BPA Fish and Wildlife FY 1997 Proposal

Section 1. Administrative
Section 2. Narrative
Section 3. Budget

see CBFWA and BPA funding recommendations

Section 1. Administrative

Title of project
Steelhead Supplementation Studies in Idaho Rivers

BPA project number   9005500

Business name of agency, institution or organization requesting funding
IDFG

Sponsor type   ID-State/Local Agency

Proposal contact person or principal investigator

BPA technical contact   ,

Biological opinion ID   

NWPPC Program number   3.1D.1

Short description
This project will help determine if supplementation can be used to rebuild depressed stocks and re-establish extirpated stocks of summer steelhead Oncorhynchus mykiss in Idaho. Different methodologies will be researched to determine the best method(s) to supplement steelhead stocks to improve natural production. We will gather life-history attributes from wild stocks to guide management decisions to match donor and recipient stocks.

Project start year   1993    End year   2004

Start of operation and/or maintenance   0

Project development phase   Implementation and Maintenance

Section 2. Narrative

Related projects
Idaho supplementation studies (ISS) (BPA Project 89-098: We share equipment, personnel, and data. Steelhead data collected from ISS study streams include parr densities, PIT-tagging steelhead at screw traps, and juvenile migration characteristics.

Performance/Stock productively impacts of hatchery supplementation (BPA Project 90-052): This study is testing our hypothesis H2 (see Testable Hypothesis section) in the Clearwater River drainage. This allows us to focus our research on hypothesis H2 in the Salmon River drainage.

Idaho Natural Production Monitoring and Evaluation (BPA Project 91-73: We share equipment, personnel, and data. We often cooperatively snorkel and PIT-tag in study streams that are in the same geographic area.

Research and recovery of Snake River sockeye (BPA Project 91-72): We share personnel, equipment, and assistance.

Project history
Experimental Design was submitted to BPA in December 1992. Field work began in 1993 and the full implementation of the project was to occur between 1993 - 1995. Because of BPA funding reductions, we have not been able to implement the full study as described in the Experimental Design.

Biological results achieved
1993

Intensively snorkeled 8 streams to obtain juvenile steelhead densities.
Recorded stream temperatures at 17 sites in 17 streams.
Install a weir in Fish Creek to collect wild adult steelhead escapement, length, sex ratio, and scales.
Collected 4,748 and PIT-tagged 2,870 juvenile steelhead in 6 streams. Obtained the length, weight, condition factor, and migration timing of juvenile steelhead from the streams.
Stocked hatchery adult steelhead in Frenchman and Beaver creeks (Objective 1)
Stocked 50,000 hatchery fingerlings in the South Fork Red River (Objective 2).
Did a stream habitat survey in 5 streams.
318 PIT-tagged fish were detected at the Snake River dams and McNary Dam.

1994

Intensively snorkeled 11 streams to obtain juvenile steelhead densities.
Recorded stream temperatures at 34 sites in 29 streams.
Install a weir in Fish Creek to collect wild adult steelhead escapement, length, sex ratio, and scales.
Collected 9,312 and PIT-tagged 6,314 juvenile steelhead in 12 streams. Obtained the length, weight, condition factor, and migration timing of juvenile steelhead from the streams.
Stocked hatchery adult steelhead in Frenchman and Beaver creeks (Objective 1)
Stocked 50,000 hatchery fingerlings in the South Fork Red River (Objective 2).
Did a stream habitat survey in 4 streams.
795 PIT-tagged fish were detected at the Snake River dams and McNary Dam.

1995

Intensively snorkeled 8 streams to obtain juvenile steelhead densities.
Recorded stream temperatures at 36 sites in 32 streams.
Install a weir in Fish Creek to collect wild adult steelhead escapement, length, sex ratio, and scales.
Recorded air temperature, relative humidity, and air pressure at Fish Creek.
PIT-tagged 3,4314 juvenile steelhead in 7 streams. Obtained the length, weight, condition factor, and migration timing of juvenile steelhead from the streams.
Stocked hatchery adult steelhead in Beaver Creek (Objective 1)
Stocked 50,000 hatchery fingerlings in the South Fork Red River (Objective 2). Did a stream habitat survey in 1 stream.
1,305 PIT-tagged fish were detected at the Snake River dams and McNary Dam.

Annual reports and technical papers
Steelhead supplementation studies in Idaho Rivers. Annual Report 1993.

Steelhead supplementation studies in Idaho Rivers. Annual Report 1994. (Final draft being reviewed by IDFG staff).

