The characterization of magnetic susceptibility (MS) has become an accepted technique in soil science. This review examines the concept of volume and mass-specific MS, magnetism, frequency dependence, and thermal behavior of MS, as they pertain to soil material. A comparison is presented of the two types of instrumentation for measuring soil MS, based on magnetic field and electromagnetic induction (EMI). These are discussed with respect to applications including magnetic granulometry, detection of pollutants, identification of organic matter, the delineation of drainage class, paleo-environmental studies, archaeology, as well as soil erosion and degradation. Instruments that use magnetic fields can precisely measure the MS of small amounts of soil, thinly deposited layers or soil exposures, but cannot effectively measure materials at distances ≥10 cm from the sensor. EMI instruments, instead, are capable of quickly measuring apparent MS of a finite volume of the soil, and are utilized in mapping of soil MS in agricultural and archaeological investigations; however, the measured apparent MS values need to be further processed to give the real volume MS values of soil layers/segments. Although both kinds of instruments are widely used in soil science, their measured data are not interchangeable. Future work should be conducted to increase the understanding of the comparability of these instruments to find better utility among soil scientists.
Introduction
Magnetic susceptibility (MS) is a physical property of matter, defined as the extent to which a material can be magnetized. The first use of the concept in soil science goes back to the work of Le Borgne (1955), who found that the MS increases from parent material to the subsoil, then to the topsoil. This MS enhancement with decreasing depth was attributed to the changes in clay-size fraction, which experienced conversion of weakly magnetic forms of iron oxides and hydroxides to maghemite, or magnetite, during successive oxidation-reduction processes within the soil. Le Borgne (1960) also observed and reported MS enhancement by fire. Since then, researchers have pursued the mechanisms by which MS is changed in the topsoil. Lukshin et al. (1968) and Vadyunina and Babanin (1972) have demonstrated how susceptibility enhancement is linked to key soil forming processes and may be utilized to get a broad understanding of the processes impacting iron minerals throughout pedogenesis, as well as particular impacts such as gleying. The influence of lithology (Mullins and Tite 1973) and climate (Tite and Linington 1975), on MS changes, has been also reported. Mullins (1977) summarized the factors affecting MS of the soil and concluded that soil MS depends on the size, shape, and concentration of magnetite and maghemite; as well as on the method of measurement. He also observed that pedogenic maghemite can be developed in soils with low MS parent material. A study of the relations between magnetic minerals and soil forming processes was done by Maher (1986). Based on isothermal measurements of mineral magnetic parameters, Maher (1986) concluded that the presence of magnetite and maghemite within the soil is widespread. It has also been shown that magnetotactic bacteria, which exist in organic matter, can generate ultrafine-grained magnetite in the presence or absence of oxygen (Løvely et al. 1987; Maher and Taylor 1988). Schwertmann (1988), Schwertmann and Taylor (1989), and Singer et al. (1996) separately studied the factors affecting MS enhancement in soils and deduced that temperature and annual precipitation, as well as parent material, are other main factors that impact soil MS.
The majority of early measurements of soil MS were based on the analysis of a magnetic hysteresis curve, in which the magnetization values of matter (M), obtained in the laboratory by employing magnetic fields (H) in different intensities, were plotted against magnetic field intensity. The volume MS of matter, commonly represented by the Greek letter kappa “κ,” was the ratio of magnetization to applied magnetic field intensity such that κ = M/H. It was also realized from magnetic hysteresis curves that MS is almost constant in low magnetic field intensities, but increases nonlinearly to a maximum in high intensity fields (Jackson et al. 1998). With innovations in computers and electronic hardware, as well as digital technologies, various MS instruments have been designed and produced by different manufacturers, and are being used by soil scientists. Two basic types of instrumentation are currently available to measure soil MS: those that employ magnetic fields and those that utilize electromagnetic induction (EMI) methods. More explanations regarding these MS instruments will be presented in Section 3.
Measurements of MS have a variety of applications in soil sciences. These include soil drainage assessments (e.g., de Jong 2002; de Jong et al. 2005; Owliaie et al. 2006; Grimley et al. 2008; Asgari et al. 2018; Gholamzadeh et al. 2019), magnetic granulometry using frequency dependence (FD) of MS (e.g., Hrouda et al. 2013; Ustra et al. 2018, 2019), soil pollution (Schibler et al. 2002; Petrovský et al. 2004; Zawadzki et al. 2010, 2012; Boadi et al. 2014; Cao et al. 2015; Liu et al. 2016; Rachwał et al. 2017; Wang et al. 2021), soil erosion (Venture et al. 2001; Jakšík et al. 2016; Menshov et al. 2018; Liu et al. 2019; Ding et al. 2020), land degradation (Sadiki et al. 2009; Łukasik et al. 2015; Magiera et al. 2019), and paleoclimate reconstruction (Maher and Thompson 1995; Maxbauer et al. 2016; Jordanova and Jordanova 2021). While almost all of these studies have focused on a specific application of a particular instrument, other publications have addressed the comparison between MS values measured using various handheld and (or) laboratory-based MS instruments for specific applications (e.g., Benech and Marmet 1999; Lecoanet et al. 1999; Simpson et al. 2009; Simpson et al. 2010; Lee and Morris 2013; Deng and Smith 2016; Grison et al. 2017). A comprehensive review of different types of instruments for measuring soil MS and their specific applications in soil science has not yet been carried out. Recently, Jordanova (2017) published a book regarding applications of MS in pedology, agriculture, and environmental sciences. Although this book dealt with applications of soil MS in detail, the MS instruments used for measuring soil MS and comparison between their applicabilities were not explained therein.
