INDUCED SEISMICITY

Assessing whether the 2017 Mw 5.4Pohang earthquake in South Koreawas an induced eventKwang-Hee Kim,1* Jin-Han Ree,2* YoungHee Kim,3 Sungshil Kim,2

Su Young Kang,1 Wooseok Seo1

Themoment magnitude (Mw) 5.4 Pohang earthquake, themost damaging event in South Koreasince instrumental seismic observation began in 1905, occurred beneath the Pohanggeothermal power plant in 2017. Geological and geophysical data suggest that the Pohangearthquake was induced by fluid from an enhanced geothermal system (EGS) site, which wasinjected directly into a near-critically stressed subsurface fault zone.The magnitude of themainshock makes it the largest known induced earthquake at an EGS site.

The injection of fluid into reservoir rocks,which facilitates oil and gas recovery, en-hances geothermal systems, and aids in thedisposal of wastewater and CO2 gas, alsohas a small chance of inducing earthquakes

[e.g., (1–4)]. Empirical and theoretical relationshipsexist to connect the maximum magnitude of aninduced earthquake and the injected fluid vol-ume (2, 5). The magnitude of an induced earth-quake may be tectonically controlled—for example,owing to the presence of a fault suitably orientedfor slip under a given stress field that is also loc-ated adjacent to one or more injection or pro-duction wells (6, 7). The earthquake nucleationitself might be controlled by the injection (6).Previously observed magnitudes of induced seis-micity at enhanced geothermal system (EGS)sites have been relatively small (8); the largestlocal magnitude (ML) reported was 3.4, in Basel,Switzerland (9), although amuch larger, possiblyinduced earthquake has been reported in theCerro Prieto geothermal field, Mexico, a tectoni-cally active area (10).A moment magnitude (Mw) 5.4 earthquake

occurred at the Pohang EGS site in southeasternKorea on 15 November 2017. The earthquakewas the most damaging and the second-largestin magnitude in South Korea since the firstseismograph was installed in 1905. The earth-quake injured 90 people, and the estimated prop-erty damage was US$52 million (11). We presentevidence that the Pohang earthquake was thelargest induced event to have occurred at anyEGS site worldwide. Moreover, this event in-dicates that injected fluid volumes much smallerthan predicted by theory can trigger a relativelylarge earthquake, at least under the right set ofconditions.

The Korean Peninsula lies within the EurasianPlate (Fig. 1), although it was composed of con-tinental magmatic arcs at a plate boundary untilthe early Tertiary period (~30 million years ago)(12). North-northeast (NNE)–striking strike-slipfaults and NNE- to NE-striking normal faultsdeveloped predominantly in southeastern Koreaand adjacent offshore areas when the East Sea(or Japan Sea) opened as a back-arc basin in theearly to middle Tertiary (~30 to 15 million yearsago), with the coetaneous formation of smaller-

scale basins, including the Pohang (12–14). Someof these faults have been reactivated as strike-slipand thrust faults in the current compressionalregime (13–15). The axes of compression deter-mined by focal mechanism solutions indicateshallow plunges to the ENE throughout thesouthern Korean Peninsula (15, 16).The Pohang basin consists of nonmarine to

deep marine sedimentary strata dating to theMiocene (~20 million years ago), with a base-ment composed of Cretaceous to Eocene sedi-mentary and volcanic rocks and late Paleozoic to

Eocene granitoids (17–19) (Fig. 2A). The geologyof the Pohang EGS site comprises (from topto bottom) Quaternary alluvia (<10 m thick),Miocene semi-consolidated mudstone (200 to400 m thick), Cretaceous to Eocene sedimentaryand igneous rocks (~1000 m thick), and Permiangranodiorite with gabbroic dykes (19–21) (Fig. 2B).One vertical injection well (4346m deep; PX2)

