Coho_Enumeration_Final

Escapement Numeration of Oncorhynchus kisutch in Chester Creek
Anchorage, Alaska
Senior Project
by
Brandon R. Forsythe
Alaska Pacific University
Department of Environmental Science
bforsythe@alaskapacific.edu
(907)688-1280
Advisor/Committee Chair: Rusty Myers
Committee member: Theresa Tanner - USFWS

Introduction
Alaska has numerous streams well known for their abundant salmon runs.
However, in several Alaskan cities salmon populations have declined due to
urbanization. An example of this would be Fish Creek (Milner & Oswood 2000) located
in Anchorage, AK. Chester Creek in Anchorage, Alaska has followed this pattern. This
small urban stream once supported several thousand salmon including Oncorhynchus
kisutch (coho) and Oncorhynchus gorbuscha (pinks), but currently supports coho,
rainbow trout, and dolly varden (Whitman 2002). It has been estimated that the adult
coho population returning to Chester Creek ranged from 0-24 within the last two
decades (Whitman 2002). No official counts of salmon escapement into Chester Creek
have been accomplished. Chester Creek flows westward from the Chugach Mountains
approximately 17 km running through the heart of Anchorage before emptying into
Westchester Lagoon (Figure 1). The creek flowed directly into the Knik Arm portion of
Cook Inlet prior to 1970 (Figure 2). In 1970 the Municipality of Anchorage (MOA)
established Westchester Lagoon by building a dam to create a recreational area. The
lagoon sits in the south western portion of Anchorage, Alaska. Within the constraints of
the dam a concrete weir (Figure 3) was built in an effort to support anadromous fish
passage into the Chester Creek drainage. The weir also provided flow control and a
sediment deposit area near the mouth of the stream (Davis & Muhlberg 2001).
Salmon populations were severely impacted due to the weir restricting
anadromous fish passage into Chester Creek (Davis & Muhlberg, 2001). The design of
the weir forced salmon to travel through two culverts 24 meters long. In order for the
salmon to enter these culverts, the tide levels need to be within -1.25 to 1.19 meters of
mean sea level (MSL) since fish avoid diving near stream entrances (USACE 2004) and
have limited ability to leap into the culverts. These culverts are routed underground
with a height difference of 12.50 m between the Knik Arm and the weir. Water leaving
the lagoon falls approximately 3.66 meters and enters the culverts and is routed to the
inlet. Coho’s known maximum vertical leap is 2 meters (Sandercock 1991). This further
limited successful escapement of coho and took away the mixing zone

Figure 1 – Chester Creek flowing from Chugach Mountains through Anchorage, AK.
Courtesy of Dustin Merrigan

Figure 2 – Chester Creek entering Cook Inlet in 1950. Westchester Lagoon currently sits to the
right of the trestle. Courtesy of Army Corps of Engineers.
that would allow juvenile salmon returning to Cook Inlet to acclimate to saltwater. It was
found that in order for the coho to successfully leap into the lagoon the incoming high
tide had to reach approximately 8.53 meters.
Entrance into the weir from Westchester Lagoon was also hindered due to
grating that was installed and blocked debris from falling into the culverts (Figure 4).
The bars on the grating were spaced 15 to 20 cm apart. The debris consisted of a thick
mat of vegetation growing at the water line and woody material from upstream. The
size of the woody debris ranged from small sticks to logs 5.12 meters in length and 15-
20 cm in diameter. The buildup of debris over time also hindered successful
escapement of coho. The stream corridor has been impacted by urbanization reducing
water quality, stream channel alterations, erosion, and poor riparian zones. Healthy
riparian zones are important to shade the stream and maintain water temperatures
sufficient for salmon habitat (Madej, 2006). Riparian zones also provide allochthonous
production required for salmon diets (Hetrick et al. 1998).