Despite the fact that both types of instruments can measure soil MS, not all devices are suitable for all of the aforementioned uses due to some constraints like measurement accuracy and speed, as well as physical differences in designing MS instruments. For instance, while EMI instruments can quickly measure apparent MS of topsoil, the measured values need to be further analyzed and inverted to give real MS values of soil layers; however, the speed of measurement makes the EMI instruments suitable for mapping soil MS, which is the main advantage of using these kinds of instruments for measuring soil MS. On the other hand, although the instruments that use magnetic fields for measuring soil MS give precise values for soil samples, such measurements are costly and time-consuming. Moreover, they are not designed for mapping purposes. Accordingly, the main objectives of this paper are: (i) to provide an overview of relevant fundamental concepts of MS, with emphasize on MS of soil constituents; (ii) to conduct a review of current analytical solutions (or instrumentation) for measuring soil MS; and (iii) to discuss the suitability of available instruments for characterizing soil MS.
Magnetic Susceptibility of Soil Constituents
Volume vs. mass-specific MS
The fundamental rules and concepts of MS have long been clarified. Each electron in an atom has magnetic moments due to its orbital and spin motions. When a material is exposed to an incident uniform magnetic field (H), it will be magnetized due to alignment of these magnetic moments with the magnetic field. The ratio of magnetization (M) to the magnetic field intensity (H) is defined as “volume MS” of that material (κ), which is a dimensionless quantity defined as:
Because this value directly depends on the amount of the material, dividing this value by the bulk density (ρ) of the material provides a “mass-specific MS” (χ), which is an intrinsic property of the matter having dimension of volume/mass, defined as (Mullins 1977):
Direct MS measurements of intact soil, such as soil core samples, are reported as volume MS. This is because the weight of the soil, which in this case interacts with instrument magnetic field, is unknown. On the other hand, for powder or crushed soil samples, a specified amount of soil sample is used, from which the mass-specific MS of the sample can easily be calculated.
Magnetism
In some atoms, the arrangement and number of electrons are such that the magnetic moments cancel each other. If such atoms are placed in a magnetic field, the rotation of electrons, according to the Lorentz force, produces a weak magnetic moment in the opposite direction of magnetic field (Evans and Heller 2003); this effect is called “Diamagnetism.” So, diamagnetic materials have weak negative MS. Quartz, feldspars, calcium carbonate, organic matter, plastic, and water are most common diamagnetic materials encountered in the soil. If the magnetic moments in the atom are partially canceled out, the atom has a permanent magnetic moment; this effect is known as “Paramagnetism.” While such atoms tend to be aligned with external magnetic fields, the thermal energy always limits this alignment. So, paramagnetic minerals have weak positive MS. Silicates and aluminosilicates containing iron, as well as biotite and olivine, are examples of paramagnetic minerals in the soil. In the pure form of some elements, like iron, nickel, and cobalt, not only there is a permanent magnetic moment in each atom, but there is also a strong interaction between the adjacent moments, which eventually results in a strong magnetic moment. This phenomenon is known as “Ferromagnetism.” Since the natural appearance of pure iron is rare in the earth’s crust and soil, the iron must occur in other mineral forms in the nature. If atomic moments in such materials occur in the opposite direction but have the same strength, the material will have zero net magnetic moment in the absence of an applied magnetic field. This effect is called “Antiferromagnetism”; goethite and hematite are two important antiferromagnetic minerals in the soil. Although these minerals are “canted” antiferromagnetic or weakly ferromagnetic above ∼250 K (or ∼−23 °C), and exhibit high MS at room temperatures due to spin or defective spin compensations, they have zero net magnetization below this temperature (Dzyaloshinsky 1958). “Ferrimagnetism” is another phenomenon of iron bearing materials, on which exchange coupling acts. In this case, the adjacent magnetic moments are in the opposite direction, but one of them has a higher strength than the other, which give rise to a strong pure magnetic moment. Magnetite, maghemite, pyrrhotite, and greigite are the most well-known ferrimagnetic minerals in the soil (Mullins 1977). By comparison, the MS of diamagnetic matter is typically a hundred times smaller than that of paramagnetic matter, and a hundred thousand times smaller than ferromagnetic matter (Evans and Heller 2003). A detailed explanation of magnetic materials can be found in Butler (1998) and Thompson and Oldfield (1986).
The MS of natural soil, which is extremely variable, is almost completely controlled by the content of ferrimagnetic minerals, their grain sizes, and their distribution in the soil. Although various attempts have been made to express the relationship between susceptibility and the content of ferrimagnetic minerals, there is still no universal agreement (Parasnis 1986). Table 1 contains mass specific MS values for common minerals, as well as volume MS values for common rocks that exist on the earth’s surface. Soils developed from igneous rocks have relatively higher volume MS values than those developed from metamorphic and sedimentary rocks. In fact, the MS value of soil directly reflects MS of parent material. Table 1 also shows some small negative mass-specific MS values, which belong to diamagnetic minerals.
Table 1.
Magnetic susceptibility values of some common soil constituents (data from Carmichael 1989; Dearing 1999; Reynolds 2011).

