Date of Award


Document Type


Degree Name

Master of Arts in Applied Archaeology



First Reader/Committee Chair

Gusick, Amy


Archaeological investigations at Edwards Air Force Base (AFB) have been ongoing for more than 40 years. Yet the findings from the vast majority of those efforts are available only as grey literature that is known only to a relative few. The primary purpose of this thesis was to investigate a late Holocene decline in obsidian frequency reported by researchers working in the Bissell Basin and Rosamond Dry Lake region of Edwards AFB near the turn of the 21st century. A secondary purpose of this thesis was to shine a light on an area of the western Mojave Desert that is not widely known despite more than four decades of research.

In order to explore the reported decline in obsidian frequency, I created an obsidian database using data gleaned from nearly 50 cultural resources management reports and supplemented those data with sourcing and hydration information for 39 additional obsidian artifacts. Those data were organized into tables, charts, and histograms to look for patterns that would support or refute the claim that obsidian use decreased significantly after the Gypsum time period (4000 to 1500 Before Present [B.P.]). Two patterns emerged from my analysis.

The first was one where the overall abundance of obsidian at Edwards AFB did not decrease during the late Holocene, thus contradicting the conclusions made in the previous research. The second was one where the obsidian abundance shifted among the various regions of the installation. Yet these shifts are nowhere near as significant as the previously reported decline. Therefore, while the total amount of obsidian that entered the archaeological record at Edwards remained relatively stable from 1500 to 100 B.P., the amount of obsidian decreased in certain regions and increased in others.

Although not within the scope of my original intent, my research also identified two areas for future research. The first involves an apparent pattern where the number of archaeological sites from which obsidian was recovered gradually decreases during the middle-Holocene even as the overall quantity of obsidian remains essentially unchanged. The second relates to the lack of a well-established hydration rate formula for obsidian recovered from archaeological sites on Edwards AFB.

Ultimately, I concluded that the previous findings that obsidian declined during the late Holocene were affected by sampling bias and faulty data organization. Most archaeologists understand that poorly implemented sampling can lead to poorly derived findings and conclusions. What may not be as well understood is that a perfectly appropriate sample where the data are not organized well can also lead to flawed results and conclusions. It is hoped that this thesis will inform archaeologists not only about how the manner in which they organize their data can affect their interpretation of past human behavior, but also about additional research opportunities at Edwards AFB.