The overall goal of the Libby Mitigation Project is to offset impacts caused by the Federal Columbia River Power System (FCRPS) operations and mitigate fish and habitat losses attributed to the construction and operation of Libby Dam.  Mitigation is accomplished using watershed-based habitat enhancement, fish passage improvements, modified dam operations and offsite fisheries habitat improvement measures.  The Libby Mitigation Project’s previous project review (FY2007-2009) identified the following objectives. 

 Basin-Level Objective:  Maintain and restore healthy ecosystems and watershed, which preserve functional links among ecosystem elements to ensure the continued persistence, health and diversity of all species including game fish species, non-game fish species, and other organisms (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p. 16). 

Objectives:

 1A:  Improve channel stability to a level equivalent to the QHA-generated channel stability scores of reference and Class 1 streams (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  32).

1B:  Improve stream habitat diversity to a level equivalent to the QHA-generated habitat diversity scores of reference streams (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  35).

1C:  Restore and provide passage to migratory fish by removing potential man caused barriers, i.e. impassable culverts, hydraulic headcuts, water diversion blockages, landslides, and impassable deltas (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  37).

Objectives:

2A:  Suppress and prevent expansions of populations of non-native fish species beyond current levels in the Kootenai Subbasin (Kootenai Subbasin Management Plan [KTOI and MFWP 2004]  p. 45 and 54).

2B:  Maintain or increase the total number of genetically pure local populations of westslope cutthroat trout, and maintain a broad distribution of local population in existing metapopulations (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  52). 

Basin-Level Objective:  Protect and expand habitat and ecosystem functions as the means to significantly increase the abundance, productivity, and life history diversity of resident fish at least to the extent that they have been affected by the development and operation of the hydrosystem  (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  16).

Objective 3:  Restore primary, secondary, and tertiary productivity rates and nutrient values downstream from Libby Dam to pre-dam condition (equal to those of inflows into Libby Reservoir, corrected for downstream lateral input).  (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  47, 56, 60, and 65).

Basin-Level Objective:  Complete assessments of resident fish losses throughout the basin resulting from the hydrosystem, expressed in terms of the various critical population characteristics of key resident fish species (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  16).

Objective 4:  Maintain or increase of local populations of native fish species in the Kootenai Basin.  This objective implies the need to monitor status and trend of native and important fish species in the Montana portion of the Kootenai Basin.

Objective 5:  Maintain or increase the total number of genetically pure local populations of redband trout in the Kootenai Subbasin.  (Kootenai Subbasin Management Plan [KTOI and MFWP 2004] p.  48).

5A:  Develop non-lethal genetic markers capable of discriminating between coastal rainbow trout, westslope cutthroat trout and redband trout. 

The following is a brief summary of work accomplishments by the Libby Mitigation Project (199500400) since the previous project review and an evaluation whether these activities are achieving stated objectives.

Objectives 1A and 1B: 

Libby Creek Demonstration Project

MFWP cooperated with the landowner, Plum Creek Timber Company, to complete the Libby Creek Demonstration Project within this area in the fall of 2001 located at approximately RM 12.3.  The restoration project reconstructed one meander length the Libby Creek stream channel (approximately 1,700 feet), and installed several bank stabilization and habitat improvement structures (Dunnigan et al. 2003).  Two of the largest point sediment sources within the Libby Creek Watershed existed above the confluence of Elliot Creek (RM 12.0), and were contributing substantial amounts of course and fine sediment to Libby Creek each year.  The largest eroding bank within the project site was over 700 feet long, averaged 80 feet high and was contributing an estimated average of 5,900 cubic yards of sediment annually to Libby Creek.  The second large unstable bank was located in the lower section of the project area and was also contributing substantial amounts of sediment to Libby Creek that increased sediment deposition; accelerated bank erosion; increased width/depth ratio and contributed to lack of habitat complexity within this section of Libby Creek.  The main objectives of this project were to: 1) Decrease coarse and fine sediment sources, 2) Decrease the stream’s width depth ratio, and 3) Return the stream channel to a properly functioning configuration able to efficiently transport bed load sediment during high discharge events; and 4) Increase the quality and quantity of fisheries habitat within this reach of Libby Creek.  During the past five years, MFWP conducted physical monitoring of the Libby Creek Demonstration Project.  The original work restoration work significantly changed the stream channel dimensions (Dunnigan et al. 2003), which ultimately resulting in a deeper and narrower channel, which translated into a significantly lower width/depth ratio after project implementation, and increased the quantity and quality of rearing habitat for native salmonids within the project reach.  Stream channel dimensions within the project area are similar to the as-built conditions (Dunnigan et al. 2011).  The project continues to meet the original objectives including limiting instream sediment from two large sources within the project area.  Stream channel instability immediately outside the project area has increased, while bank erosion within the project area has remained low.  Rainbow trout are the most abundant fish species at this site, and using the Before/After/Control/Impact (BACI) statistical design, were able to detect a significant increase in the abundance of rainbow trout at this site relative to the control site (Dunnigan et al. 2011), which represented an almost two fold increase.   Therefore, based on the condition of the physical habitat that resulted in this section of Libby Creek, and the fish population response, this effort was a success. 

Libby Creek Upper Cleveland Project

MFWP completed the Libby Creek Upper Cleveland Stream Restoration Project in the fall of 2002 (approximate river mile 22), which restored approximately 3,200 feet of stream channel.  Dunnigan et al. (2004; 2005; 2007; 2008; 2009) demonstrated that this restoration project decreased the bankfull width and bank erosion and increased stream depth, overall length, substrate mean particle size, and the quality and quantity of salmonid rearing habitat through 2006.  More recent monitoring was intended to determine if the physical changes to the habitat in this section of Libby Creek were sustained through time, especially after a rain on snow event with an approximate 20 year return interval occurred in November 2006 in the upper Libby Creek watershed (Dunnigan et al. 2011).  Despite the relatively large flow event, the riffle habitats in 2009 were significantly deeper, had a lower cross sectional area and width to depth ratio than existed prior to restoration, and although not significant, the riffles also remained narrower.    The number of pools in this section of Libby Creek also remained over two fold more abundant than existed prior to the restoration efforts, while pool dimensions have changed relatively little since the original restoration activities (Dunnigan et al. 2011).  The sustained increase in the number of pools also likely translated into increased total pool area and volume within the project area than existed prior to restoration, despite having undergone an approximate 20 year flood event.  Redband trout are the dominant fish species at this site, but we were unable to determine either a significant trend over time or a significant change in abundance since the project was completed (Dunnigan et al. 2011), despite the fact that our two most recent years of sampling resulted in the two highest abundance estimates on record, including the pre-restoration dataset. 

 Libby Lower Cleveland Project Phases I and II

The lower Cleveland property on Libby Creek is located approximately at river mile 21 (1 mile downstream of the upper Cleveland Property), and encompasses approximately 9,100 feet of stream channel.  MFWP developed a restoration strategy to implement increase stream bank stability and habitat quantity and diversity in three phases.  The first phase was completed in October 2005, and is referred to as the Libby Creek Lower Cleveland Phase I Project.  Dunnigan et al. (2007) presents a complete description of the materials and structures installed in this 2,950 feet long section of Libby Creek.  The second phase of this effort (Phase II) was completed in October 2006, and began at the downstream boundary of the Phase I project area and continued 3,273 feet downstream.  This project constructed a variety of structures intended to increase habitat diversity and bank stability (Dunnigan et al. 2007), and resulted in a more meandering (longer), narrower, and deeper stream channel.  Pool habitat also increased in both sections, as represented by the total number of pools, pool area and volume.  The winter flood event in 2006 changed the riffle and pool habitat dimensions within both project areas slightly (Dunnigan et al. 2011).  Despite the changes within these two project areas the physical changes which resulted from the initial restoration activities are being sustained through time including several years after a relatively large flood event.  The total number of pools, total pool area and volume within the Phase I project area remain 50, 148 and 215% higher than existed prior to the project.  The riffle habitats in both project areas also remain narrower and deeper than existed prior to our work.  The plan form within both project areas in 2007-2009 was relatively similar (Dunnigan et al. 2011).    

We monitored fish abundance at two control sites (above and below) and with the Phase II Project area in order to evaluate the fish community response to the restoration work.  Redband trout dominated the fish community at all three sampling locations during all years.  However, redband trout abundance at each of the sections in 2007 decreased substantially in each of the sections compared to the three years previous (Dunnigan et al. 2009).  The decrease in abundance is presumably due to the rain-on-snow event that occurred in November 2006 within the upper Libby Creek watershed.  We were unable to detect a significant trend in abundance for either redband, brook or bull trout in any of the three sections in upper Libby Creek (p > 0.05; Dunnigan et al. 2011).  We were unable to detect any significant changes in fish abundance over the pre and post periods using either the pre and post comparison or the BACI statistical design (Dunnigan et al. 2011).  We concluded that the lack of statistical power for both the student’s t-test and the BACI tests is primarily due to the limited number of observations prior to and after restoration and the relatively short duration since the work was completed that was not sufficient time for the populations to respond. 

We decided not to complete the third and final phase of the lower Cleveland Project due in part to the initial comments provided by the ISRP during the previous review process and the careful review of our monitoring results.  We plan to continue to monitor these two projects to determine if additional work is warranted. 

