Field Manual
Okanogan Monitoring and Evaluation Program
Biological Protocols
DRAFT
September 1, 2005
Prepared by
The Colville Confederated Tribes
John Arterburn
Keith Kistler
KWA Ecological Sciences, Inc.
Paul Wagner
Rhonda Dasher
Table of Contents
Introduction ...................................................................................................................................... 3
Acknowledgements ......................................................................................................................... 4
Section 1. Method For Detecting Fish Species Assemblages Using Snorkeling ............................ 4
Purpose ........................................................................................................................................ 4
Equipment .................................................................................................................................... 4
Site Selection ............................................................................................................................... 4
Sampling Duration ....................................................................................................................... 4
Permits ......................................................................................................................................... 4
Snorkeling Sampling Procedures ................................................................................................. 5
Section 2. Smolt Trapping ............................................................................................................... 7
Summary ...................................................................................................................................... 7
Purpose ........................................................................................................................................ 7
Background ................................................................................................................................ 10
Safety ......................................................................................................................................... 11
Equipment .................................................................................................................................. 11
Site Selection ............................................................................................................................. 11
Preparation and Installation ....................................................................................................... 13
Sampling Duration ..................................................................................................................... 14
Procedure................................................................................................................................... 14
Data Analysis ............................................................................................................................. 17
Estimating Total Migration ......................................................................................................... 17
Literature Cited........................................................................................................................... 22
Section 3. Adult Enumeration ........................................................................................................ 25
Purpose ...................................................................................................................................... 25
Procedure................................................................................................................................... 26
Section 4. Benthic Macro-Invertebrate Assessment ..................................................................... 26
Purpose ...................................................................................................................................... 26
Equipment .................................................................................................................................. 26
Site Selection ............................................................................................................................. 26
Sampling Duration ..................................................................................................................... 26
Procedure To Collect Kick Net Samples .................................................................................... 26
Procedure For Preparing Composite Samples for Identification ............................................... 28
Multi-Metric Index Development ................................................................................................ 29
Definition of Terms......................................................................................................................... 32
References Cited ........................................................................................................................... 32
Section 5. Method For Detecting Steelhead Redds ...................................................................... 32
Purpose ...................................................................................................................................... 32
Site Selection ............................................................................................................................. 33
Sampling Duration ..................................................................................................................... 34
Equipment .................................................................................................................................. 35
Anatomy of a Redd .................................................................................................................... 35
Multiple Pass Marked Redd Survey Method .............................................................................. 36
Peak Redd Count Method .......................................................................................................... 37
Foot Surveys .............................................................................................................................. 37
Boat /Raft Surveys ..................................................................................................................... 37
Fish Observation ........................................................................................................................ 37
Carcass Surveys ........................................................................................................................ 37
QA/QC Procedures .................................................................................................................... 38
Removal Of Flagging ................................................................................................................. 38
Estimating Total Redds and Escapment .................................................................................... 38
Literature CIted .......................................................................................................................... 40
INTRODUCTION
This Field Manual was developed by the Colville Confederated Tribes to provide specific
guidance in the evaluation and monitoring of fish populations in the Okanogan Subbasin
for the 2005 Okanogan Baseline Monitoring and Evaluation Program (OBMEP). The
OBMEP is long-term status and trend monitoring program subject to future adaptation
management. Therefore, this Field Manual should be considered to be a "living
document" with the following protocols potentially subject to some level of modification
over time as new information becomes available.
The protocols contained within this Manual are closely aligned with the Environmental
Monitoring and Assessment Program (EMAP) developed by the Environmental
Protection Agency (EPA) as adopted into the Upper Columbia Monitoring and
Evaluation Strategy. These protocols were further refined to address specific program
needs and for compatibility with the Ecosystems Diagnosis and Treatment (EDT) Model
developed by Mobrand Biometrics Incorporated. EDT is the primary assessment tool
used by subbasin planners throughout the Columbia Basin and specifically within the
Okanogan Subbasin and periodic updating of EDT input fields with compatible data will
be necessary to assess changes that may occur within the subbasin over time.
ACKNOWLEDGEMENTS
Ken MacDonald (USFS) provided valuable information pertaining to the snorkeling
protocols in use in the Wenatchee Subbasin Monitoring and Evaluation Program.
Andrew Murdock (WDFW) and Greg Volkhardt (WDFW) provided essential material
and advice used in the development of the smolt trapping protocols.
SECTION 1. METHOD FOR DETECTING FISH
SPECIES ASSEMBLAGES USING SNORKELING
Modified Protocol taken from: Rodgers (2002), Thurow (1994), and
Peck et al. (Unpubl.)
PURPOSE
Estimating the density of juvenile salmonids with a primary emphasis on summer
steelhead allows the investigator to obtain a sample over time of the change in abundance
of rearing juvenile salmonids produced in the Okanogan River basin. Collection of
information pertaining to salmonids and other species of fish may be collected but is
ancillary to the goal of estimating juvenile summer steelhead production in the Okanogan
subbasin. In addition, information pertaining to the presence of salmonids and nonsalmonids can be used to provide input for the Ecosystem Diagnosis and Treatment
Model in the Predation Risk, Fish Community Richness, and Fish Species Introductions
attribute fields.
EQUIPMENT
Persons conducting snorkel counts should be equipped with dry suits or wet suits, masks,
snorkels, rubber soled boots, hand held thermometer, field forms/data logger, and a
stopwatch. Additional equipment such as hand counters, underwater white boards, and
measuring rods are helpful for enumerating fish and determining fish lengths. A
submersible halogen light may be useful to search for fish in shaded locations.
SITE SELECTION
The sample reach is laid out according to Section 1 of the Physical Habitat Field Manual.
SAMPLING DURATION
Sampling for fish species should occur during the low flow period in late summer when
water temperatures exceed 90C.
PERMITS
Be sure that all necessary collection permits and ESA clearances are obtained and copies
of required permits are with you while in the field.
SNORKELING SAMPLING PROCEDURES
After team members don their wetsuits, a quick equipment check will ensure
that all members have the necessary tools to complete their survey. Team members need
to designate one person as the data recorder for each group in the survey. All members in
the group report fish observations to the recorder after each transect.
Step 1:
Measure the visual distance to determine the number of snorkelers needed to
survey a given reach on a given day. A snorkeler should take one end of a measuring tape
and back away from the bank while another person holds the tape. Once the snorkeler can
no longer clearly see detail along the bank. The person on shore will take a measurement
of this distance. On small streams this distance will be used to determine the number of
snorkelers needed to survey a reach. On larger streams with maximum depths greater
than 3 meters the visual distance will be measured by visual depth. A snorkeler will enter
the deepest area of the stream and drop a small unpainted lead weight attached to a string
toward the bottom until it can no longer be seen then slowly begin raising the weight,
marking the string at the depth where it first becomes visible again. The string can be
measured on shore to determine the visual distance for the reach. Once the visual distance
is determined, the number of persons needed is calculated by dividing the average width
of the reach by two times the visual depth. Snorkel surveys of large stream reaches
should be conducted during periods of maximum visibility to minimize the number of
staff required.
Step 2:
Before starting a survey, record the weather conditions and water temperature
(clear, overcast, rainy, etc). In all wadeable and most non-wadeable stream reaches,
snorkeling should involve only a single pass through areas that are deep enough to
survey. Record weather conditions, temperature, and total time required at the beginning
and end of the entire reach.
Step 3:
Note: Team members need to look downstream periodically and read the water for areas
of increased flow or change of gradient (riffles, falls, or rapids). Before making a pass
through an area of increased flow, consider possible safety issues. Objects such as
boulders and woody debris can pose a significant hazard, if team members are not
prepared. Studying the reach, before a survey, will alert team members to possible
hazards so that they can develop a plan for working through difficult areas.
Step 4: .
Wadeable Streams: Begin at the downstream boundary (Transect A) of the sample
stream reach and proceed upstream through the pools and riffles. All movements in the
water need to be slow and careful so as not to create a flight response in fish observed
(Thompson 2000) or reduce the visual distance by disturbing fine sediments. Members
will slowly proceed upstream with team members keeping each other in their line of
sight. Where possible, members should float, to avoid stirring up sediment with their feet
or hands. If there are multiple groups within a reach, each group should minimize
disturbance so that visibility downstream is not compromised.
Non-wadeable Streams and Rivers: Begin at the upstream boundary (Transect K).
Depending on flow conditions group members can stay in line by keeping each other
within the visual distance or by using a section of rope with floats on it that everyone
holds onto. Members will proceed downstream with the current, controlling their
movement enough to keep within the visual distance to the other group members. If the
number of snorkels available is less than those needed as outlined in Step 2 but visibility
is considered good for the site, a second pass maybe used. If conducting the 2-pass
strategy, have snorkelers spread from the right bank to the mid-point of the stream and
conduct the survey through all transects. Proceed to the top of the stream reach and
spread out from the left bank to the mid-point in the stream and conduct the second
survey. Sum all counts for each transect for a complete survey.
A minimum of two persons should be present during snorkel surveys. A two
person snorkeling crews can conduct snorkel surveys in many wadeable streams. In
wadeable stream reaches, one crew member should snorkel each transect while the other
crew member records the counts as they are given by the snorkeler. In non-wadeable
areas, crew members should snorkel side by side and sum their individual counts. Using
an arm chart, white board, or hand counter. Each snorkeler counts the fish to the
immediate front and to the sides opposite the other snorkeler or as designated by the team
leader to avoid duplication of counts. Each individual snorkeler should maintain a
spacing of 2-times the visual distance from the nearest snorkeler and observe fish both to
the right and left. The designated team leader will control spread as the stream channel
changes width according to pre-snorkeling instruction.
Step 5:
Counts of the number of fish should be recorded along entire transect area (AB, B-C, etc.). When enumerating and identifying fish three levels of identification are
possible. First, All fish encountered during the snorkel survey should be identified to
species this is especially critical for salmonid species. Second, non –salmonid species can
be enumerated into major taxa if species identification is not possible as may be the case
of some non-salmonids (i.e..sculpin, bass, sunfish, suckers, trout, salmon, char,
whitefish). Lastly, every effort should be made to at least classify fish as salmonids or
non-salmonids. The unidentified column on the data sheet should rarely be used.
Step 6:
Step 7: In addition to enumerating all fish, salmonids should be lumped into the following
length categories: 1) young of the year (<100 mm), 2) juveniles (100-300mm), and 3)
adults (>300mm).
