitely of hybrid origin were detected in both the middle and upper reaches of the creek that were sampled. Colt Creek, therefore, appears to contain a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.99) westslope cutthroat trout genetic contribution. Cooney Creek 3510 A single fish believed to be a rainbow trout was collected from Cooney Creek. In contrast to the expectation, alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in this fish. This fish, therefore, was undoubtedly not a rainbow trout and in fact may possibly have been a non-hybridized westslope cutthroat trout. Deborta Creek 3511 Alleles characteristic of rainbow, Yellowstone cutthroat, and westslope cutthroat trout were detected in the sample collected from Deborta Creek (Table 4). The westslope cutthroat (XV19.027; P<0.01) and Yellowstone cutthroat trout (XV23.563; P<0.01) allele frequencies were statistically heterogeneous among the diagnostic loci and they were not randomly distributed (X^;7=123.008; P<0.001) among the fish in the sample (Figure 2). Although Deborta Creek does not appear to contain a hybrid swarm, all the fish in the sample were definitely of 422 hybrid origin. From a management perspective, therefore, Deborta Creek should simply be considered to contain hybrids among westslope cutthroat, Yellowstone cutthroat, and rainbow trout with a predominant rainbow trout genetic contribution. Findell Creek 3512 Samples were collected from two reaches of Findell Creek. Between the two samples, evidence of genetic variation was detected at six loci. The allele frequencies were statistically heterogeneous between the samples only at Omml050* {X ;=4.400; P<0.05). This difference, however, is not significant at the modified level suggesting it most likely represents a chance departure from homogeneity. Since there was no compelling evidence of genetic differences between the samples, they were combined into a single Findell Creek sample for subsequent analysis. With the exception of Om55*, alleles characteristic of only westslope cutthroat trout were detected in the combined sample (Table 5). At Om55*, one fish collected from the upper reach possessed a single copy of the 799* allele. This allele is usually characteristic of rainbow trout (Table 1). Its presence in the sample, therefore, could indicate a small amount of hybridization with rainbow trout or it could simply be westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of rainbow trout. In this situation, we can not distinguish between these possibilities. With this uncertainty and a lack of conclusive evidence for hybridization, the conservative approach would be to consider Findell Creek as containing a non-hybridized westslope cutthroat trout population unless future data indicate otherwise. Holloman Creek Samples were collected from two reaches of Holloman Creek. Between the two samples, evidence of genetic variation was detected at all 14 loci analyzed. The allele frequencies were statistically heterogeneous between the samples at ten of these loci. These differences remain significant at the modified level indicating that genetic differences exist between the samples. The lower and upper Holloman Creek samples, therefore, were treated separately for further analysis. Lower Holloman Creek 3513 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from lower Holloman Creek (Table 5). With the sample size of nine, however, we have only a 90% chance of detecting as little as a one percent rainbow trout and only a 76% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in lower Holloman Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both fishes but evidence of this was not detected because of sampling error. Since the status of the trout in lower Holloman Creek is uncertain and there was no evidence of hybridization, the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. Upper Holloman Creek 3514 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at all 13 of the diagnostic loci between these species that were analyzed in the sample from upper Holloman 423 Creek (Table 3). Although the allele frequencies were statistically homogeneous {X 72=10.258; P>0.50) among the diagnostic loci, the rainbow trout alleles were not randomly distributed {X 11=923.089; P<0.001) among the fish in the sample. In contrast, significantly more fish had a hybrid index of zero and greater than five than expected by chance (Figure 3). Upper Holloman Creek, therefore, appears to have contained a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout when it was sampled. Since some of the definite hybrids had relatively low hybrid indices, reliably distinguishing the non-hybridized fish on an individual basis from the hybrids will be problematic. Thus, from a management perspective, upper Holloman Creek should simply be considered to contain hybrids between westslope cutthroat and rainbow trout. Inez Creek Samples were collected from four reaches of Inez Creek. Among the four samples, evidence of genetic variation was detected at nine of the loci analyzed. The allele frequencies were statistically heterogeneous among the samples at two of these loci. These differences remain significant at the modified level indicating that genetic differences exist among the samples. Much of this divergence was due to differences between the two lower samples and the two upper ones. When only the data from the two upper samples were analyzed, there was no evidence of genetic differences between them at the six loci showing evidence of genetic variation. These two samples, therefore, were combined into a single upper Inez Creek sample for subsequent analysis. When only the two lower most samples were compared, the allele frequencies were statistically heterogeneous between them at one of the eight loci showing evidence of genetic variation. This difference remains significant at the modified level so these were treated as separate lower and middle Inez Creek samples for further analysis. Lower Inez Creek 3515 At three of the 13 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the lower Inez Creek sample (Table 3). Although the allele frequencies were statistically homogeneous (X /2=1 8.601; P>0.05) among the diagnostic loci in the sample, the rainbow trout alleles were not randomly distributed among the fish {X j=25.172; P<0.001). In contrast, they were detected in only one fish (Figure 4). This sample, therefore, did not come from a hybrid swarm. Since the middle Inez Creek sample appears to have come from a hybrid swarm with a slight rainbow trout genetic contribution (see sample #3516), we can not conclude with any certainty that the fish in the lower sample with hybrid indices of zero are non-hybridized westslope cutthroat trout. With the estimated frequency of rainbow trout alleles being 0.015 in the middle sample, there is a six percent chance that the apparent non-hybridized fish in the lower sample actually may possess this level of hybridization but it was not detected because of sampling error. Regardless of the status of the fish in the lower sample with hybrid indices of zero, the fish definitely of hybrid origin in the sample clearly appears to be a recent migrant. Middle Inez Creek 3516 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at two of the 1 3 diagnostic loci between these species that were analyzed in the sample from middle Inez Creek (Table 2). The allele frequencies were statistically homogeneous {X i2=l0.752; P>0.50) among the diagnostic loci and the rainbow trout alleles appeared to be randomly distributed {X 7=0.472; P>0.10) among the fish in the sample. Middle Inez Creek, therefore, appears to 424 have contained a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.99) westslope cutthroat trout genetic contribution. Upper Inez Creek 3517 