Comparing single-particle analysis data of volcanic ash of the 2010 Eyjafjallajökull eruption obtained from scanning electron and light microscope images

Abstract

This study examines the influence of different analysis techniques on the results concerning the particle size distribution and mineralogical composition of airborne particles. Volcanic ash from the Eyjafjalla volcano was examined with a transmitted light microscope (TLM), a scanning electron microscope (SEM), SEM equipped with an automatic single-particle analysis program, and an electron microprobe (EMP) and the results differ considerably. Main error source is the specific particle size determination for each method. Here, a sample containing airborne particles of volcanic material was analyzed for size-differentiated particle number (n) on the evaluation area per unit time (number settling rate). The results of the TLM reveal a higher abundance of particles ≥8 μm than those obtained by manual SEM and EMP analysis, whereas the opposite is observed for particles <8 μm. When using SEM combined with automated single-particle analysis the number of identified particles is reduced compared to the manually obtained values. As error sources we identified uneven surfaces for SEM and EMP, the identification of transparent and light-colored particles under TLM, and the setting of brightness and contrast. The consequence of these discrepancies is that the TLM analysis leads to overestimation of feldspar, whereas the SEM and EMP techniques overestimate the abundances of pyroxene, quartz and other minerals in the studied sample.