and another deviated production well (4362 mdeep; PX1) were drilled into Permian grano-diorite with gabbroic dykes for the EGS, with anexpected electricity production of 1.2 MW (21).The PX1 well, only 6 m from PX2 at the surface,is 600m northwest of PX2 at the bottom (Fig. 3).The drilling began in September 2012 and wascompleted in November 2015. No earthquakeswithML > 2.0 were recorded within 10 km of thePohang EGS site between 1978 and 2015 (22); atotal of six earthquakes withML 1.2 to 1.9 weredetected in the area between 2006 and 2015. Tofurther examine the seismicity around the EGSsite, we improved the earthquake catalog byapplying a matched filter to continuous wave-forms (23) recorded by a permanent seismicstation (PHA2; Fig. 2A), located about 10 kmnorth of the EGS site, during the period between1 January 2012 and 14 November 2017. Oncedetected, waveforms were visually inspected fortime differences between P- and S-wave arrivalsconsistent with a source at Pohang (~1.54 s).The matched-filter analysis found no noticeableearthquakes at the EGS site before the comple-tion of drilling. We detected a total of 148 earth-quakes by the match filtering that all occurredafter the completion of the drilling, includingfour earthquakes with ML > 2.0.Hydraulic stimulation began on 29 January

2016 and comprised four phases of injectionwith a total volume of 12,800 m3 at injectionrates of 1.00 to 46.83 liter s−1 (Fig. 4). Fluid wasinjected into both PX1 (phases 2 and 4) and PX2(phases 1, 3, and 4). To investigate the relation-ship between seismicity and fluid injection, weused data on regional earthquakes detected bythematched filter together with earthquake dataprovided by the Ministry of Trade, Industry, andEnergy (MTIE), Republic of Korea (23). We didnot detect noticeable microearthquakes beforethe drilling. The timing of the earthquakes coin-cides with that of fluid injections. The first re-ported hydraulic stimulation (phase 1) was carriedout between 29 January 2016 and 20 February2016, followed by three additional phases offluid injection (Fig. 4). Each injection phasewas accompanied by intense seismic activitythat started only a few days after injection.Microseismic activity decreased rapidly after thetermination of fluid injection. The magnitudesof induced earthquakes tend to increase withthe net volume of injected fluid. After a ML 3.1earthquake on 15 April 2017, which was thelargest felt event near the EGS site before theML 5.4 Pohang earthquake, we deployed eighttemporary seismic stations around the EGS site.Each standalone station was equipped with athree-component velocity–type short-periodsensor. These sensors record continuous seismic

RESEARCH

Kim et al., Science 360, 1007–1009 (2018) 1 June 2018 1 of 3

1Department of Geological Science, Pusan NationalUniversity, Busan 46241, Republic of Korea. 2Department ofEarth and Environmental Sciences, Korea University, Seoul02841, Republic of Korea. 3School of Earth andEnvironmental Sciences, Seoul National University, Seoul08826, Republic of Korea.*Corresponding author. Email: kwanghee@pusan.ac.kr (K.-H.K.);reejh@korea.ac.kr (J.-H.R.)

Fig. 1. Tectonic map of northeast Asia.Saw-toothed lines with solid teeth denotesubduction zones. The broken line with openteeth represents an incipient subduction zone(30). EU, Eurasian Plate; NA, North AmericanPlate; PS, Philippine Sea Plate; PA, Pacific Plate.

Corrected 31 May 2018. See full text. on M

arch 8, 2021

http://science.sciencemag.org/

Dow

nloaded from

data at a sampling frequency of 200 Hz. Installa-tion was completed on 10 November 2017. All ofthem are still in active operation.TheML 5.4 mainshock occurred at the Pohang

EGS site on 15 November 2017 and was precededand followed by foreshocks and aftershocks, re-spectively, all of which were well recorded by ourlocal seismic array at distances of 0.6 to 2.5 km

from the mainshock epicenter (Figs. 2A and 3).We precisely relocated this earthquake sequencewith the hypoDD software package (23, 24) (fig.S2). We plotted the spatial distribution of sixforeshocks (ML ≤ 2.6), the mainshock, and 210aftershocks that occurred within 3 hours of themainshock (Fig. 3). The first two foreshocksoccurred about 9 hours before the mainshock;

the remaining four preceded it by 6 to 7 min.Most hypocentral depths fell in the range of 4to 6 km, and the mainshock depth was about4.5 km. Of note, the hypocenters of the fore-shocks andmainshockwere located immediatelyadjacent to the bottom of PX1. The hypocentraldepths of the Pohang earthquake sequence areshallower than those ofmost earthquakes in the

Kim et al., Science 360, 1007–1009 (2018) 1 June 2018 2 of 3

Fig. 2. Geologic map and column ofthe Pohang basin. (A) Map showingrocks and faults of the Pohang basinand adjacent area. One permanentseismic station operated by the KoreaMeteorological Administration (PHA2)and our eight temporary seismic stationsare represented by a dark blue squareand green triangles, respectively. Thegreen square denotes the site of thePohang enhanced geothermal system(EGS). The geologic map was compiledfrom (18, 19). (B) Geologic column of thePohang EGS site with injection well. Thegeologic column was compiled from(19, 20).