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Inez Creek (Table 5). With the sample size of 13, we have about a 97% chance of detecting as little as a one percent and considering the middle sample better than a 99% chance of detecting a 1 .5 % rainbow trout genetic contribution to a hybrid swarm. Thus, this sample very likely did not come from a population hybridized with rainbow trout. In contrast, we have only about an 88% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. We, therefore, can not reasonably exclude the possibility that the fish in upper Inez Creek may be slightly hybridized with Yellowstone cutthroat trout but evidence of this was not detected because of sampling error. Inez Creek compared to previous samples The results from the present Inez Creek samples are somewhat similar to those obtained from samples collected in 2002 and 2006 (sample #3498). The previous samples indicated a slight amount of hybridization with rainbow (0.002) and Yellowstone cutthroat trout (0.007). Thus, the major differences between the two sets of data are the presence of a recent hybrid migrant in the present lower sample, the absence of evidence of hybridization with Yellowstone cutthroat trout in all the most recent samples, and the absence of any evidence of hybridization in the recent upper sample. The lack of evidence of hybridization with Yellowstone cutthroat trout in the recent samples may simply reflect sampling error as with the estimated proportion of Yellowstone cutthroat trout alleles in the previous samples we had a 41% chance of not detecting any in the lower sample, a 57% chance of not detecting any in the middle sample, and a 23% chance of not detecting any in the upper sample. Considering all the data, therefore, with the exception of the very upper reaches possibly containing non-hybridized westslope cutthroat trout the fish in middle Inez Creek and downstream should probably be treated as being slightly hybridized with rainbow and Yellowstone cutthroat trout. Montana Creek 3518 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Montana Creek (Table 5). With the sample size of eight, however, we have only about an 88% chance of detecting as little as a one percent rainbow trout and only about a 72% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in this sample from Montana Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both fishes but evidence of this was not detected because of sampling error. This possibility, however, seems slight as previous samples of trout collected above (sample #3459) and below (sample #3460) the location of the present sample also showed no evidence of hybridization (we could not combine any of the samples as genetic differences clearly existed among all of them). Overall, therefore, Montana Creek should be considered to contain multiple populations of non- hybridized westslope cutthroat trout. 425 Murphy Creek Samples were collected from two areas in Murphy Creek. Among the two samples, evidence of genetic variation was detected at five of the loci analyzed. The allele frequencies were statistically heterogeneous between the samples at three of these loci. These differences remain significant at the modified level indicating that genetic differences exist between the samples. The lower and upper Murphy Creek samples, therefore, were treated separately for further analysis. Lower Murphy Creek 3519 At two of the eight diagnostic loci between westslope and Yellowstone cutthroat trout that were analyzed, alleles characteristic of both fishes were detected in the sample from lower Murphy Creek (Table 4). Likewise, alleles characteristic of both westslope cutthroat and rainbow trout were detected at one of the 13 diagnostic loci between these fishes that were analyzed. The rambow (XV6.203; P>0.10) and Yellowstone cutthroat trout (xVlO.505; P>0.10) allele frequencies were statistically homogeneous among the diagnostic loci and they appeared to be randomly distributed {X 2=2.3 1 8; P>0. 1 0) among the fish in the sample. Thus, lower Murphy Creek appears to contain a hybrid swarm among westslope cutthroat, Yellowstone cutthroat, and rainbow trout with a predominant (0.97) westslope cutthroat trout genetic component. Upper Murphy Creek 3520 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Murphy Creek (Table 5). With the sample size of 11, however, we have only about a 94% chance of detecting as little as a one percent rainbow trout and only about an 83% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in the sample from upper Murphy Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both fishes but evidence of this was not detected because of sampling error. With this uncertainty and a lack of evidence for hybridization, the conservative approach would be to consider upper Murphy Creek to contain non-hybridized westslope cutthroat trout unless additional data indicate otherwise. North Fork Blackfoot River above Deborta 3521 Alleles characteristic of rainbow, Yellowstone cutthroat, and westslope cutthroat trout were detected in the sample collected from the North Fork Blackfoot River above Deborta Creek (Table 4). The westslope cutthroat {X 6=8.805; P>0.10) and Yellowstone cutthroat trout (X 7=10.051; P>0.10) allele frequencies were statistically homogeneous among the diagnostic loci but the alleles did not appear to be randomly distributed {X 5=9.327; P<0.05) among the fish in the sample. Thus, this sample clearly contained hybrids among rainbow, westslope cutthroat, and Yellowstone cutthroat trout with a predominant rainbow trout genetic contribution but it does not appear to have come from a hybrid swarm. North Fork Blackfoot River below South 3522 This sample also contained alleles characteristic of rainbow, Yellowstone cutthroat, and westslope cutthroat trout (Table 4). The westslope cutthroat {X 6=5.897; P>0.10) and Yellowstone cutthroat trout {X 7=6.731; P>0.10) allele frequencies were statistically homogeneous among the diagnostic loci but the alleles did not appear to be randomly distributed 426 (X i=4.564; P<0.05) among the fish in the sample. Thus, this sample also clearly contained hybrids among rainbow, westslope cutthroat, and Yellowstone cutthroat trout with a predominant rainbow trout genetic contribution but it does not appear to have come from a hybrid swarm. North Fork Blackfoot River below Theodore 3523 This sample also contained alleles characteristic of rainbow, Yellowstone cutthroat, and westslope cutthroat trout (Table 4). In this sample, the westslope cutthroat trout (X 6=28.561; P<0.001) alleles were not statistically homogeneous among the diagnostic loci but, the Yellowstone cutthroat trout allele frequencies were (X 7=12.276; P>0.05). Also, the westslope and Yellowstone cutthroat trout alleles did not appear to be randomly distributed (X ^=1 9.020; P<0.01) among the fish in the sample. Thus, this sample also clearly contained hybrids among rainbow, westslope cutthroat, and Yellowstone cutthroat trout with a predominant rainbow trout genetic contribution but it does not appear to have come from a hybrid swarm. North Fork Blackfoot River combined The average rainbow, westslope cutthroat, and Yellowstone cutthroat trout allele frequencies did not statistically differ among the three North Fork Blackfoot River samples. Thus, they were combined into a single sample for further analysis. In the combined sample, the westslope cutthroat trout (X 6=35.289; P<0.001) alleles were not statistically homogeneous among the diagnostic loci but, the Yellowstone cutthroat trout allele frequencies were (X 7=12.740; P>0.05). Given the previous results, not surprisingly the westslope and Yellowstone cutthroat trout alleles did not appear to be randomly distributed (X 6=14.062; P<0.05) among the fish in the combined