Calibration of a New Device to Measure Water Content of Rocks

The vadose zone, which extends from the soil-atmospheric interface to the capillary fringe of the water table, is a fundamental part of the hydrologic cycle. It controls how the precipitation splits into infiltration, surface runoff, evapotranspiration, groundwater recharge; it regulates the storage, transfer, filtering, adsorption, retard and attenuation of solutes and contaminants before these reach the groundwater. Until the last decades, the vadose zone was considered as a black box that merely connect surface water and groundwater. As consequence an incomplete understanding of the complex dynamics of the vadose zone existed. Nowadays, instead, the monitoring of the hydrological processes that occur through the vadose zone, are receiving increased attention mainly with refer to contaminant transport processes. Fluxes of water and solutes strongly depend on water content thus its monitoring and estimation becomes an important issue. A wide range of methods, sensors and technologies are available for the measure of soil water content, mainly used in management of precision farming. The traditional standard method for direct measurement of soil water content is the gravi- metric method [1]. This method implies the sampling of the soil to carry out laboratory measurements, modifying the natural condition. As gravimetric method is destructive, labour intensive, not timely and costly, many alternative non-destructive methods for measuring water content have been developed. However, none of all methods measure the water content directly but only some properties, named "surrogate measure", that changes as the soil water content changes. By measuring the value of the surrogate parameter, it is possible to estimated the value of water content, using a calibration curve that represents the relationship between the surrogate measure and the soil water content values [2]. Neutron Thermalization method [3, 4], Capacitance method [5-8], Time Domain Reflectometry (TDR) method [9-13], Frequency Domain Reflectometry (FDR) method [14, 15], Impedance method [16, 17], Electrical Restistance method [18, 19] and Tensiometer method [20-22] are among the well-known technologies utilized to develop different kind of devices and probes for measuring moisture content and soil water potential, respectively. More details on these technologies and devices, advantages and limitations of each one, are given in several references [23, 24, 2]. Recently, remote sensing technologies, like passive or active radiometry, have been applied in order to provide wide-area indications of surface soil water content [25, 26]. However, the great influence of vegetation and surface structure on the quality of received signal, restricts the sampling depth to the uppermost part (2-5 cm) of soil [27], strongly limiting the usefulness of this technology in hydrological application. The use of all these methods and devices in the soil is customary by now but when the vadose zone consists of rocks the monitoring of water content become more difficult for several aspects. The main difficulty regards the installation of the probes that are often very delicate and cannot be hammered or screwed into the rock. Furthermore, after their insertion, a good contact between the rock and the sensor must be ensured in order to minimize the gap effect which causes significant errors in the investigated properties, especially for dielectric ones [28, 29]. At present, very few applications of these techniques to rocks, by creating new devices or by adapting the existing ones, are documented in literature. The first measurements in sandstone and tuff blocks by means of penetration type probes using TDR are reported in [30]. Topp's equation [9] was used to convert the dielectric constant (K) values