Photometric observations of Mt. Etna's different aerosol plumes

Abstract

We have investigated Mt. Etna's summit crater's plumes using a visible and NIR sun-photometer during eight days in July 1999. After removal of the background optical depth, we have applied the Ångstrom equation and a King-type inversion to the volcanic aerosol spectral optical depths to retrieve (1) the Ångstrom coefficients α and β, (2) the particle size, surface area and volume spectra and (3) the effective radius. Plumes supposed to contain ash have a larger effective radius (1.48±0.28 μm vs. 0.68±0.18 μm), a smaller Ångstrom exponent α (−0.02 vs. 1.63) and a more uni-modal Junge-type size distribution. Inversions of the spectral optical depth to size distributions are observed to be little sensitive to small changes in the assumed value of the refractive index. Plumes measured further downwind seem to have higher effective radii (1.68 vs. 1.48 μm) and a more distinct bi-modal type distribution, but the difference lies within experimental error.

Introduction

Volcanic aerosols can consist of a variety of different substances, in both the solid and liquid phase, including volcanic ash, solid sulphate particles, water and dilute acid droplets, with a range of 10⁻⁸–10⁻² m in diameter. They transport information useful in understanding volcanic processes from the magma to the atmosphere (Ammann et al., 1992). They are radiatively active (Lacis et al., 1992), play a central role in plume chemistry and the uptake and deposition of volatiles (Martin et al., 1986; Oppenheimer et al., 1998) and may distort current remote sensing measurements of volcanic gases (Edner et al., 1994). Some volcanic aerosols, typically fine ash, have serious environmental health impacts (Baxter et al., 1999). Although particle size is a key determinant of all these effects volcanic aerosol size distributions are poorly quantified.

Previous studies of volcanic aerosol have tended to focus on large-scale emissions into the stratosphere (e.g. Langford et al., 1995; Borrmann et al., 1995), with particular reference to their effect on global climate (e.g. Graf et al., 1993; Kerr, 1994). The effective radius (R_Eff) is of critical importance when determining the radiative forcing of stratospheric volcanic aerosols. From modelling, Lacis et al. (1992) indicate that a particle size distribution in the stratosphere with an effective radius of >2 μm will have a positive radiative forcing as absorption of outgoing terrestrial radiation exceeds the increase in albedo caused by particles of <2 μm. In contrast, the radiative effects of tropospheric volcanic aerosols are little modelled and poorly quantified.

Current analysis of volcanic gases with instruments such as COSPEC (Stoiber et al., 1983) does not take into account scattering effects within the plume. With the increased use of remote sensing instrumentation at volcanoes, including FTIR spectroscopy (Francis et al., 1998) volcanic gases are being measured further downwind. The further downwind a gas measurement is taken the more the magmatic signal is affected by interaction with the atmosphere, through different deposition, adsorption and solution rates of the different component gases. These processes are controlled by multi-phase chemistry involving aerosols, which, again, is currently poorly understood (Ravishankara, 1997). The aerosol size distribution is an important factor in understanding the processes of absorption/adsorption, coagulation and reaction chemistry, particularly in the aqueous phase, present within a volcanic plume.

There have been relatively few studies of tropospheric volcanic aerosol size distributions (Table 1). Direct sampling of volcanic aerosols using airborne sensors yields information about particles across a large radius range (0.01–>25.0 μm) but is difficult, dangerous and expensive. Another method has been direct ground-based measurement of ultrafines at Mt. Etna combining photoelectron-charging (PCP) techniques to determine the photoelectric activity and direct sampling using filters to calculate the particle size distribution using a transmission electron microscope (TEM). However, this method requires delicate and time-consuming post acquisition analysis and cannot offer a real-time measurement capability for size determination.

Remote measurements may offer a solution to real-time monitoring of volcanic aerosol particle size distributions. Remote measurements of stratospheric aerosols from El Chichon (Asano et al., 1985) and Pinatubo (Asano et al., 1993) have been made using sun-photometry and details are included in Table 1. We have taken the well-known technique of sun-photometry, previously used solely for investigation of stratospheric volcanic aerosol, and applied it to the investigation of young, near-vent volcanic aerosol. This study extends a preliminary trial (Watson and Oppenheimer, 2000) in which near-vent ash-free volcanic plumes from Mt. Etna, Sicily, were measured.

We aimed to measure tropospheric volcanic aerosol using sun-photometry in order to retrieve optical properties of Mt. Etna's aerosol plume, including the Ångstrom coefficients α and β, the particle size, surface area and volume distribution spectra and the effective radius. We took visible and near IR direct solar irradiance measurements through the NE, central and SE crater's plumes, using a solar tracking sun-photometer, to investigate the effects of differing volcanic activity upon the optical properties of the volcanic plume. We have used these measurements to assess these effects on the optical properties, and the subsequently retrieved distribution spectra, of Mt. Etna's aerosol plumes.