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
A Conceptual Spawning Habitat Model to Aid in ESA Recovery Plans for Snake River Fall Chinook Salmon

BPA project number   9406900

Business name of agency, institution or organization requesting funding
Battelle Pacific Northwest National Laboratories

Sponsor type   WA-Federal Agency

Proposal contact person or principal investigator

 NameDavid R. Geist, Research Scientist
 Mailing addressBattelle, Pacific Northwest National Laboratories
MS K6-54
P.O. Box 999
Richland, WA 99352
 Phone509/372-0590

BPA technical contact   ,

Biological opinion ID   

NWPPC Program number   7.5B.5

Short description
Investigate the physical habitat features that influence fall chinook salmon spawning site selection in the Hanford Reach of the Columbia River, and in cooperation with the other agencies, apply this information to Snake River fall chinook salmon recovery efforts.

Project start year   1994    End year   1998

Start of operation and/or maintenance   0

Project development phase   Implementation

Section 2. Narrative

Related projects
Project 91-029, Identification of the spawning, rearing, and migratory requirements of fall chinook salmon in the Columbia River Basin, U.S. Fish and Wildlife Service/National Biological Service.
Project 92-046, Upstream passage, spawning, and stock identification of fall chinook salmon in the Snake River, Washington Department of Fish and Wildlife.
Project 94-034, Assessing summer and fall chinook salmon spawning, incubation, growth, and outmigration timing in the upper Clearwater, lower Salmon, Grande Rhonde, and Imnaha rivers, Nez Perce Tribe.
Each of these projects is assessing the habitat requirements of adult fall chinook salmon spawning. Information that is collected as part of our project will directly benefit in the interpretation of results from these studies (see below).

Project history
This project was funded to compliment the research that is being conducted on the Snake River fall chinook salmon by the US Fish and Wildlife Service (USFWS). Prior to this project being funded, the research proposal was presented to the Snake River Fall Chinook Salmon Coordination Committee, and was reviewed by the USFWS, National Biological Service, Oregon Department of Fish and Wildlife, and Washington Department of Fish and Wildlife (WDFW). Reviewers of the study supported the project because it compliments other research efforts presently underway in the Snake River basin. The project is designed to provide information to the National Marine Fisheries Service to be used in their efforts to prepare recovery goals for fall chinook salmon.
One goal of the Snake River fall chinook salmon recovery effort is to develop recovery plans that are based on reasonable estimates of production potential. This requires that fall chinook salmon spawning habitat be located and characterized. This study was proposed and funded as an effort to build upon existing studies that had been conducted in the Snake River in order to provide reasonable estimates of the potential spawning habitat for Snake River fall chinook salmon.
The low population size and study limitations imposed on protected stocks of Snake River fall chinook salmon hinders the implementation of some research projects. In contrast, fall chinook salmon populations that spawn and rear in the Hanford Reach have remained relatively stable over the last decade (Dauble and Watson 1990). Thus, our approach has been to use Hanford Reach fall chinook salmon as a study surrogate to Snake River fish. This was approved by the funding agency (BPA), as well as by the management and research agencies because:
(1) the Hanford Reach population is larger than the Snake River stock. This not only reduces the likelihood of harassing an endangered fish, but provides a wider representation of the conditions important for fall chinook spawning that would otherwise not be apparent in studying the Snake River population;
(2) the life-history pattern of both stocks are similar. Both stocks spawn in October and November. Emergence for both stocks occurs from March to May, and usually peaks in late April - early May. After emergence they spend from 1-3 months in the mainstem, and migrate to the ocean in July.
(3) there is over 40 years of monitoring data from the Hanford Reach, including analyses of flood plain, surface water, and groundwater processes, ecological processes, and fall chinook spawning habitat characterization. This information is necessary to develop a model of habitat selection that takes into consideration multi-dimensional aspects of river processes. This experience will be useful in directing future data needs for the Snake River.
(4) mainstem habitat characteristics of the riverine environment of both stocks are similar. The Hanford Reach is the last free-flowing section of the mainstem Columbia River and contains 51 miles of diverse habitat that is similar to that available in the Snake River downstream from Hells Canyon Dam. In both sections, upstream flow regulation influences habitat availability.
(5) on-going studies on the Hanford Site, currently funded for other purposes, are providing unique opportunities to transfer technology to the Snake River and other locations where anadromous fish populations are depressed. Presently, studies are being conducted to characterize groundwater input to the Columbia River and techniques developed to monitor the interaction of sub-surface flows and surface water will be useful in identifying the relationships between fall chinook spawning and groundwater processes.