FD of MS
The MS of materials is frequency dependent due to the relaxation time (τ) of the magnetic dipoles, where increasing the frequency of the measuring magnetic field results in a decrease in the MS value (Debye 1929). As a range of fine to coarse grained minerals and rocks, soil includes superparamagnetic (SP), single domain (SD), and multi-domain (MD) magnetic particles, as well as organic matter. All these particles contribute to the soil MS; however, when using a low intensity magnetic field, in which there is a linear relationship between magnetic field (H) and magnetization (M), it has been observed that the soil MS decreases when the frequency of the applied magnetic field increases (Mullins and Tite 1973). This means that high frequency MS (χhf) is usually less in magnitude than low frequency MS (χlf). The phenomenon is thought to be caused by the presence of fine-grained particles in the soil. At low frequencies, almost all magnetic particles follow the direction of the applied alternating magnetic field, while at higher frequencies, the SP grains, which are typically less than ∼20 nm in diameter, cannot be fully aligned with the alternating magnetic field and result in lower MS reading (Thompson and Oldfield 1986). Thus, the difference between low and high frequency MS can be an indicator of the presence of SP grains. This frequency effect is dependent on the specific frequency used in designing experimental instruments, and commonly used to qualitatively detect soil samples containing SP grains. Most companies tune their instruments to the frequency range of 1 to 20 kHz because the relaxation of SP minerals generally occurs in this range (Ustra et al. 2018). There are no specific values to be considered as low and high values for frequencies, so the best way is to normalize this difference to the lower frequency MS value to achieve a dimensionless quantity. By convention this quantity, shown by χfd, is referred to as the “percent frequency dependence of MS” or “percentage loss of susceptibility” (Maher 1986; Dearing et al. 1996) and is used to delineate soil grain sizes, defined as:
By measuring χ in three different frequencies and utilizing an inversion procedure, Ustra et al. (2018 and 2019) could calculate real χlf and χhf, that is, the values of χ in lowest and highest frequencies, as well as relaxation time (τ) for some soil samples. The lowest and highest frequencies, as well as the relaxation time, also vary for different soil samples. FD of MS plays an important role in soil studies, mostly to discriminate between lithogenic and pedogenic soil iron oxides, and is directly related to grain size. More explanation for this discrimination is presented in Section 4.1.
Values for MS are not only a frequency-dependent feature of materials but also a complex value. This means that the measured magnetization has two components: one in-phase and another out-of-phase with the incident magnetic field (H). Studies show that out-of-phase MS can be used to distinguish the portion of magnetic grains greater than and less than ∼20 nm, in diameter (Egli 2009; Chadima et al. 2010). The relationship between in-phase (IP) and out-of-phase (or quadrature-phase, QP) MS components at only one frequency was studied, in detail, by Hrouda et al. (2013) where they concluded that this relationship is highly correlated with χfd derived from MS values measured at two different frequencies. It means that χfd can be calculated by measuring MS, even at one frequency, as long as the measurement includes IP and QP MS components.
Thermal behavior of soil MS
In the last three decades, along with the study of soil MS in different frequencies, investigations have been conducted under continuous heating, up to 700 °C, followed by subsequent cooling. This method has become commonly used among soil scientists. The observed Curie temperatures, during these processes, are used to identify magnetic minerals in natural samples (Jordanova and Jordanova 2016). The cooling curve reveals changes in the mineral composition imposed by laboratory heating. Increases in MS, at temperatures above 420–450 °C, have been observed when studying the lacustrine deposits and loesses (Deng et al. 2005; Minyuk et al. 2011). This enhanced MS is due to the conversion of iron oxides in the soil from the weakly ferrimagnetic form, hematite, to a strongly ferrimagnetic form, magnetite, when the soil is heated, followed by re-oxidation of magnetite to maghemite, due to reoxidation during cool-down (Le Borgne 1955 and 1960). The form of the heating and cooling curves of MS vs. temperature (κ-T curves), as well as the observed peaks, are interpreted as various environmental processes. These processes include the neoformation of magnetite, and other pedogenic magnetic minerals, from iron bearing silicates or clay minerals (Liu et al. 2005), reactions of Fe minerals with organic matter (Hanesch et al. 2006), and time- and temperature-dependent cation ordering in magnesioferrites (Harrison and Putnis 1999) and titanomagnetites (Bowles et al. 2013). Continuous enhancement of χfd has also been reported by Owliaie et al. (2006), which attributed this phenomenon to the formation of fine-grained SP particles with high thermal stability from destruction of coarse-grained MD minerals. This approach is a sensitive tool for identifying the magnetic, and some other iron bearing, minerals which contain information about the origin and diagenesis of soils. Since the MS of heated samples must be measured under controlled conditions, the instrumentations for measuring MS in this case are laboratory-based instruments: typically the Bartington MS2W, or the AGICO KLY and MFK series. Natural or man-made fire can also enhance MS of materials at the soil surface. Soil MS can be significantly increased by exposing soils to high temperatures by burning and incorporating burned or other high susceptibility components into the soil matrix. Consequently, MS tools may be used to study archaeological sites where soil that has been exposed to anthropogenic sources of fire often has higher MS values than surrounding soils. In-field mapping of surface MS can provide maps to assist archaeologists to find ancient habitats of humanity (Dalan et al. 2017).
Measurement of Soil Magnetic Susceptibility
Two basic types of instrumentation are currently available to measure the MS of soils: those that employ a magnetic field or those that utilize EMI methods. In this basis, different companies have designed and produced MS sensors, which have different specifications like shape, size, and type of coils, operating frequency, and recording of measured components. Some of these instruments are handheld and can operate in the field, but some are stationary and need to be used in the laboratory. Here, we briefly explain the physics behind the two different methods, which have been mostly used in instruments to measure soil MS.
Instruments that use magnetic field for measuring MS
Consider a magnetic field having the intensity of H (units of Ampere/metre; A/m) passing through free space. In this case, the magnetic field induction, B, in the free space is B = μ0H (tesla, T) in which μ0 = 4π × 10−7 T. m/A is the magnetic permeability of free space. Magnetic permeability is the physical property of a material, which describes the degree of induced magnetism the material experiences under the influence of an external magnetic field. Now, assume that a specimen, with volume MS of κ, is exposed to the magnetic field of intensity H and magnetized as M = κH. In this case, the specimen acts as a magnet and has its own magnetic field. The magnetic induction (B) is:
which can be expressed as B = μH where μ = μ0(1 + κ) is the magnetic permeability of the specimen. Now, having μ0 and measuring μ, the volume MS of the specimen can be simply calculated as:All the instruments that employ a magnetic field for measuring MS use alternating currents and a kind of RLC circuit to measure μ, which includes a coil, having resistance R and inductance L, as an inductor, and a capacitor, having capacitance C. When the coil is in free space, the circuit has a known resonance frequency. When a specimen lies in the coil, the inductance of the inductor and in turn the resonance frequency of the circuit will shift. The difference in the resonance frequencies is then related to change in magnetic permeability of the specimen. Accordingly, by measuring the difference in resonance frequencies using electronic circuitry and analyzing it, the magnetic permeability inside the inductor, and then the MS of the specimen can be determined. This is a simple explanation of the physics behind MS measurements using magnetic field-based MS instruments. It is important to realize that the variation of μ is relatively small. Therefore, any thermally induced sensor drift, due to either variations of surrounding temperature or warming up by passing alternating currents through the sensor coil, needs to be eliminated. This elimination and correction is done by occasionally measuring a new “air” value (no specimen in place), to re-establish the μ0 reference (Bartington Instruments, OM-0408/50).
