Final Report on Estimating Crustal Stress and Fracture Permeability

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 1 of 30

Grant Agreement Number 608553

IMAGE Integrated Methods for Advanced Geothermal Exploration

IMAGE-D4.04: Final Report on Estimating crustal stress and fracture permeability

Responsible author Oliver Heidbach (GFZ)

Responsible WP-leader Sæunn Halldórsdóttir (ISOR) Responsible SP-leader Gylfi Páll Hersir (ISOR) Contributions by: Moritz Ziegler, Arno Zang (GFZ) Mojtaba Rajabi (Adelaide University, Australia) Gylfi Páll Hersir, Kristján Ágústsson, Sigurveig Árnadóttir (ISOR) Andrea Brogi, Domenico Liotta (UNIBARI)

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 2 of 30

Introduction and summary This document summarizes the results of Task 4.4 that consists of the analysis of crustal stress data and data of the fracture permeability for Iceland. This is the final report of the status of data analysis and compilation. In addition to an increase in data records the report provides the integration of the findings presented in the preliminary report (IMAGE-D4.08) into a wider context.

The stress data analysis of image log data from 51 boreholes in Iceland has been performed and is summarized in the preliminary report. For the final report the number of data records was further increased by including also data from earthquake focal mechanism and geological indicators. The final result is a stress database for Iceland with ten times more data records on the stress state than previously available (increase from 38 to 485 data records) and enabled us to analyse in detail the stress pattern of whole Iceland and to identify distinct stress provinces. The results are published in the peer-reviewed journal Tectonophysics (Ziegler et al., 2016a; Appendix A) and in a new comprehensive stress map of Iceland (Ziegler et al., 2016b; Appendix B).

The study of fracture permeability’s has been performed through original field studies carried out in Geitafell (Iceland) the field area chosen as analogue of the presently active geothermal systems of Krafla presented in the preliminary report and in the Skagafjörður area. The obtained results are in agreement with data from literature and direct measurements in boreholes. Furthermore the data is an extension to the stress database.

Contents 1. Introduction ...... 3 2. Crustal Stress: A comprehensive stress map of Iceland ...... 4 2.1 Introduction ...... 4 2.2 Stress indicators ...... 5 2.2.1 Borehole breakouts and drilling induced tensile fractures ...... 5 2.2.2 Earthquake focal mechanism ...... 7 2.2.3 Geological Methods ...... 7 2.2.4 Additional Methods ...... 7 2.3 Assignment of the standardized World Stress Map quality criteria ...... 8 2.4 Stress data records ...... 9 2.4.1 Borehole data ...... 10 2.4.2 Volcanic vent alignments ...... 13 2.5 Stress pattern analysis ...... 14 2.6 Final remarks ...... 15 3. Fracture permeability ...... 16 3.1 Aims ...... 16 3.2 Background ...... 17 3.3 Methodology ...... 19 3.4 Structural Stations ...... 20 3.5 Data analysis ...... 22 3.6 Final remarks ...... 24 4. Conclusion ...... 25 5. References ...... 26 6. Appendix A ...... 29 7. Appendix B ...... 30

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 3 of 30

1. Introduction Geothermal exploration requires, amongst others, detailed information on the contemporary crustal stress field as well as on the fracture permeability. In the framework of the IMAGE Project these data are compiled using existing data from publications as well as new analysis of borehole data and field work in Iceland. With the high geothermal potential of the island the increase of information on the subsurface is beneficial for successful future geothermal projects.

Information on the contemporary crustal stress field in a reservoir is mainly required in order to estimate the criticality of potential faults and the stability of the intact rock mass. This is especially important in light of the occurrence of induced seismic events, in order to mitigate them, and to reduce any potential damage. The fracture permeability controls through the connectivity of fluid pathways the productivity of a reservoir or a borehole. The orientation of these fractures is largely controlled by the stress field at the time of fracture generation. Hence, knowledge on the stress field is at the same time knowledge of the fracture orientation in the current stress field. Furthermore, if the permeability is not enough for a sustainable production of fluid an enhancement operation/ stimulation open up new fluid pathways. Their orientation is again controlled by the stress field.

In Task 4.4 the focus is on a comprehensive analysis of the contemporary crustal stress field and the fracture permeability in Iceland. For the crustal stress a database with ten times more data records on the stress state than previously available (increase from 38 to 485 data records) is the final product. This high density data set enabled us to analyse in detail the stress pattern of whole Iceland and to identify distinct stress provinces. These new data were integrated into the new World Stress Map database release 2016 (www.world-stress-map.org) to guarantee the long-term accessibility. On several field trips to Iceland the fracture permeability and (paleo) stress data at several locations were recorded and investigated. This provides information on the permeability of recent geothermal systems that are analogy to the investigated paleo systems. Furthermore, the surface stress information increases the significance of the stress data compilation.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 4 of 30

2. Crustal Stress: A comprehensive stress map of Iceland 2.1 Introduction

The information of the contemporary crustal stress field is an important parameter to assess the reactivation potential and slip tendency of fractures and faults. Information on the stress state in terms of the orientation of the maximum horizontal stress (SHmax) is compiled in the World Stress Map (Heidbach et al., 2016). However, the amount and distribution of stress data records in the World Stress Map database release 2008 (Heidbach et al., 2010) reveals that a profound knowledge of the stress state is limited to very few areas. In many areas and regions very few to no data records are available. In the framework of the IMAGE Work Package 6 an update of the database and increase of the amount of data records of the European part of the World Stress Map was conducted (see Deliverable 6.1).

Figure 1 The new stress map of Iceland. Lines show the orientation of maximum horizontal stress SHmax. Line length is proportional to the quality of the data. Red symbols denote normal faulting (NF), green is strike-slip faulting (SS), blue is thrust faulting (TF) and black is unknown (U).

