Proximal Soil Sensing: An Effective Approach for Soil Measurements in Space and Time

C H A P T E R F I V E

Proximal Soil Sensing: An Effective Approach for Soil Measurements in Space and Time R.A. Viscarra Rossel,
V.I. Adamchuk,† K.A. Sudduth,‡ N.J. McKenzie,
and C. Lobsey
Contents 1. Introduction 1.1. Proximal soil sensing 1.2. The sampling dilema - Where to measure using proximal soil sensors? 2. Proximal Soil Sensing Techniques 2.1. γ-rays 2.2. X-rays 2.3. Ultraviolet, visible, and infrared reflectance spectroscopy 2.4. Laser-induced breakdown spectroscopy 2.5. Microwaves 2.6. Radio waves 2.7. Magnetic, gravimetric, and seismic sensors 2.8. Contact electrodes 2.9. Mechanical sensors 2.10. Telemetry—Wireless sensing 2.11. Geographic positioning and elevation 2.12. Multisensor systems 2.13. Core scanning and down-borehole technologies 3. Proximal Sensors Used to Measure Soil Properties 3.1. Soil water and related properties 3.2. Nutrients and elements 3.3. Cation exchange capacity 3.4. Carbon 3.5. pH 3.6. Clay, silt, and sand


† ‡

245 246 250 251 251 253 254 256 256 257 261 262 264 266 267 268 269 270 270 271 272 272 272 273

CSIRO Land and Water, Bruce E. Butler Laboratory, Canberra, ACT, Australia Bioresource Engineering Department, McGill University, Ste-Anne-de-Bellevue, QC, Canada USDA Agricultural Research Service, Cropping Systems and Water Quality Research, Columbia, MO

Advances in Agronomy, Volume 113 ISSN 0065-2113, DOI: 10.1016/B978-0-12-386473-4.00005-1

© 2011 Elsevier Inc. All rights reserved.

243

244

R.A. Viscarra Rossel et al.

3.7. Soil mineralogy 3.8. Soil strength, bulk density, and related properties 4. Summary 5. General Discussion and Future Aspects Acknowledgments References

273 273 274 274 281 281

Abstract This chapter reviews proximal soil sensing (PSS). Our intent is for it to be a source of up-to-date information on PSS, the technologies that are currently available and their use for measuring soil properties. We first define PSS and discuss the sampling dilemma. Using the range of frequencies in the electromagnetic spectrum as a framework, we describe a large range of technologies that can be used for PSS, including electrochemical and mechanical sensors, telemetry, geographic positioning and elevation, multisensor platforms, and core measuring and down-borehole sensors. Because soil properties can be measured with different proximal soil sensors, we provide examples of the alternative techniques that are available for measuring soil properties. We also indicate the developmental stage of technologies for PSS and the current approximate cost of commercial sensors. Our discussion focuses on the development of PSS over the past 30 years and on its current state. Finally, we provide a short list of general considerations for future work and suggest that we need research and development to: (i) improve soil sampling designs for PSS, (ii) define the most suitable technique or combination of techniques for measuring key soil properties, (iii) better understand the interactions between soil and sensor signals, (iv) derive theoretical sensor calibrations, (v) understand the basis for local versus global sensor calibrations, (vi) improve signal processing, analysis, and reconstruction techniques, (vii) derive and improve methods for sensor data fusion, and (viii) explore the many and varied soil, agricultural, and environmental applications where proximal soil sensors could be used. PSS provides soil scientists with an effective approach to learn more about soils. Proximal soil sensors allow rapid and inexpensive collection of precise, quantitative, high-resolution data, which can be used to better understand soil spatial and temporal variability. We hope that this review raises awareness about PSS to further its research and development and to encourage the use of proximal soil sensors in different applications. PSS can help provide sustainable solutions to the global issues that we face: food, water, and energy security and climate change. Keywords: Proximal soil sensing; geophysics; soil spectroscopy; gamma radiometrics; electromagnetic induction; electrical conductivity; electrochemical sensing; mechanical sensors; multisensor platform; mobile soil sensors; sensor data fusion; soil measurement; soil analysis; soil sampling

Proximal Soil Sensing for Measurements in Space and Time

245

1. Introduction Our scientific understanding of the unique qualities and functions of soils has been gained through long and arduous soil surveys complemented by careful chemical, physical, mineralogical, and biological laboratory analysis. These conventional methods continue to serve us well, but they can be expensive, complex, time consuming, and some are only qualitative. The growing demand for good quality, inexpensive soil information underlines these shortcomings. We need better soil information to solve pressing problems such as how to monitor the effects of climate change on soil, how to populate models of key processes, how to use precision agriculture for improving the sustainability and efficiency of food production, and how to assess and remediate contaminated land. These applications have prompted the development of sensors to measure soil properties and complement, or replace, the more conventional laboratory techniques used for their analyses. Sensors provide quantitative results and can be more time- and costeffective than conventional laboratory analyses. They are becoming smaller, faster, more accurate, more energy efficient, wireless, and more intelligent. Many such devices can be used for proximal soil sensing (PSS), for example, ion-sensitive field effect transistors (ISFETs) to measure soil pH and soil nutrients, or portable visiblenear-infrared (visNIR) spectrometers to measure soil properties like organic carbon content and mineral composition. Worldwide, a vast amount of research is being conducted to develop proximal soil sensors and techniques for their use in various applications. The research includes investigations on the use of frequencies across the electromagnetic (EM) spectrum (Fig. 1). For example, some are using γ-radiometrics to gain an understanding of soil patial variability and the underlying parent material; X-ray fluorescence (XRF) and laser-induced breakdown spectroscopy (LIBS) to measure soil

Figure 1 The electromagnetic (EM) spectrum. For color version of this figure, the reader is referred to the web version of this book.

246

R.A. Viscarra Rossel et al.

elemental analysis; visNIR and mid-infrared (mid-IR) energies to measure soil carbon and mineralogy; ground-penetrating radar (GPR) to measure soil water content; and electromagnetic induction (EMI) to measure soil electrical conductivity. Research on the use of ion-selective electrodes (ISEs) and ISFETs to measure soil ion activities and mechanical systems to measure soil strength is also progressing. Many of these sensors are currently in a developmental phase and are used primarily in research, while others are available commercially. The most common techniques, which account for a large portion of the PSS literature, relate to the use of EMI and soil visNIR spectroscopy. With rapid ongoing technological developments, we expect many more explorations to find new soil sensing solutions to help improve our understanding of soil processes. The aim of this review article is to discuss the current state of PSS and the potential benefits and opportunities that it presents for soil science. In it, we define PSS; provide a comprehensive review of current techniques, including multisensor platforms; report their development status and estimates of their cost; discuss the soil properties that we are interested in measuring using proximal soil sensors; finally, we provide a synthesis and discuss the future of PSS.

1.1. Proximal soil sensing We define PSS as the use of field-based sensors to obtain signals from the soil when the sensor’s detector is in contact with or close to (within 2 m) the soil (Viscarra Rossel and McBratney, 1998; Viscarra Rossel et al., 2010a). The sensors provide soil information because the signals correspond to physical measures, which can be related to soils and their properties. Our definition of PSS therefore precludes remote sensing and also laboratory measurements of soil properties with benchtop instruments. We do however, acknowledge that the development of many proximal soil sensors starts in the laboratory, and that some (e.g. vis-NIR sensors) use calibrations derived from laboratory measurements. Proximal soil sensors may be described by the manner in which they measure (invasive [in situ or ex situ] or noninvasive) the source of their energy (active or passive), how they operate (stationary or mobile), and the inference used in the measurement of the target soil property (direct or indirect) (Fig. 2). If there is sensor-to-soil contact during measurement, then the proximal soil sensor is invasive. It is noninvasive if there is no contact between the sensor and the soil. Measurements with invasive proximal soil sensors may be made in situ (i.e., the measurements are made within the soil) or ex situ (i.e., the measurements are made on excavated soil, e.g., measurements on soil cores). A proximal soil sensor is active if for the measurements it produces its own energy from an artificial source. It is

247

Proximal Soil Sensing for Measurements in Space and Time

Measurement

Energy

Operation

Inference

Non-invasive

Passive

Mobile

Indirect

PROXIMAL SOIL SENSING Invasive In situ

Active

Stationary

Direct

Ex situ

Figure 2 Proximal soil sensors may be described by their measurement being invasive and in situ or ex situ, or noninvasive, their energy source being active or passive, their operation being stationary or mobile, and their method for inferring soil properties being direct or indirect.

passive if it uses naturally occurring radiation from the sun or earth. A stationary sensor acquires measurements in a fixed manner. Mobile proximal soil sensors are those that measure soil properties while moving or “onthe-go.” Usually, mobile proximal soil sensors are used for fine-resolution soil mapping. Adamchuk et al. (2004) reviewed “on-the-go” proximal soil sensors for precision agriculture. If the measurement of the target soil property is based on a physical process, then the proximal soil sensor is said to be direct. However, when the measurement is of a proxy and inference is with a pedotransfer function, then the proximal soil sensor is indirect. Table 1 describes the different techniques for PSS and the ways in which they function. For example, from Table 1 and Fig. 2, measurements with a tine-mounted visNIR proximal soil sensor are invasive and in situ, the sensor uses an active source of energy, it has mobile operation and depending on the soil property inference might be either direct (e.g., clay mineralogy) or indirect (e.g., cation exchange capacity or CEC). Measurements of an extracted soil core with a portable visNIR proximal soil sensor might be, depending on the measurement setup, invasive and ex situ or noninvasive, the sensor might use an active (halogen bulb) or passive (the sun) source of energy, it has stationary operation and inference might be either direct or indirect. Measurements with a γ-radiometer are noninvasive, use a passive source (naturally occurring radioisotopes of Cs, K, U, Th), operation is often mobile, although stationary measurements are also possible and inference is mostly indirect. The rationale for the use of proximal soil sensors is that, although their results may not be as accurate per individual measurement as for conventional laboratory analysis (i.e., their results may be more biased and/or imprecise), proximal soil sensors facilitate the collection of larger amounts of (spatial) data using cheaper, simpler, and less laborious techniques which, as an ensemble, are greatly informative. Moreover, the measurements are made at field conditions, they are

Table 1 Proximal soil sensors are described by their measurement being invasive (In) and in situ or ex situ, or noninvasive (N), their energy source being active (A) or passive (P), their operation being stationary (S) or mobile (M), and their method for inferring soil properties being direct (D) or indirect (I) Measurement EM range wavelength (m) 212

γ-rays (10

)

X-rays (10210) UVvisIR (1028 to 1024)

Microwaves (1022) Radio waves (101 to 106)

Energy

Operation

Inference

Technique

Invasive/noninvasive Active/passive Stationary/mobile Direct/indirect

INS TNM Active γ Passive γ XRF XRD UV Vis NIR MIR LIBS Microwave TDR

N In N N N N N In In In In N In

(in situ)

(in/ex (in/ex (in/ex (in/ex

situ)/N situ)/N situ) situ)

(in situ)

A A A P A A A A/P A/P A A A A

S/M S S S/M S S S S/M S/M S S S S

D D D D/I D D D/I D/I D/I D/I D I I

Electrical resistivity Electrochemical Mechanical

FDR/capacitance GPR NMR EMI ER Gypsum/granular ISE/ISFET Implement draft Mechanical resistance Fluid permeability Acoustic

In N N N In In In In In In In

(in situ)

(in situ) (in situ) (in/ex situ) (in situ) (in situ) (in situ) (in situ)

A A A A A P P P P A A

S/M S/M S M M S S/M M S/M S/M S/M

I D/I D I I D D D D I I

Note: INS, inelastic neutron scattering; TNM, thermalized neutron methods; XRF, X-ray fluorescence; XRD, X-ray diffractometry; UV, ultraviolet; vis, visible; NIR, near infrared; MIR, mid infrared; LIBS, laser-induced breakdown spectroscopy; TDR, time-domain reflectometry; FDR, frequency-domain reflectometry; GPR, ground-penetrating radar; NMR, nuclear magnetic resonance; EMI, electromagnetic induction; ER, electrical resistivity; ISE, ion-selective electrode; ISFET, ion-sensitive field effect transistor.

250

R.A. Viscarra Rossel et al.

acquired from the surface or within the soil profile and information is produced in a timely manner; that is, almost instantly. Therefore, PSS offers advantages to soil measurement that cannot be achieved by remote sensing or destructive sampling and laboratory analyses.

1.2. The sampling dilema - Where to measure using proximal soil sensors? The decision of where to measure using proximal soil sensors will depend on whether the proximal sensor is described as direct/indirect and stationary/mobile. If the sensor measures the target soil property directly and at fixed locations, the sampling problem will be the same as conventional spatial soil sampling because it requires optimization of the geographical coverage of the measurements. If the sensor measurements are direct and are made with a mobile, “on-the-go” system, then the sampling problem might relate to the frequency (or resolution) of the measurements so as to optimize the amount of information collected. If the measurements are made indirectly, a calibration will need to be developed (using the sensor’s measurements and soil samples collected and analyzed in the laboratory) to predict the target property from the sensor measurements. In this case, a calibration sampling design that optimizes coverage of property (or feature) space will be required. Ideally, the sampling should also cover geographic space so that landscape position and other location-induced phenomena are included in the calibrations. Most of the published literature considers either geographical space sampling or property space sampling designs, not both (de Gruijter, 2002; de Gruijter et al., 2006; van Groenigen and Stein, 1998). In agricultural land, it will also be important to consider field boundaries and other transition zones to prevent situations in which the samples do not represent the same soil as the nearest measurements obtained using the proximal soil sensors. Designs for calibration sampling have been proposed by Lesch (2005) and Minasny and McBratney (2006). Christy (2008) proposes a sampling design for the calibration of visNIR spectra collected using an on-the-go system. The approach covers the range of variability in property space and considers the location of field boundaries and transition zones but only indirectly considers geographical coverage. De Gruijter et al. (2010) describe geographical and property space sampling with proximal soil sensors for fine-resolution soil mapping, and Adamchuk et al. (2008, 2011) compare designs for mobile PSS that consider geographic and property space and field boundaries and other transition zones. Further research is needed to develop optimal sampling designs for the different types of proximal soil sensors.

