MEMORANDUM

Tetra Tech, Inc.

400 Red Brook Blvd., Suite 200

Owings Mills, MD 21117

phone	410-356-8993

fax	410-356-9005

DATE: 		August 13, 2008

TO:			Paul Shriner, EPA

	

FROM:		Ann Roseberry Lincoln, Henry Latimer and Blaine Snyder 

SUBJECT:		Impingement and Entrainment Sampling Methods

We’ve prepared this memorandum in response to your request for
information on “typical” impingement and entrainment sampling
methods. There are no published “standard methods” for impingement
and entrainment sampling; however, there are some fundamental study
elements and parameters that are universally essential for the
completion of impingement and entrainment surveys. The following method
descriptions may serve as general guidance and assume that the reader
has a general understanding of impingement/entrainment definitions and
impacts.

Impingement Abundance Studies

Study Design

Impingement studies, whether for assessment of abundance or survival,
should be designed to evaluate annual, seasonal, and diel variability
exhibited by potentially impacted organisms, and hydrographic
variability that may influence the numbers or organisms present near the
intake. Young and Dey (2003) point out that “Life histories of many of
the affected species are complex, involving only temporary occurrence
near the power plants and/or long annual migrations, making them
extremely difficult to sample for some parts of the life cycle.
Invariably, all fish in a cohort do not follow the same life history
pattern.” Unless historical data indicate otherwise, this typically
requires a year-round weekly or bi-weekly sampling program. Historical
data may show that fish and shellfish activity diminishes during part of
the year and sampling effort may be less intense, or (for example) may
show that a particular species of concern is present during a particular
time and sampling effort should increase.

During sampling events, data collection should cover the entire daily
and (where appropriate) tidal cycles. Typically, impingement samples are
collected continuously for 6 or 8 hours, and repeated to cover an entire
24-hour period. Stratifying collection in this manner allows an analysis
of the diel variation exhibited by many aquatic organisms. The specific
stratification scheme should be selected to fit the specific objectives
of the study and patterns exhibited by all of the species of concern.
For example, Spicer et al. (2000) stratified sampling by “day” (8 AM
until 8 PM) and “night” (8 PM until 8 AM) in studies conducted at
the Comanche Peak Steam Electric Station in Texas, a facility located on
a man-made reservoir with little impingement due to poor habitat quality
near the intake. In contrast, Heimbuch et al. (2007) collected samples
for four 6-hour collection periods for the Charles Poletti Power Project
to investigate potential impacts to the New York-New Jersey estuary and
Long Island Sound.

The study design must include an integrated quality assurance/quality
control (QA/QC) management plan that presents sampling design and method
specifics, and addresses data quality objectives, measurement
performance criteria, analytical methods, and field and laboratory QC
requirements.

Sampling Location

Samples should be collected from the screen wash discharge trough or
fish return system. If fish return survivability studies also will be
conducted, the sampling location should be as close to where fish enter
the receiving water body as feasible, to ensure that the survivability
accounts for all impacts that fish would incur if not collected for
study. If survivability studies will not be conducted or fish are
typically (by design) not returned to the water body, collection can
occur at any accessible point in the screen wash system. Samples should
not be collected from the water body area ahead of the intake screens
(unless the intent is to obtain data describing the numbers of dead or
moribund fish and shellfish that may subsequently be impinged).

Sampling Apparatus and Collection

Samples should be collected in nets or debris baskets fitted with
suitably small mesh size (at a minimum equal to, and preferably slightly
smaller than the intake screen mesh size). The collection device should
be of sufficient size to filter the entire wash stream and to contain
the volume of debris, detritus, and impinged organisms likely over the
total collection duration. If the collection device cannot contain the
entire debris/detritus/organism volume, the collection device must be
checked and emptied at sufficient intervals to eliminate the potential
for overflow or loss of impinged organisms from the sample. Moreover, if
survival studies will be conducted, the impinged organisms should be
transferred to a holding area regularly to avoid additional damage by
compaction within the collection device. Live fish should be handled as
little as possible to minimize stress, and should be held in conditions
close to, if not identical to, their source waters. Water should be kept
oxygenated and holding tanks should be large enough to hold the fish
with minimal stress (Kelsch and Shields, 1996).

