One of the primary challenges in magnetotelluric data processing is the presence of noise and outliers (anomalous values). These disturbances often come from human-made sources such as power lines, electronic devices, and nearby infrastructure. They can significantly affect the results related to apparent resistivity and phase, leading to unreliable models of subsurface electrical resistivity.To identify and remove these outlier and noisy components, the Mahalanobis distance method is proposed as an effective solution. This approach — applied here in a four-dimensional feature space comprising the real and imaginary parts of two components of the impedance transfer function — calculates the distance of each data point from the dataset, meanwhile accounting for variances, covariances, and correlations between variables, thereby enabling the detection of anomalous points more accurately than simpler (2D) approaches.In this study, to identify outliers, we applied the Mahalanobis distance method to real data from a single MT station located at the Chassenon Forage site within the Rochechouart impact structure, France. The results demonstrate that this approach not only improves the accuracy of subsequent analyses and enables the extraction of more precise information from subsurface structures, but also reduces processing time by efficiently eliminating contaminated windows before final impedance estimation.

Altered crystalline catchments are complex to study and model, as they present multi-scale properties that control their hydrogeological behaviour and that are difficult to capture through a single geophysical imaging technique. Several volumes of interest must be sampled in order that both small-scale (porosity, layering) and large-scale (bedrock, weathering, faults) heterogeneities can be captured. We propose a geoelectrical model of the Strengbach catchment (Vosges Mountains, France), aiming at identifying the weathered structures and hydrogeological functioning of the aquifer. This is achieved through electrical resistivity tomography (ERT) and Controlled-Source Audio-Magnetotelluric (CSAMT) measurements and the use of appropriate measurement set-ups. Meters-scale shallow contrasts in the top soil, catchment-scale shallow contrasts (top 30 m), and large-scale vertical contrasts (up to 150 m) were resolved through this methodology. A structural interpretation is proposed, based on information provided by borehole measurements (gamma ray, optical images), analysis of sampled waters, and geological mapping. The limits at depth of the weathered and fractured granite, not detected by ERT, are detected by CSAMT. The analysis showed that the weathering state of the granite controls, at first order, the electrical resistivity signal. Shallow geoelectrical signal (first 30 m) is particularly driven by surface conductivity and hence by the clay content, whereas deep geoelectrical signal may arise from both the ionic content of pore waters and the clay content. A structural model is proposed and discussed. Geoelectrical contrasts revealed several qualities of weathered saprolite between the northern and the southern slopes. The inferred structural model and the distribution of weathered and unweathered crystalline units are considered for their respective effect on the hydrogeology, leading to the proposition of a new hydrogeological conceptual model of the catchment.

Space and time variability of water content in aquifers are fundamental issues to understand complex interactions taking part in the critical zone, such as land use and irrigated agricultural production. Fundamental parameters on aquifer behavior are commonly monitored through hydrogeological methods, such as piezometric levels and pumping tests in boreholes. Precisions on the water quality and residence time are provided by geochemical analyses of samples collected in surface streams and boreholes. Several studies showed how additional data can be obtained from non-invasive hydrogeophysical methods, that reveal structural heterogeneities of hydrogeological parameters filling the gaps between boreholes.We carried out a multimethod geophysical survey in the Berambadi experimental catchment (India) which is part of the M-TROPICS CZO (Multiscale TROPIcal CatchmentS Critical Zone Observatory). Two surveys including seismic, electrical, and electromagnetic methods have been repeated for contrasting piezometric levels (high in December 2019, low in May 2022) corresponding to contrasted water contents. We considered time-lapse imaging using electrical resistivity tomography (ERT) and audio-magneto-tellurics (AMT), which sensitivities apply at complementary scales. Changes in the electrical resistivity from ERT shallow cross-sections and deeper jointly inverted ERT-AMT vertical profiles are compared for the two seasons. Results are discussed in terms of water content and porosity of the regolith as well as uncertainties caused by inherent repeatability issues of time-lapse measurements. Final discussion concerns perspectives of combined time-lapse electrical and seismic velocity models to assess the impact of the spatial variability of regolith properties at the catchment scale.