Proximal Soil Sensing for Measurements in Space and Time

251

2. Proximal Soil Sensing Techniques Presently, proximal soil sensors can measure the soil’s ability to accumulate and conduct electrical charge, to absorb, reflect, and/or emit EM energy, to release ions, and to resist mechanical distortion. Using energies in the EM spectrum as the framework, below we describe currently available technologies for PSS.

2.1. γ-rays In the EM spectrum, γ-rays occur at quantum energies between 1 MeV and 124 keV and have frequencies of 10201024 Hz and short wavelengths of less than 10212 m (Fig. 1). They contain a very large amount of energy and are the most penetrating radiation from natural or man-made sources. 2.1.1. γ-ray spectrometers A γ-ray spectrometer is an instrument that measures the distribution of the intensity of γ radiation versus the energy of each photon. Most soil γ-ray spectrometers use scintillators with either thallium-doped sodium iodide or thallium-doped cesium iodide crystals, although other materials are also available (International Atomic Energy Agency (IAEA), 2003). When these are hit by the ionizing radiation, they fluoresce and a photomultiplier tube is used to measure the light from the crystal. The photomultiplier tube is attached to an electronic amplifier, which quantifies the signal. Active γ-ray sensors use a radioactive source (e.g., 137Cs) to emit photons of energy that can then be detected using a γ-ray spectrometer. The theory of operation for measuring soil properties such as water, or bulk density using active γ-rays, indicates that when these are emitted and pass through the soil, photons are transmitted following the BeerLambert law (Wang et al., 1975). The attenuation of the signal is determined by the thickness of the material, its density, and its mass attenuation coefficient. Passive γ-ray sensors (Fig. 3A) measure the energy of photons emitted from naturally occurring radioactive isotopes of the element from which they originate. While many naturally occurring elements have radioactive isotopes, only potassium (40K) and the decay series of uranium (238U and 235U and their daughters) and thorium (232Th and its daughters) have long half-lives, are abundant in the environment, and produce γ-rays of sufficient energy and intensity to be measured. The result is a γ-ray energy spectrum (Fig. 3B). Gamma-ray sensors have been more commonly used from remote sensing platforms (Minty, 1997); however, the techniques are also used to measure soil properties proximally (Viscarra Rossel et al., 2007; Wong et al., 2009). The advantage of proximal γ-ray sensing over

252

(A)

R.A. Viscarra Rossel et al.

(B) 70

Counts (s–1)

60 50 40 30 K U 1.46 1.76

20

Th 2.61

10 0 0

0.5

1

1.5 2 Energy (MeV)

2.5

3

Figure 3 (A) A proximal passive γ radiometric sensor mounted on a multisensor platform and (B) a γ-ray spectrum showing the energies of the potassium (K), uranium (U), and thorium (Th) bands.

remote sensing is that it is more directly related to soil materials and less prone to the effects of surface cover and geometry. Soil mineralogy, particle size, and the effects of attenuating materials such as water and density control the γ-ray signal. Soil parent material, the intensity of weathering, and the geometry of near-surface soil layers are therefore also important. 2.1.2. Neutron scattering methods Neutron scattering may be categorized into elastic and inelastic techniques. The most common soil sensor that uses elastic neutron scattering is the neutron probe for measuring soil water (Gardner and Kirkham, 1952). Neutrons emitted from a radioactive source into the soil are slowed by elastic collision (i.e., the emitted neutrons have the same energy as those that are injected) with the nuclei of atoms of low atomic weight, such as hydrogen. Hydrogen can slow fast neutrons more effectively than any other element present in soil and the density of the resulting cloud of slow neutrons is a function of the amount of water in the soil. Schrader and Stinner (1961) proposed inelastic neutron scattering (INS) as a technique for elemental analysis of surfaces. Wielopolski et al. (2008) proposed it for the measurement of soil carbon and other elements. INS relies on the detection of γ-rays that are emitted following the capture and reemission of fast neutrons as the sample is bombarded with neutrons from a pulsed neutron generator. The emitted γ-rays are characteristic of the excited nuclide and the γ-rays intensity is directly related to the elemental content of the sample. The detectors used are the same as those used in γ-ray spectroscopy (see above). Zreda et al. (2008) described a passive technique where the neutron source is derived from the naturally occurring neutrons generated through

Proximal Soil Sensing for Measurements in Space and Time

253

cosmic-ray interaction with the atmosphere. The sensor’s detector counts neutrons that are back scattered out of the soil and these can be correlated to soil water content. Since the method relies on a natural source, low counts are obtained and significant integration is required. As such, it is suited to stationary measurements (e.g., for monitoring changes in soil water).

2.2. X-rays X-rays have quantum energies between 124 keV and 124 eV and occur in the EM spectrum at a frequency range of 10171020 Hz and wavelengths of about 102121029 m (Fig. 1). They contain a large amount of energy and have been classified as either “hard” (shorter wavelengths) or “soft” (longer wavelengths). 2.2.1. X-ray fluorescence XRF is used to measure elements in soil samples. The technique relies on the fluorescence at specific energies of atoms that are excited when irradiated with X-rays. Detection of the specific fluorescent photons enables the qualitative and quantitative analysis of the elements in a sample. XRF spectroscopy has been used in the laboratory for many years. Kalnicky and Singhvi (2001) provide a comprehensive overview of XRF for environmental analysis. Portable, handheld XRF technology has gained acceptance as an analytical approach in the environmental community, particularly for rapid measurement of metal contaminants. 2.2.2. X-ray diffractometry X-ray diffraction (XRD) is a nondestructive technique used to acquire detailed information on the mineral composition of the soil. Moore and Reynolds (1997) provide details of the laboratory technique. The use of portable XRD systems and research prototypes are starting to appear in the literature because many of the strict hardware requirements, such as reproducible alignment of the X-ray detectors and long acquisition times, are being overcome (Gianoncelli et al., 2008). Sarrazin et al. (2005) describe the development and testing of a portable combined XRD/ XRF system that can be used in the field. The system was constructed for remote planetary exploration, but has also been used as a field tool for geological research. Gianoncelli et al. (2008) also describe the development of a portable XRD/XRF system for simultaneous elemental analysis and phase identification of inorganic materials.

254

R.A. Viscarra Rossel et al.

2.3. Ultraviolet, visible, and infrared reflectance spectroscopy Diffuse reflectance spectroscopy has been used in soil science research since the 1950s and 1960s (Bowers and Hanks, 1965; Brooks, 1952). However, it is only in around the past 20 years, most likely coinciding with the establishment of chemometrics and multivariate statistical techniques in analytical chemistry, that its usefulness and importance in soil science has been realized. Interest in using reflectance spectroscopy to measure soil properties is widespread because the techniques are rapid, relatively inexpensive, require minimal sample preparation, are nondestructive, require no hazardous chemicals, and several soil properties can be measured from a single scan (Viscarra Rossel et al., 2006a). Ultraviolet radiation possesses quantum energies of 1242.1 eV, has a frequency range of about 10151017 Hz, and wavelengths of around 10291027 m (Fig. 1). There is little in the literature on the use of ultraviolet radiation for PSS, and often the technique is combined with visible or infrared spectroscopy (Islam et al., 2003). Visible light has energies between 2.1 and 1.65 eV, frequency in the range 4 3 10147 3 1014 Hz, and wavelengths of 7 3 10274 3 1027 m (Fig. 1). Absorptions of ultraviolet and visible radiation occur under high energies due to the excitation of outer electrons. Absorption of energy by an atom or molecule involves the promotion of electrons from their ground state to an excited state. Absorptions in organic molecules are restricted to certain functional groups (chromophores) that contain valance electrons of low excitation energy. Many inorganic species, such as iron oxides in soil, show charge transfer absorptions (also called charge transfer complexes) (Schwertmann and Taylor, 1989). For a complex to demonstrate charge transfer behavior, one of its components must be able to donate electrons and another must be able to accept them. Thus, absorptions involve transfer of an electron from the donor to an orbital associated with the acceptor. A soil spectrum in the visible range is shown in Fig. 4A. The near-infrared (NIR) portion of the EM spectrum has a frequency range of 1.2 3 10144 3 1014 Hz, with wavelengths of 7 3 10272.5 3 1026 m, while the mid-IR has a frequency range of 3 3 10121.2 3 1014 Hz and wavelengths of 2.5 3 10262.5 3 1025 m (Fig. 1). Energies in the NIR range between 1.65 eV and 124 meV, while those in the mid-IR range from 124 to 12.4 meV. The farIR has an energy range of 12.41.24 meV, a frequency range of 3 3 10123 3 1011 Hz, and wavelengths between 2.5 3 1025 and 5 3 1025 m (Fig. 1); however, there are no publications on the use of farIR for PSS. Conversely, there is a vast amount of literature on the use of visNIR and mid-IR for soil analysis (Stenberg et al., 2010) and, increasingly, on the use of these techniques for PSS (Ben-Dor et al., 2008; Christy, 2008; Reeves et al., 2010; Viscarra

255

Proximal Soil Sensing for Measurements in Space and Time

(A)

(B)

(C)

2.5 Visible

Near infrared

Mid infrared

2

Log 1/R

Iron oxides

1.5

Water

Iron oxides

Clay minerals

Carbon

Quartz

Colour

1

Clay minerals Carbon Carbonate

0.5

Iron oxides

Organic matter Carbonates Water

0 400

500

600

700

1200

1700

2200

2500

7500

12,500

17,500

Wavelength (nm)

Figure 4 Typical soil spectrum in the (A) visible (vis), (B) near-infrared (NIR), and (C) mid-infrared (mid-IR) portions of the EM spectrum.

Rossel et al., 2009; Waiser et al., 2007). Infrared radiation does not have enough energy to induce electronic transitions as with ultraviolet and visible light. Their absorptions are restricted to compounds with smaller energy differences in the possible vibrational states. For a molecule to absorb infrared energy, the vibrations within a molecule must cause a net charge in the dipole moment of the molecule. The alternating electrical field of the radiation interacts with fluctuations in the dipole moment of the molecule. If the frequency of the radiation matches the vibrational frequency of the molecule, then radiation will be absorbed, causing a change in the amplitude of the molecular vibration. The positions of the molecules are not fixed and are subject to different stretching or bending vibrations. The mid-IR contains more information on soil mineral and organic composition than the visNIR, and its multivariate calibrations are generally more robust (Viscarra Rossel et al., 2006a). The reason is that the fundamental molecular vibrations of soil components occur in the mid-IR, while only their overtones and combinations are detected in the NIR (Stenberg et al., 2010). Hence, soil NIR spectra display fewer and much broader absorption features compared to mid-IR spectra (Fig. 4B and C, respectively). The adaptation of visNIR spectrometers for PSS has been ongoing for the past two decades, with the first field prototype mobile systems developed by Shonk et al. (1991) and Sudduth and Hummel (1993). Since then, other prototype mobile systems have been developed by Shibusawa et al. (2001), Mouazen et al. (2005), Stenberg et al. (2007), and Christy (2008), who described a commercially available mobile visNIR system. Alternatively, stationary PSS of visNIR reflectance has been implemented using portable instruments (Ben-Dor et al., 2008; Kusumo et al., 2011; Viscarra Rossel et al., 2009; Waiser et al., 2007). There are

256

R.A. Viscarra Rossel et al.

fewer reports of portable, mid-IR systems for PSS ( Jahn and Upadhyaya, 2010; Reeves et al., 2010).

2.4. Laser-induced breakdown spectroscopy LIBS is made possible because of lasers. The technology uses an optically focused short-pulsed laser to heat the surface of the soil sample to the point of volatilization and ablation. This results in the generation of a hightemperature plasma on the surface of the sample. It is important to note that the plasma forms over a limited area so that only a very small amount of sample is measured during each event. As it cools, the excited atomic, ionic, and molecular fragments produced in the plasma emit radiation characteristic of the elemental composition within the volatilized material. A spectrometer capable of resolving spectra in the range 200900 nm is used to detect the emitted radiation. A sample spectrum is shown in Fig. 5. LIBS has been used for elemental analysis in geochemical exploration (Mosier-Boss et al., 2002), for the analysis of soil carbon (Cremers et al., 2001) and other elements (Hilbk-Kortenbruck et al., 2001). Fiber optic technology has made it possible to develop portable (Harmon et al., 2005) and mobile LIBS systems (Bousquet et al., 2008).

2.5. Microwaves The quantum energy of microwaves ranges between B12.4 and 12.4 μeV, near frequencies of 3 3 10113 3 109 Hz, and wavelengths of 1 3 1023 5 3 1025 m (Fig. 1). Microwave sensors are typically used for remote sensing

Figure 5 Typical soil laser-induced breakdown spectroscopy (LIBS) spectrum.

Proximal Soil Sensing for Measurements in Space and Time

257

( Jackson et al., 1984), but sensors have been constructed for measuring soil water proximally (Whalley, 1991). They measure either changes in the emissivity of the soil or changes in microwave attenuation caused by changes in water content. This dependence of the soil’s emissivity on its water content is due to the large contrast between the dielectric properties of free water (k0 5 80), dry soil (k0 5 25 depending on its bulk density), and air (k0 5 1). The large dielectric constant for water results from the alignment of the electric dipole moment of the water molecule in response to an applied field. As the water content of a soil increases, its dielectric constant and attenuation increase, and changes in soil emissivity are produced. Therefore, microwave sensors measure the thermal radiation emitted by the soil, which is generated within the volume of the soil and is dependent on the water content (i.e., dielectric properties) and temperature of the soil.

