DATING OF QUATERNARY ERUPTION EVENTS USING U-SERIES AGES OF LAVA CAVE GYPSUM CRUSTS

Primary igneous anhydrite has been reported in a number of eruptions including El Chichon and Mt. Pinatubo. Similarly, sulfates have been observed to form in volcanic caves on Mt. Etna, but disappear within months of deposition. Our study of unstratified, densely crystalline gypsum crusts in lava caves in the Zuni-Bandera volcanic field in the El Malpais National Monument, New Mexico, indicates that these crusts are well-preserved primary features of eruptions. These lava tube gypsum crusts were likely formed from sulfuric acid percolation with evaporative cooling of the host basalt, and were ultimately preserved in drier parts of the caves. U-series dating of these types of crusts yield ages of lava tube formation, and therefore eruption ages. These crusts show relative enrichment of U to Th, making them suitable candidates for ²³⁴U-²³⁰Th dating. Precise radiometric dating of recent lava flows has been a problem due to the inability to resolve inherited isotopic signatures complicated by magma chamber processes. This technique may allow us to circumvent coupled burial/erosion histories with cosmogenic ages, or lack of K in basalts that limit K/Ar and Ar/Ar techniques.

The gypsum crusts have ⁸⁷Sr/⁸⁶Sr values (0.7039-0.7057) that are comparable to the Zuni-Bandera basalts (0.7030-0.7060), and distinctly lower than soils developed on the flows (0.7070-0.7120). δ³⁴S values of the crusts from two flows (~1 - 4 ‰) correspond with a magmatic source consistent with residual SO₂ degassing (δ³⁴S: -2 – +10 ‰), excluding biotic formation or alteration.

Uranium-series dates with a mean of 8.5 ± 1.2 ka from three sites in the Bandera flow are consistent with dates using other techniques (10.6-11.8 ka). One sample from the Hoya de Cibola flow, estimated to be between 15 - 20 ka, gives U-series age of 18 ± 3 ka. One cave site in the Bandera flow gives anomalously lower ages of 5 ka. Petrographic analysis shows the sample to contain abundant clay-like inclusions not observed in other crust samples and may be responsible for the age discrepancy. We are working on criteria to distinguish between pristine and altered crusts. Availability of this technique should open new opportunities in a number of fields that utilize ages of lava flows as constraints such as surface processes, archaeology and volcanic eruption intervals for hazard assessment.