FY 2003 Mainstem/Systemwide proposal 200304100

Section 1. Administrative

Proposal titleEvaluate Delayed (Extra) Mortality Associated with Passage of Yearling Chinook Salmon Smolts through Snake River Dams
Proposal ID200304100
OrganizationNational Marine Fisheries Service (NMFS)
Proposal contact person or principal investigator
NameGene M. Matthews
Mailing address2725 Montlake Blvd. East Seattle, WA. 98112-2097
Phone / email2068603251 / Gene.Matthews@noaa.gov
Manager authorizing this projectJohn G. Williams
Review cycleMainstem/Systemwide
Province / SubbasinMainstem/Systemwide /
Short descriptionDetermine if downstream migration through Snake River dams results in extra or delayed mortality.
Target speciesSnake River hatchery-reared spring/summer chinook salmon; a small percentage will be Snake River ESA-listed hatchery fish
Project location
46.659 -117.4304 Lower Granite Dam on the lower Snake River in Garfield County, Washington.
46.2501 -118.8783 Ice Harbor Dam on the lower Snake River in Franklin County, Washington
Reasonable and Prudent Alternatives (RPAs)


Hydro 188

Relevant RPAs based on NMFS/BPA review:

Reviewing agencyAction #BiOp AgencyDescription
NMFS/BPA Action 195 NMFS The Action Agencies shall investigate and partition the causes of mortality below Bonneville Dam after juvenile salmonid passage through the FCRPS.
NMFS Action 188 NMFS The Action Agencies and NMFS shall work within the annual planning and congressional appropriation processes to establish and provide the appropriate level of FCRPS funding for studies of PIT-tagged wild stocks from the lower river streams. The studies shall be used to contrast stock productivity and hydrosystem effects.
NMFS Action 195 NMFS The Action Agencies shall investigate and partition the causes of mortality below Bonneville Dam after juvenile salmonid passage through the FCRPS.

Section 2. Past accomplishments

N/A New project

Section 3. Relationships to other projects

Project IDTitleDescription
199302900 Survival estimates for the passage of juvenile salmonids Staff conducting juvenile survival analyses will determine survival of the fish detected at McNary Dam that become part of the evaluation groups for extra mortality in this proposal. They will also help in evaluations of adult returns

Section 4. Budget for Planning and Design phase

Task-based budget
ObjectiveTaskDuration in FYsEstimated 2003 costSubcontractor
1. Develop annual plan a. Determine marking and implementation strategies ongoing $20,400
1. b. Apply for federal, state, and local permits ongoing $5,300
Outyear objectives-based budget
ObjectiveStarting FYEnding FYEstimated cost
1. Develop annual plan 2004 2007 $111,100
Outyear budgets for Planning and Design phase
FY 2004FY 2005FY 2006FY 2007

Section 5. Budget for Construction and Implementation phase

Task-based budget
ObjectiveTaskDuration in FYsEstimated 2003 costSubcontractor
1. Modify fish marking facility a. modify fish marking facilities at Lower Granite Dam 1 $187,800
2. Conduct study a. mark and release fish ongoing $751,200
2. b. collect and analyze data ongoing $45,700
2. c. prepare reports and presentations ongoing $50,500
Outyear objectives-based budget
ObjectiveStarting FYEnding FYEstimated cost
1. Modify fish marking facility $0
2. Conduct study 2004 2007 $3,651,800
Outyear budgets for Construction and Implementation phase
FY 2004FY 2005FY 2006FY 2007

Section 6. Budget for Operations and Maintenance phase

Task-based budget
ObjectiveTaskDuration in FYsEstimated 2003 costSubcontractor
1. Repair and maintain marking facility a. modify, repair & maintain equipment ongoing $23,000
Outyear objectives-based budget
ObjectiveStarting FYEnding FYEstimated cost
1. Repair and maintain marking facility 2004 2007 $99,200
Outyear budgets for Operations and Maintenance phase
FY 2004FY 2005FY 2006FY 2007