Management implications
The general expectation for supplementation among management and user groups in Idaho is to use artificial propagation to improve natural production of steelhead and build self-sustaining, harvestable populations of steelhead in the Salmon River and Clearwater River drainages without adversely impacting existing wild populations. This research will be used to guide management decisions and policy on the supplementation methods attain fishable steelhead populations and to restore steelhead where extirpated. Data from smolt PIT-tag detections will be incorporated in IDFG recommendations for mainstem flow and river management polices.

Specific measureable objectives
1 . Monitor Beaver and Frenchman creeks to determine juvenile and smolt production from adult outplants.
2. Outplant wild steelhead adults (when escapement increases) and compare the production of juveniles and smolts with a hatchery stock.
3. Assess the survival from egg to smolt and from smolt to adult of wild steelhead that are spawned and reared in a hatchery and compare with the hatchery stock.
4. Determine the number of adults that return from the fingerling stockings in SF Red River and from smolt stockings in Red River. Estimate and compare juvenile production from the adults from the two groups (fingerling or smolt stocking) that spawn naturally.
5. Continue PIT-tagging juvenile steelhead to obtain smolt migration characteristics, survival rates from parr-to-smolt, survival from smolt-to-adult, and growth rates in the streams.
6. Continue snorkeling �key� steelhead production stream to tract abundance on a yearly basis.
7. Collect baseline disease and genetic data from wild steelhead streams in the Clearwater River and Salmon River drainages.

Testable hypothesis
Hl: There is no difference among brood sources to establish natural production (supplementation-restoration) in tributaries of the upper Salmon River.

H2: There is no difference in survival between a wild brood source and an established hatchery brood source from: (1) egg-to-smolt in a hatchery environment; (2) smolt-to-adult of fish reared in a hatchery and released into the natural environment; (3) egg-to-age 1 parr in the natural environment; and (4) age 1 parr to smolt in the natural environment.

H3: There is no difference between fingerling or smolt stockings in establishing natural steelhead production as measured by F, age 1 juvenile abundance in the Red River drainage.

H4: There is no difference among wild steelhead stocks in: (1) smolt age; (2) smolt length; (3) smolt migration pattern; (4) smolt condition factor; (5) juvenile growth rates and; (6) parr migration patterns.

Underlying assumptions or critical constraints
1. We need adequate wild escapement before we "mine" adults for use in experiments.
2. We receive adequate funding to continue the experiments.

Methods
The experimental design, submitted to BPA in December 1992, outlines our methods and analysis in detail.

Objective 1: Assess the performance of hatchery and wild brood sources to re-establish steelhead in streams where extirpated.

Approach: Ideally, supplementation-restoration utilizing various brood sources would be assessed at the population level by releasing hatchery fish into separate drainages and tracking fish abundance, survival between life stages, and life history attributes (sex ratio, fecundity, size, etc.) over several generations for each population. Because of the large number of streams that would be required, risks associated with supplementation uncertainties, and the desire to expedite feedback, we have partitioned this research hypothesis into four chronological components that will be tested somewhat independently: in-hatchery survival, post-release survival, reproductive success, and long-term fitness.

General

1) Collect hatchery adults at Sawtooth Hatchery and wild adults from the Middle Fork Salmon River or main Salmon River.

In-hatchery performance

2) Determine percent eye-up, survival, growth, health, and condition factor to the smolt stage of each brood source using general hatchery operating procedures. We plan to use 7 -15 pair of fish per brood source. Each brood source will be handled and reared equivalently, but in separate raceways. ANOVA will be used to test for differences among the brood sources for length, weight, and condition factor utilizing a completely random design. No statistical test for percent eye-up and survival is required since a total count of green eggs, eyed eggs, and smolts for each brood source will be made in the hatchery.

Post release survival

3) Release hatchery reared smolts at weir sites to coincide with natural smolt migration. Using PIT tag detections, measure travel time of outplanted smolts from each brood source to Lower Granite Dam. Each year's data can be analyzed with ANOVA using a Completely Random Design. When several years of data are combined, a Repeated Measures (split plot in time) ANCOVA will be used with flow as the covariate. Smolt survival data will be analyzed with log-linear models.

4) Identify the brood source of adults returning to the weir. Determine smolt-to-adult survival, sex ratio, length, weight, age, fecundity, and timing of return of adults. Results of the first release year, except survival, will be analyzed by ANOVA using a Completely Randomized Factorial Design. Since smolts will be released for 3 consecutive years, a Repeated Measures Analysis (split plot in time) will be used for subsequent analysis. Smolt-to-adult survival data will be analyzed using log-linear models.

