Steelhead at Risk Report: Assessment of Washington’s Steelhead Populations

• Background: Steelhead trout (hereafter steelhead), the anadromous form of Oncorhynchus mykiss, are an important and iconic fish in the Pacific Northwest. Steelhead are the state fish of Washington, they are of cultural and economic value to recreational tribal and non-tribal fishermen and tribal commercial fishermen, and they are an integral part of freshwater ecosystems. Despite this importance, steelhead have declined throughout their native range in North America. In Washington State, anthropogenic activities have pushed steelhead abundances to a fraction of their historical levels. Population declines led to the listing of five of the seven steelhead Distinct Population Segments (DPS) in Washington as ¡§threatened¡¨ under the Endangered Species Act (ESA).
• Purpose and Scope: In 2008, in response to declining abundance, the Washington Department of Fish and Wildlife (WDFW) Fish Program completed a Statewide Steelhead Management Plan (SSMP) that outlined policies, strategies, and actions for managing steelhead (Washington Department of Fish and Wildlife (WDFW) 2008). The SSMP is intended to direct WDFW activities towards maintaining and restoring abundance, productivity, distribution, and diversity of wild steelhead and their habitats to assure long-term health of populations. A key action stipulated in the SSMP was the production at 5-year intervals of a report on the current status of wild steelhead populations at risk of extinction. This current report fulfills that obligation and in it we 1) analyze available biological data for steelhead populations to assess status, 2) identify focal populations at high risk, 3) identify threats to viability at statewide, DPS, and focal population scales, and 4) recommend actions that can be taken to improve status and reduce extinction risks at each scale.
• Methods: We gathered available data on Washington steelhead populations for abundance, harvest, productivity, life history and genetic diversity, and spatial structure¡Xfactors that are indicative of extinction risk. We conducted a quantitative risk assessment to identify highest-risk populations using primarily adult escapement (spawner) abundance data because 1) data for other risk factors were inconsistently available and/or sparse and 2) objective methods for quantifying extinction risk based on other factors were unavailable. We used five metrics to produce total risk ratings. Available data on all factors along with the contents of existing status assessments and management and recovery plans were used to identify key threats and associated recovery actions.
• Key Results: Population Status and Risk
  • Of the 73 steelhead populations that were or are monitored and had sufficient abundance data, a majority, 38 (52%), showed decreasing trends (abundance change < -10%) since 1980, 30 (41%) showed an increasing trend (change >10%), and 5 (7%) showed a low level of change suggesting no trend. Puget Sound (18%), Olympic Peninsula (20%), and Southwest Washington (38%) DPSs had the lowest proportions of populations with increasing trends. Abundance data were not available for 41 populations.
  • Regarding population productivity, average within-DPS population growth rates (2005-2010 brood years) have been higher for Lower Columbia River, Middle Columbia River, Upper Columbia River, and Snake River Basin populations than for Puget Sound and Olympic Peninsula populations. Population growth rates above replacement for Columbia River Basin populations, which were historically reduced to very low adult abundances, are encouraging and suggest that compensatory juvenile production (increased per capita productivity at low adult abundance) may be sufficient to allow population growth at current smolt-to-adult return rates (< 5 %).
  • We report population spatial structure status in terms of habitat loss due to large, impassable dams. The percentage of populations in the following DPSs with 5% or greater habitat loss are: Puget Sound - 19%; Olympic Peninsula - 0%; Southwest Washington - 11%; Lower Columbia - 22%; Middle Columbia - 44%; Upper Columbia - 75%; Snake River Basin - 25%. With a habitat loss of 98.7%, the Baker River population (Puget Sound) is considered extirpated. Grand Coulee Dam on upper Columbia River mainstem extirpated at least six historical Upper Columbia DPS populations. Other fish passage barriers such as culverts and roads reduce a population’s spatial extent and it is likely that steelhead habitat loss from these smaller barriers is extensive. However, data were not available statewide to quantify these habitat losses.