Section 7. Budget for Monitoring and Evaluation phase

Task-based budget
ObjectiveTaskDuration in FYsEstimated 2003 costSubcontractor
Outyear objectives-based budget
ObjectiveStarting FYEnding FYEstimated cost
Outyear budgets for Monitoring and Evaluation phase

Section 8. Estimated budget summary

Itemized budget
ItemNoteFY 2003 cost
Personnel FTE: 2.96 $179,200
Fringe $41,800
Supplies $122,800
Travel $19,000
Indirect $122,000
PIT tags # of tags: 240000 $540,000
Subcontractor PSMFC / FishMarkers $59,100
Total estimated budget
Total FY 2003 cost$1,083,900
Amount anticipated from previously committed BPA funds$0
Total FY 2003 budget request$1,083,900
FY 2003 forecast from 2002$0
% change from forecast0.0%
Cost sharing
OrganizationItem or service providedAmountCash or in-kind

Reviews and recommendations

This information was not provided on the original proposals, but was generated during the review process.

Fundable only if response is adequate
Aug 2, 2002


Fundable if a response can justify this design or if the design can be modified to provide valid estimates of extra mortality.

The objective is to use empirical experiments to quantify delayed effects associated with hydrosystem passage. There is a logical need to address the problem of assessing dam passage mortality and the team possesses the experience and background to address the problem.

The ISRP questions whether this experiment will settle the issue because concern was originally for extra mortality to Bonneville and it is not clear that results from this experiment will apply. In this proposal, although the objectives are clearly defined, the methods do not appear appropriate for determining a clear answer to the hypothesis being tested. Determination of significant differences in delayed mortality due to passage through 8 dams versus passage through 4 or fewer dams will not be possible with the current study. The proposal asserts that if the null hypothesis is rejected, it is highly likely that migration through Snake River dams does cause extra mortality in spring/summer chinook salmon smolts. An assumption (unstated) is that the effect due to transport is the same for fish experiencing dam passage plus transport stress as it is for fish experiencing only transport stress. Is this assumption justified? It is possible that some fish experiencing dam passage alone would survive but due to experiencing transportation stress prior to dam passage stress, they succumb. Therefore comparing extra mortality for transportation only with extra mortality for transportation plus dam passage may not provide an unbiased estimate of the dam passage effect.

Questions: Have the authors considered conducting a study on fall Chinook instead or in addition to spring/summer Chinook? Could something be done to estimate the effects of the different impacts of spill, turbines, and bypass system instead of merging everything in dam passage as one thing? Are there procedures in place to ensure that good estimates of expected mortality at the dams for fish migrating in-river are obtained so "extra mortality" is clearly defined?

An excellent effort was made to do a power analysis in order to determine adequate sample sizes. One correction necessary is to note that a one-tailed procedure is required so alpha should be used rather than alpha/2 in the sample size formula.

Action Agency/NMFS RME Group Comments:

HYDRO SUBGROUP -- The authors indicate the proposed research provides information useful in satisfying RPAs 188 and 195.

The objective of RPA 185 is to contrast productivity and hydrosystem effects (delayed) between wild stocks in upper Snake stocks and those in the Lower Columbia Basin. To accomplish this, the RPA calls for PIT-tagging both wild population complexes with PIT tags. This proposal relies heavily on hatchery stocks from the Snake drainage as the population monitored. Thus its ability to fully satisfy the intent of RPA 188 is not readily apparent. The primary objective of this research is to identify the existence and generally quantify the magnitude of extra mortality as associated with dam passage. The linkage to the RPA is not all that pronounced.

The objective of RPA 195 is to establish how much post-Bonneville mortality is attributable to natural causes or other processes, such as hydrosystem passage or general fish fitness. This proposal is relevant to the fundamental intent of this RPA, i.e., identify delayed effects associated with hydrosystem passage. The proposed research clearly addresses the hydrosystem contribution to any extra, unexplained mortality that may exist. The experimental approach appears sound. However, the sample sizes necessary to provide the precision targets are considerable (~ 236,000 PIT-tagged @ LGR) and may be a challenge to acquire in some brood years.

