The ISRP requested a succinct summary of the fish rearing program for Lake Roosevelt since it involves three projects that rear fish, and a fourth project that is responsible for evaluating post-release survival, growth, and harvest.Sponsors of the Spokane Tribal Hatchery (1991-046-00), Sherman Creek Hatchery (1991-047-00), and Lake Roosevelt Trout Net Pen (1995-009-00) projects responded to ISRP questions in a single document and provided adequate information. Ideally, the sponsors would have text and data tables such as those in the response in concise annual reports. 

The projects producing rainbow trout and kokanee for release into Lake Roosevelt to provide resident fish substitution for lost anadromous production above Chief Joseph and Grand Coulee dams have established metrics for performance in culture (hatchery and net-pens) including egg collections, egg-to-fry survival, fry-to-release survival, fish health maintenance as well as post-release monitoring to collect survival and harvest information. Since the last review in 2006 (2007/2009 review) the co-managers have developed harvest objectives for kokanee and rainbow trout and a decision tree for kokanee egg production from Lake Roosevelt hatchery kokanee collected at Hawk Creek. The decision tree includes performance thresholds that would terminate the effort.

The data show that performance in the hatchery and net pens is adequate for both trout and kokanee. However, the percentages of released rainbow trout and yearling kokanee that are harvested are very low, averaging only 4.6% and 0.3%, respectively. These harvest levels are much lower than the harvest goals. Presumably, the harvest rate of kokanee resulting from fry releases is much lower. Are the low harvest rates associated with low survival after release, low angler effort, or both? While the hatchery program has released numerous trout and kokanee and has contributed to harvests of resident fishes, it is not clear that the program has “greatly enhanced Lake Roosevelt fishing opportunities” as stated on page 35 of the sponsor response.

The Lake Roosevelt Evaluation Project has done a good job in RME for these projects and has provided the post release metrics for these projects. Information on the harvest of wild redband trout and actions to minimize harvests of wild kokanee through harvest regulations is appreciated. 

ISRP Retrospective Evaluation of Results

Collectively, the Spokane Tribal Hatchery (199104600), WDFW Sherman Creek Hatchery (199104700), and Lake Roosevelt Net Pens (199500900) plan to rear 750,000 yearling rainbow trout (5/lb) for release into Lake Roosevelt in May after draw-down is complete. Rainbow trout will grow in the reservoir and recruit to the fishery the following fall and winter. These projects also rear 2 to 3 million kokanee fry (300/lb) and 250,000 kokanee yearlings (7/lb) for release into the reservoir. Kokanee broodstock from Lake Roosevelt are being developed using Hawk Creek as a broodstock collection location. For rainbow trout, triploid eyed eggs are obtained from the Washington Department of Fish and Wildlife’s (WDFW) Spokane Hatchery. For kokanee, Meadow Creek stock eggs are obtained from British Columbia (based on availability), and Lake Whatcom stock eggs are obtained from WDFW. Kokanee egg availability is dependent on adult run size in the source locations and is a limiting factor for achieving fry and yearling release goals.

For rainbow trout, eggs are incubated at the Spokane Tribal hatchery, and fry split between the Spokane Tribal Hatchery and Sherman Creek Hatchery. In October juvenile rainbow trout are transferred to net pens for production rearing for eventual release the following May. For kokanee, eggs are received at the WDFW Spokane Hatchery for thermal marking. Kokanee fry releases are hatched and reared at the Spokane Tribal Hatchery. Kokanee yearling releases are hatched at the Spokane Tribal Hatchery and split and reared at both the Spokane Tribal Hatchery and Sherman Creek Hatchery.

These projects have life-stage survival goals of 80% egg survival to feeding fry, 90% survival from fry to fingerlings, and 90% survival from fingerlings to yearlings.

For rainbow trout, at the Sherman Creek Hatchery there have been unaccounted losses of juvenile fish ranging from 13.5% to 19.1%. The source of these losses needs to be identified, and efforts to remedy them are warranted. The Lake Roosevelt Trout Net Pen Project released, on average, 638,000 triploid trout per year, which is slightly under the goal of 750,000 trout as a result of low numbers of fish (259,000) released in 2007.