Fig. 3. Spatial distribution of epicenters and hypocenters of the2017 Pohang earthquake sequence. (A) Epicenters of six foreshocks(red circles), mainshock (red star), and 210 aftershocks (black opencircles) recorded in the first 3 hours after the mainshock. The location ofthe Pohang EGS is indicated by a green square. Blue triangles representour eight temporary seismic stations. The red beach ball representsthe source mechanism of the mainshock. Black beach balls showthe focal mechanism solutions of representative aftershocks. Numbersabove beach balls are local magnitudes. The black beach balls of

the ML 2.4 and 3.5 aftershocks, recorded 1 and 3 days after themainshock, respectively, show strike-slip faulting. X−X′ and Y−Y′ denotethe locations of the cross sections shown in (B) and (C), respectively.(B and C) Hypocentral distributions of earthquakes, projected ontovertical planes along the lines X−X′ (B) and Y−Y′ (C) shown in(A). The red beach ball in (B) represents the focal mechanism ofthe mainshock projected onto a vertical cross section. PX1 and PX2denote production and injection wells, respectively. Other symbols arethe same as for (A).

RESEARCH | REPORT

Corrected 31 May 2018. See full text. on M

arch 8, 2021

http://science.sciencemag.org/

Dow

nloaded from

Korean Peninsula, which tend to occur at depthsof 10 to 20 km (25). For comparison, the largestinstrumentally recorded earthquake in SouthKorea, the 2016 Gyeongju event (ML 5.8), had ahypocentral depth of about 14 km (26). Thespatiotemporal distribution of hypocenters in-dicates that the rupture plane consisted of twosegments, a main southwestern segment and asubsidiary northeastern segment (Fig. 3 andfig. S3). The aftershocks tended to occur earlieron the main segment than on the subsidiarysegment (fig. S4). This observation, togetherwith the locations of the foreshocks and main-shock on the main segment, suggests that themain segment ruptured earlier than the sub-sidiary one. By statistical plane-fitting usingMATLAB software (Mathworks), we determinedthe best-fit orientations of the main and sub-sidiary rupture planes to be N36°E (strike)/65°NW (dip) and N19°E/60°NW, respectively(fig. S3). These orientations are consistent withthe nodal planes determined by focalmechanismsolutions (Fig. 3). Themain segment shows thrustfaulting with a minor strike-slip component,whereas the subsidiary segment is dominatedby strike-slip faulting with a minor dip-slip com-ponent. The compression axis trends E–W orENE–WSWwith a shallow plunge, similar to thatof other earthquakes in South Korea (15, 16). Thelocations of the foreshocks andmainshock, at thebottom of the injection well, suggest that fluidwas injected directly into the fault zone.The temporal relationship between seismicity

and fluid injection, the spatial relationship be-tween the hypocenters and the EGS site, and thelack of seismicity in the area before the EGS wasestablished all suggest that the Pohang earth-quake was induced. Furthermore, the immediateresponse of seismicity to fluid injection and thelocations of the foreshocks andmainshock at thebottom of the injection well suggest that fluidwas injected directly into a fault zone. The faultplane inferred from the spatial distribution ofhypocenters and focal mechanism solutionsstrikes NE and dips to the NW (fig. S4A), similarto Quaternary thrust faults in southeasternKorea.

In fact, a magnetotelluric survey of the EGS sitedetected a low-resistivity feature that could be thefault zone, striking NE and dipping to the NW(23, 27) (fig. S5). Reverse slip along a subsurfacefault is consistent with the current stress field.All of these lines of evidence indicate that thePohang earthquake was “almost certainly induced”in Frohlich et al.’s system of assessment (28). Ifwe use McGarr’s (2) equation for the relation-ship between the maximum magnitude and thetotal volume of injected fluid, about 4.7 × 106 m3

of injected fluid would be required to inducean Mw 5.4 earthquake, which is more than 810times the fluid volume injected at the PohangEGS site. The permeability structure of fault zonesis highly heterogeneous, and patches or layersof clay-rich gouge within the fault core act asbarriers to fluid flow (29). The pore pressure thuscan locally reach a critical value for earthquakenucleation after a relatively small volume of fluidis injected, depending on fault zone structure.Our results imply that if fluid is injected directlyinto a near-critically stressed fault, it can inducea larger earthquake than current theory pre-dicts. Detailed investigation of the geological,geochemical, and geophysical properties of thePohang EGS site will improve our understand-ing of earthquake-inducing processes.

