Imaging and Monitoring the Reykjanes Supercritical Geothermal Reservoir in Iceland with Time-Lapse CSEM and MT Measurements
M. DARNET, P. WAWRZYNIAK, N. COPPO, S. NIELSSON, A. SCHILL, G.O. FRIDLEIFSSON
[Bureau des Recherches Géologiques et Minières, France]
Surface geophysical monitoring techniques are important tools for geothermal reservoir management as they provide unique information on the reservoir development away from boreholes. For magmatic environments, electromagnetic (EM) methods are attractive monitoring tools as they allow to characterize the reservoir and hence potentially monitor changes related to fluid injection/production. Indeed, the electrical resistivity of reservoir rocks is highly dependent on the volume, chemistry and phase of the in-situ geothermal brine (e.g. liquid, vapor, supercritical). Passive EM techniques (e.g. magnetotellurics or MT) are traditionally used for geothermal exploration and a few recent studies have demonstrated its potential for monitoring reservoir development. One of the main challenges is though the presence of cultural noise and/or variability of the Earth magnetic field that can obfuscate the EM signals of interest. We have investigated the benefits and drawbacks of active EM surveying (Controlled-Source EM or CSEM) to tackle this challenge, first with a synthetic study and subsequently with an actual time-lapse survey acquired in 2016 and 2017 over the Reykjanes geothermal field in Iceland before (baseline) and after (monitor) the thermal stimulation of the supercritical RN-15/IDDP-2 geothermal well. The synthetic study showed that for geothermal fields having a resistivity structure similar to the Reykjanes field (i.e. a conductive caprock overlying a more resistive higher temperature reservoir), CSEM and MT measurements can detect resistivity changes within the deep resistive reservoir, provided that measurement errors are small. Variations in many survey parameters (e.g. errors in receiver position/orientation, differences in recording devices, variations of near surface conditions, external noise) can create significant time-lapse CSEM measurement errors. Our actual time-lapse survey showed that when similar CSEM equipment is used during the baseline and monitor surveys and systematically d-GPS positioned, the remaining key parameter controlling the survey repeatability is the level of external noise. Since the influence of external noise on CSEM data can be artificially reduced (e.g. by increasing the transmitter dipolar moment), it offers the possibility to adapt the survey design to increase the chance of detecting the time-lapse signals of interest. On the contrary, little control is possible on the MT signal to noise ratio and hence repeatability. The time-lapse EM survey acquired over the Reykjanes geothermal reservoir showed indeed that a high CSEM survey repeatability can be achieved with electric field measurements (within a few percent) but that time-lapse MT survey is a challenging task because of the high level of cultural noise in this industrialized environment. To assess the quality of our CSEM dataset, we inverted the data and confronted the resulting resistivity model with the resistivity logged in the RN-15/IDDP-2 well. We obtained a good match up to 2-3km depth, i.e. enough to image the caprock and the liquid-dominated reservoir but not deep enough to image the reservoir in supercritical conditions. To obtain such an image, we had to jointly invert legacy MT data with our CSEM data. On the monitoring aspects, the analysis of changes in electric fields did not allow to identify any CSEM signal related to the thermal stimulation of the RN-15/IDDP-2 well. One possible explanation is the weakness of the time-lapse CSEM signal compared the achieved CSEM survey repeatability as a result of a limited resistivity change over a limited volume within the reservoir. Future reservoir developments in the supercritical reservoir (e.g. hydraulic stimulation, long-term fluid circulation) will most likely generate stronger resistivity variations over a larger volume than during the thermal stimulation of the well. This calibration study provides the basic information for deciding when and how an EM monitor survey must be performed for the monitoring of the Reykjanes geothermal reservoir but also for the definition of the monitoring strategy of similar high-enthalpy geothermal reservoirs. This study was part of the DEEPEGS project, which received funding from the European Union HORIZON 2020 research and innovation program under grant agreement No 690771.
|        Topic: Geophysics||Paper Number: 13046|