For kokanee the release numbers have been variable with shortfall in release numbers owing to the unavailability of eggs.

Survival from release to harvest has not meet program goals. The co-managers and stakeholders express satisfaction with the rainbow trout program despite not having achieved the harvest targets. For rainbow trout the harvest goal is 50,000 to 150,000 fish; this has only been achieved in 2010 for the four years 2007 to 2010. The other three years had harvest of 11,547, 18,333, and 31,204. Approximately 28,200 trout have been harvested each year; the percentage of released trout that are harvested is low, averaging 4.6%. For kokanee, the goal is 18,500 fish from stocking fry and 12,500 from stocking yearlings. Table 9 in the response provided kokanee harvest for yearling hatchery production of 122; 368; 1,086; and 1,842 fish. This is a harvest yield ranging from 0.04% to 0.80%, well below the 5% goal for yearling kokanee. The harvest of wild redband trout has averaged 3,270 trout per year.

It is likely that reservoir environmental conditions including operational constraints and the biological community structure are unsuitable for rainbow trout and kokanee survival to the levels desired.

===========QUALIFICATIONS FOLLOW================

These Qualifications and Comments apply to the following projects:

Spokane Tribal Hatchery (199104600)

Sherman Creek Hatchery (199104700)

Lake Roosevelt Net-Pens (199500900) - Please note that comments for rainbow trout only, not kokanee, apply to this project.

The harvests of both net-pen reared yearling rainbow trout and kokanee fry and yearlings are substantially below the program goals. For rainbow trout the harvest goal is 50,000 to 150,000 fish. Over the period 2007-2010, this goal has been achieved only in 2010. The other three years had harvest of 11,547, 18,333, and 31,204. Kokanee have fared even worse. The kokanee goal is 18,500 fish from stocking fry and 12,500 from stocking yearlings. Harvests from yearling hatchery production from 2007 to 2010 were 122; 368; 1,086; and 1,842 fish. This is a harvest yield ranging from 0.04% to 0.80%, well below the 5% goal for yearlings.

It is likely that reservoir environmental conditions including operational constraints and the biological community structure is unsuitable for rainbow trout and kokanee survival to the levels desired.

The project sponsors should continue efforts to evaluate why harvest rates are so low on stocked trout and kokanee. The sponsors need to develop future plans for revising harvest goals for kokanee due to the continuing low harvest rates or provide plans for addressing their two major limiting factors: entrainment and predation by invasive non-native species (specifically walleye). Furthermore, in view of the partial success, developing plans for experimental fish culture work (even if modest) as part of the hatchery program to address post-release shortcomings needs consideration. Some effort to understand variation in past return to creel results would also be useful, including an assessment of past practices and their results (positive or negative). Such a scientific addition to this work could add a valuable and non-routine, adaptive management dimension to the fish-rearing.

They should also continue to evaluate whether wild redband and kokanee can withstand the harvest rates they encounter in response to harvests on hatchery fish. The attempt to fin clip 100% of yearling kokanee and trout should be evaluated after all fish have been presumably marked, because poorly marked fish may cause bias in fish metrics.

Our opinion from the current set of results with kokanee is essentially the same as our last review of the Lake Roosevelt Guiding Document. With entrainment and predation, the kokanee goals are just not being met. The kokanee stocking likely provides a forage base for predatory non-native fish in Lake Roosevelt. The ISRP believes there is a need to take a hard look at whether kokanee are a scientifically realistic fish to attempt to produce a mitigation fishery, despite past kokanee production in Lake Roosevelt and cultural values.

An economic analysis of the various stocking efforts in Lake Roosevelt and the harvest benefits would be useful. This might be a good task for the IEAB or the sponsors.

In order for the ISRP to complete evaluation and provide retrospective reporting to Council on progress since the last review, a table and brief narrative should be provided in a response that summarizes the production of trout including the number and size received from each hatchery and released from each net pen site, as well as post release survival and harvest for each year since the last ISRP review. A concise description of methods to determine these metrics should be included in the response.

The response should also include a diagram of fish transfers and relationships between the Net Pens, Spokane Tribal Hatchery, Sherman Creek Hatchery, and the Lake Roosevelt Monitoring projects.

