"Noise" during long-term continuous magnetotelluric monitoring of RN-15/IDDP-2 well engineering (Reykjanes peninsular, Iceland): A geogenic origin?. (November 2021)

• Record Type: Journal Article
• Authors: Haaf, N.; Schill, E.
• Is Part Of: Geothermics. Volume 96(2021)
• Publication Date: 2021-11
• Subjects: EGS -- Magnetotelluric -- Stimulation -- Monitoring -- Reykjanes; Hydrogeology -- Periodicals; Geothermal resources -- Periodicals; Énergie géothermique -- Périodiques; GEOTHERMAL ENGINEERING; GEOTHERMAL ENERGY; GEOTHERMAL EXPLORATION; Geothermal resources; Hydrogeology; Periodicals; Electronic journals; 621.44
• DOI: 10.1016/j.geothermics.2021.102192
• Languages: English
• ISSNs: 0375-6505

Abstract

Highlights: New Data set of continuous magnetotelluric monitoring over two months and results from seismic monitoring and the related hydraulic data of the RN15/IDDP-2 well in Reykjanes, Iceland. Changes in the electromagnetic field in the subsurface induced by measures during reservoir engineering or even natural seismicity are observed. Results show a temporal relation between low resistivity anomalies (considered as induced noise) and the geomagnetic field activity, the fluid losses up to 60 L/s, and mechanic processes occurring in the reservoir prior to clusters of induced seismicity. Abstract: To date, magnetotelluric monitoring of processes during reservoir engineering of geothermal systems have been carried out only at three sites world-wide. Here, we add a new survey at the Reykjanes peninsular (Iceland). The MT data acquisition at RN-15/IDDP-2 covered the last third of the drilling period and thermo-hydraulic stimulation. Drilling was accompanied by temporal total fluid losses of up to 60 L/s as well as randomly distributed induced seismicity. Our experimental results of a two-months magnetotelluric monitoring during the deepening of the RN-15/IDDP-2 well on the Reykjanes peninsular (Iceland) to 4'659 m are in line with earlier observations on decreasing resistivities at periods of a few s to about 20 s (or up to about 40 s) in conjunction with fluid injection. Simple models indicate however that it is not the fluid volume itself that causes the anomaly. Moreover, here, temporal decreases in the electric resistivities occur at 0.2-20 s with minima at periods of about 0.4 s 1-2 days ahead of clusters of seismic events with magnitudes up to ML < 2. Unlike in the previous studies, where resistivity decreases occur on the component that is parallel to Shmin, here they in both components (XY and YX) of the MT data. At the Reykjanes peninsular, Shmin is oriented N120°E off the XY component (N0°E). Therefore, if we rotate our data in Shmin direction, the difference between both components would be slightly smaller than that for the non-rotated ones. The decrease in resistivity (on both components) extends over > 1 order of magnitudes over a short period range and is thus considered noise induced in the subsurface. This study aims at adding an essential dataset to the general discussion on MT monitoring of reservoir processes. The results show a temporal relation between decreasing apparent resistivity and (i) the geomagnetic field activity, (ii) the fluid losses up to 60 L/s, as well as (iii) mechanic processes occurring in the reservoir before clusters of induced seismicity on the other hand.