Analysis of ballistic waves in seismic noise monitoring of water table variations in a water field site: added value from numerical modelling to data understanding

Abstract

SUMMARY Passive seismic interferometry allows to track continuously the weak seismic velocity changes in any medium by correlating the ambient seismic noise between two points to reconstruct the Green’s function. The ballistic surface waves of the reconstructed Green’s functions are used to monitor the changes of water table induced by a controlled experiment in the Crépieux-Charmy (France) exploitation field. Viscoelastic numerical modelling of the monitoring experiment reproduces quite satisfactorily the sensitivity of the surface waves to the water table previously observed with seismic noise data. This numerical approach points out that this sensitivity is controlled by mode mixing of Rayleigh waves. It also made it possible to identify the refracted P wave and to extract its anticorrelated sensitivity to water table variations. Depending on the offset between receivers, it was observed numerically that the interferences between the different waves (with different velocities) composing the seismic wavefield slightly affect the quantitative sensitivity to water table changes. This suggests the use of an optimal spatial and temporal observation window for which wave interference is minor and does not blur the quantitative response to water table variations. We were thus able to determine the relationship between velocity and water table variations for all waves involved. From numerical computations, we identify a weak signal-to-noise ratio phase in the noise correlograms, with a anticorrelated sensitivity to the water table: the reconstructed refracted waves.

A Multi‐Hydrogeophysical Study of a Watershed at Kaiwi Coast (Oʻahu, Hawaiʻi), using Seismic Ambient Noise Surface Wave Tomography and Self‐Potential Data

“…In recent years, the value of ambient noise tomography for studying high‐resolution near‐surface structures has been recognized (Chmiel et al., 2019; de Ridder & Dellinger, 2011; Lin et al., 2013; Mordret et al., 2013; Picozzi et al., 2009), but applications for hydrological purposes are still very scarce. Most hydrological studies thus far have targeted temporal monitoring of groundwater resources (Clements & Denolle, 2018; Garambois et al., 2019; Lecocq et al., 2017; Mainsant et al., 2012; Sens‐Schönfelder & Wegler, 2006; Voisin et al., 2016).…”

Protecting the Vital Groundwater Sources of Oceanic Islands

“…Seismic noise can also be used for monitoring purposes, such as water table variations (Garambois et al 2019). Self-potential signals have been successfully used to delineate contamination plumes, for example, based on the redox potential (Revil & Jardani 2013).…”

“…These data can also be integrated with self‐potential data to identify and interpret preferential groundwater or contamination plumes as controlled by the geological subsurface structure (Grobbe et al 2021). Seismic noise can also be used for monitoring purposes, such as water table variations (Garambois et al 2019). Self‐potential signals have been successfully used to delineate contamination plumes, for example, based on the redox potential (Revil & Jardani 2013).…”

Advancing measurements and representations of subsurface heterogeneity and dynamic processes: towards 4D hydrogeology

“…A grand challenge of subsurface imaging methods for dynamic hydrogeological processes (Figure 1) is to deal with systems characterized by pronounced structure and process heterogeneity, including preferential flow paths, evolving properties or geometry, unsaturated flow processes, fluctuating redox conditions, and multifunctional microbial communities. Recent breakthroughs in hydrogeophysical imaging techniques (Binley et al, 2015;Singha et al, 2015) and the emergence of interdisciplinary approaches combining new sensors such as fiber optics (Bense et al, 2016;Zhan, 2020), new experimental methodologies like ambient seismic noise correlation (Garambois et al, 2019) and coupled modelling techniques (e.g., Hinnell et al, 2010;Jardani et al, 2013;Linde and Doetsch, 2016) may profoundly change our vision and representation of the dynamics of processes that take place in these environments (Binley et al, 2015;St. Clair et al, 2015).…”