OBSIDIAN WATER, OBSIDIAN HYDRATION: INVESTIGATING VARIATION IN THE GEOCHEMISTRY OF JEMEZ RHYOLITIC GLASSES

The assertion that artifact-quality obsidian is uniformly low in water content is nearly universal in the archeological literature on obsidian hydration dating (OHD). Also ubiquitous is a linkage between obsidian artifact provenenance and source-specific rates of hydration. The resulting methodological confusion has contributed to a current pessimism regarding this once promising dating technique.

However, the literature leading to the development and introduction of OHD actually used the “less than 1%” standard for water content in rhyolitic glass as arbitrary and research-driven rather than descriptive of the natural range of variation. The relationship between “source” and hydration rate raises a different issue. In this case, the use of two concepts, 1) “source” (chemical type as defined by trace elements), and 2) “rate” (of hydration, specific to the geological source), unnecessarily conflate description and cause. That is, while artifacts of the same chemical type often do have similar hydration rates, the trace elements used to define the type do not play a causal role. In contrast, water content likely does play a role in hydration rate, where more water correlates with a faster rate. Clarity in understanding obsidian source, geochemical composition, and causes of variability in volatile content will be critical for the successful application of OHD.

Analyses of obsidians from the Jemez volcanic field in north-central New Mexico indicate strong differences in water content in obsidians in the Valles Rhyolite (VR) versus the Cerro Toledo Rhyolite (CTR). Using loss-on-ignition (LOI) and Fourier-transform infrared spectroscopy (FTIR) to measure intrinsic water content, obsidian samples from VR deposits are uniformly low in water (total LOI 0.2% - 0.6%; mean 0.4%), while samples from the CTR are variable and high in water content (total LOI range 0.3% - 1.3%; mean 0.9%). These results have important implications for OHD because artifacts of CTR glass are likely to hydrate at multiple rates depending on water content. The Jemez Mountains provide an intriguing context for modeling relationships between low water content in glasses from effusive silicic volcanism (as in VR obsidians) in contrast to the high and variable water contents in CTR obsidians where explosive eruptive activity played a greater role.