OASIS - Use of Stratified Random Sampling to Estimate the Abundance of Oregon Coastal Coho Salmon

Identification Information

Citation

Originator: S.E. Jacobs, T.E. Nickelson

Publish Date: 1998

Online Link: http://oregonstate.edu/Dept/ODFW/spawn/pdf%20files/reports/SRS.PDF

BPA Project #:

Contact Information

Agency: Oregon Department of Fish and Wildlife

Name: Mark Lewis

Job Position: Project Leader

Telephone: (541)757-4263 x241

E-Mail Address: mark.lewis@oregonstate.edu

Description

Abstract: Management of coho salmon (Oncorhynchus kisutch) in Oregon is heavily dependent upon estimates of the spawning escapement of Oregon coastal natural OCN) coho salmon. OCN coho salmon consist of an aggregate of wild coho salmon populations that originate from Oregon coastal river basins south of the Columbia River and north of Cape Blanco. OCN spawning escapement estimates are an integral component of the management of salmon fisheries in the ocean. Escapement estimates are also used to assess progress towards meeting goals addressed in Oregon's Coho Salmon Plan (ODFW 1982), the Salmon Framework Plan of the Pacific Fisheries Management Council (PFMC 1984), the Oregon Department of Fish and Wildlife (ODFW) Wild Fish Policy, the Oregon Salmon Restoration Plan (Oregon 1997) and various management plans for coastal basins. During 1985-86, the Oregon State University Department of Statistics was contracted by ODFW to review the method used to estimate OCN spawning escapement and to recommend procedures for improving the methodology (Ganio et al.1986). Five sources of bias were identified in the methodology: (1) methods of survey site selection, (2) estimates of the life span of fish in survey sites, (3) counting of nonspawning fish in survey sites, (4) under-counting of fish in survey sites, and (5)estimates of total miles of spawning habitat. The authors concluded that the method of survey site selection was probably the most serious source of bias because of evidence that survey sites were predominantly located in "better than average" spawning habitat. To improve the methodology used to estimate OCN escapement, the authors proposed a sampling plan designed to reduce survey site selection bias and to provide an estimate of the precision of the escapement estimate. The recommended sampling plan incorporated a stratified random sampling (SRS) scheme, where, within geographic 2 sampling units, survey sites are randomly selected from the estimated available miles of coho salmon spawning habitat. Although the SRS survey procedure proposed by Ganio et al. (1986) was a means of vastly improving OCN escapement estimates, more information was needed before this procedure could be adopted. Procedures needed to be developed for selecting and surveying random survey sites (Ganio et al. 1986). The feasibility of incorporating additional levels of stratification to improve the efficiency of the sampling program and account for additional escapement resulting from hatchery fish was needed. More accurate estimates of the available spawning habitat for coho salmon in coastal watersheds were necessary for the development of an effective SRS scheme and ultimately, as a means of improving absolute escapement estimates. Finally, the relationship between spawning densities estimated from the standard coho salmon survey program and spawning densities estimated from a SRS program needed to be determined.

Purpose: To determine the potential of a SRS survey program to improve estimates of OCN spawning escapement, this study was undertaken with the following objectives: 1. Develop methods needed to implement a SRS approach to making OCN spawning escapement estimates. 2. Determine the precision of OCN spawning escapement estimates produced through a SRS survey program. 3. Improve estimates of the total miles of coho salmon spawning habitat in coastal river basins. 4. Determine the relationship between the spawning density of coho salmon estimated from the standard survey program and the spawning density of coho salmon estimated from SRS. The purpose of this report is to describe the results of eight years of research to develop estimates of spawner abundance for OCN coho salmon using a SRS design.

Time Period of Content: Oct 1989 - Sept 1998

Geographic Extent: Oregon Coast - Nehalem, Tillamook, Nestucca, Siletz, Yaquina, Alsea, Siuslaw, Umpqua, Coos, Coquille, Rogue

Status: Final

Use Constraints:

Format: PDF File

Data Quality Information

Lineage-Source:

Data Information

No data information was supplied.

Entity and Attribute Information

Attributes Description:

Is a physical copy maintained for reference at Headquarters? Unknown