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 5 of 30

Task 4.4 is even more focussed on Iceland as a specific region that was previously underrepresented in the World Stress Map database. In the preliminary report we presented the analysis of new data of the crustal stress field in Iceland by analysing image logs from 51 boreholes. In order to further increase the knowledge of the stress state a comprehensive literature review was conducted to provide more data records on the stress state from different stress indicators (Figure 1). This provides an integrated view and enables the (statistical) analysis of the stress state. The new stress map of Iceland has 495 data records which are more than ten times the previous amount from the World Stress Map release 2008. In this final report we present the comprehensive stress map of Iceland (Figure 1).

2.2 Stress indicators

Various indicators for the SHmax orientation exist. Depending on the type of indicator and its individual characteristics the confidence of each stress data record varies. Nonetheless various types of indicators are used for the estimation of the SHmax orientation. The most important indicator in the World Stress Map is clearly the focal mechanism solution. However, especially in Iceland several other methods also have a significant share. They are presented in the following.

2.2.1 Borehole breakouts and drilling induced tensile fractures

Stress data can be derived from borehole breakout data (BO) and drilling induced tensile fracture (DIF). These features are observed in boreholes by means of calliper tools, acoustic image logs, or tele viewer logs. These logs are then interpreted and if such features are present they can be identified. In case of distinct features and high quality borehole images the SHmax orientation can be estimated to a high degree of certainty.

Borehole breakouts are stress-induced elongations of the wellbore. When a borehole is drilled the material removed from the subsurface is no longer supporting the surrounding rock. As a result, the stresses become concentrated in the surrounding rock (i.e. the wellbore wall). Borehole breakouts occur when the wellbore stress concentration (circumferential or hoop stress) exceeds that required to cause compressive failure of intact rock (Bell & Gough, 1979 & Figure 2). The elongation of the cross-sectional shape of the wellbore is the result of compressive shear failure on intersecting conjugate planes, which causes pieces of the borehole wall to spall off (Bell and Gough, 1979 & Figure 3). The maximum circumferential stress around a vertical borehole is perpendicular to SHmax (Kirsch, 1898). Hence, borehole breakouts are elongated perpendicular to the present-day SHmax orientation (Bell and Gough, 1979).

DIFs are caused by tensile failure of the borehole wall and form when the minimum wellbore stress concentration is less than the tensile strength of the rock. The minimum circumferential stress around a vertical borehole occurs parallel to SHmax (Kirsch, 1898). Hence, DIFs form parallel to the SHmax orientation in vertical boreholes (Figure 3). The present-day SHmax orientation was determined herein from interpretation of BOs and DIFs picked in Acoustic Borehole Imager (ABI) logs.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 6 of 30

Figure 2 Generation of DIFs and BOs Generation of DIFs and BOs as a function of rock strength from (Hillis & Reynolds, 2000). BOs occur when the circumferential stress at the borehole wall exceeds the compressive strength of the rock; the develop perpendicular to the orientation of maximum horizontal stress SHmax. DIFs occur when the circumferential stress at the borehole wall exceeds the tensile strength of the rock and fractures perpendicular to the orientation of minimum horizontal stress Shmin are generated.

Figure 3 Results of a hollow cylinder lab test simulating borehole breakout (performed by CSIRO Division of Geomechanics). Intersection of conjugate shear failure planes results in enlargement of the cross-sectional shape of the wellbore. SHmax and Shmin refer to the orientations of maximum and minimum horizontal stress, respectively.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 7 of 30

2.2.2 Earthquake focal mechanism

Earthquakes can be used as stress indicators by computing their focal mechanism which is used to estimate the stress orientation from the orientations of the principal strain axes. Even though the stress and strain axes are not identical (McKenzie, 1969) the strain axis is used as a proxy for the orientations of the principal stress axes (Célérier, 2010). The only reliable information that can be gained is that the maximum principal stress axes lie within the dilatational quadrant of the focal mechanism (McKenzie, 1969). Hence, the confidence level of the SHmax orientation determined from single focal mechanism is limited (Heidbach, 2016).

The formal inversion of several focal mechanism solutions of several seismic events which occurred in a uniform stress field provides the orientations of the principal stress axes (Gephart & Forsyth, 1984, Michael, 1987, Rivera & Cisternas, 1991). Thereby the confidence of the SHmax orientation can be enhanced compared to single focal mechanism solutions. Furthermore, an averaging of several focal mechanism solutions or the generation of composite solutions also provides basic information on the stress field but generally to a lower level of confidence.

2.2.3 Geological Methods

Analogously to earthquakes that are observed by seismometers and whose focal mechanisms provide information on the stress state, geological observations of (paleo) earthquakes, i.e. fault slip data sets observed in the field, or the inversion of several such datasets provide the orientation of the stress field (Angelier, 1979, Michael, 1984, Angelier, 1984). Fault slip data that are observed in the field are usually a manifestation of seismic events that occurred in the past. However, since the stress field has potentially changed since then, such paleo event data may indicate a paleo stress field rather than the recent stress field. In order to prevent the misinterpretation of paleo stress fields as the current one the observed fault slip has to have occurred during the Quaternary according to the WSM guidelines (Sperner et al., 2003; Zoback, 1992).

2.2.4 Additional Methods

The orientation of Quaternary intrusions and the alignment of volcanic vents are used as stress indicators (Nakamura, 1977, Nakamura et al., 1977). Volcanic vents are always related to volcanic activity or even eruptions which tend to be easier to date compared to fault slip that is required for other geological stress indicators. Often the age of volcanic eruptions and activity are known.

Further methods that can provide information on the orientation of SHmax are strain recovery methods (Teufel, 1983), borehole slotter (Corthésy et al., 1999), petal centreline fractures (Plumb & Cox, 1987), core disking (Funato et al., 2012), or shear wave splitting (Crampin & Peacock, 2005).