sample (Figure 5). A likely explanation for the heterogeneity of allele frequencies at some of the diagnostic loci and the nonrandom distribution of the westslope and Yellowstone cutthroat trout alleles among the fish is that when this reach of the North Fork Blackfoot River was sampled it contained fish from two or more hybridized populations of rainbow, westslope cutthroat, and Yellowstone cutthroat trout with different amounts of hybridization. West Fork Packer Creek 3524 With the exception of Omml050*, alleles characteristic of only westslope cutthroat trout were detected in the sample from West Fork Packer Creek. At OmmI050*, two copies of the 300* allele were detected (Table 2). This allele is usually characteristic of rainbow trout (Table 1). Normally we would be uncertain whether its presence in the sample indicated a small amount of hybridization with rainbow trout or simply represented westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of rainbow trout. In this situation, however, we strongly favor the former interpretation because a sample collected from Packer Creek near the confluence with the West Fork Packer Creek (sample #3525) clearly contained hybrids between westslope cutthroat and rainbow trout. Although the allele frequencies were statistically heterogeneous {X ^7=24. 833; P<0.05) among the diagnostic loci, the rainbow trout alleles appeared to be randomly distributed {X i=lA66; P>0. 10) among the fish in the sample. West Fork Packer Creek, therefore, appears to contain a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.99) westslope cutthroat trout genetic contribution. 427 Packer Creek Samples were collected from three reaches of Packer Creek. Among the samples, evidence of genetic variation was detected at 12 of the loci analyzed. Although the allele frequencies were statistically homogeneous among the samples at all of these loci the frequency of alleles characteristic of rainbow trout was statistically higher in the lower sample than in the other two {X 2=24.540; P<0.001). Thus, the lower sample was treated separately and the middle and upper samples were combined for subsequent analysis. Lower Packer Creek 3525 At ten of the 1 3 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the sample from lower Packer Creek (Table 3). Although the frequency of rainbow trout alleles was statistically homogeneous {X 72=8.615; P>0.50) among the diagnostic loci, the rainbow trout alleles did not appear to be randomly distributed {X 5=29.885; P<0.001) among the fish in the sample. Rather, significantly more fish had a hybrid index of zero and greater than three than expected by chance (Figure 6). Thus, lower Packer Creek may contain some non-hybridized westslope cutthroat trout. The hybrid indices, however, do not divide the potentially non-hybridized fish and hybrids into distinct classes. Thus, reliably identifying non-hybridized fish on an individual basis will be problematic and from a management perspective lower Packer Creek should simply be considered to contain hybrids between westslope cutthroat and rainbow trout with a major westslope cutthroat trout genetic component. Upper Packer Creek 3526 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Packer Creek (Table 5). With the sample size of 14, we have about a 97% chance of detecting as little as a one percent rainbow trout but only about an 89% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in upper Packer Creek may be slightly hybridized with Yellowstone cutthroat trout but evidence of this was not detected because of sampling error. Although the status of the trout in upper Packer Creek is somewhat uncertain, with the absence of evidence for hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. Plant Creek 3527 Samples were collected from two reaches of Plant Creek. Between the samples, evidence of genetic variation was detected at eight of the loci analyzed. The allele frequencies were statistically heterogeneous between the samples only at Occ36* {X i=4.328; P<0.05) but, this difference is not significant at the modified level. Since there was no conclusive evidence of genetic differences between the samples, they were combined into one for subsequent analysis. Alleles characteristic of both westslope cutthroat and rainbow trout were detected at six of the 13 diagnostic loci between these species that were analyzed in the sample from Plant Creek (Table 2). The allele frequencies were statistically homogeneous {X ^2= 16.602; P>0. 10) among the diagnostic loci and the rainbow trout alleles appeared to be randomly distributed {X i=3A07; P>0. 10) among the fish in the sample. Plant Creek, therefore, appears to have contained a hybrid 428 swarm between westslope cutthroat and rainbow trout with a predominant (0.98) westslope cutthroat trout genetic contribution. Richmond Creek Samples were collected from two reaches of Richmond Creek. Between the samples, evidence of genetic variation was detected at six of the loci analyzed. The allele frequencies were statistically heterogeneous between the samples only at OmylOOl * {X 7=25.565; P<0.001). This difference is significant at the modified level so the samples were treated separately for further analysis. Lower Richmond Creek 3528 Alleles characteristic of only westslope cutthroat trout were detected in the sample collected from lower Richmond Creek except at Om55* and Occ36* (Table 4). At Occ36*, one individual possessed a single copy of the 285* allele and at Om55* another individual contained a single copy of the 180* allele. The former allele is usually characteristic of rainbow trout and the latter is usually characteristic of Yellowstone cutthroat trout. The presence of these alleles in the sample, therefore, could indicate a small amount of hybridization with rainbow trout, Yellowstone cutthroat trout, or both fishes. Alternatively, these alleles could represent westslope cutthroat trout genetic variation that is indistinguishable from that usually characteristic of rainbow or Yellowstone cutthroat trout. Unfortunately, we can not distinguish between these possibilities in this situation but, we suspect the former interpretation is the most likely. Adhering to the latter interpretation requires postulating that this population contains two different rare variants at diagnostic loci which would be an unusual situation. Thus, tentatively we suggest that lower Richmond Creek be considered to contain a hybrid swarm among westslope cutthroat, Yellowstone cutthroat, and rainbow trout with a major (0.99) westslope cutthroat trout genetic component. Upper Richmond Creek 3529 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Richmond Creek (Table 5). With the sample size of 11, we have only about a 94% chance of detecting as little as a one percent rainbow trout and only about an 83% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in upper Richmond Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in upper Richmond Creek is somewhat uncertain, in the absence of evidence for hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. Trail Creek Samples were collected from four reaches of Trail Creek. Among the four samples, evidence of genetic variation was detected at seven of the loci analyzed. The allele frequencies were statistically heterogeneous among the samples at five of these loci. These differences remain significant at the modified level indicating that genetic differences exist among the samples. 