In 1999, the Municipality of Anchorage (MOA) and U.S. Army Corps of Engineers
proposed a two phase process to restore the ecosystem of Chester Creek, Anchorage,
Alaska (USACE, 2004). The first phase of this project removed the old weir and
culverts. The new weir will be built in the northwest corner of the lagoon. The weir will
allow anadromous fish to migrate into spawning habitat with greater ease. It will also
allow juvenile salmon returning to sea to acclimate to saline marine conditions, thereby
increasing survivorship. Phase II consists of creating a new channel from the weir to
Cook Inlet. This new channel will meander for 473.4 meters with an elevation change of
6.5 meters with an average flow of 60 ft3
/s and a depth of .8 ft. (Inter-Fluve Inc. 2008).
Figure 3 – Weir at Westchester Lagoon (April 2008). Removed in October 2008.
Boulders, woody debris, and gravel will also be put in place to help with channel stability
and provide habitat for aquatic species. The work is being accomplished in collaboration
with MOA, Anchorage Waterways Council, U.S. Army Corps of Engineers, U.S. Fish
and Wildlife Service, National Oceanic and Atmospheric Administration (Habitat
Restoration Center), and other contractors and volunteers.
Before the Chester Creek restoration project begins, the United States Fish and
Wildlife Service (USFWS) requires baseline conditions to be known. To fulfill part of this

requirement, salmon counts were conducted during July and August of 2008.
Successful escapements were recorded when the salmon jumped and swam away
Figure 4 – Right side of weir entrance showing grating and debris build up.
into Westchester Lagoon. The use of video cameras to accurately count salmon
escapements was also examined. Video surveys have been shown to detect 87-98% of
the fish (Otis and Dickinson, 2002). It is hypothesized that there will be no significant
difference between visual and video counts allowing video counting a reliable source to
estimate escapement numbers entering Chester Creek.
Methods
In order to complete salmon counts, two quantitative methods were employed: 1)
visual counts made by five trained observers and 2) video camera recordings.
Individuals were present at the weir two hours prior to high tide through two hours after

high tide in order to setup camera equipment and observe incoming salmon.
Successful escapement times and numbers were recorded in a log book.
Two cameras (Toshiba Digital Color Camera; model # IK-64WDA) were installed
with a 5-40 optical power range. Cameras were placed in a weather proof case
manufactured by Pelco, Clovis, CA (model # EH3512). Tiffin circular polarizers mounted
on the lenses reduced glare. Images were recorded on a Sanyo Digital Video Recorder
(model # DSR-3506). Two sets of halogen lights were used during dark hours providing
sufficient lighting requirements. Power supply for the video recording system and
lighting was provided by a Honda generator (model EU2000i). The generator AC output
was 120 volts, 2000 watts, and 16.7 amps. The digital recorder and generator were
housed in an old weather station box (Figure 5). A tub was placed under the generator
in the event of an oil or gas spill.
Figure 5 – Housing for generator and digital recorder.
One camera was placed on the side of the west wing wall using a metal pipe
rock bolted to the wall (Figure 6). The camera faced in a downstream direction focusing

on the left portion of the weir 3.4 meters away from the weir waterfall ledge. The
second camera was installed on the top center of the weir pointing downward focusing
on the right side of the weir 3 meters above the ledge. It was attached using u-bolts to
grating placed on top. Both sets of lights were placed on top of the weir pointed
downward towards the water at angle to reduce glare.
Total number of coho escapements visually counted were summed and
compared to totals that were counted by watching video. A chi-square test was run to
determine if a significant difference occurred between visual and video counts. A linear
regression was used to determine if video counts can significantly account for the visual
counts. Cooks distance was used to determine if outliers had an undue effect on the
regression. All analyses were accomplished using SPSS for Windows v. 15. Alpha level
was set to p <.05.
Figure 6 – Camera mounted on west wing wall of weir.