Instruments that use magnetic fields for measuring soil MS were originally designed to operate in the laboratory as stationary instruments. Some of these instruments have the ability to measure magnetic hysteresis curves for soil or rock samples by measuring magnetization of the sample in various magnetic field intensities (e.g., AGICO MFK2, having magnetic field intensities from 2 to 700 A/m), while others can operate at different frequencies. For example, the Bartington MS2D operates at 465 and 4650 Hz, while the AGICO MFK and KLY series operate at 976, 3904, and 15 616 Hz. The ability to measure MS of soil or rock samples in different temperatures is also an option for some kinds of instruments. For example, the Bartington MS2W operates in the temperature range of −200 to +850 °C, and the AGICO MFK and KLY series equipped with CS4 and CSL sensors operate at the range of −192 to +700 °C. Similar instruments have been particularly designed to measure MS of thin layers of soil by inserting their sensor in a sample hole (e.g., Bartington MS2H, and SM-400 from ZH Instruments). Many attributes, other than just MS, can be measured using different types of magnetic field-based MS instruments, such as remanent magnetization, coercive force, or Curie temperature, which are important in rock magnetism and paleomagnetism. Since the Curie temperature is different for different ferromagnetic minerals, measuring remanent magnetization of the soil samples (or equivalent MS of soil samples) while varying temperatures can assist to determine various types of ferromagnetic minerals in a soil sample (Butler 1998). Nowadays, handheld magnetic-based MS sensors are available, which can be utilized directly in the field. The penetration depth of the magnetic signal is limited and, therefore, only small amounts of soil or rock samples can be analyzed due to small loops (or pick-up coils) that these kinds of instruments utilize as a magnetic inductor. The results, however, are very precise. Table 2 presents different types of current MS instruments, which employ magnetic fields to measure MS, and their applications in soil sciences. Although the operating frequencies of such sensors are different, which can be reflected in the measurements, some authors have utilized different types of these sensors using the same soil samples, compare measurements and discuss advantages and disadvantages in their studies (e.g., Lecoanet et al. 1999; Lee and Morris 2013; Deng and Smith 2016; Grison et al. 2017). In general, if MS of the soil surface, soil exposures, or thin soil layers or horizons is required (e.g., in studying soil pollution), then handheld sensors with small loops/coils (MS2E, MS2F, MS2K, SM-20, SM-30, and KT-10) are sufficient because collection of soil samples is not required. Handheld sensors with larger coils (e.g., MS2D and Multi Kappa) are less affected by locally accumulated magnetic minerals because a larger amount of soil contributes to the measured MS values. So, these MS readings are somewhat “apparent” values, rather than MS values recorded with smaller coil sensors. On the other hand, there is a balance between the time of measurement and the resolution of recordings in MS mapping. While the smaller coil instruments yield detailed information about the soil MS, many more points need to be measured, comparing to the case using larger coil, to map MS of the same area, which is costly and time-consuming. If the study of soil MS variations with frequency or temperature is needed, laboratory-based instruments are definitely preferred since the instruments are equipped with particular sensors to measure MS of soil samples at different frequencies, temperatures, or magnetic field intensities.
Table 2.
Current instrumentsa that use magnetic field for characterization of soil MS. The specifications of each instrument as well as their applications in soil sciences are also provided.

EMI instruments
EMI instruments were initially intended to measure electrical conductivity (EC) of soil. All EMI instruments have at least one transmitter coil and one receiver coil, which are tens to hundreds of centimetres apart. An alternating current passes through the transmitter coil and produces an alternating primary magnetic field. This magnetic field penetrates the ground surface and induces electromagnetic currents in the soil materials below in response to the EC and MS of the bulk soil. These currents in turn produce the secondary magnetic field. The aggregation of primary and secondary magnetic fields is then sensed by the receiver coil. The primary magnetic field is a real absolute quantity (or a real number), but due to nature of the induction phenomenon, the secondary magnetic field is a complex number containing in-phase (or real) and quadrature-phase (imaginary or out-of-phase) components. Although the study of the effects of EC and MS, on the secondary magnetic field in homogeneous or layered earth, is straightforward, and can be easily done, the separation of EC or MS contributions to the secondary magnetic field is difficult. While this difficulty occurs when using high frequency currents in the transmitter, there are some assumptions to estimate apparent electrical conductivity (ECa) and apparent magnetic susceptibility (MSa) at low frequencies. The “skin depth” of an EM field is the depth to which the amplitude of that EM field reduces to 1/e of its value at the surface, where e ≈ 2.71 is the base of the natural logarithm. When the ratio of the transmitter–receiver distance to the skin depth of the EM field is much less than unity, this is called a “Low Induction Number” (LIN) condition. The induction number is actually proportional to the product of frequency (f), EC (σ), and magnetic permeability (μ) of the medium. It has been shown that, when using low frequencies, the MSa and ECa are proportional to the in-phase and quadrature-phase of the secondary to primary magnetic field ratio, respectively (Keller and Frischknecht 1966; McNille 1980; Thiesson et al. 2014). In this case, the apparent MS of the soil is twice that of the in-phase component of the secondary to primary magnetic field ratio (EM38 manual 2002; Thiesson et al. 2014), as long as the measurements are done at the surface of the soil. Although these are just estimations of EC and MS of the medium, it is enough to map these properties to delineate areas of low and high EC or MS. The main advantage of using EMI methods is that the measurements can be easily and quickly done because no contact is needed between the instrument and the soil surface. The instruments can also be attached to a mobile vehicle to measure the data in the field such that a high volume of data, corresponding to an extensive area, can be collected in a fast and low-cost way.