429 Much of this divergence was due to differences between the two lower samples and the two upper ones. When only the data from the two upper samples were analyzed, there was no evidence of genetic differences between them at the six loci showing evidence of genetic variation. These two samples, therefore, were combined into a single upper Trail Creek sample for subsequent analysis. When only the two lower most samples were compared, there was no evidence of genetic differences between them at the seven loci showing evidence of genetic variation. These two samples, therefore, were also combined into a single lower Trail Creek sample for further analysis. Lower Trial Creek 3530 At three of the 13 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the sample from lower Trail Creek (Table 2). Although the frequency of rainbow trout alleles was not statistically homogeneous {X 72=24.350; P<0.05) among the diagnostic loci, the rainbow trout alleles appeared to be randomly distributed {X ;=2.289; P>0.10) among the fish in the sample. Thus, lower Trail Creek appears to contain a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (0.98) westslope cutthroat trout genetic contribution. Upper Trail Creek 3531 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Trail Creek (Table 5). With the sample size of 19, we have better than a 99% chance of detecting as little as a one percent rainbow trout and about a 95% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Upper Trail Creek, therefore, very likely contains a non-hybridized westslope cutthroat trout population. Irish Creek 3532 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Irish Creek (Table 5). With the sample size of seven, however, we have only about an 84% chance of detecting as little as a one percent rainbow trout and only about a 68% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in Irish Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in Irish Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. Cache Creek 3533 Samples were collected from two reaches of Cache Creek. Between the samples, evidence of genetic variation was detected at six of the loci analyzed. The allele frequencies were statistically heterogeneous between the samples only at Omml037-1* {X ;=4.571; P<0.05). This difference, however, is not significant at the modified level so the samples were combined for further analysis. 430 At all of the loci analyzed except Om55*, alleles characteristic of only westslope cutthroat trout were detected in the Cache Creek sample (Table 2). At Om55*, a single copy of the 199* allele was detected. This allele is usually characteristic of rainbow trout. Thus, normally we would be uncertain as to whether or not the presence of this allele represented a small amount of hybridization with rainbow trout. In this case, however, we strongly favor the hybridization interpretation as a previous allozyme (sample #588) and indel analysis (sample #3462) offish collected from Cache Creek lower in the drainage clearly provided evidence of hybridization with rainbow trout. Upper Cache Creek, therefore, appears to contain a hybrid swarm between westslope cutthroat and rainbow trout with a predominant (>0.99) westslope cutthroat trout genetic contribution. Miller Creek above Holloman 3534 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from Miller Creek (Table 5). With the sample size often, however, we have only about a 93% chance of detecting as little as a one percent rainbow trout and only about an 80% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in Miller Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in Miller Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless additional data indicate otherwise. East Fork Morrell Creek 3536 Samples were collected from three reaches of East Fork Morrell Creek. Among the samples, evidence of genetic variation was detected at five loci. The allele frequencies were statistically homogeneous among the samples at all these loci. The three samples, therefore, were combined into one for subsequent analysis. Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from East Fork Morrell Creek (Table 5). With the sample size of 21, we have better than a 99% chance of detecting as little as a one percent rainbow trout and about a 97% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. East Fork Morrell Creek, therefore, very likely contains a non-hybridized westslope cutthroat trout population. The above results are very similar to those obtained from an indel analysis of fish collected in Morrell Creek just below the confluence of the East Fork (sample #3369). These fish also appeared to be non-hybridized westslope cutthroat trout. In contrast to these results, indel analysis offish collected from Morrell Creek further downstream (samples #3370 and #3381) clearly provided evidence of hybridization with rainbow trout. Thus, it appears that only the very upper portion of the Morrell Creek drainage still contains non-hybridized westslope cutthroat trout. West Fork Clearwater River 3537 Samples were collected from two areas of the West Fork Clearwater River. Between the samples, evidence of genetic variation was detected at five loci. The allele frequencies were 431 statistically homogeneous between the samples at all of these loci so they were combined into one for subsequent analysis. Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from the West Fork Clearwater River (Table 5). With the sample size of 17, we have about a 99% chance of detecting as little as a one percent rainbow trout and about a 94% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. The West Fork Clearwater River, therefore, very likely contains a non-hybridized westslope cutthroat trout population. The above results differ from those obtained from previous PINE (sample #2002) and indel (sample #3492) analyses offish collected from the West Fork Clearwater River drainage further downstream. The PESTE results suggested the sample was a mixture of 24 non-hybridized westslope cutthroat trout and one hybrid between westslope cutthroat and rainbow trout. The indel results suggested this sample came from a hybrid swarm among westslope cutthroat, Yellowstone cutthroat, and rainbow trout with a predominant (0.99) westslope cutthroat trout genetic contribution. Considering all the results, therefore, it appears that now exclusively non- hybridized westslope cutthroat trout in the West Fork Clearwater River drainage are mainly, if not solely, confined to the upper portion of the drainage. Vaughn Creek 3538 Samples were collected from two areas of Vaughn Creek. Considering both samples, evidence of genetic variation was detected at eight loci. The allele frequencies at these loci were all statistically homogeneous between the two samples so they were combined into one for further analysis. At six of the 13 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the sample from Vaughn Creek (Table 3). Although the frequency of rainbow trout alleles was statistically homogeneous {X 12=9.834; P>0.50) among the diagnostic loci, the rainbow trout alleles did not appear to be randomly distributed {X 2=36.403; P<0.001) among the fish in the sample. In contrast, they were detected in only two fish with one being in the lower and the other the upper sampling locations (Figure 7). Thus, this sample appeared to contain a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. The hybrid indices clearly separate the fish into these two groups so from a management perspective it would not be inappropriate to consider Vaughn Creek to still contain non-hybridized westslope cutthroat trout and a relatively small proportion of hybrid migrants. White Creek above South Fork 3539 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at five of the 1 3 diagnostic loci between these fishes that were analyzed in the sample from White Creek collected above the South Fork (Table 3). Although the frequency of rainbow trout alleles was Statistically homogeneous {X 1 2=9. 