After snorkeling, the underwater visibility of each study reach is ranked on a
scale of 0 to 3 where 0 = not snorkelable due to an extremely high amount of hiding
cover or zero water visibility (No snorkel survey should be conducted, <25% visibility);
1 = high amount of hiding cover or poor water clarity (25%-50% visibility); 2 = moderate
amount of hiding cover or moderate water clarity, neither of which were thought to
impede accurate fish counts (50%-75% visibility); and 3 = little hiding cover and good
water clarity (>75% visibility).
Step 8:
To calculate fish densities (fish/m2), determine the area for each reach by
utilizing the physical habitat data collected (previously described in Section 2 of the
Physical Habitat Protocols Field Manual) for average width and multiplying this by reach
length to get the area sampled. Then divide the total number of fish along with the total
Step 9:
number of all identified categories by the area sampled. Total salmonids, and total O.
mykiss should also be calculated.
Step10:
Consult Thurow (1994) for additional information.
SECTION 2. SMOLT TRAPPING
Protocol adapted from; Seiler and Volkhardt 2005, and Murdock et al. 2001.
SUMMARY
Investigators will use floating screw traps (or other appropriate traps depending on stream
conditions) to collect downstream migrating smolts to estimate the total number
(abundance) of smolts produced within a watershed or basin. Traps will operate for at
least the entire period of the smolt migration. Trapping efficiency, based on
mark/recapture will be estimated throughout the trapping period. Methods for operating
the trap, estimating efficiency, and the frequency at which efficiency tests are conducted
are described in Murdoch et al. (2000). Numbers of smolts will be reported for
populations or subpopulations. The Fulton-type condition factor will be estimated from
length and weight measurements to describe the well-being of smolts within a population
or subpopulation. Genetic samples will also be collected to characterize (via DNA
microsatellites) within- and between-population genetic variability of smolts.
PURPOSE
Operating a downstream migrant trap allows the investigator to sample the wild
salmonids produced in a watershed or tributary over time. The sample in itself is
valuable because it documents the presence/absence of migrating juveniles, enables
determination of age at migration, condition, timing, species, and genetic characteristics.
Furthermore, if the location of the trap, its placement, and hours of operation are
sufficient and held reasonably constant from year to year, catch of a given species or
catch per unit effort can be used as an index of downstream migrant production (Seiler
and Volkhardt, 2005).
More importantly, trapping information can also be used to create estimates of the total
freshwater production by using a simple mark-recapture population estimation
methodology. The rationale is simply that the proportion of marked fish appearing in a
random sample provides an estimate of the proportion marked in the total population.
The proportion captured (trap efficiency) is estimated by conducting a series of trap
efficiency experiments over the trapping season (Seiler and Volkhardt, 2005).
This protocol describes methods used to achieve estimates of wild downstream migrant
salmonid production using a rotary screw trap. Since the traps strain the upper portion of
the water column, they are generally not very useful for capturing species that migrate
along the bottom of the river (e.g., lamprey). The traps can be scaled to operate in
various sized streams, but are most commonly used in streams that are too large or
powerful to employ a fence weir (e.g., ~10 to 15-m or larger channels) (Seiler and
Volkhardt, 2005).
The rotary screw trap is used in medium to large rivers. The screw trap consists of a cone
covered in perforated plate that is mounted on a pontoon barge (Figure 1). Within the
cone are two tapered flights that are wrapped 360-degrees around a center shaft. The trap
cone is oriented with the wide end facing upstream and uses the force of the river acting
on the tapered flights to rotate the cone about its axis. Downstream migrating fish are
swept into the wide end of the cone (typically either 1.5-m or 2.5-m in diameter) and are
gently augered into a live box at the rear of the trap. Typically one or more winches are
used to adjust the fore and aft elevation of the trap. A small drum screen, powered by the
rotating cone, is located at the rear of the live box and removes organic debris (Seiler and
Volkhardt, 2005).
Figure 1. Rotary Screw Trap.
BACKGROUND
Floating inclined plane screen traps, commonly referred to as scoop traps and rotary
screw traps, have long been used by biologists to capture downstream migrating juvenile
anadromous salmonids from medium and large-sized streams (Schoeneman et al. 1961,
Seiler et al. 1981, Kennen et al. 1994). The rotary screw trap was developed by the
Center for Innovative Technology Transfer at State University of New York (SUNY) in
the late 1980’s. Rotary screw traps are anchored at a fixed point in the river channel and
intercept a portion of the downstream migrant salmonids or smolts emigrating to the sea.
Traditionally, fishery managers have relied on escapement estimates to monitor
anadromous salmonid population status and management effectiveness (Ames and
Phinney 1977; Beidler and Nickelson 1980; Hilborn et al. 1999). However, estimation of
population abundances at earlier life stages enables partitioning survival among lifestages and developing hypotheses for restoration actions (Moussalli and Hilborn 1986,
Mobrand et al. 1997). Juvenile fish traps have often been used to estimate the abundance
(Tsumura and Hume 1986, Baranski 1989, Orciari et al. 1994, Thedinga et al. 1994,
Letcher et al. 2002), timing (Wagner et al. 1963, Hartman et al. 1982), size (Orciari et al.
1994, Olson et al. 2001), survival (Schoeneman et al. 1961, Wagner et al. 1963, Tsumura
and Hume 1986, Olsson et al. 2001, Letcher et al. 2002), and behavior (Brown and
Hartman 1988, Roper and Scarnecchia 1996) of downstream migrant anadromous
salmonids. In many salmon-bearing systems, population abundance is only monitored
during the adult (spawner) stage. Additional monitoring of smolt abundance is
particularly powerful since it enables the partitioning of mortality between the freshwater
life stages (egg-to-smolt) and marine life stages (smolt-to-adult) (Seiler and Volkhardt,
2005).
While estimating smolt abundance is the most common reason for operating a scoop or
screw trap, the collection of downstream migrants also has wide utility. Traps can be
used to monitor the effects of river management on wild stocks, such as the effectiveness
of diversion, lock, and dam management. They can be used to validate assumptions
regarding the effect of watershed restoration programs and land-use policies on fish
populations. They can also be used to assess survival between life stages, such as egg-tosmolt survival or parr-to-smolt over-winter survival (Seiler and Volkhardt, 2005).
In addition to monitoring wild populations, traps are useful for evaluating hatchery
programs and hatchery/wild fish interactions. Such studies may include evaluating the
instream survival of hatchery production following release and evaluating treatments
such as rearing strategy, release timing, release location, and flow manipulation on
groups of hatchery fish. These later uses can be applied to evaluate a variety of projects
or actions ranging from hatchery supplementation strategy to avoidance of hatchery and
wild fish interactions. In addition to abundance estimates, investigators use scoop and
screw traps to collect samples of downstream migrants for such purposes as genetics
sampling, fish disease research, predation (gut content) evaluations, and wild stock
marking and tagging projects (Seiler and Volkhardt, 2005).
On the west coast of the United States and Canada, juvenile fish traps have primarily
been used to estimate the natural production of juvenile coho (Oncorhynchus kisutch),
sockeye (O. nerka), and steelhead (O. mykiss) from 5th order and smaller basins
(Nickelson 1998). Nevertheless, with careful planning reasonably accurate production
estimates have been obtained when 6th order and larger systems have been trapped
(Schoeneman et al. 1961, Thedinga et al. 1994). For example, side-by-side scoop and
screw traps have been used to successfully yield estimates of yearling coho and subyearling chinook migrants since 1990 in the Skagit River, a 7th order basin (Seiler et al.
2003).
SAFETY
When positioned in the river, screw traps represent a hazard to boaters, float tubers, and
swimmers. Signage should be positioned upstream to instruct river users how to safely
avoid the trap. Other protective measures may include flashing lights to improve the
visibility of the trap and deflectors to help prevent water users and large woody debris
from entering the trap (Seiler and Volkhardt, 2005).
A minimum of two persons shall operate the trap at any given time. Life jackets will be
worn at all times by personnel while on the water traveling to and from the trap or while
operating the trap. Standard precautions should be taken by personnel to keep hands and
loose clothing away from the cone and axel and other moving rotary screw trap parts
during trap operation.
EQUIPMENT
Trap/pontoon structure, anchor cables, boat (if necessary to reach trap), dip nets, fish
anesthetic (MS 222), marking devices (scissors, dye, etc.), buckets (for working fish),
broom/water pump for cleaning trap, flood lights (for night work)
SITE SELECTION
Selection of trapping sites should be viewed from a variety of scales. At the watershed
scale if the natural production of salmon is to be monitored, the river or stream should be
either devoid of hatchery fish or all hatchery fish should be identifiable so that wild fish
can be enumerated. Precision of the estimates increases with higher trap efficiency (i.e.
proportion of migrants captured), therefore it is generally better to select sites where a
higher proportion of the total flow can be screened through the trap. This becomes a
trade-off, however, if the trap is placed below a hatchery release site since higher trap
efficiencies can result in very large numbers of hatchery fish entering the trap following a
fish release. Where this occurs, good communication between the trap operators and
hatchery staff must be maintained to avoid a fish kill. In general, it is best to avoid these
situations when choosing a trap site. Another consideration when selecting watersheds is
the hydrologic pattern and basin landform. Rivers exhibiting a flashy hydrograph are
very difficult to trap due to high fluctuations in flow conditions and debris loads. Since
trap efficiency and migration rates often changes dramatically with flow, its much more
difficult to estimate migration where wide swings in flow occurs. Flow is dependent on
such variables as landform, geology, land cover, climate, and precipitation patterns,
which of course, cannot be controlled. The effect of these factors on the stream discharge
needs to be considered when attempting to estimate total freshwater production (Seiler
and Volkhardt, 2005).
Within a watershed, the trap should be placed as low in the watershed as practicable.
Species exhibiting a stream-type life history pattern, such as coho salmon and steelhead
often migrate within basin and rear away from their natal streams. Therefore, the smolt
production measured from part of the basin may represent a variable proportion of the
progeny from the adults that spawned upstream of the trap. Furthermore, species with an
ocean-type life history pattern often spawn lower in the watershed (e.g., pink salmon).
Estimating production for these species requires trap placement as low in the system as
possible (Seiler and Volkhardt, 2005).