EMI instruments, currently used for measuring soil MS, consist of one transmitter and one to six receivers in different separations and configurations. All EMI instruments are handheld and can be readily moved in the field. Since depth penetration of such EMI instruments depends on the distance between the transmitter and receiver, as well as the configuration of transmitter and receiver coils relative to the surface, each of the receivers senses the MS of the soil in a different way. The MS in these kinds of instruments measures the volume MS of the soil (i.e., MS of a volume of soil, for instance, 1 m × 1 m × 1 m, and referred to “apparent MS”). On this basis, EMI instruments are strictly for field use and are not able to measure the MS of soil as laboratory samples. EMI instruments can measure MS of a large area, in a short time to produce a detailed map of soil MS, which is one of the important applications of EMI instruments.
It is worth mentioning that the only way to determine the MS of different horizons or layers of soil in the field, with this type of measurements, is to invert the MS measurements. Considering the soil as a multilayered media, in which each layer has its own thickness and MS, one can mathematically calculate EMI response of the soil at the surface. This procedure is reversible so that given apparent MS measurements collected at the surface, the MS of soil layers can be calculated, as long as different configurations of transmitter and receiver have been used for measuring MS at the surface. When transmitter and corresponding receiver coils are in the same plane, which is horizontal referring to the soil surface, the configuration is called horizontal coplanar (HCP). If the plane is vertical, with respect to the soil surface, the configuration is called vertical coplanar (VCP). The configuration in which the transmitter coil is horizontal, relative to the surface, but the receiver coil is vertical, or vice versa, is called perpendicular (PERP) configuration. Table 3 presents different EMI instruments currently used for measuring soil MS and presents some of their applications in soil science. While some researchers may utilize only one of the EMI instruments (e.g., Bevan 1998; Bevan and Dalan 2003; McNeill 2012; Guillemoteau et al. 2016), other researchers have compared the apparent MS measurements utilizing different types of instruments over the same area of interest (e.g., Benech and Marmet 1999; Simpson et al. 2009; Simpson et al. 2010). Note that since the configuration and separation of transmitter and receiver coils, in different EMI instruments, affect the apparent MS measurements, so measurements from different types of such instruments are not completely comparable.
Table 3.
Electromagnetic induction instrumentsa currently used for measuring soil MS, their specifications and applications in soil sciences.

Applications of Soil MS in Soil Studies
Magnetic granulometry
The classical factors of soil formation (Jenny 1941), that is climate, soil organisms, relief, parent material, and time, have also been used to describe the formation of magnetic minerals in soil (Blundell et al. 2009). Although it seems unusual, micro-organisms can enhance soil MS through fermentation processes (Le Borgne 1955; Mullins 1977, Dearing et al. 1996). Abiotic aging of ferrihydrite to hematite, through the intermediate mineral hydromaghemite, has been found to be a major pathway of magnetic enhancement (Barrón et al. 2003; Torrent et al. 2006; Liu et al. 2008). After oxygen (O), silicon (Si), and aluminum (Al), iron (Fe) is the fourth major abundant element in earth’s crust (Jordanova 2017). The abundance of atmospheric oxygen at the earth’s surface leads to formation of iron oxides as stable magnetic minerals in nature. The type of iron oxides found in natural deposits, such as rocks, sediments, soils, and aerosols, is heavily influenced by their origin and environmental conditions under which they were formed. So, based on the origin of the magnetic iron oxide minerals, their formation can be classified as lithogenic (or primary) and pedogenic (or secondary). Lithogenic iron oxides are inherited directly from a parent rock. Thus, the mineralogical compositions of lithogenic minerals directly reflect the mineralogy of their parent rock. On this basis, soils developed on sedimentary rocks usually contain far fewer magnetic minerals than those soils developed on volcanic or intrusive parent rocks (Jordanova 2017). It has been found, however, that magnetic grains can be formed during the weathering of igneous or sedimentary rocks (Singer and Fine 1989; Vali et al. 1989; Hounslow and Maher 1996). Lithogenic iron oxides are usually coarse grained (≥20 nm), which contain SD and MD magnetic minerals that preserve remnant magnetization. They are stable against demagnetizing factors, such as alternating field (AF) and thermal demagnetizations (Dunlop and Özdemir 1997), and due to their small surface to volume ratio, they are resistive to chemical weathering (Schaetzl and Anderson 2009). On the other hand, pedogenic iron oxides are formed during soil formation and development. They are much smaller in size and have low crystallinity compared with lithogenic iron oxides, which are the two most characteristic features of these pedogenic iron oxides (Cornell and Schwertmann 2003). Pedogenic magnetic minerals are ≤ 20 nm in diameter and, therefore, these minerals are called SP minerals. These SP minerals play an important role in soil magnetism. They cannot preserve remanent magnetization, but they possess high MS so that, in some cases, their contribution in soil MS is much higher than coarse-grain minerals (Dunlop and Özdemir 1997). On this basis, the measured MS in a single frequency (χ) cannot discriminate the portion of minerals with different grain sizes. The main method for separate lithogenic and pedogenic magnetic mineral contributions in soil MS is by measuring FD of soil MS, which is actually based on the fact that fine-grained magnetic minerals cannot follow high frequency magnetic field variations, and show less MS than when the low frequency magnetic field is used, that is, χhf < χlf (Dearing et al. 1996). The difference between these two susceptibilities qualitatively reflects minerals with less than 20 nm in diameter. This phenomenon has motivated soil scientists to use FD of MS to discriminate soil grain sizes, sometimes known as “magnetic granulometry” (Dearing et al. 1996; Evans and Heller 2003; de Jong et al. 2005; Egli 2009; Chadima et al. 2010; Hrouda et al. 2013; Kodama 2013; Kodama et al. 2014). Magnetic granulometry is generally based on laboratory measurements of soil MS at different frequencies, which can be done using either Bartington MS2B or AGICO KLY and MFK series instruments. Since FD of MS is a dimensionless quantity derived from dividing the difference of low and high frequency MS measurements to low frequency MS, the low and high frequency measurements can be done as either mass-specific or volume MS.
