Summary Geothermal energy production emits significant amounts of hydrogen sulfide (H2S). A strategy to mitigate the emissions is to reinject the H2S into basaltic formations, where it reacts with the rock to form pyrite. Due to the polarization properties of pyrite, the direct current resistivity (DC) and induced polarization (IP) geophysical method (i.e., DCIP) has shown the potential for monitoring H2S mineral storage. However, field applications of DCIP monitoring have been limited by the low spatial coverage of wireline logging and by the ambiguity in interpreting IP signals due to multiple processes that contribute to the polarization response. This study integrates DCIP with field-scale reactive transport modeling, utilizing both synthetic modeling and field investigations, to assess the ability of surface and cross-hole DCIP to monitor H2S mineral storage at the Nesjavellir study site in Iceland. Two surface DCIP datasets were collected at Nesjavellir, with six months of continuous H2S injection between them. Time-lapse inversions, performed using a novel gridding scheme that accounts for electrode misplacement between the two DCIP surveys, recover no significant IP changes beyond data noise. Interpreting these results alongside the reactive transport results finds that pyrite mineralization during the six-month injection period is too small and too deep to be resolved by surface DCIP time-lapse surveying, highlighting the benefit of a joint geophysical-geochemical interpretation approach. Conversely, joint DCIP-reactive transport synthetic modeling shows the potential of cross-hole DCIP for monitoring long-term H2S mineral storage, provided that data noise is low and sufficient H2S is injected. The reactive transport models also provide insight into the mechanisms contributing to the polarization response, demonstrating that pyrite mineralization is the primary contributor to the polarization response, with minimal contribution from other minerals such as smectites and iron oxides. However, existing petrophysical relationships are simplistic, which adds uncertainty to the interpretation of the DCIP signal and the quantification of pyrite mineralization. Additionally, smectite formation has been shown to decrease both the polarization signals and the quality of the IP data due to its electrically conductive properties. At Nesjavellir, a decrease in the DC resistivity from 1925 to 325 Ωm is observed, attributed to the disposal of warm wastewater. This decrease is identified by comparing resistivity data collected in this study to historical vertical electrical sounding data collected prior to geothermal development. Lastly, petrophysical relationships linking DC resistivity, smectite content, and permeability suggest a high basalt fracture permeability of 7.9× 10−12 m2, which agrees with values recovered through flow model calibrations. This result demonstrates the value of geophysical surveying not only for monitoring but also for constraining key parameters in reactive transport simulations.
Ciraula et al. (Tue,) studied this question.