Sablefish (Anoplopoma fimbria) is a scarcely represented species in Northwest Coast archaeology, but its remains are abundant at Tse-whit-zen, a large, Lower Elwha Klallam village in modern Port Angeles, WA that was occupied over the past 2,800 years. Because sablefish flesh has high nutritional value and it can be easily captured from nearshore waters in its juvenile form, sablefish should have been pursued where it was available. Therefore, the scarcity of sablefish in many Northwest Coast archaeological sites could indicate this species was not abundant in past fisheries. However, current zooarchaeological reports do not contain sufficient information on taphonomic histories, sampling, or zooarchaeological methods to determine whether patterns of sablefish scarcity could actually explained by differential destruction of sablefish remains, sample size effects, screen size effects, or misidentification. In this thesis, I examine how each of these factors may have affected the abundance of sablefish remains in Northwest coast archaeological sites. I evaluate four hypotheses that attribute sablefish representation to zooarchaeological identification methods, screen size, sample size, and post-depositional destruction of fishbone. While I do not explicitly test whether social and ecological factors affect sablefish abundance, sociocultural and environmental variation can be considered likely explanations for the observed patterns of sablefish representation if the other hypotheses are rejected. I test my hypotheses using three scales of archaeological records. First, I reanalyzed six previously analyzed Salish Sea assemblages to assess whether criteria for sablefish identification exist, are valid, and have been applied consistently. Second, I synthesized fishbone data from 35 previously analyzed Northwest Coast assemblages to evaluate the effects of screen size, sample size, and post-depositional destruction on sablefish representation. Finally, I integrate previously unreported fishbone data from the analysis of Tse-whit-zen into the synthesis of previous studies. The Tse-whit-zen materials I report on here represent six discrete time periods in the 1,800-year history of one large area of the site, which encompasses part of a plankhouse, providing a unique opportunity to examine the effects of screening, sample size, and post-depositional destruction at an extremely fine scale. I also use data from the reanalysis of a portion of the Tse-whit-zen fishbone to verify the consistency of sablefish identification for this site. I reject all four hypotheses and conclude that the uneven distribution of sablefish is likely a true reflection of ecological factors, human decision-making, or both factors. Whether sablefish scarcity is related to distributions of sablefish in past environments, or whether humans chose not to pursue sablefish is not known from the current study. Connecting sablefish capture to specific seasons with body-size regression methods may reveal associations between sablefish acquisition and other seasonal fisheries and activities, and help evaluate whether they conflicted with sablefish procurement in some contexts. Although zooarchaeological identification and reporting methods do not appear to account for sablefish scarcity, zooarchaeologists need to include more information about their methods so that the validity of inter-assemblage comparisons can be assessed. Zooarchaeologists maximize the value of their contributions to anthropology, biological sciences, and human ecodynamics when they explicitly report the methods they use to identify animal remains. By reporting the methodological and analytic procedures they used in detail, zooarchaeologists enhance the reader's confidence in their conclusions and provide future researchers with the information that is required to replicate their results. Which elements were recorded, and the criteria that were used to make taxonomic attributions, fundamentally affect the primary faunal data that researchers use. This study is part of a growing interest among zooarchaeologists in data quality assurance and quality control, which constitute a critical part of every large-scale comparative analysis.

Evidence of large earthquakes occurring along the Pacific Northwest coast is reflected in coastal stratigraphy from Oregon to British Columbia, where there also exists an extensive archaeological record of Native American occupation. Tse-whit-zen, a large Native American village dating between 1824 and 54 cal B.P. located on the Olympic Peninsula of Washington State, was excavated with exceptionally fine stratigraphic control allowing for precise comparison of natural and cultural records. Here I report on the >10,000 fish remains from one 2x2 m excavation block; this assemblage spans one earthquake event, allowing study of changes in relative taxonomic abundance through time that may coincide with earthquakes or other environmental changes. Results indicate use of a wide range of marine fish taxa including various sculpins (Cottidae), flatfish (Pleuronectiformes), herring (Clupea pallasii) and salmon (Oncorhynchusspp.). This illustrates a highly diverse diet throughout occupation, though relative abundances of more offshore taxa decrease through time in favor of some nearshore taxa, possibly indicating the presence of a coseismic event. This thesis serves as part of a pilot study for a collaborative project that is underway. This larger project addresses human response to both gradual and abrupt environmental change through the analysis of all classes of Tse-whit-zen faunal remains, which provide a link to impacts on animal populations and in turn human subsistence.