Estimating Adult Sockeye Salmon Escapement into Akalura Lake … · 2018. 4. 4. · Author: Gareth K. VanHatten, 1390 Buskin River Road, Kodiak, Alaska, [email protected], 907-487-0230

Refuge Report 2018.1

Estimating Adult Sockeye Salmon Escapement into Akalura Lake using a remote video and time lapse system, Kodiak

National Wildlife Refuge, Alaska

2016 Progress Report

G. Kevin VanHatten

Kodiak National Wildlife Refuge Kodiak, Alaska January 2018

U.S. Fish & Wildlife Service

ii

The mission of the National Wildlife Refuge System is to administer a national network of lands and waters for the conservation, management and where appropriate, restoration of the fish, wildlife, and plant resources and their habitats within the United States for the benefit of present and future generations of Americans.

Suggested Citation:

VanHatten, G. K. 2018. Estimating Adult Sockeye Salmon Escapement into Akalura Lake using a remote video and time lapse system, Kodiak National Wildlife Refuge, Alaska: 2016 progress report. Refuge Report 2018.1, U.S. Fish and Wildlife Service, Kodiak National Wildlife Refuge, Kodiak, Alaska.

Keywords:

Sockeye salmon, Oncorhynchus nerka, Kodiak Island, time-lapse cameras, Alitak District, Akalura River

Disclaimer: The use of trade names of commercial products in this report does not constitute endorsement or recommendation for use by the federal government.

iii

Table of Contents ABSTRACT .................................................................................................................................................. 4

INTRODUCTION ........................................................................................................................................ 4

STUDY AREA ............................................................................................................................................. 4

METHODOLOGY ....................................................................................................................................... 4

Time lapse camera system ........................................................................................................................ 5

Modelling salmon escapement (abundance) ............................................................................................. 5

RESULTS AND DISCUSSION ................................................................................................................... 5

Programming video ................................................................................................................................... 5

Sockeye salmon escapement estimate ....................................................................................................... 6

Recommendations ......................................................................................................................................... 6

ACKNOWLEDGEMENTS .......................................................................................................................... 6

REFERENCES ............................................................................................................................................. 7

Refuge report 2018.1 U.S. Fish and Wildlife Service, Kodiak National Wildlife Refuge

Author: Gareth K. VanHatten, 1390 Buskin River Road, Kodiak, Alaska, [email protected], 907-487-0230 4

Estimating Adult Sockeye Salmon Escapement into Akalura Lake using a remote video and time lapse system, Kodiak National Wildlife Refuge, Alaska

2015 Progress Report

Gareth K. VanHatten

ABSTRACT The 2016 field season was the second consecutive year that a double sampling method was use to estimate adult sockeye salmon (Oncorhynchus nerka) returning to Akalura Lake. The project was conducted from July 12 to September 24, in which 30,902 ± 7,990 (95% CI) sockeye were estimated returning. There was very little difference in estimates from 2015, when the project was operated from June 12 to October 21 and 32,802 ± 7,336 (95% CI) sockeye salmon were estimated returning. The report concludes with recommended procedural improvements to increase precision, accuracy, and confidence of estimates.

INTRODUCTION The Kodiak National Wildlife Refuge continued their multi-year project on the Akalura drainage to study the run composition of returning adult sockeye salmon (Oncorhynchus nerka). The Kodiak Refuge and Alaska Department of Fish and Game fishery biologists continue to have conservation concerns about the health of the stock and the potential effect of reduced adult spawners to the system. The double sampling approach is used to estimate the number of adult salmon, run timing, and magnitude of the run from June to October. There hasn’t been enough data collected from the 2 years of study, but the estimation numbers indicate that the sockeye salmon run follows the historical data, with the exception of the run being unimodal instead of bimodal.

