High-Frequency and Very-High-Frequency (HF & VHF) Above-Ground Electromagnetic Impedance Measurements

Abstract

We have field-tested an apparatus for measuring the electromagnetic impedance above the ground at a plurality of frequencies in the 0.3 -30 MHz range. This window in the frequency spectrum, which lies between frequencies used for GPR and those used for conventional loop-loop EM soundings, has not been used because of difficulties in fielding equipment for making absolute and accurate measurements. Model and physical parameter studies however confirm that data in this frequency band can be used to construct high-resolution maps of electrical conductivity and permittivity of near-surface material. Our equipment was assembled using commercial electric and magnetic antennas. The magnetic loop source is excited by a conventional signal generator -power amplifier assembly. Signal detection is accomplished using RF lock-in amplifiers. All system elements are appropriately isolated by optic -fiber links. We estimate a measurement accuracy of about ± 10% for an 8-m separation between source and detector. Field tests were done at the University of California Richmond Field Station where the near surface electrical structure is well known. The experimental data at this site are mainly a function of electrical conductivity. In this context, we have obtained good agreement with the known local variations in resistivity both with depth and with position along a 35-m traverse. Additional tests in more resistive regimes where dielectric permittivity is not negligible yield spectral data compatible with the less well known near-surface electrical properties.

Trajectory-based modelling of broad-band electromagnetic wavefields

“…Broad-band waveforms, in the frequency range of 1 kHz to 300 kHz, are computed for 128 frequencies. As noted in Frangos et al (2004) there is some variation in conductivity with depth at the Richmond Field Station. The trajectory-based impulse response is convolved with the source wavelet and the transmitter response, which may be obtained from the data in Fig.…”

“…If one is simply interested in developing solutions that model wave-like (hyperbolic) propagation, the most direct approach involves an expansion in inverse powers or inverse fractional powers of ω (Kline and Kay 1965). It was not until relatively recently (Stewart et al 1994;Frangos et al 2004) that experimental systems were developed which could operate in an intermediate frequency range where both conduction and displacement currents are important. However, one must then neglect the term representing displacement currents in the governing equation before introducing the asymptotic series solution in inverse powers of √ ω (Virieux et al 1994).…”

“…A broad-band source would be useful for shallow geophysical investigations and crosswell studies (Stewart et al 1994;Becker et al 1999;Frangos et al 2004). A broad-band source would be useful for shallow geophysical investigations and crosswell studies (Stewart et al 1994;Becker et al 1999;Frangos et al 2004).…”

“…That is, one is interested in estimating an earth model which is compatible with a given set of observations. Furthermore, there is recognition of the advantages of broad-band data sets, spanning the kilo-hertz to mega-hertz frequency interval (Stewart et al 1994;Becker et al 1999;Frangos et al 2004). Such inverse problems require repeated solution of the forward problem and usually involve one to several orders of magnitude more computation then is necessary to solve the forward problem.…”

“…For example, because of their sensitivity to fluid properties, there is renewed interest in using electromagnetic data to characterize largescale reservoir models and for high-resolution monitoring of environmental remediation and reservoir production (Barker & Moore 1998;Binley et al 2001;Hoversten et al 2003). Data in this frequency range are particularly suited for imaging shallow and intermediate depth heterogeneity, and for inferring both conductivity and dielectric permittivity in the subsurface (Frangos et al 2004). Furthermore, there is recognition of the advantages of broad-band data sets, spanning the kilo-hertz to mega-hertz frequency interval (Stewart et al 1994;Becker et al 1999;Frangos et al 2004).…”