Goal: The goal of this project is to monitor the summer and winter steelhead populations in the Wind River to determine if restoration actions increase the abundance and productivity of these populations and subpopulations, to provide an index of ocean survival for wild summer steelhead in the Gorge Province, and provide population specific data to National Oceanic and Atmospheric Adminstration (NOAA) Fisheries so they can determine when ESA listed Wind River steelhead are a viable salmon population and can be delisted.  Scientists are also concerned with interactions between native wild steelhead and introduced hatchery spring chinook populations, therefore adult and juvenile outmigrant spring chinook are also monitored.

Background: The Wind River had one of the largest summer steelhead populations in the Lower Columbia Area.  Wild summer steelhead populations began declining after surveys were initiated in the mid-1980's.  The Wind River Restoration Team was formed in the early 1990's to determine the cause of the decline and help rebuild this wild summer steelhead population.   The group took a two pronged approach: one increase adult and juvenile monitoring to determine if the cause of the decline was occurring in the subbasin or outside of the subbasin, and to restore habitat to increase the steelhead population.  In the late 1990's BPA began funding this cooperative effort between WDFW, UCD, USFS, USGS, and USFWS.

Location: The Wind River subbasin is located in the Columbia Gorge Province and this river enters the Columbia River approximately 15 miles upstream of Bonneville Dam.  

BPA Furnished Property or Services:  none.

As part of the Wind River partnership both WDFW and USGS agreed to participate in objective 1.  The WDFW tasks are covered below:

Objective 1.  Monitor the physical habitat conditions and natural production of juvenile, smolt, and adult steelhead in the Wind River subbasin (WDFW, USGS).

Task 1.1: Conduct sampling to derive annual estimates of steelhead smolt production in key areas of the subbasin (WDFW)

Methods
We propose to develop annual estimates of smolt production from the Wind River basin.  Proposed trap sites are located to estimate total basin production (mouth of Wind River) and production from key watersheds (Trout Creek, Panther Creek, and Upper Wind River).  Rotary screw traps will be fished from March 15 to June 15, which coincides with Wind River smolt migration (Rawding, 1997).  Traps will be checked daily and fish will be enumerated.  After fish are anesthetized, we will obtain fork lengths and weights from all fish and scale samples will be obtained from a weekly subsample.  Ages from these fish will be used in conjunction with fork length frequencies to determined age composition for the smolt outmigration.

Smolt estimates will be determined by the trap efficiency method of releasing marked fish upstream of each trap (Thedinga et al. 1994).  All fish will be tattooed and marks will be rotated weekly to determine changes in trap efficiency.  In season trap efficiency will be determined by Bailey's (1951) modification of the Petersen estimator.  Short-term survival and mark retention will be measured and used to adjust trap efficiency (Murphy, 1997).  Confidence intervals will be determined using the bootstrap method (Efron and Tibshirani, 1986). Juvenile spring chinook outmigrants will be monitored using the same methodology.  


Captured wild steelhead smolts will be coded-wire-tagged to help determine mark retention.  In addition, adults sampled at the Wind River and Trout Creek traps will be scanned for wire tags.  Using a Bailey's (1951) modification to the Petersen estimator we will be able to determine the smolt production and smolt to adult survival through back-calculation (Seiler et al. 1995).


Task 1.2: Conduct sampling to derive annual estimates of adult escapement of steelhead and spring  chinook to the subbasin. (WDFW)

Methods
Currently, expanded redd surveys are used to develop adult escapement estimates based on assumptions on differences in winter and summer steelhead distribution and differences in hatchery and wild spawning time (WDFW, 1997).  Knudsen (1997) rated these methods as fair to poor for determining escapement estimates.  Therefore, we propose to trap adults in the Shipherd falls and Trout Creek fish ladder to improve the accuracy of adult escapement estimates.

