Introduction to Ground Penetrating Radar E-Book

Raffaele Persico

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 536

Veröffentlichungsjahr: 2014

Beschreibung

A real-world guide to practical applications of ground penetrating radar (GPR) The nondestructive nature of ground penetrating radar makes it an important and popular method of subsurface imaging, but it is a highly specialized field, requiring a deep understanding of the underlying science for successful application. Introduction to Ground Penetrating Radar: Inverse Scattering and Data Processing provides experienced professionals with the background they need to ensure precise data collection and analysis. Written to build upon the information presented in more general introductory volumes, the book discusses the fundamental mathematical, physical, and engineering principles upon which GPR is built. Real-world examples and field data provide readers an accurate view of day-to-day GPR use. Topics include: * 2D scattering for dielectric and magnetic targets * 3D scattering equations and migration algorithms * Host medium characterization and diffraction tomography * Time and frequency steps in GPR data sampling * The Born approximation and the singular value decomposition The six appendices contain the mathematical proofs of all examples discussed throughout the book. Introduction to Ground Penetrating Radar: Inverse Scattering and Data Processing is a comprehensive resource that will prove invaluable in the field.

Leseprobe

IEEE Press445 Hoes LanePiscataway, NJ 08854IEEE Press Editorial BoardTariq Samad, Editor in Chief

George W. Arnold

Mary Lanzerotti

Linda Shafer

Dmitry Goldgof

Pui-In Mak

MengChu Zhou

Ekram Hossain

Ray Perez

George Zobrist

Kenneth Moore, Director of IEEE Book and Information Services (BIS)

INTRODUCTION TO GROUND PENETRATING RADAR

Inverse Scattering and Data Processing

Raffaele Persico

Cover Design: WileyCover Images: Francesco Gabellone/Courtesy of the author

Copyright © 2014 by The Institute of Electrical and Electronics Engineers, Inc.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey. All rights reserved.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762–2974, outside the United States at [phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Persico, Raffaele, 1969– Introduction to ground penetrating radar: inverse scattering and data processing/Raffaele Persico.  pages cm Includes bibliographical references and index.

 ISBN 978-1-118-30500-3 (hardback)1. Ground penetrating radar. I. Title.  TK6592.G7P47 2014 621.3848′5–dc23

                     2013039737

To my wife Grazia Maria and my daughter Luisa Anna

Foreword

Ground penetrating (or probing) radar (GPR) is a vital technique on which the day-to-day safety of literally millions of people depend. The technology allows a very wide range of verifications, the most common being the safe and accurate location of the position of buried pipes and utilities, investigating the reinforcement and condition of roads, bridges, and airport runways, and identifying the structural integrity of buildings. Other important applications include locating buried potential hazards such as mine shafts and voids, investigating environmental and geological conditions (both of natural and man-made origin), studying glaciology, locating, identifying, and investigating archaeological sites, and uncovering forensic evidence including buried human remains and weapons.

Although the technology is widely used, it is a highly specialized area that requires a good understanding of the underlying science if it is to be applied successfully. In addition to the technical journals that regularly carry scientific papers on both the theory and application of GPR, there are two major biennial international conferences, namely the International GPR Conference and, in the intervening years, the International Workshop on Advanced GPR. In terms of books, the fundamental cornerstone of GPR in all its applications has long been David J. Daniels’ Ground Penetrating Radar. However, one book, regardless of how well it is researched and written and how comprehensively it addresses its subject matter, cannot cover all aspects of the science to the equal satisfaction of all users. This current volume is not intended for the general reader or for anyone for whom this is a completely new subject. Rather it is aimed primarily at doctoral and post-doctoral students who wish to develop their understanding of the technology and, in particular, how the results may be developed and interpreted.

Dr. Persico has been a practitioner and researcher in GPR for many years, primarily interested in the resolution of inverse problems, with particular application to archaeological investigations and the conservation of cultural heritage. From his background in Physics and Mathematics and the expertise he has built up, he applies processing and interpretative techniques to GPR data collected with the primary aim of conserving, preserving, and rehabilitating buildings of historical importance and also archaeological remains. In 2010 he chaired the 13th International GPR Conference, held in Lecce, Italy. He is also an active member of the European GPR Association within which he has worked to build a virtual library, accessible to the membership through the Association website (www.eurogpr.org). His considerable expertise makes him an ideal candidate to share his knowledge and understanding of GPR interpretative techniques with other researchers and users.