Prior to sampling, the intake screens and screen wash discharge trough
or fish return sluiceway should be carefully washed to remove any
previously impinged and trapped organisms. The collection device should
be inspected for holes or other damage that would allow impinged
organisms to escape, and then secured in place in the discharge stream.
The time that sampling begins should be noted, along with initial
conditions. If the volume of impinged organisms (and/or detritus) is
very large, it may be necessary to empty the collection device during
the sampling event to prevent overtopping of the device and loss of
organisms. At the conclusion of the sampling event, the intake screens
should again be carefully washed to ensure that all organisms impinged
during the sampling time frame are collected. The end time and
conditions for the sampling event should be recorded. 

The sample should be carefully transferred to an appropriate holding
area for immediate taxonomic identification. Impinged organisms can then
be identified (species-level identification is preferred; or the lowest
feasible taxon for damaged organisms), measured, weighed, and their
condition (live, dead, injured, etc.) recorded before being returned to
the receiving water body. In rare cases, sample sizes may be too large
to measure and weigh each individual. Only in those cases, it may be
acceptable to subsample; subsampling should be avoided under most
circumstances. If subsampling may be required, a statistician should be
consulted to ensure that the subsampling design provides statistically
useful results.

Data collected may vary depending upon the specific facility and study
requirements, but should include some description of environmental
conditions, hydrologic conditions, intake system operational conditions,
and organism-specific information. Example primary data elements or
parameters include the following:

Environmental data: water temperature, dissolved oxygen, pH,
conductivity or salinity (if applicable), and turbidity

Hydrologic conditions: river flow or tide stage

Intake system operational conditions: intake pump flow rate, screen
spray wash pressure and schedule, traveling screen change rate

Organism data: species, lifestage, length, weight, and condition (live
or dead)

Additional data might include: air temperature, cloud cover, wind speed,
wind direction, current direction, precipitation levels or recent
precipitation events, degree of icing, and general observations such as
debris loading, debris type, and whether or not confounding conditions
occur (such as unrelated local fish kills observed in the vicinity).
Representatives of each species and lifestage should be preserved (or
photographed) as vouchers for QA evaluation purposes.

Measuring Collection Efficiency and Re-impingement Rates

Measurements of impingement abundance may be underestimated if the
study’s collection efficiency is less than 100 percent, or
overestimated if re-impingement occurs. Fortunately, both can be
measured relatively easily with some forethought.

If collection efficiency is to be measured, it is important to obtain a
suitable sample prior to the start of the study. The collection
efficiency sample should represent the variety of species and organism
lifestages anticipated in the impingement study. Ideally this control
sample will be collected about the same time as the impingement study,
in proximity to the intake structure. The sample should be collected
using collection gear types suitable for the capture of species likely
to be impinged (e.g., trawl, seine, or electrofishing unit). Organisms
should be identified, measured, counted, and marked with fin clips, dye,
or another method that allows for rapidly identifying the organism as a
collection efficiency organism. The marked individuals are then released
immediately ahead of the intake screens under operating conditions
identical to those anticipated during the impingement study. The sample
is collected using the same apparatus and methods as for the impingement
study, and the marked individuals identified to species, counted, and
measured.  The collection or recovery results are compared to the
released records to determine the collection efficiency.

Reimpingement may occur if tidal or other currents transport fish from
the discharge or fish return system back to the intake. Reimpingement
rates can be determined by marking impinged individuals, releasing them
into the discharge or fish return system, and determining the number
that are collected a second time. 

Entrainment Abundance Studies

Study Design

Like impingement studies, entrainment studies for abundance or survival
should be designed to evaluate annual, seasonal, and diel variability
exhibited by potentially entrained organisms, and hydrographic
variability that may influence the numbers or organisms present near the
intake. According to Snyder (1983), “Most larval fish investigations
require sampling from twice a week to twice a month depending on
objectives. Except for exploratory purposes, monthly or less-frequent
sampling is of very limited value—entire species or cohorts may be
missed, even when sampling is somewhat more frequent.” On the other
hand, regional fish spawning information suggests that entrainment may
be very low during cold months in northern and inland areas in the
United States (see attached figures). Therefore, unless historical data
indicate otherwise, a well-designed entrainment abundance study requires
a year-round weekly or bi-weekly sampling program, with increased
sampling frequency when egg and larval entrainment increases.

