In-situ measurement of tephra deposit load based on a disdrometer network at Sakurajima volcano, Japan

Abstract

In recent years, optical disdrometers have been used to observe tephra sedimentation at several volcanoes, but a method for calibrating disdrometer observations to accurately match corresponding samples has yet to be determined. In this study, tephra sedimentation samples were taken and disdrometer measurements were made simultaneously for more than 100 eruptions at Sakurajima volcano. Collected tephra samples were sieved and classified into two groups, larger than or smaller than 0.25 mm, the assumed detection threshold of the disdrometer used in this study. A comparison between disdrometer observations and collected samples revealed that particles smaller than 0.25 mm were detected when they formed aggregates or when many particles fell close enough together to be falsely registered as a single particle, even though they were individually smaller than the detection threshold. Two particle groups can be distinguished by their effective densities (assuming spheroid particles). Using samples collected during 44 collection periods (which could each consist of multiple eruptions), the tephra deposit load per particle for each combination of diameter and settling velocity classified by the disdrometer was calculated using multivariate linear regression. Compared to simpler approaches the conversion formulas presented here were found to lead to more accurate estimates. The tephra deposit load values estimated using this method for real-time simultaneous observations were constrained to be within two and six times the sample load. Although the formula is developed based on data from Sakurajima volcano, it can be applied to other volcanoes with similar activity and expected tephra morphology and the methodology presented here can be replicated to produce a tailored formula given enough input data.