106; P>0.50) among the diagnostic loci, the rainbow trout alleles were not randomly distributed {X 2=10.321; P<0.01) among the fish in the sample. Rather, they were detected in only one fish (Figure 8). Thus, this sample appeared to contain a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. The hybrid indices clearly separate the fish into these two groups so from a management perspective it would not be inappropriate to consider this reach of White Creek to 432 still contain non-hybridized westslope cutthroat trout and a relatively small proportion of hybrid migrants. The above results are different from those obtained from a previous indel analysis of fish collected from White Creek (sample #3461). In this sample of nine fish collected further downstream than the present one, alleles characteristic of only westslope cutthroat trout were detected. Because of the small sample size we were not sure whether or not this indicated an absence of hybridization in the drainage. The present results and those obtained from a sample in the South Fork (sample #3540) clearly indicate that hybrids between westslope cutthroat and rainbow trout exist in the White Creek drainage. The majority of the fish in the drainage, however, still appear to be non-hybridized westslope cutthroat trout. South Fork White Creek 3540 At two of the 1 3 diagnostic loci between westslope cutthroat and rainbow trout that were analyzed, alleles characteristic of both fishes were detected in the sample from South Fork White Creek (Table 3). Although the frequency of rainbow trout alleles was statistically homogeneous {X 12=10.687; P>0.50) among the diagnostic loci, the rainbow trout alleles were not randomly distributed {X ;=4.500; P<0.05) among the fish in the sample. In contrast, they were detected in only one fish (Figure 9). Thus, this sample appeared to contain a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. The hybrid indices tend to separate the fish into these two groups so from a management perspective it would not be inappropriate to consider South Fork White Creek to still contain non-hybridized westslope cutthroat trout and a relatively small proportion of hybrid migrants. North Fork Fish Creek 3541 Alleles characteristic of both westslope cutthroat and rainbow trout were detected at four of the 13 diagnostic loci between these fishes that were analyzed in the sample from North Fork Fish Creek (Table 3). Although the frequency of rainbow trout alleles was statistically homogeneous {X 72=9. 106; P>0.50) among the diagnostic loci, the rainbow trout alleles were not randomly distributed {X 2=12.690; P<0.01) among the fish in the sample. Rather, they were detected in only one fish (Figure 10). Thus, this sample appeared to contain a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. Indel analysis of two previous samples (# 3480 and #3481) collected from North Fork Fish Creek upstream from the present sample detected no evidence of hybridization. Likewise, indel analysis offish collected from Fletcher Gulch (#3482) and Greenwood Creek (#3484), tributaries to North Fork Fish Creek above the present sample, detected alleles characteristic of only westslope cutthroat trout. In contrast to these results, indel analysis of fish collected from French Creek (#3483), another North Fork tributary above the present sample, appeared to be a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. Considering all the data, therefore, it appears that the North Fork Fish Creek drainage mainly contains non-hybridized westslope cutthroat trout and a small proportion of hybrids between westslope cutthroat and rainbow trout. It is also likely that the hybrids may be fairly widely distributed throughout the drainage and failure to detect them in some locations may be a result of sampling error rather than their actual absence. East Fork Indian Creek 3542 433 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from East Fork Indian Creek (Table 5). With the sample size of six, we have only about a 79% chance of detecting as little as a one percent rainbow trout and only about a 62% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in East Fork Indian Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in East Fork Indian Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. West Fork Indian Creek 3543 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from West Fork Indian Creek (Table 5). With the sample size of five, we have only about a 73% chance of detecting as little as a one percent rainbow trout and only about a 55% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in West Fork Indian Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in West Fork Indian Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless future data indicate otherwise. Upper Indian Creek drainage Samples have now been analyzed from the East Fork, Middle Fork (#3497), and West Fork of Indian Creek and Indian Creek below the forks. In all the samples, alleles characteristic of only westslope cutthroat trout were detected. Thus, it appears that the Indian Creek drainage in the West Fork Fish Creek drainage mainly, if not solely, contains non-hybridized westslope cutthroat trout. We can not, however, completely rule out the possibility that this portion of the West Fork Fish Creek drainage may contain a small proportion of hybrids between westslope cutthroat and rainbow trout but they were not detected because of sampling error. This is a possibility as hybrids between these fishes have been detected in other portions of the West Fork Fish Creek drainage (e.g. see samples #3483 and 3541). Upper Marshall Creek 3544 With the exception of Ssa408*, alleles characteristic of only rainbow trout were detected in the sample from upper Marshall Creek. At Ssa408*, a single copy of the 199* allele, which is usually characteristic of Yellowstone cutthroat trout, was detected. Thus, this sample may have come from a hybrid swarm between rainbow and Yellowstone cutthroat trout with a predominant rainbow trout (>0.99) genetic contribution. Alternatively, this could be a non-hybridized rainbow trout population with an unusual Ssa408* variant. Regardless of the case, this is certainly an introduced non-native trout population. Indel analysis of two previous samples from Marshall Creek collected downstream from the present one found no evidence of hybridization in one location (#3490) but hybrids between westslope cutthroat and rainbow trout were definitely present in the lower most sample (#3491). 434 Considering all the data, we now suspect that the absence of evidence for hybridization in the middle sample (#3490) was probably more likely due to sampling error than the actual absence of hybridization. From a management perspective, therefore, Marshall Creek should simply be considered to contain non-native trout. Mill Creek Samples were collected from three reaches of Mill Creek. Among the samples, evidence of genetic variation was detected at four loci. The allele frequencies were statistically heterogeneous among the samples only at Omml037-1 * {X 2=8.452; P<0.05). This difference remains significant at the modified level indicating that genetic differences exist among the samples. Much of this divergence was due to differences between the lower sample and the other two. When only the data from the two upper samples were analyzed, there was no evidence of genetic differences between them at the three loci showing evidence of genetic variation. These two samples, therefore, were combined into a single upper Mill Creek sample for subsequent analysis and the lower sample was treated separately. Lower Mill Creek 3545 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from lower Mill Creek (Table 5). With the sample size of six, we have only about a 79% chance of detecting as little as a one percent rainbow trout and only about a 62% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in lower Mill Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in lower Mill Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. Upper Mill Creek 3546 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Mill Creek (Table 5). With the sample size of 15, we have about a 98% chance of detecting as little as a one percent rainbow trout and about a 91% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, the fish in upper Mill Creek very likely are non-hybridized westslope cutthroat trout. Mormon Creek Samples were collected from three areas of Mormon Creek. Among the samples, evidence of genetic variation was detected at six loci. The allele frequencies were statistically heterogeneous among the samples at two of these loci. These differences remain significant at the modified level indicating that genetic differences exist among the samples. Much of this divergence was due to differences between the upper sample and the other two. When only the data from the two lower samples were analyzed, there was no evidence of genetic differences between them at the six loci showing evidence of genetic variation. These two samples, therefore, were combined into a single lower Mormon Creek sample for subsequent analysis and the upper sample was treated separately. 435 Lower Mormon Creek 3547 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from lower Mormon Creek (Table 5). With the sample size of 14, we have about a 97% chance of detecting as little as a one percent rainbow trout but only about an 89% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the trout in lower Mormon Creek may be slightly hybridized with Yellowstone cutthroat trout but evidence of this was not detected because of sampling error. Since the status of the trout in lower Mormon Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless additional data indicate otherwise. Upper Mormon Creek 3548 Alleles characteristic of only westslope cutthroat trout were detected at all the loci analyzed in the sample from upper Mormon Creek (Table 5). With the sample size of seven, we have only about an 84% chance of detecting as little as a one percent rainbow trout and only about a 68% chance of detecting as little as a one percent Yellowstone cutthroat trout genetic contribution to a hybrid swarm that once was non-hybridized westslope cutthroat trout. Thus, we can not reasonably exclude the possibility that the fish in upper Mormon Creek may be slightly hybridized with rainbow trout, Yellowstone cutthroat trout, or both but evidence of this was not detected because of sampling error. Since the status of the trout in upper Mormon Creek is somewhat uncertain, in the absence of evidence of hybridization the conservative approach would be to consider this reach of the creek to contain non-hybridized westslope cutthroat trout unless further data indicate otherwise. Middle Thompson Creek #3549 This sample was collected from the middle reach of Thompson Creek above a supposed barrier. Although a previous indel analysis of trout collected from this reach (#3488) provided no evidence of hybridization, in the present sample alleles characteristic of both westslope cutthroat and rainbow trout were detected at seven of the 13 diagnostic loci between these fishes that were analyzed (Table 3). Although the frequency of rainbow trout alleles was statistically homogeneous {X ] 2=7. 189; P>0.50) among the diagnostic loci, the rainbow trout alleles did not appear to be randomly distributed {X 2=9.401; P<0.01) among the fish in the sample. Rather, they were detected in only one fish (Figure 11). Thus, the present sample appeared to contain a mixture of non-hybridized westslope cutthroat trout and hybrids between westslope cutthroat and rainbow trout. There are two possible explanations for the difference between the past and present sample. Hybrids were actually present in this reach when it was first sampled in 1999 and they were not detected because of sampling error or hybrids have only fairly recently invaded this portion of the stream. If the latter is the case, then it is possible that the hybrids came from downstream as a previous indel analysis offish collected from below the supposed barrier (#3487) definitely contained hybrids between westslope cutthroat and rainbow trout. If this is the case, then the supposed barrier is not absolute. Robb Leary 436 Steve Amish Literature Cited Ostberg, C. O., and R. J. Rodriguez. 2004. Bi-parentally inherited species-specific markers identify hybridization between rainbow trout and cutthroat trout subspecies. Molecular Ecology 4:26-29. Ostberg, C. O., and R. J. Rodriguez. 2006. Hybridization and cytonuclear associations among native westslope cutthroat trout, introduced rainbow trout, and their hybrids within the Stehekin River drainage. North Cascades National Park. Transactions of the American Fisheries Society 135:924-942. Ostberg, C. O., S. L. Slatton, and R. J. Rodriquez. 2004. Spatial partitioning and asymmetric hybridization among sympatric coastal steelhead {Oncorhynchus mykiss irideus), coastal cutthroat trout {O. clarki clarki) and interspecific hybrids. Molecular Ecology 13:2773-2788. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. 437 Table 1 Alleles at the diagnostic indel and microsatellite loci that usually differentiate among westslope cutthroat, Yellowstone cutthroat, and rainbow trout. Alleles in bold are occassionally shared between or among taxa. Taxa and alleles Locus Westslope Yellowstone Rainbow Indels Occ34 225 225 215 Occ35 230 230 200 Occ36 324 325 275 325 275 285 Occ37 268 270 260 270 Occ38 175 175 150 Occ42 160 190 160 190 Om55 220 180 199 221 200 Microsatellites Ssa408 183 199 170 195 174 226 178 282 182 183 186 190 194 198 202 206 210 214 218 222 226 230 234 238 246 250 254 262 266 282 438 Table 1 -continued Taxa and alleles Locus Westslope Yellowstone Rainbow Microsatellites Oki10 93 141 99 99 149 101 97 153 103 101 157 105 103 161 109 105 165 113 109 169 117 113 173 121 117 125 121 129 125 133 129 137 133 153 137 161 141 145 149 153 157 Omm1037-1 127 127 139 131 143 135 159 139 163 143 167 147 171 151 175 155 179 163 183 187 191 195 199 203 Omm1037-2 104 106 98 106 100 102 439 Table 1 -continued Taxa and alleles Locus Westslope Yellowstone Rainbow Microsatellites Omm1050 226 235 238 227 240 230 244 231 246 234 250 235 254 236 256 238 258 260 262 266 268 269 270 271 272 273 274 276 278 280 281 282 284 285 286 289 291 292 293 296 300 302 304 306 308 310 312 316 322 324 325 326 328 330 331 335 338 340 347 348 365 440 Table 1 -continued Taxa and alleles Locus Westslope Yellowstone Rainbow Microsatellites Omy0004 77 173 99 183 178 101 181 103 183 105 189 109 191 113 193 117 195 121 197 125 199 129 201 131 205 133 225 135 227 137 229 139 231 141 233 142 235 143 239 145 241 149 245 151 249 153 155 157 159 165 OmylOOl 219 212 159 228 216 174 232 220 176 236 224 178 240 242 182 242 258 184 244 262 186 248 266 190 252 270 192 254 274 194 256 278 196 258 282 198 260 286 200 262 290 202 264 294 204 266 298 206 268 306 208 270 310 210 272 318 214 276 218 280 222 284 226 230 441 Table 2 Allele frequencies at the diagnostic loci between westslope cutthroat and rainbow trout in samples from what appear to be hybrid swarms between these fishes collected from Bertha Creek, Blind Canyon Creek, Boles Creek, Broadus Creek, Cache Creek, Colt Creek, Devils Creek, Eustache Creek, middle Inez Creek, Plant Creek, lower Trail Creek, and West Fork Packer Creek. Alleles in bold are characteristic of rainbow trout. Locus Sample and allele frequencies alleles Bertha Blind Boles Broadus Cache Colt 215 1.000 0.020 225 1.000 1.000 1.000 1.000 0.980 200 0.875 230 1.000 1.000 1.000 0.125 1.000 1.000 275 0.019 0.875 285 0.096 0.060 324 0.139 0.146 0.154 0.300 0.020 325 