GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 1.3 225 as electrical sources that are heterogeneously distributed in the whole volume responsible of electric potential generated. In particular the electric signal derived from considering only the streaming potential originated from groundwater flow has been firstly studied and compared with experimental data recorded at Soultz. After the effect of temperature has been taken into account to complete our study of SP in that area and we underline the effect that bring major contribution, within sources considered. In the end the obtained distribution of electric potential has been compared with the density of seismic events recorded at Soultz-sous-Forets for the same sequence of injection. The self potential method. In recent years, the self-potential (SP) method has been applied to study the dynamics of fluid flow within natural water and/or hydrocarbon reservoirs. This technique was first applied as mapping tool of water flow in aquifers (e.g. Ogilvy et al. , 1969; Bogoslovsky and Ogilvy, 1970, 1973) and in geothermal areas (e.g. Corwin and Hoover, 1979). The SP method was later used as a monitoring tool of fluid flow within geothermal reservoirs (Ishido et al. , 1983; Kawakami and Takasugi, 1994; Ushijima et al. , 1999; Murakami et al. , 2001) and more recently within aquifers (Perrier et al. , 1998; Pinettes et al. , 2001). Moreover, several theoretical and laboratory studies (Ishido and Mizutani, 1981; Morgan et al. , 1989; Sprunt et al. , 1994; Jouniaux and Pozzi, 1995; Bernabe, 1998; Revil et al. , 1999a, 1999b; Marino et al. , 1999; Lorne et al. , 1999a, 1999b; Reppert, 2000) were conducted in order to better understand electrokinetic phenomena in rocks for various chemical and thermal conditions. Numerical modeling schemes were also developed (Sill, 1983; Wurmstich and Morgan, 1994; Ishido and Pritchett, 1999) in order to combine all these studies and better interpret SP anomalies induced by fluid flow in reservoirs. The main contributions to the self- potential signals are (1) the streaming potential related to groundwater flow, (2) the diffusion potential related to gradients of the chemical potential of ionic species, (3) the thermoelectric effect related to the influence of the temperature upon the chemical potential of charge carriers, and (4) the electro-redox effect associated with bodies and contaminant plumes that are rich in organic matter. A general formulation of this problem for a porous material has been developed recently by Révil et al. (2002). Considering a porous media composed by grains of mineral and pores filled by an electrolyte in chemical equilibrium with grains, the chemical interaction between grains and electrolyte originates an electrical double layer. In the case of the majority of minerals composed by grains with beam > 0.1 μm the thickness of this layer can be assumed smaller than the rays of pores and grains (electrical double layer plane and fine). In addition, the fluid flow through a porous media can be considered as laminar. The constitutive relationship of electrical current density and velocity of the fluid are (Révil et al., 2002): (1) (2) V is the electric potential [V], P is the fluid pressure [Pa], g is the gravity acceleration [m/s 2 ] and ρ f is volumetric mass of the fluid . The terms l ij , inside Eqs. (1) and (2), represent the flux coupling coefficients, generally considered as constant to account the linearity experimentally observed between fluid flux and its generating forces. The coefficient l 12 and l 21 can be assumed as equals (Révil et al. , 2002) due to the reciprocity law of Onsangar (De Groot and Mazur, 1962). l is the electrokinetic coupling coefficient, determining both the intensity of fluid mass generated by a gradient of electrical potential (electroosmosis) and the electrical flux induced by a pressure gradient (electrokinetism). In the case of homogeneous isotropic porous media the l coefficient is: (3)