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 8 of 30

2.3 Assignment of the standardized World Stress Map quality criteria All data in the World Stress Map (WSM) database are quality ranked to allow the comparison of data records that are from different stress indicators for the SHmax orientation such as borehole breakouts, drilling-induced tensile fractures, overcoring, hydraulic fracturing, earthquake focal mechanism solutions and geological indicators (Heidbach et al., 2010, Sperner et al., 2003, Zoback, 1992). An excerpt of the quality ranking for stress orientations is presented in Table 1.

Table 1 Excerpt from the latest version of the World Stress Map quality criteria for borehole breakouts from image logs, drilling induced fractures, focal mechanism solutions, fault slip data, and volcanic vent alignments, s.d. = standard deviation (Heidbach et al., 2016). Indicator A-Quality B-Quality C-Quality D-Quality E-Quality type Borehole Wells that have Wells that have Wells that have Wells that have Wells with no Breakouts 10 or more at least 6 at least 4 less than 4 reliable (from distinct breakout distinct distinct breakouts zones breakouts image logs) zones with a breakout zones breakouts zones or a combined detected or combined length with a combined with a combined length < 20 m with extreme > 100 m length > 40 m length > 20 m or with scatter of and with and with and with s.d. > 25° breakout s.d. ≤ 12° s.d. ≤ 20° s.d. ≤ 25° azimuth or with s.d. > 40° Drilling ≥ 10 distinct ≥ 6 distinct ≥ 4 distinct < 4 distinct Wells without Induced fracture zones in fracture zones fracture zones in fracture zones in fracture zones fractures a single well in a single well a single well a single well or or s.d. > 40° with a combined with a combined with a combined a combined length ≥ 100 m length ≥ 40 m length ≥ 20 m length < 20 m and s.d. ≤ 12° and s.d. ≤ 20° and s.d. ≤ 25° and s.d. ≤ 40° Focal Formal inversion Formal Well constrained Well constrained Mechanism mechanism of ≥ 15 well inversion of ≥ 8 single event single event with P,B,T constrained well constrained solution, M ≥ 2.5 solution, M < 2.5 axes all single event single event (e.g. CMT plunging 25°- solutions in solutions in solutions) 40° close close or geographic geographic Mechanism proximity and proximity and with P and T s.d. or misfit s.d. or misfit axes both angle ≤ 12° angle ≤ 20° plunging 40°- 50° Fault slip Inversion of ≥ 25 Inversion of ≥ Inversion of ≥ 10 Inversion of ≥ 6 fault-slip data 15 fault-slip fault-slip data fault-slip data with a data with a with a with a fluctuation ≤ 9° fluctuation ≤ 12° fluctuation ≤ 15° fluctuation ≤ 18° for ≥ 60% of the for ≥ 45% of the for ≥ 30% of the for ≥ 15% of the whole dataset whole dataset whole dataset whole dataset Volcanic ≥ 5 Quaternary ≥ 3 Quaternary Single well- Volcanic Vent vent alignments vent alignments exposed alignment Alignment or "parallel" or "parallel" Quaternary dike inferred from < 5 dikes with s.d. ≤ dikes with s.d. ≤ or vents 12° 20° Single alignment with ≥ 5 vents

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 9 of 30

2.4 Stress data records

The update of the World Stress Map for Iceland led to an increase of the amount of available data records to now 495 (Figure 1). This is more than ten times the previously available information. The data records are from a variety of different stress indication methods such as from boreholes, geologic observations, or focal mechanism solutions. Each data record is assigned a quality from A (highest) to E (lowest) depending on the confidence level of the stress indication method and the individual data record. Table 2 provides a summary of the different methods and qualities present in the Iceland stress map.

Table 2 The quality and type of data records in Iceland. FMF: Formal focal mechanism inversion, FMS: Focal mechanism single, FMA: Focal mechanism averaging, BO: Borehole breakouts, DIF: Drilling induced tensile fractures, HF: Hydraulic fracturing, OC: Overcoring, GFI: Geological fault slip data, GVA: Vent alignment Type A B C D E Total FMF 15 7 14 36 FMS 63 22 90 175 FMA 9 9 BO 6 13 30 49 DIF 1 3 1 15 1 21 HF 1 2 6 9 OC 25 25 GFI 1 11 40 63 14 129 GVA 1 11 25 2 3 42 Total 18 33 137 130 177 495

As indicated in Table 2 and shown in Figure 4 most of the stress data records in Iceland are from focal mechanism solutions or geologically observed fault slip data. This is exceptional since geological fault slip data is overrepresented in Iceland compared to the rest of the World (Heidbach et al., 2010, 2008). The quality distribution is usual with a few high quality data records from stress inversions and boreholes and a large amount of data with a lower confidence level.

Type Quality

FMS FMF A BO B DIF C GFI D GVA E Other

Figure 4 Type and Quality of data records in Iceland.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 10 of 30

2.4.1 Borehole data

In 2002 to 2013 the Iceland GeoSurvey (ÍSOR) ran borehole image logs in geothermal wells mainly close to settlements at the shores, in the south Iceland lowlands and around Akureyri and Krafla in the North (see Figure 5 for location). We collected and analysed 34 km acoustic image logs of 51 geothermal and scientific boreholes to interpret borehole breakouts (BOs) and drilling induced tensile fractures (DIFs, Figure 6).

Figure 5 The location of geothermal boreholes with acoustic image logs. The black and grey triangles denote the location of boreholes with and without stress indicators (based on our image log analysis) respectively. The white triangles show the location of borehole HJ-20 Hjalteyri, ST-16 Sigtún, and HN-16 Hellisheiði.

From the 51 Icelandic boreholes with image logs 23 boreholes contained at least one stress indicator, i.e. either a drilling induced tensile fracture or a borehole breakout (Table 3). The indicators are mainly found between the surface and a depth of 1 km. Some few indicators are located in deeper sections of the boreholes. The deepest indicators are found in an average depth of 2.06 km in well HN-16 in Hellisheiði. There, six sets of drilling induced tensile fractures are found between 2021 m and 2187 m depth. Even though the standard deviation is low (12°) the added length of the feature of 9 m only qualifies it for a D quality. The shallowest seismic events recorded in Iceland are in a depth of 3 km. Thus borehole indicators bridge the gap between surface (geological) indicators and focal mechanism solutions.