2.6. Radio waves Radio waves occur in the EM spectrum at frequencies less than 3 3 109 Hz with wavelengths greater than 1 3 1023 μ and energies less than 12.4 μeV (Fig. 1). 2.6.1. Time- and frequency-domain reflectometry and capacitance Time-domain reflectometry (TDR), frequency-domain reflectometry (FDR), and capacitance sensors use the dielectric properties of soils, that is, their permittivity, to measure water content. TDR instruments consist of a transmission line (TL) and a waveguide made up of two or three parallel metal rods that are inserted into the soil. The instrument produces a series of precisely timed electrical pulses with frequencies of 2 3 1073 3 109 Hz, which travel along the TLs and waveguide. These frequencies provide a response that is less dependent on soil-specific properties like texture, salt content, and temperature. Impedance along the waveguide varies with the dielectric constant of the bulk soil. The soil bulk dielectric constant is determined by measuring the time it takes for the electromagnetic pulse to propagate along the TL and waveguide surrounded by the soil. Since the propagation velocity is a function of the soil bulk dielectric constant, the latter is proportional to the square of the transit time out and back along the TL and waveguide. Because the dielectric constant of soil depends on the amount of water present, soil volumetric water content can be inferred from the reflected measurements. Noborio (2001) provides a good overview of the use of TDR for the measurement of soil water content and electrical conductivity. FDR and capacitance probes consist of two or more capacitors (rods, plates, or rings) that are inserted into the soil. Plates are usually annuli arranged concentrically to facilitate borehole measurements (Dean et al., 1987). These capacitors use the soil as a dielectric and hence depend on the

258

R.A. Viscarra Rossel et al.

soil water content. When the capacitor is connected to an oscillator to form an electric circuit, changes in soil water can be detected by changes in the circuit’s operating frequency. In FDR, the oscillator frequency is controlled within a certain range to determine the resonant frequency at which the amplitude is greatest, which is a measure of the soil water content. In capacitance, a measure of the soil’s permittivity is determined by measuring the charge time of the capacitor in the soil. There are three basic parts to a capacitance sensor (Dally et al., 1993): the target plate, the air space, and the sensor head. The target plate accumulates the voltage that will eventually dissipate across the air space. The air space is the gap between the target plate and the sensor head. This space can be filled with soil to increase or decrease the voltage being accumulated on the target plate. The sensor head measures the voltage accumulated by the target plate and dissipated through the soil. Whalley et al. (1992) developed one such sensor, which was also found to be sensitive to fluctuations in soil bulk density. Liu et al. (1996) and later Andrade-Sanchez et al. (2007) and Adamchuk et al. (2009) also evaluated a dielectric-based moisture sensor under dynamic conditions by incorporating it into a nylon block attached to an instrumented tine (Fig. 6). A series of studies demonstrated that salinity, texture, and temperature also affected measurements. Soil-specific calibrations are recommended because the operating frequency of these devices is generally below 1 3 108 Hz. At these frequencies, the bulk permittivity of the soil may change and measurements are more affected by texture, salinity, and bulk density.

Figure 6

A capacitance soil water content sensor prototype.

Proximal Soil Sensing for Measurements in Space and Time

259

2.6.2. Nuclear magnetic resonance Nuclear magnetic resonance (NMR) is based on the interaction between nuclear magnetic moments and applied static and radiofrequency magnetic fields (Matzkanin and Paetzotd, 1982). Paetzold et al. (1985) developed a tractor-mounted NMR instrument and used the technique to measure soil water content. The instrument detected and measured the NMR signal from the hydrogen in water. By adjusting the strength of the radiofrequency and static magnetic fields, the researchers were able to measure soil water content at depths of 38, 51, and 63 mm. Their findings suggested that the NMR signal is a linear function of volumetric water content and is not affected by clay mineralogy, soil organic matter, or texture. Furthermore, the technique can distinguish between water that is bound by clay particles and not available to plants, and that which is available for plant use. Magnetic resonance sounding (MRS) uses the NMR principle that is used in medical brain scanning (i.e., MRI or magnetic resonance imaging) to measure subsurface free water and hydraulic properties (Lubczynski and Roy, 2003). It is also known as surface NMR and can be used to measure water content and porosity to depths up to 1500 m. 2.6.3. Ground-penetrating radar GPR uses the transmission and reflection of high-frequency (106109 Hz) electromagnetic waves in the soil. They have transmitter and receiver antennas that can be moved across the soil surface (Fig. 7). Much like with TDR sensors, the primary control on the transmission and reflection of the electromagnetic energy is the dielectric constant. Because of the large contrast between the dielectric constants of water, air, and minerals, GPR can be used to measure variations in soil water content (Lambot et al., 2004). Unlike TDR, however, GPR measurements are non invasive and the sensors can measure soil water content of relatively large volumes of soil. The resolution of GPR images can be varied through the use of different antennae frequencies. Typically, higher frequencies increase the resolution at the expense of depth of penetration. Daniels et al. (1988) describe the fundamental principles of GPR. Knight (2001) provides an overview of GPR in environmental applications and Huisman et al. (2003) review its use for soil water determinations. The penetration depth of GPR measurements is affected by the electrical conductivity of the soil. Good penetration depth of up to about 15 m can be achieved in dry sandy soils or massive dry materials such as granite and limestone. As the conductivity increases, penetration depth decreases because the electromagnetic energy is more quickly dissipated into heat, causing a loss in signal strength at depth. In highly conductive soils, such as those with large amounts of clay, water and/or salt, penetration depth can be only a few centimeters. Slowly

260

R.A. Viscarra Rossel et al.

Figure 7

A ground-penetrating radar (GPR) system.

changing water contents are also difficult to detect with GPR, and water profiling is generally not possible with most types of instruments. More abrupt changes, such as wetting fronts, are easier to detect and this use of GPR is more appropriately applied in irrigated regions. 2.6.4. Electromagnetic induction EMI is a highly adaptable noninvasive technique that measures the apparent bulk electrical conductivity of soil (ECa). The instruments commonly have a transmitter and a receiver. Using a varying magnetic field of relatively low frequency (kHz), the technique induces currents in the ground in a way that ensures their amplitude is linearly related to the conductivity of the soil. The magnitude of these currents is determined by measuring the magnetic field that they generate. McNeill (1980) provides a good account. EMI has been used extensively in mapping soils since De Jong et al. (1979) first reported it. It has been particularly useful for mapping saline soils (Rhoades, 1993) and for precision agriculture (Corwin and Lesch, 2003). EMI coupled with a global positioning system (GPS) provides a rapid soil-mapping tool and, until now, may be the most commonly used proximal soil sensor (Fig. 8). Because most soil and rock minerals are very good insulators, the electrical conductivity sensed by an EMI unit is electrolytic and it takes place through the porewater system. The following factors are therefore important: shape, size, and connectivity of the pore system; water content; concentration of dissolved electrolytes in the soil water; temperature and phase of the pore water; and amount and composition of colloids (Rhoades et al., 1989).

Proximal Soil Sensing for Measurements in Space and Time

Figure 8

261

An on-the-go soil electromagnetic induction (EM-38) system.

While clay content, electrical conductivity of the soil solution, and water content are often recognized as the controlling factors that must be accounted for when calibrating EMI measurements (Williams and Baker, 1982; Williams and Hoey, 1987), it is not that simple. It is the pore system and its contents rather than the clay content per se that should be considered. Soils with significant clay content usually have a pore geometry dominated by finer-sized pores. In comparison to a sandy soil, greater proportions of these pores are filled and connected at comparable water contents, and this gives rise to their larger electrical conductivity. The bulk density of the soil should also be considered because it determines total porosity. Clay soils in most cropping areas usually have a substantial CEC, and cations in solution are in equilibrium with the charged clay surface—these cations also contribute to the electrolyte concentration. Finally, colloids—particularly those associated with organic matter—may also contribute to the measured conductivity. Effective measurement depth is a function of coil spacing and, under some conditions, frequency. For commercial EMI sensors, the depth of measurement can range from B0.37 m to more than 60 m. There are a number of operational issues with EMI sensors, including temperature effects (Robinson et al., 2004; Sudduth et al., 2001) and spurious signals due to nearby metal objects (Lamb et al., 2005).

2.7. Magnetic, gravimetric, and seismic sensors 2.7.1. Magnetics Magnetic sensors, or magnetometers, measure variations in the strength of the earth’s magnetic field and the data reflect the spatial distribution of

262

R.A. Viscarra Rossel et al.

magnetization in the ground. Magnetization of naturally occurring materials and rocks is determined by the quantity of magnetic minerals and by the strength and direction of the permanent magnetization carried by those minerals (Hansen et al., 2005). Typically, magnetics has been used for the detection of geological bodies; however, there is increasing use of the technique for near-surface applications. For example, for mapping field drainage for hydrologic modeling (Rogers et al., 2005); to better understand soil genesis and formation (Mathe and Leveque, 2003); to detect anthropogenic pollution on topsoils; through their associations with iron oxides (Schibler et al., 2002); and for rapid identification and mapping of soil heavy metal contamination ( Jordanovaa et al., 2008). 2.7.2. Gravity Gravity data can be collected using gravimeters (or gravitometers) and provide information on the local gravitational field. There are two types of gravimeter: relative and absolute. A relative gravimeter measures relative differences in the vertical component of the earth’s gravitational field based on variations in the extension of an internal spring in the gravimeter. The technique has typically been used to determine the subsurface configuration of structural basins, aquifer thickness, and geological composition. An absolute gravimeter measures the acceleration of free fall of a control mass. Absolute gravimetry can be used to measure mass water balances at regional or local scales (Nabighian et al., 2005). 2.7.3. Seismology Seismic reflection methods are sensitive to the speed of propagation of various kinds of elastic waves. The elastic properties and mass density of the medium in which the waves travel control the velocity of the waves and can be used to infer properties of the earth’s subsurface. Reflection seismology is used in exploration for hydrocarbons, coal, ores, minerals, and geothermal energy. It is also used for basic research into the nature and origin of rocks that make up the Earth’s crust (McCarthy and Thompson, 1988). It can be used in near-surface applications for engineering, groundwater and environmental surveying (Harry et al., 2005). A method similar to reflection seismology, which uses electromagnetic instead of elastic waves, is GPR.

2.8. Contact electrodes This section refers to techniques for measuring electrical properties of soils, such as their resistivity and dielectric, using direct injection of current into the soil using electrodes.

Proximal Soil Sensing for Measurements in Space and Time

263

2.8.1. Electrical resistivity Electrical resistivity (ER) can be used to determine the resistivity distribution of the measured soil volume. Measurements of ER usually require four electrodes: two to inject the current (current electrodes) and two to measure the resulting potential difference (potential electrodes). The ER of the soil is determined from this and measurements of the apparent electrical conductivity (ECa) are possible because resistivity is the reciprocal of conductivity. The technique has long been used in geophysics, and various configurations of electrodes can be used to control the volume and depth of measurement. The soil properties that affect measurements of soil with EMI instruments (see above) also affect resistivity measurements. Samoue¨lian et al. (2005) provide a good review on the use of ER in soil science. 2.8.2. Induced polarization Induced polarization (IP) measurements are essentially an extension of the four-electrode resistivity technique. IP operates by first injecting an electric current between a current electrode pair and the resulting voltage induced in the soil is measured between a potential electrode pair. However, IP captures both the charge loss (conduction) and the charge storage (polarization) characteristics of the soil. IP instruments have been used in hydrogeophysical applications, for example, to look at hydraulic properties of soil in the vadose zone (Binley et al., 2005; Bo¨rner et al., 1996). 2.8.3. Electrochemical sensors Electrochemical sensors have been developed to measure specific ions in solution. The most common of these are electrodes to measure pH; however, their uses for measuring various other ions are increasing in environmental applications. Their durability, portability, fast response, and ability to measure in unfiltered soil slurries are key advantages allowing direct measurements of soil chemical properties. PSS using electrochemical sensors is currently an active area of research with particular focus on the development of mobile soil pH, lime requirements, and nutrient sensing (Adamchuk et al., 1999; Adsett and Zoerb, 1991; Birrell and Hummel, 2001; Sibley et al., 2009; Viscarra Rossel and McBratney, 1997; Viscarra Rossel et al., 2005). Kim et al. (2009a) provide a recent review. 2.8.4. Ion-selective electrodes ISEs are potentiometric sensors that use ion-selective membranes to measure the concentration of the target species. When submerged in the solution to be analyzed, an electromotive force is generated at the sensing surface proportional to the log of the ion activity. The electromotive force

264

R.A. Viscarra Rossel et al.

can then be measured using a suitable reference system (e.g., a reference electrode). ISEs selective for many useful soil nutrients (nitrate, sodium, potassium, calcium) are commercially available and phosphate-selective electrodes for soil phosphorus are also being developed (Kim et al., 2007a). 2.8.5. Ion-sensitive field effect transistors ISFETs combine ISE technology with that of the field effect transistor (FET). The construction of the ISFET is as for the standard FET; however, the gate is replaced with a separate electrode (in contact with the analysis electrolyte) and the exposed insulating oxide (commonly SiO2 but also Al2O3, Ta2O5) is also left in contact with the electrolyte being analyzed. The charge developed on the oxide surface (due to proton interaction) now controls the sourcedrain current of the FET, which is then indicative of the electrolyte. Key advantages of pH ISFETs over standard glass pH electrodes are small size, increased durability, fast response, and the ability to mass produce using microelectronic manufacturing techniques. They have been used for proximal sensing of soil pH (Viscarra Rossel and Walter, 2004) and lime requirement (Viscarra Rossel et al., 2005). ISFETs can be chemically modified by depositing membrane layers on the oxide surface to produce CHEMFETs selective for other ionic species. CHEMFETs selective for nitrate, calcium, and potassium have been developed and evaluated for use in soil nutrient sensing (Artigas et al., 2001; Birrell and Hummel, 2000, 2001). 2.8.6. Metal electrodes Metal electrodes are also being explored for PSS applications to address a need for increased physical durability. Antimony electrodes are being researched as a durable alternative to glass electrodes in direct contact soil pH measurement (Adamchuk et al., 2009; Viscarra Rossel and McBratney, 1997). Kim et al. (2007a) explored the use of cobolt rod-based ISEs for measuring soil phosphates.