Reproductive Success

5) Outplant 7 - 10 pair of adults from each brood source into vacant habitat in two streams that are suitable for steelhead production. Enclose the section of stream where adults are stocked with picket weirs. Determine pre-spawning mortality by daily carcass census in the enclosed stream reach and track the number of redds built. Estimate egg retention from carcass necropsies and calculate the number of eggs deposited in redds.

6) Trap juvenile steelhead immigrants and emigrants with 2-way weirs. Measure stream flow, temperatures, conductivity, and other habitat characteristics.

7) Estimate parr production by snorkel surveys and trapping juvenile emigrants.

8) Estimate reproductive success of each brood year from the number of fry emigrants and age 1 parr densities produced from known adult spawning escapement. The null hypothesis is no differences between brood sources (tested with ANOVA using a split-plot in time repeated measures design). Covariates in the analysis could include the number of spawners, conductivity, temperature, and habitat area.

Long-term fitness

9) PIT tag 500 - 800 juvenile steelhead when they reach 70 mm length. Fish will be collected by seining, electrofishing, minnow traps, and emigrant traps.

10) Trap smolts leaving the streams, determine age, length, weight, time of migration, and yield. Interrogate smolts for PIT tags and tag untagged fish to estimate their travel time and survival to Lower Granite Dam.

11) Estimate long-term fitness as a function of smolt production to Lower Granite Dam and emigration characteristics. Null hypothesis of no differences between brood stocks (tested with ANOVA using a split-plot in time repeated measures design). Covariates in the analysis could include the number of spawners, conductivity, temperature, and habitat area.

12) Use the results obtained from experiments to propose guidelines for donor stock selection
in restoration rivers.


Objective 2: Evaluate the ability of returning adults from hatchery smolt and fingerling releases to produce progeny in natural streams.

Our research will focus on life stage of release using an established hatchery stock. We will compare the abundance of age 1 steelhead parr produced by naturally spawning adults that were released as hatchery fingerlings or smolts. Because of limited research opportunities we will use only an established hatchery broodstock(s) to test this hypothesis. Fingerlings (rather than fry) were chosen since the fish can be fin marked and PIT tagged prior to release into the study streams. The primary evaluation point will be age 1 parr abundance produced from the returning adults This objective is being investigated in the upper Red River drainage.

1) Release 50,000 marked fingerlings (5,000 PIT-tagged, 45,000 CWT) into the South Fork Red River for four consecutive years. Stocking began in 1993.

2) Release 5,000 PIT-tagged smolts in Red River, upstream of the South Fork Red River, for four consecutive years. Begin stocking in 1996.

3) Snorkel the SF Red River and Red River yearly to obtain juvenile steelhead densities.

4) Determine smolt production from PIT tag detections at Lower Snake River dams from smolt and fingerling releases.

5) Trap returning adults at the Red River weir. Determine the origin of the fish and allow them to spawn naturally. We may use radio telemetry to follow fish to spawning areas.

6) Estimate reproductive success to age 1 using densities obtained from snorkel surveys. Null hypothesis of no differences between smolt or fingerling release can be tested with ANOVA using a split-plot in time repeated measures design. Covariates in the analysis could include the number of spawners, conductivity, temperature, and habitat area.

Objective 3: Assess the abundance, habitat, and life history characteristics of existing steelhead populations in the Salmon and Clearwater river drainages.

1 ) For each study stream record daily mean, maximum, and minimum temperatures; develop stream hydrograph, estimate proportion of each habitat type, measure conductivity, and gather other pertinent stream and geological data. Snorkel study streams to obtain summer juvenile steelhead densities. PIT tag juvenile steelhead and estimate survival during freshwater rearing and to Lower Snake River dams.

2) Estimate adult steelhead escapement into Fish Creek, Clear Creek, Pahsimeroi River, and the upper Salmon River streams with weirs. Record length of adults, sex ratio, age, and date of return. Develop stock-recruit curves relating spawners with juvenile density and smolt yield.

3) Trap emigrating parr and smolts at screw traps operated by this project and the Idaho Supplementation Studies chinook project. Length, weight, age, and date captured are recorded for outmigrating steelhead. PIT tag fish for survival and travel time estimates to Lower Granite Dam. ANOVA will be used to test for differences among rivers for smolt length and weight.

4) Obtain baseline genetic and pathogenic data on steelhead populations. Measure allelic frequency for naturally produced steelhead using electrophoresis. ANOVA will be used to test for stream differences and differences between hatchery and wild steelhead. Values used in the analysis will be the arcsin of the square root of the frequency of the common allele at each locus. In addition, cluster analysis techniques will be applied to Nei's genetic distance to develop a denogram showing the relations among steelhead stocks within Idaho.