Biological results achieved
The physical characteristics (depth, substrate, velocity, and slope) of two fall chinook salmon spawning areas in the Hanford Reach have been described based on data collected in 1994 and 1995. Comparisons to fall chinook salmon spawning areas in the Snake River are presently being conducted. Preliminary results suggest that these “simple” descriptors of habitat do not adequately describe the amount of fall chinook spawning that occurs in the Hanford Reach. This suggests that other descriptors are needed to provide accurate estimates of potential habitat for fall chinook salmon spawning in large rivers. We have initiated efforts to describe the spatial and temporal relationship(s) between geomorphic features of river channels (i.e., meander lengths, geologic land forms, gravel bars, etc.), channel hydraulics, and salmon spawning locations.
One characteristic that appears to be critical for successful reproduction in many salmonids is sub-surface (also called the hyporheic zone) flow. In 1995, new techniques were developed for installing devices to monitor the hyporheic zone near fall chinook spawning areas in the Hanford Reach. This was previously not possible because of the difficulty in installing monitoring devices into the bottom of rivers with large substrate. Differences in hydraulic head between the hyporheic zone and water surface of the river were found. Further investigation into the relationship of sub-surface flows to reproductive success of fall chinook salmon is needed and proposed here.

Annual reports and technical papers
The installation of piezometers in large cobble substrate of large rivers. D. R. Geist, M. Joy, and D.R. Lee. Paper that will be presented at the North American Benthological Society Meeting in Kalispell, Montana, June, 1996.
Mapping groundwater discharge in the Hanford Reach of the Columbia River. D.R. Lee, D. Hartwig, D.R. Geist, T.A. Cooper, and K.A. Saldi. Paper that will be presented at the Canadian Geophysical Engineers Meeting, Canada.
Measuring current velocity, discharge, and river bathymetry with an acoustic doppler current profiler. D.R. Geist and G.E. Johnson. Paper that will be presented at the North American Benthological Society Meeting in Kalispell, Montana, June, 1996.
Rebuilding strategies for endangered stocks of Pacific salmon. D.D. Dauble and D.R. Geist. Paper that will be presented at the 2nd annual World Fisheries Conference, Brisbane, Australia, July, 1996.

Management implications
Considerable effort is presently underway to rebuild and enhance native salmon and steelhead populations in the Columbia River basin. In order to formulate recovery goals and prepare recovery plans for Snake River salmon stocks, the production potential of salmon and steelhead under existing operating conditions is needed (SRSRT 1994; NMFS 1995a). Snake River salmon recovery plans include habitat protection and restoration (SRSRT 1994; NMFS 1995b). However, recovery efforts cannot be developed until we increase our understanding of the factors that limit the various life history stages (Rondorf and Miller 1993; SRSRT 1994). Presently there is a lack of information on the physical habitat features that limit production of salmon in large, free-flowing rivers (NPPC 1994; SRSRT 1994; NMFS 1995a). This study is attempting to identify the physical factors that influence spawning of fall chinook salmon in the Columbia River Basin.
Water velocity, substrate size, and water depth are the principle variables used to describe mainstem spawning habitat for fall chinook salmon. However, these habitat characteristics vary considerably, both between and within major spawning areas. Suitable spawning habitat is more limited in many rivers than superficial observations of depth, velocity, and substrate would suggest. Models that predict suitable spawning habitat based on only these simple measures may tend to overestimate production potential. Inaccurate estimates of chinook salmon production may lead to unrealistic recovery goals for listed stocks.
Our general approach has been to use the fall chinook salmon of the Hanford Reach of the mid-Columbia as an analog of Snake River stocks. We are presently using characteristics of spawning habitat (both “simple descriptors” and more complex variables) from the Hanford Reach fall chinook stock to validate a model that the USFWS uses to predict fall chinook spawning habitat in the Snake River. This validation process is expected to result in a better model that can be used to provide more accurate assessments of production potential which will lead to reasonable recovery goals for Snake River fall chinook salmon.