This book consists of 15 chapters and 7 appendices, the aim being to introduce the reader to the complexities of using and interpreting GPR data step by step. An important feature of this book is the inclusion of questions at the end of almost all of the chapters, allowing the reader to assess his or her progress in understanding the subject. The answers to questions are in Appendix G.

Beginning with a general definition and description of GPR technology, this book goes on to consider an important basic characteristic of GPR operation, namely the interdependence of the nature of the survey medium with the transmission of the electromagnetic pulses through that medium. Chapter 3 considers the time and frequency aspects of GPR transmission and their implications. The following two chapters concentrate on the mathematical aspects of GPR transmission including Maxwell’s equations, the effects of incident fields, and the relevant scattering equations considered in two dimensions for dielectric and magnetic materials (Chapters 4 and 5).

Continuing into methods of making the interpretation of GPR data more accessible, the next section describes a number of mathematically based constructs that can be used for this purpose. Chapter 6 introduces the reader to the nature of inverse scattering problems and the associated mathematical uncertainties to be addressed in resolution. In Chapter 7, data processing steps typically associated with improving raw GPR data are described in detail and their effects are illustrated.

The Born approximation has become a standard algorithm to apply to GPR data and has also formed the basis of much of Dr. Persico’s own research work. The Born approximation and its application to magnetic targets and also to weak and strong signal reflectors is described in full in Chapter 8. Leading on from this, the theoretical basis and application of diffraction tomography is the next topic, including: the consideration of horizontal and vertical resolution of targets; related sampling issues in space, frequency, and time; and the radiation characteristics of radar antennas (Chapter 9). Chapter 10 deals with two-dimensional migration algorithms in the frequency domain and the time domain.

Chapter 11, contributed by Drs. Lo Monte and Solimene together with Dr. Persico, extends data treatment into the development of three-dimensional scattering equations based on Maxwell’s equations and includes particular consideration of Green’s function applied to three-dimensional space. This is followed by extending the analysis of diffraction tomography into three dimensions with careful consideration of the sampling parameters required for the application of the technique in order to reconstruct the target(s) reliably from the GPR data (Chapter 12). This is important because improving target resolution from appropriate survey parameters is of primary importance in all GPR applications.

Building on the previous chapters, Chapter 13 considers the corresponding derivation of three-dimensional migration algorithms, first in the frequency domain and subsequently in the time domain before comparing two-dimensional and three-dimensional migration formulas in the time domain by means of a worked and illustrated example.

Chapter 14 considers the alternative technique of singular value decomposition. The same careful mathematical logic is used to derive the singular value decomposition before considering its application to GPR data.

Finally, Chapter 15 provides a variety of worked examples and exercises to illustrate some of the concepts covered in the preceding chapters. This begins with an examination of measuring propagation velocities followed by two sets of exercises dealing with target resolution, namely the interrelation of spatial sampling and horizontal resolution and the effects of frequency sampling on vertical resolution. Both of these latter two concepts are integral to understanding the capability of any GPR. The next set of exercises is concerned with trialing the number and categories of unknowns treated in the equations in order to optimize the quality of the GPR data without undue processing simply for the sake of it. This is followed by exercises examining frequency and spatial content of the data and consideration of the effects of measurement from above the soil–air interface (instead of directly coupled). Extending the area of investigation provides the basis for the next set of exercises, an important consideration given that this varies extensively from one survey to another. This is followed by exercises using background removal, the single most extensively applied processing technique for all GPR surveys since, by definition, it is the anomalous material that forms the targets. The added complication of complementary data sets in different orientations is then considered, based on a real archaeological example and including contributions from Drs. Ciminale, Leucci, and Matera. Lastly, the results of further 2D and 3D inversion techniques are compared (with the collaboration of Dr. Catapano).

