Post‐emplacement cooling and contraction of lava flows: InSAR observations and a thermal model for lava fields at Hekla volcano, Iceland. Issue 2 (13th February 2017)

• Record Type: Journal Article
• Authors: Wittmann, Werner; Sigmundsson, Freysteinn; Dumont, Stéphanie; Lavallée, Yan
• Is Part Of: Journal of geophysical research. Volume 122:Issue 2(2017)
• Page Start: 946
• Page End: 965
• Publication Date: 2017-02-13
• Subjects: InSAR -- lava -- Hekla -- thermal -- contraction -- Iceland; Geomagnetism -- Periodicals; Geochemistry -- Periodicals; Geophysics -- Periodicals; Earth sciences -- Periodicals; 551.1
• DOI: 10.1002/2016JB013444
• Languages: English
• ISSNs: 2169-9313

Abstract

Abstract: Lava flows contract as they cool, causing progressive subsidence of the flow surface. Here we study this process by measuring and modeling the deformation of emplaced lava flows and the surrounding substrate. The temporal trend of vertical lava movements was investigated using interferometric analysis of synthetic aperture radar (InSAR) images from the 1991 and 2000 Hekla eruptions, covering periods of 23 and 12 years, respectively. Data from six tracks from three satellites, including both ascending and descending passes, were used to create 99 interferograms, from which trends of accumulated subsidence and subsidence velocities were derived. Subsidence rates are similar for both lava flows and decay approximately exponentially from about 20 mm/yr 5 years after emplacement to about 2 mm/yr 15 years after emplacement. A one‐dimensional, semi‐analytical model was fitted to the observed subsidence rates, with subsidence due to phase change calculated analytically and subsidence due to thermal contraction calculated numerically using dilatometric constraints obtained experimentally. The initial thicknesses of the 1991 and 2000 lava fields, D 1991 and D 2000, scaled thermal expansivity, γ α, and thermal diffusivity, κ, are the crucial parameters influencing lava subsidence and subsidence rate. Inversion for these parameters reveal linear correlations between them. Best fitting results of inversions for D 1991 range from 10 m to 27 m, for D 2000 from 10 m to 30 m, γ α = 9–24 ×10 −6 K −1, and κ = 1–7 ×10 −7 m 2 s −1 . Key Points: Lava subsidence can be derived from InSAR, when observations over lava fields are referenced to areas adjacent to the lava A simple thermal model based on heat conduction can reproduce the InSAR observations on lava fields Combined InSAR observations up to 23 years suggest exponential decay of the subsidence rate of lava fields