G32A-03
The 2010 Eyjafjallajökull and 2011 Grimsvötn ash plumes as seen by GPS

Wednesday, 16 December 2015: 10:50
2002 (Moscone West)
Ronni Grapenthin, New Mexico Institute of Mining and Technology, Socorro, NM, United States
Abstract:
The injection of a volcanic plume introduces a dynamic, localized, short-term heterogeneity in the atmosphere. Satellite-imagery based remote sensing techniques provide good spatial coverage for the detection of such plumes, but slow satellite repeat times (>30 minutes) and cloud cover can delay, if not entirely prevent, the detection. GPS, in turn, provides excellent temporal coverage, but requires favorable satellite-station-geometry such that the signal propagates through the plume if it is to be used for plume detection and analysis. Two methods exist to detect / analyze ash plumes with GPS: (a) Ash-heavy plumes result in signal dispersion and hence a lowered signal-to-noise ratio (SNR). A lowered SNR, recorded by some receivers, can provide useful information about the plume, such as location and velocity of ascent. These data can be evaluated directly as they are recorded by the receiver; without the need of solving for a receiver’s position. (b) Wet plumes refract the GPS signals piercing the plume and hence induce a propagation delay. When solving for a receiver position GPS analysis tools do not model this localized phase delay effect and solutions for plume-piercing satellites do not fit the data well. This can be exploited for plume analysis such as the estimation of changes to the atmospheric refractivity index.

We analyze GPS data of the ~2 month 2010 Eyafjallajökull erption and the week-long 2011 Grímsvötn eruption to infer a first order estimate of plume geometry and its progression. Using SNR and phase delay information, we evaluate the evolution of the partitioning of wet versus dry parts of the plume. During the GPS processing we iteratively solve for phase-delay and position and fix other parameters, hence reducing the mapping of least-squares misfit into position estimates and other parameters. Nearly continuous webcam imagery provides independent observations of first-order plume characteristics for the Eyafjallajökull event.