2.9. Mechanical sensors Another family of proximal soil sensors quantify soil properties by measuring the mechanical interaction between the sensor and the soil. Although there are no widely used commercial systems, a number of prototypes are being developed and include mechanical, acoustic, and fluid permeability sensors. 2.9.1. Integrated draft Soil strength, or mechanical resistance to failure, has been widely used to estimate the degree of soil compaction. Soil compaction and soil strength

Proximal Soil Sensing for Measurements in Space and Time

265

can be measured using tine-based sensors (Hayhoe et al., 2002; Lapen et al., 2002). A method to determine soil physical properties using specific draft measurements was proposed by Van Bergeijk et al. (2001). In their study, information gathered automatically during plowing was used to predict the spatial distribution of topsoil clay content. Sirjacobs et al. (2002) developed a soil strength sensor that was later evaluated by Hanquet et al. (2004). It consisted of a single chisel shank pulled through the soil at constant speed and a depth of 30 cm. In addition to the integrated (bulk) measure of draft, sensors have also been used to measure vertical variation in soils to identify hardpan layers. These sensors measure by moving the tool up and down while traveling across a field (Hall and Raper, 2005; Manor and Clark, 2001; Pitla et al., 2009; Stafford and Hendrick, 1988). 2.9.2. Mechanical resistance Soil penetrability is a measure of the effort required to force an object through the soil. Penetration resistance of soil is relatively easy to measure and is governed by several soil properties, including shear strength, compressibility, and friction between the soil and the metal. Numerous tip-based penetrometers have been developed (Fig. 9), including the standardized vertically operating cone penetrometer (ASABE, 2009), the single-tip horizontal soil impedance sensor (Alihamsyah et al., 1990), the multiple-tip horizontal soil impedance sensor (Chukwu and Bowers, 2005; Chung and Sudduth, 2006; Chung et al., 2006, 2008), and the vertically oscillating shank with a horizontal single-wedge sensor (Hall and Raper, 2005). While the vertically operated sensor provides the conventional means for measuring soil strength, horizontally operated tip-based sensors have been used for mobile, on-the-go sensing. In addition to multiple-tip soil impedance sensors, several attempts have been made to use tine-based sensors to perform mobile measurements of the entire profile. There are two approaches: (i) using an array of strain gauges mounted on a rigid tine (Adamchuk et al., 2001a, 2001b; Glancey et al., 1989) and (ii) multiple active cutting edges supported by independent load cells (Andrade-Sanchez et al., 2007, 2008; Khalilian et al., 2002). Hemmat and Adamchuk (2008) reviewed proximal soil sensor prototypes to measure compaction. 2.9.3. Fluid permeability Many soil processes depend on the effects of soil structure indirectly through hydraulic conductivity, air porosity, bulk density, and other relevant properties. Therefore, measuring the pressure required to inject a constant flow of air into the soil, as an indication of the relative soil pore space and the continuity of the pores, can provide a measure for soil compaction (Clement and Stombaugh, 2000). Air was forced into the soil

266

R.A. Viscarra Rossel et al.

Figure 9

A five-probe soil penetrometer system.

at a depth of 30 cm using a subsoiler shank and the measured pressure resistance was related to the air permeability of the subsoil. The sensor could detect changes in soil structure/compaction, moisture content, and soil type. Later, Koostra and Stombaugh (2003) redesigned the first version of the air permeability sensor to minimize the soil disturbance induced by the wide point of its shank. 2.9.4. Acoustic sensors The interaction between an implement and the soil creates noise. Thus, Liu et al. (1993) tested an acoustic method for determining soil texture. A shank with a rough surface and hollow cavity was equipped with a microphone that recorded the sound produced through the interaction of soil and shank. The frequency of the resulting sound was used to distinguish different types of soil. In a system developed by Tekeste et al. (2002), sound waves were used to detect compaction layers. A small microphone installed inside a horizontal cone attached to a tine was pulled through the soil. The amplitude of sound in a selected frequency range was compared to the cone index obtained at different depths in the soil profile. The instrument could successfully detect a prepared hardpan at a particular depth; however, in both studies, the authors needed it was necessary to account for background noise.

2.10. Telemetry—Wireless sensing Wireless sensor networks can be used for continuous and real-time monitoring of soil properties such as soil water and nutrients for irrigation. Commercial systems for monitoring soil water using wireless telemetry are currently available, for example, capacitance probes linked to mobile

Proximal Soil Sensing for Measurements in Space and Time

267

telephone systems or radio networks are being used in irrigated agriculture (Vellidis et al., 2008). Wireless sensor technologies are increasingly being used to monitor the condition of the environment (Zerger et al., 2010). With wireless systems, it is possible to obtain information about soil matric potential, water content, temperature, and other properties from remote locations in real time (Kim et al., 2009b; Vellidis et al., 2008). This improves the accuracy and convenience of monitoring soil water content. Irrigation systems manager can then use the data collected to optimize the use of resources in response to dynamic changes in soil condition and reduce the risk of water stress in crops (Han et al., 2009; Lamm and Aiken, 2008; Miranda et al., 2005). Ramanathan et al. (2006) describe a series of wireless networking case studies for monitoring soil CO2, temperature, and moisture. These systems also incorporate ISEs selective for ammonium, calcium, carbonate, chloride, pH, reductionoxidation, and nitrate. Lemos et al. (2004) describe a system that uses potassium ISEs along a PVC tube at various depths with real-time data relayed wirelessly to a base station. The main problems with wireless sensing using ISEs are durability, large sensor drift, and difficulties with in situ calibration. An alternative wireless technique that provides greater spatial coverage and reduced cost is ad hoc wireless networking or “mesh” networking. It is suited to situations where small rates and volumes of data exchange are required. It is based on the deployment of a large number of sensor “nodes” that are battery or solar powered and equipped with low power and low cost radio systems. Various configurations are possible in a network (with each node measuring one or more soil properties) and nodes can be used for relaying information to extend their range through self-configuring ad hoc networks. An example of such a system is the farm-based wireless sensor network developed by Sikka et al. (2006), which is part of a wider network and contains 12 soil moisture nodes using up to five gypsum blocks to measure soil moisture through the profile (to 1 m depth).

2.11. Geographic positioning and elevation In addition to locating sensor measurements on the landscape, the availability of differential global positioning systems (DGPS) and real-time kinematic (RTK) GPS systems make it possible to collect low cost, accurate digital elevation data. This data can then be used to create a digital elevation model (DEM) and provides information on surface geometry (e.g., slope, aspect, various curvatures, and wetness indices), which is an important descriptor of soil. Local variations in terrain control the movement of sediments, water, and solutes in the landscape. Soil formation is strongly influenced by these processes and the DEM and related attributes can be used to help characterize the spatial distribution of

268

R.A. Viscarra Rossel et al.

soil properties (Moore et al., 1993). A DEM also provides the landscape framework for interpreting results from other sensors (e.g., EM, γ-ray survey, and GPR) (Gish et al., 2005). Global positioning and the collection of elevation data are imperative for PSS, particularly for mobile and multisensor systems.

2.12. Multisensor systems As every soil-sensing technology has strengths and weaknesses and no single sensor can measure all soil properties, the selection of a complementary set of sensors to measure the required suite of soil properties is important. Integrating multiple proximal soil sensors in a single multisensor platform can provide a number of operational benefits over single-sensor systems, such as: robust operational performance increased confidence as independent measurements are made on the same soil extended attribute coverage increased dimensionality of the measurement space (e.g., different sensors measuring various portions of the EM spectrum). There are few reports of multisensor systems directed at PSS in the literature. For example, Christy et al. (2004) reported the use of a mobile sensor platform that simultaneously measures soil pH and ECa. An NIR sensor has also been recently added to this multisensor platform (Christy, 2008). Taylor et al. (2006) reported the development of a multisensor platform consisting of two EMI instruments, ER and pH sensors, a γ-radiometer, and a DGPS (Fig. 10A). Adamchuk and Christenson (2007) described a system that simultaneously measured soil mechanical resistance, optical reflectance, and capacitance (Fig. 10B). Yurui et al. (2008) reported the development of a multisensor technique for measuring soil physical properties (soil water, mechanical strength, and electrical conductivity). Other sophisticated integrated sensor systems have been developed for various applications. For example, the United States Army’s site characterization and analysis penetrometer system (SCAPS) is mounted on a 20-ton truck. Down-hole determinations are made to 50 m using realtime video; γ-ray spectrometers to detect radioactive waste; sensors to measure water content, pore water pressure, liquid and gas samplers; laser-induced fluorescence sensors to detect hydrocarbons; mass spectrometers to detect volatile organic compounds; LIBS to measure various metals; and XRF for measuring heavy metals. Eight SCAPS trucks are operated by three federal agencies in the United States and millions of dollars have been saved in site investigation and cleanup costs (USAEC, 2000).

269

Proximal Soil Sensing for Measurements in Space and Time

(A)

(B)

Figure 10 Multisensor platforms. (A) A multisensor platform with EMI, passive γ, electrical resistivity, and pH sensors and (B) one with mechanical, electrical, and optical sensors.

2.13. Core scanning and down-borehole technologies Core scanning and borehole sensors can be used to measure soil profiles, for example, to measure soil carbon stocks, determine subsoil constraints to root growth (e.g., subsurface acidity), or to characterize the soilwater regime. Some of the PSS technologies described above provide bulk measurements to a specific depth. For example, EMI sensors provide a depth-weighted ECa reading to a depth proportional to their coil spacing (Sudduth et al., 2010). However, there is still a need to develop core scanning and down-borehole technologies that can characterize the entire soil profile layer by layer to at least 1.5 m. Undisturbed soil cores can be readily collected with small drill rigs using either push-tubes or core samplers. There is an excellent opportunity to apply many of the methods considered above to an automated scanning system for soil cores. Commercial units have been developed for sediment and rock cores (Geotek, 2001) that include active γ-ray attenuation for measuring water content and bulk density, ER, magnetic susceptibility, and digital photography (Fig. 11). Research prototypes that allow core scanning for ECa (Myers et al., 2010) and for visNIR reflectance (Kusumo et al., 2011) have also been reported. Use of such rapid core measurement systems would allow soil surveys to be undertaken in a far more efficient manner and would be a natural complement to vehicle-mounted sensor systems. Down-borehole sensor systems also provide a means for characterizing soil profiles. Measurements of electrical conductivity in particular can be made at a well-defined depth and the sensor can integrate over a realistic volume of soil to reduce the effects of short-range variation (Myers et al., 2010). Cone penetrometers or other specialized probes can also be modified to contain sensors or fiber optic probes for visNIR spectroscopy (Ben-Dor et al., 2008; Hummel et al., 2004; Kweon et al., 2009), XRF (Elam et al., 1998) and LIBS (Mosier-Boss et al., 2002).

270

R.A. Viscarra Rossel et al.

Figure 11

A core scanning multisensor system.

3. Proximal Sensors Used to Measure Soil Properties Many soil properties can be measured with different proximal soil sensors. This section describes and gives examples of alternative techniques that are available for measuring soil properties.

3.1. Soil water and related properties Several sensor systems for measuring water content have been developed. Soil water content has been measured using active γ-ray attenuation (Pires et al., 2005), visNIR (Sudduth and Hummel, 1993; Whiting et al., 2004) and mid-IR spectroscopy ( Janik et al., 2007a), tine-mounted microwave sensors (Whalley, 1991), TDR and FDR, capacitance (Paltineanu and Starr, 1997), GPR (Huisman et al., 2003), and EMI and ER (Sudduth et al., 2005). Although total soil water content as measured by these sensors is useful, measurements of plant-available water capacity (PAWC) are more important for agriculture. PAWC is determined in the field by measuring differences between volumetric water content at the drained upper and lower limits after complete extraction of water by the plants. Whalley et al. (1992) evaluated multisensor capacitance probes in the nontraffic interrows of agricultural fields to monitor soil water dynamics over the growing season. Wireless sensor networks (Vellidis et al., 2008) can also be used for

Proximal Soil Sensing for Measurements in Space and Time

271

this. EMI surveys measured at different times may be used to approximate PAWC ( Jiang et al., 2007; Wong et al., 2006), but local calibration and careful interpretation are imperative. Rapid measurements of bulk density are needed to convert water content measurements to a volumetric basis (see below).

3.2. Nutrients and elements Soil nutrients are important for healthy plant growth. The macronutrients (nitrogen, potassium and phosphorus) are required in large quantities and are therefore managed and replaced as fertilizer on a crop-by-crop basis. They represent a significant input cost of food production both financially and environmentally: excessive nitrogen fertilizer can subsequently leach soil nitrates into waterways and have direct consequences on human and environmental health and water quality. There are various options for proximal sensing of plant nutrients and elements in soil (Kim et al., 2009a); however, their measurement is not straightforward because these properties show large variability in both space and time. This is particularly true for nitratenitrogen, which has been measured using mid-IR spectroscopy ( Jahn et al., 2006), LIBS (Harmon et al., 2005), and electrochemical techniques using ISEs and ISFETs (Adsett and Zoerb, 1991; Artigas et al., 2001; Birrell and Hummel, 2001; Davenport and Jabro, 2001; Kim et al., 2006; Sethuramasamyraja et al., 2008; Sibley et al., 2009). Measurement of soil phosphorus is difficult. Most indices estimate readily available (or labile) phosphates that occur in soil solution. These occur as freshly precipitated forms or as anions that can be readily removed from positively charged sites on clay and organic surfaces. However, most of the phosphorus in soil is very slowly available (or less labile). Apart from electrochemical methods, proximal soil sensors for measuring soil phosphorus are indirect and return variable results, although good correlations using visNIR spectroscopy have been reported in the literature (Bogrekci and Lee, 2005). Janik et al. (1998) also reported good results for phosphorus sorption using mid-IR spectroscopy, but not for available phosphorus. Kim et al. (2007a, 2007b) evaluated the ability of ion-selective membranes and cobalt-rod electrodes to quantify available phosphorus and reported relatively good success with cobalt electrodes. Potassium can be measured using passive γ-radiometry (Wong and Harper, 1999) and electrochemically (Kim et al., 2006; Sethuramasamyraja et al., 2008). Measurements of potassium using visNIR and mid-IR spectroscopy have also been reported but with variable results. Other major nutrients such as calcium and magnesium, however, appear to correlate well with both visNIR and mid-IR spectra (Lee et al., 2009; Viscarra

272

R.A. Viscarra Rossel et al.

Rossel and McBratney, 2008). Minor nutrients and elements can be measured directly using XRF (Kalnicky and Singhvi, 2001) and LIBS (Hussain et al., 2007) and electrochemically using ISEs or ISFETs (Artigas et al., 2001; Davenport and Jabro, 2001). Heavy metal contamination in soils can be measured using XRF, visNIR and mid-IR spectroscopy (Bray et al., 2009), and LIBS (Hilbk-Kortenbruck et al., 2001). Salinity and sodicity can be measured electrochemically with ISEs or ISFETs (Artigas et al., 2001; Davenport and Jabro, 2001) as well as with EMI and ER (Corwin et al., 2003).

