A hybrid correlation denoising method based on complex resistivity and application on spread spectrum induced polarization data

In electrical exploration techniques, an effective suppression method for Gaussian and impulsive random noise in spread spectrum induced polarization (SSIP) continues to be challenging for conventional denoising methods. Remnant noise influences the complex resistivity spectrum and damages the subsequent interpretation of geophysical surveys. We present a hybrid method based on a correlation function and complex resistivity, which introduces the correlation analyses between the transmitting source, the measured potential, and the injected current signal. According to the analyses, reliable results for complex resistivity spectra can be calculated, which can be further used for noise suppression. We apply the hybrid method to both numerical and field experiments to process measured SSIP data. Simulation tests show that the hybrid method not only suppresses the two types of noise but also improves the relative error of the complex resistivity spectrum. Field data processing shows that the hybrid method can minimize the standard deviation of the data and possess a greater ability to distinguish adjacent objects, which can improve the reliability of the data in subsequent processing and interpretation.

Complex resistivity inversion using controlled source electromagnetic data

<p>In some Earth materials, induced polarization (IP) phenomena are occurring when an electric perturbation is applied. These mechanisms are described by a frequency dependent complex resistivity. The study of relaxation model parameters describing these phenomena allows to access indirectly to several properties of interest of the underground, as properties linked to the pore space geometry, fluid content or presence and discrimination of disseminated metallic particles. Nevertheless, complex resistivity is usually studied using electrical method with a direct current hypothesis, neglecting by the way electromagnetic induction that can occurs in the data. Thus, strong limitations appear to recover a complex resistivity image as EM induction increase with frequencies and larger offset.</p><p> </p><p>We implemented a frequency dependent complex resistivity in POLYEM3D, a 3D finite-difference modelling and inversion code for controlled-source electromagnetic data (CSEM) in order to fully recover IP information contained in EM data. CSEM method is a resistivity imaging technique using multi-frequency electromagnetic signals fully taking into account EM induction with larger investigation depth. Following a preliminary sensitivity study, a multi-stages inversion strategy was defined to undertake the multi-parameters problem. Furthermore, to manage the increasing number of parameters, a second order polynomial parametrization is used to describe frequency variation of complex resistivity.</p><p> </p><p>We show through 1D synthetic data inversions and preliminary 3D results that we are able to recover a complex resistivity and its frequency variation from CSEM data in the IP/EM coupling domain, when IP signals are sufficiently large compared to EM induction. Our inversion strategy allows then to access to IP parameters of the medium in an extended frequency domain as well as for greater depth of investigation. A 3D CSEM survey was undertaken in December 2020 on the former mining site of La Porte-Aux-Moines (Côtes-d'Armor, France) presenting strong IP responses, to validate our inversion method for a 3D in-situ dataset.</p>

Petrophysical study of different rock types from the mining area of Freiberg, Germany

<p>The ore mining area of Freiberg is located in the federal state of Saxony in the east part of Germany and is characterized by hydrothermal ore mineralization.</p><p>The present petrophysical study concentrates on three different rock types from the research mine “Reiche Zeche”. The set of samples contains rocks from the metamorphic host rock - Freiberger Gneiss (FG), from a hydrothermal alternated gneiss (hG) and from a pyrit-galenit rich ore vein (ore). The investigations include the determination of density and porosity as well as laboratory measurements of the Spectral Induced Polarization (SIP) in the frequency range from 10<sup>-3</sup> to 10<sup>4</sup> Hz. Furthermore, measurements of the magnetic susceptibility and P-wave velocity were performed.</p><p>For the determination of P-wave velocity by ultrasonic measurements, the rock samples were cut into cubes (5 cm x 5 cm) under preservation of their spatial orientation. The sample set contains 17 cubes (FG - 8 cubes, hG -3 cubes and ore – 6 cubes).</p><p>The determination of the complex resistivity was performed in a SIP – measuring cell on cylindric plugs with a length of 3 cm and a diameter of 2 cm. For the SIP-measurements 19 plugs (FG – 11 plugs, hG – 2 plugs and ore – 6 plugs) were available.</p><p>Density and magnetic susceptibility measurements were performed on 10 samples of crushed material for each rock type.</p><p>The data sets of complex resistivity and P-wave velocity measurements from rock samples of the metamorphic host rock and the ore vein were analysed with focus on possible anisotropic behaviour.</p>