At the site scale, water velocity, depth, and proportion of the flow screened are also
important considerations for trap placement. Velocity is an especially important
consideration if trapping strong swimming species such as steelhead trout, and becomes
less important when trapping newly emerged fry. For most species, water velocities of at
least 1-mps are desirable for scoop trap operation and over 2-mps may be required to
capture and retain most steelhead smolts. Similar velocities are recommended for screw
trap operation. Screw traps should rotate at least 5-6 rotations per minute for retention of
larger smolts. Care must be taken that the water depth under the trap and live well will
be sufficient over all flow conditions expected during the outmigration period. It is
usually best to select a site where a relatively high proportion of the total flow can be
screened through the trap in order to achieve the highest trap efficiency. The requirement
for adequate velocity, depth, and trap efficiency usually argues for placing the trap in the
thalweg of the channel. Consideration must be given, however, to the number of
migrants captured. The investigator may opt to operate the trap in a slightly less
advantageous position to avoid causing stress or predation in the live well by capturing
and holding too many migrants (Seiler and Volkhardt, 2005).
Screw traps are inherently noisy simply due to the rotation of the trap about its central
axis. Migrants will avoid the trap if they are aware of its presence; therefore, it is best to
select a site where the trap noise can be masked in order to maintain higher trap
efficiency. Fortunately, higher velocity reaches are also noisy reaches. In smaller rivers,
these conditions are encountered at the head end of a pool or chute where water velocities
over an elevation drop (e.g., riffles, cascades, or falls) can be directed into the trap. In
larger rivers, channel constrictions may afford the best sites (Seiler and Volkhardt, 2005).
In addition to the above-mentioned criteria, consideration must be given to anchoring the
trap in the stream. Scoop and screw traps can be anchored by cables to the base of stout
trees on each bank, to anchors affixed to bridge abutments, retaining walls, or bedrock, or
to a high lead suspended across the river. In the early 1960’s, the mainstem Columbia
River was trapped using a series of scoop traps cabled to large concrete blocks
submerged in the river (Schoeneman et al. 1961 as cited in Seiler and Volkhardt, 2005).
Finally, investigators need to consider access and security when selecting trapping
locations. Traps anchored in the river are a curiosity, which can undergo theft or
vandalism when not attended. Ideally, the trap site would be located near a
launch/recovery site to ease trap installation and removal (Seiler and Volkhardt, 2005).
PREPARATION AND INSTALLATION
Before trapping can begin, all equipment and supplies must be assembled to accomplish
projects objectives. At a minimum, these include the trap/pontoon structure and anchor
cables, a means to get to the trap (e.g., boat or gangplank), dip nets for removing and
handling fish, data forms, fish anesthetic, a marking device (e.g. scissors, dye, etc.), tanks
or buckets for working up captured fish, a trap cleaning device (e.g. brooms, water pump
and nozzle), and lights for night work. Permits may also need to be secured for
placement of the trap and/or handling fish from various jurisdictions. Sufficient time
must be allotted during the planning period to secure permits (Seiler and Volkhardt,
2005).
The approach for trap installation depends on the size and weight of the trap used. Small
traps that use lightweight aluminum pontoons can be transported disassembled in pickup
beds and assembled on-site. Components of larger, heavier traps can be trucked to the
site using a low-boy trailer. In this case, on-site assembly requires the use of a loader or
other heavy equipment to move the components into place. A third option is to truck an
assembled trap to the site and position it at the water’s edge using a boom truck or crane
(Seiler and Volkhardt, 2005).
Once assembled at the water’s edge, the trap is ready to be positioned in its fishing
location. The approach used to accomplish this will depend on the size of the trap and
stream, and the distance from the launch site to the fishing site. Small traps operating on
small streams can be pushed and pulled into position by hand. Bow-mounted cables or
ropes can simply be attached to trees or other anchoring structures on the banks.
Movement of the trap into its final position can be accomplished using hand winches or
chainfalls. If the trap is anchored to trees, some method should be used to spread the load
over the trunk and prevent girdling. Fabric straps make useful attachments (Seiler and
Volkhardt, 2005).
Larger traps may use bow winches, mounted port and starboard, to store the attachment
cable or rope. The most direct approach is to run the cabling out to the attachment points
and pull the trap into position using the winches. Another approach is to attach the
cabling directly from the trap to a highline that has been strung over the river. For larger
traps (e.g., 2.4 meter cone diameter rotary screw trap), the trap should be secured in the
river with 10 mm aircraft cable attached to a 13 mm aircraft cable and pulley system
strung above the river between two large trees or bridge pilings on either bank (Murdoch
et al. 2001). The position of the trap can be adjusted by the tension of the highline and
length of the bow cables that are attached to it using a chainfall or similar device. The
use of bow-mounted winches is the preferred approach since it makes repositioning the
trap much easier (Seiler and Volkhardt, 2005).
In some cases, the launch point may be some distance from the fishing site. In this
situation, the trap can be “walked” into position by alternating port and starboard
attachment points either upstream or downstream and tightening or loosening the bow
cables as necessary using winches. In navigable waters, a boat can be used to push the
trap to a point near the trap site where one of the above methods can be used to secure the
trap to its fishing position (Seiler and Volkhardt, 2005).
SAMPLING DURATION
The time frame for operation of the trap varies with the target species and trapping
location. Table 1 provides general migration timing for Washington rivers.
Downstream migration timing in specific watersheds can vary from these general
guidelines. Timing may need to be investigated during the first year of monitoring where
it is not well known (Seiler and Volkhardt, 2005).
Table 1. Generalized migration timing for anadromous salmonids in Washington state.
Species
Age
Chinook
0, 1
Coho
1
Sockeye
0
Chum
0
Pink
0
Steelhead
2
Cutthroat
0, 1, 2
* Migration timing for cutthroat vary widely.
Migration Period
January – July/August
April – June
January – May
February – April
January - May
March – May
January – December*
In order to estimate production, traps should be operated throughout the migration period
for the target species. Migration rates for most species are often highest at night,
however, daytime migration rates can also be high on some streams, particularly where
turbidity levels are high. At a minimum, the investigator should stratify trapping periods
to reflect different migration/capture rates. This often means checking the trap and
processing the catch at dawn and at dusk to measure day and night catch rates. This
doesn’t infer that these are the only times to check the trap. Catch rates and debris loads
determine the frequency of trap maintenance. Stratification facilitates subsampling and
estimating catches during periods when trapping is suspended (Seiler and Volkhardt,
2005).
PROCEDURE
Trap Operation
The screw trap is lowered into its fishing position by cables attached to the forward
and/or aft ends of the trap structure. Typically, a single hand winch or chainfall is used to
raise and lower each end. The forward end of the screw should be lowered until the axle
is at the water’s surface. The aft end is lowered so that fish can swim from the aft screw
chamber into the live well, but not so low that they can ride the debris drum over the back
of the trap (Seiler and Volkhardt, 2005).
Since the screw is constantly rotating, relatively little debris builds up on the screw’s
outer screen. As the debris drum removes much of the debris entering the trap, this gear
requires less cleaning than a scoop trap. During each trap check, debris remaining in the
live well is removed and captured fish are dip-netted out. The trap can usually remain in
operation during this procedure. The date and time of the trap check is recorded. If the
trap is outfitted with a counter to record rotations, the count is recorded. Rotations per
minute are also recorded during the trap check. These later data are used to estimate the
time fished if debris stops the screw between trap checks. Catch is enumerated by
species and other data/samples are taken as required by the study (Seiler and Volkhardt,
2005).
Daily Capturing Procedure
Traps are checked as often as necessary to provide for the safe holding and handling of
captured fish, and maintain the efficient operation of the gear. At a minimum, the trap
should be checked at dawn and at dusk in order to evaluate day vs. night capture rates.
When operated during period of high discharge, the trap will be checked and cleaned
three times a day. Where subyearlings are captured, holding these in close proximity to
larger piscivorous fish such as cutthroat smolts and sculpins increases the likelihood that
catch counts on the subyearlings will be biased low due to live-box predation (Seiler and
Volkhardt, 2005).
Some investigators have placed tree branches or other debris in the live well to provide
refuge for small fish. Care must be taken when using this approach since the debris may
cause de-scaling as turbulence in the live well increases. The safest approach for
maintaining fish health and minimizing predation is to frequently check and remove fish
from the trap (Seiler and Volkhardt, 2005).
Fish will be removed from the livebox every morning and placed in an anesthetic solution
of MS-222. Fish will be identified to species and counted. Incidental species and small
chinook fry will be allowed to fully recover in fresh water prior to being released in an
area of calm water downstream from the smolt trap. Juvenile target salmonid species will
be held in separate live boxes attached to the end of the main pontoons for use during
mark/recapture efficiency trials conducted in the evening.
Length and Weight
Length and weight measurements will be recorded for all target species, except on days
when high numbers are captured, then only target species used in mark/recapture
efficiency trials were measured and weighed. Fork length to the nearest millimeter and
weight to the nearest 0.1 g will be measured. A Fulton type condition factor (W×105/FL3)
will be calculated for all target species sampled. The degree of smoltification (parr,
transitional, or smolt) will be determined by visual examination. Juvenile Chinook,
sockeye, and steelhead O. mykiss will be classified as parr if parr marks are distinct,
transitional if parr marks are not distinct, and smolts if parr marks are not visible and the
fish exhibited a silvery appearance.
Condition
The Fulton-type condition factor describes the well-being of smolts within a population
or subpopulation. Smolts collected with traps will be measured (fork length; mm) and
weighed (g). Fulton-type condition will be estimated with methods described in
Anderson and Neumann (1996).
Genetics
Genetic characterization (via DNA microsatellites) describes within- and betweenpopulation genetic variability of smolts. DNA samples from a systematic sample of
smolts1 will be collected and analyzed according to the WDFW protocols contained in
Appendix 1.
Trap Efficiency Tests
Trap efficiency is measured by the rate that marked fish released above the trap are
recaptured. Mark/recapture efficiency trials will be conducted throughout the trapping
season when a minimum of 30 individual fish of a given target species are captured
within a three day period. If less than 30 fish are captured within a three day period, all
fish will be released unmarked. A variety of techniques can be used to mark fish for trap
efficiency testing. Probably the simplest approach is to anesthetize the fish and apply a
partial fin clip (e.g., upper or lower caudal lobe, posterior/anterior anal lobe, various
caudal punches, etc.). Other approaches include dying, freeze branding and PIT tagging.
Fish will be placed in a live pen to recover for at least 8 h before being transported in 5
gallon buckets to the release site. Fish should be fully recovered from the anesthetic prior
to release (Seiler and Volkhardt, 2005).
