GNGTS 2014 - Atti del 33° Convegno Nazionale

after the samples were weighted and water content was estimated. A linear regression equation with R- square = 0.73 and RMSE = 2.23 % was used to calibrate the TDR readings. Ground penetrating radar (GPR) and electrical resistivity tomography (ERT) provide observations at the spatial scale and resolution convenient to gain information about the hydrological processes of the topsoil. The use of these techniques has become ever more widespread for their ability to measure indirectly changes in soil moisture content and obtain images of soil water distribution during irrigation (Butnor et al. , 2001, 2003; Hagrey et al. , 2002, 2007; Huisman et al. , 2003; Michot et al., 2003; Ursino et al., 2014) Geophysical surveys were performed at the soil surface along transects longitudinal and parallel with the bean rows with the aim to characterize the soil profile in correspondence of crop roots locations. In particular, electrical resistivity survey permits to determine the soil resistivity distribution, which provides an indirect measurement of various geological parameters such as the solid and fluid content, porosity and water content. Electrical resistivity tomography (ERT) is an active sensing technique based on the injection of an electrical current I in the subsoil by means of a pair of electrodes (A and B) and the subsequent measurement of the electrical potential ΔV by using another pair of electrodes (M and N). There are several possible electrode arrangements as Wenner, Schlumberger and dipole- dipole arrays. Whatever the used array, the ERT is based on the measure of the soil electrical potential while the current is injected. In this circumstance, it is possible calculate the apparent resistivity ρa = k (ΔV/I),where ρa is expressed in Ωm and k is a geometrical factor depending on the adopted array configuration. The apparent resistivity value is not the true resistivity of the soil, but an “apparent” value that is the resistivity of a homogeneous soil which will give the same resistance value for the same electrode arrangement. The apparent resistivity provides, in fact, a first preliminary image of the electrical subsurface structure denominated as the ‘pseudo-section’ (Loke, 1999). The estimation of the true resistivity required the adoption of the inversion algorithm, RES2DINV, proposed by Loke (1999) and Loke and Barker (1996). For the surveys at hand, a Wenner-Schlumberger array survey were performed with the resistivity meter Syscal R1 (Iris Instruments, Orleans, France) equipped with 48 electrodes, aligned on the soil surface along a line parallel to the bean rows, with an electrode spacing of 0.15 m and a total length of 7.0 m. Ground penetrating radar (GPR) is a powerful tool for subsurface remote sensing and in particular is a noninvasive geophysical technique that provides images space-time called radargram, of the subsoil. The ground-penetrating radar (GPR) is an active electromagnetic technique whose working principle is that of a common radar system. In particular, a transmitting antenna radiates an electromagnetic wave, usually a microwave pulsed signal The radar produces a short pulse of electromagnetic wave whose spectrum varies from 10 MHz to some GHz, it is transmitted into the soil generating a wave front which propagates downward through the sub-surface materials at the velocity determined by the soil dielectric permittivity. When this propagating wave encounters any change in the bulk electrical properties of different subsurface lithologies, mineralogy, and/or the character of the sediment interface, some of the energy is reflected back to the surface (Davis and Annan, 1989) The amount of energy that is reflected by an sediment interface is approximately proportional to the difference in dielectric constants of the two layers and these contrasts produce strong and identifiable reflections on radar records. The dielectric permittivity of soil materials is principally dependent upon moisture content (Annan et al. , 1991). The GPR surveys in this experiment, were performed with an GSSI SIR 2000 acquisition unit equipped with a mono-static equipment, and two 400 MHz and 1500 MHz central frequency antenna was dragged along the same survey lines of the electrical resistivity survey; GPR data were post-processed with Reflex software (Sandmeier Scientific Software, Karlruhe, Germany). In order to display radargrams as a function of depth, the change of the time sections into depth was performed through a hyperbola approximation GNGTS 2014 S essione 3.3 229