We propose to operate the adult trap year round.  Adult steelhead will be floy tagged and released upstream.  Shipherd Falls is believed to be a total barrier to winter steelhead, and the winter escapement estimate would equal the trap count.  Since summer steelhead can successfully jump the falls, snorkel surveys and the Trout Creek trap will be used to determine the total number of tagged and untagged fish.  Adult run size will be estimated using Bailey's (1951) modification to the Peterson estimator.  Floy tag loss will be estimated through the use of double marking.  In addition to receiving a floy tag, a small hole will be placed in the caudal fin with a paper punch.  As fish are recaptured upstream at the Trout Creek trap floy tag loss will be estimated.  In addition, the caudal fin tissue from the paper punch hole will be archived for future genetic analysis (DNA).  Trout Creek data provided from operation of the trap by WDFW and the USFS will be used for escapement into this subbasin.

There has been a growing concern regarding the potential interaction between hatchery spring chinook salmon and ESA list steelhead in the Wind River.  Currently, the estimate of naturally spawning spring chinook salmon is derived from the peak count of live and dead chinook salmon between Paradise Creek and Beaver Camp.  Implicit are the assumptions that all live and dead salmon are visible, all spawners are present at the peak, no fish have been removed by predators or scavengers, and all fish spawn within the index area.   To account for the violation of these assumptions WDWF developed a correction factor of 1.5.  The total escapement is estimated by multiplying the peak count by the correction factor.  This correction factor was developed based on the professional opinion of biologists familiar with chinook spawning ground escapement estimates and the spring chinook surveys in the Wind River.  To better estimate the spring chinook escapement WDFW is initiating a mark-recapture experiment on salmon carcasses between Paradise Creek and Beaver Camp.  Estimates will be calculated using the Jolly-Seber method.  Carcasses are bio-sampled on each survey and sex, length, and fin marks are recorded for each recovered fish. A missing adipose fin identifies the fish as a CWT fish and snouts are collected from these fish to determine the stock origin.  We will compare the estimates from the carcass tagging with the estimates derived from redd surveys, Area-Under-the-Curve (AUC), and the current peak count method.  The redd surveys are used to estimate escapement using the cumulative redd method, which is the sum of all new redds counted during the spawning season.  The AUC is the sum of the area from the first count to the last count and is estimated by trapezoidal approximation expressed as fish * days.  This escapement estimate is calculated by dividing the AUC by the mean residence time.

References

Bailey, N. T. J.  1951.  On estimating the size of mobile populations from recapture data.  Biometrika 38:293-306.

Efron, B., and R. Tibshirani.  1986.  Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy.  Statistical Science 1:54 77.

Knudsen, E.E.  1997.  Managing Pacific Salmon escapements: the gaps between theory and reality.  U.S. Geological Survey - Biological Resources Division.  Anchorage, AK.  In press.

Murphy, M.L., J. F. Thedinga, and J. J. Pella.  1996.  Bootstrap confidence intervals for trap efficiency estimates of migrating fish. U.S. Department of Commerce, National Marine Fisheries Service, Alaska Fisheries Science Center.  In press.

Rawding, D.  1997.  Wind River smolt monitoring report.  Washington Department of Fish and Wildlife.  Vancouver, WA.

Seiler, D., P Hanratty, S. Neuhauser, P. Topping, M. Ackley, and L. Kishimoto. 1995.  Wild Salmon Production and Survival Evaluation - Annual Performance Report, October 1993 - September 1994.  Washington Department of Fish and Wildlife Contract No. F-112-R-2.

Thedinga, J. F., S. W. Johnson, K V. Koski, J. M. Lorenz, and M. L. Murphy.  1993.  Potential effects of flooding from Russell Fiord on salmonids and habitat in the Situk River, Alaska.  U.S. Department of Commerce, National Marine Fisheries Service, Alaska Fisheries Science Center Processed Report 93 01.

Washington Department of Fish and Wildlife. 1997.  Preliminary stock status for steelhead in the Lower Columbia River, Washington.  Olympia, WA.