Detailed mathematical workings in support of the content of the chapters are provided in full in the appendices.

This is not a volume for a beginner, but it is a careful and comprehensive enumeration and explanation of the mathematical concepts inherent to GPR. It should provide a useful platform for those who wish to delve deeper into the subsurface of the technology and equip themselves with the mathematical tools for handling their own data sets.

Erica Carrick UtsiNovember 2013

Acknowledgments

I sincerely thank the six contributors—Dr. Ilaria Catapano, Professor Marcello Ciminale, Dr. Giovanni Leucci, Dr. Lorenzo Lo Monte, Dr. Loredana Matera, and Dr. Raffaele Solimene—who helped me write this book. Their contributions were of the utmost importance to me, and the interaction with them has enhanced and deepened my knowledge of the covered topic. Moreover, I am grateful for the fruitful and interesting scientific discussions with Professor Massimiliano Pieraccini of the University of Florence and with his research group, with a special mention to Dr. Filippo Parrini and Dr. Devis Dei. These discussions inspired me to think about the calculation of the Hermitian images in a closed form, detailed in Chapter 3. I also would like to thank Dr. Jacopo Sala of 3d-Radar for our scientific collaboration, which in particular inspired me to write Section 15.7. I also would like to thank Dr. Erica Utsi of Utsi Electronics Ltd, Chair of the European GPR Association. She wrote the Foreword for this book, and this is an honor for me. Finally, I am grateful to Dr. Francesco Gabellone for his valuable aid in conceiving the cover for this book.

Raffaele Persico

About the Author

Raffaele Persico was born in Napoli, Italy in 1969. After humanistic secondary school studies, he achieved his degree in Electronic Engineering from the University of Napoli Federico II and then his Ph.D. in Information Engineering from the Second University of Napoli. He has been Fellow Student at the Second University of Napoli, Research Scientist at the Consortium CO.RI.S.T.A., and then Research Scientist at the Institute for the Electromagnetic Sensing of the Environment IREA-CNR. Since 2007, he has been affiliated with the Institute for Archaeological and Monumental Heritage IBAM-CNR.

Dr. Persico’s research activity has been devoted to microwave imaging, inverse problems, GPR data processing, and GPR systems. He has co-authored about 150 papers in international journals and conference proceedings and holds an Italian patent on the reconfigurability of the GPR systems. He is reviewer for several international journals and is Associate Editor of Near Surface Geophysics. He chaired the 13th International Conference on Ground Penetrating Radar, held in Lecce, Italy in 2010, and has devised a prototypal reconfigurable stepped-frequency GPR system together with the IDS Corporation and the University of Florence. Since 2009 he has been a Member of the EuroGPR Association.

Contributors

1INTRODUCTION TO GPR PROSPECTING

1.1 WHAT IS A GPR?

Ground-penetrating radar (GPR), also known as surface-penetrating radar (SPR) (Daniels, 2004), is literally meant as a radar to look underground. Actually, it is used to look into both soils and walls and, recently, even beyond walls.1

In principle, the GPR can be viewed as composed by a central unit, a transmitting antenna, a receiving antenna, and a computer. The central unit generates an electromagnetic pulse or, more generally, an electromagnetic signal that is radiated into the soil by the transmitting antenna. The signal is radiated in all directions, but most energy is radiated within a conic volume under the antenna, as shown in Figure 1.1. When the electromagnetic waves meet any buried discontinuity (a buried object but also the interface between two geological layers, a cavity, a zone with different humidity, etc.), they are scattered in all directions according to some pattern depending on the buried scenario. Consequently, they are partially reflected also toward the receiving antenna, again according to Figure 1.1. More precisely, Figure 1.1 is intended to be the central cut of a three-dimensional scenario.

Figure 1.1. The working principle of a GPR.

Usually, both the transmitting and the receiving antennas are incorporated into a rigid structure2 and move together. The gathered signal is customarily represented in real time on the screen of the computer3 and is stored in the hard disk of the computer. It is implicit that the equipment of a GPR also includes suitable cables to connect the central unit, the antennas, and the computer, along with a device to provide energy in the field. The energy is usually supplied by rechargeable batteries in the form of a zero-frequency electrical voltage. The central unit transforms this energy into a signal in the microwave frequency range. Modern systems are often also equipped with a GPS, in order to geo-reference the probed areas.

