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
Gas Bubble Disease Monitoring and Research of Juvenile Salmonids

BPA project number   9602100

Business name of agency, institution or organization requesting funding
National Biological Service

Sponsor type   

Proposal contact person or principal investigator

 NameDr. Alec G. Maule
 Mailing addressColumbia River Research Laboratory
5501A Cook-Underwood Road
Cook, WA 98605

BPA technical contact   Pat Poe, EWI 503 230-4043

Biological opinion ID   NMFS BO RPA 2, 11, 16, 17, 18

NWPPC Program number   

Short description
This project is composed of a monitoring objective, a laboratory research objective and a field research objective: I. At dams and/or in-river, we will non-lethally monitor juvenile chinook and steelhead for signs of GBT. II. We will conduct laboratory experiments to determine the progression and diminution of GBT signs in juvenile chinook and steelhead exposed to water with high total dissolved gas (TDG) . We will also determine the effects of water with high TDG on the ability of juvenile salmonids to resist disease, respond to stress and develop full smolt characteristics. III. We will use a miniature pressure-sensing radio tag to determine the exposure histories of individual juvenile steelhead migrating through reservoirs with high TDG.

Project start year   1997    End year   2000

Start of operation and/or maintenance   

Project development phase   Implementation

Section 2. Narrative

Related projects
This study was part of work conducted under Objective 6 (“Conduct research and monitoring of GBT in juvenile salmonids migrating in the Columbia and Snake rivers during times of high TDG”) of BPA project 87-401, “Assessment of smolt condition for travel time analysis”.

“Symptoms of gas bubble disease induced in salmon by gas supersaturation” (Tom Backman and Dennis Rondorf) . We will work in conjunction with this study (investigating the signs of GBT in purse-seined fish and the distribution of TDG and fish) to determine if (1) fish from the smolt collection systems of hydropower projects are the equivalent of fish in the river and (2) and if fish implanted with pressure-sensitive transmitters behave similarly to untagged fish in depth and location in the reservoir using hydroacoustics. This study had not been funded at the time of this writing.

Project history

Biological results achieved
Under project 87-401, we developed a non-lethal method for assessing GBT, monitored fish at dams in 1995, and began work on some objectives of this project. In 1996, we monitored fish at seven dams, continued lab work and began work on the pressure-sensitive radiotag.

Annual reports and technical papers
New project. Draft 1995 reports on laboratory work and monitoring are available.

Management implications
I. Monitoring GBT signs in salmonids provides biological criteria upon which water and fisheries managers can base in-season spill regulation.

II. Results of the laboratory research (development of GBT signs and the sublethal effects of exposure to water with high TDG) will provide information upon which managers can base monitoring protocols and biological criteria for water regulation. Analysis of physiological data and behavioral data (Objective III) can be used in a model to predict the level of mortality expected based on water quality and GBT signs observed in migrating fish.

III. Results of this project can be used as a base of information about the depths used by individual juvenile steelhead in the Columbia River. This knowledge is essential to understand the effects high TDG may have on these animals. Each meter of depth allows compensation for approximately 10% of supersaturation (Weitkamp and Katz 1980). Data from this study can be used 1) as a model of individual fish exposure to TDG that can be duplicated in laboratories to determine the risk different levels of TDG have on juvenile salmonids migrating in the river; 2) as a source of a horizontal and vertical component in geographical information systems (GIS) housing TDG data for the prediction of the risk of TDG on juvenile salmonids; 3) as a source of definitive fish depth information in reservoirs and near dams for the design of surface collectors for the passage of fish around dams; and 4) to refine current survival models such as CRiSP, the TDG component of which is based on a simplistic, untested hypothesis of a constant depth during migration.

Specific measureable objectives
I. Non-lethally determine the prevalence and severity of GBT signs in juvenile salmonids migrating in the Snake and Columbia rivers and report data to the FPC on a daily basis from April through August.