Specific measureable objectives
Examine if the spatial and temporal variability in the vertical hydraulic gradient of the hyporheic zone is correlated with the spatial and temporal distribution of fall chinook spawning locations in the Hanford Reach.
Specific questions this research will attempt to answer: Can the spatial and temporal variability of the vertical hydraulic gradient of the hyporheic zone be used to explain the relationship between landform features and channel hydraulics that are important to where fall chinook salmon redds are located? Can this relationship be used to predict where future fall chinook salmon spawning will occur? Are the geomorphic processes in the Snake River similar to the Hanford Reach, and if so, would an evaluation of the spatial and temporal variability in the vertical hydraulic gradient of the hyporheic zone improve production estimates of fall chinook salmon?

Testable hypothesis
(1) There are no differences in the measurements of depth, substrate, and velocity in areas where fall chinook salmon spawn and areas where they do not spawn. (2) There are differences in the vertical hydraulic gradient of the hyporheic zone between locations where salmon spawning occurs and locations where there is no spawning. (3) The physical and chemical characteristics of the hyporheic zone are important in the reproductive success of fall chinook salmon.

Underlying assumptions or critical constraints
The location of fall chinook salmon nests can be determined using aerial photography and/or underwater video technology. There are no underlying geologic features that would prevent the installing of piezometers. The hydrology of 1996 and 1997 is similar to the average hydrology of 1994 and 1995.

Methods
We propose to continue investigating two study sites in the Hanford Reach that were monitored in 1994 and 1995; a high density fall chinook salmon spawning location and a low density fall chinook spawning location. Each study site is ~3.5 km long and 300-400 m wide. The basic physical habitat features at each of these sites were described using data collected in 1994 and 1995. Further investigation into the relationship of sub-surface flows to reproductive success of fall chinook salmon is proposed here.
We are proposing to continue monitoring existing piezometers (~35 between the two sites) and install approximately 50 more in 1996-1997. Piezometers consist of a perforated steel pipe (dia. ~4 cm) that is driven into the river bottom approximately 2 m. Measurements of water surface elevation, conductivity, and temperature will be taken within the piezometer and within the river immediately adjacent to the piezometer. These measurements will be taken using a method randomly stratified over the sampling period (year-round). At key locations, sensors attached to data loggers will be deployed within piezometers and adjacent river stations to continuously monitor changes in elevation, temperature, and conductivity. This information will be used to calculate vertical hydraulic gradients and predict groundwater flow lines. Data on elevation changes will be used to create contour maps of hyporheic elevations in the spawning areas.
Data on river elevation (stage), water depth, and water velocity will be collected along eight channel cross-sections at each site during a low and high flow. Near-bed and mean column velocities will be collected using an acoustic Doppler current profiler (ADCP). This information will be used to improve existing stage-discharge relationships at each site. Hourly discharge data is also available from Priest Rapids Dam.
The physical features of the river channel in the vicinity of the spawning areas will be used to model complex hydraulic variables. For example, complex hydraulic variables may include turbulence in the free flow (Froude number), turbulence close to the river bottom (Reynolds number), force of flow prevailing at the stream bottom (Shields criterion/critical shear stress), and a measure of substrate roughness (Manning roughness coefficients).
Aerial photographs of fall chinook salmon spawning locations will be taken at weekly intervals beginning in mid-October continuing through mid-November. Locations of deep water redds will be mapped using underwater video and Global Positioning System technology. Redd locations will be digitized into an ArcInfo Geographical Information System.
Data from the investigation of the channel hydraulics and hyporheic zone will be compared to spawning locations using quadrat methods and/or regression analysis.
Data layers for salmon spawning and measures for the physical variables will be compared using spatial models developed in a GIS (e.g., ArcInfo). A likely scenario would involve using aerial photographs of salmon spawning from 1991, 1994, and 1995 to develop a relationship with explanatory variables, and then use this relationship to predict spawning locations in 1996. This assumes physical variables are constant year to year at any given discharge. Aerial photographs taken in 1996 would be used to test the accuracy of the predictions. Goodness of fit (e.g., chi square) tests will be done to compare how well predictions match observed (Ludwig and Reynolds 1988).

