Near-surface effects in seismic wavefields at Krafla volcano (NE Iceland): characterization and mitigation.

Seismic imaging of small-scale geological features in volcanic regions is challenging due to the often heterogeneous subsurface, causing extensive wave scattering and limited coherence in the wavefields. A well-known example that illustrates this problem is the unexpected encounter of magma at 2.1 km depth at Krafla (NE Iceland) during geothermal drilling in 2009. Despite numerous geophysical studies, the magma body remained undetected prior to drilling. In the summer of 2022, we deployed ~100 short-period seismic nodes in a reflection seismic configuration at Krafla. Using the known reflector as a guide, our goal is to investigate and enhance seismic imaging in complex geological settings. Analyses of ~300 local earthquakes (magnitudes < 1.5) show that the wavefields at Krafla are largely dominated by wave scattering and reverberations within the uppermost ~100m of the subsurface, causing little coherency in the data even among neighbouring stations spaced at 30 meter intervals. Using auto – and cross-correlation techniques, we address the reverberations and construct transfer functions characterising the seismic response of the sites at each station. This is followed by time-dependent deconvolution. The deconvolved wavefields show increased coherency, as the influence of the near-surface could be considerably reduced. Coherent phases emerge in the wavefields which were previously obscured by reverberating waves. Different imaging techniques such as common-depth-point (CDP) binning and stacking will be applied to the cleaned wavefields in order to resolve potential layer boundaries and magma pockets, ultimately improving our understanding of the Krafla geothermal system.