II. (1) Determine the progression and diminution of external and internal signs of GBT in chinook salmon and steelhead held in shallow tanks containing water with high TDG (110, 120 & 130%) at various temperatures (8, 12, 15 C). (2) Determine the sub-lethal effects of exposure to gas supersaturated water on the ability of juvenile chinook salmon to resist disease, respond to stress, and develop full smolt characteristics.

III. Determine the vertical and horizontal distribution of juvenile steelhead migrating in the Snake and Columbia rivers at times of high TDG. (1) Test a newly developed depth-sensing miniature radio transmitter for use in juvenile salmonids. (2) Determine the horizontal and vertical distribution of individual migrating juvenile steelhead in a reservoir. (3) Determine the near-dam horizontal and vertical distribution of migrating juvenile steelhead.

Testable hypothesis
I. Non-lethal measures of GBT can be used to estimate risk of mortality in migrating salmonids.

II. (1) The progression of GBT signs leading to mortality is predictable for given species of fish under specific conditions of TDG, temperature and available depth.
(2) Juvenile salmonids ability to resist disease, respond to stress, and develop smolt characteristics is affected by exposure to water with high TDG

III. (1) The radio transmitter consistently indicates the correct depth to the nearest 0.3 meters.
(2) Depths of juvenile steelhead are greater than the compensation depth, based on TDG at fish locations.
(3) Juvenile steelhead migrate deeper in the water column in areas of high TDG than in areas of low TDG.
(4) Depths of migrating juvenile steelhead are constant.

Underlying assumptions or critical constraints
River management will continue in its current form requiring the monitoring of migrating fish to ensure operations are optimal for fish survival. The Smolt Monitoring Program continues to monitor emigrating salmonids. Spill at dams and regulation of water flow continue to be management actions. We will continue to have test fish available for laboratory and field studies. TDG will continue to be greater than 100% in the Snake and Columbia rivers

I. GBT Monitoring
(1) Develop methodologies, protocols and experimental designs to monitor and conduct research on gas bubble disease in juvenile salmonids migrating through the Snake and Columbia rivers. This objective was originally added to project number 87-104 prior to the 1995 juvenile salmonid migration season. We examined juvenile spring/summer chinook salmon and steelhead for external signs of GBT at Lower Granite, Little Goose and Lower Monumental dams on the Snake River and Rock Island, McNary, John Day, and Bonneville dams on the Columbia River three days per week during the outmigration of juvenile salmonids. Fish are collected at the separator, anesthetized, examined, and released back to the collection system at each dam. Dissecting microscopes with magnification of 10X-40X are used to examine the unpaired fins, opercula, eye, and lateral line for bubbles.

(2) The examinations at each site are completed through cooperation of one employee from the National Biological Service and one employee from the Smolt Monitoring Program. Data recorded on a data sheet is transferred to a computer, checked for correctness, and transferred electronically to the Fish Passage Center daily.

(3) A total of 200 fish are examined at each site on each day. This number is composed of no more than 100 juvenile yearling or subyearling chinook salmon and 100 juvenile steelhead. A sample size of 100 results in an estimate of GBT in the population at large with an associated 95% confidence interval of ±6%. Based on this plan, we will examine approximately 5500 juvenile steelhead and 10,500 juvenile chinook salmon each year.