Results
Observations occurred during a 42 day period starting on July 22 and ending on
September 2nd
. Several fish were first seen on July 22nd
but no successful
escapements. The first successful escapements occurred on July 30th
. The highest
high tide during that period was 9.97 meters and the lowest was 7.25 meters (NOAA
2008). There were 497 coho visually counted that successfully escaped and 394 were
counted using video. This resulted in no significant difference between visual and video
counts ( 2
36 = 33.30, p = .05). Assumptions for the regression model were not met with
two outliers outside of ± 2 standard deviations. However, Cooks distance was less than
1 indicating the outliers had no undue effect on the model. The regression model
showed a significant improvement over the mean (F1, 35 = 1333.41, p = .00). The
regression slope (t35 = 36.52, p = .00, Table 1) is significantly different than zero
indicating that the video counts (observed) can predict the visual counts (expected).
Video counts accounted for 98% of the visual counts (R2
= .98, Table 2).
Table 1- The slope is significantly greater than zero indicating that video counts can predict the
visual counts would be (t35 = 36.52, p = .00).
Coefficientsa
.189 .621 .305 .762 -1.072 1.450
1.244 .034 .987 36.516 .000 1.175 1.313
(Constant)
Video counts
Model
1
B Std. Error
Unstandardized
Coefficients
Beta
Standardized
Coefficients
t Sig. Lower Bound Upper Bound
95% Confidence Interval for B
Dependent Variable: Visual countsa.
Table 2 – R2
= .98 indicating that 98% of the visual counts are accounted for using video counts.
Model Summaryb
.987a .975 .974 2.977 .975 1333.406 1 34 .000
Model
1
R RSquare
Adjusted
RSquare
Std. Error of
the Estimate
RSquare
Change F Change df1 df2 Sig. F Change
Change Statistics
Predictors: (Constant), Video countsa.
Dependent Variable: Visual countsb.

Discussion
The abundance of salmon that successfully escaped was much higher than
originally thought. Weir design, broken fish ladder, and previous anecdotal
observations had led us to expect fewer than 100 coho successfully escaping into the
lagoon. If the debris wasn’t cleaned out prior to the start of this project, it is believed that
there would have been fewer successful escapements. It is unknown if MOA had a
schedule to keep the debris cleaned out in the past. If the debris wasn’t cleaned out,
especially prior to the salmon run, it may have contributed to past low escapement
estimates.
The method of using a video camera is a good predictor of the expected or visual
counts. This is concluded since 98% of the variability in visual counts could be
explained by video counts (Table 2, Figure 7). The other 2% are explained by chance.
These results support the hypothesis that there is no significant difference between
visual and video counts. Figure 8 shows a daily comparison of visual and video
escapement numbers. An 80% success rate using video counts was calculated from
Figure 8.
Although the difference between the total visual and video counts seems high
(103), the chi-square determined there was no significant difference between the total
daily observations of the two modes of counting. The regression model then looks at
differences on each day of observations. The variability in the differences is small
allowing us to obtain a linear regression value.
Reliable power supply was a key aspect using video technology. The generator
was chosen for this project due to time and cost constraints as well as aesthetics in the
highly visited recreational area. If this type of project were to continue, a solar powered
system would be optimal for supplying power. Since Westchester Lagoon is a high
visibility area we felt that the solar panel would be vandalized rendering the video
system inoperable. Additionally, since this was only occurring during one season,

establishing a solar powered system didn’t seem economically feasible. Technical
issues with the generator did cause two days of no video. Human error also caused
another day of missing video. Noise levels of the weir drowned out the generator noise
(~ 59 db).
Figure 7 – 98% of the visual counts are explained using video counts. The equation, Visual
counts = .19 + 1.24*Video count can be used to make the prediction.
An ideal situation to count salmon escapement would have been at a location
further upstream near other culverts. This would have included a v-shaped weir to allow
salmon to pass through upstream avoiding the possibility of double counting. However,
MOA declined our request to put scientific equipment into the stream. It was possible to
have double counted salmon escaping at the weir since there was no equipment

installed to prevent any salmon swimming back through the weir. However, during
observation times, no salmon were noted returning back through the weir.
The new weir will be constructed with a built in capability of housing a camera
system and will be wired to supply an electrical source for future fish monitoring (HDR
2008). Other improvements for future monitoring consist of a support rack to house
cameras, recorders, lights, and framing to hold white-board that will increase the
contrast of passing fish. The option of using a contrasting background was examined
for this project but do to safety and a limitation set by MOA its use was not granted.
Figure 8 – Daily comparison of visual and video counts showing an 80% success rate
using video counts.