Restoration of lower Grave Creek

In 1996, a watershed analysis was completed to support development of a Total Maximum Daily Load (TMDL) for the Grave Creek watershed (DEQ 2005), which identified lower Grave Creek as having fish habitat limitations linked to excess sediment loading and low large woody debris.  Additional problems included an over widened channel, eroding banks and a reduction in function of the riparian corridor.  The restoration efforts completed on lower Grave Creek began in 2001 as a cooperative effort between a number of agencies, organizations and individuals, with much of the work facilitated through the Kootenai River Network (a local watershed group).  The two primary objectives of the initial restoration work were to increase bank stability and enhance salmonid habitat of 8,200 feet of stream channel in three phases:  Demonstration Phase (1,000 feet), Phase I (4,200 feet), and Phase II (3,000 feet), which were completed in 2001, 2002, and 2004, respectively.  Dunnigan et al. (2003) describes the initial work that was completed in the Demonstration and Phase I project areas, and Dunnigan et al. (2007) describes the initial work completed in the Phase II project area.  Substantial effort has been devoted to monitor and evaluate the restoration activities (Dunnigan et al. 2011; River Design Group 2011), and generally concur that the physical changes that  resulted from the initial stream restoration activities have been sustained through time and meet the two objectives of increased streambank stability, reduced instream sediment sources and increased habitat diversity.  

MFWP only monitors fish abundance within the Demonstration Phase of the Grave Creek project area.  We compared mean fish abundance during pre and post project periods at the site.  Bull trout and rainbow trout were the two most abundant species, and mean abundance of each species was higher after the project was implemented.  Rainbow trout abundance increased substantially (184%) after project construction.  However, this difference was not significant (p = 0.18; two-tailed test), and the observed power of the test was very low (Dunnigan et al. 2011).  Bull trout abundance after project completion also increased over 3.5 fold after project completion, which represented a nearly significant increase (p = 0.068; two tailed test), but the power of this test was also low (0.46).   Brook trout and westslope cutthroat trout abundance were low during all years, but nearly identical before and after project completion (Dunnigan et al. 2011).  We were unable to use the more powerful Before/After/Control/Impact (BACI) statistical design at this site due the lack of an adequate control site nearby in Grave Creek.  Annual variability of fish abundance within the project area is relatively large, and likely limits our ability to detect changes that result from restoration efforts.  Given the lack of a control site, additional years of fish population monitoring prior to restoration may have also improved the statistical power of most of our tests to detect change relative to the restoration activities. 

Although our monitoring results suggest that the initial stream channel work is increasing channel stability and improving fish habitat, and that those changes have been sustained through time, we concluded that our initial revegetation efforts, implemented in 2001, resulted in limited success due to several site constraints (Geum Environmental 2008a).  Successful revegetation strategies are key to achieving project objectives long-term.  Therefore, efforts since the initial channel restoration have been focused primarily on restoring a functioning riparian community that will ultimately sustain streambank stability and diverse habitat for fish.   

MFWP (through Project 1995-004-00) has contributed to the following revegetation efforts within the Grave Creek project area either through direct funding or in-kind contributions. 

2001:  Revegetation treatments implemented in conjunction with stream channel restoration work included whole sod and shrub transplants, containerized root stock, dormant sprigs and pole plantings, broadcast seeding, and organic compost application.

2005:  Supplemental riparian revegetation including stream bank bioengineering  (vegetated soil lifts), planting a small number of containerized shrubs and trees, and enhancement of floodplain areas through construction of small swale features and placement of wood to promote natural floodplain processes and natural recruitment of riparian vegetation within entire project area. 

2006:  Installation of 13 vegetated soil lifts on high priority streambanks throughout the entire project area. 

2008:  In 2007, a revegetation plan was completed that included all three project phases.  This plan included an inventory of existing and desired conditions, a set of alternatives, a project implementation plan, project schedule, and a monitoring plan.  The plan also acknowledged the importance of restoring a functioning riparian community in order to provide the long-term bank stability and maintenance of diverse fishery habitat.

In 2008, a collaborative effort was made to begin implementing the revegetation plan (Geum Environmental 2008a; 2008b).    In summer 2008, effectiveness monitoring of completed revegetation treatments was conducted and results were compared with monitoring conducted in late 2007.   Effectiveness monitoring activities included: riparian planting area survival estimates, percent cover of woody vegetation on vegetated soil lifts, and monitoring of plant community succession on treated point bars (Geum Environmental 2009a).  The monitoring results were used to guide selection, design and implementation of revegetation treatments in the fall of 2008.  Revegetation treatments implemented in Fall 2008 included:  installation of riparian fencing in the upper half of the project area to temporarily limit deer and elk browse of the project area, maintenance of existing riparian planting areas, construction of additional floodplain swales and woody debris placement, planting and seeding within constructed swales on inside meanders, installation of bioengineering treatments (vegetated soil lifts, coir log fascines and buried coir/willow fascines), weed control, construction of a vegetated set back bank in one location, and repair of damaged stream channel grade control structures (Geum Environmental 2009a). 

2009:  A collaborative effort was made to implement additional revegetation treatments in the Demonstration Project area (Geum Environmental 2009).  These treatments are briefly summarized in Table 1.  

2009 (Continued):  Geum Environmental (2009a) also completed effectiveness monitoring in 2009 of revegetation treatments implemented  between 2005 and2008.  Important findings of that monitoring are summarized below: 

Containerized Planting: Containerized planting units were monitored to estimate plan survival, percent cover of herbaceous species, and evidence of browse.  In 2005, containerized shrubs were installed on outer meander bends where initial revegetation efforts did not work.  The survival rates of the planting units from 2005 were relatively low due to well-drained substrate, competition from aggressive grass species and bank erosion.

Containerized plants were installed on inside meander floodplain surfaces in 2008. Survival of containerized plants within these treatments was 100%.  For plants installed within the electric fence exclosure no browse was observed.  Slight to moderate browse of plants installed outside the fence was observed.  On average, plants within the fence were one to two feet higher than those outside.  Woody debris placed around planted shrubs in combination with the steep sides of the constructed swales is relatively effective at preventing extensive browse outside of the fenced area.  Seeded grasses, shrubs and forbs, as well as naturally recruited shrubs and forbs were recorded in constructed swales, and some swales had very high density of naturally recruited cottonwood seedlings.  Weed species were recorded in some swales, but were not dominant species.  This treatment is effective at creating safe microsites and desirable conditions for establishment of riparian shrubs and trees.  Containerized planting on outer meanders is only marginally effective as a revegetation treatment, but installing containerized plants within natural or created microsites and other protected locations within the floodplain is an effective revegetation technique at this site. 

Streambank Bioengineering Treatments:  Monitoring of the bioengineering treatments installed in 2005, 2006, and 2008 yielded the following observations.  Average shoot height (new growth) recorded for willows in structures installed in 2005 and 2006 structures ranged from 12 to 72 inches.  Maximum height recorded during the 2008 monitoring was 36 inches.  Average shoot height for willows in the 2008 structures ranged from two to 18 inches on coir log fascines and six to 18 inches on the vegetated soil lifts.  Total percent willow cover on the 2005 and 2006 structures ranged from 23 to 93%, 53% on the 2008 soil lifts, and 37% on the 2008 coir logs fascines.  There was no evidence of browse on any of the structures located within the electric fence, but browse was evident on structures outside the fence. Browse was more extensive in 2008 than in 2009.  Monitoring indicates that these treatments are creating stable areas along the channel to allow establishment of woody vegetation. 

Monitoring of the buried coir/willow fascines resulted in the following observations and conclusions.  In general, willow survival and growth is good, but browse may affect growth over the long term.  Overall, minimal amounts of organic matter and woody debris have accumulated around the fascines.  This may be an effective treatment to establish islands of willow on constructed point bars, but continued monitoring will be required to determine if these treatments can provide long term stability and promote successional processes.

Point Bar/Inside Meander Enhancement: Constructed inside meander floodplains and point bars were also monitored in 2009, and resulted in several observations.  Treatments including swales, woody debris, seeding, and containerized plantings on point bars and inside meanders appear to provide the structure that supports ecological processes necessary for desired pioneer plant species to colonize and plant community succession to occur.  The survival of containerized plants is high, native shrubs, trees and forbs are colonizing the swales and other microsites, and flood deposited sediment and debris is accumulating around woody debris.

Vegetated Set Back Bank:  One vegetated set back bank treatment was constructed within the project area to allow continued migration of the channel.  This site was also monitored in 2009.  Willow survival along the edges of the trench was estimated to be 80 to 90%.  Containerized plant survival was also high (>80%) within the treated area.  The majority of the cottonwood pole cuttings were re-sprouting from the base while other have new growth along the entire pole.  Weed cover within the trench is low, but the surrounding area has high densities of knapweed and oxeye daisy.  Seeded grass cover is low, but grasses were beginning to germinate in the bottom of the trench.  Pasture grasses were also present.

2009 (Continued):An integrated weed management plan was completed for the project in 2009 (Geum Environmental 2009c), which included an inventory of existing weed species within the project area, weed management strategies, and an adaptive management framework for monitoring and managing weed species.  Weed treatments, including hand-pulling and herbicide application were completed in 2009 targeting state listed noxious weed species.

2010:  Based on the results of 2009 effectiveness monitoring, additional revegetation treatments were implemented in 2010 and 2011 in high priority areas.  A total of 244 large container sized shrubs and trees were installed at nine sites within the project reach to promote long term riparian function, including bank stability.  A variety of plant material sizes were installed including: 10 gallon, 15 gallon, and 6 to 8-foot tall, 1 to 2-inch diameter ball and burlap.  A total of 8,840 feet of riparian fence was installed in October, 2010.  Fence types designed to exclude deer from entering the project area included 7-strand electric slant fence and 8 foot tall 15-strand vertical electric fence (Geum Environmental 2011).

2011:  As a continuation of treatments identified in 2010, a total of 67, 16-inch high density coir logs were installed at three sites along Grave Creek between August 15 and 16, 2011 (Geum Environmental 2011), for a total of 340 linear feet of streambank treatment. A total of 620 containerized plants were also installed between coir logs and on the banks behind the coir logs at each location.  Effectiveness monitoring including survival monitoring of nine large containerized planting sites originally planted in 2010.   Total survival for all sites is 87 percent.