In 28 of the 51 examined wells no stress indicators were found. Some of the wells were quite shallow and did not reach areas with stresses high enough that borehole wall failure occurs. In other instances the image quality was not sufficient to find enough evidence to constrain stress indicators.

Most of the studied wells are slightly deviated from vertical (<4°) which still allows the interpretation of stress related features in every stress regime (Mastin 1988; Tingay et al. 2005; Peška & Zoback 1995). Of the Icelandic wells eight contain both BOs and DIFs. In these cases the indicators are generally in a good agreement with each other. BOs or DIFs are observed in seven and eight wells respectively (Figure 5 & Figure 6).

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 11 of 30

Table 3 Stress indicators from the analysed acoustic borehole images. All the information required for the WSM quality ranking is included in the Table. Azimuth: Interpreted orientation of SHmax. Number: The amount of recognised fracture pairs in a single well. S.D.: Standard deviation calculated according to the circular statistics of bi-polar data by Mardia (1972) with a weighting depending of the length (short: L) of the feature. Length: The added length of the fractured borehole sections. Top and Bottom: The depth of the uppermost and lowermost stress indicator found in the borehole. Depth: The mean between top and bottom. Date: Date of the tool run. le ID Boreho Bottom [m] Depth [km] Length [m] Length Longitude Weighting Location Azimuth Latitude Number Top [m] Quality Type Date S.D.

HH-08 63.425023 -20.25904 133 BO 1.05 C Vestmannaeyjar 20050415 11 13 22 L 789 1719

KH-34 63.98881 -20.44006 67 BO 0.04 D Kaldárholt 20050322 1 0 2 L 38 40

KH-34 63.98881 -20.44006 109 DIF 0.2 D Kaldárholt 20050322 2 2 3 L 55 390

SO-01 63.995165 -21.13729 47 DIF 0.32 D Sogn/Ölfus 20050322 3 13 6 L 314 325

HE-21 64.008906 -21.3438 41 BO 1.67 D Hellisheiði 20060215 11 14 16 L 1608 1748

HE-21 64.008906 -21.3438 67 DIF 1.35 B Hellisheiði 20060215 53 14 123 L 912 1812

HN-01 64.026124 -21.45102 45 BO 0.9 C Hellisheiði 20050405 20 22 26 L 866 977

HN-01 64.026124 -21.45102 44 DIF 0.85 D Hellisheiði 20050405 7 18 10 L 768 977

HK-15 64.041 -20.81377 8 BO 0.1 C Grímsnes 20060303 33 15 25 L 37 183

HN-12 64.044597 -21.38636 84 DIF 1.5 D Hellisheiði 20101021 7 21 11 L 1152 1878

HN-16 64.045106 -21.3862 86 DIF 2.06 D Hellisheiði 20101018 6 12 9 L 2021 2187

HF-01 64.391916 -15.34195 151 DIF 0.6 D Hoffell 20130221 10 11 17 L 424 805

ASK-29 64.393174 -15.34205 130 BO 0.11 D Hoffell 20120926 6 16 6 L 103 123

ASK-57 64.393898 -15.34267 4 BO 0.28 D Hoffell 20120926 1 0 1 L 283 284

HB-02 65.04501 -22.77176 60 BO 0.36 D Stykkishólmur 20070215 1 0 4 L 366 370

ST-16 65.5519 -18.07022 127 BO 0.35 C Sigtún/Eyjafjörður 20050126 28 9 37 L 111 671

ST-16 65.5519 -18.07022 140 DIF 0.4 D Sigtún/Eyjafjörður 20050126 5 7 16 L 329 508

BO-3 65.562966 -18.10464 107 DIF 0.07 D Botn 20130122 3 13 10 L 60 80

KV-01 65.692163 -16.81934 29 BO 1.43 D Krafla 20060803 1 0 1 L 1435 1437

KV-01 65.692163 -16.81934 164 DIF 1.43 D Krafla 20060803 2 8 2 L 1432 1435

K-18 65.702026 -16.73063 17 BO 0.74 D Krafla 20081118 2 4 6 L 733 750

HJ-17 65.855115 -18.2105 151 DIF 0.15 D Hjalteyri 20020221 2 11 2 L 122 170

HJ-13 65.855337 -18.21303 145 DIF 0.06 D Hjalteyri 20020220 1 0 3 L 62 65

HJ-20 65.856089 -18.21142 141 BO 1 D Hjalteyri 20050202 4 8 12 L 784 1176

HJ-20 65.856089 -18.21142 144 DIF 0.75 A Hjalteyri 20050202 60 11 136 L 352 1346

HJ-15 65.859457 -18.21754 154 DIF 0.2 D Hjalteyri 20020223 1 0 2 L 204 207

ARS-32 65.931479 -18.33783 163 BO 0.75 D Árskógsströnd 20060608 6 19 6 L 668 842

ARS-32 65.931479 -18.33783 173 DIF 0.55 C Árskógsströnd 20060608 17 14 36 L 206 713

SK-28 65.997822 -19.33668 143 BO 0.5 C Hrolleifsdalur 20051008 55 25 137 L 240 821

SD-01 66.127507 -18.96229 146 BO 0.45 D Skarðdalur/Tröllaskagi 20100925 2 3 3 L 430 537

SD-01 66.127507 -18.96229 140 DIF 0.5 B Skarðdalur/Tröllaskagi 20100925 20 11 69 L 319 687

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 12 of 30

The majority of the indicators are ranked with a D quality according to the World Stress Map guidelines for image logs (Tingay et al. 2008). Still A, B, and C quality indicators do exist. This distribution of quality is partly related to the challenges of well-logging in a geothermally active and igneous environment which results in a poor image quality which prevents the detection of some features. Special tools adapted to high temperatures are required and may only remain in the well for a short time period (Ásmundsson et al. 2014). In some cases the well diameter was too large for the tool so that no clear picture could be recorded. In addition, in some of the wells image tools were not centralised which creates different types of artefacts which renders the images impossible to interpret for stress indicators.