The ISRP understands and appreciates that the Lake Roosevelt Net Pen Project uses monitoring data collected and analyzed by the Lake Roosevelt Monitoring Project. Even if these data and analyses are presented in the monitoring project, they should have been incorporated into the accomplishments and adaptive management sections of this proposal. For example, what is the harvest rate on native redband trout in the rainbow trout fishery, and is this harvest rate detrimental to the native population? At this point in the review process, these items should be included in a response to the ISRP.

The Council’s 1999 Artificial Production Review (NWPCC 1999-15) established that evaluating hatcheries based on numbers or pounds of fish produced and released was inadequate and that goals and objectives were required for post-release performance. The ISRP looks for clear metrics for performance in the hatchery or net pen including in-pen survival and growth, disease monitoring or other health inspections, percentage of triploid trout, net pen water quality compliance inspections, and food conversion as well as post-release performance including survival for stated intervals, harvest, and fish condition. These should all be identified and reported in the response for the time period since last ISRP review.

In the latest annual report, the sponsors note that due to high snowpack and resulting short water retention time in Lake Roosevelt many fish were entrained through Grand Coulee dam last year. This comment should be expanded upon and any estimates of entrainment provided in the response.

See the programmatic comments on fish stocking.

 1. Purpose: Significance to Regional Programs, Technical Background, and Objectives

Significance to Regional Programs: The proposal provides an adequate presentation and linkages to the Fish and Wildlife Program and individual subbasin plan elements.

Technical background: A lengthy explanation of the history of kokanee and rainbow trout artificial production programs within the Lake Roosevelt watershed is provided. The summary includes information on the original justification for selecting kokanee as a focal species, challenges the program has encountered since its inception with entrainment from Lake Roosevelt, and predation by walleye, and more recently by smallmouth bass, and difficulties with assessment as well as current production targets and harvest goals. The length of text and information extending beyond the proposal distracted from the primary elements of what was being proposed. The background section would be improved by including a chart of fish entering and exiting the net pens each year, and their growth within the pens. 

Objectives: The overall objective of providing harvest for rainbow trout is consistent with the Fish and Wildlife Program and this project. 

2. History: Accomplishments, Results, and Adaptive Management (ISRP Review of Results)

Accomplishments and results: The information provided is inadequate. A clear description of the metrics used for evaluation of the net pens and post release performance including survival, harvest, and interactions with native resident fishes needs to be presented in tabular form with a brief narrative. The extensive text in the problem statement provides a comprehensive narrative about the history of the program but does not identify metrics for the project phases (hatchery and post release) and an indication that performance standards have (or have not) been achieved. Some of this information does appear to be included in annual reports. The ISRP needs this concise presentation for both evaluation of the proposal and retrospective reporting to Council. The questions are: What was the survival while rearing in the net pens, growth, fish health, net pen monitoring for sediment quality, etc.? What were the growth, survival, condition factor, and harvest rate after release into Lake Roosevelt? What was the harvest rate on native redband trout during the fishery on rainbow trout? This should be presented for each year since the project was last reviewed. 

Adaptive Management: The explanation of the project history in the problem statement provides a reasonable summary of the changes in management. 

ISRP Retrospective Evaluation of Results

To be completed by the ISRP following the sponsors’ response.

The harvest goals for rainbow trout (20% of release or 150,000 harvested fish) needs to be considered in comparison to other large reservoir systems so that the ISRP, Council, BPA, and stakeholders can put this program in context with other similar put-grow-and-take efforts such as Lahontan cutthroat trout in Pyramid Lake, Nevada and rainbow trout in Flaming Gorge, Wyoming/Utah.

3. Project Relationships, Emerging Limiting Factors, and Tailored Questions for Type of Work (hatchery, RME, tagging)

Project Relationships: Information in the proposal is adequate for consideration of Lake Roosevelt BPA projects. Additional consideration is needed with Washington trout stocking programs, and whether this project is consistent/compliant with State of Washington policies on hatchery operations, fish release, harvest yields, and native species interactions.