Implementing the list of treatments above, integrating monitoring and monitoring results into decision making has required a partnership committed to the long-term success of the project.  The partnerships on the Grave Creek Restoration Project have built synergy in achieving project objectives through a collaborative effort of the diverse working group to achieve project goals.  The purpose of the revegetation treatments implemented over the last six years is to restore the riparian and floodplain environment along Grave Creek within the project area. The intent of the revegetation treatments is to create conditions which support the establishment of diverse plant communities capable of sustaining floodplain ecological processes. These ecological processes include: plant community succession, sediment storage, flood water retention, and long-term channel stability.  Adapting a long-term, adaptive approach to selection and installation of revegetation treatments that support these processes will assist project partners in preserving valuable natural resources in Grave Creek including threatened and sensitive fish species, wildlife, and water quality.  Our restoration approach relies heavily on monitoring and the adaptive management approach.  Because of the length of time required for vegetation to reach maturity and the dynamic nature of riparian ecosystems, restoration of riparian vegetation requires substantial time and considerable patience in order to evaluate the site response to implemented treatments.  At the largest scale, Grave Creek’s potential natural community within the project reach is the Picea/Cornus stolonifera (spruce/red-osier dogwood) habitat type (Geum Environmental 2008a).  There are many remnant areas of this mature state within the project reach.  The desired future condition includes a mosaic of habitat types with later successional stages dominated by the spruce/red-osier dogwood habitat type.  Although this vegetative state has not yet been achieved within constructed areas, monitoring is showing that the revegetation treatments have created conditions necessary to achieve this long-term desired state and that various early successional stages of this and other desirable habitat types are present.   The partners are committed to continuing to monitor site response long-term and use the results to shape future management activities to ensure long-term success of this project.    

Restoration of Therriault Creek

MFWP partnered with The Kootenai River Network (KRN), the USFWS Partners for Wildlife and the local landowner to complete the Therriault Creek Restoration Project in 2004-2005 (Dunnigan et al. 2006).  Prior to the restoration work, the lower section of Therriault Creek was extensively modified through land cover disturbance, riparian vegetation clearing, and physical stream straightening prior to the mid-1900s.  These past activities resulted in an incised stream channel, accelerated bank erosion, channel degradation, and poor fish habitat.  This project reconstructed a total of 9,100 feet of entirely new stream channel which approximately doubled the length of stream, and restored the stream channel dimensions, pattern, and profile that could be sustained through time, with stream channel dimensions remaining relatively unchanged (<10% annually) since the channel work was implemented (Dunnigan et al. 2011).  The original stream channel work achieved the original goals for this project including: 1. Reduction of nonpoint sediment sources within the project area, 2. Elimination of an existing partial fish barrier (perched culvert), 3. Restoration and creation of approximately 55 acres of wetland from converted agricultural land, and 4. Improved quantity and quality of resident fish habitat. 

MFWP monitors fish abundance in Therriault Creek at three sections including two controls sites located above and below the restoration project area, and a single site within the project area (Dunnigan et al. 2011).  Rainbow and brook trout are the most common fish species at all three sites, and bull trout are also encountered at lower abundance during some years.  We compared the abundance of rainbow, brook and bull trout within the restoration project area to control sites located below and above the restoration project using the Before/After/Control/Impact (BACI) statistical design.  The mean difference (Control  - Treatment) in rainbow, brook and bull trout abundance between the lower control site and the project are decreased in each comparison, suggesting that the abundance of each species within the project area decreased after project implementation.  Comparisons were significant using the more conservative two-tailed test for rainbow trout (p = 0.048), and brook trout (p = 0.018), but not significant for bull trout (p = 0.110; Dunnigan et al. 2011).  This trend was nearly the opposite scenario when we used the upper control site as part of this evaluation.  However, these comparisons were not significant (p > 0.05).  The results of a student’s t-test for rainbow trout abundance within the restoration project area, which compare pre and post restoration fish abundance, support the hypothesis that rainbow trout abundance decreased at this site after the project was completed (mean prior = 78.4, mean after = 29.3; p = 0.002 for a two tailed test).  Bull trout abundance within the project site relative to the upper control site also significantly decreased the following project completion (p = 0.005).  The results of a student’s t-test for bull trout abundance within the project area support this observation also (mean prior = 26.0, mean after = 4.1; p = 0.054 for a two tailed test; Dunnigan et al. 2011). 

The project partners acknowledge that the stability of the channel is tied to the structure and composition of riparian vegetation, which provides rooting structure to maintain lateral channel stability by preventing accelerated lateral erosion.  Initial revegetation efforts associated with the restoration work in 2005 included the installation of 5,000 riparian shrubs, 10,000 dormant willow cuttings and seeding of disturbed areas.  However, poor survival of the plants installed in the initial phase of the restoration project, which ultimately resulted from a poor understanding of limiting factors for riparian vegetation at this site, led to the need for additional revegetation work at the site to achieve project goals.  

In 2006 project partners worked collaboratively to develop a riparian revegetation plan for this site (Geum Environmental 2007a).   Data collection for this effort included: identification of existing plant communities and major weed infestation areas; assessment of natural vegetation recruitment potential; and evaluation of ecological processes occurring on the site, which were taken into consideration during the development of implementation strategies, techniques and effectiveness monitoring. 

In 2007, MFWP began implementing the revegetation plan in a multi-year phased effort to convert the project reach to desired riparian plant communities. This included protection of the remaining containerized plants initially installed at the site, and implementation of additional revegetation treatments aimed at restoring site conditions capable of supporting native riparian woody vegetation along the restored Therriault Creek channel.  The specific actions implemented in the fall of 2007 included those actions listed below, and are described in greater detail in Geum Environmental Consulting (2007). 

 1.  Residual Shrub Protection: 

This treatment included protecting approximately 500 surviving shrubs planted during initial (2004 and 2005) restoration activities at the site by installing browse and vole protectors since many of these surviving shrubs had been heavily browsed by ungulates or killed due to girdling by voles.  In addition, most shrubs were being subjected to high levels of competition from aggressive grasses or other invasive species, so 2’ x 2’ brush blankets were installed around the base of each plant.

2.  Containerized Planting:

This treatment included planting a total of 1,000, one-gallon shrubs and 50, five-gallon containerized native shrubs and trees in select locations along the channel (see Geum Environmental 2007b for species list).  Planting was concentrated in ‘islands’ along outer meander bends where woody vegetation was not establishing naturally.  Each one-gallon shrub was protected  from ungulate browse using 4-foot tall rigid plastic browse protector, protected from girdling by voles using 6-inch by 6-inch corrugated pipe fit around the base of the plant and  filled with ~3 inches of bark mulch), and protected from competition with aggressive grasses and other invasive species using a 2-foot x 2-foot photodegradable plastic mulch mat.. 

3.  Solarization Treatments:

The site was dominated by introduced pasture grasses prior to restoration.  The purpose of this treatment was to heat kill these aggressive grasses to open up niches for desirable herbaceous species, such as sedges, rushes and native forbs to establish. . A total of 0.5 acres (0.4 acres temporary and 0.1 acres long-term) within the project reach was treated with solarization. Long-term (up to 3 growing seasons) solarization planting sites, were located on outer meanders and planted through using one-gallon shrubs.  Five temporary and two long-term solarization areas were installed as a pre-planting treatment or as a weed barrier treatment to completely eradicate small areas of invasive species. 

4.  Vegetated Soil Lifts: 

A streambank bioengineering technique, vegetated soil lifts, were installed at two sites along outer meander bends near the upstream end of the project where the risk of channel avulsion is present due to the proximity of the new channel with the abandoned, incised channel.  Vegetated soil lifts are a revegetation and bank construction technique that combines layers of dormant willow cuttings with fabric-wrapped soil to revegetate and stabilize stream banks.  The length of each soil lift is approximately 50 feet (100 feet total).  Approximately 500 live willow cuttings were placed within the soil lifts. 

5.  Live Willow Fascines:

Live willow fascines are willow cuttings tied together to form a linear bundle and anchored.  This treatment was installed at recent depositional areas to promote woody vegetation establishment within and along the channel below the dense aggressive grass rooting zone.  A total of 800 feet of live willow fascines were installed within the project area.  

6.  Large Woody Debris Structures: 

The lower one third of the project transitions to a lower gradient channel type.  To encourage a high late season water table in the floodplain and support the transition from introduced grasses to a more diverse composition of wetland graminoids in the lower portion of the project reach, channel-spanning large woody debris structures were constructed in five locations.  Materials for this aspect of the project included whole logs and trees with limbs, branches and root wads attached. Approximately 20 whole trees were used for this treatment. 

7.  Herbicide Application:

A number of invasive species are present in the project reach that are limiting potential for the site to achieve revegetation goals.  To control invasive species, herbicide has been applied at the site targeting three primary species: Canada thistle, Yellow toadflax and reed canarygrass. 

 In 2008, MFWP completed effectiveness monitoring for all revegetation treatments implemented in 2007. The monitoring methods, results and discussion are reported for each treatment in Geum Environmental (2008c).  Monitoring includes several metrics used to evaluate site response and effectiveness of each treatment including vegetation height, survival, growth metrics, inundation depth (LWD treatment), percent cover by species (vegetated soil lift and coir log treatments), and general observations.  Based on the results of effectiveness monitoring, recommendations for treatment maintenance and the next phase of revegetation treatments were made (Geum Environmental 2008c).   