3.3. Cation exchange capacity CEC determines the nutrient supply in soils, with cation nutrients in higher CEC soils generally more available to plants. CEC increases with increasing pH, clay, and organic matter in the soil. It also varies with the type of clay, with smectites having the highest CEC, followed by illites and kaolinites. CEC can be inferred using visNIR and mid-IR spectra (Sudduth and Hummel, 1993; Viscarra Rossel et al., 2006b). Reports of good correlations using EMI and ER instruments do exist (Sudduth et al., 2005). Since CEC is affected by soil texture, mineralogy, and organic matter content, it may be more accurately measured by combining measurements from different proximal sensors.

3.4. Carbon Carbon plays a key role in improving soil physical properties, increasing CEC and water-holding capacity, and improving soil structure. Soil carbon is thus considered important in assessing soil quality (Andrews et al., 2004). Furthermore, the ability of soils to sequester carbon is of increasing interest as a potential way to mitigate greenhouse gases in the atmosphere. Soil carbon can be measured using charge-coupled devices (Viscarra Rossel et al., 2008), visNIR and mid-IR (Viscarra Rossel et al., 2006b), LIBS (Cremers et al., 2001), and INS (Wielopolski et al., 2008). Carbon fractions can also be measured using visNIR (Cozzolino and Moron, 2006), but measurements appear to be more accurate using mid-IR spectroscopy ( Janik et al., 2007b).

3.5. pH As a measure of acidity, the level of soil pH is important in many processes, including availability of plant nutrients and efficacy of herbicides. Soil pH, buffering capacity, and lime requirement can be measured using ISE or ISFET systems (Adamchuk et al., 1999; Viscarra Rossel and McBratney, 1997; Viscarra Rossel and Walter, 2004; Viscarra Rossel et al., 2005).

Proximal Soil Sensing for Measurements in Space and Time

273

These properties can also be inferred using visNIR and mid-IR spectroscopy, the latter producing more accurate results (Viscarra Rossel and McBratney, 2008). Because the relationship between soil pH and lime requirement in different types of soil is not constant, measurements of lime requirement can be made using either a lime requirement buffer (Viscarra Rossel et al., 2005) or by combining sensors that account for the buffering capacity of the soil and pH itself.

3.6. Clay, silt, and sand Soil texture (clay, silt, and sand content) has been measured using γ-radiometrics (Viscarra Rossel et al., 2007), visNIR and mid-IR spectroscopy (Lee et al., 2009; Viscarra Rossel et al., 2006b, 2009), and EMI and ER ( James et al., 2003; Sudduth et al., 2005). However, measurements of silt are often less accurate than for clay. Sand content has also been measured with mid-IR spectroscopy (Viscarra Rossel et al., 2006b), where strong fundamental vibrations of siliconoxygen bonds exist. There is no response to sand in the visNIR spectrum as quartz is insensitive in this region, although sand may affect soil albedo and might also show some response due to iron oxide coatings on sand grains. EMI and ER have also been used to measure sand and determine soil textural boundaries (Carrol and Oliver, 2005; James et al., 2003).

3.7. Soil mineralogy Mineralogy strongly affects the physicochemical processes occurring in soils. In particular, the mineralogy of soil clays relates to soil fertility through CEC and also to soil water dynamics by virtue of the shrinkswell nature of the various clays. Clay mineralogy can be measured in situ by portable XRD and XRF instruments (Sarrazin et al., 2005), visNIR spectroscopy (Viscarra Rossel et al., 2006a, 2009), and mid-IR spectroscopy (Nguyen et al., 1991). Iron and its oxides have been measured using γ-radiometrics (Viscarra Rossel et al., 2007), XRF (Kalnicky and Singhvi, 2001), ultraviolet and visNIR spectroscopy (Viscarra Rossel et al., 2010b), and LIBS (Hussain et al., 2007).

3.8. Soil strength, bulk density, and related properties Soil strength can be determined using mechanical sensors for measuring soil mechanical resistance (Adamchuk and Christenson, 2007; AndradeSanchez et al., 2007; Chung et al., 2006; Hemmat et al., 2008). Simultaneous measurements of soil water content are often necessary to account for its relationship to soil strength. Measurements of soil bulk density using conventional techniques are slow and subject to large

274

R.A. Viscarra Rossel et al.

measurement error. The development of proximal sensors to measure bulk density is important because it makes more sense to provide measurements of the soil profile in a volumetric rather than gravimetric basis, for example, for reporting organic carbon, water content, and lime requirements. Bulk density and compaction can be inferred using active γ-ray attenuation measurements (Oliveira et al., 1998) and mechanically by measuring draught, depth, and soil water content (Mouazen and Ramon, 2006).

4. Summary Table 2 provides a summary of this review, showing the approximate frequency, energy, and wavelengths at which these sensors operate and whether the measurement is direct or indirect. For most soil properties, multiple sensing options can be used. For example, soil pH can be measured directly using ISFETs or indirectly using visNIR spectroscopy. There is widespread interest in the use of diffuse reflectance spectroscopy for PSS because several soil properties can be measured from a spectrum. Largely, however, the techniques are indirect (Table 2) and to be useful quantitatively, spectra must be related to a set of known reference samples through calibration. Successful generalization of indirect proximal soil sensor calibrations will depend on the type of soil: its mineralogy, particle-size distribution, presence of segregations (e.g., iron oxides and oxyhydroxides), soluble salts, water content, and the abundance and composition of organic matter. Inference using indirect techniques may be strong or weak (Table 2), but their measurements are invariably less accurate than direct methods. However, indirect methods are generally less expensive, technologically and methodologically better developed and more readily available to users. The different proximal soil sensors described in this article are in various stages of development, with some relying on expensive instruments designed for the laboratory, and others on purpose-built, lowercost, portable sensors designed for field application. Table 3 indicates the developmental status of various proximal soil sensors and their approximate costs.

5. General Discussion and Future Aspects PSS is not entirely new, although its development and that of new technologies is ongoing. The earliest reported use of a proximal soil

Table 2

Proximal soil sensors used to measure soil attributes

EM Range

γ-rays

X-rays

Frequency (Hz)

1022

10218

1016 1015 1013

Wavelength (m)

10212

10210

1028 XRD UV

INS TNM Active γ

Passive γ

XRF

Total carbon

D

i

D

Organic carbon

I

Technique

UVvisibleinfrared

Micro and radio waves

1012

1010

108

107

106

102

1026

1025

1024

1022

101

102

103

106

Vis

NIR

mid-IR

LIBS Micro WSN TDR

FDR Capac

GPR

EMI

ER

ECh Mech

I

D

D

D

I

D

D

I

D

D

I

I

D

D

i

I

i

i

D

Biochemical

Inorganic carbon I Total nitrogen

D

D

D

Nitratenitrogen Total phosphorus D

D

I

Extractable

I

I

D

phosphorus Total potassium

D

Extractable

D

D

I

I

I

I

I

I

D

D

i

I

D

D

i

I

I

D

D

D

D

i

I

I

I

I

I

D D

potassium Other major

D

D

D

D

nutrients Micronutrients,

D

elements Total iron

D

Iron oxides Heavy metals CEC

i

D

i D

D D

I

D i

i

(Continued)

Table 2

(Continued )

EM Range

γ-rays

X-rays

Frequency (Hz)

1022

10218

1016 1015 1013

Wavelength (m)

10212

10210

1028

XRF

XRD UV

Technique

INS TNM Active γ

Soil pH

Passive γ

UVvisibleinfrared

Micro and radio waves

1012

1010

108

107

106

102

1026

1025

1024

1022

101

102

103

106

Vis

NIR

mid-IR

FDR Capac

GPR

EMI

I

Buffering

I

I

I

I

LIBS Micro WSN TDR

ER

D

ECh Mech

D I

capacity and LR Salinity and

D

D

D

sodicity Physical Color Water content

D D

D

D

I

i

D

D

D

D

D

D

D

D

I

Matric potential Clay

I

Silt

I

Sand

I

Clay minerals

I

i

I

D

i

I

I

I

I

I

I

i

i

I

D

I

I

D

D

i

i

I

I

I

Soil strength Bulk density

D/I

D I

I

D

I

I

I

Porosity Rooting depth

I D

D

We denote the measurement as either physically based and direct (D) or correlative and indirect (I). Lower case “i” indicates weak inference. Note: INS, inelastic neutron scattering; TNM, thermalized neutron methods; XRF, X-ray fluorescence; XRD, X-ray diffractometry; UV, ultraviolet; vis, visible; NIR, near infrared; mid-IR, mid infrared; LIBS, laser-induced breakdown spectroscopy; Micro, microwaves; WSN, wireless sensor networks; TDR, time-domain reflectometry; FDR, frequency-domain reflectometry; Capac, capacitance; GPR, ground-penetrating radar; EMI, electromagnetic induction; ER, electrical resistivity; ECh, electrochemical; Mech, mechanical.

Table 3

Current development status of proximal soil sensors and approximate costs in US dollars (USD)

EM range wavelength (m) 212

γ-rays (10

)

X-rays (10210 m) UVvisR (1028 to 1024)

Microwave (1022) Radiowave (101 to 106)

Technique

Development status a

Approximate costs (USD)

INS TNM Active γ Passive γ

Research Commercial/research Commercial/research Commercial

XRF

Commercial/research

XRD UV Vis NIR

Commercial/research Commercial Commercial Commercial/research

MIR LIBS

Commercial/research Commercial/research

Microwave TDR FDR and capacitance GPR NMR EMI

Research Commercial Commercial

800,0001,500,00 10,00015,000 10,000 10,00070,000 (depends on crystal size and sensitivity) 800040,000 (OEM to commercial handheld units) 75,000 (portable combined XRF/XRD) 3000 (combined UVvisNIR 250900 nm) 10005000 (combined visNIR 4001000 nm) 10,000100,000 (visNIR depends on range and portability) 10,00075,000 (field and laboratory instruments) 15,00040,000 (dependent on the number of channels)  5001500 (sensor with display) 100500 (sensor only)

Commercial Research Commercial

80,000  10,00040,000 (depends on number of coils) (Continued)

Table 3

(Continued )

EM range wavelength (m)

Technique

Development status

Approximate costs (USD)

Electrical resistivity

ER

Commercial

Gypsum

Commercial

Electrochemical

Electrochemical

Commercial/research

Mechanical

Tillage Penetrometers

Research Commercial

Acoustic Pneumatic

Research Research

2003000 (for handheld/portable sensor); 7000 (for on-the-go sensor) 5100 (single sensor—dependent on type and quality) 501000 (with data logger—depends on sensor capabilities) 1001000 (single sensor—depends on ion, quality, reference electrode) 20010,000 (with logger/interface, e.g., interfacing multiple sensors to a computer)  15005000 (hand-operated device with digital data storage)  

Note: INS, inelastic neutron scattering; TNM, thermalized neutron methods; XRF, X-ray fluorescence; XRD, X-ray diffractometry; UV, ultraviolet; vis, visible; NIR, near infrared; MIR, mid infrared; LIBS, laser-induced breakdown spectroscopy; TDR, time-domain reflectometry; FDR, frequency-domain reflectometry; GPR, ground-penetrating radar; NMR, nuclear magnetic resonance; EMI, electromagnetic induction; ER, electrical resistivity; ISE, ion-selective electrode; ISFET, ion-sensitive field effect transistor. a Commercial INS systems are likely to appear in 23 years. Their cost will be determined largely by the cost of the neutron generator and the number of detectors used.

Proximal Soil Sensing for Measurements in Space and Time

279

sensor was in the 1920s when an instrumented drawbar dynamometer was used to discern spatial variation in soil compaction (reported in McBratney and Minasny, 2010). PSS gained prominence in soil science in around the past 30 years because of the realization that sensed data could provide good quality soil information more efficiently than laboratory methods of soil analysis, which can be expensive and time consuming. Some of the earlier reports using sensors to measure soil properties were given by Bowers and Bowen (1975), who measured electrical resistance to detect drying fronts; Rhoades and Corwin (1981), who used EMI to detect soil salinity; Perumpral (1987), who used standardized penetrometers for measuring soil compaction; and Dean et al. (1987), who used capacitance for measuring soil water. In the 1990s, the development and use of sensors for soil measurement gained momentum and various technologies were being reported, for example, GPR (Whalley et al., 1992), microwave sensing (Whalley, 1991), visible and NIR reflectance (Ben-Dor and Banin, 1995; Shonk et al., 1991; Sudduth and Hummel, 1993; Viscarra Rossel and McBratney, 1998), ISEs (Adsett and Zoerb, 1991), ISFETs (Birrell and Hummel, 1997; Viscarra Rossel and McBratney, 1997), mobile penetrometers (Alihamsyah and Humphries, 1991), acoustic sensors (Sabatier et al., 1990), and odor sensors to determine soil air composition (Persaud and Talou, 1996). Recognizing the increasing interest in soil sensing, Viscarra Rossel and McBratney (1998) used “proximal soil sensing” to describe measurement of soil properties with ground-based sensors. The development of PSS coincided with that of precision agriculture, which for some time appeared to be the application most suited to the use of proximal soil sensors. Interest in PSS is now more widespread (Viscarra Rossel et al., 2010a), and currently a wide range of technologies can be used for it. By its own merit, PSS is becoming a new discipline and is a topic of considerable interest in the soil, agricultural and environmental sciences, and engineering communities. The efficiency with which PSS can obtain soil data makes it naturally suited to many situations that require large amounts of quantitative data at fine spatial and/or temporal resolutions, for example, digital soil mapping, soil monitoring, precision agriculture, the assessment of contaminated sites, and measurement of subsurface hydrology. Although the fundamental scientific principles of the sensors that were reported early on remain the same, for the most part we understand them better and therefore are better placed to use them. For some sensors, technology has improved considerably, for example, with visNIR array-based detectors that increase instrument portability and ruggedness. Research has also refined our understanding of how we can best apply these sensors to measure soils and their properties. Our ability to extract useful information from the sensed data and to analyze large spatial

280

R.A. Viscarra Rossel et al.

