Electrical resistivity tomography and time-domain induced polarization field investigations of geothermal areas at Krafla, Iceland: comparison to borehole and laboratory frequency-domain electrical observations

Interaction of H2S and basaltic rocks in volcanic geothermal areas can originate from natural up-flow of magmatic fluids or H2S artificial re-injection in relation to geothermal exploitation, both causing pyrite mineralization. We study the possibility to track these processes with electrical impedance field measurements. Electrical Resistivity Tomography (ERT) and TimeDomain Induced Polarization (TDIP) measurements were performed along thirteen 1.24 km long profiles, at three different sites around the eastern caldera rim of the Krafla caldera: (i) a 'cold altered' site affected by past hydrothermal circulations, (ii) a hot active site and (iii) a 'cold un-altered' site, unaffected by hydrothermal circulations. We present 2-D inversions of direct current (DC) resistivity, maximum phase angle of the electrical impedance (MPA) and relaxation time. The maximum depth of investigation for the MPA is 200 m, obtained in zones of high resistivity, corresponding to fresh and recent unaltered basalt. At the hot and cold altered sites, the field resistivities are compared to in situ borehole logs and laboratory complex resistivity measurements on rock samples from the boreholes. The laboratory complex resistivity was measured at six different pore water conductivities, ranging from 0.02 to 5 Sm-1, and frequency in the range 10(-2)-10(6) Hz. The time-range investigated in our field TDIP measurements was approximately 0.01-8 s. At the cold altered site, the inverted resistivity is consistent with both borehole observations and laboratory measurements. At the hot site, resistivity from field inversion and borehole logs are consistent. Comparing inversion results and borehole logs to laboratory resistivity measured on core samples at room temperature reveals that a correction coefficient for the effect of temperature on resistivity of 6 per cent per degrees C is appropriate at investigated depths. This exceptionally high temperature correction coefficient suggests a dominant influence of interface and interfoliar conduction, characteristic of smectite-rich rocks, compared to electrolyte conduction. High MPA is attributed to the presence of pyrite at the hot site and of iron-oxides at the cold unaltered site, through joint consideration of MPA together with DC resistivity and relaxation time. TDIP measurements offer the possibility to detect the presence of metallic minerals at shallow depth and distinguish between pyrite and iron-oxides. The abundance of highly conductive smectite in altered volcanic rocks represents a challenge for resolving IP parameters, because the low resistivity created by abundant smectite limits the data quality of the measured voltage discharge.