Grays Harbor Fall Chum Salmon (Oncorhynchus keta) Abundance and Distribution: Final Report, 2015-2020

Background

Pacific Salmon (Oncorhynchus spp.) are an important economic and ecological resource in the lower Chehalis River system (Nelson and Reynolds 2014). The goals of this project were to improve estimates of Grays Harbor Fall Chum Salmon (O. keta) spawner abundance and describe the distribution of this species throughout the sub-basins of Grays Harbor. Grays Harbor Chum Salmon populations include spawners within the Chehalis River Basin, Humptulips River Basin, and smaller tributaries that empty into the south of Grays Harbor. Washington Department of Fish and Wildlife (WDFW) and the Aquatic Species Enhancement Plan Technical Committee of the Chehalis Basin Strategy (Aquatic Species Enhancement Plan Technical Committee 2014) identified spawner abundance, distribution, and mapping areas of key spawning habitats of Grays Harbor Fall Chum Salmon as key information gaps in the Chehalis River basin. The current method for estimating Chum spawner abundance, developed in the 1970s, is based on a biased and assumed correlation between the overall escapement goal and the relative proportion of spawners in each index section. The method does not account for changes in spawning habitat or distribution of Chum Salmon within the basin. Pacific Salmon (Oncorhynchus spp.) are an important economic and ecological resource in the lower Chehalis River system (Nelson and Reynolds 2014). Chum Salmon spawn in large aggregations and provide important marine-derived nutrients to the ecosystem. Identifying Chum Salmon spawning “hot spots” in the Chehalis River basin can inform restoration activities by:

1. Ensuring that restoration activities enhance these critical “hot spots”.
2. Highlighting areas that will benefit Chinook and Coho Salmon and Steelhead juvenile production through protection of habitats with large nutrient imports.
3. Targeting areas for Chum Salmon restoration to sustain populations in the future.

Methods

Several approaches were employed to determine Chum Salmon escapement. These included (1) area-under-the-curve (AUC), a method of using periodic counts over the entirety of the Chum Salmon run to generate season totals in set areas, (2) carcass mark-recapture (CMR), a method for independently generating an estimate from tagging and tag recovery based on a Jolly-Seber abundance estimator for open populations, and (3) peak expansion, a method of comparing Chum Salmon observed within the weekly index surveys with supplemental surveys (one-time surveys) during peak spawning performed throughout the rest of basin to expand the estimate from the indexes.

Survey life is the apparent residence time of Chum Salmon in the observed stream and includes both the actual residence time of the fish and the observer’s efficiency. Estimates of survey life are crucial for escapement estimation. By pairing AUC and CMR methods in the same indexes we were able to generate independent estimates of survey life. The independent survey life values were then used in combination with the other AUC indexes to generate an estimate for all indexes within the basin.

Escapement Estimate = AUC Fish - days / Survey Life

Three approaches were used to refine escapement estimates. First, index reaches were selected to estimate survey life in variable stream size classes – side channel, small, medium, and large. These were then applied to the remainder of AUC indexes based on their stream size. Second, live counts were separated into spawner and holder counts or total live counts. Spawner counts represented Chum Salmon holding or actively spawning on spawning habitat, while holders represented Chum Salmon holding in pools or not actively spawning. Total live counts represented a combination of holder and spawner counts and were used when counts were not separated. Escapement was calculated from spawner counts and total live counts separately.

Finally, due to the size of the Grays Harbor basin, it was divided into four sub-basins: Humptulips, Wynoochee, Satsop, and “Other” (i.e., all other Chum Salmon streams including Wishkah, Hoquiam, Cloquallum, and Black rivers, and South Bay tributaries). One sub-basin was selected as the focus each year; in 2020 that was the “Other” sub-basin. In the non-focus basins, annual indexes were surveyed for other species, but Chum Salmon counts were leveraged from these surveys and distribution information applied from a previous year when the basin was the focus to apply a ratio of Chum Salmon inside these indexes to those not covered.

Results

The new or “updated” 2020 escapement estimate for Grays Harbor Basin Chum Salmon spawner abundance was 48,294 (CV = 2.2%) with 95% probability interval of 44,597 and 52,250. The historic or “current” method produced a point estimate of 23,457 with no estimate of precision. The “current” method was 49% lower than the “updated” method. Of the 2020 focus streams using the “updated” method, Wishkah sub-basin had the highest contribution with just over 4,300 Chum Salmon, while Hoquiam and Cloquallum sub-basins both contributed less than 500 chum each. Both Black River and the South Bay tributaries each contributed less than 30 fish to the total Chum Salmon abundance estimate.

Important Study Findings

• Results from monitoring Chum Salmon over the study period suggest that the historic, or “current” method used to produce Grays Harbor Chum Salmon spawner abundance over the last 40 years may underestimate escapement by up to 50%.
• Chum Salmon distribution is widespread in the lower basin, however 90% of the population is spawning within the Humptulips, Wynoochee, and Satsop sub-basins.
• Survey life (measured in days) is the value used to define Chum Salmon residence time in streams and is a parameter in the AUC equation used to calculate abundance (fish). We found survey life was typically smaller in larger streams with greater average bankfull widths; survey life values in larger streams were nearly half those in smaller streams. Values of survey life incorporate residence time and observer efficiency.
• Using counts of fish on spawning habitat (spawner counts) as opposed to counts of all fish (total live counts) produced more accurate survey life estimates; therefore, we recommend that all counts separate spawners and holders.
• It is possible to generate escapement estimates in a large basin like Grays Harbor (4,184 km of river) with precision by measuring density, distribution, and unique survey life values at the sub-basin scale, then updating those values regularly to account for spawning abundance and distribution changes over time.

The findings from 2020 and previous years of the study may be used to inform restoration efforts for other salmon species within Grays Harbor. High-density Chum Salmon spawning locations should be considered when planning and implementing restoration activities. Low-density areas should be candidates for improving habitat for all salmon species. High density spawning areas provide important ecological role by spawning in high densities and providing marine-derived nutrients which benefit all aquatic species. Our work demonstrates that there are alternative methods that are accurate and repeatable with a measure of precision for estimating Chum Salmon spawner abundance in the Grays Harbor basin. In particular, this work indicates that studies based on carcass mark-recapture generate more accurate estimates of survey life and spawner abundance then the historic, or “current” method, and updating distribution information regularly is needed to maintain accuracy. This work provides critical guidance for practitioners looking to update and improve Chum Salmon escapement estimation methodology for management and conservation purposes, especially in large basins where intensive surveys are often impractical.