Using physical and mathematical models, Hrouda (2011) has shown that FD of MS can be utilized to determine different soil grain sizes, but he also introduced a method to compensate the discrepancy of using different frequencies to calculate the FD of MS. Recently, Grison et al. (2017) used this method to detect pedogenic magnetic minerals in volcanic soils developed on basalt utilizing two different magnetic-based instruments, Bartington MS2D and AGICO MFK1-FA.
Detection of pollutants
Anthropogenic processes such as atmospheric deposition of fly ash from industrial emissions and motor vehicle exhaust, mining activities, irrigation with polluted or sewage water, or liquid slags from industrial activities may be detected in soil using MS instrumentation. These pollutants partly include some of magnetic iron oxides, which enhance the MS in the polluted area. As such, MS measurements can be used as a proxy to trace environmental pollutants. On the other hand, most of anthropogenic pollutant factors are related to heavy metals such as Al, Cd, Cr, Cu, Fe, Ni, Pb, and Zn. The existence of these heavy metals, in the soil, commonly results in enhancement of soil MS that can then be measured in the field or in the laboratory using soil samples. In fact, anthropogenic pollution results in a strong magnetic signature of the soil such that magnetic techniques, which are nondestructive, fast, and cost-effective, can be used to monitor these kinds of pollutants.
Numerous studies have reported positive relationships between soil MS and the contents of heavy metals in soil around roads or industrial sites (e.g., Kapička et al. 1999; Matzka and Maher 1999; Jordanova et al. 2003; Hu et al. 2007; El Baghdadi et al. 2012; Brempong et al. 2016; Orosun et al. 2020). In-field mapping of soil MS has also been utilized to monitor soil pollution resulted from industrial emissions (Hay et al. 1997; Heller et al. 1998; Hoffmann et al. 1999; Boyko et al. 2004; Duan et al. 2009). While the magnetic particles are well preserved in the topsoil of forest soils, cultivation of arable lands results in mixing the topsoil with high MS with underlying subsoil with low MS (Declercq et al. 2019). More recently, the soil MS has been used to detect heavy metal pollution in coastal areas (Tholkappian et al. 2019; Devanesan et al. 2020). The agricultural soil pollution can also be associated with the spread of heavy metals due to mining activities (Sheoran and Sheoran 2006). Mining wastes, produced by leaching ore bodies, are highly enriched in heavy metals that may penetrate nearby farming lands, groundwater, or streams. Dlouhá et al. (2013) utilized soil MS to investigate alluvial soils in river valleys contaminated by potentially toxic elements derived from mining and metallurgical industry. Hu et al. (2014) studied the effect of mining wastewater discharge on the quality in four representative paddy soils. The enhancement of soil MS signals around a tungsten mine is also reported to be linked to the contamination by arsenic (Ouyang et al. 2020). Unfortunately, EMI do not see applicability in all instances because the instruments are not sensitive to thin layers, containing fly ash or heavy metals for instance. They also cannot precisely measure MS of individual soil constituents; however, if the amount and special distribution of contamination are substantial, EMI techniques can help to delineate contaminated areas more accurately.
Impact of organic matter
Organic matter also affects soil MS. As a key indicator of soil quality (Owliaei et al. 2006), soil organic matter provides the energy needed for the reduction of iron from a hydroxide solution (Mullins 1977; Schwertmann et al. 1989; Bedard-Haughn 2011), which leads to formation of magnetite or maghemite and enhances the MS of the soil. Maher (1988) reported a positive correlation between MS and soil organic carbon content. Rijal et al. (2010 and 2012) observed enhancement of MS in hydrocarbon-contaminated soils and found a relationship between the amount of total nonpolar hydrocarbon (TNPH) and enhancement of soil MS. By studying the role of a kind of bacteria in the reduction of iron (hydr)oxides, Atekwana et al. (2014) observed correlation between zones with high concentrations of hydrocarbons and enhancement of MS and suggested the coupling of iron reduction with organic carbon oxidation in such zones. Zhang et al. (2013) observed an enhancement of low and high frequency MS in topsoil in farmland soils irrigated by sewage water. By comparing soil MS values of two adjacent agricultural sites, one irrigated by solely groundwater and other by sewage water, Yang et al. (2015) showed enhancements of MS and organic matter in the soils irrigated by sewage water and concluded that soil MS is significantly correlated with organic matter content. Another possibility for enhancing MS in soils irrigated by sewage water may be related to metals applied with the sewage water, which has not been addressed by Yang et al. (2015).
Almost all studies regarding the effect of organic matter on soil MS have been carried out by utilizing laboratory-based instruments, which use magnetic field for measuring soil MS. EMI instruments have not yet been used for studying the effect of organic matter on soil MS.
