Figure 1 - uploaded by Jan Igel
Content may be subject to copyright.
Picture of sandy soil demonstrating soil heterogeneity on an area of 1 m × 1 m. The turf has been removed and the pattern corresponds to the soil moisture (dark parts = moist soil, light parts = dry soil), pers. comm. Schmalholz, TU Berlin.
Source publication
The small-scale variability of physical soil properties has a negative influence on ground exploration with physical sensors. This particularly holds true for small target objects like landmines. Studies were carried out to determine magnetic susceptibility, electric conductivity and dielectric permittivity of natural soils. The spatial variability...
Contexts in source publication
Context 1
... question is to what extend physical soil heterogeneities influence the measurements which are used to detect landmines. Figure 1 shows a picture of a sandy soil whereas the turf has been removed. The pattern corresponds to the moisture distribution and one can recognise high variability in the range of decimeters. ...
Context 2
... resistivity (inverse of conductivity) distribution at a depth of 0.1 m as a result of a 3D dipole-dipole measurement (longitudinal and equatorial) on an area of 1.5 m × 1.5 m after inversion. One can recognise a high spatial variability mainly caused by moisture inhomogeneity with a spatial distribution very similar to the pattern illustrated in Fig. 1. Figure 7 shows the conductivity distribution along a 15 m long 2D profile. The ground is distinctly layered: heterogeneous topsoil overlays homogeneous subsoil. The interface between both soils at a depth of 0.25 m corresponds to the former ploughing horizon. Topsoil heterogeneity is probably caused by vegetation and root distribution ...
Context 3
... the measurement is not influenced by vegetation and a proper coupling of the antennas to the ground is ensured. Figure 10 shows the measuring grid and the deduced permittivities for the profiles in y-direction. Permittivity distribution shows a stripe pattern in direction of approx. ...
Context 4
... is not influenced by vegetation and a proper coupling of the antennas to the ground is ensured. Figure 10 shows the measuring grid and the deduced permittivities for the profiles in y-direction. Permittivity distribution shows a stripe pattern in direction of approx. 30 • with regard to the x-axis. The probability density functions is plotted in Fig. 11. In contrast to conductivity, permittivity is almost normally distributed (ε r = 5.3 ± 0.85). Due to the stripe pattern which can be seen in Fig. 10, variogram analysis was performed in different directions (Fig. 11). The correlation length in x-direction is longer than in y-direction. An anisotropic variogram model was fitted with the ...
Context 5
... for the profiles in y-direction. Permittivity distribution shows a stripe pattern in direction of approx. 30 • with regard to the x-axis. The probability density functions is plotted in Fig. 11. In contrast to conductivity, permittivity is almost normally distributed (ε r = 5.3 ± 0.85). Due to the stripe pattern which can be seen in Fig. 10, variogram analysis was performed in different directions (Fig. 11). The correlation length in x-direction is longer than in y-direction. An anisotropic variogram model was fitted with the maximum correlation length a max being measured parallel to the stripes (30 • ) and the shortest correlation length a min in the perpendicular ...
Context 6
... shows a stripe pattern in direction of approx. 30 • with regard to the x-axis. The probability density functions is plotted in Fig. 11. In contrast to conductivity, permittivity is almost normally distributed (ε r = 5.3 ± 0.85). Due to the stripe pattern which can be seen in Fig. 10, variogram analysis was performed in different directions (Fig. 11). The correlation length in x-direction is longer than in y-direction. An anisotropic variogram model was fitted with the maximum correlation length a max being measured parallel to the stripes (30 • ) and the shortest correlation length a min in the perpendicular direction whereby the anisotropy factor is a max /a min = 5. The ...
Context 7
... was fitted with the maximum correlation length a max being measured parallel to the stripes (30 • ) and the shortest correlation length a min in the perpendicular direction whereby the anisotropy factor is a max /a min = 5. The determined parameters of the anisotropic variogram model are used for kriging the data. The results are depicted in Fig. 12( left). The periodical structure and the resulting anisotropy are probably caused by the former cultivation of the area. Until two years before the measurements were carried out on the grassland, the area had been used as cropland and the direction of the regular pattern corresponds to the former direction of cultivation. The periodical ...