  • Evaluation of life history diversity status within and among DPSs was limited due to few populations with appropriate data. Freshwater age composition from smolts was available for 17 populations or subpopulations statewide, and the majority of smolts left freshwater after two years, with average outmigration age ranging from 1.6 to 2.4 years. Based on data from 21 winter-run and 15 summer-run populations or subpopulations, winter-run steelhead average ocean age ranged from 2.3 to 2.5 years, and summer-run steelhead average ocean age ranged from 1.4 to 2.0 years. Similarly, evaluation of genetic diversity status was limited statewide due to only 50% of populations having available baseline data. Genetic data availability was lowest for Olympia Peninsula and Southwest Washington DPSs (16.1% and 21.1% of populations, respectively).
  • The size and type of hatchery steelhead releases served as indicators of diversity risks from potential gene flow between hatchery and wild steelhead. Total average annual smolt releases for 2009-2013 were lower than those for 2000-2008 in all DPSs except for Southwest Washington. The largest reductions in smolt releases between the two time periods occurred in Puget Sound DPS, followed by Olympic Peninsula and Snake River Basin DPSs. Average percent of off-site hatchery smolt releases was highest in Olympic Peninsula, Middle Columbia, and Snake River Basin DPSs in both time periods. In Puget Sound DPS, average percent of off-site hatchery smolt releases for 2009-2013 was substantially lower (16.4%) than that for 2000-2008 (40.4%).
  • Statewide, among populations with available data, the long-term abundance trends of only 10 populations (14%) declined by > 55% (high risk criterion 1). Eight of those populations were in Puget Sound DPS and there was one each in Olympic Peninsula and Lower Columbia DPSs. Regarding risk from short-term (12 year) decline, only five populations (7%) showed a growth rate significantly less than zero (high risk criterion 2). These included one Puget Sound population, two Olympic Peninsula populations, and two Southwest Washington populations.
  • Of the 69 populations with abundance data and a defined quasi-extinction threshold (QET), 18 (26%) had a > 20% probability of abundance falling below their QET at least once in the next 20 years (high risk criterion 3). The Lower Columbia DPS had the lowest percentage of populations (17%) that met this extinction risk criterion and the Upper Columbia DPS had the highest percentage (75%).
  • Of the 71 populations with defined escapement or recovery goals and appropriate abundance data, 51 (72%) did not have abundance values above their escapement or recovery goal in seven or more of the recent 10 years (high risk criterion 4).
  • Out of the 21 populations rated at high risk over all criteria, we selected 15 as focal populations in these DPSs: Puget Sound - 9; Lower Columbia - 2; Middle Columbia - 1; Upper Columbia - 2; Snake River Basin - 1. We provide detailed descriptions of threats and actions for the focal populations based on existing management and recovery plan documents and highlight WDFW opportunities for actions.
  • Historical changes in steelhead abundance have resulted from changes in habitat (freshwater and marine), dams and passage barriers, hatcheries, and harvest. However, the contribution of each factor to recent changes in abundance has been unequal. Recently, variable marine survival has influenced adult abundance trends among DPSs. Additionally, higher juvenile freshwater survival through Columbia Basin hydropower dams in recent years has provided relatively synchronous improvements in adult abundance for Upper Columbia River, Middle Columbia River, and Snake River Basin DPSs. The effects of changes in hatchery management along with freshwater spawning and rearing habitat conditions on short-term abundance trends are less clear because population abundance response rates are likely slower, less synchronous among populations, and/or harder to measure.
• Key Results: Threats and Actions
  • Hatcheries: Ongoing hatchery reform efforts are focused on reducing threats to wild populations posed by hatcheries. WDFW has designated areas as wild steelhead gene banks in Olympic Peninsula and Lower Columbia DPSs and is working on designations in other DPSs. Statewide risk-reduction action recommendations included: 1) continue hatchery reform efforts to ensure hatcheries are compliant with HSRG recommendations; 2) improve population-scale monitoring of hatchery genetic and ecological impacts; 3) implement volitional steelhead smolt release; and 4) incorporate demographic analysis into conservation hatchery planning to determine whether the demographic boost intended by such programs is actually needed and to facilitate appropriate sizing of these programs.