 In 2009, MFWP completed additional effectiveness monitoring of revegetation treatments installed at the project site and also implemented vegetation maintenance and supplemental revegetation efforts based on adaptive management guided by previous monitoring.  Geum Environmental (2009d) provides a complete summary and discussion of these efforts, but a brief summary is presented below.  Monitoring in 2009 included all metrics used in 2008 and compared these results to the previous year.  Based on the results of 2008 and 2009 effectiveness monitoring, the following maintenance tasks and supplemental revegetation treatments were identified and completed during September and October 2009.  All containerized plants and protected residual shrubs were watered with a minimum of five gallons of water in August and September.  Browse protectors were expanded, re-secured and straightened in all planting units and residual shrub protection areas, and for additional stability, a second four-foot wooden stake was added to each expanded browse protector.   Browse protectors were expanded on about 900 plants that had outgrown previous protectors, and because shrubs installed in 2005 that were protected in 2007 had such a significant growth response, an additional 60 protectors were installed on plants not previously protected.  The solarization fabric from one Temporary Solarization Plot was removed and the area where grass had been effectively killed was re-seeded with a native seed mix consisting of shrubs, grasses, and forbs.  Fabric removed from the plot was placed along the edges of the plot to create a buffer around the newly exposed bare soil, resulting in an additional 2,370 square feet solarization treatment.  Maintenance was completed on the two additional temporary solarization plots and two planted solarization plots, and included re-securing staples and fabric edges, and weeding around plants.   A total of 115 supplemental willow cuttings were installed at seven coir log fascine sites where  willow cutting survival was poor.  Two herbicide applications were completed in 2009 (August and October), directed at controlling sulfur cinquefoil, houndstongue, Canada thistle, reed canarygrass and yellow toadflax.  Based on the results of 2009 effectiveness monitoring, Geum Environmental (2009d) also provided a list of recommendations for additional maintenance and revegetation efforts for 2010. 

 In 2010, MFWP repeated effectiveness monitoring of the revegetation treatments installed at the project site (Geum Environmental 2010).  The following is a brief description of those efforts.  General observations were recorded for all previous revegetation treatments, photopoints of revegetation treatments established in previous years were repeated, survival monitoring data was collected for two containerized planting units and two planted solarization plots, the need for browse control on vegetated soil lifts and coir logs was evaluated, revegetation treatment maintenance needs were document, and the location of a riparian fence constructed by project partners was evaluated and re-alignment options developed. Geum Environmental (2010) presents a detailed summary and discussion of these results.  Based on the monitoring results and adaptive management recommendation, several additional vegetation treatments were implemented in fall 2010.  These treatments included the following.  Browse protectors were installed on all of the remaining 2005 shrubs that could be located and that hadn’t been previously protected.  A total of 1,100 containerized shrubs and trees were planted within twenty-two separate planting units in the lower half of the project reach, including one long-term, solarization site. All planting units were located along outside meander bends to promote long-term channel stability. Species mixes was determined by topography and depth to water table at each site.  Each plant was fitted with a vole protector, brush blanket, and four-foot tall, sixteen-inch diameter browse protector.  A total of 1,580 square feet of solarization fabric was installed at one site to kill reed canarygrass dominated the planting site.  .  A total of 39 plants were planted through the fabric at this site.  Solarization fabric was removed from a single temporary solarization site installed in 2007 and the exposed mineral soil was seeded with a native seed mix.  Two herbicide applications occurred throughout the project area during July and September 2010.  Both applications targeted Canada thistle, reed canarygrass, yellow toadflax, houndstongue and sulphur cinquefoil, and were applied with care to minimize damage non-target species.

 MFWP’s most recent work on Therriault Creek occurred in summer and fall of 2011.  This work included effectiveness monitoring, vegetation maintenance and limited implementation of vegetation treatments.  Effectiveness monitoring in 2011 of vegetation treatments installed prior to 2010 included photo documentation of all treatments, survival monitoring of the four planting units monitored in 2010, qualitative observations of all treatments, and documentation of all maintenance needs.  Monitoring of vegetation treatments installed in 2010 included survival monitoring of approximately 40% of the total number of plants installed, photo documentation of all treatments, and documentation of all maintenance needs.  Minor maintenance of all previously installed treatments was completed in August and included watering all shrubs installed in 2010 and select watering of shrubs installed earlier, browse protector repair and expansion, and solarization treatment maintenance (re-securing fabric edges and loose staples, and hand-pulling weeds from around the base of plant in long-term plots).  Additional revegetation treatments were completed in fall 2011 and were based on the results from previously completed effectiveness monitoring.  Additional revegetation treatments included the following.  One temporary solarization plot was removed and seeded.  Another temporary solarization plot was partially removed, seeded and willow cuttings installed in the exposed soil..  Fabric was removed from two long term solarization plots and the bare mineral soil was seeded.   A single herbicide application was conducted in July 2011 to target infestations of noxious weeds and reed canary grass.  A report that summarizes all monitoring, maintenance and revegetation treatments completed in 2011 will be completed in December 2011. 

MFWP believes that the stream channel restoration work that re-connected the channel with a large area of former floodplain created the hydrologic and geomorphic conditions necessary to sustain ecological processes to support long term establishment and maintenance of a mosaic of riparian plant communities throughout the project area.  The revegetation treatments that have been implemented over the last six years have addressed additional factors limiting revegetation potential and established a diverse number of riparian tree and shrubs species that have shown consistently high survival.  The distribution and density of noxious weeds is declining at the site and diversity of native species in the herbaceous plant communities are increasing. The site is on a trajectory towards achieving the desired future condition necessary to support project goals long-term.  Project partners remain committed to monitoring, maintenance and adaptively managing the site as it responds to the treatments that have been implemented and dynamic watershed inputs. 

Pipe Creek

Pipe Creek is a third order tributary that enters the Kootenai River at river mile 201.  It provides habitat for a migratory population of bull trout, westslope cutthroat trout, rainbow trout, and brook trout.  The lower one mile section of Pipe Creek has been subject to residential and urban infrastructure encroachment, removal of riparian vegetation, inadequate bridge capacities, and channelization.  These activities have increased bank instability and erosion and reduced the quantity and quality of fish habitat within this section on Pipe Creek.  MFWP worked collaboratively with a group of nine landowners on lower Pipe Creek to develop a stream restoration design plan (RDG 2009).  The purpose of the restoration work was to increase stream channel stability, enhance fisheries habitat, reestablish a healthy riparian vegetation community, and maintain existing floodway conveyance.  MFWP implemented the first phase of this project in the fall of 2010, which encompassed approximately 1,100 feet of stream channel.  A summary of the types and quantity of structures installed is presented in Table 2. 

A total of one log vane and seven engineered logjams were installed throughout the project area in 2010.  The log vanes and engineered logjams are both designed to provide bank stabilization by reducing near-bank stress and redirecting flow away from the bank.  The structure was designed to allow fish passage at all flow levels and dissipate stream energy in the form of a downstream scour pool.  A total of six boulder energy dissipater structures were installed throughout the project area.  The design intent of these structures is to reduce flow energy in meanders and redirect flow away from the banks.  These structures were designed to provide interim streambed grade control in run features until natural streambed armoring/sorting processes develop and control long-term vertical stability. A matrix of large, immobile, and irregularly placed boulders forms the foundation of the structure.  Gaps between boulders were filled with smaller, mobile alluvium, thus maintaining bed load transport through the structure.  A total of four boulder grade control structures were installed in this section of Pipe Creek.  The design intent of these structures was to ensure that floodwaters access the floodplain at or near the design bankfull discharge and fish passage is maintained.  The structure was designed to provide interim streambed grade control in glide and riffle features until natural armoring/sorting processes develop and control long-term vertical stability.  This project installed approximately 750 linear feet of single and double vegetated soil lifts, which are structures designed to provide site conditions directly along the stream channel that are suitable for growing riparian vegetation.  Vegetated soil lifts are used in conjunction with other larger bank stabilization structures (engineered logjams and log vanes) to reduce bank erosion rates.  This project also installed a total of two sod brush trenches, which are a revegetation technique used to secure the back edge of sod mats to the edge of the bankfull channel along riffle and run sections of the stream channel.  The brush trench will provide roughness along the channel bank and reduce potential for bank erosion.  During overbank flows, the densely branching willows in the trench will trap sediments and native seed, and provide an environment for seed to germinate and grow. 

 MFWP recognizes that the restoration techniques used in this project require effectiveness monitoring to evaluate project success and to shape future restoration activities.  Therefore we developed a monitoring protocol for this project that included the following methods.  We established photo-points will be established throughout the project area, including all structures.  Photos documentation occurred upon construction and after all bankfull events.  A longitudinal profile survey of the project area was surveyed after construction and will be repeated after each bankfull event the project experiences in order to evaluate vertical stability within the project area.  Cross-sectional surveys throughout the project area will also be completed before and at least five years after construction.  We stratify these surveys into either riffle or pool habitat.  Within the pool type habitats, we measure the mean width, length and maximum bankfull depth, total length and total surface area of all pools within the project area annually.  Within the riffle habitats, we measure the bankfull width, maximum and mean depths, cross sectional area, and slope.  These cross-sectional profiles were surveyed once prior to construction in 2010, and will be resurveyed after project completion and also after the first bankfull event for at least five years annually.  In addition to monitoring the physical dimensions of the stream channel, Montana FWP also remains committed to monitoring the status and abundance of the fisheries resources that inhabit Pipe Creek.  Montana FWP’s current annual monitoring activities in Pipe Creek include two electrofishing sites.  One site is located upstream of the project area, and the other site is located within the project area.  Bull trout redd surveys are also conducted annually in Pipe Creek.  These activities will continue for several years.  The results of these monitoring activities will be included in Project 1995-004-00’s next annual report. 