datasets (Cressie and Kang, 2010) has improved because of advances in mathematical and statistical methods. Improvements in electronics and mobile computing, fueled by the consumer and automotive sectors, have made it feasible (and often relatively inexpensive) to interface with sensors in a user-friendly manner. The recently established International Union of Soil Sciences (IUSS) Working Group on Proximal Soil Sensing (www.proximalsoilsensing.org) aims to provide the framework for greater interaction between scientists and engineers with a common interest in developing proximal sensing technologies and mathematical and statistical techniques to better understand soil processes and spatiotemporal soil variability. Two large ongoing multinational European projects—the iSoil (Werban et al., (2010)) and Digisoil (Grandjean et al., 2010) projects that aim to develop PSS for digital soil mapping—present a step in the right direction and are the largest current investment in PSS research. The future of PSS lies in such interactions and multidisciplinary collaborations. Below, we short-list general considerations for future work. Development of soil sampling (measuring) designs for PSS—considering both geographic and property spaces. Research to define the most suitable technique or combination of techniques for measuring key soil properties, for example, bulk density, plant-available water, soil carbon, and carbon fractions. Research the often-complex interactions between the soil matrix and sensor signals. Research the underlying mechanisms that allow prediction of soil properties from indirect proximal soil sensors to develop theoretical calibrations that use soil knowledge. This will lead to improved accuracy, robustness, and applicability. Research the use of local versus global sensor calibrations. This might be soil property specific. Develop better signal processing and signal reconstruction methods. Often the methods used to process data from a proximal soil sensor are chosen ad hoc based on the experience of the particular investigator. Better, more widely applicable methods that could lead to standardization would help advance collaborative research and PSS. Develop data fusion methods that combine data from multiple sensors to produce useful soil information. Research the application of proximal soil sensors for diverse applications, for example, the use of multisensor platforms for digital soil mapping, soil monitoring, assessment of soil carbon, contaminated site assessment, and soilplant relationships. PSS provides soil scientists with an effective approach that can be used to learn more about soils. Proximal soil sensors allow rapid and

Proximal Soil Sensing for Measurements in Space and Time

281

inexpensive collection of precise, quantitative data at fine (spatial and temporal) resolutions, which can be used in more meaningful analyses to better understand soils and the spatiotemporal variability of their properties. Soil science needs PSS to device sustainable solutions to the global issues that we face today: food, water, and energy security and climate change. Our intent here is to raise awareness about PSS to further its research and development and to encourage the use of proximal soil sensors in different applications.

ACKNOWLEDGMENTS Dr. Viscarra Rossel would like to thank the CSIRO Division of Land and Water Capability Development Fund—“Which soil sensors do we use where?” for supporting this work.

REFERENCES Adamchuk, V. I., & Christenson, P. T. (2007). Development of an instrumented blade system for mapping soil mechanical resistance represented as a second-order polynomial. Soil and Tillage Research, 95, 7683. Adamchuk, V. I., Hempleman, C. R., & Jahraus, D. G. (2009). On-the-go capacitance sensing of soil water content. St. Joseph: American Society of Agricultural and Biological Engineers. Adamchuk, V. I., Hummel, J. W., Morgan, M. T., & Upadhyaya, S. K. (2004). On-the-go soil sensors for precision agriculture. Computers and Electronics in Agriculture, 44, 7191. Adamchuk, V. I., Morgan, M. T., & Ess, D. R. (1999). An automated sampling system for measuring soil pH. Transactions of the American Society of Agricultural Engineers, 42, 885891. Adamchuk, V. I., Morgan, M. T., & Sumali, H. (2001a). Application of a strain gauge array to estimate soil mechanical impedance on-the-go. Transactions of the American Society of Agricultural Engineers, 44(6), 13771383. Adamchuk, V. I., Morgan, M. T., & Sumali, H. (2001b). Mapping of spatial and vertical variation of soil mechanical resistance using a linear pressure model. St. Joseph: American Society of Agricultural Engineers. Adamchuk, V. I., Viscarra Rossel, R. A., Marx, D. B., & Samal, A. K. (2008). Enhancement of on-the-go soil sensor data using guided sampling. In R. Kholsa (Ed.), Proceedings of the ninth international conference on precision agriculture, Denver, Colorado, July 2023, 2008. Fort Collins: Colorado State University(CD publication, 13 pages). Adamchuk, V. I., Viscarra Rossel, R. A., Marx, D. B., & Samal, A. K. (2011). Using targeted sampling to process multivariate soil sensing data. Geoderma, 163, 6373. Adsett, J. F., & Zoerb, G. C. (1991). Automated field monitoring of soil nitrate levels. In Automated agriculture for the 21st century, Proceedings of the 1991 symposium (pp. 326335). St. Joseph: American Society of Agricultural Engineers. Alihamsyah, T., & Humphries, E. G. (1991). On-the-go mechanical impedance measurements. In Automated agriculture for the 21st century, Proceedings of the 1991 symposium (pp. 300306). St. Joseph: American Society of Agricultural Engineers. Alihamsyah, T., Humphries, E. G., & Bowers, C. G., Jr (1990). A technique for horizontal measurement of soil mechanical impedance. Transactions of the American Society of Agricultural Engineers, 33(1), 7377.

282

R.A. Viscarra Rossel et al.

Andrade-Sanchez, P., Upadhyaya, S. K., & Jenkins, B. M. (2007). Development, construction, and field evaluation of a soil compaction profile sensor. Transactions of the American Society of Agricultural and Biological Engineers, 50, 719725. Andrade-Sanchez, P., Upadhyaya, S. K., Plouffe, C., & Poutre, B. (2008). Development and field evaluation of a field-ready soil compaction profile sensor for real-time applications. Applied Engineering in Agriculture, 24, 743750. Andrews, S. S., Karlen, D. L., & Cambardella, C. A. (2004). The soil management assessment framework: A quantitative soil quality evaluation method. Soil Science Society of America Journal, 68, 19451962. Artigas, J., Beltran, A., Jimenez, C., Baldi, A., Mas, R., & Dominguez, C., et al. (2001). Application of ion sensitive field-effect transistor based sensors to soil analysis. Computers and Electronics in Agriculture, 31, 281293. ASABE (2009). ASAE Standard S313.3 FEB1999 (R2009): Soil cone penetrometer. St. Joseph: American Society of Agricultural and Biological Engineers. Ben-Dor, E., & Banin, A. (1995). Near-infrared analysis as a rapid method to simultaneously evaluate several soil properties. Soil Science Society of America Journal, 59, 364372. Ben-Dor, E., Heller, D., & Chudnovsky, A. (2008). A novel method of classifying soil profiles in the field using optical means. Soil Science Society of America Journal, 72, 11131122. Binley, A., Slater, L., Fukes, M., & Cassiani, G. (2005). The relationship between frequency dependent electrical conductivity and hydraulic properties of saturated and unsaturated sandstone. Water Resources Research, 41, 13. Birrell, S. J., & Hummel, J. W. (1997). Multi-sensor ISFET system for soil analysis. In J. V Stafford (Ed.), Precision agriculture ’97, Vol. I: Spatial variability in soil and crop (pp. 459468). Oxford: Bios Scientific Publishers. Birrell, S. J., & Hummel, J. W. (2000). Membrane selection and ISFET configuration evaluation for soil nitrate sensing. Transactions of the American Society of Agricultural Engineers, 43, 197206. Birrell, S. J., & Hummel, J. W. (2001). Real-time multi ISFET/FIA soil analysis system with automatic sample extraction. Computers and Electronics in Agriculture, 32, 4567. Bogrekci, I., & Lee, W. S. (2005). Spectral phosphorus mapping using diffuse reflectance of soils and grass. Biosystems Engineering, 91, 305312. Bo¨rner, F. D., Schopper, J. R., & Weller, A. (1996). Evaluation of transport and storage properties in the soil and groundwater zone from induced polarization measurements. Geophysical Prospecting, 44, 583602. Bousquet, B., Travaille´, G., Ismae¨l, A., Canioni, L., Michel-Le Pierre`s, K., & Brasseur, E., et al. (2008). Development of a mobile system based on laser-induced breakdown spectroscopy and dedicated to in situ analysis of polluted soils. Spectrochimica Acta Part B, 63, 10851090. Bowers, C. G., & Bowen, H. D. (1975). Drying front sensing and signal evaluation for planters. Transactions of the American Society of Agricultural Engineers, 18, 10511056. Bowers, S. A., & Hanks, R. J. (1965). Reflection of radiant energy from soils. Soil Science, 100, 130138. Bray, J., Viscarra Rossel, R. A., & McBratney, A. B. (2009). Diagnostic screening of urban soil contaminants using diffuse reflectance spectroscopy. Australian Journal of Soil Research, 47, 433442. Brooks, F. A. (1952). Atmospheric radiation and its reflection from the ground. Journal of Meteorology, 9, 4152. Carrol, Z. L., & Oliver, M. A. (2005). Exploring the spatial relations between soil physical properties and apparent electrical conductivity. Geoderma, 128, 354374. Christy, C. D. (2008). Real-time measurement of soil attributes using on-the-go near infrared reflectance spectroscopy. Computers and Electronics in Agriculture, 61, 1019.

Proximal Soil Sensing for Measurements in Space and Time

283

Christy, C. D., Collings, K., Drummond, P., & Lund, E. (2004). A mobile sensor platform for measurement of soil pH and buffering. St. Joseph: American Society of Agricultural and Biological Engineers. Chukwu, E., & Bowers, C. G. (2005). Instantaneous multiple-depth soil mechanical impedance sensing from a moving vehicle. Transactions of the American Society of Agricultural Engineers, 48, 885894. Chung, S. O., & Sudduth, K. A. (2006). Soil failure models for vertically operating and horizontally operating strength sensors. Transactions of the American Society of Agricultural and Biological Engineers, 49, 851863. Chung, S. O., Sudduth, K. A., & Hummel, J. W. (2006). Design and validation of an on-the-go soil strength profile sensor. Transactions of the American Society of Agricultural and Biological Engineers, 49, 514. Chung, S. O., Sudduth, K. A., Plouffe, C., & Kitchen, N. R. (2008). Soil bin and field tests of an on-the-go soil strength profile sensor. Transactions of the American Society of Agricultural and Biological Engineers, 51, 518. Clement, B. R., & Stombaugh, T. S. (2000). Continuously-measuring soil compaction sensor development. St. Joseph: American Society of Agricultural Engineers. Corwin, D. L., Kaffka, S. R., Hopmans, J. W., Mori, Y., Lesch, S. M., & Oster, J. D. (2003). Assessment and field-scale mapping of soil quality properties of a saline-sodic soil. Geoderma, 114, 231259. Corwin, D. L., & Lesch, S. M. (2003). Application of soil electrical conductivity to precision agriculture: Theory, principle and guidelines. Agronomy Journal, 95, 455471. Cozzolino, D., & Moron, A. (2006). Potential of near-infrared reflectance spectroscopy and chemometrics to predict soil organic carbon fractions. Soil and Tillage Research, 85, 7885. Cremers, D. A., Ebinger, M. H., Breshears, D. D., Unkefer, P. J., Kammerdiener, S. A., & Ferris, M. J., et al. (2001). Measuring total soil carbon with laser-induced breakdown spectroscopy (LIBS). Journal of Environmental Quality, 30(6), 22022206. Cressie, N., & Kang, E. L. (2010). High-resolution digital soil mapping: Kriging for very large datasets. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 4963). New York, NY: Springer-Verlag. Dally, J. W., Riley, W. F., & McConnell, K. G. (1993). Instrumentation for engineering measurements. Hoboken, NJ: John Wiley & Sons. Daniels, D. J., Gunton, D. J., & Scott, H. F. (1988). Introduction to subsurface radar. IEEE Proceedings—F Radar and Signal Processing, 135, 278320. Davenport, J. R., & Jabro, J. D. (2001). Assessment of hand held ion selective electrode technology for direct measurement of soil chemical properties. Communications in Soil Science and Plant Analysis, 32, 30773085. de Gruijter, J. J. (2002). Sampling. In J. H. Dane, & G. Clarke (Eds.), Methods of soil analysis. Part 4: Physical methods (pp. 4577). Madison, WI: Soil Science Society of America. de Gruijter, J., Brus, D., Bierkens, M., & Knotters, M. (2006). Sampling for natural resource monitoring. Berlin: Springer. de Gruijter, J. J., McBratney, A. B., & Taylor, J. (2010). Sampling for high resolution soil mapping. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 314). New York, NY: Springer-Verlag. De Jong, E., Ballantyne, A. K., Cameron, D. R., & Read, D. W. L. (1979). Measurement of apparent electrical conductivity of soils by an electromagnetic induction probe to aid salinity surveys. Soil Science Society of America Journal, 43, 810812. Dean, T. J., Bell, J. P., & Baty, A. T. P. (1987). Soil moisture measurement by an improved capacitance technique: Part 1. Sensor design and performance. Journal of Hydrology, 93, 6778.

284

R.A. Viscarra Rossel et al.