Context 8
... pattern corresponds to the former direction of cultivation. The periodical structure might be a relict of ploughing which induced compaction of the soil. Else, it might be related to the grass roots and an augmented evapotranspiration. Identical measurements were carried out on an other location which was used as grassland for at least 35 years ( Fig. 12 (right)). Variogram analysis (not shown here) yields an isotropic spatial pattern with a correlation length of 0.35 m. Table 1. Electromagnetic properties of the mines and the soils used for FD simulations. The mean, the standard deviation (std), the coefficient of variation (cv) and the range a of the exponential variogram function are listed ...
Context 9
... model and might change the back-scatter cross section of the mine but not the principle results of the analysis. The physical properties of soils and mines are itemised in Table 1. In model a)-c) permittivity is constant and conductivity is spatially variable whereas in model d)-f) permittivity is variable and conductivity is held constant (see Fig. 13). In model a) conductivity and in model e) permittivity variations are close to the variability determined in field (see ...
Context 10
... FD calculation is performed to simulate GPR measurements with an antenna offset of 10 cm and a centre frequency of 1.5 GHz. 21 Besides a gain function to counterbalance for geometrical spreading, no further processing was applied to the data to assure a proper comparability of the results (Fig. 13, right). If soil permittivity is homogeneous and conductivity is low as is typical for sandy soils, the three mines can clearly be recognised ...
Context 11
... Su] [ 55] [ Su] [ Su] [ Su] their diffraction hyperbolas (Fig. 13 a). If conductivity is heterogeneous and relatively high, the signals of the three mines are damped differently. The mines can still be detected at midrange conductivities (Fig. 13 b) which are typical for e.g. silty soils. Mine detection will be difficult in soil c) as the signal of the left mine is completely damped because it is placed ...
Context 12
... Su] [ 55] [ Su] [ Su] [ Su] their diffraction hyperbolas (Fig. 13 a). If conductivity is heterogeneous and relatively high, the signals of the three mines are damped differently. The mines can still be detected at midrange conductivities (Fig. 13 b) which are typical for e.g. silty soils. Mine detection will be difficult in soil c) as the signal of the left mine is completely damped because it is placed in a highly conductive region with σ > 0.1 S/m. These are typical values for wet salty or clayey soils. If soil permittivity is heterogeneous, the form and absolute traveltime of ...
Context 13
... which interfere with the signals from the mines. The reflections caused by permittivity variations of the models are stronger than the reflections caused even by considerable conductivity variations. If the contrast of the mines to the soil is high as is the case for moist soil with high permittivities, the mine signal is still clearly visible (Fig. 13 f). For decreasing water content, the contrast is getting smaller and mines are difficult to detect (Fig. 13 e). This model corresponds to permittivity variations which were determined in situ for sandy soils. If water content keeps on decreasing, the mines cannot be detected any more (Fig. 13 ...
Context 14
... models are stronger than the reflections caused even by considerable conductivity variations. If the contrast of the mines to the soil is high as is the case for moist soil with high permittivities, the mine signal is still clearly visible (Fig. 13 f). For decreasing water content, the contrast is getting smaller and mines are difficult to detect (Fig. 13 e). This model corresponds to permittivity variations which were determined in situ for sandy soils. If water content keeps on decreasing, the mines cannot be detected any more (Fig. 13 ...
Context 15
... the mine signal is still clearly visible (Fig. 13 f). For decreasing water content, the contrast is getting smaller and mines are difficult to detect (Fig. 13 e). This model corresponds to permittivity variations which were determined in situ for sandy soils. If water content keeps on decreasing, the mines cannot be detected any more (Fig. 13 ...
Citations
... We assume that the surrounding medium is homogeneous and has a relative permittivity ε r = 5, a relative permeability µ r = 1 and an electric conductivity σ 0 = exp(−3) ∼ 0.05. These values correspond to the permittivity, permeability and conductivity of the soil [61]. The size ofΩ is taken as 2 m in width and 2 m in depth. ...
... We work with different types of inclusions, still following the parameters from [61]: ...