During sampling events, data collection should cover the entire daily
and (where appropriate) tidal cycles. Entrainment samples are collected
over an entire 24-hour period, although collection of each individual
sample may last as little as 5 to 10 minutes in the case of towed
plankton nets or plankton trawls, or from 1 to 3 hours for pumped
samples. For example, Heimbuch et al. (2007) collected at least 100 m3
of water using a gasoline-powered trash pump during 1 hour of pumping,
and repeated that effort for a total of four samples over 24 hours at
the Charles Poletti Power Project in New York. On the other hand, Melton
and Serviss (2000) conducted paired-net plankton tows lasting 30 minutes
each in the discharge canal of the Florida Anclote Power Plant, which
exhibits low flow conditions and required lengthy tows to ensure
adequate sample collection.

The study design must include an integrated quality assurance/quality
control (QA/QC) management plan that presents sampling design and method
specifics, and addresses data quality objectives, measurement
performance criteria, analytical methods, and field and laboratory QC
requirements.

Sampling Location

Samples should be collected from the cooling water intake system,
downstream of the intake screens, if possible. Sampling directly after
the intake screens ensures that the sample contains organisms actually
entrained and that the organisms have suffered the least possible
physical damage, thereby facilitating proper identification. However, if
entrainment survivability studies also will be conducted, the sampling
location should be as close to where the cooling water discharges as is
feasible, to ensure that the survivability accounts for all impacts that
entrained organisms would incur if not collected for study. Samples
should not be collected from the water body area ahead of the intake
screens (unless the intent is to obtain data describing the numbers of
dead fish/shellfish eggs and larvae that may subsequently be entrained).
For all entrainment sampling it is important to collect samples from all
depths (e.g., by oblique net tows or pumping from multiple depth
intervals), as ichthyoplankton frequently exhibit patchy distribution
vertically (as well as horizontally and temporally).

Sampling Apparatus and Collection

Samples may be collected directly from the intake screen discharge
stream, or water (taken downstream from the intake screens) may be
diverted and pumped to the sampling location. In either case, samples
should be collected in plankton nets with suitably small mesh size and
sufficient length to adequately filter the water and to collect the
entire sample. Care should be taken to minimize damage to the sample due
to extrusion or overloading, resulting in compaction of the collected
organisms. Kelso and Rutherford (1996) recommend a minimum net area to
mouth area ratio of at least 3:1 and preferably 5:1 to minimize net
clogging and extrusion. Typical mesh sizes range from 333 to 505 µm. If
the sample is collected directly from the discharge stream, the net
should be fitted with a flow meter in its mouth so that the volume of
water filtered can be calculated. The net may be towed within the stream
or it may be fixed to a suitable physical structure, if available. If a
pump is used instead, the piping leading from the pump to the net should
be fitted with an in-line flow meter, and the pump should be selected to
minimize damage to entrained organisms (Kelso and Rutherford 1996).
Bowles and Merriner (1978) present a thorough review of entrainment
sampling gear. 

Prior to sampling, the collection nets should be carefully washed to
remove any attached organisms, inspected for holes or other damage, and
a removable collection cup should be attached to the terminal end of the
net. Flow meters should to be set to zero or their initial readings
recorded. If a pump is used, the pump should be flushed thoroughly with
clean water prior to sampling, to ensure that no ichthyoplankton remain
within the chamber. The pump is then fitted into place with the inlet
withdrawing from the intake screen discharge stream. If the pump cannot
be flushed with clean water, then it should be run for several minutes
prior to starting collection to ensure that the sample collected
represents the current conditions. The net is then positioned to collect
the pumped water sample. To minimize any additional damage to the
entrained organisms, the net is placed vertically into a container of
water with an overflow outlet. The sample flow is directed into the
center of the net. During the sampling period, the pump inlet should be
raised/lowered periodically to obtain a depth-integrated sample. The
time and operating conditions are recorded when sample collection begins
and ends. 

If a towed plankton net is used, the clean net (with flow meter
attached) is deployed in the intake screen discharge stream and towed
against the current for a known time (e.g., from 5 to 10 minutes at 1 to
2 meters per second). Towing speed should be fast enough to minimize
larval avoidance, but slow enough to ensure adequate filtering. Multiple
depths should be sampled, either by deploying multiple nets at specified
depths or by employing stepped oblique tows.

At the completion of the pump or net tow sampling event, the time and
flow meter readings are recorded (to allow calculation of sampling
duration and sample volume) and the collection net washed so that the
entire sample flows into the removable collection cup at the end of the
net. The collection cup is removed and all contents are placed into a
sample jar (or the sample transferred). Samples are fixed or preserved
by adding 5-10% formalin.