  • Harvest: Long-term average harvest rates have been greatest in the Olympic Peninsula DPS (27%) and lowest in the Lower Columbia DPS (5%), with moderate harvest rates in Puget Sound (7.2%) and Southwest Washington (6.8%) DPSs. However, the population-specific impacts of these harvest rates are difficult to interpret without accompanying demographic analyses, since risks posed by harvest depend on a population¡¦s productivity. Harvest threats to wild steelhead are managed through mark-selective sport fisheries that require wild steelhead release, by legal prohibition of non-tribal commercial sale of steelhead, and through tribal treaty fisheries¡¦ operations designed to achieve conservation objectives. However, harvest-related threats remain, including 1) the inability to completely account for fishery non-retention mortality in all areas outside the Columbia Basin; 2) existing methods producing inaccurate harvest impact estimates; and 3) unknown levels of mortality from illegal harvest or unreported catch. Recommended threat-reducing actions included: 1) implement the practice of systematically and consistently estimating total harvest mortality on populations statewide; 2) initiate and continue studies to measure non-retention mortality and sub-lethal impacts to steelhead released from fisheries; 3) quantify risks by developing methods to estimate harvest mortality that include measures of precision and bias; and 4) undertake studies to quantify illegal harvest of wild steelhead and work to increase enforcement where necessary.
  • Dams, barriers, and fish passage: Dams and other fish passage barriers continue to present some of the greatest threats to steelhead viability statewide. Besides blocking or restricting upand down-stream migration, barriers interrupt wood and sediment transport, hindering maintenance and creation of critical habitat. WDFW estimated that 18,000-20,000 barriers to salmon and steelhead exist in Washington streams. In the Columbia Basin, threats to all life stages from passage at mainstem dams and reservoirs include longer and delayed migration times, greater exposure to predation, elevated water temperatures, and physical harm. Action recommendations included: 1) continue to work with all dam operators to ensure adult and juvenile passage survival targets (required by licenses, settlement agreements, and Habitat Conservation Plans) are met; 2) improve downstream passage of pre-spawn adults, kelts, and juveniles at all dams and irrigation diversions, 3) continue active engagement on the Columbia River Technical Management Team, Fish Passage Advisory Committee, and Comparative Survival Study; 4) improve mapping of small fish passage barriers and potential upstream habitat; and 5) complete the statewide removal of all artificial fish passage barriers, including hatchery facility structures, on WDFW-owned lands by 2026. Activities related to many dams, barriers, and fish passage actions described in this document have already begun and we hope that they will be continued.
  • Habitat: Freshwater habitat that steelhead depend on is degraded throughout the state due to a legacy of natural resource extraction, agricultural practices, increased surface flow diversion, and development. Habitat loss is a primary factor limiting steelhead abundance and recovery in all DPSs. Actions that would increase freshwater habitat capacity (quality and quantity) may be highly effective for increasing adult abundance if, as suggested by our productivity analysis, the number of smolts per spawner is density limited. Recommended actions include: 1) continue to develop tools to link life stage-specific survival bottlenecks directly to habitat restoration actions; 2) continue to support habitat restoration by expediting Hydraulic Project Approval (HPA) permits, providing environmental engineering expertise, and offering steelhead-specific planning and design assistance; 3) continue to invest, including via acquisition and easement, in instream flow enhancement where feasible including instream water transfers, irrigation or water-use efficiency projects, and fish screens at instream diversions to help achieve adequate streamflow levels, guard against critical temperature exceedance, and prevent fish entrainment; 4) continue to collaborate with habitat restoration and protection efforts statewide, especially cold water refuges, riparian, and nearshore habitats; and 5) continue to protect existing freshwater habitat through HPA program, Forest Practices, Priority Habitats and Species (PHS) riparian management recommendations, and acquisition and easement programs. Activities related to many habitat actions described in this document have already begun and we hope that they will be continued.
• Key Results: Monitoring and Data Gaps
  • The lack of abundance, productivity, and diversity data was most the common impediment to conducting wild steelhead status assessments statewide. For example, even basic adult abundance data are unavailable for 33 of 114 (29%) extant populations, and smolt-to-adult survival estimates exist for only 16 of the 114 (7%). Implementing methods that will increase the availability of monitoring data, as well as their accuracy and precision, and improve future evaluations of population status should be prioritized. High-quality monitoring data are needed to measure effectiveness of management actions intended to improve status.