After the construction of the new weir, coho and other fish will be able to migrate
into the Chester Creek drainage with greater ease. Juvenile survivability should
increase with a mixing zone present and improved habitat around the mouth of Chester
Creek. Similar small streams in Oregon have had an average of 40% annual increase
(House 1996). Katz (2007) suggests that for a restoration project to be effective, winter
and spring habitat areas also need to be restored as well as a more natural migration of
fish. Once the restoration of the mouth of Chester Creek is complete, efforts should be
exerted toward habitat improvements near the headwaters.
Overall the goals of this project were achieved; however, several improvements
could have made this project a greater success. Measurements of the tide flux within
the weir would be useful in determining the exact water levels required in order for the
coho to successfully escape. Water temperature and turbidity levels in Cook Inlet may
have shown a correlation in these parameters and the run timing. Recent studies also
suggest examining air pressure, stream discharge, and precipitation amounts in
determining run timing of salmon (Keefer et al. 2004).
Acknowledgements
I would like to thank all the people that had assisted in making this project a
success. Without these people and organizations I would not have been to complete
this project. The individuals include Rusty Myers, committee chair & advisor, Theresa
Tanner, USFWS, the observers; Dustin Merrigan, Shawna Nieraeth, Jenn Gonzalez,
and James Whitlock, and statistics professor Dr. Leslie Cornick. The organizations that
directly helped me include USFWS, Anchorage Waterways Council, US Army Corps of
Engineers, HDR Inc., and NOAAs’ Habitat Restoration Center. Again thank you all for
your tremendous help throughout this project.

References
Anderson, J. L. and N. J. Hetrick (2004). “Carrying Capacity of Habitats Used
Seasonally by Coho Salmon in the Kametolook River, 2002-2003, Alaska
Peninsula National Wildlife Refuge”. Alaska Fisheries Technical Report
Number 73.
Anderson, J. L., N. J. Hetrick, et al. (2006). “Design and Performance of a Digital Video
Monitoring Station Incorporated in a V-Shaped Resistance Board Weir”. USFWS.
King Salmon, Alaska.
Bureau of Reclamation (2007). Coho Salmon Production Potential in the Cle Elum River
Basin, Storage Dam Fish Passage Study, Yakima Project, Washington,
Technical Series No. PN-YDFP-007, Boise, Idaho.
Christopher C. Caudill, W. R. D., Matthew L. Keefer, Charles T. Boggs,, B. J. B. Michael
A. Jepson, Richard W. Zabel, Theodore C. Bjornn, and, et al. (2007). "Slow dam
passage in adult Columbia River salmonids associated with unsuccessful
migration: delayed negative effects of passage obstacles or condition-dependent
mortality?" Canadian Journal of Fisheries & Aquatic Sciences. 64: 979-995.
Davis, J. C. and G. A. Muhlberg (2001). “Chester Creek Stream Condition Evaluation”.
Technical Report No. 01-7: 47.
Engineers, U. S. A. C. (2004). “Chester Creek Aquatic Ecosystem Restoration
Draft Report and Environmental Assessment Anchorage, Alaska”. D. o. t. Army.
Estensen, J. L. and M. Cartusciello (2005). “Salmon Enumeration in the Nome River
Using Video Technology”. Fishery Data Series. Anchorage, Alaska Department
of Fish and Game.

