| |
Recommendation to Study Effects of Load Following on Juvenile Salmon
Migratory Behavior and Survival
April 29, 2005 | document ISAB 2005-3
Introduction
The ISAB recommends to the Council, NOAA Fisheries, and CRITFC that
during 2005 there be a study of the effects of load following (flow
interruption) on survival of outmigrant smolts in the Snake River and
perhaps the lower Columbia River. We understand that there is a new
Biological Opinion (BiOp) that bears on flow management, that this new
BiOp is being challenged in court, and that there is a specific legal
challenge seeking an injunction to increase flow in the Lower Snake River.
Any or all of these may constrain the ability to manage flow during the
period of outmigration of salmonid smolts this year.
Nevertheless, 2005 presents an opportunity to answer critical questions
concerning the effects of flow interruption, brought about by load
following by the hydrosystem, on survival of migrating juvenile chinook
and steelhead during extreme low flow conditions. This critical question
is not explicitly discussed in the BiOp or the present challenges to it.
The prospects for ?no spill? this summer would simplify the design of
the experiment.
We are sorry that our recommendation comes so late in the season. The
timeliness of inspiration cannot always be regulated. Although it is
probably too late to include spring chinook and steelhead in the study,
and estimating survival of fall chinook with sufficient precision to
detect an effect of load flow interruption may be problematic,
nevertheless the study should proceed, at least as a pilot study. At the
very least it will provide useful information on whether flow fluctuations
do or do not affect migratory behavior.
Full recommendation
To:
Melinda Eden, Chair, Northwest Power and Conservation Council
Olney Patt, Jr., Executive Director, Columbia River Inter-Tribal Fish
Commission
Usha Varanasi, Director, NOAA-Fisheries Northwest Fisheries Science
Center
D. Robert Lohn, Regional Administrator, NOAA Fisheries
From: Eric J. Loudenslager, Chair, Independent
Scientific Advisory Board
The ISAB recommends to the Council, NOAA Fisheries, and CRITFC that
during 2005 there be a study of the effects of load following (flow
interruption) on survival of outmigrant smolts in the Snake River and
perhaps the lower Columbia River. We understand that there is a
new Biological Opinion (BiOp) that bears on flow management, that this
new BiOp is being challenged in court, and that there is a specific
legal challenge seeking an injunction to increase flow in the Lower
Snake River. Any or all of these may constrain the ability to manage
flow during the period of outmigration of salmonid smolts this
year. Nevertheless, 2005 presents an opportunity to answer
critical questions concerning the effects of flow interruption,
brought about by load following by the hydrosystem, on survival of
migrating juvenile chinook and steelhead during extreme low flow
conditions. This critical question is not explicitly discussed in the
BiOp or the present challenges to it. The prospects for ?no spill?
this summer would simplify the design of the experiment. We are sorry
that our recommendation comes so late in the season. The timeliness of
inspiration cannot always be regulated. Although it is probably too late
to include spring chinook and steelhead in the study, and estimating
survival of fall chinook with sufficient precision to detect an effect
of load flow interruption may be problematic, nevertheless the study
should proceed, at least as a pilot study. At the very least it will
provide useful information on whether flow fluctuations do or do not
affect migratory behavior.
The ISAB views the question of fish survival during exceptionally low
flow years as an important one. Although managers tend to think of
extreme drought years as climatic anomalies that occur infrequently,
recent evidence suggests that drought years tend to be somewhat
clustered and that multiple low-flow years can occur over a short time
period. Gedalof et al. (2004), using tree ring data to model
Columbia River flows, concluded that low flow episodes have happened a
number of times since 1750. The following figure from their paper
illustrates the distribution of drought years.
As these authors point out, the period 1950-1987 is unusual in the
context of this flow record in having no multi-year drought
events. Yet the Columbia River experienced a severe drought in
2001 and is apparently experiencing another in 2005. If we are
entering a period of drought-prone years (especially one as severe as in
1835-1850), the issue of migrant survival during exceptionally low
flows, with and without load following, becomes particularly important.
The neglected question, which could be addressed in this experiment,
is whether low in- river survival rates and migration speeds often
observed during extreme low flow conditions are in large part a
consequence of abnormal cycles of flow interruption (and possible
within-pool flow reversal) induced by load following at the dams, or a
consequence of low flow itself (other effects such as temperature and
season aside). The essential hypothesis justifying the experiment is
that a pattern of relatively constant low flow would prove less damaging
to in-river survival rate and migration speed than would the pattern of
load following imposed on the same average flow. This issue was most
recently discussed in the ISAB's report, Review of Flow
Augmentation: Update and Clarification (ISAB
2003-1), which was an
examination of flow augmentation. More detailed information is given
there. The essential design of the experiment would involve temporal
switching from ?treatment? flow management (load following) to ?control?
flow management (constant flow with the same average flow), and
back, at intervals of time that are long enough to satisfy the
requirements of the fish marking method identified as being most likely
to produce the ?best? survival estimates. PIT tags and radio
tags are the two marking technologies that we know are available. The
choice of method for the existing circumstances would best be selected
by agreement among the contractors presently engaged in studies in the
Snake and Columbia rivers that could be modified to collect the relevant
data. Information we have available at this time suggests that
radio tagging would be most likely to produce the desired results this
year. We discuss this suggestion further below.