In Figure 1.2, a photograph of a GPR is shown, and the main components are put into evidence. The trolley is facultative, but extremely useful for prospecting on the soil. Usually, the antennas are also equipped with an odometer that allows us to measure the covered distance.

Figure 1.2. Photograph of a GPR (a Ris Hi Mode system) equipped with a double antenna at 200 and 600 MHz antenna.

The odometer is an important detail, because it allows us to compensate for the natural nonuniformity of the velocity of the human operator while driving the GPR: that is, it allows us to achieve a uniform sampling of the GPR signal along the observation line.

However, in some cases it is impossible to make use of the odometer—for example, because the prospecting is on a sandy area that hinders the rotation of the wheel. In these cases, periodical markers have to be recorded along the observation line, which is partitioned into segments of known length. The velocity of the instrument (and thus also the sampling) is considered constant within each segment but not along the entire observation line.

The working principle of a GPR is the same as that of a conventional radar. However, there are meaningful differences between the two instruments, in terms of technologies, exigencies, applications, and frequency bands (Daniels, 2004; Levanon, 1988). In particular, unlike the conventional radar, usually a GPR has to identify static targets, and in most cases the interpretation of the data is not requested in real time. On the other hand, in GPR prospecting the electromagnetic waves do not propagate in air but instead propagate in more complicated host media, customarily lossy and inhomogeneous, possibly dispersive, and in some cases anisotropic and/or magnetic (Daniels, 2004; Jol, 2009; Conyers, 2004). Last but not least, in GPR prospecting the characteristics of the host medium are usually not known a priori and have to be estimated from the data, as described in more depth in the next chapter.

1.2 GPR SYSTEMS AND GPR SIGNALS

There are essentially two kinds of GPR systems: the pulsed one and the stepped-frequency one. A pulsed GPR system radiates and receives the echoes to electromagnetic pulses. On the other hand, a stepped-frequency GPR decomposes the electromagnetic pulse into its spectral components and radiates them sequentially. Consequently, it radiates and receives trains of sinusoidal signals. The soil and the buried targets usually have a linear behavior with respect to the radiated GPR signal, in the sense that the signal scattered by the buried targets is a linear quantity (more details will be given in Chapter 6) with respect to the incident signal. Moreover, the soil can usually be considered a time-invariant medium within the time needed for the GPR measurement campaign. This makes the pulsed and the stepped-frequency GPRs theoretically equivalent. In practical terms, however, stepped-frequency systems are generally claimed more performing (Noon, 1996), even if the pulsed systems are quite more common and their technology has been assessed for a longer time. So, the debate about what kind of system is really the best one (or at least the most convenient one in dependence of the application) is still open. In this text we will not enter such a debate, which is mainly based on technological aspects, but will deal with both some analytical and practical aspects of the GPR prospecting in relationship with both systems. In particular, whatever the system, the GPR signal can be regarded as a function of the spatial point and of the time or the frequency indifferently, because of course we can Fourier transform pulsed GPR data in frequency domain and we can back-Fourier transform stepped-frequency GPR data in time domain.

Following a widely accepted terminology (Daniels, 2004), the GPR data relative to a single spatial point will be labeled as an A-scan or just a GPR trace, and the comprehensive set of GPR traces relative to an entire scanned line will be labeled as a B-scan. A B-scan corresponds to a matrix of numbers: N time (or frequency) samples times M spatial positions—that is, M traces each of which discretized into N samples. This is equivalent to assuming that the GPR “stops” in each A-scan position, gathers the data in that position, and goes on to the next position. Actually, in most cases the data are gathered in continuous mode—that is, the GPR gathers the data while moving—but the model “stop-gather-and-go-on” is in most cases acceptable because of the huge difference between the velocity of propagation of the electromagnetic signal and the velocity of the human operator, even if the time required to store an A-scan is actually quite longer than its formal time bottom scale. This happens because of several reasons such as sequential sampling (for pulsed systems), integration time of the harmonics (for stepped frequency systems), and stacking (for both). Here, we will not focus on these aspects, which are mainly technological and already explained elsewhere (Noon, 1996; Daniels, 2004; Jol, 2009). Let us just restrict ourselves to say that a nonexcessive (the quantification is case-dependent) and constant velocity during the data acquisition is always a good rule of thumb. The comprehensive set of GPR data relative to a series of parallel B-scans is usually labeled as a C-scan. In general, what is immediately visualized in the field is a B-scan in some color or gray tone scale. These data, usually called raw data, can allow us to identify targets of interest, but in general the image and its interpretation can improve meaningfully after a suitable processing.