STUDY AREA Akalura Lake has a surface area of 4.9 km2, a mean depth of 9.9 m, and a maximum depth of 22 m (Schrof et al. 2000). There are four main creeks that drain into the lake: Mud Creek, Falls Creek, Eagle Creek, and Crooked Creek (Edmundson et al. 1994). The outlet stream (Akalura Creek) is 1.9 km long and drains into a small saltwater lagoon before entering Olga Bay. Akalura Creek is a small, clear water creek and has a single 2.8 km tributary, Humpy Creek (Figure 1).

METHODOLOGY The methods used for estimating adult sockeye salmon followed VanHatten (2017).

mailto:[email protected]



Time lapse camera system The camera/video system was installed and operational on July 12, 2016, which was 30 days later than in 2015. The later installation date did not seem to dampen the escapement estimates as the Akalura sockeye salmon run usually starts towards the end of July.

We used the same time-lapse camera model in 2016 as we did in 2015 to record images of passing salmon. We followed the protocol as stated in VanHatten (2017), for the use of time lapse and infrared lights to capture salmon movement (Figure 2 and 3).

Modelling salmon escapement (abundance) An initial sub goal is initial development an expansion factor that provides a comparative basis for estimation of total salmon return exclusively from salmon counts subsampled by time-lapse camera (Deacy et al. 2016).

In 2016, we encountered a major malfunction in the video system—no data were recovered. The source of the malfunction was not determined. In response, we used the expansion estimator from the results generated in 2015.

RESULTS AND DISCUSSION The results from 2015 study and the time-lapse pictures from 2016 enabled us to estimate 30,902 ± 7,990 (95% CI).The double sampling method used for our study failed in 2016. The video component of the study did not record the intended data for the entire season. The time-lapse camera did meet our expectations as data was collected on 72 of 74 days (97% of study period). Due to the lack of both sets of data collected, video and time-lapse pictures; we resorted to the calibration results from 2015 to estimate daily and seasonal sockeye salmon movement (Figure 4, 5 and 6).

The method provided to be a useful method for estimating sockeye salmon on small systems because the use of time-lapse pictures and video is relatively inexpensive and low labor costs. With weekly visits to the project location to service the system and to check to make sure it was operational, very little maintenance was required for this study. These benefits will be a useful tool for managers to quantify salmon runs in streams otherwise too difficult or expensive to monitor by other means.

Although the time lapse camera system worked as planned, recording 3 photos per minute, 24 hours per day, we encountered some problems during the study. Due to the large amounts of data being recorded we needed to exchange Secure Digital (SD) memory cards every 10 days, otherwise the SD cards would fill up and recording would stop. There was only one block of time where photographs were not recorded throughout the field season. This occurred from September 12, 1:49 am until September 13, 10:20 am. It is unknown why the time-lapse camera failed to record images as the system worked before and after these times. Also, the SD data memory card was not full and seemed to be working fine when brought in from the field.

Programming video It was our intent to obtain a statistically reliable estimate of sockeye salmon escapement estimate by modeling the relationship between time-lapse photographs and video counts (Deacy 2016). In 2016 we




did not accomplish this goal because our DVR hard drive was damaged sometime during the field season. If we had used the collected video data, we could refined our model and improved the confidence of estimates.

Sockeye salmon escapement estimate The results from the 2016 field season as well as the calibration results from 2015 enabled us to use the regular linear regression model to estimate sockeye salmon movement. The results from the plot reinforces the need to collect more video data and in particular, to attempt to capture good video of the big pulses of passing fish. Once this type of data is collected, the polynomial might provide better predictions.

In results from 2016 showed the rate of salmon passage was much higher than the values modeled in 2015. In response, we extrapolated our models beyond the values used to fit them. In general, extrapolation tends to produce poor predictions. The results from the 2016 data show that the extrapolations of the polynomial model produce unrealistic predictions at high run intensities.

