Advances in Agronomy

1. Auflage 2015
ISBN: 978-0-12-802345-7
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, Band Volume 131, 354 Seiten

Reihe: Advances in Agronomy

Advances in Agronomy continues to be recognized as a leading reference and a first-rate source for the latest research in agronomy. Each volume contains an eclectic group of reviews by leading scientists throughout the world. Five volumes are published yearly which ensures that authors' contributions are disseminated to the readership in a timely manner. As always, the subjects covered are varied and exemplary of the myriad of subject matter dealt with by this long-running serial. - Timely and state-of-the-art reviews - Distinguished, well recognized authors - A venerable and iconic review series - Timely publication of submitted reviews

Autoren/Hrsg.

Leseproben

2. Technologies and Applications: Sensing of Soil- and Vegetation-Specific Properties
2.1. Proximal Soil Sensing
Proximal soil sensing has been defined 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 et al., 2011). Proximal soil sensors have been extensively used in DSM to cost-efficiently augment the number of observation sites in order to better characterize the spatial variation of physical and chemical soil properties in an area. They make use of a large set of technologies, from mechanical to electromagnetic, with modern hardware and software, and, by definition, need to be field portable (although many sensors are used in the laboratory, and many sensors are not hand-portable needing to be mounted on a vehicle). It is also assumed that they rapidly measure soil properties with minimum use of resources. Applications of proximal soil sensors in soil and related sciences include soil mapping and monitoring, precision agriculture and farm decision making, environmental pollution assessment, among others. In this review we focus on electromagnetic sensors, but present literature on other sensors, mostly mechanical, when we discuss sensor fusion. 2.1.1. Electrical Conductivity Sensors Electrical conductivity (EC) is the ability of a material to conduct electrical current. Soil is a three-phase system composed of water, air, and organic and mineral solids that offer multiple pathways with different EC and an undefined boundary for electrical current to flow. In this case, it is feasible to measure the ECa. There are two types of field sensors that measure ECa, an electrode-based and an electromagnetic induction-based. In the former, four electrodes placed in a linear array are inserted in the soil using probes or coulters. The sensor measures the potential difference that arises between the two internal electrodes when electrical current is applied to the two external electrodes, and calculates the electrical resistivity of the soil. Soil ECa is calculated as the inverse of electrical resistivity. The depth of reading depends on the spacing between internal and external electrodes. In the latter system, a transmitter coil generates time-varying primary magnetic fields that induce electrical current in the soil, which in turn creates a secondary magnetic field. A receiver coil measures the magnitude of the primary and secondary magnetic fields to calculate the ECa of the soil. The depth of reading depends on the current frequency, distance between transmitter and receiver coils, and dipole mode (horizontal or vertical). In soils, the EC is affected, among other things, by the concentration and ensemble of ions and the amount of water in the soil. The higher the ion concentration and moisture, the higher the EC. Thus, EC is directly correlated with and has been used to estimate soil salinity (Li et al., 2013; Akramkhanov et al., 2014), cation exchange capacity (CEC; Bronson et al., 2005; Sudduth et al., 2005; Kweon et al., 2013), ion concentrations (Bronson et al., 2005), and soil moisture (Sudduth et al., 2005; Hossain et al., 2010). Soil EC has been also used to estimate a number of indirectly correlated soil physical and chemical properties such as particle size distribution (Bronson et al., 2005; Sudduth et al., 2005; Heil and Schmidhalter, 2012), topsoil depth (Sudduth et al., 2013), depth to clay layer (Vitharana et al., 2008b; Saey et al., 2011), compaction (Al-Gaadi, 2012), water holding capacity (Abdu et al., 2008), and organic matter/carbon (Sudduth et al., 2005; Kweon et al., 2013). A review on this topic is presented by Corwin and Lesch (2005), whereas Friedman (2005) discusses soil properties that influence ECa. They are part of a special volume of Computers and Electronics in Agriculture on “Applications of Apparent Soil Electrical Conductivity in Precision Agriculture.” In a different direction, some studies were interested in estimating soil ECa based on other soil properties for diverse purposes. To assess vertical soil textural distribution, Saey et al. (2009) predicted ECa based on clay content alone (R2 = 0.81), or clay content plus organic matter and land use (R2 = 0.86). In another study, Ekwue and Bartholomew (2011) assessed the influence of different soil properties on ECa. They predicted ECa as a linear function of bulk density, and clay, peat and water contents, obtaining an R2 of 0.89. Finally, Peralta and Costa (2013) proposed the use of ECa to delineate management zones. They predicted ECa as a function of principal components derived from laboratory soil data (pH, CEC, organic matter, EC, and nutrients), also obtaining good results (R2 > 0.90). 