  • Steelhead populations in western Washington are generally less-rigorously monitored than those in the central and eastern part of the state. Monitoring of Columbia Basin populations has improved in the past decade. It is critically important that ESA-listed populations statewide be monitored sufficiently to detect progress towards delisting. Specific recommendations included: 1) initiate comprehensive population-scale monitoring, including adult and juvenile abundance and age composition, for one or more moderate- to large-sized populations in Puget Sound, Olympic Peninsula, and Southwest Washington DPSs and 2) develop improved redd monitoring designs that include representative sampling, estimates of females per redd, observer efficiency, redd life, and methods to account for error in subsequent abundance estimates.
  • Harvest monitoring data gaps include few current mortality estimates for wild steelhead released in sport fisheries or from net drop-out in tribal fisheries along with a lack of uncertainty estimates for mortality due to other fisheries. We suggest that modifying the sport catch record card (CRC) system to accommodate unbiased estimation of wild steelhead release mortality would greatly reduce creel survey costs and increase spatial and temporal coverage of sport fishery impacts. We recommend that methods be developed to estimate numbers of wild steelhead released by sport fishers and to incorporate measurement uncertainty in CRC-based harvest estimates. We make additional suggestions, above, about improving our ability to monitor harvest mortality.
  • Monitoring hatchery impacts is hindered by currently the low capacity to, and high cost of, estimating gene flow or percent of hatchery-origin steelhead spawning in the wild. Gene flow assessments are possible using genetic stock identification (GSI) for segregated hatchery stocks, but due to high costs this method has only been implemented in a small subset of stocks and years. Gene flow assessment is particularly difficult for integrated hatchery programs due to 1) genetic similarities of hatchery and wild fish, which negates the use of GSI as a monitoring tool, and 2) difficult observational conditions on spawning grounds, which complicate visual identification of fin-clipped hatchery-origin spawners. For most western Washington populations, no methods currently exist to obtain monitoring data required for integrated hatcheries to be in compliance with Hatchery and Genetic Management Plans. For integrated hatcheries, we recommended the development of parentage-based or field methods for quantifying reproductive interactions between wild and hatchery steelhead. Limited efforts have been made to quantify ecological impacts of hatcheries on wild populations, including competition, predation, and disease. We recommend future efforts to better quantify these impacts.
  • Improvements are needed in the ways WDFW¡¦s monitoring data are stored, managed, and made available to managers and others. Specific data management recommendations include: 1) expand and develop data entry and data analysis capabilities of the WDFW-managed juvenile salmonid database and 2) develop a standardized harvest reporting database where steelhead exploitation rates and supporting metadata are entered and stored.
• Key Results: Progress since 2008: Scott and Gill (2008) contained a list of 35 recommendations on eight topics: biology, habitat, artificial production, management, population structure, diversity and spatial structure, and abundance and productivity. We report attention and action directed towards 30 of these recommendations.
• Intended Use and Relationship to Other Evaluations: This report is intended as a statewide review of steelhead population status, threats, and recovery actions. Its scope is sufficiently large to encompass goals of population assessments and recovery planning processes conducted at other spatial scales (e.g., NOAA DPS-scale status reviews and watershed/population-scale habitat restoration, fishery, and hatchery management plans). As such, this report does not supersede other efforts but rather complements them by synthesizing their findings and rendering them comparable at a statewide scale. In order to accomplish this we had to develop standardized methods to compare populations statewide regardless of differences in data availability and methods used in previous regional reviews and management processes. The risk assessment is at a relatively coarse scale in order to maintain comparability among populations, and it does not make use of more extensive data available for smaller subsets of populations.

This report presents a high-level synthesis of population status, threats, and recovery actions rather than a detailed recovery or action prioritization plan. We focused on near-term recovery and management actions that 1) could improve status and viability of individual steelhead populations, entire DPSs, and statewide and 2) WDFW has a greater ability to influence or implement over shorter time frames with less dependency on external support. For this second reason, habitat restoration-related actions were not a principal focus of this report.