The essential design issues are:
- Obtaining a good enough match between ?treatment? time blocks
and ?control? time blocks so that extraneous environmental
factors (other than the treatment with load following or the control
with stable flows ) that might operate differently at those
times do not confound the results. Possible confounding factors
include temperature, turbidity, average flow, smoltification status,
and condition of the released fish. The match (and the effects of
mismatch) will be affected by the choice of the time period for the
experiment and the number of treatment/control pairs that can be
established.
- Providing a large enough sample size of releases to allow
survival rates and migration speed estimates to be made with
sufficient precision from the recoveries so that the treatment
effect can be resolved against the noise of sampling variation and
the possible effects of confounding factors (time block effects).
Existing information about survival rates, migration rates, detection
rates, and the variation in these rates should be adequate for arriving
at a design with useful anticipated statistical performance (power). It
would not be worthwhile to proceed with an expensive experiment bearing
on a possibly controversial management issue, unless the design shows
prospects for delivering reasonably conclusive results.
The goal of the experiment is to obtain data that would inform the
region whether flow interruption substantially affects survival or
migration behavior of tagged smolts migrating in the river(s). The study
objectives would include description of behavior and if feasible
estimation of relative survival of tagged groups of hatchery chinook
and/or steelhead: 1) with flow interruption and 2) without flow
interruption. We assume that hatchery fall chinook during summer
would be the best test fish and time period for the study this year.
Prediction of the effects on wild populations would only require the
assumption that the relative rates estimated for the two tagged groups
are approximately the same. If the hypothesis is correct, the experiment
would benefit all in-river migrants, including any wild fish migrating
in-river. The design we outline does not specifically attempt to
generate estimates for unmarked fish migrating in river. Carrying
out the experiment should not raise any ESA issues.
We suggest the design use fixed alternating periods (e.g., with a
length of one week), following a random start to avoid the possibility
of clumping of particular test or control groups at the beginning
or end of the study. This would take into account, insofar as possible
the known fact that survival shows a trend with time due to temperature
and other factors, including degree of smoltification, turbidity, and
volume of flow.
Although PIT tagged fish might be used to estimate survival, there is
a possible complication in that as a batch of test fish moves through
the hydrosystem they tend to migrate at different rates. This is
particularly true for fall chinook. Thus, the PIT tagged fish in a
batch may not all be subject to the same treatment or control. We
suggest using radio tags for the survival estimates, because their use
would make it possible to regroup the fish into treatment and control
groups based upon their known locations through the study period.
In any case, the information provided on their migration routes will
indicate whether there is an effect of the different flow conditions.
We understand that NOAA Fisheries, the USFWS and NPT will be
conducting a transportation study in the Snake River and will be making
weekly survival estimates through the period of outmigration, using
recoveries of PIT tagged fish at detectors located at the dams. These
estimates may be useful as is, but NOAA Fisheries scientists should be
consulted. In addition, we understand that other agency employees will
be conducting radiotracking studies of juvenile salmonids during the
migration period. It would be necessary to bring key agency scientists
together to formulate and approve an appropriate study design. We
foresee little additional costs associated with this study beyond
existing budgets, except in two areas: 1. Responsibility for submitting
to the Council a written study design, and a completion report providing
an analysis of results should be assigned to one or more of the
participating agencies; 2. Monitoring of flows between the dams included
within the study reach should be done with the objective of detecting
any unusual effects of flow interruption on downstream movement of the
river, such as the seiches observed in the ISAB 2003-1 report. We know
of no existing study that might add this to their list of tasks, so a
new project might need to be established.
We have thought of the Lower Snake River as a logical location for
this test. The four lower Snake River hydropower projects are
operated more or less as a unit because of their limited storage
capacity combined with limitations of fluctuations in reservoir
elevations specified in the BiOp. That being the case, to accomplish the
study objectives in that reach would require close cooperation of Idaho
Power Company in the operations of the Hells Canyon complex to provide
storage and release of water according to the schedule in the study
design.
Action Steps:
- Assign agency personnel currently involved in survival or behavior
studies of juvenile salmonids the task of developing a detailed
study plan to accomplish the objective of measuring the effect of
flow fluctuations associated with hydrosystem load following.
- Commence discussions with the hydrosystem operators to develop a
schedule for load following alternating with no load following to
fit the study design.
- Fund a project to monitor hydraulic conditions in the reservoirs
for the purpose of detecting any unusual patterns of flow that might
result from flow fluctuations due to load following.
- Fund, if necessary, a project specifically designed to coordinate
the collection of necessary data and to provide a summary report
focused upon the question whether flow fluctuations affect survival,
and if so to what degree.
References Cited
Gedalof, Z., D. L. Peterson, and N. J. Mantua. 2004. Columbia River
flow and drought since 1750. Journal of the American Water
Resources Association, December 2004:1-14.
ISAB 2003. Independent Scientific Advisory Board. (11 members).
Review of Flow Augmentation: Update and Clarification. Report to the
Northwest Power Planning Council, Portland, OR. ISAB 2003-1,
February 10, 2003. 69 pp. www.nwcouncil.org/library/isab/isab2003-1.htm
^ top
|
|
home