2.1.2. Ground Penetrating Radar and Reflectometers Ground penetrating radar (GPR), time-domain reflectometer (TDR), and frequency-domain reflectometer (FDR) measure the relative permittivity, or dielectric constant, of the soil, which describes how easily an electric field is generated in the soil when electric current is applied to it. They follow the same principle of the ECa meter, but operate based on electric field (flux) instead of electric current. Both reflectometers require contact with the soil using probes, which emit and receive electric pulses that are processed by a control unit. Time-domain reflectometer measures the time it takes for the emitted electric pulse to interact with the soil and then come back. Frequency-domain reflectometer measures the frequency differences between the emitted electric pulse and the received pulse after interaction with the soil. They have the advantage of measuring both ECa and volumetric water content with the same probes and in the same sampling volume (Wraith et al., 2005). The GPR does not require probing the soil, but instead it emits and receives a microwave (radar) pulse that interacts with the soil and is modified (attenuated) depending on the EC and dieletric constant of the soil, where the depth of reading is inversely proportional to the EC. The operation of the GPR requires dragging it across an area of interest to produce a continuous cross-sectional profile of the soil plotting the distance travelled versus depth. This is called a radargram. Permittivity meters exploit the large difference in dielectric constant between air (~1), water (~80) and solid constituents (~4–8) in the soil (Wraith et al., 2005) to estimate volumetric water content (TDR—Bittelli et al., 2008; Inoue et al., 2008; Arsoy et al., 2013; FDR—Böhme et al., 2013; GPR—Lunt et al., 2005; Weihermüller et al., 2007; Ardekani, 2013), obtaining RMSE from 0.01 to 0.1 in volumetric content units (volume volume-1). In addition, they have been used to estimate other soil properties, including ECa (FDR—(Wilczek et al., 2012), nitrate (TDR—Wraith and Das, 1998), pore water salinity (FDR—(Wilczek et al., 2012), and bulk density (FDR—Al-Asadi and Mouazen, 2014). 2.1.3. VIS-NIR-MIR Diffuse Reflectance Sensors Visible/near-infrared/mid-infrared diffuse reflectance spectroradiometers measure the reflectance of the soil, i.e., the proportion of the incoming (incident) light that is reflected by the soil. Incident energy (light), usually emitted from an artificial light source, interacts with the soil, and the reflected energy (light) is measured and processed to generate a continuous spectral curve by plotting the reflectance against the wavelength or its inverse, the wavenumber. The reflectance of the soil depends on how its constituents interact with (i.e., absorb, transmit or reflect) the received energy, and is affected by the soil mineral and organic matter ensembles, particle size distribution, and moisture, among other things. The most common spectral ranges, i.e., wavelength intervals, sensed by reflectance sensors include the visible (VIS; ~400–700 nm), near-infrared (NIR; ~700–2500 nm), and mid-infrared (MIR; ~2500–25,000 nm). Ultraviolet (~200–400 nm) and far-infrared (~25,000–350,000 nm) sensors are less common in soil science. Simpler spectrometers are available that operate on the same measurement principles, but aggregate the response across the sensed spectral range (depending on the instrument) into a single number instead of a spectral curve. For example, this is the case of colorimeters and photosynthetically active radiation meters. The most common reflectance sensors applied in soil science include an energy (light) source, a detection apparatus composed of fiber optic cables, mirrors and sensors for specific spectral ranges, and a processing unit. They can be field portable, handheld, or mounted in the laboratory. Depending on the technology employed, they can require much, little or no sample preparation, including drying, grinding, and removing or adding materials to the soil. Different from other sensors that output a single value, VIS-NIR-MIR reflectance sensors generate a curve storing hundreds to thousands of reflectance readings that can be used as...