<p class="x_x_MsoNormal" align="left"><span lang="EN-US">Water transfer through the unsaturated zone, in terms of upward or downward water fluxes, is a critical term for estimation of the water budget. As fluid flow modifies diffusive heat transfer through advective processes, since the early 90s several studies have attempted to deduce vertical water flow from soil temperature series. Likewise, if information on the water content profiles is known, bulk thermal properties can be inferred from thermal time series at different depths.</span></p> <p class="x_x_MsoNormal" align="left"><span lang="EN-US">In this study we compare two field sites in the Paris Basin Area, with two different types of soil and vegetation. We present our preliminary results from two approaches aiming at retrieving inferring soil bulk thermal parameters, namely heat capacity and conductivity, as well as vertical water flow.</span></p> <p class="x_x_MsoNormal" align="left"><span lang="EN-US">On the one hand, thermal measurements until a depth of 1.8 m have been carried out in a managed crop field. Using frequency decomposition of the thermal series, the upward and downward flows are determined. The water fluxes are compared with high-frequency EM time-lapse maps in an attempt to spatialize the variations.</span></p> <p class="x_x_MsoNormal" align="left"><span lang="EN-US">On the other hand, the thermal properties of a wetland area are inferred from soil thermal time series inversion using the thermo-hydrodynamic code suite Ginette, and are compared with spatial distribution of vegetation derived from remote sensing imagery.</span></p> <p><span lang="EN-US">The two approaches are compared and discussed with their respective caveats and abilities.</span></p>

The detection of water leakage along its transportation network has important societal impacts, such as avoiding a large volume of water wasted along the waterways or preventing water-related chemical or physical surrounding media deterioration. Among the vast domain of destructive techniques, Ground-Penetrating Radar (GPR) is a common and efficient tool used for detection in many near-surface contexts, and it is particularly efficient in civil engineering cases, such as utility detection, due to its fine resolution and the ease of data acquisition. A peculiar form of signal enhancement appears in GPR profiles recorded over spheres and cylinders where velocity contrasts exist between the body’s material and the surrounding medium. We used this enhancement to detect potential water leakages in water pipes. After exhibiting the signal enhancement effect in a laboratory sandbox experiment using a spherical glass ball, we verified the results with numerical experiments with varied sphere and cylinder sizes and dielectric properties. We then investigated field and numerical experiments of GPR transects above a “real life” water-leaking PVC pipe. Our results show that the water cylinder and water infiltration bulb produced a characteristic signal that could be used for detecting water leakages along water pipes. The largest amplitude in the GPR signal is caused by a bottom pipe reflection enhanced by the water bulb and not by the top of the pipe. We stress the risk of miscalculating the pipe’s depth during velocity estimation when amplitude enhancement conditions are met. Beyond civil-engineering impacts, knowledge on signal amplification phenomena can help GPR data interpretations in sedimentology and hydrogeology studies.

Abstract The 205 Ma Rochechouart impact structure (France) is characterized by various impactite formations overlying the Hercynian crystalline basement. New constraints from downhole logging and surface electrical resistivity measurements along >100 m long profiles reveal that the top melt‐bearing breccia layer is more conductive and porous than the underlying melt‐poor breccia layer. The stratigraphy within the impactite and the transition with the basement are irregular at small (∼1–10 m) and medium (>100 m) scales, with vertical amplitude up to 40–50 m. At larger scale (>1 km), audio‐magnetotelluric observations are able to map the lateral and vertical extent of fracturing/brecciation in the basement, reaching 200 m below the surface nearby Chassenon, in the northern part of the structure. Our results also unveil that the impactite deposits and the brecciated basement of the Rochechouart impact structure may have been shifted laterally and vertically during the modification stage of the impact event through displacements of megablocks, which may be associated with the collapse of a central uplift.

