First Advisor

Virginia L. Butler

Term of Graduation

Winter 2020

Date of Publication


Document Type


Degree Name

Master of Science (M.S.) in Anthropology






Archaeology -- Oregon -- Sauvie Island, Archaeology -- Lower Columbia River (Or. and Wash.), Sauvie Island (Or.) -- History, Wetland ecology -- Lower Columbia River (Or. and Wash.)



Physical Description

1 online resource (xiv, 145 pages)


There is a growing interest in anthropology towards identifying and documenting the ways in which people have modified landscapes and ecosystems through time. Previous research has focused predominantly on terrestrial modification, whereas recently, research has turned towards aquatic environments. Examples range from the tidal fish pens of Hawai'i to fish weir complexes and clam gardens on the Northwest Coast. Scholars are beginning to apply the term human ecosystem engineering to the practices linked to wetland landscape modifications. Evidence of these practices can contribute to understanding optimization, cultivation, and modification of aquatic environments on a landscape level. However, some regions have received more consideration than others; the backwater wetlands of the Lower Columbia are minimally studied.

Lower Columbia archaeological and ethnohistoric records highlight the importance of aquatic plants and mammals that inhabit wetland environments; freshwater fish remains from families Catostomidae and Cyprinidae are prominent in regional archaeological site assemblages. This raises questions as to how wetland resources in the region were used and potentially optimized in the broader context of the backwater ecosystem, and how this ecosystem has been modified by humans. A wood stake alignment located on Sauvie Island in Virginia Lake, a seasonally flooded backwater, offered a starting place to examine these questions.

I conducted archival research, pedestrian survey, site mapping, subsurface testing, and laboratory analysis to evaluate hypotheses related to the feature's age (precontact, historic, and multicomponent), cultural affiliation (Indigenous, Euro-American, or both), and function (fish weir, causeway/pier, hunting platform or blind, post and line structure for straightening/storing cedar planks, boundary line/fence, or multi-use). Results from Virginia Lake were compared to four sites containing wood stake/post alignments in Lower Columbia River wetlands. Niche Construction Theory and Historical Ecology informed my analyses.

Fieldwork and site documentation were conducted at two scales; landscape-level and site-specific, and included pedestrian survey, metal detector survey, and excavation of 11 subsurface shovel probes, and two 1 x .5 m test units. Six wood samples and one sediment sample were submitted for AMS dating and macrobotanical identification. Toolmarks were analyzed to determine if the stakes were shaped using metal or stone tools.

The alignment (northeast/southwest orientation) consists of a total of 23 wooden stakes that extend approximately 60 m from the lake edge. Stakes average 5.9 cm in diameter, and 25.6 cm height above ground surface, but stakes are not uniform in size or condition. Stakes are spaced an average of 2.5 m apart. The five sampled stakes are from western red cedar (Thuja plicata). Except for two small metal fragments found within two stakes, additional artifacts were not observed in association with the alignment. One of the removed stakes with an intact distal end appears to have been shaped with a metal tool, likely a chisel or axe. Calibrated median AMS dates fall within a tight range, between 1847 and 1854 CE. AMS dating of humin fraction of a peat sample obtained during excavations returned a calibrated date of between 969 and 1035 CE (925-981 BP). This date suggests that Virginia Lake has been separate from the main Multnomah Channel for at least the last approximately 1000 years.

Though the AMS ages indicate an historic era construction, the feature could reflect Euro-American or Indigenous affiliation, or links to both, given that archival records document a complex and ongoing relationship between Indigenous people and Euro-American settlers in the region through the 19th century. Research eliminated four hypotheses related to function, leaving the possibility that the alignment was a fish weir or a pier. The Virginia Lake feature is distinct from other wood stake and post alignments documented on the Lower Columbia River making it difficult to place it in a larger system of ecosystem engineering and wetland modification.

My thesis contributes to the understanding of the history of Virginia Lake and its formation, and facilitates future work on wood stake features in backwater systems, which have been under-studied in the Pacific Northwest. Researchers cannot adequately test for specific site types, if there is little precedent for them to exist in a given environment. My thesis provides a methodological template for evaluating wetland landscape modifications, and more specifically stake sites, increasing replicability and a richer understanding of their role in human-modified landscapes.


In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

Persistent Identifier

Included in

Anthropology Commons