Emerging Limiting Factors: Consideration of predation on project fish by walleye, smallmouth bass, and northern pike are discussed in detail in the problem statement. Reservoir operations and other environmental considerations, such as climate change, are briefly identified. The proposal notes that these coastal rainbow trout are planktivores, not predators on native fishes, and that zooplankton levels are adequate for hatchery and native salmonids. The proposal indirectly notes the potential effect of recreational harvests on native redband trout but acknowledges little data exist. This is an important data gap that needs to be filled. Some genetic and tagging efforts are intended to improve the knowledge of interactions, and the project is using triploid rainbow trout to avoid introgression. If consequential impacts are detected, it could influence the fate of the project.

Tailored questions: 

1. Describe opportunities to restore or reintroduce resident native fish: The response indicates that other projects are involved in sturgeon and redband trout restoration and habitat enhancement. If so, descriptions and linkages should be provided more clearly in the proposal.

2. A resident fish loss assessment has not been completed and is needed.

3. Impacts of non-native fish releases on native fishes need to be more clearly identified and discussed. The sponsors’ statement that the rainbow trout released by the project are "native", may be technically true, but operationally it is not. The rainbow trout are a stock derived from the coastal California subspecies. The sponsors include an adequate discussion of the operating hypothesis that stocked rainbow trout and kokanee are primarily planktivores. The monitoring plan for the program needs to continue to evaluate the potential for impacts on native kokanee and redband trout, and other non-game fish. Impacts to forage fish species could have trophic affects that would require management decisions.

4. Deliverables, Work Elements, Metrics, and Methods

Deliverables are presented in a straightforward manner as 750,000 triploid rainbow trout released from net pens.

The metrics for fish production (life-stage survival, condition factor, fish health) and facility operations (sediment quality, etc.) are not presented and need to be included.

The RM&E protocols and methods section states: "There are no RM&E protocols identified for this proposal." The ISRP questions this, and believes M&E needs to be sufficient to meet the Council Program’s Artificial Production standards.

4a. Specific comments on protocols and methods described in MonitoringMethods.org

No information on protocols and methods was provided in the proposal or on MonitoringMethods.org. As noted above, the some basic metrics need to be measured and methods need to be described. 

Table 3.      Rainbow trout: Release year, stock, broodyear, transfers dates to net pens, rearing and release site, number of fish transferred, size (fish/lb.) at transfer, release date, number released from net pens, size (fish/lb.) at release, and the percent mortality while rearing in net pens from 2007-2011.

Table 4.      Redband Rainbow trout: Release year, stock, broodyear, date transferred to SCH (Sherman Creek Hatchery) from WDFW Colville Hatchery, number transferred, size (fish/lb.) at transfer, date of released, number released, size (fish/lb.) at release, number of mortalities, percent mortality, number carried over, release location, and adipose fin clip percent of Phalon Lake redbands released into Lake Roosevelt, WA from 2007-2011.

Table 5.      Release year, total number of fish released from all net pen locations, total harvest and condition factor of hatchery rainbow trout, mean growth from tag date to capture date, and mean number of days between tagging and capture for rainbow trout from Lake Roosevelt, WA.  Means followed by standard deviation in parenthesis.  * Harvest and K_TL for 2010 is preliminary.

Kokanee

A total of 5 million kokanee eggs are needed to meet annual release goals of 3.7 million kokanee fry and 250,000 kokanee yearlings.  Egg availability is dependent on annual adult return run size and is an ongoing limiting factor.  Currently, three stocks of kokanee are used to meet annual release goals.  In order of preference, they are the Lake Roosevelt mixed stock, Meadow Creek stock, and Lake Whatcom stock. 

A limited number of eggs are collected from mature kokanee returning to tributaries in the fall. This stock, Lake Roosevelt Mixed stock, is comprised of spawning returning Meadow Creek stock fish.  Past attempts at collecting eggs from other collection sites have been unfeasible, but adult returns to Hawk Creek have shown the potential for a locally adapted stock of kokanee.  Up to 3.2 million kokanee eggs are potentially available from the British Columbia Freshwater Fisheries Society Meadow Creek Kokanee Program.  This stock is preferred over the Lake Whatcom stock because it is an upper Columbia River origin stock and has been proven to have higher survival rates in the reservoir when compared to the Lake Whatcom stock (McLellan et al. 2002, 2003).  Up to 1.8 million kokanee eggs are provided by the Washington Department of Fish and Wildlife’s Lake Whatcom Hatchery Program. .  Egg availability from all three sources is dependent on annual adult returns.