Ancillary Benefits. These tagged yearling chinook will also yield inriver survival estimates. The large sample sizes all but ensure improved precision over most extant smolt survival estimates. This could be advantageous to the extent these estimates can be incorporated into survival Performance Standards tests prescribed in the BO. The proposal does not discuss the suitability of these estimates for such evaluations. The RME Hydro Work Group encourages the authors to explore this application and incorporate it as a section in the proposal.

ISRP Remarks on RME Group Comments: The ISRP does not agree with the RME Workgroup comments that the experimental approach appears sound as noted above. Other RME Workgroup comments relate to RPA connections and ancillary benefits that are not noted in the ISRP review comments.

High Priority
Oct 24, 2002


Before this study proceeds, the project sponsors and co-managers implementing tagging studies on the Columbia River need to meet and discuss this proposal and determine an appropriate statistical and logistical study design to resolve this critical uncertainty identified in the 2000 NMFS BiOp. CBFWA members will provide specific comments to the project sponsor. This task is considered an urgent need for the Basin although this proposal is not adequate for funding at this time. CBFWA recommends that the project sponsor work closely with the CSS Oversight Committee to develop a regionally accepted study design and revised proposal. The proposal should only be funded following approval from the AFC and CBFWA Members.
High Priority
Oct 24, 2002


I have reviewed the budget for the proposal "Evaluate Delayed (Extra) Mortality Associated with Passage of Yearling Chinook Salmon Smolts Through Snake River Dams", Project # 35047. The budget is appropriate for the proposed research and tasks. A reduction in the funding request without a corresponding reduction in the scope of work would be inappropriate.
Nov 5, 2002


Fundable; agree with CBFWA's high priority ranking. The ISRP commends the proponent for efforts to address estimation of "extra mortality", which has been a contentious issue for many years. Of necessity, any study to address what the proposal refers to as a "hypothetical value" will be complex and require careful thought, analysis and planning. The proposal exhibits these features.

It has been argued by process of elimination that "extra mortality" is attributable to conditions experienced by smolts of Snake River stocks of chinook in their outmigrations through the hydrosystem. Arguments purporting to demonstrate the existence of "extra mortality" are primarily based on unconvincing comparisons with SARs of Lower Columbia River Basin stocks that are outside the Snake River. Since the differences observed favor the lower river stocks to a larger degree than expected based upon the "measured" in-river losses of upriver stocks, it was postulated that there might be an element of "delayed or extra mortality" that occurred in the ocean after the fish passed Bonneville Dam, the point of the last "measurement". In fact, the "measurements" consist of projections of average losses measured in the Snake River that are assumed to apply from McNary Dam to below Bonneville Dam. In our view, the assumptions and the methods used to develop the projections are of highly questionable validity.

The present proposal is adopting a definition of "extra mortality" as a component of mortality in a MCN-BON SAR attributed to conditions experienced previously in the in-river smolt passage LGR-MCN. This has a narrower focus than the PATH definitions, since it does not include the MCN-BON reach for smolt passage.

This is a design with two treatments and no control. The design measures MCN-to-BON SAR for two treatment groups: fish that were truck-transported as smolts LGR-ICE, and fish that spent equivalent truck transportation time going nowhere followed by in-river passage LGR-ICE. The truck-only treatment group is subject to "delayed mortality," and the truck-followed-by-in-river treatment group is subject to "delayed mortality" plus "extra mortality." This experiment provides information about delayed effects of transport plus passage through a differing number of dams for the two groups. It does not provide information about the effect of differing number of dam passages alone. There is an implicit assumption of additivity for these effects. With the additivity assumption, and the assumption that "delayed mortality" does not begin to be expressed until the fish are below MCN, one could calculate "extra mortality" as the difference between the mortality rates measured in the MCN-BON SAR for the two treatment groups. Regardless of the assumptions, there would be no opportunity just with this design to estimate "delayed mortality." Delayed mortality and extra mortality might be isolated by providing an additional experimental group not transported, as recommended in item 2 below.