Figure 6 The acoustic image of borehole sections. The depth is scaled 1:25 m. Left: Borehole breakouts in well ST-16 Sigtún close to Akureyri. The inferred overall orientation of SHmax from BOs is 127° in this well. Right: Drilling induced tensile fractures in well HJ-20 Hjalteyri close to Akureyri. The inferred overall SHmax orientation of from DIFs is 144° in this well.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 13 of 30

2.4.2 Volcanic vent alignments

The high volcanic activity in Iceland allows inclusion of young eruptive fissures, vent alignments and dykes from the Quaternary. We therefore included 17 GVAs produced by recent volcanic activities (even in historic times, Thordarson and Larsen, 2007). The data originates in geological mapping campaigns and is also displayed in the Geologic Map of Iceland — Bedrock (Hjartarson and Sæmundsson, 2014). Table 4 shows the stress orientations inferred from eruptive fissures mainly deduced from geologic mapping also presented in the map by Hjartarson and Sæmundsson (2014). They are quality ranked according to the WSM criteria shown in Table 1.

Table 4 Newly included volcanic vent and fissure alignments. Latitude Longitude Azimuth Quality Number S.D. Vents

Type

Location 63.43 -20.2 45 C Vestmannaeyjar 1 5 vents

63.82 -18.83 18 C Eldgjá (South) 1 6 fissures

63.9 -21.8 56 B Reykjanes 4 5 21 vents

63.94 -18.65 43 C Eldgjá (Middle) 1 5 fissures

64.1 -18.3 35 C Eldgjá (North) 1 9 fissures

64.25 -18.6 33 B Veiðivötn 4 13 67 fissures

64.29 -20.84 43 C Þjófahraun 1 11 fissures

64.4 -20.5 47 C Langjökull 2 3 10 vents

64.75 -16.6 30 C Kverkfjöll 1 7 fissures

64.8 -17.3 22 B Dyngjuháls 3 6 28 fissures

65.0 -17.15 29 C Trölladyngja/ Frambruni 1 8 fissures

65.15 -16.6 21 C Askja 1 14 vents

65.4 -16.8 9 C Fremrinámur 1 9 vents

65.5 -16.45 8 C Nýjahraun 2 6 16 fissures

65.6 -16.8 8 B Reykjahlíð 4 2 16 fissures

65.7 -16.8 6 C Krafla 1 10 fissures

65.9 -16.35 11 C Hólssandur 1 7 fissures

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 14 of 30

2.5 Stress pattern analysis

The updated World Stress Map database segment for Iceland has now sufficient data records in order to identify significant patterns in the SHmax orientation. Thus four distinct stress provinces are identified that can be attributed to the tectonic features of the island (Figure 7). The results are presented in more detail in the publication (see Appendix A).

Furthermore, a statistical analysis of the data records on an evenly spaced grid provides enhanced information on the mean SHmax orientation. Yet, the amount of data required for such an analysis within a certain area and on a grid with a given size is not sufficient in many areas. After the presented update of the Iceland stress map an analysis on a 0.5° grid is now feasible (Figure 7). In addition to the mean orientation of SHmax its variance as a measure of uncertainty, as well as the stress field wavelength, or the relation of the SHmax orientation to the plate motion can be computed (Ziegler & Heidbach, 2017a, 2017b). These analyses provide not just a robust insight into the stress pattern but also in its spatial variability. Thus, an idea of the stress state in areas with no data records is provided. Furthermore, the confidence level of this derived stress state is provided with the variance and wavelength information.

Figure 7 Left: Four Icelandic stress provinces. Right: The mean SHmax azimuth (lines) and its variance (colour coded).

However, both analyses show that the lack of data in the uninhabited areas of Iceland (mainly the Highlands) remain a challenge. Neither the identification of stress provinces nor the application of a statistical analysis on a regular grid can compensate this.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 15 of 30

2.6 Final remarks

The stress data compilation and update of the World Stress Map for Iceland performed in the framework of the IMAGE project allows to draw the following conclusion for the estimation of the crustal stress.

• A thorough literature review provides useful information on the stress field. Especially in a unique geological and tectonic setting such as Iceland large quantities of data can be gained.

• The interpretation of existing borehole logs potentially provides high quality stress data records. Even though it can be challenging in settings with mainly volcanic rock material.

• A statistical analysis provides information on the stress field that are to some extent in areas with no data records. However, a statistical estimation is only justified within small boundaries and “ground truth” is required.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 16 of 30

3. Fracture permeability 3.1 Aims

The activity here described is framed in Task 4.4 and implies collection and analysis of data for evaluation of the present stress field in geothermal areas in Iceland. UNIBA is engaged for collection of structural and kinematic field data in key areas.

The fieldwork was planned following indications deriving from the recent publication by Ziegler et al. (2016a), where it has been highlighted that the northern and central part of Iceland resulted with scarce information (Figure 8). The main aim was therefore to implement the stress field data set those areas. However, we will see that further and interesting considerations may derive by the collected data.

The fieldwork was developed in cooperation with Sigurveig Árnadóttir (ISOR).

investigated area

Figure 8 Stress field map of Iceland after Ziegler et al. (2016a). The study area is also indicated.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 17 of 30

3.2 Background

Ziegler et al. (2016a) have considered deformation occurred in rocks younger than 2 Ma. We have followed the same criterion, looking for structures involving geological bodies of the indicated age. However, these young rocks are often affected by intense erosional processes (mainly glacial and meteoric), thus making difficult the preservation of outcrops where fault zones with their structural and kinematic data are well preserved, and that are necessary for the goals of our task.