Cole-Cole Model Parameter Estimation from Multi-frequency Complex Resistivity Spectrum Based on the Artificial Neural Network

In near surface electrical exploration, it is often necessary to estimate the Cole-Cole model parameters according to the measured multi-frequency complex resistivity spectrum of ore and rock samples in advance. Parameter estimation is a nonlinear optimization problem, and the common method is least square fitting. The disadvantage of this method is that it relies on initial value and the result is unstable when data is confronted with noise interference. To further improve the accuracy of parameter estimation, this paper applied artificial neural network (ANN) method to the Cole-Cole model estimation. Firstly, a large number of forward models are generated as samples to train the neural network and when the data fitting error is lower than the error threshold, the training ends. The trained neural network is directly used to efficiently estimate the parameters of vast amounts of new data. The efficiency of the artificial neural network is analyzed by using simulated and measured spectral induced polarization data. The results show that artificial neural network method has a faster computing speed and higher accuracy in Cole-Cole model parameter estimation.

Complex Electrical Resistivity and Dielectric Permittivity Responses to Dense Non-aqueous Phase Liquids' Imbibition and Drainage in Porous Media: A Laboratory Study

The effective techniques for remediation of sites polluted by dense non-aqueous phase liquids (DNAPLs) remains a challenge. Among the various technical monitoring methods, there is an increasing interest in studying the geophysical characteristics of contaminated soils, as indicators of the progress in clean-up programs. This work sought to investigate the variation of the electrical complex resistivity and the relative permittivity by analyzing the results obtained from spectral induced polarization (SIP) and time domain reflectometry (TDR). Different series of measurements during drainage and imbibition of DNAPLs in porous media were done to validate the clean-up process on sites polluted by DNAPLs. Therefore, a methodology based on laboratory work was designed and carried out to study the electrical complex resistivity (both in magnitude and phase) in the frequency range 0.183 Hz to 20 kHz, and the relative dielectric permittivity at 70 MHz. The experiments were done on small 1D cells. In these cells, glass beads were used as a porous medium. Two different fluid couples, i.e., coal tar (CT)/water and canola oil (CO)/salty ethanol (SE), were used to produce two-phase flow. Our findings highlight that due to the high resistivity of CO and CT, an increase in water saturation led to decrease in amplitude and phase. Saturation change of SE had the same effect on resistivity but no relationship was found for phase and saturation for the mixture CO and SE. It is also showed that the complex resistivity and relative permittivity measurements were compatible with generalized Archie's law and complete complex refractive index method (CRIM) model as two empirical models for defining the correlation between the electrical resistivity, relative permittivity, and saturation of each phase in the multiphase porous medium.

Experimental Study on Characteristics of Reactance Dispersion of Water Bearing Coal

As an emerging geophysical exploration method, complex resistivity method is a potential non-invasive one for evaluating hydrofracturing effect. Reactance X is an important parameter of complex resistivity method. Compared with the traditional parameter resistance R, reactance has advantages such as distinct dispersion characteristics, clear conduction mechanism and rich information. In this paper, reactance X of four kinds of coal sample with different water saturation was tested, their dispersion characteristics were analyzed, and their conductive mechanism was studied. The results show that, (1) the characteristic curve of reactance dispersion presents a three-stage law when the water saturation is low, and a two-stage law when the water saturation is high,to some extent, the water content of coal body and the effect of hydraulic fracturing are distinguished;(2) polarization is the cause of dispersion, in which low frequency is induced polarization, high frequency is dielectric polarization, and electromagnetic induction is the interference phenomenon in the frequency band of induced polarization; (3) spectrum extremum frequency point is sensitive to changed water saturation. Extremum points of different coal rank are concentrated in 100-1000Hz. This frequency band can serve as the dominant frequency band for evaluating coal seam water saturation. This paper gives new insight into the evaluation of coal seam hydrofracturing effect by complex resistivity method.