In addition, as part of the permitting requirements for this project mandated by Lincoln County, MFWP was required to prepare a hydraulic analysis report (RDG 2010) that modeled a 100 year flood event to evaluate if this project would increase flood risk in the area, and evaluate if the engineered log jams would be stable.  The results of this analysis indicate that, for a 1% annual flood discharge of 1,900 cubic feet per second, proposed conditions will decrease the water surface elevation within the project area by more than five feet and intermittently increase the water surface elevation at select locations as a result of changes in the vertical profile required to reconnect the proposed channel with its adjacent floodplain.  Results of an inundation analysis indicate that despite the slight increases in water surface elevations, the overall extent of flood inundation remains generally unchanged and does not increase the flood risk for existing structures adjacent to the study reach.  In addition, basic stability checks for the proposed engineered logjam structures indicated sufficient factors of safety are maintained for failure due to buoyancy, sliding, and scour during the 1% annual flood (RDG 2010).  Despite these results, Lincoln County required MFWP to complete a Letter of Map Revision (LOMR) for submission to FEMA.  The survey work and hydraulic modeling required to complete the LOMR will be completed in fall 2011.  

Objective 1C:

Libby Creek Fish Screen

Libby Creek provides important spawning, rearing, and a migratory corridor for redband, westslope cutthroat, and bull trout.  The largest substantial irrigation diversion located on Libby Creek exists at approximately river mile 13, where two water right holders divert water up to 3.4 cfs of water from Libby Creek through a previously unscreened diversion.  The intent of the project was to install a functional fish screen near the point of diversion capable of delivering the legal volume of water for these two water right holders in order to prevent fish entrainment into the irrigation system. 

Montana FWP worked with the landowner to develop a collaborative project to upgrade the existing system in order to improve ease of operation, eliminate fish entrainment and decrease maintenance at the point of diversion.   The system was designed by the Montana Design and Construction Bureau, the FWP Libby staff and was installed in 2006.  The system consisted of a box control structure, a turbulent fountain screen, and conveyance pipe.  The control structure is an 8’x8’x5’ concrete box with a round 18 inch Waterman screw gate on the intake side, and round 15 inch Waterman screw gate on the screen side.  The screen is a 3 feet diameter turbulent fountain constructed of 20 squares per inch stainless steel mesh, that is  enclosed in a 6’x6’x6’steel catch box.  The conveyance system consists of 220 feet of 18 inch diameter corrugated metal pipe above the control structure, 200 feet of corrugated HDPE pipe between the control structure and the screen, and 500 feet of 12 inch diameter HDPE pipe from the screen to the irrigation ditch.

Young Creek Fish Screen

Young Creek is a 17 km long tributary to Koocanusa Reservoir, 5 km south of the Montana-British Columbia border that drains a 119 km² basin of the Purcell Mountains.  Young Creek is one of the most important westslope cutthroat trout spawning tributaries to Koocanusa Reservoir because it represents one of the last known genetically pure populations of westslope cutthroat trout in the US portion of the reservoir and is one of the most potentially productive tributary streams upstream of Libby Dam.  Brook trout also reside in Young Creek, and although bull trout are not known to spawn in Young Creek, juvenile bull trout rearing in the reservoir occasionally enter lower Young Creek.  The largest irrigation diversion on Young Creek is located approximately 1.7 miles upstream from Libby Reservoir and diverts a maximum of approximately 8 cfs of water from Young Creek.  This system was originally constructed in the 1970s and lacked a functioning fish screen.  The system consisted of a diversion structure with headgate that diverted water into a 480-foot long open ditch, and then feed two irrigation pipes.  This system represented the largest single loss of fish due to entrainment within the drainage. 

MFWP worked with the irrigation user group to develop plans to upgrade the existing system in order to improve the headgate, install a fish screen that would eliminate fish entrainment of all age classes and replace the open ditch with buried pipe.  The system was designed by the Montana FWP Design and Construction Bureau and was installed in the spring of 2009.  The old headgate and diversion structure was replaced with a rock structure and a new waterman 30-inch diameter headgate valve and delivery pipe was installed to deliver water to the fish screen.  The fish screen was 6-foot diameter turbulent fountain fish screen with a 20 mesh per inch screen size, and was buried at ground level.  A 12-inch diameter fish return line routed screened fish back into Young Creek.  A 24-inch diameter and 15-inch diameter water delivery pipes exit the fish screen and connect to the existing lines.  These pipes were buried and eliminated the previous open ditch system.  The new screen and associated system conserves water, prevents fish entrainment and requires less maintenance. 

Deep Creek Fish Screen

Deep Creek is third order 17.7 km long tributary that enters the Tobacco River at Rkm 32.8, draining approximately 48 km² of the Whitefish Mountains.  The largest irrigation diversion on Deep Creek is located approximately 1.7 miles upstream from the confluence, did not have a functional fish screen and represented the largest single loss of fish due to entrainment within this watershed.  Montana FWP worked with the landowner to develop a collaborative project to upgrade the existing system in order to improve ease of operation, eliminate fish entrainment and decrease maintenance at the point of diversion.   The system was designed by the Montana FWP Libby staff and was installed in the spring of 2010.  This project installed a new trash rack in front of the existing headgate, removed 26 feet of existing 18 inch diameter corrugated metal pipe and installed a prefabricated fish screen and 16 feet of new 18 inch diameter CMP.  The fish screen structure consisted of a 4-foot diameter turbulent fountain fish screen with a 20 mesh per inch screen size, and was buried at ground level.  A 12-inch diameter fish return line routed screened fish back into Deep Creek approximately 40 feet away from the screen structure.  The screen was also fitted with a hinged cover for safety purposes.  The new screen and associated system prevents fish entrainment and requires less maintenance.

Objectives 2A and 2B:

Kilbrennan Lake

Kilbrennan Lake is located approximately 10 miles north of Troy, and has a surface area of 57 acres and a maximum depth of 22 feet.  Feeder Creek and a small, unnamed spring at the south end of the lake are the only major sources of surface water entering the lake.  Kilbrennan Creek flows out of the lake through private and public land for about 4 miles to the Yaak River.  FWP believes that redband trout were likely the dominant salmonid species historically present in Kilbrennan Lake.  However, black bullheads were illegally planted in Kilbrennan Lake sometime in the late 1960s or early 1970s, and since that time black bullheads have become a dominant species in the lake.  Yellow perch were illegally introduced into Kilbrennan Lake in approximately 1995.  The presence of these two exotic species has dramatically changed the species composition and population dynamics within the lake.  In 2006, brook trout and rainbow trout numbers and decreased to point of near eradication, while stunted populations of black bullhead and yellow perch were very abundant.  As a result of the change in species composition within Kilbrennan Lake, angler use of the lake had also declined.  In 1991, FWP estimated that Kilbrennan Lake received 1,391 angler-days per year (McFarland 1992), but in 2003, angling pressure on Kilbrennan Lake dropped to 240 angler-days per year (McFarland 2004).

The objectives of this project were to expand the current distribution of redband trout in the Kootenai River Subbasin while restoring a popular recreational fishery.  The goals of this project were to construct a fish barrier on Kilbrennan Creek downstream of Kilbrennan Lake, remove all fish residing in Kilbrennan Lake, Feeder Creek, and Kilbrennan Creek downstream to the constructed barrier and restock these waters with genetically pure redband trout.

The first phase of this project included enhancing the spawning habitat in Feeder Creek and the unnamed spring and constructing a fish barrier on Kilbrennan Creek approximately 0.8 miles downstream of the lake, in order to prevent non-native fish species from re-colonizing after treatment.  These activities were completed in mid-October 2006.  MFWP applied approximately 841 gallons of 5% Prenfish to Kilbrennan Lake using a combination of venturi pumps from motor boats and backpack sprayers to achieve an estimated 4.9 ppm rotenone concentration within the lake, Feeder Creek and the unnamed spring entering the lake.  We also operated a detoxification station located at the constructed fish barrier continuously from October 31, 2006 to March 9, 2007.  During this period we added 3,217 pounds of dry potassium permanganate directly to Kilbrennan Creek to neutralize the rotenone using an auger powered by a 4000-watt generator.  Kilbrennan Creek was detoxified on March 9, 2007.  The MFWP Hatchery Division (not related to Project 1995-004-00) restocked the lake with redband trout in the early summer of 2007.   

 MFWP last gillnetted Kilbrennan Lake in June 2011, and the average catch per net was 14 redband trout, with an average total length of 224 mm.  MFWP estimated that Kilbrennan Lake received 4,107 angler-days of pressure in 2009 (McFarland and Dykstra 2010).  MFWP conducted redd counts and electrofishing surveys in Feeder Creek and the unnamed spring to determine if the hatchery origin redband trout are successfully spawning in these areas and we concluded that natural reproduction is occurring. 

Loon Lake

Loon Lake is located in the Pipe Creek watershed, approximately 15 miles northwest of Libby, Montana, and has a surface area of 33 acres and a maximum depth of 25 feet.  Westslope cutthroat trout were likely the dominant salmonid species historically present in Loon Lake.  However, prior to this project, black bullheads were illegally introduced in the late 1960s and MFWP stocked brook trout in the lake on seven occasions from 1948-1961.  The objectives of this project were to expand the current distribution within the historic range of westslope cutthroat trout in the Kootenai River Subbasin while improving angling opportunity within the Pipe Creek watershed.  The goals of this project were to remove all fish residing in Loon Lake and the unnamed outlet stream downstream to the existing natural barrier falls and restock these waters with genetically pure westslope cutthroat trout.