A large outcrop of magmatic rocks included in the indicated time-range is exposed in the western promontory of the Skagafjörður area (Figure 9). Previous geological surveys indicate clear morphological lineaments (Evert, 1975), that we have carefully checked. None resulted in a fault- zone with preserved kinematic data.

Figure 9 Part of the bedrock geological map of Iceland (1:600000) centred on the Skagafjörður area. Rocks < 2 Ma are indicated by the green colour. Thermal springs are indicated by coloured (different temperature) triangles.

As an example, the most prominent morphological lineament is indicated in the western side of the promontory. This is delimiting a deep and large glacial valley, initially tectonically controlled by a west-dipping normal fault (Evert, 1975). Reasonably, this paleo-tectonic scarp (< 2 Ma) addressed the glacial erosion, determining the fjord and, consequently, the drawback of the original fault scarp, now resulting in a morphological lineament, possibly parallel to the original fault-zone whose expression is buried under the Pliocene-Quaternary sediments of the valley.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 18 of 30

Consequently, a collection of data based only considering the age of the rocks can result time- consuming and providing a scarce amount of data. In this view, we adopted a joined criterion that is based on the investigation of those geological structures associated to thermal springs, since geothermal fluid flow circulation implies permeable rocks volumes that can be maintained only in active tectonic settings. By this, we have investigated the outcrops in the surroundings of thermal springs, as indicated in the 1:600.000 geological map (Figure 9).

Furthermore, in the southern part of the Skagafjörður area, our survey was based on the geological map provided by ISOR (Figure 10).

Figure 10 Structural and lithological map (1:50.000) provided by ISOR and considered as a basis for the southern part of the Skagafjörður area.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 19 of 30

3.3 Methodology

The adopted methodology followed the classical approach of structural geology, based on structural stations located where outcrops showing clear fracture systems < 2 Ma (i.e., associated to geothermal fluid or lava flow or affecting rocks < 2 Ma) and kinematic indicators on fault surfaces are detectable (Figure 11). The kinematic indicators that were recognized on the fault-slip zones mainly consist of slickenlines and grooves. More rarely calcite or Fe-hydroxides shear veins have been identified. The geometric features of almost vertical extensional fractures have been also reconstructed through evidences of geothermal manifestations. In particular, strike of these structures have been collected when extensional fractures clearly control gas vents, thermal springs and hydrothermal alteration zones, or controlled lava flow (lava fissure systems). The amount of data is directly connected to the width and accessibility of the outcrops. In general, five- six measurements in a 4 m2 fault-surfaces are considered sufficient if data are coherent among them. The strike of extensional fractures in geothermal springs have been measured each 2 m, whereas from 6 to 10 measurements were taken on account for lava fissures, 20 m long.

Figure 11 Examples of geological structures taken in account: A) normal fault systems with (B) kinematic indicators on the faults-slip surface; (C) strike of extensional fracture system connected to thermal springs,; (D) fault with (E) kinematic indicators, associated to thermal springs; (F) strike of extensional fractures associated to lava fissure systems (G).

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 20 of 30

3.4 Structural Stations

The locations of the investigated structural stations are indicated in the Figure 12. Stations labelled with SK, are distributed through the Skagafjörður; differently, CV-stations are referred to the Hveravellir area and V-stations to Kerlingarfjöll area.

Table 5 indicates the main structural elements investigated for each structural station. The bulk data are from geological structures linked to geothermal manifestations.

Figure 12 Location of the structural stations

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 21 of 30

Table 5 Structural stations main features. STRUCTURAL FAULTS ASSOCIATED FRACTURES FAULTS AFFECTING STATION WITH GEOTH.SPRING ASSOCIATED TO LAVA < 2 Ma GEOLOGICAL FISSURE SYSTEMS BODIES

SK1 X

SK2 X

SK3 X

SK4 X

SK5 X

SK6 X

SK7 X

SK8 X

SK9 X

SK10 X

SK11 X

SK12 X

SK13 X

SK14 X

CV1 X X

CV2 X X

CV3 X X

V1 X

V2 X

V3 X

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 22 of 30

3.5 Data analysis

Meso-fault structural and kinematic data have been inverted to focal-plane diagrams following the methodology proposed by Angelier (1979). The reliability of the results depends on the coherence among the collected data in each structural station and on the number of data. Both of these factors are however controlled by width, accessibility and weathering conditions of study outcrops. Our data are influenced by all these factors; nevertheless, a WNW-ESE extensional direction through the Skagafjörður area is suggested, although this is only a first glance on the collected data (Figure 13). A cumulative diagram is also tentatively proposed (Figure 14). Significantly, the distribution of stress axes is in agreement with the almost NNE-SSW orientation of the fissure lava swarms and alignments of thermal springs recognized at Hveravellir (Figure 15).

Figure 13 Stereonets (lower hemisphere, equal area projections) illustrating the collected data of meso- faults and related kinematic indicators for each structural station. The focal mechanism is therefore derived by field data. Symbols: 1) direction of the compressional axis; 2) direction of the intermediate axis; 3) direction of the extensional axis.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 23 of 30

Figure 14 Stereonet (lower hemisphere, equal area projections) illustrating all the collected data of meso- faults in the Skagafjörður area. Symbols are as in Figure 13.

Figure 15 Rose diagrams illustrating the preferential direction of extensional fractures connected to lava fissures and alignment of thermal springs, in two structural stations at Hveravellir. A) 14 data, circle = 30% of data; B) 10 data, circle = 36% of data.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 24 of 30

3.6 Final remarks

The main points found in the fieldwork related to fracture permeability and geological stress indicators are:

• The geological survey highlighted the occurrence of faults controlling the hydrothermal fluids flow and the location of the thermal springs. This implies tectonic activity and an active stress field in the Skagafjörður area.

