The purposes of this study were to examine 1) understandings of science expressed in formal salmon recovery policy discourse; 2) rhetorical practices employed to justify or undermine claims about salmon policy 3); and patterns of understandings of science and associated rhetorical practices between social categories of actors. This research contributes to scholarship in public understanding of science, discourse studies, and natural resource policy. A constructivist discourse analysis was conducted using qualitative methods to analyze transcripts from over one hundred congressional hearing witnesses representing a wide diversity of stakeholder groups. Multiple coders organized discourses into analytic categories, achieving a final proportional agreement of 80% or greater for each category, at the finest scale of analysis. Stakeholders employed a collection of prototypical understandings of the nature of science, boundaries of science, and roles of science in decision-making. The process of science was described as impartial and ideal, a way to reduce uncertainty through consensus and peer-review, and subject to changing paradigms. Scientific knowledge was sometimes represented as "truth" and other times as tentative, and scientists were portrayed as independent and objective as well as captured and interest-driven. Witnesses described science as separate from and superior to politics and management. Testimony included descriptions of science?s role in developing decision alternatives, selecting among alternatives, and evaluating and legitimating alternatives. Stakeholders used these understandings of science to construct justifications to support their claims about salmon policy and undermine opposing claims. Science-based justifications included externalizing devices, construction of consensus, category entitlement, and extreme case formulations. Other justifications invoked local control, treaty rights, and local knowledge, or relied on interest management. This study has extended the theory and method of empirical discourse analysis, and produced a taxonomy of understandings that should be transferable to studies of similar policy settings. Additionally, conclusions from this study about differences between social groups in the presence, distribution, and frequency of expression of the discourses might be developed into propositions for further testing. Finally, the study has implications for communication about the role of science in collaborative natural resource decision-making processes. / Ph. D.

Scotts Creek, in Santa Cruz County, Calif., supports the southernmost extant population of Coho Salmon (Oncorhynchus kisutch) in North America. In 1997, the California Department of Fish and Game (now Fish and Wildlife) conducted an extensive habitat typing survey of mainstem Scotts Creek, describing all habitat units from the top of the estuary to the limit of anadromy approximately 12 km upstream. I repeated this survey in 2013 to (1) assess changes in the quantity and quality of instream habitat, (2) compare the current condition to goals and standards established in the federal Central California Coast (CCC) Coho Salmon Recovery Plan, and (3) identify opportunities for possible future restoration. A comparison of the two surveys revealed an overall increase in mean canopy cover, mean bank vegetation, mean percentage instream cover, pool depth diversity, and percentage riffles since 1997, and decreases in mean residual pool depth, percentage flatwater, and number of primary pools. Overall, the percentage of the total mainstem classified as pool habitat did not change between the two survey periods. Results for individual habitat metrics were more variable when the stream was broken into discrete reaches delineated by major tributary junctions. Although a large woody debris (LWD) survey was not conducted as part of the 1997 survey, contrasting our results with data collected during intervening years indicated that instream LWD has become more abundant, primarily due to increases in hard-wood species (i.e., red alder [Alnus rubra] and California bay [Umbellularia californica]). When compared to habitat goals established in the federal CCC Coho Salmon Recovery Plan, Scotts Creek has adequate canopy cover and percentage pools, but is lacking in percentage riffles, instream cover, key pieces of LWD per100 m, and percentage primary pools.

xiii, 236 p. : ill. (some col.) / Over the past two decades, contemporary state-level watershed management burgeoned in the Pacific Northwest. This research offers a comparative analysis of contemporary state-level watershed management frameworks in the Pacific Northwest. The four case study areas consist of Idaho, Oregon, Washington, and Northern California. This study begins with a historical analysis of the greater watershed movement in natural resource management. Next, document analysis and key informant interviews are utilized to detail the watershed management framework of each state. Finally, this study explores a comparative analysis of each state framework. Results indicate that while the case study areas share many characteristics endemic to the bioregion, the watershed management framework of each state differs substantially. Key informant interviews indicate that these differences often reflect the unique sociopolitical climate of each state. Results additionally indicate the vital importance of stable state-derived funding for the establishment and resilience of watershed management organizations. / Committee in charge: Dr. Michael Hibbard, Chairperson; Dr. Richard Margerum, Member; Dr. Max Nielsen-Pincus, Member