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
Individual - Based Approach to Analyzing Pit-Tag Data

BPA project number   5500500

Business name of agency, institution or organization requesting funding
U. S. Department of Energy

Sponsor type   TN-Federal Agency

Proposal contact person or principal investigator

BPA technical contact   Tom Morse, EWN 503/230-3694

Biological opinion ID   Numbers 13, 13a, 13f, 17

NWPPC Program number   

Short description
The project will examine the efficacy of an individual-based analytical approach and models, in contrast to a group-based, statistical approach, for obtaining useful information from PIT-tag mark-recapture data in the mainstem Snake and Columbia rivers.

Project start year   1997    End year   1998

Start of operation and/or maintenance   0

Project development phase   Implementation

Section 2. Narrative

Related projects
93-29 (Survival estimates for the passage of juvenile salmonids
through Snake River dams and reservoirs). This project uses
Cormack-Jolly-Seber group statistical methods for obtaining survival
information from PIT-tagged fish. The individual-based approach uses the
inherent individualized data of PIT tags to better advantage and may lead
to very different types of information, as attested by the usefulness of
individual-based models for other fisheries studies.
The entire PIT-tag program is related in that the data from it can
be approached from the individual perspective for an alternative data
analysis approach.
BPA is a member of EPRI, which has supported development of the
Individual-based approach to fish population models at ORNL (COMPMECH).

Project history

Biological results achieved

Annual reports and technical papers

Management implications

Specific measureable objectives
The objective is to extract new and novel information from existing
PIT-tag data on the Snake and Columbia rivers by use of an individual-based
approach, which more fully uses the inherently individual-based PIT tags,
rather than currently-used group statistical mark-recapture analytical
approaches. This new information can be synthesized into individual-based
models that may be a more realistic presentation of population dynamics
during outmigration than current statistical models.

Testable hypothesis
The hypothesis is that new and novel information on survival,
growth, migration rates, and other features of the downstream migration of
juvenile salmonids can be obtained when data from the basin-wide
mark-recapture tagging of migrants with PIT tags is analyzed from an
individual perspective rather than from group statistics.

Underlying assumptions or critical constraints
A massive mark-recapture experiment is underway in the Columbia
River basin. Juvenile salmonids are individually marked with unique
identifiers in the form of passive transponder (PIT) tags and then
"recaptured" (often several times) at detection stations at several
downstream dams. The mark and detection data are stored in a centralized
computerized data file accessible for data analysis. There have been
several years of PIT-tagging and more such work is conducted each year.
Although tagging is inherently individual-based because of the unique fish
identifier, most analyses have used group mark-recapture statistics. For
example, a large effort is now underway to test the use of
Cormack-Seber-Jolly group statistics for estimating survival of juvenile
salmonids during outmigration.
There is a current perspective in ecology that more useful
information is obtainable from treating ecological processes as embodied in
the responses of individuals rather than of groups. Individual-based
models, pioneered at ORNL, are seeing increased use in analyzing ecological
information ranging from plant communities to fish populations. Use of
these models has given insight into the new and novel information.
Without conducting additional PIT tagging, it should be possible to
use the existing (and continually obtained) PIT tag data for application in
individual-based analyses. After the empirical data are analyzed from an
individual-based perspective, then appropriate individual-based models can
be constructed to represent juvenile salmonid migrations.

Methods
Response must include 1) brief experimental design including a
description of equipment, techniques and materials; 2) statistical
analysis; 3) type and number of fish to be used.

The project will be undertaken in three phases. The first phase
will entail ORNL's individual-based modelers becoming familiar with the
available PIT tag database. Details such as marking times and places,
recovery times and places, and other biological data will give these staff
insights into the kinds of novel information that could be extracted from
the database.
The second phase will entail analysis of selected PIT tag data for
features that are particularly amenable to an individual-based approach.
These analyses should yield new information about characteristics of the
migration. A third phase would entail development of a preliminary
individual-based model of juvenile salmonid outmigration in the
Columbia-Snake river basin.
The number of fish "used" would be limited only by the number of
data records in the database. No new fish would be tagged. We anticipate
using data on fall chinook salmon underyearlings, spring chinook yearlings,
and steelhead yearlings. Other species may be used, if sufficient records
are available.

Brief schedule of activities
October-December 1996: Familiarize staff with the PIT-tag database.
January-July 1997: Conduct data analyses using an individual-based
approach.
August-September 1997: Summarize work to date in progress report
and plan the next year's analyses.
October 1997-July 1998: Continue PIT-tag analyses of selected
topics and construct preliminary individual-based models of these
processes.
August-September 1998: Prepare final progress report and
manuscript(s) for open-literature publication.

Biological need
For preservation and recovery of endangered salmonids in the basin,
we desperately need to better understand the basic processes that occur
during juvenile outmigration, such as the relationships between flows,
migration rates, and survival. Experience with many ecological systems,
including fish populations, has shown that an individual-based approach can
offer new and novel insights into such important processes. The current
approach to improving our understanding of these processes is largely group
statistics. Because it is inherently individual-based, the PIT-tag
methodology of mark and recapture in the basin would lend itself to an
individual approach.

Critical uncertainties
The critical uncertainty is how much of a gain in knowledge about
juvenile salmonid migrations we might derive from an individual-based
approach to PIT-tag analyses and information synthesis. We do not know
with certainty whether this approach is better than group statistics for
developing our understanding of fish migrations (and thus having a better
knowledge base for management decisions).

Summary of expected outcome
There should be two principal outcomes from the use of this
approach with existing PIT-tag data. One would be a judgement about
whether an individual-based approach to PIT-tag mark-recapture data is
useful for getting new insights into relevant questions about fish
migrations. The second would be specific gains in knowledge about juvenile
salmonid growth, survival, and movement during ourmigration.

Dependencies/opportunities for cooperation
The ORNL project would depend on the PIT-tag database for its
primary information. This, in turn, is dependent on the field tagging and
detection activities of many investigators, many funded by BPA. There
would be a period of instruction of ORNL staff required from the custodians
of the PIT-tag database. Individual-based modeling and data analysis
techniques are available at ORNL from on-going projects. Because
individual-based modeling is being undertaken for other purposes at
in-basin locations (e.g., Jim Peterson at the NBS Cook lab), collaborations
with scientists there could provide rapid transfer of techniques.

Risks
There would be no risks to either fish populations or humans from
this new analysis of existing data.

Monitoring activity
The project's outcomes would be monitored by annual progress
reports, presentations to interested parties, and open-literature
manuscripts. The ultimate outcome of the project will be monitored by the
usefulness of the analytical approach for obtaining new insights into
migration processes that aid in managing the Columbia-Snake river basin for
protection and recovery of endangered salmonids and for other multiple
purposes.

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   System Policy

Recommendation    Tier 3 - do not fund