The release point selected should be far enough upstream as to provide for a similar
distribution across the channel compared to unmarked fish (at least 2 pool/riffles
sequences), but not so far upstream that predation on marked fish is substantial. Murdoch
et al. recommends that the release point be located at least 1 km upstream of the trap
located at least 1 km upstream of the trap. Each group of marked fish should be released
evenly across the river to avoid biasing their lateral distribution and along approximately
100 m of the bank in pools or in calm pockets of water where possible. To reduce
predation subsequent to recapture, marked fish should be released during the time strata
that they migrate (Seiler and Volkhardt, 2005).
Mark groups can be comprised of hatchery fish or fish that have been previously captured
in the trap. However, using hatchery fish complicates the study since one must assume
their probability of capture is the same as for naturally reared fish. Groups of marked
fish representing each targeted species are released upstream of the trap over the period
of their migration.
While hatchery fish used for calibration may be of the same species and age as their wild
counterparts, they may be larger, behave differently, and consequently, may be captured
at higher or lower rates than wild fish. Rates of instream predation and residualism are
likely higher for hatchery fish. For these reasons, trap efficiency estimates resulting from
release groups using hatchery fish may be biased low (Seiler and Volkhardt, 2005).
Flow is the dominant factor affecting downstream migrant trapping operations in any
system. It affects trapping efficiency and migration rates since high flows often stimulate
1
The total number of smolts needed to characterize within and between-population genetic variability is
presently unknown. Therefore, “k” (i.e., the kth smolt sampled) remains undefined.
fish to migrate. Therefore, minimal trap efficiencies may occur at the same time that
peak flow events are causing migration rates to increase (Seiler and Volkhardt, 2005).
Visibility, fish size, and noise are other factors that affect trap efficiency. Larger
downstream migrants, especially steelhead and coho, may be able to avoid capture when
the trap is visible by swimming around the trap or back out of the mouth of the trap,
especially where velocities are low. Some portion of ocean-type chinook salmon may
rear upstream for a short period of time and grow prior to migration; therefore, efficiency
for a species may change over time. Fish behavior may also be important. Some species
may primarily migrate down the thalweg of the channel whereas a higher proportion of
others may use the channel margins. Noise created by the trap causes an avoidance
response. This is mitigated through proper site selection as discussed above (Seiler and
Volkhardt, 2005).
These factors indicate that efficiency tests should, if possible, be conducted over the
entire migration period, over a range of flows and turbidity levels, and for each species
whose production is to be estimated (Seiler and Volkhardt, 2005).
Emigration estimates can be calculated using estimated daily trap efficiency derived from
the regression formula using trap efficiency (dependent variable) and
discharge (independent variable) as described in Murdock et al. 2001.
A valid estimate requires the following assumptions to be true concerning the trap
efficiency trials:
1) All marked fish passed the trap or were recaptured during time period i.
2) The probability of capturing a marked or unmarked fish is equal.
3) All marked fish recaptured were identified.
4) Marks were not lost between the time of release and recapture.
Incidental Species
When time permits, incidental species should be measured and weighed as described for
target species. All incidental species will be released downstream of the trap.
DATA ANALYSIS
ESTIMATING TOTAL MIGRATION
Estimating migration for any period, whether a short time interval or an entire season,
requires a catch and an estimate of trap efficiency. Estimating abundance from a set of
trapping data is not always straightforward. A variety of approaches have been used. In
many cases the most appropriate approach will not become apparent until after all of the
field work is completed and the data is analyzed. The biologist needs to always temper
his/her decision on the approach with knowledge of the behavior of the targeted species.
A plausible rationale should be developed to explain and support these decisions. Four
general approaches are outlined in this section (Seiler and Volkhardt, 2005).
.
1. Estimating discreet outmigration periods using individual trap efficiency
estimates. This approach estimates migration for discreet time periods, typically a day or
a week, using a single test to estimate trap efficiency or by pooling several efficiency
trials to develop a mark-recapture based estimate of the migration for the time period.
Migration over the discreet period, Ni, is found using the simple equation;
MC
(1)
Nˆ i  i i
Ri
Bias in this estimate can be reduced using the Peterson mark-recapture equation;
 ( M  1)(Ci  1) 
Nˆ i   i
 1
( Ri  1)


(2)
Where
Mi
Ci
Ri
=
=
=
Number of fish marked and released during discreet period i,
Number of unmarked fish captured during discreet period i, and
Number of marked fish recaptured during discreet period i.
The variance, V(Ni), of the Peterson estimate can be calculated using;
V ( Nˆ i )  Nˆ i2
(Ci  Ri )
(Ci  1)( Ri  2)
(3)
Total juvenile production is estimated by the sum of the estimated migrations over
discreet periods and the variance of the total production is the sum of the variances. The
95% confidence interval (CI) is  1.96(sd).
This approach assumes each estimate of trap efficiency is an accurate measure of the
proportion of downstream migrants caught in the trap. Since each test actually represents
a single measure, it would be expected to include error. Assuming error is normally
distributed, this approach argues for estimating discreet periods of short duration (e.g., 1
day) since cumulative error from many samples should approach zero. We cannot
assume error is normally distributed where trap efficiencies are low, however. Estimates
of efficiency that are lower than the true efficiency cannot offset those that are higher as
the true value approaches zero (Seiler and Volkhardt, 2005).
A variation of this approach is to use another trap upstream to capture and mark migrants
over the trapping season. The recapture of these migrants in the downstream trap over
the season represents a single mark-recapture experiment. Since both marked and
unmarked fish should have an equal chance of being captured over time, the timing
distribution of marked releases should reflect the migration timing for the species.
Therefore, a weir trap located in a tributary is the best choice for this second trap since it
is designed to catch 100% of the passing migrants over the entire season. Total
production is estimated using Equation 1, substituting the total migration (N), total catch
of marked and unmarked fish (C), total marked releases (M), and total recaptures (R), for
Ni, Ci, Mi, and Ri in the equations. Variance, V(N), is estimated by the variance of the
trap efficiency estimate, R/M, which is a binomial multiplied by the C2 over (R/M)4.
This reduces to:
R
R
1  
C2
C 2 M ( M  R)
M M
V (N ) 
*

(4)
4
M
R3
 R
 
M 
2. Modeling Trap Efficiency. This approach estimates trap efficiency from an
independent variable, typically stream flow. A series of trap efficiency tests are
conducted over a range of flows and analyzed to determine if a significant relationship
can be established (Figure 8). When using regression analysis, it has been suggested that
the observed F should exceed the chosen test percentage point by a factor of four or more
for the relationship to be considered of value for predictive purposes (Draper and Smith
1998).
16%
Trap Efficiency (%)
Y=0.477-0.0585(lnX)
12%
8%
4%
0%
10
15
20
25
30
35
40
45
50
Mean Daily Discharge (cms)
Figure 2. Age 0 sockeye trap efficiency and 95% confidence intervals as a function of stream
discharge, Cedar River, Washington USA (Source: Seiler and Volkhardt, 2005).
Using this approach, migration on day i, Ni, and its variance, V(Ni), are estimated by;
C
Nˆ i  i
eˆi
(5)
2
C
V ( Nˆ i )  V (eˆi ) 4i
eˆi
(6)
If linear regression is used to estimate trap efficiency, its variance is estimated by;


( X  X )2
1
V (eˆi )  MSE 1   n i
 n
( X i  X )2


i 1







(7)
where:
eˆi  The trap efficiency predicted on day i by the regression equation, f(X i ),
MSE  The mean square error of the regression ,
n  The number of trap efficiency tests used in the regression , and
X i  The independen t variable on day i.
3. Stratifying Trap Efficiency. Like #2, this approach also predicts trap efficiency
using an independent variable. In this case, efficiencies are fairly constant over some
range of the independent variable or a condition class. Then as the independent variable
passes some threshold or another condition class occurs, efficiencies change or “step” to
a new level. For example, if the trap is placed in a “U”-shaped channel adjacent to a
wide gravel bar, trap efficiencies may be at one level when flows are contained in the
channel and another when higher discharge causes a substantial portion of the flow to
spread out across the gravel bar. Fish size may change over the trapping season causing
changes in trap efficiency by time strata. Turbidity levels may cause changes in
efficiencies as well. In some locations, fish are better able to avoid traps during day
fishing periods. In this case, efficiency data would be stratified by condition class (i.e.,
day and night periods) (Figure 3). Mean trap efficiency is calculated for each strata
(Seiler and Volkhardt, 2005).
60.00%
Trap Efficiency
50.00%
40.00%
30.00%
20.00%
10.00%
0.00%
Day
Night
Diel Condition Class
Figure 3. Range and mean trap efficiencies stratified by diel fishing periods, Issaquah Creek,
Washington USA (Source: Seiler and Volkhardt, 2005).
Migration is estimated for discreet periods when the independent variable is within a
defined stratum by dividing the sum of the catch by the mean trap efficiency for the
stratum. The variance of the estimate is calculated using Equation 6, substituting the
mean trap efficiency for the stratum, ēj, for the predicted trap efficiency on day i. (Seiler
and Volkhardt, 2005).
4. Back-Calculating Production. Using this approach, fish captured in the screw
trap are marked or tagged and released downstream. Recapture occurs at another location
and/or life stage and a Peterson estimate of production is made. Typically, recaptures
occur when the returning adults are sampled in a fishery, upon the spawning grounds, or
at another sampling location such as a trap. The term “back-calculating production”
generally refers to calculating downstream migrant production from the recapture of
adults marked as downstream migrants captured in the trap. However, production
estimates could also be achieved using this method by sampling marked juveniles from
the lower river or estuary (Seiler and Volkhardt, 2005).
Production is estimated using the same equation described for the variation of approach
#1 above. The variance is estimated by Equation #4. This approach is most useful where
trap efficiency estimates are difficult to make. If mark or tag sampling occurs while the
juvenile fish are still on their seaward migration, then this approach could be used for all
species. If sampling will not occur until the adults return, then this method is more easily
applied where nearly the entire cohort returns in a single year (e.g. coho). Age sampling
would be required for this approach to work for species that return to spawn in multiple
year classes (Seiler and Volkhardt, 2005).
LITERATURE CITED
Anderson, R. O. and R. M. Neumann. 1996. Length, weight, and associated structural
indices. Pages 447-482 in: B.R. Murphy and D.W. Willis, editors. Fisheries Techniques,
2nd edition. American Fisheries Society, Bethesda, MD.
Ames, J., and D.E. Phinney. 1977. Puget Sound summer-fall chinook methodology;
escapement goals, run size forecasts, and in-season run size updates. Washington
Department of Fisheries Technical Report 29, Olympia, WA.
