GNGTS 2015 - Atti del 34° Convegno Nazionale

150 GNGTS 2015 S essione 3.3 a 2D geometry is assumed to approximate a limited strike length slab (Wannamaker et al. , 1984,1997; Livelybrooks et al. , 1996). We propose now a 3D refinement of the 2D section in Fig. 2, using the same resistivity sequence and Cole-Cole parameters as in the 2D model and assigning after trial-and-error a final strike length of 15 km to the horizontal slab, shorter than the width of the infinite length slab in the original 2D model. Fig. 3 shows the TM and TE original and synthetic apparent resistivity pseudosections and the misfit between them. The misfit has been plotted by assigning at each point of the pseudosections the modulus of the discrepancy index r i , given by (Troiano et al ., 2014) r i =|(d i −m i )/ε i | , (2) where d i are the observed data, m i are model responses and ∑ i are the data errors with i=1,2,...,M , M being the total number of measured data. The normalized root mean square ( rms ) misfit has been calculated using the formula (Gabàs and Marcuello, 2003) rms=√(∑ Μ r i 2 /M). (3) Average rms values have been obtained equal to 2.6 for the TM section and 3.1 for the TE section. Following a previously suggested interpretation paradigm (Mauriello et al. , 1996, 2004), m and τ seem to play an important role mostly in the evaluation of IP effects in volcano-geothermal areas. In fact, m can be associated to the degree of alteration and mineral particle deposition Fig. 2 – At the top the NW-SE section of the interpreted Snake River Plain structural model across the MT profile by Stanley (1982), The 3D conductive slab (15 kmwide in the direction normal to the section) is assumed to be affected by resistivity frequency dispersion. At the bottom the Snake River Plain MT TM and TE field pseudosections with the MT stations (redrawn after Stanley, 1982). The color scale is in Ω m.