0.861 0.854 0.731 0.125 0.700 0.920 260 0.019 1.000 270 1.000 1.000 0.981 1.000 1.000 150 0.021 0.019 1.000 175 1.000 0.979 0.981 1.000 1.000 160 0.875 190 1.000 1.000 1.000 0.125 1.000 1.000 199 0.500 0.031 200 0.019 0.500 220 1.000 0.917 0.654 0.938 0.960 221 0.083 0.327 0.031 0.040 178 0.125 182 186 195 0.917 1.000 1.000 1.000 1.000 199 0.056 210 0.125 218 0.250 234 0.250 250 0.250 282 0.028 Occ34* Occ35* Occ36* Occ37* Occ38* Occ42* Om55* Ssa408* 442 Table 2-continued Alleles Sample and allele frequencies Locus Bertha Blind Boles Broadus Cache Colt 0mm 1037-1* 139 147 0.250 0.375 0.042 0.442 0.125 0.438 0.380 151 0.750 0.583 0.462 0.563 0.620 155 0.096 159 171 0.625 187 0.250 0mm 1037-2* 100 0.875 104 1.000 1.000 1.000 1.000 1.000 106 0.125 OmmlOSO* 230 0.167 0.038 0.031 0.160 234 1.000 0.833 0.942 0.875 0.800 236 0.094 0.040 238 270 0.125 273 0.019 281 0.125 292 300 304 0.250 308 0.500 Omy0004* 77 131 137 139 141 159 1.000 1.000 0.981 0.019 0.125 0.125 0.500 0.125 0.125 1.000 1.000 443 Table 2-continued Alleles Sample and allele frequencies Locus Bertha Blind Boles Broad us Cache Colt OmylOOr 186 190 192 202 206 222 0.028 0.021 0.125 0.125 0.500 0.125 228 0.056 0.042 0.031 0.020 232 0.021 0.019 0.094 0.020 236 0.639 0.333 0.346 0.031 0.140 240 0.083 0.019 0.125 0.220 244 0.219 0.040 248 0.063 0.040 252 0.083 0.058 0.040 254 0.125 256 0.028 0.083 0.250 0.188 0.100 260 0.146 0.192 0.094 0.080 264 0.056 0.063 0.096 0.219 0.180 268 0.111 0.125 0.019 0.060 272 0.021 0.040 276 0.020 Average Westslope 0.991 0.997 0.984 0.067 0.998 0.994 Average Rainbow 0.009 0.003 0.016 0.933 0.002 0.006 444 Table 2-continued Alleles Sample and allele frequencies Locus Devils Eustache Inez Packer Plant Trail Occ34* 215 0.062 0.075 0.150 225 0.938 1.000 1.000 1.000 0.925 0.850 Occ35* 200 230 1.000 1.000 1.000 1.000 1.000 1.000 Occ36* 275 285 0.062 0.071 324 0.750 0.100 0.175 0.050 325 0.188 0.929 0.900 1.000 0.825 0.950 Occ37* 260 0.062 270 1.000 0.938 1.000 1.000 1.000 1.000 Occ38* 150 0.062 175 1.000 0.938 1.000 1.000 1.000 1.000 Occ42* 160 0.062 190 1.000 0.938 1.000 1.000 1.000 1.000 Om55* 199 200 0.062 0.025 0.025 220 1.000 0.875 0.900 0.714 0.925 0.850 221 0.062 0.100 0.286 0.025 0.150 Ssa408* 178 182 186 0.062 0.062 195 0.875 1.000 0.900 1.000 1.000 1.000 199 0.100 210 218 234 250 282 mm 1037-1* 139 147 0.438 0.750 0.400 0.929 0.350 0.025 0.800 151 0.562 0.250 0.500 0.071 0.575 0.150 155 159 0.050 171 0.100 0.050 187 445 Table 2-continued Alleles Sample and allele frequencies Locus Devils Eustache Inez Packer Plant Trail 0mm 1037-2* 100 0.062 0.050 104 1.000 0.938 1.000 1.000 0.950 1.000 106 Omm1050* 230 0.150 0.050 234 0.875 0.500 1.000 0.786 0.750 0.900 236 0.071 0.050 238 0.500 270 273 281 0.050 292 0.062 300 0.062 0.143 0.050 304 308 Omy0004* 77 131 137 139 141 159 1.000 0.875 1.000 1.000 0.125 1.000 1.000 446 Table 2-continued Sample and allele frequencies Locus Alleles Devils Eustache Inez Packer Plant Trail OmylOOr 190 0.125 0.025 186 0.214 190 0.125 192 0.143 202 206 222 228 232 0.375 0.214 236 0.125 0.286 240 0.188 0.143 244 248 252 0.125 254 256 260 264 0.062 268 272 276 0.300 0.700 0.071 0.357 0.143 0.100 0.400 0.286 0.150 0.071 0.025 0.050 0.125 0.200 0.275 0.150 0.071 0.075 0.200 0.125 0.050 0.025 0.025 Average Westslope 0.962 0.934 0.985 0.989 0.977 0.981 Average Rainbow 0.038 0.066 0.015 0.011 0.023 0.019 447 Table 3 Allele frequencies at the diagnostic loci between westslope cutthroat and rainbow trout in samples showing evidence of hybridization between these fishes that do not appear to be hybrid swarms collected from Camp Creek, lower Inez Creek, upper Holloman Creek, Ninemile Creek, lower Packer Creek, North Fork Fish Creek, South Fork White Creek, Thompson Creek, Vaughn Creek, and White Creek. Alleles in bold are characteristic of rainbow trout. Alleles Sample and allele frequencies Locus Camp Inez Holloman Ninemile Packer Occ34* 215 0.083 0.300 0.125 0.071 225 0.917 1.000 0.700 0.875 0.929 Occ35* 200 0.042 0.200 0.071 230 0.958 1.000 0.800 1.000 0.929 Occ36* 275 0.300 0.167 0.071 324 0.167 0.071 0.500 325 0.833 0.929 0.700 0.333 0.929 Occ37* 260 0.350 0.071 270 1.000 1.000 0.650 1.000 0.929 Occ38* 150 0.300 0.125 175 1.000 1.000 0.700 0.875 1.000 Occ42* 160 0.300 0.071 190 1.000 1.000 0.700 1.000 0.929 Om55* 199 200 0.100 0.250 0.200 220 0.625 1.000 0.650 0.800 0.786 221 0.375 0.214 Ssa408* 178 182 190 0.042 0.125 0.150 0.125 195 0.958 0.875 0.850 0.875 1.000 0mm 1037-1* 135 0.083 0.071 139 0.292 0.500 0.650 0.667 0.714 143 0.071 147 0.042 151 0.542 0.500 0.300 0.167 0.071 159 179 0.042 0.050 0.167 0.071 0mm 1037-2* 98 0.125 100 0.083 0.150 0.333 0.071 104 0.917 0.875 0.850 0.667 0.929 448 Table 3-continued Alleles Sample and allele frequencies Locus Camp Inez Holloman Ninemile Packer 0mm 1050* 230 0.167 0.143 0.050 234 0.792 0.857 0.750 0.875 0.786 236 0.125 238 256 0.050 269 0.050 273 280 0.042 0.071 281 0.100 300 0.071 328 0.071 my 0004* 77 137 139 141 143 145 153 165 1.000 0.875 0.063 0.063 0.750 0.100 0.050 0.100 0.713 0.214 0.071 0.929 0.071 Omy1001* 174 176 178 182 186 190 192 218 232 0.042 0.050 0.100 0.050 0.050 0.250 0.167 0.083 0.143 0.071 236 0.333 0.143 0.200 0.083 0.429 240 0.292 0.429 0.050 0.167 0.071 244 0.083 248 252 0.083 0.214 256 0.042 0.250 0.071 260 0.208 0.357 0.100 264 0.167 268 0.071 272 276 0.150 449 Table 3-continued Alleles Sample and allele free juencies Locus NF Fish SF White Thompson Vaughn White Occ34* 215 0.056 225 0.944 1.000 1.000 1.000 1.000 Occ35* 200 0.050 0.028 0.067 230 1.000 1.000 0.950 0.972 0.933 Occ36* 275 0.063 0.050 324 0.056 0.063 0.100 0.083 0.067 325 0.944 0.875 0.850 0.917 0.933 Occ37* 260 0.050 0.056 0.067 270 1.000 1.000 0.950 0.944 0.933 Occ38* 150 0.050 175 1.000 1.000 0.950 1.000 1.000 Occ42* 160 190 1.000 1.000 1.000 1.000 1.000 Om55* 199 200 0.063 0.050 0.028 220 0.889 0.563 0.850 0.944 1.000 221 0.111 0.375 0.100 0.028 Ssa408* 178 182 190 195 1.000 1.000 1.000 1.000 1.000 0mm 1037-1* 135 139 0.500 0.438 0.400 0.611 0.188 143 147 0.063 151 0.444 0.500 0.600 0.361 0.750 159 0.028 0.063 179 0.056 0mm 1037-2* 98 100 0.056 104 0.944 1.000 1.000 1.000 1.000 450 Table 3-continued Sample and allele frequencies Locus Alleles NF Fish SF White Thompson Vaughn White 0.050 0.083 0.900 0.861 0.933 0.028 0.028 0.067 0mm 1050* 230 0.111 234 0.833 1.000 236 238 256 269 273 280 0.056 281 300 328 my 0004* 77 137 139 141 143 145 153 165 1.000 1.000 Omy1001* 174 176 178 182 186 190 192 218 232 236 0.444 240 0.167 0.125 244 0.111 248 252 256 0.111 0.313 260 0.056 0.188 264 0.056 0.188 268 0.188 272 276 0.056 0.050 1.000 1.000 1.000 0.050 0.028 0.200 0.111 0.188 0.400 0.333 0.375 0.200 0.250 0.125 0.056 0.067 0.083 0.188 0.050 0.028 0.056 0.067 0.100 0.028 0.028 451 Table 4 Frequencies of alleles characteristic of westslope cutthroat trout, Yellowstone cutthroat trout, both westslope and Yellowstone cutthroat trout (cutthroat), and rainbow trout in samples showing evidence of hybridization among these fishes collected from Deborta Creek, lower Murphy Creek, North Fork Blackfoot River above Deborta Creek, North Fork Blackfoot River below South Creek, North Fork Blackfoot River below Theodore Creek, and lower Richmond Creek. Averages are given only if the sample appears to have come from a hybrid swarm. Note averages do not sum to one because they are estimated only using a subset of all the loci analyzed. Sample and allele frequencies North Fork Blackfoot Locus Alleles Deborta Murphy Deborta South Theodore Richmond 0.200 1.000 0.250 1.000 1.000 1.000 0.083 1.000 0.875 1.000 1.000 1.000 0.100 1.000 1.000 0.250 1.000 Cutthroat Alleles Occ34* 225 Occ35* 230 Occ36* 324 325 Occ37* 270 Occ38* 175 Occ42* 190 452 Table 4-continued Sample and allele frequencies North Fork Blackfoot Locus Alleles Deborta Murphy Deborta South Theodore Richmond Westslope