Baranski, C. 1989. Coho smolt production in ten Puget Sound streams. State of
Washington Department of Fisheries technical report #99. Olympia, WA.
Beidler, W.M., and T.E. Nickelson. 1980. An evaluation of the Oregon Department of
Fish and Wildlife standard spawning fish survey system for coho salmon. Oregon
Department of Fish and Wildlife Information Report Series 80-9, Portland, OR.
Brown, T.G. and G.F. Hartman 1988. Contribution of seasonally flooded lands and minor
tributaries to the production of coho salmon in Carnation Creek, British Columbia.
Transactions of the American Fisheries Society 117:546-551.
Draper, N.R. and H. Smith. 1998. Applied regression analysis, 3rd edition. John Wiley &
Sons, Inc.
Hartman, G.F., B.C. Andersen, and J.C. Scrivener. 1982. Seaward movement of coho
salmon (Oncorhynchus kisutch) fry in Carnation Creek, an unstable coastal stream in
British Columbia, Canadian Journal of Fisheries and Aquatic Sciences 39:588-597.
Hilborn, R., B.G. Bue, and S. Sharr. 1999. Estimating spawning escapements from
periodic counts: a comparison of methods. Canadian Journal of Fisheries and Aquatic
Sciences 56:888-896.
Johnson, D.H., B.M. Shrier, J.S. O’Neal, J.A. Knutzen, X. Augerot, T.A. O’Neil, I.G.
Cowx. 2005. Measuring and Monitoring Biological Diversity – Standard Methods for
Freshwater Fishes. Chapter X. Scoop and Rotary Screw Traps. Measuring Juvenile
Anadromous Salmonid Production in Boatable River Systems. Contributing Authors:
Gregory C. Volkhardt and David E. Seiler. Smithsonian Institution Press. In prep.
Kennen, J.G., S.J. Wisniewski, N.H. Ringler, and H.M. Hawkins. 1994. Application and
modification of an auger trap to quantify emigrating fishes in Lake Ontario tributaries.
North American Journal of Fisheries Management 14:828-836.
Letcher, B.H., G. Gries, and F. Juanes. 2002. Survival of stream-dwelling Atlantic
salmon: effects of life history variation, season, and age. Transactions of the American
Fisheries Society 131:838-854.
Mobrand, L.E., J.A. Lichatowich, L.C. Lestelle, and T.S. Vogel. 1997. An approach to
describing ecosystems performance “through the eyes of salmon”. Canadian Journal of
Fisheries and Aquatic Sciences 54:2964-2973.
Moussalli, E., and R. Hilborn. 1986. Optimal stock size and harvest rate in multistage life
history models. Canadian Journal of Fisheries and Aquatic Sciences 43:135-141.
Murdoch A., K. Petersen, M. Tonseth, and T. Miller. 1998b. Freshwater production and
emigration of juvenile spring chinook salmon from the Chiwawa River in 1997. Report
No. H98-05. Washington Department of Fish and Wildlife, Olympia WA
Murdoch, A., K. Petersen, T. Miller, M. Tonseth, and T. Randolph. 2000. Freshwater
production and emigration of juvenile spring chinook from the Chiwawa River in 1999.
Prepared for: Public Utility District Number 1 of Chelan County. Wenatchee, WA.
Murdoch A., K. Petersen, T. Miller, M. Tonseth, and T. Randolph. 2001. Freshwater
production and emigration of juvenile spring chinook salmon from the Chiwawa River in
2000. Washington Department of Fish and Wildlife, Olympia WA. 45 pages +
appendices.
Nickelson, T.E. 1998. ODFW coastal salmonid population and habitat monitoring
program. Oregon Department of Fish and Wildlife. Salem, OR.
Olsson, I.C., L.A. Greenberg, and A.G. Eklov. 2001. Effect of an artificial pond on
migrating brown trout smolts. North American Journal of Fisheries Management 21:498506.
Orciari, R.D., G.H. Leonard, D.J. Mysling, and E.C. Schluntz. 1994. Survival, growth,
and smolt production of Atlantic salmon stocked as fry in a southern New England
stream. North American Journal of Fisheries Management 14:588-606.
Petersen, K., R. Eltrich, A. Mikkelsen, and M. Tonseth. 1995. Downstream movement
and emigration of chinook salmon from the Chiwawa River in 1994. Report No. H95-09.
Washington Department of Fish and Wildlife, Olympia, WA.
Roper, B. and D.L. Scarnecchia. 1996. A comparison of trap efficiencies for wild and
hatchery age-0 chinook salmon. North American Journal of Fisheries Management
16:214-217.
Schoeneman, D.E., R.T. Pressey, and C.O. Junge, Jr. 1961. Mortalities of downstream
migrant salmon at McNary Dam. Transactions of the American Fisheries Society 90:5872.
Seiler, D.E. and G.C. Volkhardt. 2005. Scoop and Rotary Screw Traps. Measuring
Juvenile Anadromous Salmonid Production in Boatable River Systems. Chapter of:
Measuring and Monitoring Biological Diversity – Standard Methods for Freshwater
Fishes. David H. Johnson, B.M. Shrier, J.S. O’Neal, J.A. Knutzen, X. Augerot, T.A.
O’Neil, I.G. Cowx. 2005. Smithsonian Institution Press. In prep.
Seiler, D., S. Neuhauser, and M. Ackley. 1981. Upstream/downstream salmonid trapping
project, 1977-1980, progress report #144. State of Washington Department of Fisheries.
Olympia, WA.
Seiler, D., S. Neuhauser, and L. Kishimoto. 2003. 2002 Skagit River wild 0+ chinook
production evaluation annual report. FPA 03-11. Washington Department of Fish and
Wildlife. Olympia, WA.
Thedinga J.F., M.L. Murphy, S.W. Johnson, J.M. Lorenz, and K.V. Koski. 1994.
Determination of salmonid smolt yield with rotary-screw traps in the Situk River, Alaska,
to predict effects of glacial flooding. North American Journal of Fisheries Management
14:837-851.
Tsumura, K., and J.M.B. Hume. 1986. Two variations of a salmonid smolt trap for small
rivers. North American Journal of Fisheries Management 6:272-276.
Wagner, H.H., R.L. Wallace, and H.J. Campbell. 1963. The seaward migration and return
of hatchery-reared steelhead trout, Salmo gairdneri Richardson, in the Alsea River,
Oregon. Transactions of the American Fisheries Society 92:202-210.
APPENDIX 1
Guidlines for Non-Lethal Fry and Smolt Sampling for DNA
Analysis
The goal is to take a small enough piece of a non-critical tissue (e.g., fin) to have little or
no impact on the subsequent survival of the fish but that is adequate to allow genetic
analysis. DNA analysis is ideal for this for two reasons: 1) all living cells of an organism have
essentially the same DNA composition (unlike the tissue specific expression characteristic of
allozymes and other proteins), so that tissues such as fin and opercle can provide adequate
samples, and 2) amplification of the resulting DNA from such samples via the PCR (polymerase
chain reaction) provides the sensitivity of detection to enable working with very small pieces of
tissue and small amounts of DNA. [For mammals, this approach has be used successfully to
characterize animals by analyzing DNA extracted from hair follicles, blot spatters, and scat
samples.]
•
The minimum amount of tissue that is needed is approximately the size of this circle: (a
piece of tissue with the same approximate surface area as a 1.5mm diameter disc). Failure to
take a large enough tissue sample can prevent successful DNA analysis. The
recommended sources of such a tissue sample are any of the following:
1) A distal portion of the dorsal lobe of the caudal fin
2) A distal portion of one of the pelvic fins
3) Smaller distal portions of both pelvic fins
4) One entire pelvic fin
By sampling only the distal portion of a fin, we expect that the fish will successfully regenerate
the entire fin over time. In contrast, removing an entire fin often results in little or no fin
regeneration, presumably leaving the fish at a selective disadvantage.
When sampling larger fish, a larger sample is preferred (e.g., a piece of tissue approximately
the size of one of these circles:
[approx. 3mm diameter] or
[approx. 4.5mm diameter]),
because this will provide more material (DNA). The “extra” tissue provides a reserve that can be
used to overcome some types of analytical problems in the lab by repeated analysis and/or it
provides material that can be used for subsequent analyses (for example to examine additional
loci at a future date) or can be shared with other laboratories/agencies.
Live fish should be handled appropriately before, during, and after sampling. This will
probably involve: a) anesthetization prior to handling for tissue sampling (and taking of
measurements or other biological samples such as scales), b) careful handling during sampling to
avoid injury and scale/mucous loss, and c) holding fish in a recovery vessel after sampling (until
the anesthetic has worn off) before releasing them in a way that minimizes immediate mortality
due to predation of other effects.
Each tissue sample should be placed in a vial that contains DNA preservative solution
(and an appropriate label -- preprinted by WDFW [preferred] or written in pencil) immediately after
it is taken. We recommend using vials that are approximately 3/4 full of preservative solution and
never adding more than 1/5 of this volume of tissue (to ensure adequate preservation). Please
rinse forceps, scissors, etc. (with fresh water) and dry them between fish to minimize the chance
of cross- contamination of samples. Such preserved samples should be stored at ambient
temperatures (20-80oF) until they are returned to the WDFW Genetics Laboratory in Olympia.
If you have questions or need additional information, please telephone Todd Kassler (360-9022722), Sewall Young (360-902-2773), or the Genetics Lab at 360-902-2775).
SECTION 3. ADULT ENUMERATION
PURPOSE
PROCEDURE
SECTION 4. BENTHIC MACRO-INVERTEBRATE
ASSESSMENT
PURPOSE
The health of a stream can be determined from the species of macro-invertebrates
present. Some species of aquatic insects are very sensitive to water quality problems and
others are affected by sedimentation or temperature. The purpose of this protocol is to
provide for a standard method of measuring changes in the macro-invertebrate
assemblages of streams in the Okanogan subbasin. A macro-invertebrate index is
calculated based upon previous studies and used to compare results against future
measures at the same site and other sites.
EQUIPMENT
Modified kick net (D-Frame with 500 micro meter mesh) and 4 ft handle (Wildco # 425C50), stop watch, plastic buckets (8-10 qt), sieve with 500 micro meter mesh openings.
Forceps, wash bottle, spatula, spoon or scoop, funnel with large bore spout, sample jars,
ethanol (95%), rubber gloves, cooler, composite benthic sample labels with preprinted ID
numbers (barcodes), blank labels on waterproof paper for inside of jars, sample collection
form, clear packing tape for sealing jars, plastic electrical tape, scissors, appropriate field
forms.