• The preliminary results support for an almost E-W oriented direction of extension.

• This is also in agreement with the direction of the fissure swarms and thermal spring alignment in the volcanic area of Hveravellir.

These results can be integrated in the Ziegler et al.’s (2016a) map, without having significant differences from the present knowledge. However, the consideration of the northern-central part of Iceland as an inactive rift zone should be re-considered.

A comparison with earthquake focal mechanisms in the same area would be of crucial interest. This requires the setup of a seismometer network and the high-precision location and analysis of recorded seismic events. Furthermore, a deeper analysis of the collected data will be performed, having consideration of the local context of each structural station.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 25 of 30

4. Conclusion

The two parts of Task 4.4 Estimation of Crustal stress and estimation of fracture permeability are largely interlaced and in synthesis provide an integrated picture of the geomechanical state of an area and its suitability for geothermal exploitation. For Iceland this means that after the update it is the country with one of the largest amount of stress data records per area in the World Stress Map.

The stress state is a crucial component for the generation and opening of fractures and their orientation. These fractures may then become fluid pathways. Knowledge of their orientation and permeability is imperative for a successful geothermal exploration and application. Especially in a tectonically highly active area like Iceland the stress state may significantly fluctuate with time. Therefore it is important to know the stress state at the time of fracture generation. The estimation of fracture permeability and crustal (paleo) stress from surface indicators is required to ensure the identification of potential paleo stress fields which allow the estimation of fracture orientation. If the surface indicators are recent they in turn provide valuable information on the contemporary stress state. This is valuable information for the stimulation of reservoirs and enhancement of existing fractures. Furthermore, the stability of the borehole and adjacent faults depend on the stress state.

In summary the knowledge of the stress state and the fracture permeability is highly important for the safety and sustainability of any subsurface operations like geothermal energy production.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 26 of 30

5. References

Angelier, Jacques. 1979. Determination of the mean principal directions of stresses for a given fault population. Tectonophysics, 56(3-4), T17–T26. Angelier, Jacques. 1984. Tectonic analysis of fault slip data sets. Journal of Geophysical Research, 89(B7), 5835–5848. Ásmundsson, R.; Pezard, P.; Sanjuan, B.; Henninges, J.; Deltombe, J.-L.; Halladay, N.; Lebert, F.; Gadalia, A.; Millot, R.; Gibert, B.; Violay, M.; Reinsch, T.; Naisse, J.-M.; Massiot, C.; Azais, P.; Mainprice, D.; Karytsas, C. & Johnston, C. High temperature instruments and methods developed for supercritical geothermal reservoir characterisation and exploitation - The HiTI project Geothermics, 2014, 49, 90-98 Bell, JS. 1996. In situ stresses in sedimentary rocks (Part 1): Measurement Techniques. Geoscience Canada, 23(2), 85–100. Bell, J.S., & Gough, D.I. 1979. Northeast-southwest compressive stress in Alberta - evidence from oil wells. Earth and Planetary Science Letters, 45(2), 475–482. Célérier, Bernard. 2010. Remarks on the relationship between the tectonic regime, the rake of the slip vectors, the dip of the nodal planes, and the plunges of the P, B, and T axes of earthquake focal mechanisms. Tectonophysics, 482(1-4), 42–49. Corthésy, R., He, Guang, Gill, DE D.E., & Leite, MH M.H. 1999. A stress calculation model for the 3D borehole slotter. International Journal of Rock Mechanics and Mining Sciences, 36(4), 493–508. Crampin, Stuart, & Peacock, Sheila. 2005. A review of shear-wave splitting in the compliant crack- critical anisotropic Earth. Wave motion, 41(1), 59–77. Everts, P., 1975: Die Geologie von Skagi und der Ost-Küste des Skagafjords (Nord-Island). Sonder veröffentlichung des Geologischen Instituts der Universität Köln. Funato, Akio, Ito, Takatoshi, & Shono, Taito. 2012. Laboratory verification of the Diametrical Core Deformation Analysis (DCDA) developed for in-situ stress measurements. In: 46th US Rock Mechanics/Geomechanics Symposium. American Rock Mechanics Association. Gephart, John W., & Forsyth, Donald W. 1984. An improved method for determining the regional stress tensor using earthquake focal mechanism data: Application to the San Fernando Earthquake Sequence. Journal of Geophysical Research, 89(B11), 9305–9320. Heidbach, O.; Tingay, M.; Barth, A.; Reinecker, J.; Kurfeß, D. & Müller, B. The 2008 release of the World Stress Map 2008 Heidbach, Oliver. 2016. Scientific Technical Report 16-01 – WSM quality ranking scheme, database description and analysis guidelines for stress indicator. http://www.world-stress- map.org/fileadmin/wsm/pdfs/WSM_STR_16_01.pdf. Accessed: 07.04.2017. Heidbach, O.; Rajabi, M.; Reiter, K.; Ziegler, M. & the WSM Team World Stress Map Database Release 2016 GFZ Data Services, 2016, doi: 10.5880/WSM.2016.001 Heidbach, Oliver, Tingay, Mark, Barth, Andreas, Reinecker, John, Kurfeß, Daniel, & Müller, Birgit. 2010. Global crustal stress pattern based on the World Stress Map database release 2008. Tectonophysics, 482(1-4), 3–15. Hillis, RR, & Reynolds, SD. 2000. The Australian stress map. Journal of the Geological Society, 157(5), 915–921. Hjartarson, Á. & Sæmundsson, K. Geologic Map of Iceland. Bedrock. 1 : 600 000 Iceland GeoSurvey, Iceland GeoSurvey, 2014 Kirsch, E. G. 1898. Die Theorie der Elastizität und die Bedürfnisse der Festigkeitslehre. Zeitschrift des Vereines Deutscher Ingenieure, 42, 797–807.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 27 of 30

