An overview of the velocity and attenuation distribution in the Krafla volcanic system

Natural resources, such as geothermal energy, are usually allocated in the uppermost part of the crust. The exploration and exploitation of these resources rely on a proper understanding of the rock’s physical and rheological properties. For instance, the reduction in the rock’s permeability with depth depends on the progressive porosity decrease, due to the lithostatic pressure increase and the transition from brittle to ductile deformation (BDT), related to the rising temperature. Therefore, the characterization of underground conditions is crucial for planning explorative studies in geothermal systems. One way to retrieve subsurface information is through the analysis of the propagation of seismic waves, which provides information on physical rocks’ behavior and an alternative assessment of the BDT depth [1]. In particular, the decay of the amplitude of the seismic waves (i.e. seismic attenuation), usually described by a “quality factor”, depends on the seismic frequency, temperature, water content, and grain size of the rocks. Depending on the seismic scale, it could be used as an indicator of subsurface heterogeneities. In this study, we investigate the seismic velocity and attenuation sensitivity to the crustal heterogeneities in the volcanic system of Krafla, an area affected by young tectonics and hot thermal conditions. To this aim, we implement a 𝑄 seismic tomography, using as input a published seismic tomography model [2] and estimate the seismic quality factors of basalt rocks sampled in the Krafla area. To retrieve the 𝑄𝑃 perturbations, we implement a method consisting of a combination of a spectral decay technique to derive the attenuation operator (𝑡∗) and seismic tomographic inversion [3]. The distribution of seismic wave velocities is obtained from a 3D tomographic inversion, using 1453 earthquakes detected from a local seismic network (2009-2012) [2]. 𝑄𝑃 inversion is performed with the simul2014 algorithm [3], while a linearized technique solves a nonlinear problem that uses a damped least-squares inversion for model perturbations. We obtain a map of 𝑄𝑃 variations for the first 4 km, which we jointly interpret with the seismic wave velocities [2]. In this way, we can discriminate between temperature anomalies and compositional heterogeneities. We also test the possibility to detect the BDT depth on the base of the reduction of the 𝑄𝑃 , related to hot temperatures/melt conditions. In order to estimate the seismic quality factors at the laboratory scale, we first develop a numerical method to extract seismic attenuation information from ultrasonic signals in the frequency domain (Spectral Ratio Method). The technique is implemented and tested on suitable synthetic attenuated models [4]. The aim of this test is to assess the reliability and robustness of the method in diverse signal noise conditions and absorption levels. Additionally, the test helps identify the frequency ranges satisfying the assumption of constant seismic quality factor and assess the influence of the windowing effect on providing an accurate seismic quality factor. Subsequently, the method is applied to real ultrasonic waves acquired in the basalts during the execution of a hydrostatic test. The results help constrain the 𝑄𝑃 variations previously obtained. This study will contribute to understanding the dynamics of the tectonic features and help plan explorative investigations of high enthalpy geothermal systems, adding constraints to the correlation between viscous rocks’ deformation and their seismic attenuation.