SITE SELECTION
The sample reaches should be laid out according to procedures laid out in Section 1,
Physical Habitat Protocols.
SAMPLING DURATION
Sampling should occur at the same time that other samples are taken from the stream
reach for fish assemblages and for habitat measures.
PROCEDURE TO COLLECT KICK NET SAMPLES
Protocol modified from: Peck et al. (2001) Table 11-3 and 11-4 Targeted Riffle Sample
Step 1: Before sampling, survey the stream reach to estimate the total number (and
area) of riffle habitat units contained in the defined stream reach. To be considered as a
unit the area must be greater than 1 square foot.
A. Do not sample poorly represented habitats. If the reach contains less than 8 ft2
of riffle macrohabitat, then do not collect a targeted riffle sample.
B. If the reach contains more than one distinct riffle macrohabitat unit but less
than eight, allocate the eight sampling points among the units so as to spread
the effort throughout the reach as much as possible. You may need to collect
more than one kick sample from a given riffle unit.
C. If the number of riffle macrohabitat units is greater than eight, skip one or
more habitat units at random as you work upstream, again attempting to
spread the sampling points throughout the reach.
Step 2: Begin sampling at the most downstream riffle unit, and sample units as they are
encountered to minimize instream disturbance.
Step 3: At each unit exclude “margin” habitats by constraining the potential sampling
area. Margin habitats are edges, along the channel margins or upstream or downstream
edges of the riffle macrohabitat unit. Define a core area for each riffle unit as the central
portion, visually estimating a “buffer” strip circumscribing the identified unit. In some
cases, the macrohabitat unit may be so small that it will not be feasible to define a core
area and avoid and edge.
Step 4: Visually lay out the core area of the unit sampled into 9 equal quadrants (i.e
3X3 grid). For each macrohabitat type, select a quadrant for sampling at random from
the following list of locations (right and left are determined as you look downstream)
Lower right quadrant
Lower center quadrant
Lower left quadrant
Right center quadrant
Center quadrant
Left center quadrant
Upper right quadrant
Upper center quadrant
Upper left quadrant
Step 5: Beginning at the most downstream riffle unit within the sampling reach, locate
the sampling point within the microhabitat as described in Steps 3 and 4.
Step 6: Attach the 4 ft handle to the kick net. Make sure that the handle is on tight or
the net may become twisted in a strong current, causing the loss of part of the sample.
Step 7: With the net opening facing upstream, position the net quickly and securely on
the stream bottom to eliminate gaps under the frame. Avoid large rocks that prevent the
sampler from sealing properly on the stream bottom.
Step 8: Holding the net in position on the substrate, visually define a rectangular
quadrant that is one net width wide and one net width long upstream of the net opening.
The area within this quadrant is 0.09 m2 (1 ft.2). Alternatively place a wire frame of the
correct dimensions in front of the net to help delineate the quadrant to be sampled.
Step 9: Hold the net in place with your knees. Check the quadrant for heavy
organisms, such as mussels and snails. Remove these organisms from the substrate by
hand and place them into the net. Pick up any loose rocks or other larger substrate
particles in the quadrant. Use your hands or a small scrub brush to dislodge organisms so
that they are washed into the net. Scrub all rocks that are golf ball sized or larger and
which are over halfway into the quadrant. Large rocks that are less than halfway into the
quadrant are pushed aside. After scrubbing, place the substrate particles outside the
quadrant.
Step 10: Keep holding the sampler securely in position. Start at the upstream end of
the quadrant, vigorously kick the remaining finer substrate within the quadrant for 30
seconds (use a stopwatch).
Step 11: Pull the net out of the water. Immerse the net in the stream several times to
remove fine sediments and to concentrate organisms at the end of the net. Avoid having
any water or material enter the mouth of the net during this operation.
Step 12: Invert the net into a plastic bucket marked “TARGETED RIFFLE” and
transfer the sample. Inspect the net for any residual organisms clinging to the net and
deposit them into the bucket as well. Use forceps if necessary to remove organisms from
the net. Carefully inspect any large objects (such as rocks, sticks, and leaves) in the
bucket and wash any organisms found off the objects and into the bucket before
discarding the object. Remove as much detritus as possible without loosing organisms.
Step 13: Record the nearest transect location in the box for the sample on the Sample
Collection Form. Place an “X” in the appropriate substrate type box for the transect on
the Collection Form.
Fine sand:
Not gritty (silt/clay/muck < 0.06mm diam.) to gritty, up to ladybug
sized (2 mm diam.)
Gravel:
Coarse:
Other:
Fine to coarse gravel (ladybug to tennis ball sized; 2mm to 64 mm
diam.)
Cobble to boulder (tennis ball to car sized; 64mm to 4000 mm).
Bedrock (larger than car sized; > 4000 mm), hardpan (firm,
consolidated fine substrate, wood of any size, aquatic vegetation,
etc.). Note type of “Other” substrate in comments on field form.
Step 14: Thoroughly rinse the net before proceeding to the next sampling location.
Step 15: Repeat steps 1-14 at subsequent riffle sampling points until 8 kick samples
have been collected and placed in the “TARGETED RIFFLE” BUCKET.
PROCEDURE FOR PREPARING COMPOSITE SAMPLES FOR
IDENTIFICATION
Protocol modified from: Peck et al. (2001), Table 11-5
Step 1: Pour the entire contents of the “TARGETED RIFFLE” bucket through a sieve
with 500 micro-meter mesh. Remove any large objects and wash any clinging organisms
back into the sieve before discarding.
Step 2: Using a wash bottle filled with stream water, rinse all organisms from the
bucket into the sieve. This is the composite reach-wide sample for the site.
Step 3: Estimate the total volume of the sample in the sieve and determine how large a
jar will be needed for the sample. Avoid using more than one jar for each composite
sample.
Step 4: Fill in a “TARGETED RIFFLE” sample label with the stream ID and date of
collection. Attach the completed label to the jar and cover it with a strip of clear packing
tape.
Step 5: Wash the contents of the sieve to one side by gently agitating the sieve in the
water. Wash the sample into the jar, using as little water from the wash bottle as possible.
Use a large bore funnel if necessary. If the jar is too full, pour off some water through
the sieve until the jar is not more than ¼ full, or use a second jar if a larger one is not
available. Carefully examine the sieve for any remaining organisms and use forceps to
place them into the sample jar.
Step 6: Place a waterproof label with the following information inside each jar:
 Project number
 Worksite description
 Type of sampler and mesh size used
 Name of stream
 Date of collection
 Collector’s name
 Number of transects sampled composited
Step 7: Completely fill the jar with the 96% ethanol (no headspace) so that the final
concentration of ethanol is between 75 and 90%. It is very important that sufficient
ethanol be used, or the organisms will not be properly preserved.
Step 8: Replace the cap on each jar. Slowly tip the jar to a horizontal position, then
gently rotate the jar to mix the preservative. Do not invert or shake the jar. After mixing,
seal each jar with plastic electrical tape.
Step 9:
Store labeled composite samples in a container until transport to the laboratory.
MULTI-METRIC INDEX DEVELOPMENT
Protocol taken from: Wiseman (2003), Tables 1, 8, and 9
Step 1: Obtain results from laboratory analysis of species present in the REACHWIDE
composite sample and their relative abundance.
Step 2:
Determine the following metrics from the laboratory sample:
 Percent of the family Chironomidae of the total sample count
 Percent of the Orders Ephemeroptera, Plecoptera, and Trichoptera of
the total sample count
 Percent of the Order Ephemeroptera of the total sample count












Step 3:
(2003).
Hilsenhoff Biotic Index (HBI) which is calculated by multiplying the
number of individuals of each species by its assigned tolerance value,
summing these products, and dividing by the total number of
individuals
Total number of taxa
Number of highly intolerant taxa, as defined by Wiseman (1998)
Percent of clinger taxa of the total sample count
Number of clinger taxa
Number of intolerant taxa with a tolerance value less than 3 (TV3)
Percent of the tolerant taxa of the total sample count with a tolerance
value greater than 7 (TV7)
Percent of the top 3 abundant taxa of the total sample count
Percent of the filter taxa of the total sample count
Percent of the predator taxa of the total sample count
Percent of the scraper taxa of the total sample count
Number of long-lived taxa
Score each indicator based upon the following tables taken from Wiseman
Table 4. Scoring criteria for Puget lowland area MMI.
Category
Richness
Richness
Richness
Richness
Tolerance
Tolerance
Tolerance
Trophic/habitat
Trophic/habitat
Voltinism
Metric
Total richness
Ephemeroptera richness
Plecoptera richness
Trichoptera richness
Intolerant richness (bi)
% tolerant (TV7)
% top 3 abundant
% predators
% clingers
Long lived richness
1
<24
<4
<3
<4
<2
>19
>70
<11
<26
<3
Scoring Criteria
3
5
24-33
>33
4-6
>6
3-5
>5
4-6
>6
2
>2
11-19
<11
54-70
<54
11-19
>19
26-47
>47
3-5
>5
Table 5. Scoring criteria for Cascade MMI.
Category
Composition
Richness
Richness
Richness
Richness
Tolerance
Tolerance
Tolerance
Trophic/habitat
Trophic/habitat
Metric
%
Ephemeroptera
Total richness
Plecoptera
richness
Trichoptera
richness
Clinger richness
Intolerant
richness (bi)
% tolerant (bi)
HBI
% filterers
% clingers
1
<35
Scoring Criteria
3
5
35-57
>57
<37
<5
37-52
5-9
>52
>9
<9
9-12
>12
<12
<6
12-16
6-9
>16
>9
>23
>3.8
>28
<36
12-23
2.8-3.8
15-28
36-54
<12
<2.8
<15
>54
Scoring criteria for eastern Washington have recently been developed by the Department
of Ecology.
DEFINITION OF TERMS
REFERENCES CITED
Peck, D.V., J.M. Lazorchak, and D.J. Klemm (editors). Unpublished draft.
Environmental Monitoring and Assessment Program - Surface Waters: Western
Pilot Study Field Operations Manual for Wadeable Streams. EPA/XXX/XXX/XXXX. U.S. Environmental Protection Agency, Washington, D.C.
Rodgers, J.D. (2002). Abundance monitoring of juvenile salmonids in Oregon coastal
streams, 2001. Mon. Rpt. No. OPSW-ODFW-2002-1. Oregon Dept. Fish and
Wildlife. Portland, OR. 51p.