II. GBT Laboratory Research

(1) In the laboratory, we will conduct several studies over the next five years: (1) assess the progression and severity of GBT in juvenile salmonids exposed to various TDG levels and water temperatures; (2) assess the effects of GBT on infectious disease progression and stress responses of juvenile salmonids; and (3) assess the effects of GBT on smolt development and ability to survive in saltwater. For all of these studies, we will use fish obtained from local hatcheries (about 2000-3000 fish annually). For study #1, we will expose groups of fish to standard TDG levels and water temperatures. During their exposure, fish will be sampled at selected time intervals and examined for progression and severity of GBT in the lateral line, fins, body surface, and gills. Data will be used to construct time series analyses of GBT progression and to relate the severity of GBT signs to the onset of mortality. For study #2, we will expose fish to high TDG and subsequently challenge them with a dose of BKD. We will then assess the effects of TDG exposure on BKD progression and mortality. As a corollary, we will expose fish that already have BKD to high levels of TDG and assess the progression of signs and subsequent mortality. Finally, for study #3, we will expose fish that are actively smolting to high levels of TDG. Fish will be sampled for physiological indicators of smoltification and also examined for the progression and severity of GBT. As a final test, seawater challenges may be done to assess the effects of GBT (and perhaps BKD) on the ability of juvenile salmonids to survive in seawater.

(2) Much of the data generated by the progression of GBT signs research is descriptive in nature; hypothesis testing is not relevant. Within each time interval, we will average lateral line and gill data, determine standard errors and prevalence, and plot the data over time. For the fins, we will plot average and maximum severity rankings and prevalence over time using data from all fins combined or data from selected fins. Mortality will be plotted as a cumulative percentage over time. We will fit a curve through the points by eye and estimate the time to 50% mortality (i.e., the LT50) by extrapolation. We will examine the potential relationship between GBT signs and mortality by using correlation and regression analysis. At certain levels of mortality, e.g. the LT5 or LT10, we will characterize the average GBT signs in the sample population at that time.

We will also determine sublethal effects of exposure to supersaturated water by conducting disease challenges similar to that described by Maule et al. (1987, 1989) for Vibrio anquillarum and that developed by R. Pascho (NBS, Northwest Biological Science Center, Seattle) for Renibacterium salmoninarum. Physiological measures of smoltification and stress will be monitored and assayed based on protocols described by Beeman et al. (1994, 1995), Haner et al. (In press) and Schrock et al. (1994).

(3) Fish to be used in laboratory research will be hatchery chinook salmon and steelhead and will most likely be obtained from lower-river federal hatcheries.

III. Vertical and Horizontal Distribution

(1) Much of the work in 1997 will be of a preliminary nature as this depth tag was recently developed and has not been field tested. Laboratory experiments will be conducted prior to field work to determine the best method of attachment/implantation and if the radio transmitter affects the buoyancy of tagged fish. Tests of reported depth versus the actual pressure of tags in fish and buoyancy of tagged and untagged fish will be conducted. The transmitter itself will be evaluated to determine accuracy and precision of reported depths using computer-controlled pressure chambers at Battelle Pacific Northwest Laboratories (Richland, WA) in cooperation with Dr. Tom Carlson.

Juvenile steelhead of hatchery origin will be tracked through the Ice Harbor Dam-to-McNary Dam reach. This reach was selected because gradients of TDG occur due to the different TDG levels in the Columbia and Snake rivers. Fish collected at McNary, Lower Monumental, or Ice Harbor dams will be implanted with pressure-sensitive radio transmitters. Fish will be sequestered in the Ice Harbor Dam tailrace at depths sufficient to prevent gas bubble disease (GBT; approximately 5 m) for a period of at least 24 hours. Fish releases will be made during spill (evening hours) and non-spill (daylight hours) in an effort to track fish in “plumes” of water with high and low TDG.

We propose to release five-to-ten fish every three-to-five days. This schedule is based on a median travel time of approximately three days between Ice Harbor and McNary dams (estimated from Fish Passage Center 1993, 1994). The actual number of releases will depend on the travel times of the tagged fish in each group. We plan to release a maximum of 100 fish during the study period in each year.

Tagged fish will be monitored in the Snake River below Ice Harbor Dam and in McNary reservoir from boats and in the forebay near McNary Dam with telemetry equipment mounted on the dam. The equipment at McNary Dam will provide detailed information about fish near the dam and will serve to confirm when fish have left the study area.