MFWP applied 285 gallons of the CFT-Legumine rotenone formulation to Loon Lake in September 2007 using venture and trash pumps on a small motorboat and backpack sprayers around the lake margins to achieve an overall lake concentration of 5.1 ppm rotenone to remove the brook trout and black bullheads inhabiting the lake.  We treated the tributaries of the lake and outlet stream with powdered rotenone to remove fish from these waters and ensure fish could not use these areas as refugia.  In addition, we applied CFT-Legumine the outlet stream proper using a combination of drip stations and backpack sprayers to achieve an overall concentration of 2 ppm rotenone.  Rotenone treatments occurred downstream to an existing barrier on the outlet stream approximately 1 mile upstream from the confluence with Pipe Creek.  In order to prevent the rotenone laden water flowing out of Loon Lake from reaching Pipe Creek, we operated a detoxification station located near an existing fish barrier continuously from September 19 to October 19, 2007.  During this period we added 272.1 kg of dry potassium permanganate directly to the outlet stream to neutralize the rotenone.  We used bioassays to determine when the lake effluent contained non-lethal levels of rotenone, and determined the lake was capable of supporting trout on November 30, 2007.  We used gillnetting and backpack electrofishing in the spring of 2008 to ensure that all fish had been removed from the lake and outlet stream, respectively.  These surveys found no living fish in either waters.  The MFWP Hatchery Division (not related to Project 1995-004-00) began stocking the lake annually with age 0 westslope cutthroat trout in 2008.

MFWP last gillnetted Loon Lake in June 2011, and the average catch per net was 29 cutthroat trout, with an average total length of 269 mm.  MFWP estimated that Loon Lake received only 82 angler-days per year in 2003 (McFarland 2004), but angling pressure on Loon Lake increased to 3,095 angler-days per year in 2009 (McFarland and Dykstra 2010).

Boulder Lake/Creek

Boulder Lake is located approximately 15 miles southwest of Eureka, Montana, and has a surface area of six acres and a maximum depth of approximately 13 feet.  Boulder Creek begins at the outlet of Boulder Lake and flows approximately eight miles across public land (USFS) before flowing into Lake Koocanusa.  The Boulder Creek watershed was likely historically fishless due primarily to the presence of a natural falls barrier located approximately 2 miles upstream from Lake Koocanusa.  Montana FWP stocked Boulder Lake in 1954 with an undesignated strain of cutthroat trout, and Lower Boulder Lake was stocked the following year with a similar group of fish.  Boulder Creek was stocked with rainbow trout in 1944 and once with an undesignated strain of cutthroat trout in 1946.  These single stocking events resulted in self sustaining populations with individuals containing characteristics from Yellowstone, westslope cutthroat and rainbow trout ancestry. 

The objectives of this project were to expand the current distribution within the historic range of westslope cutthroat trout in the Kootenai River Subbasin while continuing to provide angling opportunity within the Boulder Creek watershed.  The goals of this project were to remove all fish residing in Boulder Lake and Boulder Creek from the lake outlet downstream to the existing natural barrier falls and restock these waters with genetically pure westslope cutthroat trout. 

In September 2009, Montana FWP airlifted a 12-foot long aluminum boat, motor, piscicide and dispersal equipment into the lake using a helicopter operated by a Montana FWP pilot.  We distributed 10.5 gallons of CFT-Legumine, a commercial formulation that contains 5% rotenone as the active ingredient, to Boulder Lake using a venturi pump system from the motor boat to achieve 1 part per million concentration within the lake.  We used a continuous drip device and packets of powdered rotenone to remove all fish from the inlet stream.  We also treated Boulder Creek using a combination of eight drip stations and backpack sprayers to achieve a target concentration of 1 ppm.  We used caged cutthroat trout to measure the toxicity of the water in the lake and creek to ensure the objectives were met.  Although we were prepared to use a detoxification station below the barrier falls to neutralize the piscicide, we determined that no active rotenone had reached this site.         

Effectiveness monitoring in June 2010 that included electrofishing in the creek and gillnetting in the lake, indicated that the single application achieved a complete removal of all fish within the project area.  The MFWP Hatchery Division (not related to Project 1995-004-00) stocked the lake and creek with westslope cutthroat trout fry in July 2010.  Stocking will continue for at least one to two additional years.  We will continue to monitor and evaluate growth and relative survival of the hatchery over the next several years.  This project expanded the distribution of genetically pure westslope cutthroat trout in the Montana portion of the Kootenai watershed upstream of Libby Dam by approximately twenty percent.  The stocked hatchery cutthroat trout fry will be catchable sized fish next year, and should be sufficiently abundant to support a viable recreational fishery in both the lake and the creek.  

Objective 3:

Idaho Fertilization Project Control Site Monitoring

MFWP collaborates and participates in the Idaho Fertilization Project (Projects 1988-065-00 and 1995-049-00) by participating in planning meetings and conducting field monitoring at a single control site (KR10) located on the Kootenai River located approximately 6 river miles upstream from the fertilization addition site.  We coordinate closely with Idaho Department of Fish and Game and Kootenai Tribe of Idaho project personnel in our annual monitoring, adhering to the  sampling protocols established by our three agencies (Gidley 2009).  We have collected data at this site annually since 2002 using jet boat electrofishing to capture fish at six replicated sites at this location.  MFWP completes the field inventories, converts the field data into electronic format, summarizes the data, presses a sub-sample of mountain whitefish and other trout scales, and delivers these data to the Idaho Fish and Game Project Biologist for analysis.  A summary of mean catch per unit of effort for the six most commonly observed fish species at this site is presented in Table 3.  Rainbow trout catch rates at this site have exhibited a significant positive trend since 2002 (r² = 0.65; p = 0.005).  The other five most common species have not exhibited a significant trend (r² < 0.37; p > 0.05). 

 Objective 4 and 5:

Resident Fish Population Trend/Status Monitoring: 

MFWP also conducts monitoring (funded through Project 1995-004-00) to assess the status and trend of several resident populations in the Kootenai Basin.  The following is a brief description and summary of those monitoring efforts.

MFWP annually monitors fish species composition, and species size and relative abundance within Koocanusa Reservoir using spring and fall gill netting.  Dunnigan et al. (2011) present these results and trend analyses for 11 fish species.  The following is a brief summary of these results.  Kokanee mean length is significantly negatively correlated to catch rate in the fall nets.  The catch rates of cutthroat trout and mountain whitefish during the past several years has remained low and not differed significantly from a stable population, but rainbow trout catch rates have exhibited a significantly increasing trend since 1994.  Bull trout catch rates on Libby Reservoir peaked in 2000 at 6.71 bull trout per net, and have generally exhibited a declining trend since.  Catch rates of inland rainbow trout in the fall gillnets has been relatively low since these fish were introduced into the reservoir.  However, the catch rate for these fish is significantly and positively correlated with the number of hatchery Inland rainbow trout stocked in the reservoir the previous year, especially yearling fish releases from 1989 to present. 

MFWP has also monitored zooplankton species composition, abundance and size of zooplankton within the reservoir since the construction and filling of Libby Dam (Dunnigan et al. 2011).  Zooplankton abundance, species composition, and size distribution have also all been relatively stable during the second half of the reservoir’s history.  Cyclops has been the most abundant genera of zooplankton present in the reservoir since 1997, and Daphnia was the second most abundant genera of zooplankton within the reservoir in most years.  Annual zooplankton abundance within the reservoir varies by month, with the monthly abundance peaks over the past ten years remaining relatively consistent.   Trends in zooplankton abundance in Koocanusa Reservoir have generally been of decreasing abundance of the larger and more numerous genera of zooplankton and an increasing abundance of smaller zooplankton since the late 1970s.  Two of the most abundant and larger zooplankton, Daphnia and Diaptomus have both negatively decreased in mean annual abundance since 1977, while the smaller zooplankton, Bosmina and Cyclops have both significantly increased in abundance since reservoir.  This increase in the smaller zooplankton (especially Cyclops) was primarily responsible for the observed overall significant increase in total zooplankton since 1977 in Koocanusa Reservoir (Dunnigan et al. 2007) 

MFWP conducts annual bull trout redd counts in all nine known migratory bull trout populations in the Montana portion of the Kootenai Basin.  Dunnigan et al. (2011) presents these data and provides additional discussion.  However, the following is a brief summary of recent results.  Grave Creek and the Wigwam River have both exhibited significant positive trends since the mid 1990s.  However, both populations have decreased in recent years.  Bull trout core areas in the Kootenai River downstream of Libby Dam include Quartz, Pipe, Bear (Libby Creek drainage), O’Brien creeks and the West Fisher Creek.  Bull trout redd counts within these individual core streams have been variable over the past several years, and have not increased in proportion to bull trout redd counts upstream of Libby Dam.  Three of the four populations between Libby Dam and Kootenai Falls have been below average in recent years.  MFWP also monitors fine sediment (<6.35 mm) levels annually in eight bull trout spawning tributaries within the Montana portion of the Kootenai Basin using core sampling as an index of potential incubation/emergence success.  Dunnigan et al. (2011) presents the most recent results from these efforts. 

MFWP has used nighttime jetboat electrofishing to conduct annual mark recapture population estimates of adult bull trout in the Libby Dam tailrace since 2004 (Dunnigan et al. 2004; 2006; 2007; 2008; 2009; 2010; 2011).  Estimates are conducted during April or May within this 3.5-mile section of the Kootenai River, and have ranged from 176 bull trout in 2006 to 918 bull trout in 2004 (Table 4).  We collected tissue and scale samples from all bull trout we handled, and marked all fish with PIT tags, which allowed us to obtain capture histories and estimate growth across years for many fish (Dunnigan et al. 2007; 2008; 2009; 2010; 2011). 