Hart, PJB, Reynolds, JD. (2002). “Handbook of Fish Biology and Fisheries”. Volume 1.
Oxford, UK: Blackwell Science Ltd. 413 p.
Hatch, D. R., J. K. Fryer, et al. (1998). "A Computerized Editing System for Video
Monitoring of Fish Passage." North American Journal of Fisheries Management.
18(3): 694-699.
Hatch, D. R., M. Schwartzberg, et al. (1994). "Estimation of Pacific Salmon Escapement
with a Time-Lapse Video Recording Technique." North American Journal of
Fisheries Management. 14(3): 626-635.
HDR, Alaska Inc. (2008). Technical Memorandum.
Hetrick, N. J., K. M. Simms, et al. (2004). “Feasibility of using Video Technology to
Estimate Salmon Escapement in the Ongivinuk River, a Clear-Water Tributary of
the Togiak River”. U. S. F. W. Service. King Salmon, Alaska.
House, R. (1996). "An evaluation of stream restoration structures in a coastal Oregon
stream, 1981-1993”. North American Journal of Fisheries Management. 16(2):
272-281.
Irvine, J. R., B. R. Ward, et al. (1991). "Evaluation of a Method to Count and Measure
Live Salmonids in the Field with a Video Camera and Computer”. North American
Journal of Fisheries Management. 11: 20-26.
Johnson, DH, Schrier, BM, O’Neal, J.S., Knutzen, J.A., Augerot, X., Pearsons, T.N.
(2007). “Salmonid Field Protocols Handbook: Technique for Assessing Status
and Trends in Salmon and Trout Populations”. American Fisheries Society.
Bethesda, Maryland.

Katz, S. L., K. Barnas, et al. (2007). "Freshwater Habitat Restoration Actions in the
Pacific Northwest: A Decade's Investment in Habitat Improvement”. Restoration
Ecology. 15(3): 494-505.
Keefer, M.L., Peery, C.A., Jepson, M.A., Tolotti, K.R., Bjornn, T.C. (2004). “Stock-
Specific Migration Timing of Adult Spring-Summer Chinook Salmon in the
Columbia River”. North American Journal of Fisheries Management. 24:1145-
1162.
Madej, M. A., C. Currens, et al. (2006). "Assessing possible thermal rearing restrictions
for juvenile coho salmon (Oncorhynchus kisutch) through thermal infrared
imaging and in-stream monitoring, Redwood Creek, California". Canadian
Journal of Fisheries & Aquatic Sciences. 63(6): 1384-1396.
Milner, A.M., Oswood, M.W. (2000). “Urbanization gradients in streams of Anchorage,
Alaska: a comparison of multivariate and multimetric approaches to
classification”. Hydrobiologia. 422/423, 209-223.
National Oceanic and Atmospheric Administration. Revised September 26, 2008.
Retrieved on September 27, 2008. http://tidesandcurrents.noaa.gov/index.shtml.
Otis, E. and M. Dickson (2002). “Improved Salmon Escapement Enumeration Using
Remote Video and Time-Lapse Recording Technology”. Restoration Project
00366.
Purcell, A. H., C. Friedrich, et al. (2002). "An Assessment of a Small Urban Stream
Restoration Project in Northern California”. Restoration Ecology. 10(4): 685-694.
Richter, A. and S. A. Kolmes (2005). "Maximum Temperature Limits for Chinook, Coho,
and Chum Salmon, and Steelhead Trout in the Pacific Northwest”. Reviews in
Fisheries Science. 13(1): 23-49.

Salkind, N.J.(2004). “Statistics for people who (think they) hate statistics”. 2nd
ed. Sage
Publications. Thousand Oaks, California.
Sutherland, WJ. (2006). “Ecological Census Techniques”. Cambridge University Press.
Cambridge, UK: 432 p.
Sandercock, F.K. (1991). “Life History of Coho Salmon (Oncorhynchus kisutch)”.
Pacific Salmon Life Histories. Pg. 397-445 In C. Groot and L. Margolis (Eds.)
University of British Columbia Press, Vancouver, B. C.
Thompson, W. L. (2003). "Hankin and Reeves’ Approach to Estimating Fish
Abundance in Small Streams: Limitations and Alternatives Transactions”.
American Fisheries Society. 132(1): 69-75.
Whitman, M. S. (2002). “A Protocol For Assessing The Impacts Of Urbanization On
Coho Salmon with Application to Chester Creek, Anchorage, Alaska”. Fairbanks,
University of Alaska Fairbanks. Master of Science: 79.

Coho_Enumeration_Final

More Related Content

What's hot (7)

Similar to Coho_Enumeration_Final (20)

Coho_Enumeration_Final