Elam, W. T., Adams, J. W., Hudson, K. R., McDonald, B., & Gilfrich, J. V. (1998). Subsurface measurement of soil heavy-metal concentrations with the SCAPS X-ray fluorescence (XRF) metals sensor. Field Analytical Chemistry and Technology, 2, 97102. Gardner, W. H, & Kirkham, D. (1952). Determination of soil water by neutron scattering. Soil Science, 73, 391401. Geotek (2001). Multi-sensor core logger. Daventry: Geotek. Gianoncelli, A., Castaing, J., Ortega, L., Dooryhee, E., Salomon, J., & Walter, P., et al. (2008). A portable instrument for in situ determination of the chemical and phase compositions of cultural heritage objects. X-ray Spectrometry, 37(4), 418423. Gish, T. J., Walthall, C. L., Daughtry, C. S. T., & Kung, K. J. S. (2005). Using soil moisture and spatial yield patterns to identify subsurface flow pathways. Journal of Environmental Quality, 34, 274286. Glancey, J. L., Upadhyaya, S. K., Chancellor, W. J., & Rumsey, J. W. (1989). An instrumented chisel for the study of soil-tillage dynamics. Soil and Tillage Research, 14(1), 124. Grandjean, G., Cerdan, O., Richard, G., Cousin, I., Lagacherie, P., & Tabbagh, A., et al. (2010). DIGISOIL: An integrated system of data collection technologies for mapping soil properties. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 89101). New York, NY: Springer-Verlag. Hall, H. E., & Raper, R. L. (2005). Development and concept evaluation of an on-the-go soil strength measurement system. Transactions of the American Society of Agricultural Engineers, 48(2), 469477. Han, Y. J., Khalilian, A., Owino, T. O., Farahani, H. J., & Moore, S. (2009). Development of Clemson variable-rate lateral irrigation system. Computers and Electronics in Agriculture, 68, 108113. Hanquet, H., Sirjacobs, D., Destain, M. F., Frnkinet, M., & Verbrugge, J. C. (2004). Analysis of soil variability measured with a soil strength sensor. Precision Agriculture, 5 (3), 227246. Hansen, R. O., Racic, L., & Grauch, V. J. S. (2005). Magnetic methods in near-surface geophysics. In D. K. Butler (Ed.), Near-surface geophysics: Investigations in geophysics no. 13 (pp. 151175). Tulsa, OK: Society of Exploration Geophysicists. Harmon, R. S., De Lucia, F. C., Miziolek, A. W., McNesby, K. L., Walters, R. A., & French, P. D. (2005). Laser-induced breakdown spectroscopy (LIBS)—An emerging field-portable sensor technology for real-time, in-situ geochemical and environmental analysis. Geochemistry: Exploration, Environment, Analysis, 5, 2128. Harry, D. L., Koster, J. W., Bowling, J. C., & Rodriguez, A. B. (2005). Multichannel analysis of surface waves generated during high-resolution seismic reflection profiling of a fluvial aquifer. Journal of Environmental and Engineering Geophysics, 10, 123133. Hayhoe, H. N., Lapen, D. R., & McLaughlin, N. B. (2002). Measurements of mouldboard plow draft: I. Spectrum analysis and filtering. Precision Agriculture, 3, 225236. Hemmat, A., & Adamchuk, V. I. (2008). Sensor systems for measuring spatial variation in soil compaction. Computers and Electronics in Agriculture, 63, 89103. Hemmat, A., Adamchuk, V. I., & Jasa, P. (2008). Use of an instrumented disc coulter for mapping soil mechanical resistance. Soil and Tillage Research, 98(2), 150163. Hilbk-Kortenbruck, A., Noll, R., Wintjens, P., Falk, H., & Becker, C. (2001). Analysis of heavy metals in soils using laser-induced breakdown spectrometry combined with laser-induced fluorescence. Spectrochimica Acta Part B: Atomic Spectroscopy, 56(6), 933945. Huisman, J. A., Hubbard, S. S., Redman, J. D., & Annan, A. P. (2003). Measuring soil water content with ground penetrating radar: A review. Vadose Zone Journal, 2, 476491. Hummel, J. W., Ahmad, I. S., Newman, S. C., Sudduth, K. A., & Drummond, S. T. (2004). Simultaneous soil moisture and cone index measurement. Transactions of the American Society of Agricultural Engineers, 47, 607618.

Proximal Soil Sensing for Measurements in Space and Time

285

Hussain, T., Gondal, M. A., Yamani, Z. H., & Baig, M. A. (2007). Measurement of nutrients in green house soil with laser induced breakdown spectroscopy. Environmental Monitoring and Assessment, 124, 131139. International Atomic Energy Agency, IAEA (2003). Guidelines for radioelement mapping using gamma ray spectrometry data. Vienna: IAEA. Islam, K., Singh, B., & McBratney, A. (2003). Simultaneous estimation of several soil properties by ultra-violet, visible, and near-infrared reflectance spectroscopy. Australian Journal of Soil Research, 41, 11011114. Jackson, T. J., Schmugge, T. J., & O’Neill, P. (1984). Passive microwave remote sensing of soil moisture from an aircraft platform. Remote Sensing of Environment, 14, 135151. Jahn, B. R., Linker, R., Upadhyaya, S. K., Shaviv, A., Slaughter, D. C., & Shmulevich, I. (2006). Mid-infrared spectroscopic determination of soil nitrate content. Biosystems Engineering, 94, 505515. Jahn, B. R., & Upadhyaya, S. K. (2010). Determination of soil nitrate and organic matter content using portable, filter-based mid-infrared spectroscopy. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 143152). New York, NY: Springer-Verlag. James, I. T., Waine, T. W., Bradley, R. I., Taylor, J. C., & Godwin, R. J. (2003). Determination of soil type boundaries using electromagnetic induction scanning techniques. Biosystems Engineering, 86, 421430. Janik, L. J., Merry, R. H., Forrester, S. T., Lanyon, D. M., & Rawson, A. (2007a). Rapid prediction of soil water retention using mid infrared spectroscopy. Soil Science Society of America Journal, 71, 507514. Janik, L. J., Merry, R. H., & Skjemstad, J. O. (1998). Can mid infrared diffuse reflectance analysis replace soil extractions? Australian Journal of Experimental Agriculture, 38, 681696. Janik, L. J., Skjemstad, J. O., Shepherd, K. D., & Spouncer, L. R. (2007b). The prediction of soil carbon fractions using mid-infrared partial least square analysis. Australian Journal of Soil Research, 45, 7381. Jiang, P., Anderson, S. H., Kitchen, N. R., Sudduth, K. A, & Sadler, E. J. (2007). Estimating plant-available water capacity for claypan landscapes using apparent electrical conductivity. Soil Science Society of America Journal, 71, 19021908. Jordanovaa, N., Jordanovaa, D., & Tsachevab, T. (2008). Application of magnetometry for delineation of anthropogenic pollution in areas covered by various soil types. Geoderma, 144, 557571. Kalnicky, D. J., & Singhvi, R. (2001). Field portable XRF analysis of environmental samples. Journal of Hazardous Materials, 83, 93122. Khalilian, A., Han, Y. J., Dodd, R. B., Sullivan, M. J., Gorucu, S., & Keskin, M. (2002). A control system for variable depth tillage. St. Joseph: American Society of Agricultural Engineers (ASAE Paper No. 02-1209). Kim, H. J., Hummel, J. W., & Birrell, S. J. (2006). Evaluation of nitrate and potassium ion-selective membranes for soil macronutrient sensing. Transactions of the American Society of Agricultural and Biological Engineers, 49, 597606. Kim, H. J., Hummel, J. W., Sudduth, K. A., & Birrell, S. J. (2007a). Evaluation of phosphate ion-selective membranes and cobalt-based electrodes for soil nutrient sensing. Transactions of the American Society of Agricultural and Biological Engineers, 50, 415425. Kim, H. J., Hummel, J. W., Sudduth, K. A., & Motavalli, P. P. (2007b). Simultaneous analysis of soil macronutrients using ion-selective electrodes. Soil Science Society of America Journal, 71, 18671877. Kim, H. J., Sudduth, K. A., & Hummel, J. W. (2009a). Soil macronutrient sensing for precision agriculture. Journal of Environmental Monitoring, 11, 18101824.

286

R.A. Viscarra Rossel et al.

Kim, Y., Evans, R. G., & Iverson, W. M. (2009b). Evaluation of closed-loop site-specific irrigation with wireless sensor network. Journal of Irrigation and Drainage Engineering, 135, 2531. Knight, R. (2001). Ground penetrating radar for environmental applications. Annual Review of Earth and Planetary Sciences, 29, 229255. Koostra, B. K., & Stombaugh, T. S. (2003). Development and evaluation of a sensor to continuously measure air permeability of soil. St. Joseph: American Society of Agricultural Engineers. Kusumo, B. H., Hedley, M. J., Hedley, C. B., & Tuohy, M. P. (2011). Measuring carbon dynamics in field soils using soil spectral reflectance: Prediction of maize root density, soil organic carbon, and nitrogen content. Plant Soil, 338, 233245. Kweon, G. Y., Lund, E., Maxton, C., Drummond, P., & Jensen, K. (2009). Soil profile measurement of carbon contents using a probe-type visNIR spectrophotometer. Journal of Biosystems Engineering, 34, 382389. Lamb, D. W., Mitchell, A., & Hyde, G. (2005). Vineyard trellising with steel posts distorts data from EM soil surveys. Australian Journal of Grape and Wine Research, 11, 2432. Lambot, S., Rhebergen, J., van den Bosch, I., Slob, E. C., & Vanclooster, M. (2004). Measuring the soil water content profile of a sandy soil with an off-ground monostatic ground penetrating radar. Vadose Zone Journal, 3, 10631071. Lamm, F. R., & Aiken, R. M. (2008). Comparison of temperature-time threshold and ET-based irrigation scheduling for corn production. St. Joseph: American Society of Agricultural and Biological Engineers. Lapen, D. R., Hayhoe, H. N., Topp, G. C., McLaughlin, N. B., Gregorich, E. G., & Curnoe, W. E. (2002). Measurement of mouldboard draft: II. Draft-soil-crop and yield-draft associations. Precision Agriculture, 3, 237257. Lee, K. S., Lee, D. H., Sudduth, K. A., Chung, S. O., Kitchen, N. R., & Drummond, S. T. (2009). Wavelength identification and diffuse reflectance estimation for surface and profile soil properties. Transactions of the American Society of Agricultural and Biological Engineers, 52, 683695. Lemos, S. G., Nogueira, A. R. A., Torre-Netto, A., Parra, A., Artigas, J., & Alonso, J. (2004). In-soil potassium sensor system. Journal of Agricultural and Food Chemistry, 52, 58105815. Lesch, S. M. (2005). Sensor-directed spatial response surface sampling designs for characterizing spatial variation in soil properties. Computers and Electronics in Agriculture, 46(1), 153180. Liu, W., Gaultney, L. D., & Morgan, M. T. (1993). Soil texture detection using acoustic methods. St. Joseph: American Society of Agricultural Engineers. Liu, W., Upadhyaya, S. K., Kataoka, T., & Shibusawa, S. (1996). Development of a texture/ soil compaction sensor. In P. C. Robert, R. H. Rust, & W. E. Larson (Eds.), Proceedings of the third international conference on precision agriculture (pp. 617630). Minneapolis, MN: University of Minnesota, Precision Agriculture Center. Lubczynski, M., & Roy, J. (2003). Magnetic resonance sounding: New method for ground water assessment. Ground Water, 42, 291302. Manor, G., & Clark, R. L. (2001). Development of an instrumented subsoiler to map soil hardpans and real-time control of subsoiler depth. St. Joseph: American Society of Agricultural Engineers. Mathe, V., & Leveque, F. (2003). High resolution magnetic survey for soil monitoring: Detection of drainage and soil tillage effects. Earth and Planetary Science Letters, 212, 241251. Matzkanin, G. A., & Paetzotd, R. F. (1982). Measuring soil water content using pulsed nuclear magnetic resonance. St. Joseph: American Society of Agricultural Engineers.

Proximal Soil Sensing for Measurements in Space and Time

287

McBratney, A. B., & Minasny, B. (2010). The sun has shown here antecedently. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 6776). New York, NY: Springer-Verlag. McCarthy, J., & Thompson, G. A. (1988). Seismic imaging of extended crust with emphasis on the western United States. Bulletin of the Geological Society of America, 100, 13611374. McNeill, J. D. (1980). Electromagnetic terrain conductivity measurement at low induction numbers. Ontario: Geonics Ltd. Minasny, B., & McBratney, A. B. (2006). A conditioned Latin hypercube method for sampling in the presence of ancillary information. Computers and Geosciences, 32(9), 13781388. Minty, B. R. S. (1997). Fundamentals of airborne gamma-ray spectrometry. AGSO Journal of Australian Geology and Geophysics, 17, 3950. Miranda, F. R., Yoder, R. E., Wilkerson, J. B., & Odhiambo, L. O. (2005). An autonomous controller for site-specific management of fixed irrigation systems. Computers and Electronics in Agriculture, 48, 183197. Moore, D. M., & Reynolds, R. C., Jr. (1997). X-ray diffraction and the identification and analysis of clay minerals (2nd ed.). New York, NY: Oxford University Press. Moore, I. D., Gessler, P. E., Nielsen, G. A., & Peterson, G. A. (1993). Soil attribute prediction using terrain analysis. Soil Science Society of America Journal, 57, 443452. Mosier-Boss, P. A., Lieberman, S. H., & Theriault, G. A. (2002). Field demonstrations of a direct push FO-LIBS metal sensor. Environmental Science and Technology, 36, 39683976. Mouazen, A. M., De Baerdemaeker, J., & Ramon, H. (2005). Towards development of on-line soil moisture content sensor using a fibre-type NIR spectrophotometer. Soil and Tillage Research, 80, 171183. Mouazen, A. M., & Ramon, H. (2006). Development of on-line measurement system of bulk density based on on-line measured draught, depth and soil moisture content. Soil and Tillage Research, 86, 218229. Myers, D. B., Kitchen, N. R., Sudduth, K. A., Grunwald, S., Miles, R. J., & Sadler, E. J., et al. (2010). Combining proximal and penetrating soil electrical conductivity sensors for high resolution digital soil mapping. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 233243). New York, NY: SpringerVerlag. Nabighian, M. N., Ander, M. E., Grauch, V. J. S., Hansen, R. O., LaFehr, T. R., & Li, Y., et al. (2005). The historical development of the gravity method in exploration. Geophysics, 70, 63ND89ND. Nguyen, T. T., Janik, L. J., & Raupach, M. (1991). Diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy in soil studies. Australian Journal of Soil Research, 29, 4967. Noborio, K. (2001). Measurement of soil water content and electrical conductivity by time domain reflectometry: A review. Computers and Electronics in Agriculture, 31, 213237. Oliveira, J. M. C., Appolonia, C. R., Coimbra, M. M., Reichardt, K., Bacchi, O. O. S., & Ferraza, E., et al. (1998). Soil structure evaluated by gamma-ray attenuation. Soil and Tillage Research, 48, 127133. Paetzold, R. F., Matzkanin, G. A., & De Los Santos, A. (1985). Surface soil-water content measurements using pulsed nuclear magnetic resonance techniques. Soil Science Society of America Journal, 49, 537540. Paltineanu, I. C., & Starr, J. L. (1997). Real-time soil water dynamics using multisensor capacitance probes: Laboratory calibration. Soil Science Society of America Journal, 61(6), 15761585. Persaud, K. C., & Talou, T. (1996). Hunting the black truffle. Land Technology, 3(1), 45.

288

R.A. Viscarra Rossel et al.