A power to detect a 20% difference, with 95% confidence, 80% of the time, for the data from one year may be inadequate. This scenario is probably a best-case calculation, since variation between the within-year replicates may turn out to be large. If possible, the sample size should be increased.

The ISRP strongly recommends:

  1. reaching an agreement so that the PIT tagged fish from the truck-followed-by-in-river treatment group from this project will always be returned to migrate in-river, rather than taken into transport, whenever they hit a bypass detector. This can be done automatically with the "sort by code" hardware at the collector dams (LGO and LMO). This will preserve sample size for this treatment group, and it will also present opportunities for stratifying on the number of bypass detections between LGR and MCN.
  2. developing a method to estimate "delayed mortality" by using PIT tagged fish from outside this study but with coordinated release dates from the hatcheries to match batches of treatment groups from this study. An alternative would be to include an additional real control group of PIT tagged fish that are allowed to migrate in river from LGR to MCN, without the truck detour. This would permit estimation of "delayed mortality," which is not possible with the present design.
  3. creating a CD of the "consensus interpreted data" as part of its annual reporting process. This would allow statistical researchers to try various statistical methods for analyzing these data, without the confounding issue of how the different researchers made their decisions about which data to cull.
Further recommendations are:
  1. increasing the budget to create another treatment group that is collected at LGR and barged to the tail race of ICE. This would allow investigation of the question concerning whether the "delayed mortality" from truck transportation is the same as from barges.
  2. estimating mortality from recoveries of fish released below Lower Granite Dam and recovered downstream. Similarly, estimating mortality for fish released below Ice Harbor Dam and detected at McNary Dam. This would allow directly estimating the mortality in transportation itself rather than assuming the usual 5% mortality rate used in modeling transportation.

Finally, it should be noted that the ISRP recommends that NMFS, CBFWA, and the Corps of Engineers concentrate on development of better estimates of SARs that will directly answer critical questions for recovery of endangered stocks in the Columbia Basin. For example, direct answers are needed to questions concerning return rates of transported versus in-river migrating fish, adequacy of return rates needed to recover stocks, indirect mortality, extra mortality, "D", etc. The ISRP has recommended elsewhere in this report, and again emphasizes the importance of installation of a PIT tag reader in the corner collector at Bonneville 2 (project 199302900). Data collected there will provide information that would more directly estimate SARs for groups of in-river migrants than the efforts in this project.

If funded, this project should be coordinated with other monitoring projects to ensure compatibility of objectives, common methods and protocols. This coordination could be accomplished under the favorably reviewed CBFWA proposal #35033.

Jan 21, 2003


Statement of Potential Biological Benefit
Indirect. Experimentally test for extra mortality caused by Snake River dam passage by comparing smolt to adult survival of fish that pass 4 dams (Columbia) compared to 8 (Snake and Columbia). This could help explain why despite the substantial gains realized in direct smolt survival, adult return rates of Snake River spring/summer chinook salmon have not increased.

NMFS proposal. Comment is inappropriate

Already ESA Required?


Fund (Tier 1)
Jun 11, 2003


1. Council Staff preferred projects that fit province allocation

Correct budgets are needed.

Aug 4, 2003


The only project that explicitly addresses extra mortality (one of the primary critical uncertainties identified in the Biological Opinion). Budget revised by BPA.
NW Power and Conservation Council's FY 2006 Project Funding Review
Funding category:
May 2005
FY05 NPCC start of year:FY06 NPCC staff preliminary:FY06 NPCC July draft start of year:
$1,200,000 $1,200,000 $1,200,000

Sponsor comments: See comment at Council's website