Delineation of drainage class
It has been suggested that the extremely reducing conditions, present in hydric soils, significantly enhance dissolution of soil ferrimagnetic minerals such as magnetite and maghemite (Maher 1986; Williams 1992). Since the MS of soils is mainly controlled by magnetite and maghemite concentrations (Mullins 1977; Maher 1986), MS values are typically very low in hydric, that is, poorly drained or gleyed, soils (Le Borgne 1955; Vadyunina and Babanin 1972; Yu et al. 1986). This phenomenon has widely been utilized by soil scientists to characterize soil drainage conditions (e.g., de Jong 2002; de Jong et al. 2005; Grimley et al. 2004 and 2008; Asgari et al. 2018; Gholamzadeh et al. 2019), which all have utilized the instruments that use magnetic fields to measure soil MS (Bartington MS2B and MS2D). Although measuring and (or) mapping of soil MS by utilizing EMI instruments can easily reveal drainage conditions of the soil, these kinds of instruments have rarely been used for discriminating soil drainage conditions (Shirzaditabar et al. 2021; Shirzaditabar and Heck 2021).
Paleo-environmental studies
The effect of climate, including temperature and rainfall, on soil MS has received a lot of attention because it is a proxy and a key factor in reconstructing past climate (e.g., Begét et al. 1990; Maher and Thompson 1991, 1992 and 1995; Geiss et al. 2008). By recording the MS of loess sequences in China, Maher and Thompson (1995) reported increasing MS with mean annual rainfall peaking at around 1500 mm, which was followed by decreasing MS to around 3000 mm. Singer et al. (1996) observed a similar behavior in increasing soil MS by annual precipitation of around 1000 mm in Kohala peninsula island, in Hawaii, with an associated decline of MS to around 2500 mm. They also reported enhancement of soil MS with increasing mean annual temperature, to a maximum of 18 °C, followed by a decrease. These researchers also reported enhancement of soil MS with time. More recently, by studying MS of loess soils in China and Europe, Jordanova and Jordanova (2021) also observed the influence of mean annual precipitation and temperature on MS of loess soils. All of these studies were done using laboratory-based instruments (Bartington MS2B and AGICO KLY and MFK series), because they require precise measurements of soil MS, which cannot be achieved by utilizing EMI instruments directly in the field.
Archaeology
Unnatural impacts of human activities on soil, as well as anthropogenic processes, have made the archaeological sites very complex environments. Therefore, combined multidisciplinary approaches are required to generate thorough reconstruction of these sites (Walkington 2010; Canti and Huisman 2015). Although chemical methods have been shown to be useful in archaeological studies (Pastor et al. 2016), geophysical methods are still the most important methods to detect archaeological sites (Gaffney 2008). These geophysical methods include EC measurements, to map conductivity (σ) changes in the soil, ground penetrating radar (GPR), to reveal detailed layering of the subsurface (mostly based on soil electrical permittivity, ε) and magnetometry, to disclose and map soil MS variations. While EC measurements are useful in archaeological investigations, moisture content of the soil can quickly affect EC values. On the other hand, it has been shown that MS is less affected by soil moisture (Maier et al. 2006), which suggests the soil MS as a reliable feature in archaeological studies. Indeed, many surveys in archaeological studies utilize soil MS to discover archaeological features (De Smedt et al. 2014; Fassbinder 2015). The early work of Le Borgne (1955) suggested that high temperatures produced by fires can cause enhancement in soil MS. In such circumstances, non-ferrimagnetic oxides, produced by weathering of the soil, are reduced to magnetite and maghemite. It has been also suggested that in poorly drained soils, dehydration of lepidocrocite, at temperatures between 275 °C and 400 °C, may cause formation of maghemite (Thompson and Oldfield 1986; Maher 1986). Fires may be produced either naturally or intentionally by humans. Inspired by this suggestion, the first attempts were done to explore archaeological sites (Aitken et al. 1958; Colani and Aitken 1966). Tite and Mullins (1971) confirmed that heating by fire has a strong impact on soil MS enhancement. Although iron tools, like swords, provide high magnetic signature, and can be easily detected using magnetic measurements, some human activities, like pottery production or metal workings. They also leave certain magnetic signatures behind in the landscape because of alteration of magnetic minerals in the soil or even distribution of waste pottery products. Evidence also shows that repetitive use of local fires, or intentional burning of a landscape, has left a distinctive magnetic trace on the soil (Linford 2005). Therefore, for the above-mentioned reasons, the most surveys in archaeological investigations nowadays use soil magnetic properties, especially MS, to reveal signatures from ancient civilizations (e.g., Bevan 2000; Bevan and Dalan 2003; Jordanova et al. 2001; Dalan 2006, 2008 and 2009; Sipmson et al. 2009; Bonsall et al. 2013; Benech et al. 2016; Heil and Schmidhalter 2017; Dalan et al. 2017; Gibson 2017; Lowe et al. 2020). Although using instruments that employ a magnetic field may be useful for archaeological investigations, EMI instruments are much more popular in such investigations because they are nonintrusive and have the ability to measure and map soil MS in a very fast and efficient way.
Soil erosion and degradation
As a natural phenomenon, soil erosion affects all types of landforms and the high level of land degradation in some geographical regions is a serious threat for sustainable agricultural activities (Bouhlassa and Bouhsane 2019). Soil erosion, in agriculture, is related to the removal and redistribution of topsoil, and sometimes subsoil, by natural forces, such as water and wind, or anthropogenically by tillage (Jordanova et al. 2014; Jakšík et al. 2016).
As a physical property of soil, MS has proved to be an effective tool in soil erosion and degradation studies. The distribution of ferrimagnetic minerals in soils has shown to be an indicator of soil degradation (Sadiki et al. 2009; Łukasik et al. 2015). Measuring soil MS for evaluation of soil redistribution along the slope in semiarid regions was validated by Karchegani et al. (2011). Jordanova et al. (2011), in a detailed field and laboratory study showed that the magnetic methods are efficient in soil erosion assessment; even in a highly magnetic parent material, where they observed a significant amount of soil loss and related it to tillage practices. By measuring soil MS on a sloped site of agricultural land in the loess terrain, Kapička et al. (2013) observed the highest values of MS on flat up-slope areas, where the original top horizon remained, and the lowest values of MS on the steep side slopes, and attributed this to erosion on steep slopes. Results of studying MS of Chernozem soils, from an agricultural land, showed that tillage was the main cause of soil erosion on upper slopes (Jordanova et al. 2014). By analyzing the variation of soil MS in different soil profiles on forested, cultivated, and pasture lands, having the same lithology and climatic conditions but different land uses and slope gradients, Bouhlassa and Bouhsane (2019) found that MS decreases as χforested > χpasture > χcultivated, and by observing positive correlation between χlf and χfd, they concluded that the loss of fine magnetic particles is associated to decrease in χlf. Highly correlated with the humus content and erosion index, MS is an indicator for identifying soil erosion rates and depths. Soil MS investigations, in addition to providing the benefit of rapid and low-cost measurements, allow for the formation of a dense grid of sampling and the identification of slope erosion structure (Menshov et al. 2018).