Telemetry equipment used from boats will consist of one receiver and 6-element antenna on each of two boats. Additional equipment will include global positioning systems and total dissolved gas meters. We will track fish using two boats each operating two 10 h shifts per day from the time of release until the time the fish pass McNary Dam. We will use a tracking protocol designed to maximize the number of fish contacts as well as the amount of data collected from individual fish. We will alternate between attempting to locate as many fish as possible (Method 1) and intensely following individuals (Method 2). For example, we may spend several hours locating as many fish as possible and then spend 40 minutes following one randomly-selected individual. This process will be repeated until the fish have passed McNary Dam. This method will provide large amounts of data about within-fish variability in vertical and horizontal movements of individuals while allowing us to measure between-fish variability from a larger number of fish than if we intensely tracked a small number of fish.

At each fish contact, location via a global positioning system (GPS), TDG, water temperature, and fish depth will be recorded. When fish are contacted using Method 1 depths will be recorded over a period of 15 minutes with TDG measured at the end of this period. This time is based on the minimum time required for the TDG meter to equilibrate. We will collect several hundred observations of fish depth during this time. Data will be collected continuously when individuals are located using Method 2. All efforts will be made to avoid affecting fish behavior during data collection by keeping an adequate distance from the fish during tracking. Total dissolved gas meters will be tested on a regular schedule (at least weekly) to ensure they are operating within established criteria.

The equipment mounted at the dam will consist of five data-logging telemetry receivers collecting data from 4 to 8 4-element yagi antennas each. The antennas will be placed approximately 60 m apart across the earthen dam on the Oregon shore and between the navigation lock and the Washington shore, and at every other turbine unit on the powerhouse and every other spill gate across the spillway. This configuration will enable us to receive data from fish located up to approximately 400 m upstream from the dam over the entire length of the dam.

Hydroacoustics will be used to determine if the depths and locations of tagged fish represent other fish in the reservoir. The horizontal and vertical distribution of tagged fish will be compared with locations of other fish in the reservoir identified using hydroacoustics to relate observational data on individual fish and the distribution of the population at large. Hydroacoustic information will be provided by a study titled “Symptoms of gas bubble disease induced in salmon by gas supersaturation”, headed by the NBS (Dennis Rondorf, Cook, WA) and the Columbia River Intertribal Fish Commission (CRITFC; Tom Backman, Portland, OR). No further details of this cooperation are known at this time because the hydroacoustic project had not received funding as of this writing.

(2) Data collected will be analyzed using several methods. Distributions, measures of central tendency and associated errors in fish depth and TDG will be calculated for each fish and correlations between fish depth and TDG from the fish monitored using Method 2 will be examined for statistical significance. Within-fish variability in depth will also be estimated from this data. Data will be examined for diurnal trends and to determine if juvenile steelhead travel at a constant depth during migration or if their depth varies. Previous studies of the vertical distribution of juvenile salmonids indicate a diurnal change in vertical and horizontal distribution (Smith 1974; Ledgerwood et al. 1991). Correlations between fish depth, TDG, reservoir depth, and reservoir temperature will be examined from data collected using Method 1. Between-fish variability in depth will be estimated from this data. Fish tracks will be plotted using a GIS.

Estimates of the sample sizes required to achieve a specified level of precision in the depth estimates can not be drawn as the data on which to base such calculations do not exist. We know of no previous studies using this method to assess the depth of juvenile salmonids. All depth-sensing radio transmitters have previously been too large for use in juvenile salmonids.

(3) This study will use juvenile steelhead of hatchery origin. The numbers of fish needed for work in 1997, 1998, and 1999 will up to 300 in each year, including fish for laboratory studies; no fish will be required in the year 2000. Fish will be taken from the collection facilities at either McNary, Ice Harbor, or Lower Monumental dams, depending on the availability of fish.