The remediation of a polluted site relies, as a first stage, on the proper delineation of the contamination sources. In classical investigations, soil and water samples are collected throughout the field. These measurements allow a quantitative characterization of the gathered materials but only provide information about the medium in the vicinity of the points where they were collected. On the other hand, geophysical techniques can provide a quasi-continuous coverage of the investigated field. This paper describes a geophysical survey that was performed on an industrial site impacted by a chlorinated DNAPL. The precise location of the contamination was needed for the treatment of the saturated zone, while the unsaturated zone was remediated by general excavation of the sediments, followed by separate treatment. As this excavation allowed to get closer to the saturated zone, geophysical measurements were conducted at the bottom of the pit. Whereas Electrical Resistivity Tomography measurements only brought little information, Ground Penetrating Radar drew the remediation operations towards an area that preliminary point measurements had not identified as a possible source location.

The Lodève landslide is a slow moving (3 to 4 mm/yr) and deep (60 m) rotational instability, located in the South-East of France, 60 km North from Montpellier (Hérault department). It is located in the Lodève basin, a set of connected steep head valleys marking the southern limit of the karstic Larzac plateau, and particularly prone to hydraulically triggered landslides. The unstable slope was progressively formed by the erosion of the upper limestone and sandstone units. The local tectonics build up resulted in a series of vertical North/South faults and fissures, allowing the water to infiltrate down to the deeper Triassic clay and evaporite layers. During heavy rainfall events, an amount of the meteoritic water infiltrates along these flow paths, down to the clay and evaporite layers from the Norian and Rhaetian era, leading to the rapid recharge of the units, the onset of high pressure in the confined layers and the decrease of the cohesion of the rock material and of the shear strength. The Controlled Source Audio-frequency Magneto-Telluric (CSAMT) method is a low-impact, non-invasive active frequency domain electromagnetic sounding technique, deriving from the Magneto-Telluric (MT) method. An electromagnetic signal is produced a few km away from the studied site, and the electric and magnetic transfer functions of the plane wave signal are recorded at multiple frequencies, permitting the computation of far-field MT impedance tensor. CSAMT is characterized by a good vertical resolution and large depths of investigation, but poor sensitivity to the first tens of meters. For these reason, it is expected to be a good candidate method to conduct time-lapse studies in the context of pseudo-1D layered subsurface. CSAMT data were acquired at the landslide from November 2018 to March 2019 at 8 different stations. The landslide is assumed to be a pseudo-1D medium with a tilted flat surface topography. The aim was to observe the variations of electrical resistivity related to the hydrogeological response to the heavy rains observed during the monitoring period. Sensitivity tests were realized with the software custEM. Measurements were taken at ten fundamental frequencies ranged from 510 to 9600 Hz with a Phoenix’s System-2000.net equipment and were repeated every months except in February. The data quality is uneven from one station to another next. Most station showed significant variations in apparent resistivity. The observed variations were interpreted in a one-dimensional context, revealing lateral variations in the hydrogeological response of the slide. Complementary TDIP and DC data and high temporal geochemical and geophysical monitoring of properties at two boreholes were used to constraint the CSAMT interpretation.