Brief schedule of activities
FY 97 Schedule:
October, 1996: Install piezometers within study sites.
October, 1996 - April, 1997: Monitor the piezometers; measure channel cross-sectional depth, slope, and velocity during low and high flow period; take aerial photographs of salmon spawning during 1996 spawning period.
May - August, 1997: Digitize aerial photographs and perform spatial and temporal analysis of salmon spawning relative to physical features monitored above.
September, 1997: Prepare annual report.
Out-year activities:
These would be dependent on results obtained in FY 1997 and would likely include further investigation of specific findings from FY 97 following a similar schedule as that in FY 97.

Biological need
Water velocity, substrate size, and water depth are the principle variables used to describe mainstem spawning habitat for fall chinook salmon. However, these habitat characteristics vary considerably, both between and within major spawning areas. Suitable spawning habitat is more limited in many rivers than superficial observations of depth, velocity, and substrate would suggest. Models that predict suitable spawning habitat based on only these simple measures may tend to overestimate production potential. Inaccurate estimates of chinook salmon production may lead to unrealistic recovery goals for listed stocks. Improved characterization of potential spawning sites could also lead to better protection measures for critical habitat.
Our general approach for this project has been to incorporate into existing models the characteristics of fall chinook salmon spawning habitat that are measured at various spatial and temporal scales. One characteristic that appears to be critical for successful reproduction in many salmonids is hyporheic flow. Investigating the relationship between hyporheic flow and fall chinook salmon spawning has previously not been done. We made significant progress toward this effort in 1995. However, more information is needed before a clear relationship can be established.
Therefore, our effort in FY97 will focus on the relationships between the spatial and temporal variability of hyporheic flow and salmon spawning locations. We will attempt to explain how land-form features and the channel hydraulics within salmon spawning site are related via the hyporheic zone. We will then evaluate the use of these findings to improving management of Snake River fall chinook populations.

Critical uncertainties
Spawning habitat in large rivers is more limited than superficial observations of depth, substrate, and velocity would suggest. Inaccurate estimates of the amount of spawning habitat that is available to fall chinook salmon would lead to unrealistic recovery goals.

Summary of expected outcome
This study is designed to provide information toward increasing our understanding of the factors that limit production of fall chinook salmon in large rivers. This information will be essential in defining critical habitat areas and recovery plans for salmon stocks throughout the basin, especially Snake River fall chinook salmon.

The multi-dimensional aspects of river systems are not well understood. We believe this study will provide evidence that rivers are not bound systems, and that connectivity within the landscape can be used to improve our definition of salmon habitat. This study will provide a mechanism to include these aspects in future efforts to model production potential of salmon.

Although the hyporheic zone has been studied, little is known about the importance of the hyporheic zone to chinook spawning habitat. This study is expected to provide a better understanding of the geomorphic processes (including hyporheic flow) in a large river. In the process of studying these processes, new techniques that can be used to analyze spatial data and examine relationships between hyporheic flow and fish habitat will also be developed.

Dependencies/opportunities for cooperation
No actions or events that would affect the timing of this project are known at this time.

Risks
No risks associated with implementing this project are known at this time.

Monitoring activity
Inherent in the project design is a comparison of predicted to actual spawning. The accuracy and robustness of the model in making these predictions will be a primary measure of the project’s outcome.

Section 3. Budget

Data shown are the total of expense and capital obligations by fiscal year. Obligations for any given year may not equal actual expenditures or accruals within the year, due to carryover, pre-funding, capitalization and difference between operating year and BPA fiscal year.

Historic costsFY 1996 budget data*Current and future funding needs
1994: 229,770
New project - no FY96 data available 1997: 165,000
1998: 165,000

* 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 2 - fund when funds available

Recommended funding level   $165,000

BPA 1997 authorized budget (approved start-of-year budget)   $165,000