REFERENCES AND NOTES

1. W. L. Ellsworth, Science 341, 1225942 (2013).2. A. McGarr, J. Geophys. Res. Solid Earth 119, 1008–1019

(2014).3. A. Zang et al., Geothermics 52, 6–21 (2014).4. M. D. Zoback, S. M. Gorelick, Proc. Natl. Acad. Sci. U.S.A. 109,

10164–10168 (2012).5. A. McGarr, J. Geophys. Res. Solid Earth 81, 1487–1494 (1976).6. N. J. van der Elst, M. T. Page, D. A. Weiser, T. H. W. Goebel,

S. M. Hosseini, J. Geophys. Res. Solid Earth 121, 4575–4590(2016).

7. E. L. Majer et al., Geothermics 36, 185–222 (2007).8. G. Grünthal, Geothermics 52, 22–35 (2014).9. M. O. Häring, U. Schanz, F. Ladner, B. C. Dyer, Geothermics 37,

469–495 (2008).10. D. T. Trugman, A. A. Borsa, D. T. Sandwell, Geophys. Res. Lett.

41, 8767–8774 (2014).11. National Disaster and Safety Status Control Center, Ministry of

the Interior and Safety, Republic of Korea, press release(6 December 2017).

12. S. K. Chough, S.-T. Kwon, J.-H. Ree, D. K. Choi, Earth Sci. Rev.52, 175–235 (2000).

13. J.-H. Ree et al., Isl. Arc 12, 1–12 (2003).14. S. H. Yoon, Y. K. Sohn, S. K. Chough, Mar. Geol. 352, 70–88

(2014).15. H. Choi, T.-K. Hong, X. He, C.-E. Baag, Tectonophysics

572–573, 123–133 (2012).16. J.-C. Park, W. Kim, T. W. Chung, C.-E. Baag, J.-H. Ree,

Geophys. J. Int. 169, 1103–1114 (2007).17. Y. K. Sohn, C. W. Rhee, H. Shon, Sediment. Geol. 143, 265–285

(2001).18. Y. K. Sohn, M. Son, Sedimentology 51, 1387–1408 (2004).19. T. J. Lee, Y. Song, D.-W. Park, J. Jeon, W. S. Yoon, “Three

dimensional geological model of Pohang EGS pilot site, Korea,”in Proceedings of the World Geothermal Congress, Melbourne,Australia, 19 to 25 April 2015; https://pangea.stanford.edu/ERE/db/WGC/papers/WGC/2015/31025.pdf.

20. K.-S. Yoon, J.-S. Jeon, H.-K. Hong, H.-G. Kim, A. Hakan,J.-H. Park, W.-S. Yoon, “Deep drilling experience for PohangEnhanced Geothermal Project in Korea,” in Proceedings of theWorld Geothermal Congress, Melbourne, Australia, 19 to25 April 2015; https://pangea.stanford.edu/ERE/db/WGC/papers/WGC/2015/06034.pdf.

21. M. Kim, B. Yoon, C. Lee, K. G. Park, W.-S. Yoon, Y. Song,T. J. Lee, “Microseismic monitoring during hydraulicstimulation in Pohang (Korea) for EGS pilot project” [abstractS23B-0804], American Geophysical Union Fall Meeting, NewOrleans, LA, USA, 11 to 15 December 2017.

22. The Korea Meteorological Administration catalog is available athttp://necis.kma.go.kr/.

23. Supplementary materials.24. F. Waldhauser, W. L. Ellsworth, Bull. Seismol. Soc. Am. 90,

1353–1368 (2000).25. Korea Meteorological Administration earthquake data

service; www.weather.go.kr/weather/earthquake_volcano/domesticlist.jsp.