Recommendations Weather – The glare of the sun on the water did not hinder visibility across the contrast panels. The one factor that we encountered was high water. Starting on August 7 and lasting till the end of the project, rain events raised the water level to a stage where images were not viewable at night. The timeframe for these periods was roughly from 23:00 hours to 0600 hours in August and from 2200 hours to 0800 hours in mid-September. A way to possibly fix this problem would be to install waterproof lights at the top of, as well as lights that zig zag across the contrast panels (Deacy et al. 2016). Placing lights on or around the contrast panels have been experimented with at other locations within the Karluk Lake drainage.

Mixed Stock species – Starting on September 15, 2016, to the end of the study period pink salmon (O. gorbuscha) were detected passing the panels. Fish differentiation between sockeye and pink salmon could easily be detected during the day. In addition, sex differentiation for pink salmon was easily detected. The pink salmon run was not very high, as was in 2015, and it is unknown if fish were moving during the night.

Equipment specifications – To eliminate black and white video which we experienced in 2015, we programmed the video camera to record in color. Unfortunately, we were not able to see if this change helped in differentiating between species as the DVR hard drive was damaged.

ACKNOWLEDGEMENTS The work conducted and analyses of the data in this report are the results of previous work conducted by Kodiak National Wildlife Refuge volunteers and employees. We would like to thank Will Deacy for his input in the analysis of the data. We would also like to thank those individuals who helped in construction and maintenance for the camera system, Laura Kromrey and William Leacock. We would also like to thank the numerous individuals within the Kodiak National Wildlife Refuge office for their support and dedication to this project; these include Cinda Childers, Lectia Monzon, Tevis Underwood and Ann Marie Larosa.




REFERENCES

Deacy, W.W., W.B. Leacock, L.A. Eby and J.A. Sanford. 2016. A time-lapse photography method for monitoring salmon (Oncorhynchus spp.) passage and abundance in streams. PeerJ 4:e2120; DOI 10.7717/peerj.2120.

Edmundson, J.A., L.E. White, S.G. Honnold, and G.B. Kyle. 1994. Assessment of sockeye salmon production in Akalura Lake. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report NO. 5J84-17.

Schrof, S.T., S.G. Honnold, C.J. Hicks, and J.A. Wadle. 2000. A summary of salmon enhancement, rehabilitation, evaluation, and monitoring efforts conducted in the Kodiak Management Area through 1998. Alaska Department of Fish and Game, Division of Commercial Fisheries, Regional Information Report No. 4K00-57.

VanHatten, G. K. 2017. Estimating Adult Sockeye Salmon Escapement into Akalura Lake using a remote video and time lapse system, Kodiak National Wildlife Refuge, Alaska: 2014 progress report. Refuge Report 2017.1, U.S. Fish and Wildlife Service, Kodiak National Wildlife Refuge, Kodiak, Alaska.



8

Figure 1. Video monitoring project location, Akalura Lake, Kodiak, Alaska 2016.


9

Figure 2. Time lapse and video camera setup, Akalura River, Alaska 2016.


10

Figure 3. Contrast panels for detection of fish movement, Akalura River, Alaska 2016.


11

Figure 4. Daily estimated sockeye salmon count, Akalura River, Alaska 2016.

0

500

1,000

1,500

2,000

2,500

3,000

3,500

12-Jul 19-Jul 26-Jul 2-Aug 9-Aug 16-Aug 23-Aug 30-Aug 6-Sep 13-Sep

Est

imat

ed n

umbe

r of

sock

eye

Date


12

Figure 5. 2015 and 2016 daily sockeye salmon estimates, Akalura River, Alaska.

0

500

1,000

1,500

2,000

2,500

3,000

3,500D

aily

Soc

keye

Sal

mon

Est

imat

e

Date

2015 2016


13

Figure 6. Cumulative sockeye salmon escapement estimates, Akalura River, Alaska.

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000C

umul

ativ

e So

ckey

e Sa

lmon

Est

imat

e

Date

2015

Documents

Estimating Adult Sockeye Salmon Escapement into Akalura Lake … · 2018. 4. 4. · Author: Gareth K. VanHatten, 1390 Buskin River Road, Kodiak, Alaska, [email protected], 907-487-0230