We propose to explore the Time-Reversed Absorbing Condition (TRAC) method in the case of dissipative homogeneous media. In previous work, the TRAC method was derived from the time-reversibility of the (undamped) wave equation and proved to be efficient in both the time domain and the frequency domain. Namely, two main utilisations of the TRAC method have been probed: (a) redatuming, i.e., moving virtually the measurements by reconstructing the wavefield and (b) tracking down the location of a possible inclusion inside the domain.
In this paper, we focus on the redatuming application and investigate the feasibility of the TRAC method in the case of dissipation. In particular, we will see that performing the classical TRAC method, i.e., ignoring the dissipation, may give satisfactory results, even for larger values of dissipation.
An analysis is provided in the frequency-domain and one-space dimension and shows satisfactory updated versions of the TRAC method. Moreover, a systematic error study in two-space dimension is illustrated via numerical examples.
... Dielectric Remote Sens. 2019, 11, 1232 3 of 16 permittivity of the soil is important since a strong contrast between soil and mines makes detection possible [10][11][12][13]. For successful humanitarian demining, it is necessary to understand the range and lateral variability in electromagnetic properties of the local soils [14]. General data on local soils are often available in many parts of the world, but for demining, specific and detailed data are required for detection system design, and during data processing for field operations. ...
... To make measurements at a frequency closer to the design COF of 2 GHz, and to address the partitioning of the transmitted radar signal at the ground surface for the air-coupled impulse GPR, another estimate of permittivity was desired. There are several GPR-based methods for measuring soil permittivity [30]: Recording of signals scattered by small conductive objects in the subsurface [31,32], common midpoint and common-offset reflection methods [33], and the groundwave technique [14,34]. A new method for location of subsurface objects and simultaneous estimation of electromagnetic wave velocity in the ground (which is as important for mine location than permittivity alone) based on analyses of times-of-flight for signals from an impulse GPR transmitter (Tx) to a subsurface reflector Remote Sens. 2019, 11, 1232 6 of 16 and back to four receiver (Rx) antennas distributed about the transmitter was proposed in [35]. ...
... To make measurements at a frequency closer to the design COF of 2 GHz, and to address the partitioning of the transmitted radar signal at the ground surface for the air-coupled impulse GPR, another estimate of permittivity was desired. There are several GPR-based methods for measuring soil permittivity [30]: Recording of signals scattered by small conductive objects in the subsurface [31,32], common midpoint and common-offset reflection methods [33], and the groundwave technique [14,34]. A new method for location of subsurface objects and simultaneous estimation of electromagnetic wave velocity in the ground (which is as important for mine location than permittivity alone) based on analyses of times-of-flight for signals from an impulse GPR transmitter (Tx) to a subsurface reflector and back to four receiver (Rx) antennas distributed about the transmitter was proposed in [35]. ...
To design holographic and impulse ground penetrating radar (GPR) sensors suitable for humanitarian de-mining in the Donbass (Ukraine) conflict zone, we measured critical electromagnetic parameters of typical local soils using simple methods that could be adapted to any geologic setting. Measurements were recorded along six profiles, each crossing at least two mapped soil types. The parameters selected to evaluate GPR and metal detector sensor performance were magnetic permeability, electrical conductivity, and dielectric permittivity. Magnetic permeability measurements indicated that local soils would be conducive to metal detector performance. Electrical conductivity measurements indicated that local soils would be medium to high loss materials for GPR. Calculation of the expected attenuation as a function of signal frequency suggested that 1 GHz may have optimized the trade-off between resolution and penetration and matched the impulse GPR system power budget. Dielectric permittivity was measured using both time domain reflectometry and impulse GPR. For the latter, a calibration procedure based on an in-situ measurement of reflection coefficient was proposed and the data were analyzed to show that soil conditions were suitable for the reliable use of impulse GPR. A distinct difference between the results of these two suggested a dry (low dielectric) soil surface, grading downward into more moist (higher dielectric) soils. This gradation may provide a matching layer to reduce ground surface reflections that often obscure shallow subsurface targets. In addition, the relatively high dielectric deeper (10 cm–20 cm) subsurface soils should provide a strong contrast with plastic-cased mines.