Introduction: The Rochechouart impact structure, located nearby Limoges in France, corresponds to a ~100 km2 area showing some remains of impact breccia formations overlying the Hercynian crystalline basement. The geological mapping already revealed a variety of breccia compositions and a complex stratigraphy, which is confirmed by the preliminary analyses of the cores sampled from a recent drilling campaign [1]. The age of the impact event is now well resolved: 204-207 Ma [2]. However, there are still some uncertainties among which the size of the initial crater (d = 20 km ? 30 km or more ?), its associated morphology (central uplift ? peak-ring? basin?), and/or the occurrence of an impact-generated tsunami, etc.: all these open questions will be investigated by a series of research projects submitted to the CIRIR and mainly focusing on the core samples [1]. Indeed studying Rochechouart helps to investigate the composition, geometry and physical rock properties at the transition between the allochtonous (when preserved) and autochtonous formations of complex impact structures. Objectives and methods: our group aims at revealing the thickness and lateral extension of the breccia lenses, as well as to map the impact effects on the basement formations. In parallel to and after the drilling campaign, we mainly investigated the electrical properties of the rocks from the Chassenon area (NW part of the remaining breccia lenses) in the drill holes (electrical resistivity downhole logging), on subsurface (2D electrical resistivity tomography) and at kilometer scale (audio-magnetotelluric, AMT). Preliminary results: by downhole logging in the SC2 Chassenon drill hole, 4 electrical units were defined [3]: (1) the top 40 m with resistivities are between 80 and 100 Ohm.m; (2) between 40 and ~85 m where resistivities range from 90 to 300 Ohm.m; (3) below ~85 m, resistivities are always greater than 200 Ohm.m, slightly increasing to 1000 Ohm.m at 110-115 m; (4) after a sharp decrease nearby 115 m, very large resistivities up to ~3000 Ohm.m are observed. These 4 electrical units well correlate with the SC2 lithologies defined from cores: Unit 1 (0-40m) – melt-rich suevite; U2 (40-88m) – melt-poor suevite; U3 (88-114m) – brecciated/fractured gneiss; U4 (114-120m) – gneiss [1]. Using electrical resistivity tomography along a 630 m profile centered on the SC2 hole, the vertical resistivity contrasts show a similar, but smoothed, stratigraphy/architecture (Figure 1), while it evidences that the suevite/gneiss transition depth laterally varies (amplitude of ~50 m). Then, at larger scale, the preliminary results of the AMT measurements reveal that, deeper than 500-600 m, the gneiss appears more resistive than 2000 Ohm.m, meaning that it is probably unaffected by the impact at this depth. We will show similar preliminary results for other drill holes and breccia lenses.

Atmospheric electromagnetic waves created by global lightning activity contain information about electrical processes of the inner and the outer Earth. Large signal-to-noise ratio events are particularly interesting because they convey information about electromagnetic properties along their path. We introduce a new methodology to automatically detect and characterize lightning-based waves using a time-frequency decomposition obtained through the application of continuous wavelet transform. We focus specifically on three types of sources, namely, atmospherics, slow tails and whistlers, that cover the frequency range 10 Hz to 10 kHz. Each wave has distinguishable characteristics in the time-frequency domain due to source shape and dispersion processes. Our methodology allows automatic detection of each type of event in the time-frequency decomposition thanks to their specific signature. Horizontal polarization attributes are also recovered in the time-frequency domain. This procedure is first applied to synthetic extremely low frequency time-series with different signal-to-noise ratios to test for robustness. We then apply it on real data: three stations of audio-magnetotelluric data acquired in Guadeloupe, oversea French territories. Most of analysed atmospherics and slow tails display linear polarization, whereas analysed whistlers are elliptically polarized. The diversity of lightning activity is finally analysed in an audio-magnetotelluric data processing framework, as used in subsurface prospecting, through estimation of the impedance response functions. We show that audio-magnetotelluric processing results depend mainly on the frequency content of electromagnetic waves observed in processed time-series, with an emphasis on the difference between morning and afternoon acquisition. Our new methodology based on the time-frequency signature of lightning-induced electromagnetic waves allows automatic detection and characterization of events in audio-magnetotelluric time-series, providing the means to assess quality of response functions obtained through processing.

