Advisor

Virginia L. Butler

Date of Award

Winter 3-30-2015

Document Type

Thesis

Degree Name

Master of Science (M.S.) in Anthropology

Department

Anthropology

Physical Description

1 online resource (x, 80 pages)

Subjects

Chinook salmon -- Age determination – Pacific Northwest, Animal remains (Archaeology) -- Pacific Northwest, Vertebrae -- Growth

DOI

10.15760/etd.2216

Abstract

Incremental growth rings in X-rays of salmon vertebrae have been used since the 1980s to age Pacific salmon (Oncorhynchus spp.) remains from archaeological sites in the Pacific Northwest. These age estimates, paired with generalized life history patterns, have been used to determine salmon species, season of capture and in turn season of site occupation. This approach relies on a variety of assumptions, the most fundamental of which is that rings represent true years. Archaeologists using vertebral age determination techniques have failed to adequately test this assumption and present their methodologies. This thesis assesses the validity of using incremental growth structures in Chinook salmon (Oncorhynchus tshawytscha) vertebrae to determine the age at death of fish represented in archaeological sites. This project develops criteria and a protocol for the identification of true annuli and tests these identifications on a collection of modern Chinook salmon of known age. Finally, this protocol is applied to archaeological remains of Chinook from Cathlapotle (45CL1).

Three collections of modern known age fish (N=121) were used to evaluate and test approaches to aging Chinook salmon with vertebrae. These collections contained juvenile and adult Chinook from throughout Washington and Oregon. I evaluated a variety of methods for viewing rings including magnified surface images, X-ray images and thin sections to determine which is the most accurate, reliable and efficient, also considering the extent of specimen destruction. Rings visible in X-rays were found to reflect the internal structure of vertebrae rather than annular growth. The number of these internal walls did not correspond to the known ages of fish and are therefore not true annuli.

Criteria previously described by salmon fisheries biologist were used to isolate annuli (on the centrum surface) on the Hanford reach collection (N=46). In a test for accuracy 39 (85%) were aged correctly. In a test of reliability utilizing five additional readers all but 14 cases showed discrepancies among readers. Results of the test of reliability were not as successful as other researchers in fisheries biology but given the high accuracy rate the method remains valid.

The final goal of this project was to determine the feasibility of applying the surface ring method of age determination to archaeological collections. Archaeological salmon vertebrae from Cathlapotle (45CL1) on the lower Columbia River, Washington state were utilized. One hundred salmon vertebrae were selected and classified to species according to Huber et al. (2011); 89 were identified as Chinook. Of these, 39 had sufficient preservation of the surface to view and interpret incremental rings. Three ages were identified 3, 4 and 5 year olds. This ageing protocol can be applied to archaeological Chinook salmon vertebrae to estimate age of ancient Chinook salmon. Additional work is needed on other salmon species to demonstrate the methods validity across all salmonid species.

This analysis has great potential for modeling salmon paleo-life history by contributing data from salmon populations prior to the major impacts of the 19th and 20th century. This is particularly valuable for salmon conservation because information on size, the timing and duration of freshwater emigration as well as the age of spawning and death is critical to the management of hatchery and wild salmon populations.

Persistent Identifier

http://archives.pdx.edu/ds/psu/15182

Share

COinS