... The influence of the homogeneity of the soil on GPR for landmine detection is another usual handicap that must be taken into account (Igel and Preetz 2009). In this frame, Takahashi et al. (2012) explain how landmine detection by GPR becomes challenging when soil is inhomogeneous. ...
This paper is a review of methods related to assessment of construction details and material properties using of GPR. The focus is on recent research activities of the Project “Innovative procedures for effective GPR inspection of construction materials and structures” (project 2.4) in COST Action TU1208. The electromagnetic properties of investigated media are interesting because they reflect physical features of the materials (e.g. their composition), enabling a non-invasive inspection of their condition. Moreover, the assessment of electromagnetic properties (e.g. wave velocity) is an inherent part of any GPR structural study necessary for correct depth determination or amplitude interpretation. As a result of the review major directions of research are highlighted and some benefits and limits of different approaches are described.
... One of the improvements made in recent years, is the joint application of metal detectors together with ground penetrating radar sensors (GPR), which was initially used in landmine detection systems (Igel et al., 2009). Since IED's are often made from non-metallic material, electromagnetic induction sensors (EMI) like metal detectors are not applicable anymore. ...
... Especially in high-frequency investigations, as for IED and landmine detection, the intrinsic attenuation plays an important role. In previous studies we observed that frequency-dependent clutter causes unwanted energy reflections from soil heterogeneities (Igel et al., 2009). Apart from that unwanted energy scattering, it is intrinsic attenuation that detracts the radar wave energy from the system. ...
One of the soil properties influencing the sensing depth of ground-penetrating radar (GPR) is intrinsic attenuation. Especially in high-frequency investigations, as for explosive ordnance and landmine detection, it plays a major role. In many cases attenuation is the limiting factor that determines the applicability of the GPR system. In order to investigate the frequency-dependence of electromagnetic soil properties, different soil samples from Germany and Afghanistan were analyzed. For this purpose, we used a coaxial transmission line together with a vector network analyzer in the laboratory. Two coaxial line cells were used in order to determine the complex dielectric permittivity in the 1 MHz to 10 GHz frequency range. The complex permittivity curves were fitted by a generalized model, which accounts for different dielectric relaxation mechanisms. The inversion of the model was carried out by means of the Geophysical Inversion and Modelling Library (GIMLi). Splitting the measured complex dielectric permittivity data using the generalized model made it possible to investigate the different energy loss contributions to the intrinsic attenuation, which are in turn attributed to certain soil components. Based on the laboratory results, we intend to create a simple method for the prediction of the GPR performance in the field.
... Van Dam et al. (2004) carried out measurements along transects in different tropical/subtropical countries and found the magnetic susceptibility to vary by a factor of 2-5. This finding is consistent with our measurements of the spatial distribution of soil susceptibility on former landmine fields in Mozambique, showing the same variability with correlation lengths between one and a few metres (Igel & Preetz 2009). To the best of our knowledge, no measuring devices are available that can be used to determine the frequency dependence of susceptibility in the field, and there is a lack of measurements on the spatial variability of this property. ...
Electromagnetic induction (EMI)-based metal detectors are the most widely used sensing techniques in landmine clearance operations; however, they are negatively influenced by frequency dependence of magnetic susceptibility. A total of 466 rock and soil samples collected from across the tropics are investigated in this study. The data show that frequency-dependent susceptibility depends on the parent material as well as on the degree of weathering. Ultramafic and mafic rocks and their derivatives have higher susceptibility and absolute frequency dependence than material originating from intermediate, felsic and sedimentary rocks. Within each parent material group, absolute frequency dependence increases steadily with increasing alteration from unweathered rock to topsoil. This effect is likely due to either the residual enrichment of weathering resistant ferrimagnetic minerals including superparamagnetic (SP) grains, the comminution of larger ferrimagnetic minerals or the neoformation of SP minerals during soil formation.
