Advances in time–domain induced polarisation tomography: Data acquisition, processing and modelling

Per-Ivar Olsson

Research output: ThesisDoctoral Thesis (compilation)

296 Downloads (Pure)

Abstract

What would you find below your feet – how do you find out? Perhaps you could look at a geological map, drill, or dig? That could work, but sometimes the maps are not detailed enough, and digging everywhere to find out is impractical. Imagine if you could develop a method that allows you to see straight through the ground, as x–rays through the body! To some extent, such methods already exist; you can for example send electric current into the ground that tells you what lies beneath when it returns. The current does not see the subsurface as we do–no soil, boulders, water or bedrock – but it can tell us about its view of the underground. Its image is in terms of electrical resistivity and chargeability and can be difficult to understand if it is not translated. The translation is accomplished by comparing the electric image with, for example, geological maps and information from boreholes. With this method, we obtain more reliable and more detailed models of the subsurface. Additionally, electrical surveys can help us to determine where we need more subsurface information and where it would be most interesting to drill or dig.

When we plan and build structures below or above ground, search for suitable places for wells, or remediate contaminated areas, we need good and reliable information about the ground below. Incorrect or incomplete information about the subsurface can lead to unexpected problems. These problems can in turn lead to delays and reduced sustainability in the implementations. One example of such a project is the train tunnel through Hallandsåsen in Southern Sweden, which suffered several delays and took 23 years to complete, and its final price tag was approximately ten times the initial estimate.

This thesis addresses how we can develop and improve the use of electrical current to investigate the subsurface. The method has been used and developed for more than one hundred years, but bottlenecks remain that limit its use. One example is in cities where electrical installations and a complex environment in the subsurface distort the current. We then need to filter the image to make use of the results from the measurements. The method may also be limited by lack of resources needed to conduct the investigations, or to make proper interpretations of its information. By refining and optimising the method, its usefulness can be increased. For example, by enabling its use in urban areas or for projects with limited resources.

This thesis describes how we can process signals from electrical surveys and handle interference from other electrical installations, similar to a pair of noise–cancelling headphones. The processing allows us to retrieve more information about the subsurface and increase the reliability of the results. Another improvement that is introduced is a change of the shape of the current that is sent into the ground. The change of waveform results in a reduction in the time required for a survey, while the magnitude of the signals is increased, similar to completing a podcast in half of the time with better audio quality.

Another way to improve the method is to increase our understanding of what types of responses we can expect from the measurements. This thesis describes how measurement results that were previously considered erroneous can be explained, and that these are actually physically possible. By not rejecting such results, we can obtain more information from the measurements, more reliable models of the subsurface, and post–processing of the measurements is simplified. In addition, it describes how we can compensate for the effects of varying duration of current transmissions. If the effects are not considered properly, different electrical images of the same subsurface are obtained depending on whether the current is sent just one second longer or shorter.

The optimisations of the thesis are exemplified with, among others, results from a major survey that mapped a geologic site in terms of resistance and chargeability down to 200 metres below ground. Such information is important and can help us to take better decisions, for example in connection with infrastructure projects for a more sustainable future. Hopefully, the work in this thesis can increase the use of electrical surveys, ensuring we can make more informed decisions in the future.
Original languageEnglish
QualificationDoctor
Awarding Institution
  • Faculty of Engineering, LTH
Supervisors/Advisors
  • Dahlin, Torleif, Supervisor
  • Fiandaca, Gianluca, Assistant supervisor, External person
  • Auken, Esben, Assistant supervisor, External person
Award date2018 Nov 30
Publisher
Print ISBNs978-91-7753-850-9
Electronic ISBNs978-91-7753-851-6
Publication statusPublished - 2018 Nov 5

Bibliographical note

Defence Details
Date: 2018-11-30
Time: 10:15
Place: lecture hall B, building V, John Ericssons väg 1, Lund University, Faculty of Engineering LTH, Lund
External reviewer(s)
Name: Schmutz, Myriam
Title: Professor
Affiliation: Université Bordeaux, France
---

Subject classification (UKÄ)

  • Engineering and Technology

Keywords

  • Tomography
  • Signal–to–noise ratio
  • Waveform
  • Duty–Cycle
  • induced polarisation
  • Electrical properties

Fingerprint

Dive into the research topics of 'Advances in time–domain induced polarisation tomography: Data acquisition, processing and modelling'. Together they form a unique fingerprint.

Cite this