Injection-induced surface deformation and seismicity at the Hellisheidi geothermal field, Iceland

• Published in: Journal of Volcanology and Geothermal Research
• Main Authors: Juncu, D., Arnadottir, Th, Geirsson, H., Gudmundsson, G. B., Lund, B., Gunnarsson, G., Hooper, A., Hreinsdottir, S., Michalczewska, K.
• Other Authors: Nordic Volcanological Center, Institute of Earth Sciences, University of Iceland, Reykjavik, Iceland, Icelandic Meteorological Office, Reykjavik, Iceland, Department of Earth Sciences, Uppsala University, Uppsala, Sweden, OR–Reykjavik Energy, Reykjavik, Iceland, COMET, School of Earth and Environment, University of Leeds, Leeds, United Kingdom, GNS Science, Lower Hutt, New Zealand
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: Elsevier BV 2020
• Subjects: Man-made deformation; Induced seismicity; Fluid injection; Geodesy; Geothermal energy; Iceland
• Online Access: http://hdl.handle.net/10754/678693; https://doi.org/10.1016/j.jvolgeores.2018.03.019

Induced seismicity is often associated with fluid injection but only rarely linked to surface deformation. At the Hellisheidi geothermal power plant in south-west Iceland we observe up to 2 cm of surface displacements during 2011–2012, indicating expansion of the crust. The displacements occurred at the same time as a strong increase in seismicity was detected and coincide with the initial phase of geothermal wastewater reinjection at Hellisheidi. Reinjection started on September 1, 2011 with a flow rate of around 500 kg/s. Micro-seismicity increased immediately in the area north of the injection sites, with the largest seismic events in the sequence being two M4 earthquakes on October 15, 2011. Semi-continuous GPS sites installed on October 15 and 17, and on November 2, 2011 reveal a transient signal which indicates that most of the deformation occurred in the first months after the start of the injection. The surface deformation is evident in ascending TerraSAR-X data covering June 2011 to May 2012 as well. We use an inverse modeling approach and simulate both the InSAR and GPS data to find the most plausible cause of the deformation signal, investigating how surface deformation, seismicity and fluid injection may be connected to each other. We argue that fluid injection caused an increase in pore pressure which resulted in increased seismicity and fault slip. Both pore pressure increase and fault slip contribute to the surface deformation. The authors would like to thank: Einar Gunnlaugsson from Reykjavik Energy (RE) for providing injection data from Hellisheidi; Bjarni Reyr Kristjánsson (RE) for discussions and advice regarding the geothermal power plant operations and tracer tests; Sarah Minson (U.S. Geological Service) for help with CATMIP and letting us use her code; Elías Rafn Heimisson (Stanford University, USA) for advice regarding Coulomb failure stress calculations; David Bekaert (University of Leeds, UK) for helpful advice on the InSAR processing; Sigurjón Jónsson (King Abdullah University of ...