Relative frequency dependence is generally lower than 15 per cent for the investigated samples with a few exceptions. It increases with alteration for igneous rocks but remains at the initially high level for sediments. This finding indicates that the relative concentration of SP minerals changes with respect to the total magnetic fraction for igneous rocks but remains constant for sediments. Soils derived from ultramafic, mafic and intermediate rocks show low relative frequency dependence, and their magnetic susceptibility is mainly the result of multidomain lithogenic minerals. In contrast, soils derived from felsic rocks and sediments show the highest values, and their susceptibility is due to SP minerals that are either formed during pedogenesis or residually enriched.
The average and extreme values of the absolute frequency dependence within each subgroup, based on parent material and alteration grade, are used to design a prognosis system for assessing the impact of the subsurface on EMI sensors for landmine detection. In general, intermediate, felsic and sedimentary rocks have no influence on the detectors and only a weak influence in the most extreme cases. Soils derived from these rocks typically have no influence; however, they can have a very severe influence in a few cases. In contrast, ultramafic and mafic rocks typically have a moderate influence and a very severe influence in extreme cases, with the associated soils resulting in a severe influence in general. The deduced prognosis system can be used by demining organizations to help them predict metal detector performance in tropical regions on the basis of geologic and/or soil maps, which do not supply information on electromagnetic properties. In this way, such a system may eventually help with the planning of demining campaigns and selection of appropriate sensors.
... Van Dam et al. (2004) carried out measurements along transects in different tropical/subtropical countries and found the magnetic susceptibility to vary by a factor of 2-5. This finding is consistent with our measurements of the spatial distribution of soil susceptibility on former landmine fields in Mozambique, showing the same variability with correlation lengths between one and a few metres (Igel & Preetz 2009). To the best of our knowledge, no measuring devices are available that can be used to determine the frequency dependence of susceptibility in the field, and there is a lack of measurements on the spatial variability of this property. ...
Magnetic susceptibility of soils is mainly determined by their content of ferrimagnetic minerals whereas titanomagnetite, magnetite and maghemite being the most important ones. Titanomagnetite and magnetite are of magmatic origin, i.e. they crystallise during cooling of iron-rich magma and are part of many igneous rocks. Maghemite and sometimes magnetite are of pedogenic origin. They develop by crystallisation of dissolved iron during soil forming processes. Ferrimagnetic minerals that are smaller than some tens of nanometres are superparamagnetic (SP) and show frequency dependent susceptibility. SP minerals crystallise if magma cools down rapidly (e.g. volcanic magmas, glasses and ashes) and are frequently formed during pedogenesis. In order to investigate the origin and formation of SP minerals in tropical soils, we analyse magnetic properties of 594 samples from the entire tropics comprising the whole range of weathering states from unweathered rock to highly weathered soil. Tropical soils are subject to intense chemical weathering and are rich in ferrimagnetic and in particular SP minerals. The process leading to a high content of these minerals is either residual enrichment due to their weathering resistance or neo-formation. In this study we focus on the frequency dependent susceptibility (absolute and relative) of the samples and classify it according to the parent material and alteration. We observe that • within each parent-material group, rock material shows in general lower susceptibility and absolute frequency dependence than soil material • ultrabasic and basic/intermediate rocks and soils developed from these rocks show high absolute frequency dependent susceptibility and, in contrast, acid rocks and sediments show lower absolute frequency dependence • absolute frequency dependence increases from unweathered rock to weathered rock, and from subsoil to topsoil material within every group of parent material • relative frequency dependence rises successively with weathering for ultrabasic, basic/intermediate and acid igneous parent material, but, it tends to decrease for clay/clay slate and sandstone. Based on the above observations we conclude that the content of SP minerals depends on both: parent rock and alteration of the material. The total amount of SP minerals rises during weathering, regardless of the parent material. The process is either preferential accumulation of weathering resistant magnetic minerals, including the ultra-fine grained fraction, or neo-formation of new magnetic minerals. The increase of relative frequency dependence of igneous rocks is a clear indication that SP minerals are formed during soil genesis. However, for some sedimentary rocks, the amount of SP minerals is already high and is not subsequently increased further during weathering. Electromagnetic induction (EMI) based metal detectors are the most widely used sensing techniques in landmine clearance operations. They are negatively influenced by magnetic susceptibility and its frequency dependence. In particular tropical soils show to have a negative impact on EMI sensors. Besides, the tropics are the regions which are most affected by landmines where most of the humanitarian demining-activities concentrate. Currently, no soil classification system exists that helps to predict the influence of frequency dependent susceptibility on landmine detection. We deduce a system that can be used to predict the soil impact depending on parent material and weathering. Our system can be consulted by demining organisations to predict metal detector performance in tropical regions based on geologic and soil maps. Ultra-basic, basic and intermediate igneous rocks have a moderate influence on EMI detectors in average cases and a very severe influence in extreme cases. Soils developed from these rocks have a severe or very severe influence. In contrast, acid igneous rocks and sediments do not influence EMI detectors severely. Soils developed from these rocks have no influence in average cases; however, they may have a very severe influence in extreme cases.