Bull/Brook Trout Genetic Analyzes in Bull Trout Natal Streams

In order to determine if there is evidence of hybridization between bull trout and brook trout, in the four core bull trout streams in the Montana portion of the Kootenai Subbasin that also contained brook trout, and develop a baseline dataset for future trend monitoring, we randomly collected fin clips from approximately 50 Salvelinus spp.  from the bull trout juvenile monitoring sections (see Hensler and Benson 2006) from Keeler, Pipe, West Fisher, and O’Brien creeks for genetic analyses using a backpack electrofisher.  The remaining bull trout core streams in the Montana portion of the Kootenai Subbasin that do not contain brook trout (Grave, Quartz, Libby, and Bear creeks) were previously genetically tested and found no evidence of hybridization with brook trout (Ardren et al. 2007).  DNA was extracted from fin clips, and using the polymerase chain reaction (PCR), each fish’s genotype was determined at seven microsatellite loci that distinguish bull trout from brook trout (Dunnigan et al. 2008).  We also calculated a hybrid index for each stream.

All four streams contained hybridized fish, with hybridization especially problematic in O’Brien and Keeler creeks.  However, bull trout were rare in Pipe Creek.  Overall, we were able to correctly visually identify 91% of the fish collected for analysis.  The most common visual misidentification involved misidentification of later generation hybrids as being either brook trout or bull trout (Dunnigan et al. 2008).

Bull trout entrainment and genetics within the Kootenai/ay Basin

Montana FWP initiated a study in the summer of 2005 to investigate the genetic structure of bull trout within the Kootenai Basin using genetic analysis (Ardren et al. 2007) with the objective to examine how genetic diversity was distributed both within and among bull trout populations in the Kootenai River Basin, and ultimately to develop a methodology to use genetic techniques to assign natal tributary to individual bull trout of unknown origin.  This study was initiated after a similar proposed study using rare earth elemental signatures on scales and otoliths proved non-workable. 

Montana FWP collected juvenile bull trout tissue samples from 15 tributaries located in British Columbia and Montana for this study.  DNA was extracted from all samples, and then all individuals were genotyped at a core set of 12 microsatellite loci (Ardren et al. 2007).  Adren et al. (2007) presented measures of genetic diversity including numbers of alleles and observed and expected heterozygosity for each tributary population.  Pairwise estimates of Fst among all population pairs were also calculated and tested for statistical significance, and a neighbor joining tree based on Cavalli-Sforza and Edwards chord distance was generated.

Ardren et al. (2007) performed a jackknife on the baseline dataset from the Kootenai River Basin using a maximum likelihood estimator in order to assess the ability to correctly assign unknown fish to their population of origin, and found that we were able to assign unknown fish captured in the Kootenai River as originating either upstream or downstream of Libby Dam within an overall accuracy level of 95% or greater.  When we applied these techniques to the 62 unknown adults collected from below Libby Dam  in 2004, exactly half of these individuals assigned having originated upstream of Libby Dam (Ardren et al. 2007). 

DeHaan et al. (2008) continued to assign origin of 262 individual adult bull trout collected below Libby Dam to natal tributary for fish collected during our bull trout population estimate from 2004-2007 (see above).  We estimated that the proportion of fish originating above Libby Dam ranged from 62.7% in 2004 to 49.2% in 2006.  Although there was a high degree of accuracy for assigning individuals to above or below Libby Dam, our assignment accuracy to individual tributary was somewhat lower,  (e.g. Skookumchuck River; ), the rate of correct assignment to other populations was much lower (e.g. West Fisher; 44% accuracy). These results prompted further investigation in 2009-2010. 

 DeHaan and Adams (2011) attempted to increase assignment accuracy by increasing the number of diagnostic loci used during the genotyping to 16 and comparing alternative computer software (Bayesian approach) used to assign individuals to each tributary or group (above/below Libby Dam).  We also collected additional juvenile fish from several of the tributaries to help determine if the genetic signatures were stable through time and increase the sample size of our baseline dataset. 

 DeHaan and Adams (2011) concluded that assignment accuracy increased as a result of the additional 4 loci, and they observed 90% or greater correct assignment to population of origin for most populations in the leave-one-out assignment tests.  Furthermore, for most populations, the two analysis methods predicted similar assignment accuracy based on leave-one-out tests.  However, they chose to use WHICHRUN (maximum likelihood approach) for genetic assignments because there was little difference in assignment success between approaches in the leave-one-out tests, WHICHRUN had been used for previous genetic assignments (Ardren et al. 2007; DeHaan et al. 2008), and WHICHRUN had a slightly greater assignment success rate for the blind samples.  DeHaan and Adams (2011) further suggested that there may be temporal genetic variation within some of the baseline tributaries, which may result from small populations with low effective size or varying rates of gene flow among populations from one year to the next.  This issue should be investigated if additional study is proposed.  When we applied the genetic assignments to above/below Libby Dam for those fish collected during our population estimates below Libby Dam, we estimated that approximately 64% of the fish present in both 2008 and 2009 originated above Libby Dam. 

Burbot monitoring in Kootenai River and Koocanusa Reservoir

MFWP has annually monitored the relative abundance of burbot in the stilling basin below Libby Dam using hoop traps since 1994 (Dunnigan et al. 2004; 2005; 2006; 2007; 2008; 2009; 2010; 2011), but catch rates have declined significantly since.   The highest catch rates occurred in 1995-96 and 1996-97.  We speculate that burbot abundance in the Kootenai River directly below Libby Dam is strongly influenced by entrainment.   

MFWP began monitoring the relative abundance of burbot in Koocanusa Reservoir using baited hoop traps during the 2003/2004 winter (Dunnigan et al. 2004; 2005).  However, instead of annual monitoring we utilize a pulse monitoring strategy, where we fish randomly determined locations throughout the reservoir during the winter months for two consecutive years followed by a period of two years with no monitoring.   We observed a mean catch per unit effort of 0.064 burbot/trap-day during the 2003/2004 season (Dunnigan et al. 2004), which was similar to the mean catch rate we observed during the 2004/2005 season (Dunnigan et al. 2005).  During the winter of 2007/2008, we fished 702 trap-days and captured 39 burbot for an overall catch rate of 0.057 burbot per trap-day (Dunnigan et al. 2009).   However, during the 2008/2009 season we fished a total of  612 trap-days and captured a total of 62 burbot during the trapping season for an overall mean catch-per-unit-effort for the season of 0.103 burbot/trap/day, which was significantly higher than the previous year (Dunnigan et al. 2010).  Trap depth was significantly correlated to burbot catch rates during all years. 

In addition to trapping, we also used radio and sonic telemetry during the 2003/2004 and 2004/2005 winters to quantify burbot movements and home range (Dunnigan and Sinclair 2008).  Fish were marked from November 2003 to April 2004, and tracking efforts were conducted approximately weekly during daylight hours on Koocanusa Reservoir using a motor boat during the period of mid December 2003 to July 2005, which spanned two burbot spawning seasons.  Six (15%) tagged burbot were never relocated after release, and an additional seven (17.5%) of the tagged burbot either died or shed their tag within 60 days after release.  We relocated 34 marked burbot (855%) on at least 3 occasions (mean = 10.0).  Burbot frequently used the deepest portions of the reservoir, especially during the summer months.  We estimated that the mean 50, 75 and 90% kernel home ranges were 14.6 km², 22.6 km² and 32.3 km², respectively.  We found no evidence that burbot home range differed within the two years of this study.  Although we did observe burbot moving a maximum distance of up to 64.3 km, we were unable to discern any general movement patterns by month or capture location.  Individual burbot tagged in this study had relatively high fidelity to the original capture side of the reservoir, and almost half of our marked burbot were never relocated on the opposite side of the Kootenai River thalweg, and those that did cross the old thalweg do so infrequently.  One burbot was entrained through Libby Dam approximately 134-276 days after capture and tagging, respectively.   

Creel survey

Montana FWP designed and implemented a creel survey to estimate fishing effort, catch and harvest of trout in the Kootenai River downstream of Libby Dam during the 2009/2010 fishing season which included the period June 1, 2009 to March 31, 2010 (Dunnigan et al. 2011).  This creel survey targeted mostly crepuscular trophy sized rainbow and bull trout fishery in the tailrace area.  We conducted angler interviews to estimate angling success, and we conducted visual counts of boat and bank anglers to estimate fishing effort (pressure).  The majority of the fishing effort was concentrated near Libby Dam downstream to David Thompson Bridge, followed in decreasing order by the Dredge Cut Area, the Dunn Creek boat ramp area, and the Alexander Creek Campground area.  The most common type of terminal gear used by bank anglers were lures, followed in descending order by bait, combination bait and lure, and artificial fly.  The vast majority of the bank anglers were residents of Lincoln County.   Bank angler effort differed by month, with the highest effort occurring in July, and the lowest effort occurring during November.  The total effort for the season was 4,079 hours (1,467 trips).  Bank angler catch rates of rainbow trout > 24 inches were low, averaging only 0.007 fish/hour (151 hours/fish).  Harvest rates of rainbow trout > 24 inches were similar to catch rates, indicating most fish angled in this size class by bank angler were harvested.  Bank angler bull trout catch rates were relatively high, and averaged 0.045 bull trout/hour (22 hours/fish).  We estimated that bank anglers caught a total of 27 rainbow trout >24 inches and 185 bull trout during the season.  Boat angler effort was substantially lower than bank effort, but generally showed a similar pattern.   Boat effort was lowest from September through December, but increased to the highest effort in January to the end of the season in March.  Total boat effort for the season was 262 boat hours (411 boat angler hours), which represented 74 boat trips.  Boat angler catch rates of rainbow trout > 24 inches averaged 0.020 fish per boat hour (77 hours/fish).  The estimated total catch and harvest for the season of rainbow trout > 24 inches was relatively low (5 and 3 fish, respectively).  Bull trout catch rate for boats averaged 0.151 fish per boat hour (11 hours/fish).  We estimated boats angler caught 39 bull trout during the season. 