alleles Om55* 220 221 Ssa408* 195 Omm1037-r 139 147 151 0mm 1037-2* 104 Omm1050* 230 234 Omy0004* 77 OmylOOr 228 232 236 240 244 256 260 264 268 0.400 Average Westslope 0.818 0.091 1.000 0.318 0.682 1.000 0.909 1.000 0.136 0.136 0.091 0.091 0.182 0.227 0.045 0.091 0.974 0.100 0.500 0.250 0.200 0.100 0.958 1.000 0.333 0.667 1.000 0.042 0.958 1.000 0.042 0.042 0.208 0.417 0.208 0.042 0.042 0.994 453 Table 4-continued Locus Sample and allele frequencies North Fork Blackfoot Alleles Deborta Murphy Deborta South Theodore Richmond Yellowstone alleles Om55* 180 0.500 0.091 Ssa408* 199 0.600 Oki10* 165 0.100 0mm 1037-1* 127 0.100 0mm 1037-2* 106 0.400 0mm 1050* 235 0.045 Omy0004* No Yellowstone alleles detected OmylOOl* No Yellowstone alleles detected Average Yellowstone 0.017 0.200 0.042 0.200 0.200 0.100 0.100 0.250 0.005 454 Table 4-continued Sample and allele frequencies North Fork Blackfoot Locus Alleles Deborta Murphy Deborta South Theodore Richmond Rainbow alleles Occ34* Occ35* Occ36* Occ37* Occ38* Occ42* Om55* Ssa408* Omm1037-r Omm1 037-2* Omm1050* 215 0.800 200 1.000 275 1.000 285 260 1.000 150 1.000 160 1.000 199 0.400 200 0.100 170 178 182 186 190 218 0.200 234 0.200 250 139 0.200 159 171 0.700 183 100 0.600 240 0.200 270 274 276 0.100 281 0.200 282 286 296 300 304 308 0.500 0.045 1.000 1.000 1.000 1.000 0.900 1.000 0.300 0.700 0.100 0.750 1.000 1.000 1.000 1.000 0.750 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.500 0.300 0.100 0.250 0.250 0.200 0.300 0.250 0.500 0.300 0.100 0.100 0.250 0.100 0.300 0.500 0.200 0.400 0.250 0.200 0.900 0.750 1.000 0.100 0.200 0.100 0.100 0.250 0.250 0.100 0.250 0.200 0.800 0.250 0.400 0.042 455 Table 4-continued Locus Sample and allele frequencies North Fork Blackfoot Alleles Deborta Murphy Deborta South Theodore Richmond Rainbow alleles Omy0004* 131 137 0.600 139 141 153 OmylOOr 186 190 0.600 192 0.200 198 0.100 200 202 206 0.100 214 0.100 0.600 0.750 0.800 0.100 0.250 0.100 0.300 0.600 0.250 0.200 0.500 0.100 0.100 0.100 0.100 0.100 0.250 0.100 0.200 0.100 Average Rainbow 0.003 0.003 456 Table 5 Allele frequencies at the loci showing evidence of genetic variation in samples from what appear to be non-hybridized westslope cutthroat trout collected from Butler Creek, Canyon Creek, East Fork Indian Creek, East Fork Morrell Creek, Findell Creek, lower Holloman Creek, upper Inez Creek, Irish Creek, lower Mill Creek, upper Mill Creek, Miller Creek, Montana Creek, lower Mormon Creek, upper Mormon Creek, upper Murphy Creek, upper Packer Creek, upper Richmond Creek, upper Trail Creek, West Fork Clearwater River, and West Fork Indian Creek Alleles Sample and allele frequencies Locus Butler Canyon EF Indian EF Morrell Findell Holloman Occ36* 324 0.450 0.360 0.024 0.136 325 0.550 0.640 1.000 0.976 0.864 1.000 Om55* 199 0.023 220 1.000 1.000 1.000 1.000 0.864 0.944 221 0.114 0.056 Ssa408* 195 282 0.983 0.017 1.000 1.000 1.000 1.000 1.000 Oki10* 97 101 0.117 105 0.400 0.024 0.273 0.167 109 0.833 0.238 0.045 0.111 113 0.217 0.260 0.167 0.286 0.159 0.278 117 0.567 0.320 0.024 0.091 121 125 0.119 0.023 0.056 129 0.020 0.024 0.205 0.056 133 0.068 0.111 137 141 0.100 0.114 145 0.214 0.023 0.??? 149 153 0.048 157 0.024 Omm1037-r 139 147 0.967 0.300 0.500 0.500 0.048 0.636 0.667 0.056 151 0.033 0.480 0.417 0.452 0.341 0.278 155 0.020 0.083 0.023 163 0.200 0mm 1050* 230 0.429 0.278 234 1.000 1.000 0.429 0.909 0.722 235 236 0.091 238 0.143 457 Table 5-continued Alleles Sample and allele frequencies Locus Butler Canyon EF Indian EF Morrell Findell Holloman OmylOOr 219 228 232 236 0.467 0.283 0.020 0.083 0.667 0.167 0.227 0.250 0.056 240 244 0.150 0.060 0.167 0.083 0.357 0.071 0.227 0.159 0.167 0.222 248 252 0.120 0.048 256 260 0.340 0.240 0.048 0.095 0.091 0.222 0.111 264 268 0.100 0.100 0.120 0.048 0.045 0.111 272 276 0.024 0.143 0.056 0.056 458 Table 5-continued Alleles Sample and allele frequencies Locus Inez Irish LMill UMill Miller Montana Occ36* 324 0.077 0.143 0.083 0.500 0.125 325 0.923 0.857 0.917 1.000 0.500 0.875 Om55* 199 220 0.577 1.000 1.000 1.000 1.000 0.750 221 0.423 0.250 Ssa408* 195 282 1.000 1.000 1.000 1.000 1.000 1.000 OkHO* 97 0.063 101 0.071 0.083 0.063 105 0.357 0.167 0.250 0.063 109 0.577 0.071 0.150 0.313 113 0.308 0.500 0.500 0.800 0.600 0.438 117 0.115 0.083 0.200 0.063 121 125 129 0.083 133 0.083 137 141 145 149 153 157 0mm 1037-1* 139 147 0.538 0.500 0.417 0.833 0.500 0.313 0.688 151 0.462 0.500 0.583 0.167 0.500 155 163 0mm 1050* 230 0.063 234 1.000 0.714 1.000 1.000 0.950 0.938 235 0.286 236 238 0.050 459 Table 5-continued Alleles Sample and allele frequencies Locus Inez Irish LMill UMill Miller Montana OmylOOr 219 228 232 236 0.038 0.071 0.083 0.833 0.667 0.200 0.125 240 244 0.077 0.143 0.143 0.300 0.033 0.250 0.063 248 252 0.269 0.050 256 260 0.154 0.154 0.571 0.150 0.100 0.063 0.313 264 268 0.115 0.192 0.071 0.083 0.050 0.125 0.188 272 276 0.150 0.050 0.125 460 Table 5-continued Om55* Oki10* Omm1037-r Sample and allele frequencies Locus Alleles L Mormon U Mormon Murphy Packer Richmond Trail Occ36* 324 0.067 0.143 0.158 1.000 0.857 1.000 0.842 Ssa408* 195 1.000 1.000 1.000 1.000 1.000 1.000 324 0.067 325 0.933 1.000 199 220 0.967 1.000 221 0.033 195 1.000 1.000 282 97 101 105 0.536 0.429 109 0.071 113 0.250 0.500 117 121 125 129 0.036 0.071 133 137 0.071 141 0.036 145 149 153 157 139 0.464 1.000 147 151 0.464 155 0.071 163 230 0.071 234 0.679 0.929 235 236 0.321 238 0.727 0.679 0.955 0.947 0.273 0.321 0.045 0.053 0.182 0.818 0.158 0.250 0.184 0.071 0.273 0.184 0.214 0.727 0.211 0.036 0.036 0.143 0.071 0.179 0.158 0.079 0.026 0.864 0.679 0.227 0.342 0.026 0.136 0.321 0.773 0.631 0mm 1050* 230 0.071 0.107 0.184 1.000 0.857 1.000 0.763 0.036 0.053 461 Table 5-continued Sample and allele frequencies Locus Alleles L Mormon U Mormon Murphy Packer Richmond Trail OmylOOr 0.045 0.036 0.429 0.393 0.182 0.105 0.273 0.263 0.079 0.227 0.105 0.053 0.036 0.105 0.682 0.071 0.591 0.079 0.036 0.053 0.079 0.026 0.053 0.250 0.214 0.214 0.071 0.500 0.286 0.036 0.429 462 Table 5-continued Sample and allele freqi jencies Locus Alleles WF Clearwater WF Indian Occ36* 324 0.294 325 0.706 1.000 Om55* 199 220 0.706 1.000 221 0.294 Ssa408* 195 282 1.000 1.000 Oki10* 97 0.118 101 0.088 105 0.029 109 0.382 0.900 113 0.294 0.100 117 0.088 121 125 129 133 137 141 145 149 153 157 0mm 1037-1* 139 147 0.735 151 0.265 0.400 155 0.600 163 OmmlOSO* 230 234 1.000 1.000 235 236 238 463 Table 5-continued Sample and allele frequencies Locus Alleles WF Clearwater WF Indian OmylOOr 219 228 0.088 232 236 0.324 0.700 240 0.353 0.300 244 248 0.029 252 256 0.029 260 0.118 264 0.059 268 272 276 464 Camp 9 n 8 7 - |6 •5 5 1 4 30 z 2 1 n ■ Observed D Expected J i 1 u ^ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Hybrid Index Figurel. Observed and expected random distribution of hybrid indices among the fish showing evidence of hybridization between westslope cutthroat and rainbow trout collected from Camp Creek. Note the observed distribution significantly differs (P<0.05) from the expected random distribution suggesting the sample did not come from a hybrid swarm. 465 1.2 1 I 0.8 o >- 0.6 I 0.4 0.2 Deborta ■ Observed D Expected Hybrid Index Figure 2. Observed and expected random distribution of hybrid indices among the fish showing evidence of hybridization among westslope cutthroat, Yellowstone cutthroat, and rainbow trout collected from Deborta Creek. Note the observed distribution significantly differs (P<0.05) from the expected random distribution suggesting the sample did not come from a hybrid swarm. 466 (0 3.5 3 2.5 •5 2 I 1.5 E I 1 0.5 Upper Holloman ■ Observed D Expected L % V - 4 0) n E 3 3 Lower Inez 2 1 H ■ Observed D Expected n, t>^ - 4 E 3 ^ 2 1 n nJ U 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 b ^,0 - 4 E 3 ^ 2 n 1 n I U 1 1 1 1 1 1 o q. ix