Since erosion affects topsoil more than subsoil, and the MS of soil must precisely be measured to be used in erosion studies, all instrumentations used for this case are laboratory instruments, mostly Bartington MS2B and AGICO KLY series, which are instruments that use magnetic fields for measuring MS. While EMI instruments have not yet been used for investigating soil erosion and degradation, the ability of measuring MS in different frequencies has made the laboratory instruments the only, and the most appropriate, option to be used in such studies.
Summary and Future Directions
The MS measurements in soils depend on grain sizes and the concentration of magnetic minerals. Measurements are also highly dependent on the instrument used to measure it. Almost all MS instruments that use magnetic fields to measure MS, utilize some kind of coil to pick up the returning magnetic signal from the soil and display or record the soil MS. The geometry of these coils, and electronics behind different MS instruments, affects the MS measurements. The coils of the MS instruments can excite certain amounts of soil to contribute to MS measurements. For example, a larger coil diameter is capable of influencing a larger volume of soil for the MS recordings (or more depth of investigation). The recorded MS in this case is a mean value of MS of the constituents within the volume of excited soil. Therefore, since the magnetic minerals are randomly distributed in the soil context, handheld instruments with larger coils have the advantage that their results are less affected by local accumulation of magnetic minerals and yield the mean MS of the soil.
The frequency in which the instrument operates also affects the measurement of soil MS. Due to relaxation of fine-grained magnetic minerals in the soil, MS recordings at higher frequencies are less than MS recordings at lower frequencies. This behavior is used to discriminate between lithogenic and pedogenic soil minerals. Furthermore, the FD of MS is a key factor to delineate soil drainage classes because intermittent water saturation and redox processes lead to formation of fine-grained magnetic minerals, which respond better to low frequency fields. The FD of MS can also be used in soil pollution studies where, for example, fine-grained fly ash from industry or motor vehicle exhausts are deposited on the soil surface. It should be noted that when the data are acquired using instruments with different operating frequencies over the same area, where the increases or decreases in MS over a profile or in a map are comparable, the individual values of soil MS, as well as percentage FD, are not the same over the same position. This means that the comparison between different types of instruments is meaningful only when they have the same design and geometry, and operate at the same frequency.
The ability to measure simultaneously out-of-phase components, as well as in-phase components, is the advantage of using some kinds of MS instruments, which use magnetic fields (e.g., AGICO KLY and MFK models). Using the in-phase and out-of-phase components in just one frequency is enough to calculate FD of MS of a sample. Some types of instruments have the capability to measure MS in a range of temperatures (AGICO KLY and MFK, Bartington MS2W), which can be used to reveal mineralogy of soil samples.
Contrary to instruments that use magnetic fields, the separation between transmitter and receiver in EMI instruments determines the amount of soil that contributes to the MS measurement. Therefore, MS values recorded by EMI instruments are essentially apparent values because they are mean values of MS based on a hemisphere of soil below the instrument, for example, 1 m diameter. Moreover, the sensitivity of EMI instruments to MS of the soil is different for different configurations. While the sign of apparent MS remains the same in VCP configuration mode, it changes from positive to negative, or vice versa, in HCP and perpendicular (PERP) configurations, because these two latter modes are sensitive to the distance from the magnetic layer (or magnetic horizon). Thus, mapping with EMI instruments may reveal negative apparent MS locations, especially when using HCP and PERP configurations, which are not necessarily the indication of diamagnetic zones. Although inversion methods can be utilized to recover MS of soil layers, or horizons, from apparent MS measurements taken from EMI instruments, these methods have limited accuracy when they are given limited input data. The best results are achieved if data from various configurations, and possibly different heights above surface, are used as input data.
The great advantage of EMI instruments against those that use magnetic fields is their high speed of recording. This is because the EMI instruments does not need to contact the earth’s surface and the signal can be sensed by receiver coil of instrument even it is suspended above the soil surface. So, if the objective of MS measurements is to map the MS variations in a certain area, for instance in archaeological or agricultural studies, the EMI instrument is preferred because the measurements can be done quickly, and the resulting MS map reveals high and low MS zones. On the other hand, if grain size or mineralogy of the soil is desired, instruments that use just magnetic fields are preferred, especially laboratory instruments which have the ability to measure FD or out-of-phase components of MS.
This review showed that both kinds of MS instruments have been widely used in soil science; however, they cannot be simply interchanged with each other because they use different methods for measuring soil MS. Limitations of designing handheld EMI instruments to operate at low frequencies for laboratory measurements, or vice versa, have limited comparison. Moreover, the lack of available three-dimensional, and even one-dimensional, inversion software for calculating MS of different segments of soil has deterred soil scientists from utilizing EMI instruments widely in their research. Therefore, designing new instruments to be used in the field and (or) laboratory that are able to operate for a range of frequencies, and comparison of the results of applying both kinds of instruments in the same area, might be a future work to see which kind of instrument can find better status among soil scientists.
Acknowledgements
The authors would like to thank the editor, Dr. James Warren, and two anonymous reviewers for their helpful remarks as well as thorough and detailed editing recommendations for improving the article.