This Chapter deals with a compilation of published works in the frame of the applications of the GPR for humanitarian assistance and security. The fields of application, in which the technique has experienced more advances, are the detection of mines and unexploded ordnances, as well as the location of underground spaces. Different types of defensive constructions were built throughout history to protect people from natural catastrophes, aerial bombardments and other attacks. Moreover, the use of the GPR technology in rescue operations is considered by including the main contributions in locating human remains or living victims in disaster areas. An overview of the main GPR works in humanitarian missions and their results are therefore mentioned. Specific systems, methodologies and processing algorithms developed in these applications are also analysed. As result, the method has shown significantly benefits when compared to other traditional searching methods.
Metal detector has commonly been used for landmine detection and ground-penetrating radar (GPR) is about to be deployed as dual sensor that is in combination with metal detector. Since both devices employ electromagnetic techniques, they are influenced by magnetic and dielectric properties of soil. To observe the influence, various soil properties as well as their spatial distributions were measured in four types of soil where a field test of metal detectors and GPRs took place. By analyzing soil properties these four types of soil were graded based on the estimated amount of influence on the detection techniques. The classification was compared to the detection performance of devices obtained from the blind test and a clear correlation between the difficulty of soil and the performance was observed; the detection and identification performance were degraded in soils that were classified as problematic. Therefore, it was demonstrated that the performance of metal detector and GPR for landmine detection can qualitatively be assessed by geophysical analyses.
Metal detectors have commonly been used for landmine detection, and ground-penetrating radar (GPR) is about to be deployed for this purpose. These devices are influenced by the magnetic and electric properties of soil, since both employ electromagnetic techniques. Various soil properties and their spatial distributions were measured and determined with geophysical methods in four soil types where a test of metal detectors and GPR systems took place. By analysing the soil properties, these four soils were classified based on the expected influence of each detection technique and predicted soil difficulty. This classification was compared to the detection performance of the detectors and a clear correlation between the predicted soil difficulty and performance was observed. The detection performance of the metal detector and target identification performance of the GPR systems degraded in soils that were expected to be problematic. Therefore, this study demonstrated that the metal detector and GPR performance for landmine detection can be assessed qualitatively by geophysical analyses.
Knowledge of ground wave penetration depth and methods for facilitating ground wave velocity analysis are important practical aspects to consider when measuring soil water content with surface ground penetrating radar. A field study was conducted to optimize the wide angle reflection and refraction and fixed offset methods of measuring the ground wave velocity and to find the effective ground wave sampling depth under irrigation and drainage conditions. In this study, a PulseEkko 1000 GPR system with 450 MHz antennas was used at a sandy loam soil site. Water contents measured with time domain reflectometry (TDR) were used to determine the sampling depth of GPR based water content estimates. Cumulative irrigation and drainage calculated with GPR were found to be more closely related with cumulative irrigation and drainage measured with shorter TDR probes, suggesting that the ground wave sampling depth was in the 0.2–0.5 m range. During the infiltration phase the depth of the ground wave penetration was found to be in the 0–0.56 m range, assuming a sharp boundary between wet and dry sand. Comparison of water contents measured with the WARR and FO methods revealed that an antenna separation distance of 1.5–2.0 m for the FO method was required to obtain similar results between the two methods.
