TRACING VIOLENT VOLCANIC ERUPTIONS: DISTANCE INDUCED SHIFTS IN GEOCHEMICAL ASH FINGERPRINTS

The long distance correlation of volcanic ash deposits from plinian eruptions is a critical step in understanding the recurrence rates and sizes of eruptions and their associated impacts. This study is designed to provide a new tool for distal correlations by looking at variations in the geochemical fingerprints of ash by particle size divisions for distal samples with extreme distances simulated. The 1912 eruption of Novarupta, the largest plinian eruption of the last century, was used in this study because of the availability of well correlated and well sampled distal and proximal tephras. The Novarupta eruption consisted of multiple eruptive events, similar to other large eruptions, which are preserved in several layers near the vent and 4 layers at a distance of approximately100 km from the vent. Eleven tephra samples from these layers, 7 from within 10 km of the vent and 4 from a distance of approximately100 km, were provided by the USGS for the use in this study.

Step one of this project, the geochemical analysis of the proximal and distal samples, was done to establish a baseline geochemical fingerprint of the eruption, to determine if there were any changes in the composition of the ashes with distance, and to determine if there were any trace element ratios that could be correlated across the multiple layers of the eruption. Ash samples were cleaned of any secondary contaminates and then processed using magnetic and heavy liquid methods to ensure a nearly pure glass phase before microwave-assisted dissolution in strong acids in preparation for ICP-MS analysis.

Step two of the project seeks to simulate even further distances from the vent by breaking each of the distal samples into four sieve fractions to represent increasingly smaller size fractions as might be expected to be found at increasing distance from the vent. ICP-MS analyses of these ash samples will be conducted and the results compared between the individual size fractions and the original sample, and between the sieved samples and the proximal samples. Should there be a noticeable shift in the geochemical fingerprint between the different size fractions and locations it would imply that present methods for the correlation of ash deposits to specific volcanic eruptions may be limited by distance and thus would require a new method that takes this shift into account.