Fish and habitat assessment in Rock Creek, Klickitat County, southeastern Washington, 2018

Abstract

Executive Summary Native steelhead (anadromous form of rainbow trout [ Oncorhynchus mykiss ]) and bridgelip sucker ( Catostomus columbianus ) were historically used by the Kah-miltpah (Rock Creek) Band for sustenance, trade, and traditional practices in Rock Creek, a tributary to the Columbia River in southeastern Washington State. Rock Creek flows south to the Columbia River at river kilometer (rkm) 368 and is an intermittent stream of great significance to the Yakama Nation and to the Kah-miltpah Band in particular. Concern over declines in the abundance of these fish in Rock Creek prompted a research and monitoring program to better understand habitat conditions, population status, and limiting factors. In addition to steelhead and bridgelip sucker, coho salmon ( Oncorhynchus kisutch ) and resident rainbow trout are also present and monitored. Rainbow trout and steelhead will be collectively referred to as O. mykiss . Streamflow is a limiting habitat factor in this system, but steelhead and coho salmon still successfully return to spawn, rear, outmigrate, and survive over summer in many of the isolated pools that provide important refuge for juvenile rearing. We completed a habitat survey during autumn 2018 to assess the perennial pools during low-flow conditions. In Rock Creek, the overall percentage of habitat recorded as dry was 41, non-pool wet was 42, and pool was 17. The number of pools ( n =93) recorded was less than during previous years’ survey efforts (2015–17). The percentage of non-pool wet habitat was generally higher in 2018 than in previous years. This is a likely result of habitat reaches, which in the past, were considered pools but have become shallower and smaller and are now categorized as non-pool wet habitat. However, the fewer habitat reaches categorized as pools in 2018 now have an average length, area, and depth that are generally greater than in past years. In Walaluuks Creek, the percentage of habitat recorded as dry was 53, non-pool wet was 40, and pool was 7. The percentage of pool habitat was the lowest of all years surveyed since 2015. Fish sampling occurred during autumn after habitat surveys were completed from October 1 to November 9. Fish species distribution, relative abundance, length-frequency distribution, and pool fish density were determined using backpack electrofishing in stratified, systematically selected pools. During fish sampling, 855 O. mykiss were handled and 662 were tagged with a passive integrated transponder (PIT) tag, and 718 coho salmon were handled and 567 were PIT tagged. A total of 536 bridgelip suckers and largescale suckers ( Catostomus macrocheilus [ n =6]) were handled and 294 were PIT tagged. In Rock Creek, pool abundance estimates were calculated for six pools for both O. mykiss age classes (age 0 and age 1 or older [age 1+]) and one additional pool for age 1+. For pools where age-0 O. mykiss were present, the average pool population abundance was 0.144 ( n =6; range: 0.052–0.208) fish per square meter. For age-1+ O. mykiss , the average pool population abundance was 0.045 ( n =7; range: 0.002–0.179) fish per square meter. For age-0 O. mykiss in Walaluuks Creek, the average pool abundance was 0.207 fish per square meter ( n =7; range: 0.038–0.416), and for age-1+ fish, the average pool abundance was 0.382 fish per square meter ( n =6; range: 0.009–0.761). In Rock Creek, coho salmon were more abundant than O. mykiss in pools except for three pools upstream from rkm 20. The average pool abundance for coho salmon was 0.256 fish per square meter ( n =8; range: 0.019–0.756) in Rock Creek pools. In Walaluuks Creek, coho salmon were captured in four pools and were not captured in the upstream pools sampled. The average pool abundance for coho salmon in the four lower pools was 0.488 fish per square meter ( n =4; range: 0.417–0.548). Bridgelip suckers were captured in all pools in Rock Creek except the pool sampled at rkm 21.8. The average pool abundance for bridgelip suckers was 0.552 fish per square meter ( n =7; range: 0.015–1.554) in Rock Creek. Bridgelip suckers were captured in three downstream pools in Walaluuks Creek, and the average abundance was 0.044 fish per square meter ( n =3; range: 0.024–0.085). Overwinter and reach survival probabilities were estimated for O. mykiss and coho salmon using a Cormack-Jolly-Seber modeling approach. The best fit survival model for the O. mykiss and coho salmon was a reach only model. The upstream reach includes overwinter survival probability because fish are tagged and released in autumn and primarily migrate the following spring. During 2018, coho salmon (0.568, standard error [SE]=0.027) had a significantly higher probability of overwinter survival than O. mykiss (0.276, SE=0.019). The reach survival probability was higher for O. mykiss than coho salmon in the downstream migratory reaches. Survival was not modeled for bridgelip suckers. For bridgelip suckers, 147 were detected of 294, that were PIT tagged and released in the Rock Creek subbasin (50.0 percent). Information provided in this report increases our understanding of the status and trends of these populations. It further documents how intermittent streams can support salmonid populations. It also provides insight into potential management and restoration actions that could be beneficial and timing and allocation of resources. Ongoing monitoring work of this population will inform progress towards Rock Creek species recovery goals and contribution to recovery goals for the steelhead Middle Columbia River Distinct Population Segment.