24 th EM Induction Workshop, Helsingør, Denmark, August 12-19, 2018 2 / 4 THEISTAREYKIR

Summary Magnetotelluric (MT) is classically used in geophysical exploration for imaging electrical conductivity structures and is being developped as a monitoring technique. In geothermic during fluid injections and stimulation experiments, MT is used in addition to microseismic observations and can provide critical information to geothermal fluid flows because the electrical conductivity is related with temperature, porosity, water content and minerals of rocks. Some experiments have shown that such MT signals might be difficult to observe because they are at periods of 1–10 s, within the MT dead-band. We show actual data from Northern France then consider the sensitivity of MT monitoring by forward modelling. We use ModEM open source code to build a 3-dimensional model which includes topography and simple sedimentary conductive layers. Modelling allows us to simulate different changes that could be caused by brine and/or acid injection within fractures at depth and show subsequent MT monitoring parameters, particularly the phase tensor. From these models, it seems that MT monitoring in a sedimentary environment at 20 □.m could be sensitive to an increase of conductivity in a fault area at geothermal depths of 2–3 km if the size of the disturbed domain reaches about 10×0.3×2 km3.

Summary Audio-Magnetotellurics (AMT) is an electromagnetic geophysical method based on passive measurements of the induced electric currents in the ground by atmospheric sources which mostly originate from global lightning activity. Each lightning strike generates different waves with distinct time-frequency properties. Two major difficulties arise in AMT acquisition and processing. The first one lies in the relatively low signal-to-noise ratio of natural signals compared to anthropogenic signals. The second one is the so-called AMT dead-band, a specific frequency band (generally from 1 kHz to 5 kHz) where the level of natural signals energy remains low. Using the continuous wavelet transform, we identify electromagnetic (EM) waves in the time-frequency plane and we then invert the response function. Two criteria are used for detection: high signal-to-noise ratio and specific shape of local maxima in the time-frequency plane. The determination of AMT response functions are based on a hierarchical bootstrap scheme. We illustrate this methodology on AMT data acquired near Chambon-La-Forêt magnetic observatory. By using the new procedure on this dataset, we are able to, on the one hand, drastically reduce the AMT dead-band width, and on the other hand, greatly reduce the confidence interval of the AMT response functions.

Summary Far from its sources and in the presence of conductivity variations in the ground, the vertical and horizontal components of the VLF magnetic field can be linked by a linear equation: H_z=AH_x+BH_y. The components A and B are called the tipper vector and inform us about the ground conductivity changes. By mounting an ADU data logger and a SHFT antenna from Metronix geophysics on a homemade trolley, we investigate the possibility of profiling and mapping with the tipper vector. We carried out a survey with seven profiles on a site next to Reims in France. The low electromagnetic pollution of the area allow us to record many VLF frequencies (nine) and to run many tests, considering the direction of the various VLF sources. Finally, after choosing two frequencies, we calculated the tipper vector on the whole survey area. The first results show a good spatial resolution and a strong capability of detection. We were able to precisely map a buried pipe. Moreover, the comparison with a magnetic map showed a very good fit between the data.

An increasing interest in magnetotelluric monitoring of hydraulic stimulation experiments and of natural earthquakes currently requires a detailed investigation of the limits of this method. Our study contributes to this discussion with a unique quasi -continuous and long-term magnetotelluric monitoring of different injection and production experiments at the Rittershoffen geothermal site in Alsace (France). Here, we provide a first evaluation of the data and an outlook on more specific challenges. The geothermal doublet at Rittershoffen is operated through two wells GRT1 and GRT2. The magnetotelluric monitoring covers the end of drilling phase of GRT2, mostly, production from, but also injection into this well, injection into GRT1 and a circulation experiment. Magnetotelluric data were processed successfully using remote referencing. Transfer functions show particular variation pattern for different operations, i.e. an increase in uncertainty, conductivity and phase during test operation with a preferential direction sub-parallel to Shmin, i.e. perpendicular to the expected extension of the fractures controlling the reservoir. In particular fluid injection, either into GRT2 or GRT1 causes a strong decrease in resistivity by up to one order of magnitude in the YX component between about 8-25 s of period. However, the comparison between the temporal distribution of the occurring anomalies and the magnetic field intensity reveals an amplifying effect on the anomalies. In conclusion, we confirm earlier findings on the influence of the magnetic field intensity on magnetotelluric monitoring and at the same time show the lower limit of monitoring in particular monitoring of injection into deep geothermal reservoirs.