STATISTICAL ANALYSIS

The first-year data of some experiments, for example smolt travel time and survival to Lower Granite Dam, will be analyzed with a completely randomized factorial design. This design can be used to test for in-hatchery growth, condition factor, and health differences among broodstocks.

Since most of the experiments in this study will be repeated on a yearly basis for three to four years, the data will be analyzed upon completion with ANOVA using a repeated measures design (split plot in time) with two or three treatments.

Log-linear models will be used to analyze adult return data. Log linear models are a good method for analysis of survival data of known numbers of smolt releases since we are dealing with ordinal (count) data. Analysis of the adult return from one smolt release can be done using the standard chi-square test for independence. For example, adults from b broodstocks released at I weirs can be arranged into a b x I table. When all the data has been collected from the y release years, a b x I x y multidimensional contingency table is formed for analysis. Two models, the product multinominal and super multinominal, will be used for the analysis of multidimensional contingency tables. The product multinominal model is used for hypothesis testing. The model is a treated as a product multinominal (rows are treatments and row totals are fixed). This procedure will determine what independent variables are necessary to predict the logic. In the adult return from smolt release example, brood source is the response variable, and location and year are the independent variables. A test of significance would be made for location, year, and the interaction between location and year. A significant effect implies that a factor is necessary to explain the observed adult return rate of the brood sources. The super multinominal model is not concerned with dependent and independent variables or hypotheses testing. Unlike the product multinominal model, row and column margins are not fixed. Only the total number of observations (N), is fixed. This method is a goodness of fit test to pick a model which best explains the data.

Several nonparametric statistics will be used for data analysis. Kolomogorov-Smirnov statistics will be used to test for differences in distribution functions of spawning time, time of emigration, and time of smolt migration. A Cox-Stuart test for trend will be used to test for changes in adult escapement and juvenile densities if changes are made at existing weir sites and the binomial test can be used to test for changes in sex ratios.

Brief schedule of activities
1) Outplant hatchery and wild (dependent upon escapement) into streams for Objective 1 experiments.
2) Begin evaluation of wild stock in a hatchery as outline in Objective 1 methods.
3) Stock smolts in Red River and monitor the weir for returning adults from previous stockings for Objective 2 experiments.
4) Operate screw traps and PIT-iag steelhead parr and smolts.
5) Snorkel study streams for juvenile steelhead densities.
6) Collect wild steelhead from various streams for genetic analysis and disease screening.
7) Data analysis and reporting.

Biological need
To determine the best methods to increase production of naturally spawning steelhead and to reestablish steelhead in streams where extirpated.

Critical uncertainties
The uncertainties this research addresses include: 1) the relative ability of hatchery and wild steelhead broodstocks to re-establish naturally producing populations where extirpated, 2) the ability of hatchery steelhead released at different life stages to produce adults and the ability of those adults to produce progeny in the wild, 3) assessing the abundance and life history characteristics of steelhead in various streams and habitats in the Salmon River and Clearwater River drainages and 4) the effects (if any) of steelhead supplementation on existing wild populations of steelhead, chinook salmon, and resident species.

Summary of expected outcome
Objective 1: Wild brood sources will perform better compared with hatchery sources in the natural stream environment. Hatchery fish will perform better in the hatchery environment.

Objective 2: Smolt stockings will produce more adult returns per fish stocked than fingerling stocking, but the adults returning from a fingerling stocking will produce more progeny per female in the natural stream environment.

Objective 3: There will be differences among the wild stocks in smolt age, length, and condition factor. Streams that are warmer (low elevation) will produce smolts that are younger and smaller than colder streams (high elevation). Genetic analysis will show that there are differences among stocks throughout the state, in general fish that are in geographical proximity will be more similar than fish that are more distant.

Dependencies/opportunities for cooperation
Potential listing of ldaho's summer steelhead under ESA.

Wild adult escapements must be large enough to allow us to "mine' some for our Objective 1 experiments.

Risks
The biological consequences of not continuing the project are: (1) failure to identify and evaluate potential recovery tools for depressed or locally extirpated wild stocks, (2) use of supplementation as a recovery tool without identifying genetic and population viability risks, thus adversely impacting natural populations.

Monitoring activity
This was outlined in the methods section.

Section 3. Budget

* For most projects, Authorized is the amount recommended by CBFWA and the Council. Planned is amount currently allocated. Contracted is the amount obligated to date of printout.

Funding recommendations

CBFWA funding review group   Snake River

Recommendation    Tier 1 - fund

Recommended funding level   $220,000

BPA 1997 authorized budget (approved start-of-year budget)   $220,000