Thurow, R.F. (1994). Underwater methods for study of salmonids in the Intermountain
West. U.S. Forest Service. Gen Tech Rept. INT-GTR-307. 29 p.
SECTION 5. METHOD FOR DETECTING
STEELHEAD REDDS
PURPOSE
Estimates of adult spawner abundance and/or redd counts will allow investigators to
monitor changes in the spawner abundance and redd distribution. Redd surveys in the
Okanogan River basin have occurred for years using index sites for summer/fall Chinook
and sockeye salmon by WDFW and ONA respectively and these surveys are expected to
continue under a separate set of protocols not included in this section. The protocols
herein will be used by the Colville Tribes and ONA specifically to conduct surveys for
summer steelhead redds throughout the Okanogan River basin as part of the OBMEP.
The protocols contained within this section were specifically developed to conduct
steelhead redd surveys in the Okanogan River basin. These protocols may require some
modification if applied to other species or locations.
This information is useful to:




Determine if the spatial distribution of spawning fish changes over time;
Determine if the abundance of returning adults is changing over time;
Identify and map preferred spawning habitat areas;
Conduct surveys that provide census level precision and accuracy at the lowest
possible cost.
SITE SELECTION
The Upper Columbia Strategy (Hillman 2004) calls for conducting a census of redds in
the target watershed, if possible, but also provides for conducting these surveys at
established index areas or using a random sample of fixed length reaches. In the
Okanogan River basin a mixture of these approaches is most appropriate (i.e. index
surveys for sockeye salmon and random samples for habitats that may become available
in the future). To establish potential spawning areas first you need to established the
distribution of redds across all habitats and eliminate areas that fail to meet basic habitat
parameters (Bjornn and Reiser 1991). Habitats that are eliminated represent large areas
where habitats typify conditions under which redds would not be constructed and this
condition is unlikely to change significantly over time (i.e. substrate is >60% fines and
dominated by particles less than 0.6 cm in diameter, water depths are greater than 6cm
but less than 300cm, velocities are less than 10cm/s, and gradients are less that 0.1%).
The lower 23-miles of the Okanogan River consist of waters created from the influence
from the Well’s pool and represent a good example of habitats with no spawning
potential. The inundated reach represents considerable logistic challenges due to large
size and poor visibility. Consensus of local biologist’s provides the best information
possible for this reach and provides a solid scientific basis for eliminating from the
potential sampling universe for redd surveys. The upper extent of the Wells pool effect is
agreed to be located at the confluence with Chiliwist Creek. The same logic can be used
to eliminate all lake habitats from further consideration. The Okanogan River from
upstream of the town of Ellisforde (Washington) to the town of Tonasket (Washington)
represents an extremely low gradient reach with mostly fine sediment substrates and
provides virtually no potential spawning habitat. The reach from Salmon Creek
downstream to the town of Malott (Washington) represents a reach with very minimal
spawning habitat resulting from low gradient and either large cobble/boulder or fine
sediment substrates. Extremely small patches of spawning habitats do represent some
minor spawning potential but these areas represent less than 1% of all potential habitats
available along the Okanogan River main-stem.
Based upon the best available scientific data the areas presented in Table 1 should be
considered a complete census for the Okanogan River and Similkameen River main-stem
reaches in the United States. Reaches in Canada will be determined in the future as more
information is collected and most tributaries will be sampled in there entirety unless preexisting index areas exist (i.e. Omak Creek below Mission Falls). Steelhead redd count
surveys will be conducted at EMAP sites located in areas that have the potential to be
accessed by fish in the near future in order to establish a before treatment baseline. Redd
surveys in areas using EMAP sites will use the downstream extent of the randomly
selected EMAP site as the farthest point downstream to begin a 1km reach. As additional
information becomes available we will focus on reducing costs where a census design
can be maintained through reduced field efforts.
Table 1: Long-term redd survey reaches for Okanogan basin monitoring and evaluation project.
Reach
Reach
Reach (rkm)
length
Code
Main-stem Habitats
(km)
S1
Similkameen/Okanogan Confluence(0) to Enloe Dam (14.6)
14.6
O1
Okanogan River at Chiliwist Creek(24.4) to Malott bridge (26.5)
2.1
O2
Okanogan River at Salmon Creek(41.4) to CCT F&W Office(52.3)
10.9
O3
Okanogan River at CCT F&W Office(52.3) to Riverside(66.1)
13.8
O4
Okanogan River at Riverside(66.1) to Janis Bridge(84.6)
18.5
O5
Okanogan River at Janis bridge(84.6) to Tonasket park(91.4)
6.8
Okanogan River at Oxbow Lake(106.6) to Confluence with
O6
Similkameen(119.5)
12.9
O7
Okanogan River at confluence(119.5) to Zoesel dam(127.0)
7.5
TU1
B1
N1
OM1
OM2
OM12
OM48
OM366
OM361
TO1
Tributary Habitats
Tunk cr @Okanogan river Confluence (0) to High water mark (0.2)
Bonaparte Creek/Okanogan River confluence (0) to Bonaparte Falls
(1.6)
Ninemile Creek/Okanogan River confluence (0) to Eder land (1.7)
Omak Creek/Okanogan River confluence (0) to Omak Lake Road Bridge
(2.0)
EMAP Site 19 lower (5.1) to Mission Falls (8.2)
Jim Cr rd bridge(29.4) to EMAP Site 12 lower (30.4)
Staploop cr (26.8) to 500 meters below site 48 lower (27.8)
end of forest road at Dutch Anderson bridge (21.5) to Dutch Anderson
br.(22.5)
Above Mission Falls(10.75) to EMAP site 361 upper (11.75)
Tonasket Creek/Okanogan River confluence (0) to Tonasket Falls (3.5)
0.2
1.6
1.7
2.0
3.1
1.0
1.0
1.0
1.0
3.5
SAMPLING DURATION
Sampling should occur beginning with the earliest anticipated spawning for the target
species (steelhead ) and should continue until the end of the normal spawning period
unless prohibited by environmental conditions (i.e. high flows, increased turbidity, etc.).
The spawning period for summer steelhead typically begins when water temperatures
reach 39◦F and concludes shortly after water temperatures reach 49◦F (begin last week in
March and conclude no later than May 15). Surveyors should be aware that stream flow
conditions can alter the timing, visibility, and distribution of spawning activity from one
year to the next and spawner distribution within a stream system may be different for
early versus late spawners. Redd life (visibility) estimates should not normally be needed
if main stem surveys are conducted frequently enough so that new redds are readily
distinguishable and these issues are addressed in the protocols that follow. Main-stem
surveys will be conducted at intervals of no greater than once every two weeks. Mainstem surveys will begin in March but likely will be completed due to visibility
restrictions and the ascending limb of the hydrograph in a normal water year prior to May
1.
Based upon information from previous survey years (Arterburn and Fisher 2005, 2004)
tributary surveys are most likely to be conducted in May. Fish collection at the Omak
Creek weir/trap will be used to determine the most appropriate time to conduct tributary
surveys. Redd surveys will be conducted 2-weeks the ratio of kelt steelhead is higher than
new adult collections at the trap located on Omak Creek (rkm 1.5). Single pass surveys
conducted during the peak spawning period in the tributary habitats are not likely to
capture information regarding either early or late spawners and therefore will represent
only a conservative index. However, the costs associated with collecting these data are
reduced by two thirds by using this method and the total number of redds observed in the
tributaries. The authors believe that the number of redds missed as a result of following a
single pass methodology will only represent a small percentage of the total count.
EQUIPMENT













Thermometer
Waders with non-slip soles
Multi-colored flagging
Field notebook
Pencils and sharpie
Waterproof field record form
Trimble GPS data logger
Vest or day pack
Polarized glasses
Stream map to indicate location of spawning activity 
Drinking water and food
2-one man Skookum cata-rafts
2-way radios 
ANATOMY OF A REDD
Summer Steelhead redds are considered to typically cover 4.4 to 5.4 square meters (Bjorn
and Reiser 1991). Redds are typically found in areas of down-welling (were water is
hydraulically forced into the substrate) and where water depths are >24 cm (Smith 1973),
with velocities of 40 to 91 cm/sec (Smith 1973) and substrate diameters range from 0.6 to
10.2 cm (Hunter 1973). The investigator should be familiar with the size of the redds
produced by other species of fish that may be spawning at the time the surveys are
conducted (Figure 1).
Figure 1: Steelhead Redd on the Okanogan River near Driscoll Island dug on 4/22/05. Note: This
photo clearly illustrates tail spill, pit, substrate size, and area of disturbance if these terms are
unfamiliar please reference page 93 of Meehan 1991.
MULTIPLE PASS MARKED REDD SURVEY METHOD
Summation of the number of new redds counted throughout the entire spawning season
will be the method used on the main-stem Okanogan and Similkameen Rivers. By
marking redds, old but still visible redds are not counted twice. Individual redds (or
groups of redds, in the case of superimposition) are to be flagged and documented.
Redds are marked by GPS and by flagging tied to bushes or trees on the stream bank
adjacent to the area were redds are observed. Each flag will be marked with the date,
coordinate and distance, flag number, total number of redds represented by the flag, and
surveyor initials. This same information will be captured electronically by entering it into
a Trimble data logger. Redds will be flagged and numbered consecutively as they are
encountered during each survey. The color of the flagging should be changed for each
survey. Incomplete redds or test digs should not be flagged and not counted. On
subsequent surveys, all redds should be counted and every attempt should be made to
locate all flags. Potential biases can result from redd superimposition or removal of
flagging by people. Surveyors will check all flags from previous surveys as they search
for new redds and note missing flags by a gap in the numbering sequence. If a flag is
found to be missing, the surveyor will note it on the field form and re-flag redds based on
the previous GPS location. Re-flagged redds will not be counted as new redds.
PEAK REDD COUNT METHOD
The Peak Redd Count Method is the primary method used on tributary habitats. This
method is most often used where individual redds are unlikely to be missed due to high
visibility resulting from narrow width, shallow depths and good water clarity
characteristic of most small tributaries. Under The peak count method, all redds are
simply counted during a single foot survey. The total redd count per stream or stream
reach is used to estimate total redd deposition in that reach. However, this method is
likely to be biased low due to two factors; 1) redds constructed earlier in the spawning
season but not counted because they are no longer discernable or, 2) redds constructed
after the survey is completed by late spawning fish (Hahn et al. 2001).
FOOT SURVEYS
Under most conditions foot surveys are the most appropriate method for counting redds
and detecting adult spawners. Where possible, foot surveys will be conducted on all sites
where water depths do not explicitly require a boat to obtain complete counts of live
and/or dead spawners and redds.
