Retrieval of tephra size spectra and mass flow rate from C-band radar during the 2010 Eyjafjallajökull eruption, Iceland

• Published in: IEEE Transactions on Geoscience and Remote Sensing
• Main Authors: MEREU, LUIGI, MARZANO, FRANK SILVIO, MONTOPOLI, MARIO, Bonadonna, C.
• Other Authors: Mereu, Luigi, Marzano, FRANK SILVIO, Montopoli, Mario
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2015
• Subjects: Radar data; tephra; volcanic eruption; Eyjafjallajokull; Eyjafjallajökull; Iceland
• Online Access: http://hdl.handle.net/11573/763383; https://doi.org/10.1109/TGRS.2015.2427032

The eruption of the Eyjafjallajokull volcano in April-May 2010 was continuously monitored by the Keflavik C-band weather radar. The Keflavik radar is located at a distance of about 156 km from the volcano vent, and it has sensitivity of about -5 dBZ at 2-km range resolution over the volcanic area. The time series of radar volume data, which was available every 5 min, is quantitatively analyzed by using the Volcanic Ash Radar Retrieval (VARR) technique. The latter is a physically based methodology that is applied to estimate ash-fall rate and mass concentration within each radar volume. The VARR methodology is here extended, with respect to the previous formulation, to provide an approximate estimate of both mean particle diameter and airborne tephra particle size distribution under some assumptions. Deposited tephra at ground is also extrapolated together with an estimate of the magma mass flow rate (MFR) at the volcano vent, derived from the implementation of the mass continuity equation in the radar reference system. The VARR-based retrievals are compared with those derived from a direct tephra sampling at the ground, experimentally carried out in terms of ash grain size and loading during the Eyjafjallajokull eruption activity on May 5-7, 2010. VARR-based particle diameter estimates may suggest that a sorting of airborne particles during the downwind transport is taking place without observing aggregation processes during the ash fall. VARR-derived daily ash mass loadings in the period between April 14 and May 10 are also evaluated with respect to integrated ground and model-based data in the Eyjafjallajokull area. VARR-retrievedMFRs are finally compared with corresponding values obtained from analytical 1-D eruption models, using radar-estimated plume height and radio-sounding wind fields. A fairly good agreement is obtained, thus opening the exploitation of weather radar retrievals for volcanic eruption quantitative studies and ash dispersion model initialization.