The effect of local distortions on time‐domain electromagnetic measurements

Abstract

Based on the time‐domain integral equation, we derive expressions for the effect of an anomalous body close to the receiver or close to the transmitter on transient electromagnetic measurements. Similar to magnetotellurics, the distortion of electric fields at late times can be described by a constant distortion tensor relating the secondary electric field to the primary field components that would be obtained in the absence of the body. The distortion of a single electric field transient is a static shift only for particular configurations over a layered half‐space. In the general case, the perturbation is time dependent because the direction of the total electric field vector varies with time. The theory nicely explains spatial variations in electric field transients measured during a high‐redundancy long‐offset transient electromagnetics (LOTEM) survey over an underground gas storage site. An inversion example with synthetic data illustrates how distortion can be corrected. The elements of the distortion tensor are determined simultaneously with the model parameters. Ambiguity is reduced by a regularization of the distortion parameters. In the example, the background model is recovered well, even for the difficult case where only one transmitter is used. The distortion of the magnetic field time derivatives caused by bodies close to the receiver is proportional to the time derivative of the primary electric step response. The distortion is generally not limited to early times and cannot be neglected in general. Transmitter overprint effects resulting in static shifts of vertical magnetic field time derivatives may also be understood from the theory.

A resolution study of buried valleys using laterally constrained inversion of TEM data

“…During the years, a number of papers have dealt with the effect of 3D structures in 1D interpretation of TEM data, e.g. Newmann et al, (1987), Goldman et al (1994), Auken, (1995 and Hördt and Scholl, (2004). In general, these studies show that, if the geological environment consists of gently varying 3D structures with moderate resistivity contrasts, the 1D inversion approach in many cases gives a good recovery of the true model.…”

Breaks in lithology: Interpretation problems when handling 2D structures with a 1D approximation

“…Most of the available literature shows that interpretation of isotropic and slowly varying geoelectrical data is possible, using full inversion methods with constraints and fast conductivity transforms ͑Newman et al., 1987;Goldman et al, 1994;Hördt and Scholl, 2004;Auken et al, 2008;Viezzoli et al, 2008;Christensen et al, 2010͒. Because of the volume of data, AEM surveys usually are interpreted using stitched 1D models, although most interpreters are aware that geologic structures have a 3D spatial distribution. Despite successful 2D/3D EM modeling and inversion ͑Haber et Wilson et al, 2006͒, these multidimensional algorithms are very computationally demanding, which limits their applications on large data sets such as those acquired for AEM surveys, typically sampled every 3 -10 m and resulting in millions of soundings per survey.…”

Coastal aquifer assessment based on geological and geophysical survey, northwestern Crete, Greece

“…Presently, several works, e.g., Newman et al (1987), Goldman et al (1994a), Auken (1995) and Hördt and Scholl (2004), show that if the geological condition consists of smoothly varying 3D structures with moderate resistivity contrasts, the 1D inversion approach gives a satisfactory reconstruction of the true model. Of course, in case of high-resistivity geological contacts and/ or contrast and strong 3D features, an 1D approximation is strongly influenced by 3D effects and in many cases will provide unreliable models.…”