1.3 GPR Application Fields

A meaningful overview about the GPR applications is beyond the purposes of this book and is not its goal. Notwithstanding, for sake of self-consistency, a brief outlook is provided.

Within the field of the archaeology (Conyers, 2004), GPR can allow us to identify the areas with alleged interesting buried remains, which in turn allows us to avoid an exhaustive and expensive (sometimes too expensive) excavation. Another issue of interest is the field of the preventive archaeology—that is, the preventive prospecting of areas where something is going to be built (a road, a building, an underground station, etc.). This mitigates the risk of destroying archaeological sites and also mitigates the economic risk that the works will be stopped by some Cultural Heritage Institution.

Monitoring of monuments as historical buildings, statues (Sambuelli et al., 2011), ancient fountains, historical bridges (Solla et al., 2011), and so on, is another subject of interest. In particular, GPR monitoring (possibly integrated with other geophysical investigations) can give information about the state of preservation of the monuments and can provide useful information in order to address a restoration project properly. In some cases, information of historical interest can also be achieved—for example, about the presence of walled rooms, crypts, hypogeum rooms, tombs, hidden frescoes, and so on (Pieraccini et al., 2006; Grasso et al., 2011).

GPR prospecting is also exploited in civil engineering (Grandjean et al., 2000; Utsi and Utsi, 2004). In particular, it can be used to identify structural damages and to investigate about hidden structures like sewers or water and gas pipes, whose presence is in many cases not precisely documented.

Demining is another important application. In particular, modern mines are customarily built with plastic materials with only little or even no metallic parts. Therefore, they are often hardly visible or completely invisible to a metal detector. Moreover, a metal detector is not able to provide all the details possibly available from a GPR system, namely the position (in particular the depth), the size, and (among certain limits) the shape of the buried target. Demining has been dealt with for years within the GPR community (Groenenboom and Yarovoy, 2002), and it has also been successfully performed many times (Sato and Takahashi, 2009).

GPR prospecting is also exploited for asphalt monitoring.4 In particular, it is possible to identify subsidences or damaging before they become worse or even dangerous for drivers and pedestrians. These problems are even more pressing in areas where the roads frost in the winter and thaw in the spring (Hugenschmidt et al., 1998; Villain et al. 2010).

In several application fields, it can be particularly useful to make use of advanced GPR systems equipped with a large array of antennas (Sala and Linford, 2010; Böniger and Tronicke, 2010). These systems can gather simultaneously several (up to 14 and more) measurement lines with a unique going through. On the other hand, these systems need a quite flat scenario to provide good performances, because the arrays are rigid and possibly quite large (up to 2 m and larger).

GPR prospecting is also applied with regard to mines and pits. In particular, it can help to identify shallow veins of the mineral of interest (Ralston and Hainsworth, 2000; Francke, 2010) and can even help with regard to some safety issues. In fact, fractures, water infiltrations, or just obsolescence can badly affect the stability of the structure, in mines as well as in tunnels (Grodner, 2001; Cardarelli et al., 2003). In some cases, even explosive gases trapped in natural cruets can be met while digging, especially in coal mines (Cook, 1975).

Another application of interest is GPR prospecting on the ice (Arcone et al., 2005). In particular, polar ice contains information about the geological history of our planet and can also provide information about the occurring climatic and environmental changes. GPR prospecting can be successfully performed on both fresh and salty ice.