Perumpral, J. V. (1987). Cone penetrometer applications: A review. Transactions of the American Society of Agricultural Engineers, 30, 939944. Pires, L. F., Bacchi, O. O. S., & Reichardt, K. (2005). Soil water retention curve determined by gamma-ray beam attenuation. Soil and Tillage Research, 82, 8997. Pitla, S. K., Wells, L. G., & Shearer, S. A. (2009). Integration of an extended octagonal ring transducer and soil coulterometer for identifying soil compaction. Applied Engineering in Agriculture, 25, 647652. Ramanathan, N., Schoellhammer, T. D., Estrin, M. H., Harmon, T., Kohler, E., & Srivastava, M. (2006). The final frontier: Embedding networked sensors in the soil. Los Angeles: Center for Embedded Networked Sensing, UCLA. Reeves, J. B., McCarty, G. W., & Hively, W. D. (2010). Mid-versus near-infrared spectroscopy for on-site analysis of soil. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing. New York, NY: Springer-Verlag. Rhoades, J. D. (1993). Electrical conductivity methods for measuring and mapping soil salinity. Advances in Agronomy, 49, 201251. Rhoades, J. D., & Corwin, D. L. (1981). Determining soil electrical conductivity-depth relations using an inductive electromagnetic soil conductivity meter. Soil Science Society of America Journal, 45, 255260. Rhoades, J. D., Manteghi, N. A., Shouse, P. J., & Alves, W. J. (1989). Soil electrical conductivity and soil salinity: New formulations and calibrations. Soil Science Society of America Journal, 53, 433439. Robinson, D. A., Lebron, I., Lesch, S. M., & Shouse, P. (2004). Minimizing drift in electrical conductivity measurements in high temperature environments using the EM-38. Soil Science Society of America Journal, 68, 339345. Rogers, M. B., Cassidy, J. R., & Dragila, M. I. (2005). Ground-based magnetic surveys as a new technique to locate subsurface drainage pipes: A case study. Applied Engineering in Agriculture, 21, 421426. Sabatier, J. M., Hess, H., Arnott, W. P., Attenborough, K., Romkens, M. J. M., & Grissinger, E. H. (1990). In situ measurements of soil physical properties by acoustic techniques. Soil Science Society of America Journal, 54, 658672. Samoue¨lian, A., Cousin, I., Tabbagh, A., Bruand, A., & Richard, G. (2005). Electrical resistivity survey in soil science: A review. Soil and Tillage Research, 83, 173193. Sarrazin, P., Blake, D., Feldman, S., Chipera, S., Vaniman, D., & Bish, D. (2005). Field deployment of a portable X-ray diffraction/X-ray fluorescence instrument on Mars analog terrain. Powder Diffraction, 20, 128133. Schibler, L., Boyko, T., Ferdyn, M., Gajda, B., Ho¨ll, S., & Jordovana, N., et al. (2002). Topsoil magnetic susceptibility mapping: Data reproducibility and compatibility, measurement strategy. Studia Geophysica et Geodaetica, 46, 4357. Schrader, C. D., & Stinner, R. J. (1961). Remote analysis of surfaces by neutron gamma-ray inelastic scattering technique. Journal of Geophysical Research, 66, 19511956. Schwertmann, U., & Taylor, R. M. (1989). Iron oxides. In J. B. Dixon, & S. B. Weed (Eds.), Minerals in soil environments (2nd ed.). Madison, WI: Soil Science Society of America(pp. 379438).. Sethuramasamyraja, B., Adamchuk, V. I., Dobermann, A., Marx, D. B., Jones, D. D., & Meyer, G. E. (2008). Agitated soil measurement method for integrated on-the-go mapping of soil pH, potassium and nitrate contents. Computers and Electronics in Agriculture, 60, 212225. Shibusawa, S., Made Anom, S. W., Sato, H. P., & Sasao, A. (2001). Soil mapping using the real-time soil spectrometer. In G. Gerenier, & S. Blackmore (Eds.), ECPA 2001 (Vol. 2, pp. 485490). Montpellier: AGRO Montpellier.

Proximal Soil Sensing for Measurements in Space and Time

289

Shonk, J. L., Gaultney, L. D., Schulze, D. G., & Van Scoyoc, G. E. (1991). Spectroscopic sensing of soil organic matter content. Transactions of the American Society of Agricultural Engineers, 34, 19781984. Sibley, K. J., Astatkie, T., Brewster, G., Struik, P. C., Adsett, J. F., & Pruski, K. (2009). Field-scale validation of an automated soil nitrate extraction and measurement system. Precision Agriculture, 10, 162174. Sikka, P., Corke, P., Valencia, P., Crossman, C., Swain, D., & Bishop-Hurley, G. (2006). Wireless ad hoc sensor and actuator networks on the farm. In IPSN ’06 proceedings of the 5th international conference on information processing in sensor networks. New York, NY: Association for Computing Machinery. Sirjacobs, D., Hanquet, H., Lebeau, F., & Destain, M. F. (2002). On-line soil mechanical resistance mapping and correlation with soil physical properties for precision agriculture. Soil and Tillage Research, 64, 231242. Stafford, J. V., & Hendrick, J. G. (1988). Dynamic sensing of soil pans. Transactions of the American Society of Agricultural Engineers, 31(1), 913. Stenberg, B., Rogstrand, G., Bolenius, E., & Arvidsson, J. (2007). On-line soil NIR spectroscopy: Identification and treatment of spectra influenced by variable probe distance and residue contamination. In J. V. Stafford (Ed.), Precision agriculture ’07. Papers presented at the 6th European conference on precision agriculture, Skiathos, Greece, June 36, 2007 (pp. 125131). Wageningen: Wageningen Academic Publishers. Stenberg, B, Viscarra Rossel, R. A., Mouazen, A. M., & Wetterlind, J. (2010). Visible and near infrared spectroscopy in soil science. Advances in Agronomy, 107, 163215. Sudduth, K. A., Drummond, S. T., & Kitchen, N. R. (2001). Accuracy issues in electromagnetic induction sensing of soil electrical conductivity for precision agriculture. Computers and Electronics in Agriculture, 31, 239264. Sudduth, K. A., & Hummel, J. W. (1993). Soil organic matter, CEC, and moisture sensing with a portable NIR spectrophotometer. Transactions of the American Society of Agricultural Engineers, 36, 15711582. Sudduth, K. A., Kitchen, N. R., Myers, D. B., & Drummond, S. T. (2010). Mapping depth to argillic soil horizons using apparent electrical conductivity. Journal of Environmental and Engineering Geophysics, 15, 135146. Sudduth, K. A., Kitchen, N. R., Wiebold, W. J., Batchelor, W. D., Bollero, G. A., & Bullock, D. G., et al. (2005). Relating apparent electrical conductivity to soil properties across the north-central USA. Computers and Electronics in Agriculture, 46, 263283. Taylor, J. A., McBratney, A. B., Viscarra Rossel, R. A., Minansy, B., Taylor, H., & Whelan, B., et al. Development of a multi-sensor platform for proximal soil sensing. 18th World congress of soil science. Philadelphia, PA: World Congress of Soil Science. Tekeste, M. Z., Grift, T. E., & Raper, R. L. (2002). Acoustic compaction layer detection. St. Joseph: American Society of Agricultural Engineers. USAEC. (2000). The tri-service site characterization and analysis penetrometer systemSCAPS. Report number SFIM-AEC-ET-TR-99073. Maryland United States Army Environmental Center, Aberdeen Proving Ground. Van Bergeijk, J., Goense, D., & Speelman, L. (2001). Soil tillage resistance as tool to map soil type differences. Journal of Agricultural Engineering Research, 79(4), 371387. van Groenigen, J. W., & Stein, A. (1998). Constrained optimization of spatial sampling using continuous simulated annealing. Journal of Environmental Quality, 27(5), 10781086. Vellidis, G., Tucker, M., Perry, C., Kvien, C., & Bednarz, C. (2008). A real-time wireless smart sensor array for scheduling irrigation. Computers and Electronics in Agriculture, 61, 4450. Viscarra Rossel, R. A., Bui, E. N., de Caritat, P., & McKenzie, N. J. (2010b). Mapping iron oxides and the color of Australian soil using visiblenear-infrared reflectance spectra. Journal of Geophysical Research, 115, F04031.

290

R.A. Viscarra Rossel et al.

Viscarra Rossel, R. A., Cattle, S., Ortega, A., & Fouad, Y. (2009). In situ measurements of soil colour, mineral composition and clay content by visNIR spectroscopy. Geoderma, 150, 253266. Viscarra Rossel, R. A., Fouad, Y., & Walter, C. (2008). Using a digital camera to measure soil organic carbon and iron contents. Biosystems Engineering, 100, 149159. Viscarra Rossel, R. A., Gilbertsson, M., Thyle´n, L., Hansen, O., McVey, S., & McBratney, A. B. (2005). Field measurements of soil pH and lime requirement using an on-the-go soil pH and lime requirement measurement system. In J. V. Stafford (Ed.), Precision Agriculture ’05. Wageninen: Wageninen Academic Publishers. Viscarra Rossel, R. A., & McBratney, A. B. (1997). Preliminary experiments towards the evaluation of a suitable soil sensor for continuous “on-the-go” field pH measurements. In J. V. Stafford (Ed.), Precision agriculture ’97, Vol. II: Technology, IT and management (pp. 493502). Oxford: Bios Scientific Publishers. Viscarra Rossel, R. A., & McBratney, A. B. (1998). Laboratory evaluation of a proximal sensing technique for simultaneous measurement of clay and water content. Geoderma, 85, 1939. Viscarra Rossel, R. A., & McBratney, A. B. (2008). Diffuse reflectance spectroscopy as a tool for digital soil mapping. In A. E. Hartemink, A. B. McBratney, & L. Mendonc¸a-Santos (Eds.), Digital soil mapping with limited data (pp. 165172). Amsterdam: Elsevier Science(Chapter 12). Viscarra Rossel, R. A., McBratney, A. B., & Minasny, B. (Eds.) (2010a). Progress in soil science series. New York, NY: Springer-Verlag. Viscarra Rossel, R. A., McGlynn, R. N., & McBratney, A. B. (2006b). Determining the composition of mineral-organic mixes using UVvisNIR diffuse reflectance spectroscopy. Geoderma, 137, 7082. Viscarra Rossel, R. A., Taylor, H. J., & McBratney, A. B. (2007). Multivariate calibration of hyperspectral γ-ray energy spectra for proximal soil sensing. European Journal of Soil Science, 58, 343352. Viscarra Rossel, R. A., & Walter, C. (2004). Rapid, quantitative and spatial field measurements of soil pH using an ion sensitive field effect transistor. Geoderma, 119, 920. Viscarra Rossel, R. A., Walvoort, D. J. J., McBratney, A. B., Janik, L. J., & Skjemstad, J. O. (2006a). Visible, near-infrared, mid-infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties. Geoderma, 131, 5975. Waiser, T. H., Morgan, C. L. S., Brown, D. J., & Hallmark, C. T. (2007). In situ characterization of soil clay content with visible near-infrared diffuse reflectance spectroscopy. Soil Science Society of America Journal, 71, 389396. Wang, C. H., Willis, D. L., & Loveland, W. D. (1975). Characteristics of ionizing radiation. In C. H. Wang, D. L. Wills, & W. D. Loveland (Eds.), Radiotracer methodology in biological environmental and physics sciences (pp. 3973). Englewood Cliffs, NJ: Prentice Hall. Werban, U., Behrens, T., Cassiani, G., & Dietrich, P. (2010). iSOIL—An EU-level project to integrate geophysics, digital soil mapping and soil science. In R. A. Viscarra Rossel, A. B. McBratney, & B. Minasny (Eds.), Proximal soil sensing (pp. 103110). New York, NY: Springer-Verlag. Whalley, W. R. (1991). The development and evaluation of a microwave soil moisture sensor for incorporation in a narrow tine. Journal of Agricultural Engineering Research, 50, 2532. Whalley, W. R., Dean, T. J., & Izzard, P. (1992). Evaluation of the capacitance technique as a method for dynamically measuring soil water content. Journal of Agricultural Engineering Research, 50(1), 2533. Whiting, M. L., Li, L., & Ustin, S. L. (2004). Predicting water content using Gaussian model on soil spectra. Remote Sensing of Environment, 89, 535552.

Proximal Soil Sensing for Measurements in Space and Time

291

Wielopolski, L., Hendrey, G., Johnsen, K. H., Mitra, S., Prior, S. A., & Rogers, H. H., et al. (2008). Nondestructive system for analyzing carbon in the soil. Soil Science Society of America Journal, 72, 12691277. Williams, B. G., & Baker, G. C. (1982). An electromagnetic induction technique for reconnaissance surveys of soil salinity hazards. Australian Journal of Soil Research, 20, 107118. Williams, B. G., & Hoey, D. (1987). The use of electromagnetic induction to detect the spatial variability of the salt and clay content of soils. Australian Journal of Soil Research, 25, 2127. Wong, M. T. F., Asseng, S., & Zhang, H. (2006). A flexible approach to managing variability in grain yield and nitrate leaching at within-field to farm scales. Precision Agriculture, 7, 405417. Wong, M. T. F., & Harper, R. J. (1999). Use of on-ground gamma-ray spectrometry to measure plant-available potassium and other topsoil attributes. Australian Journal of Soil Research, 37, 267277. Wong, M. T. F., Oliver, Y., & Robertson, M. J. (2009). Gamma-radiometric assessment of soil depth across a landscape not measurable using electromagnetic surveys. Soil Science Society of America Journal, 73, 12611267. Yurui, S., Schulze-Lammers, P., Daokun, M., Jianhui, L., & Qingmeng, Z. (2008). Determining soil physical properties by multi-sensor technique. Sensors and Actuators A: Physical, 147, 352357. Zerger, A., Viscarra Rossel, R. A., Swain, D. L., Wark, T., Handcock, R. N., & Doerr, V. A. J., et al. (2010). Environmental sensor networks for vegetation, animal and soil sciences. International Journal of Applied Earth Observation and Geoinformation, 12(5), 303316. Zreda, M., Desilets, D., Ferre, T. P. A., & Scott, R. L. (2008). Measuring soil moisture content non-invasively at intermediate spatial scale using cosmic-ray neutrons. Geophysical Research Letters, 35, L21402.