All kokanee eggs are thermally marked at Washington Department of Fish and Wildlife’s Spokane Hatchery.  Once marked, eggs are transferred to the Spokane Tribal Hatchery for subsequent rearing.  Kokanee fry are released (350/lb.) in June at the deep-water release site (Seven Bays).  Approximately 300,000 fry are carried over to meet annual yearling kokanee release goals.  To alleviate space constraints at the Spokane Tribal Hatchery yearling kokanee (currently approximately 100,000) to be released in June are transferred to the Sherman Creek Hatchery and released in June.  The release goals for the kokanee program are 250,000 yearlings to be released at the Fort Spokane boat launch and the remaining 2-4.5 million Whatcom/Meadow Creek fry to be released in the open water at Seven Bays.  In years when Meadow Creek eggs are unavailable Lake Whatcom eggs are used to meet annual yearling release goals.  A flow chart has been provided that displays the transfer of kokanee between the two hatcheries (Figure 2).

The total number and stock of kokanee eggs received from 2007-2011 are presented in Table 6.  The percent survival to feeding fry, number of fry reared to distribution, percent survival to distribution, number of fry released, size (fish/lb.) of fry at release, release site, and number of fry carried over to meet annual yearling production goals are also displayed in Table 6. 

Release year, stock, broodyear, and number of kokanee reared to distribution from 2007 through 2011 are displayed in Table 7.  Percent survival to distribution, number of yearling kokanee transferred to Sherman Creek Hatchery, number of kokanee yearlings released from the Spokane Tribal hatchery, size (fish/lb.) at release, and release location are presented in Table 7.

Yearling kokanee production from 2007 through 2011 at Sherman Creek Hatchery is outlined in Table 8.  The date the yearlings were received at Sherman Creek Hatchery, the number transferred, size (fish/lb.) at transfer, date of release, number of yearling kokanee released, size (fish/lb.) at release, release location, percent mortality, unknown loss, percent fin clipped, and type of fin clip are presented in Table 8.

Release year, total number of adipose fin clipped yearling kokanee released, harvest of adipose fin clipped kokanee, condition factor of adipose fin clipped kokanee, and mean growth (total growth for four months)for adipose fin clipped kokanee from time of release till time of capture  is presented in Table 9.

Table 6.      Kokanee salmon: Year, stock (WHAL = Lake Whatcom stock, MEAD = Meadow Creek stock), broodyear, number of eggs received at the Spokane Tribal Hatchery, number of kokanee eggs hatched, percent survival to feeding fry, number of fry reared to distribution, percent survival to distribution, number of fry released, release site, and number of kokanee fry held over for yearling releases into Lake Roosevelt, WA from 2007-2011.

Table 7.      Kokanee salmon: Year, stock (WHAL = Lake Whatcom stock, MEAD = Meadow Creek stock, ROOS = Lake Roosevelt mixed stock), number reared to distribution, percent survival to distribution, number transferred to SCH (Sherman Creek Hatchery), number released, size (fish/lb.) at release, and release site for kokanee yearlings released into Lake Roosevelt, WA from 2007-2011.

Table 8.      Kokanee salmon: Release year, stock, broodyear, date transferred to Sherman Creek Hatchery, size (fish/lb.) at transfer, date released, size (fish/lb.) at release, percent mortality, unknown loss/overrun, release site, percent marked, and markt ype (AD = adipose fin clip, LV = left ventral fin clip, RV = right ventral fin clip) for kokanee salmon released into Lake Roosevelt, WA from 2007-2011.

Table 9.      Kokanee salmon: Release year, number of yearling kokanee released, annual hatchery kokanee harvest, condition factor (SD), and mean growth (calculated for a four month time span from when fish are released to recaptured) for yearling kokanee released in Lake Roosevelt, WA from 2007-2011.  Harvest and condition factor data for 2010 are preliminary.