26. K.-H. Kim et al., Geosci. J. 20, 753–757 (2016).27. T. J. Lee, Y. Song, T. Uchida, Mulli-tamsa 8, 145–155 (2005).28. C. Frohlich et al., Seismol. Res. Lett. 87, 1022–1038 (2016).29. J. S. Caine, J. P. Evans, C. B. Forster, Geology 24, 1025–1028

(1996).30. T. Seno, S. Stein, A. E. Gripp, J. Geophys. Res. Solid Earth 98,

17941–17948 (1993).

ACKNOWLEDGMENTS

We thank Representative S. S. Kim of the National Assemblyand the Ministry of Trade, Industry and Energy, Republic of Korea,for providing fluid injection data. We are grateful to the KoreaMeteorological Administration for providing continuous waveformsused in the study. We also thank two anonymous reviewersfor their constructive comments. Funding: This work wassupported by the Nuclear Safety Research Program throughthe Korea Foundation of Nuclear Safety (KoFONS) using financialresources granted by the Nuclear Safety and Security Commission(NSSC) of the Republic of Korea (no. 1705010). Authorcontributions: K.-H.K.: conceptualization, formal analysis, fundingacquisition, methodology, original draft, and review and editing.J.-H.R.: conceptualization, formal analysis, funding acquisition,methodology, original draft, and review and editing. Y.K.:investigation, methodology, validation, and review and editing.S.K.: data curation, formal analysis, investigation, and software.S.Y.K.: data curation, formal analysis, investigation, and software.W.S.: data curation, formal analysis, investigation, and software.Competing interests: The authors declare no conflicts of interest.Data and materials availability: Our earthquake catalog,including the earthquake source parameters (locations andtimes) and waveforms used in this paper, is available athttps://zenodo.org/record/1218738#.WtPQROTEaRE. Earthquakewaveform data recorded at PHA2 can be acquired from theNational Earthquake Comprehensive Information System, KoreaMeteorological Administration (http://necis.kma.go.kr/; lastaccessed April 2018). Continuous data may be obtained from thewebsite on request.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/360/6392/1007/suppl/DC1Materials and MethodsFigs. S1 to S5References (31–38)

18 March 2018; accepted 16 April 2018Published online 26 April 201810.1126/science.aat6081

Kim et al., Science 360, 1007–1009 (2018) 1 June 2018 3 of 3

Fig. 4. History of fluid injection volume. Volume of injected and bleed-off fluid (left axis) and netfluid volume (right axis) as a function of time. Red circles denote times of seismic events from dataprovided by the MTIE (63 events; magnitude information is only available for M > 1). Dark blue circlesdenote seismic events determined by matched-filter analysis (148 events). The right axis also givesthe local magnitudes of seismic events.

RESEARCH | REPORT

Corrected 31 May 2018. See full text. on M

arch 8, 2021

http://science.sciencemag.org/

Dow

nloaded from

5.4 Pohang earthquake in South Korea was an induced eventwMAssessing whether the 2017 Kwang-Hee Kim, Jin-Han Ree, YoungHee Kim, Sungshil Kim, Su Young Kang and Wooseok Seo

originally published online April 26, 2018DOI: 10.1126/science.aat6081 (6392), 1007-1009.360Science 

, this issue p. 1003, p. 1007Sciencefuture development of this geothermal resource.coincidentally at the EGS site location, but the aftershock distribution and other lines of evidence are concerning for probably or almost certainly anthropogenically induced. The possibility remains that the earthquake occurredobservations, teleseismic waveform analysis, and stress modeling leads to the assessment that the earthquake was Pohang, as an induced event. The combination of data from a local seismometer network, well logs, satellitepresent seismic and geophysical evidence that may implicate the second of these earthquakes, which occurred in

et al. and Kim et al.magnitude-5 earthquakes recently occurred in South Korea during EGS site development. Grigoli Enhanced geothermal systems (EGSs) provide a potentially clean and abundant energy source. However, two

Triggering quakes in a geothermal space

ARTICLE TOOLS http://science.sciencemag.org/content/360/6392/1007

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2018/04/25/science.aat6081.DC1

CONTENTRELATED http://science.sciencemag.org/content/sci/360/6392/1003.full

REFERENCES

http://science.sciencemag.org/content/360/6392/1007#BIBLThis article cites 29 articles, 7 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on March 8, 2021

http://science.sciencem

ag.org/D

ownloaded from