Spill Monitoring

Spill at Libby Dam has been an infrequent event since the fourth turbine unit went online in 1976.  However, The Federal Action Agencies have used spill at Libby Dam used spill at Libby Dam in 2002, 2006 and 2010 with the intent to benefit the Kootenai River white sturgeon.  MFWP monitored fish response to supersaturated gas conditions during each of these events.  During the 2002 spill event, we used three general approaches including the examination of captive fish and fish captured via electrofishing for signs of gas bubble disease, and radio telemetry to assess fish displacement and behavior changes (Dunnigan et al. 2003).  The following points briefly summarize the results of the 2002 spill event. Signs of gas bubble disease developed rapidly in the captive fish, and quickly escalated to 100% incidence, relative to fish captured via nighttime electrofishing. Approximately 86% of the rainbow trout 80% of the bull trout and 31% of the mountain whitefish collected via electrofishing during the peak total discharge and spill at Libby Dam exhibited signs of gas bubble disease.  Results from the radio telemetry work suggests that most radio tagged rainbow trout, bull trout, or mountain whitefish did not move substantially during the spill activities at Libby Dam, and remained within the general vicinity of Libby Dam (RM 221.7) downstream to Dunn Creek (RM 219.8), with the center of gravity more near Libby Dam. We developed relatively simple models to predict fish response to spill.     

Marotz et al. (2007) summarizes the results of our monitoring during the 2006 spill event at Libby Dam.  We visually examined fish captured by nighttime electrofishing for external symptoms of gas bubble trauma (GBT).  Fish captured along the right and left banks were recorded separately and correlated with varying exposures to supersaturated water. GBT was observed in rainbow trout, westslope cutthroat trout, kokanee, bull trout and mountain whitefish. Symptoms in trout were observed on the fourth day of spill and increased in frequency as spill continued.  GBT was greater on along the left downstream bank where gas supersaturation was greatest.  After 11 days of spill, all bull trout and westslope cutthroat trout captured had GBT, including multiple hemorrhages on the ventral surface of the body, bubbles in fins, eyes, dermis on the operculum and split fins. Hemorrhaging on the ventral body surface increased when gas saturation approached 131 percent, then apparently reduced when dissolved gas concentrations reduced toward 124 percent. Observed frequency of GBT in mountain whitefish and rainbow trout increased to 92 to 93 percent. Population estimates before and after the 2006 spill event did not detect impacts to trout populations in the Kootenai River. Comparisons of length frequencies and recaptures of tagged fish indicated that few if any fish were displaced downstream during the high discharge event.

Dunnigan et al. (2011) summarizes the results of our monitoring during the 2010 spill event at Libby Dam.  The Federal Action Agencies conducted a spill test in June 2010 at Libby Dam that lasted seven days which was intended to benefit the Kootenai River white sturgeon.  Discharge from the turbines at Libby Dam was held constant at 27,000 cubic feet per second (cfs) throughout the spill test.  Spill discharge peaked at 9,000 cfs on June 15, 2010 for two hours at 36,000 cfs total discharge from Libby Dam.  Our monitoring effort in 2010 differed slightly from the 2002 and 2006 spill events, in that we focused much of our effort on attempting to locate dead or morbid fish attributable to the elevated total dissolved gas conditions on the Kootenai River immediately downstream of Libby Dam.  We conducted day and night visual searches for dead or dying fish, expending a total effort of 103.5 boat-hours.  We did not observe any fish mortality attributable to elevated gas levels.  However, we did recover five species of fish, whose deaths using our visual criteria could not be attributed to gas-related injuries. In an effort to estimate search efficiency of dead or morbid fish, we released a total of 39 dead and individually marked bull trout in the Kootenai River.  We recovered a total of 12 (30.8%) bull trout during our search efforts.  The spatial recovery pattern of the test fish was not randomly distributed, most of the relocated test fish were recovered on the river bottom of the back eddy associated with the pool located near Big Bend (RM 217.4).  The visual recovery of test fish was likely biased towards larger individuals during daylight hours.  Montana FWP captured fish via jetboat electrofishing on two occasions after spill had ceased in order to determine if fish exhibited symptoms of gas bubble trauma.  The day after spill had ceased, we estimated that 26.5% of the mountain whitefish examined had GBT symptoms.  We also captured two rainbow trout, but none of these fish exhibited GBT symptoms.  Almost six days after spill activities had ceased at Libby Dam, we captured and examined rainbow trout, bull trout, mountain whitefish, kokanee salmon, and brook trout.   However, none of these fish exhibited readily apparent external GBT symptoms.  We also present fish population estimates derived from mark recapture electrofishing for rainbow trout on three sections of the river and bull trout from a single section located immediately downstream of Libby Dam.

Macroinvertebrate Ecology in the Kootenai River below Libby Dam

Changes in hydro operations at Libby Dam during the previous decade and the presence of the nuisance diatom Didymosphenia geminata  prompted MFWP to initiate a study in 2005 (Marshall 2007).  The goals of this study were to describe the differences among the macroinvertebrate community in the Kootenai River downstream of Libby Dam communities since two previous studies were completed in the 1980s and mid 1990s (Perry and Huston. 1983, Hauer and Stanford 1997), and to quantify the effects of D. geminata on the benthos.  The study collected benthic macroinvertebrate and algae samples from five locations below Libby Dam, along with ancillary habitat data.  

This study kept the sites and methods consistent with earlier river surveys (Perry and Huston. 1983, Hauer and Stanford 1997), but used a flow-standardized sampling design to quantify downstream gradients in community structure below the dam, and quantify the effects of D. geminata—independent of dam-related station effects.

Marshall (2007) found that Libby Dam continues to exert a major influence on the structure and function of downstream benthic food webs. However, community dissimilarity analysis suggested that the direct influence of Libby Dam was somewhat less intense than indicated by sampling in the 1990’s (Hauer and Stanford 1997).   Marshall (2007) concluded that a combinations of factors including (1) landscape or climate changes unrelated to the dam operations, (2) homogenizing habitat quality by D. geminata coating the substrata and reducing diversity in community composition, (3) changing river-flow management, and (4) the sample design that sampled the same flow velocity at all sites (which differed from earlier studies ) reduced the effect of dam-related changes in benthic community structure. Modeling results found that for most invertebrate response variables, could be significantly explained by habitat variables, but the average Ashfree- dry-mass (AFDM) of epilithon (which consisted mostly of D. geminata) explained more variation in any other habitat variable, including flow, and after AFDM variation was explained,  the site effect was no longer significant.  These results indicate that many aspects of community structure were strongly correlated with the thickness of epilithic biofilms.  

This study found that functional groups including shredders and scrapers, which feed on biofilm, were less abundant in areas with moderate D. geminata growth, and developed a nonlinear regression to fit a dose-response curve to the abundance of scrapers and shredders relative to the mass of epilithic biofilms.  Marshall (2007) also identified a threshold level of 3-5mg/cm² for epilithic biofilms required to maintain natural ecosystem function, and that concentrations of organic material exceeding 8mg/cm² completely excluded shredders and dramatically reduced scrapers.

Redband identification using non-lethal DNA analysis

Native inland populations of rainbow trout, particularly resident populations, often hybridize with introduced populations of the widely-cultured coastal, form of the species. The inland and coastal subspecies genetically differ from each other by allozyme polymorphisms at lactate dehdyrogenase (LDH-B2* ) and superoxide dismutase (sSOD-1* ) loci which can be detected using protein electrophoresis. However the protein electrophoresis methods require whole lethal fish samples and preservation of tissue samples.  Both of these requirements have limitations.  In addition, fewer laboratories are now using allozyme technology and most genetic studies from wild organisms are now being conducted using DNA rather than protein analyses.  Project 1995-004-00 collaborated in an effort to identify the single nucleotide polymorphism (SNP) differences responsible for the protein variations by sequencing the cDNAs for the LDH-B2* and sSOD-1* genes in a large number of individuals whose genotypes were also determined by protein electrophoresis (Brunelli et al. 2008). The genetic differences causing the allozyme polymorphisms were converted into SNP allelic discrimination assays.  This methodology allows simple, efficient tests using smaller non-lethal tissue samples (fin clips) that are much easier to preserve in the field to be used in a large number of laboratories as aids in assessing levels of hybridization between inland and coastal rainbow trout.  This effort also allows DNA studies to be more directly related to previous allozyme studies.

 Impacts of Kootenai River discharge regulation on resident fishes

In 2010, Project 1995-004-00 collaborated on a study to quantify the impacts of alternative flow management strategies at Libby Dam on resident fish habitat in Kootenai River (Muhlfeld et al. 2011).  Seasonal and life-stage specific habitat suitability criteria were combined with a one-dimensional habitat simulation model to assess how discharge affects the amount of usable habitat of bull and rainbow trout.  Telemetry data used to construct seasonal habitat suitability curves revealed that subadult bull trout move to shallow, low-velocity, shoreline areas at night, which are most sensitive to flow fluctuations.  Habitat time series analyses comparing the natural flow regime (e.g., pre-dam, 1911-1971) to four post-dam flow management strategies (1972-2009) supports that natural flow conditions optimize critical habitats and that the current strategy (e.g., Mainstem Amendments, 2003-2009) best resembles natural flow conditions of all post-dam periods.  Late summer flow augmentation for anadromous fish recovery in the lower Columbia River Basin, however, produces higher variable flows, which reduces the availability of useable habitat during this critical growing season. These results suggest that past flow management policies that created sporadic flow fluctuations were likely detrimental to native trout populations, and that natural flow management strategies that optimize and stabilize the availability of channel margin habitats are likely beneficial to the recovery and conservation of native salmonids in the Kootenai River Basin.