1.1 Theoretical background

GOME-2 SO₂ retrieval

SO₂ columns are retrieved using data from the Second Global Ozone Monitoring Experiment (GOME-2) on MetOp-A. GOME-2 is a nadir-scanning UV-VIS spectrometer with a spectral coverage of 240 – 790 nm and a spectral (FWHM) resolution between 0.26 nm and 0.51 nm. It measures the back-scattered radiation from the earth-atmosphere system. In addition, a direct sun spectrum is recorded as reference once a day. The pixel size is 80 km x 40 km. With the normal operation mode near global coverage is achieved at the equator in one day.

SO₂ slant columns are retrieved from the UV backscatter measurements of sunlight using the Differential Optical Absorption Spectroscopy (DOAS) method (Platt 1994) in the wavelength region between 315 – 326 nm. Input parameters for the DOAS fit include the absorption cross-section of SO₂, for which the temperature is adjusted depending on the assumed height of the volcanic SO₂ plume, and the absorption cross-sections of interfering gases, ozone and NO₂. A further correction is made in the DOAS fit to account for the ring effect (rotational Raman scattering).

In the 315 – 326 nm wavelength range used for the retrieval, there is a strong interference of the SO₂ and ozone absorption signals, especially at high solar zenith angles. Therefore, an interference correction is applied to the SO₂ slant column values. The corrected slant column densities of SO₂ are converted to geometry-independent vertical column (VC) amounts through division by an appropriate air mass factor (AMF) as VC = SC/AMF (Valks and Loyola 2008).

For SO₂, the AMF is strongly dependent on measurement geometry, surface albedo, clouds, aerosols, and most importantly, the shape of the vertical SO₂ profile in the atmosphere. For the AMF calculations, an a priori volcanic SO₂ profile is assumed with a predefined central plume height and a Gaussian SO₂ distribution. As the correct plume height is rarely available at the time of measurement, the SO₂ column is computed for three different assumed volcanic plume heights: 2.5 km, 6 km and 15 km above ground level. The lowest height represents passive degassing of low volcanoes, the second height effusive volcanic eruptions or passive degassing of high volcanoes and the third height explosive eruptions. The AMFs are calculated using the radiative transfer model LIDORT (Spurr et al. 2001).

MODIS true colour images

The MODIS (Moderate Resolution Imaging Spectroradiometer) instruments are onboard the two satellites AQUA and TERRA. MODIS scans a swath of 2330 km across track by 10 km along track and provides global coverage every 1 to 2 days. It acquires data in 36 spectral bands between 0.4 µm and 14.4 µm. Bands 1 and 2 are imaged at a spatial resolution of 250 m at nadir, bands 3-7 at 500 m and the remaining 29 bands at 1 km.

The MODIS Rapid Response System makes use of the combination of different MODIS spectral bands to create true- or false-colour images. These combinations of various wavelengths bring out different characteristics of the atmosphere, for land as well as ocean surface. For creating MODIS Rapid Response true colour imagery an algorithm developed by Jacques Descloitres at NASA/GSFC is applied to the Level 1B radiances in order to remove gross atmospheric effects (Gumley et al. 2010). MODIS true colour images are based on the three spectral bands at 670 nm, 565 nm and 479 nm. These wavelengths correspond to the red, green and blue region of the light spectrum and the resulting image is similar to what a human eye would see. Therefore it enables the visual detection of volcanic ash.

Trajectory matching technique

In order to identify the source-receptor relationship the GOME-2 SO₂ observations are stochastically analyzed with the kinematic 3D trajectory model FLEXTRA (Stohl et al. 1999) by releasing ensembles of backward trajectories using the temporal and spatial characterization of the SO₂ observation. The backward trajectories are started at different pressure levels from ground up to 20 km. The starting point is not confined to the centre coordinate of the GOME-2 pixel. It is rather defined as an air mass (e.g. a volume).

Due to the vertical wind shear as a result of the reduction of the friction force with the altitude or changes in pressure gradients, emitted particles and gases are transported in different directions and with various velocities depending on their emission height. If the volcanic source and the eruption time are known, the backward trajectories that pass the volcano’s coordinates can be filtered in dependency of their heights and time resulting in a distribution of the emission height for the moment of the eruption. If the emission source is unknown, the backward trajectories are summed up and normalised for every grid cell. The result of this is a trajectory density map indicating the most probable source of an SO₂ or ash plume. The underlying assumption is that the source-receptor relationship can be definitively determined by taking advantage of coherent trajectories. Kolmogorov-Smirnoff tests show a predominantly correct solution of the source-receptor relationship and that the eruption parameters can be identified accurately enough. Limitations of the technique occur for adjoining volcanoes with a distance lower than 100km and for a small vertical wind shear (Seidenberger 2009).

FLEXTRA is driven by meteorological wind and temperature fields from the numerical weather prediction models of the European Center for Medium Range Weather Forecast (ECMWF) and the National Oceanic and Atmospheric Administration (NOAA) respectively. In order to reconfirm the estimated volcanic source and the appropriate emission height, ensembles of forward trajectories are released at the volcano coordinates for different levels between 0 km and 20 km above mean sea level. By matching these forward trajectories with the SO₂ retrieval of GOME-2 the backward trajectory results can be verified.

As trajectory models deliver rather qualitative information and neglect processes like convection, diffusion and turbulence, the more advanced Lagrangian Particle Dispersion Model (LPDM) FLEXPART (Stohl et al. 2005) is used for validation tasks and further analysis of the satellite observations regarding the quantification of volcanic emission strengths and the investigation of long-range transport processes of volcanic SO₂ and ash (Maerker et al. 2008). FLEXPART has been disposed in many studies of the long-range atmospheric transport also of volcanic ash (Prata et al. 2007). It was validated on the basis of continental-scale tracer experiment data (Stohl et al. 1998).

The LPDM is initialized using source term parameters such as emission source, height, time and duration determined by the stochastic trajectory analysis. The model results provide a three dimensional calculation of the transport of volcanic emissions and are therefore important to address the climate impact of eruptions. They are currently also applied to investigate the predominant loss processes of SO₂ and ash in the atmosphere and to quantify the sensitivity of the GOME-2 SO₂.