Preserved entrainment samples are sent for laboratory identification by
an ichthyoplankton taxonomy specialist. Entrained organisms are picked
from detritus in the laboratory using side lighting and a high-contrast
background for better visibility, and are stored in 5% buffered
formalin. Ichthyoplankton are identified with the aid of a dissecting or
low-power microscope. All fish and shellfish eggs and larvae are
identified to the lowest feasible taxon (species preferred). 

Data collected may vary depending upon the specific facility and study
requirements, but should include some description of environmental
conditions, hydrologic conditions, intake system operational conditions,
and organism-specific information. Example primary data elements or
parameters include the following:

Environmental data: water temperature, dissolved oxygen, pH,
conductivity, and turbidity

Hydrologic conditions: river flow or tide stage

Intake system operational conditions: intake pump flow rate

Organism data: species, life stage, length

Additional data might include: air temperature, cloud cover, wind speed,
wind direction, current direction, precipitation levels or recent
precipitation events, degree of icing, and general observations
including any confounding conditions. 

Measuring Collection Efficiency and Re-entrainment

Entrainment rates can be underestimated if collection efficiency is less
than 100%, and overestimated if re-entrainment occurs. It is possible to
measure both collection efficiency and re-entrainment if suitable fish
eggs and larvae are available for the study. The ichthyoplankton sample
used to measure these potential sources of error can be collected from
source waters and should be stained with a suitable biological stain
(e.g., rose Bengal) prior to the study.

If collection efficiency is to be measured, a stained and counted
ichthyoplankton sample is released immediately in advance of the intake
screens, such that 100% entrainment will occur. Entrainment sample
collection proceeds as described above and the number of collected
organisms compared to the number released to determine collection
efficiency. To measure re-entrainment, stained and counted
ichthyoplankton are released into the discharge stream near its terminus
so that the sample discharges to the receiving waterbody. Entrainment
sample collection proceeds as described above and the number collected
subsequently is the number re-entrained.

References

Bowles, R.R. and J.V. Merriner. 1978. Evaluation of Ichthyoplankton
Sampling Gear Used in Power Plant Entrainment Studies. Pages 33–43 in
L.D. Jensen, editor. Fourth National Workshop on Entrainment and
Impingement, 5 December 1977, Chicago, Illinois. Sponsored by Ecological
Analysts, Inc., Melville, NY. Published by EA Communications, Melville,
NY.

Heimbuch, D.G., D.J. Dunning, Q.E. Ross, and A.F. Blumberg. 2007.
Assessing Potential Effects of the New York–New Jersey Harbor Estuary
and Long Island Sound. Transactions of the American Fisheries Society
136: 492–508.

Kelsch, S.W. and B. Shields. 1996. Care and Handling of Sampled
Organisms. Pages 121–144 in B.R. Murphy and D.W. Willis, editors.
Fisheries Techniques, 2nd Edition.  American Fisheries Society,
Bethesda, MD.

Kelso, W.E. and D.A. Rutherford. 1996. Collection, Preservation, and
Identification of Fish Eggs and Larvae. Pages 255–302 in B.R. Murphy
and D.W. Willis, editors. Fisheries Techniques, 2nd Edition.  American
Fisheries Society, Bethesda, MD.

Melton, B.R. and G.M. Serviss. 2000. Florida Power Corporation-Anclote
Power Plant Entrainment Survival of Zooplankton. Environmental Science
and Policy 3: S233–S248.

Snyder, D.E. 1983. Fish Eggs and Larvae. Pages 165–197 in L.A. Nielsen
and D.L. Johnson, editors. Fisheries Techniques. American Fisheries
Society, Bethesda, MD.

Spicer G., T. O’Shea, and G. Piehler. 2000. Entrainment, impingement
and BTA evaluation for an intake located on a cooling water reservoir in
the southwest. Environmental Science and Policy 3: S323–S331.

Young, J.R. and W.P. Dey. 2003. Uncertainty and Conservatism in
Assessing Environmental Impact under §316(b): Lessons from the Hudson
River Case. Pages 30–39  in D.A. Dixon, J.A. Veil, and J. Wisniewski,
editors. Defining and Assessing Adverse Environmental Impact from Power
Plant Impingement and Entrainment of Aquatic Organisms. Swets &
Zeitlinger B.V., Lisse, The Netherlands.

 PAGE   

 PAGE   7 

