Monitoring of Grays River Fall Chinook Salmon using an Instream Weir, 2008-2010

The Lower Columbia River (LCR) Chinook Evolutionarily Significant Unit (ESU) was listed as threatened under the Endangered Species Act (ESA) on March 24, 1999. Since the initial listing, the threatened status has twice been reaffirmed, once in 2005 and again in 2014. The Grays River population of Chinook salmon (Oncorhynchus tshawytscha) was designated as a "primary" population within the ESU by the Lower Columbia River Fish Recovery Board (LCFRB) in 2004. This designation was changed to a "contributing" population in 2010, and as a result, the viability goal was reduced from 1,400 to 1,000 adult natural-origin spawners.

The Hatchery Scientific Review Group (HSRG) has stated that one of the factors limiting naturally spawning populations is interaction with hatchery-origin fish on the spawning grounds. The HSRG recommended less than 5% hatchery-origin spawners for primary populations and less than 10% for contributing populations (HSRG 2009). Historically, the use of peak count expansion (PCE) and coded wire tag (CWT) expansion have provided a means to estimate spawner abundance and stock composition, respectively. Between 1995 and 2006, PCE abundance estimates ranged from 14-745 Chinook salmon with out-of-basin stray rates ranging from 0 to 41.6% (Jenkins 2006).

In an effort to improve escapement monitoring and to promote recovery of the Grays River fall Chinook salmon population through removal of non-local Chinook salmon (hatchery strays), a resistance board weir was installed at river kilometer 17.2 in the Grays River in September 2008. This was the first resistance board weir used for fall Chinook salmon management in the LCR. The weir was operated through October 2008 before it was removed. It was reinstalled and operated from mid-August through the last week of October in 2009 and from mid-August through mid-October in 2010.

In these three years of operation, a total of 328 Chinook salmon, 1,621 coho salmon, 8 chum salmon, 1 pink salmon, and 3 steelhead were captured. Biological data was collected from all salmonids trapped. The proportion of marked Chinook salmon (defined as having an adipose and/or left ventral fin removed) captured at the weir was 78.7% in 2008, 79.8% in 2009, and 94.9% in 2010. Select Area Bright (SAB) stock fall Chinook represented 93.7% of the marked Chinook salmon trapped. These fish were identified as SAB stock based on the presence of a left ventral (LV) fin clip. Ten CWTs were recovered from Chinook salmon at the weir (three in 2008, five in 2009, and two in 2010) and all were decoded as SAB stock fall Chinook salmon released from the Oregon's South Fork Klaskanine River.

Adult Chinook salmon (defined as a fork length 60 cm and larger) spawner estimates were derived using a Bayesian framework using three independent methods: binomial Lincoln-Petersen, area-under-the-curve (AUC) based on live counts of fish identified as spawners, and redd expansion. Low sample sizes resulted in high CVs when Lincoln-Petersen estimates were generated (56.1% and 55.5% for 2008 and 2009, respectively) and did not allow for a Lincoln-Petersen estimate to be generated in 2010. AUC abundance estimates relied on estimates of apparent residence time (ART) derived from other populations as we did not have adequate mark recapture estimates and/or complete spatial and temporal coverage to develop basin-specific estimates of ART (Parken et al. 2003). We believe there could be bias in applying ART estimates from populations without weirs to populations with weirs as ART may be different due to weir effects. Therefore, we chose to report spawner abundance using redd expansion methods for 2008-2010. Apparent females per redd was derived from other LCR fall Chinook salmon population monitoring programs and was applied to 2008-2010 Grays River Chinook salmon redd and sex ratio data to develop a spawner abundance estimate. The mean value and 95% credible interval (CI) of the posterior distribution for adult Chinook salmon spawner abundance using redd expansion was 95 (95% CI 76 - 123) in 2008, 555 (95% CI 417 - 756) in 2009, and 159 (95% CI 114 - 233) in 2010.

The proportion of marked Chinook salmon spawners, or pMark, (based on a visual cue of having the adipose and/or left ventral fin removed) based on carcass recoveries was much higher than recommended by the HSRG each year. The mean value and 95% credible intervals of the posterior distribution for pMark for adult Chinook salmon was 63.9% (95% CI 47.7 - 78.5%) in 2008, 61.5% (95% CI 49.6 - 73.1%) in 2009, and 54.2% (95% CI 34.6 - 73.4%) in 2010.

Applying the estimated proportion of unmarked carcasses to estimates of spawner abundance yields an estimate of unmarked Chinook salmon spawners. Unmarked Chinook salmon are likely a mix of Tules, naturalized SABs, and their hybrids (Roegner et al. 2010). The mean value and 95% credible intervals of the posterior distribution for unmarked adult Chinook salmon was 33 (95% CI 19 - 52) in 2008, 210 (95% CI 132 - 315) in 2009, and 70 (95% CI 36 - 118) in 2010.

To evaluate weir effectiveness, we used our redd-based Chinook salmon spawner estimates and added sport harvest above the weir based on catch record card data and pre-spawn mortalities based on carcass recovery data to estimate the total number of Chinook salmon that passed the weir site. Weir efficiency was estimated as the proportion of upstream Chinook salmon population captured at the weir. The mean value and 95% credible intervals of the posterior distribution for weir efficiency for adult Chinook salmon was 44.1% (95% CI 36.4 - 52.5%) in 2008, 23.7% (95% CI 17.9 - 29.4%) in 2009, and 15.7% (95% CI 10.1 - 21.6%) in 2010.

Our analysis suggests that weir efficiencies would need to be in excess of 90% to achieve the proportion of hatchery-origin spawners goal of 10%. There are three areas of emphasis that could improve weir efficiencies in the future: 1) improving the substrate rail design; 2) adding additional floatation to the resistance panels; and 3) adding bulkheads where the resistance board weir transitions to the river bank. While these three improvements should improve weir efficiency, it should be noted that the Grays River is a very dynamic system. If weir operations are to continue into the foreseeable future, land acquisition for a suitable weir site, preferably at the upper end of tidal influence, and construction of a concrete sill should be considered to improve weir operations and efficiencies in order to meet HSRG standards.