Observations are made from the banks and by walking into the stream as needed to
confirm redds and/or species of fish. The observer should wear Polaroid sunglasses,
carry a “write-in-the-rain” notebook to record data, and use surveyor’s plastic flagging to
mark redds. Weather conditions, water clarity and number of redds are also recorded.
BOAT /RAFT SURVEYS
A boat will be required to conduct redd surveys in large deep stream reaches that cannot
be safely waded. In areas where boat access is limited, surveyors should exit the boat
and conduct foot surveys where safe to do so. The rivers were divided into reaches to
allow for reference points regarding distribution maps (see table 1). When conducting
redd surveys using boats two people in two 1-man, 10’ Skookum Steelheader catarafts is
recommended as the most cost effective use of resources. .
FISH OBSERVATION
During redd surveys all fish observed need to be documented. If the summer steelhead is
dead please see carcass survey protocols below. Live summer steelhead observed need to
be documented to the greatest extent possible so observers need to be careful to not spook
fish before data can be collected. To help this all foot surveys are to be conducted
working upstream. Steelhead observed should be identified to species, adipose fin
observed to determine if it has been clipped or is present. Gender determination is
typically not possible unless fish are observed actively spawning (females are identified
by digging activity and males are identified by chasing and milting activity).
CARCASS SURVEYS
Generally this procedure applies only to salmon spawner surveys. However, steelhead
carcasses may be occasionally encountered during redd count surveys in which case the
following protocol should be used.
Carcass sampling should be conducted as part of any adult spawner survey in order to
obtain an accurate estimate of the total abundance of males and females. Carcass surveys
consist of counting dead steelhead or impinged kelts. The belly of the carcass will be
opened to observe gonad condition and this open belly is also to used to keep from
recounting the same carcasses. Carcass counts should be conducted concurrently with
redd counts throughout the sampling period. Specific information to be collected from
carcasses is include the location of where the carcass was found, length, sex, condition of
gonads (spent, unspent), adipose fin (present or absent).
The otoliths of the fish should be removed from the head and placed in a container with a
tag documenting the date, stream and reach collected, collectors initials, the presence or
absence of an adipose fin. Each contained should be labeled with a number that starts
with the last two digits of the year, month and day, then the reach code from Table 1 and
lastly the number of the sample collected that day (i.e. 050505S11 would represent the
first sample collected from the Similkameen River on May 5, 2005). If you are surveying
main-stem habitats keep other parties informed when samples are collected to keep from
having duplicate sample numbers. If the heads of fish are brought back to the office
remove the otoliths immediately upon return and place into vial with ethyl alcohol.
QA/QC PROCEDURES
In general, steelhead spawning areas can be surveyed by foot or rafts. In the Okanogan
basin redds enumerated during surveys will be ground verified by at least two trained,
knowledgeable staff until consensus is reached in order to maintain quality control. In the
event that disagreements arise, disputed redds will be recorded by GPS and revisited by
an additional senior biologist to make a final determination.
REMOVAL OF FLAGGING
Flagging used to mark redds should be removed at the conclusion of each field season.
This can be accomplished either during the final redd count survey or during summer
habitat or snorkel surveys. Biodegradable flagging, which will not require manual
removal, may be used as an alternative.
ESTIMATING TOTAL REDDS AND ESCAPMENT
Because all redds are marked, they represent a total count and not an estimate. However
this count only represents the area examined. Data can be extrapolated after all counts
are concluded to develop a number of redds for each sub-watershed and these combined
to estimate total redds for the Okanogan River. However, caution should be used in
deriving these estimates as often spawning areas are not evenly distributed throughout a
river reach. Total redd estimates in combination with spawner escapement sex ratio data
can be further expanded to provide estimates of total spawner escapement for the
watershed or sub-watershed. The number of redds is also critical in deriving production
estimates based upon fecundity and survival estimates.
“Spawning escapement” can be estimated as the number of redds times a “fish-per-redd”
estimate. WSRFB (2003) uses 1.23 chinook per redd, assuming one redd per female.
For steelhead, they assume 1.23 redds per female. A more accurate method currently
used by WDFW in the Upper Columbia Basin is based on the sex ratio of broodstock (not
recovered carcasses) collected randomly over the run (A. Murdoch, personal
communication, WDFW). For example, if the sex ratio of a random sample of the run is
1.5:1.0, the expansion factor for the run would be 2.5 fish/redd. This method is used for
all supplemented stocks within the Upper Columbia Basin. Another method, which can
be used if the sex ratio is unknown, is the “Modified Meekin Method” (A. Murdoch,
WDFW, personal communication). This method takes the 2.2 adults/redd (from Meekin
1967) and increases it by the proportion of jacks in the run. For example, if jacks make
up 10% of the run, the modified adults/redd would be 2.42 (2.2 x 1.1 = 2.42 adults/redd).
Summer-run steelhead spawning escapement estimates will be calculated under the
OBMEP based upo sex ratios as described above. Sex ratios will be obtained from the
Well’s dam broodstock collection efforts and from data collected at the Omak Creek
weir. If these data vary then escapement values will be calculated and reported using both
ratios. Both total redds and spawning escapement will be reported as “whole” numbers.
Redds will be reported in number and density by reach code. Fish data collected during
redd surveys will be used to supplement data collected from other sources (i.e. traps and
video counts) to help determine age, origin, sex, pre-spawn mortality, and other estimates
for each population.
LITERATURE CITED
Bjornn, T.C.and D.W.Reiser, 1991. Habitat requirements of salmonids in streams.
American Fisheries Society Special Publication 19:83-138.
Hahn, P., C. Kraemer, D. Hendrick, P. Castle, and L. Wood (2001). Washington State
Chinook salmon spawning escapement assessment in the Stillaguamish and Skagit
Rivers, 1998. Washington Department of Fish and Wildlife. Olympia, WA. 165p.
Hillman, T.W. 2004. Monitoring Strategy for the Upper Columbia Basin. Draft Report.
February, 2004. Preparded for the Upper Columbia Regional Technical Team,
Upper Columbia Salmon Recovery Board. Wenatchee Washington. 107pp.
Hunter, J. W. 1973. A discussion of game fish in the state of Washington as related to
water requirements. Report by the Washington State Department of Game, Fishery
Management Division, to the Washington State Department of Ecology, Olympia.
Jacobs, S.E. and T.E. Nickelson (1999). Use of stratified random sampling to estimate the
abundance of Oregon coastal coho salmon. Final report. Oregon Department of
Fish and Wildlife. Portland, OR. 29p. Version 5/18/2004 26
Mosey, T. R and L.J. Murphy.(2002). Spring and Summer Chinook Spawning Ground
Surveys on the Wenatchee River Basin, 2001. Chelan County Public Utility
District, Fish and Wildlife Operations.
Meehan, W. R., editor. 1991. Influence of forest and rangeland management on salmonid
fishes and their habitats. American Fisheries Society Special Publication 19.
Smith, A. K. 1973. Development and application of spawning velocity and depth criteria
for Oregan salmonids. Transaction of the American Fisheries Society 102:312-316.
Taylor, R.N(editor) (1997). Aquatic Field Protocols Adopted by the Fish, Farm, and
Forest Communities (FFFC) Technical Committee. Version 1.1
APPENDIX A
Instructions for Completing EMAP Site Redd Survey Field Form
1. Stream - Print the stream name
2. Observers - Enter the names of the persons doing the survey
3. EMAP Site # - Enter the EMAP site number to be surveyed.
4. Date of survey - Enter the day’s date: nm/dd/yy
5. Weather- Make a check mark to indicate weather conditions: clear, overcast,
rain. If weather conditions change during the survey, note this in the remarks
section at the end of the page.
6. Water clarity -Estimate water clarity at the beginning of the survey: clear, slight
moderate, or heavy. If water clarity changes during the survey, note this in the
remarks section at the end of the page.
7. Water temperature -Water temperature is taken in degrees Fahrenheit at the
beginning and end of the survey.
8. Time - Time when temperatures were taken.
9. Redd location-Indicate the location of each new redd as either within the lower
(transects A through F) or upper (transects F through K) section of the sampling
reach.
10. Flag color - Record the color of flagging used to mark new redds in the current
survey.
11. Number of live fish observed - Enter the number of live steelhead. If positive
identification is not possible, record the fish as an unknown.
12. Number of carcasses examined - Identify all carcasses to:
a. Species (Assumed to be steelhead unless otherwise notes) and sex,
b. The presence or absence of gametes and note pre-spawn mortality or fully
spawned out.
c. Examine all carcasses for adipose fin clips or any other fin clip.
d. By opening up the carcasses along the abdomen to check for the presence
or absence of gametes all the carcasses will be marked after examination.
13. Number of skeletons observed - Any fish that cannot be measured, or any
identifiable parts of fish found are considered skeletons.- If it is possible to
identify the species, record it appropriately; if not, record it as unknown.
14. Remarks - Add any, information discovered during the. survey such as barriers,
landslides, etc. Include any information necessary to clarify other entries on the
field form. Note that if a complete barrier exists that would prohibit adult
steelhead passage then enter a zero in the count section of the field form and fillout information as if the survey were completed but note no water or other reason
for the barrier. In main-stem sites where the river bottom can not be clearly
observed in riffle or side channel areas note visibility precludes survey and peak
count method applies. Note if this is likely to change in the future and describe
reason for visibility difficulty (i.e. too deep, turbidity, …, etc.)
SRFB MC-10
Example data form to be completed for each EMAP site redd survey.
REDD SURVEY FORM
Page _____of_______
Stream Name
_________________
OBSERVERS________________________
Reach Code
_________________
DATE(M/DD/YYYR)_____________________
EMAP SITE #
Water
Temperature
_________________
Start
End
________
________
DATA LOGGER FILE NAME__________________________________________
Time of
temperatures
________
_________
REDD INFORMATION
GPS way point
Version 2/12/2016
Flag #
Flag color
# Redds
Weather
Clear
overcast
Water
clarity(Turbidity)
Clear
slight
moderate
REDD INFORMATION
Flag
Flag #
color
GPS way point
rain
Heavy
# Redds
43
Species
Version 2/12/2016
Live or
Dead
Sex (M/F)
FISH/CARCASS OBSERVATIONS
Adipose
Spawned
Fin Present
Other marks
(Y/N)
(Y/N)
or tags
Length
(mm)
DNA #
Otolith #
44
Version 2/12/2016
45