Brief schedule of activities
I. Monitor migrating chinook salmon and steelhead at dams

II. Complete laboratory studies of progress of GBT signs
III. Tag evaluation; release fish and track from boats

I. Monitor migrating chinook salmon and steelhead at dams
II. Begin experiments on effects of high TDS on disease resistance of juvenile salmonids
III. Release fish and track them from boats and with fixed gear at McNary Dam

I. Monitor migrating chinook salmon and steelhead at dams
II. Complete experiments on effects of high TDS on disease resistance of juvenile salmonids and begin experiments on effects on stress response and smolt development.
III. Release fish and track them from boats and with fixed gear at McNary Dam

I. Monitor migrating chinook salmon and steelhead at dams
II. Complete experiments on effects of high TDS on stress response and smolt development.
III. Analyze data and write final report to BPA and articles for publishing peer-reviewed journals.

I. Monitor migrating chinook salmon and steelhead at dams
II. Analyze data and write final reports and publications. Determine feasibility of incorporating data and analyses into predictive model.

Biological need
If we are to regulate flows in the Snake and Columbia rivers to optimize survival of salmonids, we must know what effect water manipulations will have on those fish. Furthermore, we must develop protocols for monitoring fish and establish criteria to determine when to alter water regulation. There is currently much debate over the extent of the role high TDG, from NMFS BO #2, plays in fish survival in the Columbia River basin. Current monitoring programs are based on the examination of passively (collected at dams) and actively (collected with purse seines) captured fish. These programs are needed for monitoring purposes, but do nothing to help understand the root of the issue: How deep are the fish and what is their relative exposure to TDG?. This question is of the utmost importance because fish depth is one of the most important factors determining the presence and severity of GBT. Data on the exposure histories, including depths and TDG at fish locations, will enable researchers and managers to assess and predict the effects of TDG on juvenile salmonids. Repeated measures from individuals is the most appropriate method for this purpose.

Critical uncertainties
I. (1) Fish collected at dam bypass systems and at in-river smolt traps are representative of the fish in the river.
(2) Non-lethal signs of GBT can be used to predict risk.

II. (1) Current technology is adequate to detect significant effects of exposure to GBT.

III. (1) Tagged fish behave as untagged fish in the reservoir.
(2) Radio signals can be detected at depths of tagged fish.
(3) Tagged fish can be located and followed without affecting fish behavior.

Summary of expected outcome
I. Information on the prevalence and severity of GBT signs will help managers make decisions about spill.

II. We will describe the progression of GBT signs leading to mortality in juvenile salmonids. We will determine the risk that exposure to water with high TDS poses for juvenile salmonids.

III. We expect fish depths are not constant during their migration, as variation in biological systems is the rule rather than the exception. We expect approximately 50-75% of juvenile steelhead will be in the upper 10 meters of the water column, based on studies using hydroacoustics or vertical gill nets. We expect we will be able to locate and track fish at these depths with the proposed radio transmitter. The study we propose will produce information about the variation and range of depths used by individual fish; we have no information on which to base predictions of this data. This knowledge is required to understand the exposure of the fish to TDG and possible effects on survival.

Dependencies/opportunities for cooperation
Funding must be received no later than December 1 to allow purchases of non-expendable equipment. Much of this equipment must be ordered 2-4 months in advance of the field work to ensure timely delivery.

There are safety concerns when operating boats in the Columbia and Snake rivers.

The risks inherent in this project will change annually. For example, in 1996 there is a risk that the GBT signs selected for monitoring or the locations where fish are sampled do not accurately reflect problems in the river. Research conducted in our lab and proposed research by others will address that risk and the monitoring program will be adapted as needed.

Monitoring activity
Annual summaries of information will be prepared for BPA, culminating in a final report after the 2000 project year. Articles prepared for publication in fishery journals will receive peer-review.

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
(none) New project - no FY96 data available 1997: 750,000
1998: 586,000
1999: 609,000
2000: 432,000
2001: 450,000

* 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   Mainstem

Recommendation    Tier 1 - fund

Recommended funding level   $750,000

BPA 1997 authorized budget (approved start-of-year budget)   $851,000