Mardia, K. Statistics of Directional Data: Probability and Mathematical Statistics Academic Press, 1972, 357 Mastin, L. Effect of borehole deviation on breakout orientations Journal of Geophysical Research, 1988, 93, 9187-9195 McKenzie, DP. 1969. The relation between fault plane solutions for earthquakes and the directions of the principal stresses. Bulletin of the Seismological Society of America, 59(2), 591–601. Michael, Andrew J. 1984. Determination of stress from slip data: Faults and folds. Journal of Geophysical Research: Solid Earth, 89(B13), 11517–11526. Michael, Andrew Jay. 1987. Use of focal mechanisms to determine stress: A control study. Journal of Geophysical Research, 92(B1), 357–368. Nakamura, K. 1977. Volcanoes as possible indicators of tectonic stress orientation - principle and proposal. Journal of Volcanology and Geothermal Research, 2, 1–16. Nakamura, Kazuaki, Jacob, Klaus H., & Davies, John N. 1977. Volcanoes as possible indicators of tectonic stress orientation - Aleutians and Alaska. Pure and Applied Geophysics PAGEOPH, 115(1-2), 87–112. Peška, Pavel, & Zoback, Mark D. 1995. Compressive and tensile failure of inclined well bores and determination of in situ stress and rock strength. Journal of Geophysical Research, 100(B7), 12791–12811. Plumb, Richard A, & Cox, John W. 1987. Stress directions in eastern North America determined to 4.5 km from borehole elongation measurements. Journal of Geophysical Research: Solid Earth, 92(B6), 4805–4816. Plumb, Richard A., & Hickman, Stephen H. 1985. Stress-induced borehole elongation: A comparison between the four-arm dipmeter and the borehole televiewer in the Auburn Geothermal Well. Journal of Geophysical Research, 90(B7), 5513–5521. Rivera, L., & Cisternas, A. 1991. Stress tensor and fault plane solutions for a population of Earthquakes. Bulletin of the Seismological Society of America, 80(3), 609–614. Sperner, B., Müller, B., Heidbach, O., Delvaux, D., Reinecker, J., & Fuchs, K. 2003. Tectonic stress in the Earth’s crust: advances in the World Stress Map project. Geological Society, London, Special Publications, 212(1), 101–116. Teufel, L.W. 1983. Determination of In-Situ Stress From Anelastic Strain Recovery Measurements of Oriented Core. In: SPE/DOE Low Permeability Gas Reservoirs Symposium. Society of Petroleum Engineers. Thordarson, T. & Larsen, G. Volcanism in Iceland in historical time: Volcano types, eruption styles and eruptive history. Journal of Geodynamics, 2007, 43, 118-152 Tingay, M. R. P.; Hillis, R. R.; Morley, C. K.; Swarbrick, R. E. & Drake, S. J. Present-day stress orientation in Brunei: a snapshot of 'prograding tectonics' in a Tertiary delta Journal of Structural Geology, 2005, 162, 39-49 Tingay, M.; Heidbach, O.; Davies, R. & Swarbrick, R. Triggering of the Lusi mud eruption: Earthquake versus drilling initiation Geology, Geological Society of America, 2008, 36, 639-642 Ziegler, M, Rajabi, M., Heidbach, O, Hersir, G. P., Ágústsson, K., Árnadóttir, S., & Zang, A. 2016a. The Stress Pattern of Iceland. Tectonophysics, 674, 101–113. Ziegler, M. O., & Heidbach, O. 2017a. Manual of the Matlab Script Stress2Grid. WSM Technical Report 17-02. Potsdam: GFZ German Research Centre for Geosciences. Ziegler, M. O., & Heidbach, O. 2017b. Matlab script Stress2Grid. http://doi.org/10.5880/- wsm.2017.002.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 28 of 30

Ziegler, Moritz, Rajabi, Mojtaba, Hersir, Gylfi, Ágústsson, Kristján, Árnadóttir, Sigurveig, Zang, Arno, Bruhn, David, & Heidbach, Oliver. 2016b. Stress Map Iceland 2016. http://doi.org/- 10.5880/WSM.Iceland2016. Zoback, Mary Lou. 1992. First- and second-order patterns of stress in the lithosphere: The World Stress Map Project. Journal of Geophysical Research, 97(B8), 11703–11728.

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 29 of 30

6. Appendix A

Ziegler et al. (2016a) The Stress Pattern of Iceland

Doc.nr: IMAGE-D4.04 Version: 2017.09.14 Classification: public Page: 30 of 30

7. Appendix B

Stress Map Iceland 2016

Tectonophysics 674 (2016) 101–113

Contents lists available at ScienceDirect

Tectonophysics

journal homepage: www.elsevier.com/locate/tecto

The stress pattern of Iceland

Moritz Ziegler a,b,⁎, Mojtaba Rajabi c,OliverHeidbacha,Gylﬁ Páll Hersir d, Kristján Ágústsson d, Sigurveig Árnadóttir d,ArnoZanga,b a Helmholtz Centre Potsdam, German Research Centre for Geosciences, Telegrafenberg, 14473 Potsdam, Germany b University of Potsdam, Institute of Earth and Environmental Science, Karl-Liebknecht-Str. 24–25, 14476 Potsdam-Golm, Germany c Australian School of Petroleum, The University of Adelaide, Adelaide, SA 5005, Australia d Iceland GeoSurvey (ÍSOR), Grensásvegur 9, 108 Reykjavík, Iceland article info abstract

Article history: Iceland is located on the Mid-Atlantic Ridge which is the plate boundary between the Eurasian and the North Received 13 August 2015 American plates. It is one of the few places on earth where an active spreading centre is located onshore but Received in revised form 29 January 2016 the stress pattern has not been extensively investigated so far. In this paper we present a comprehensive compi- Accepted 2 February 2016 lation of the orientation of maximum horizontal stress (S ). In particular we interpret borehole breakouts and Available online 12 February 2016 Hmax drilling induced fractures from borehole image logs in 57 geothermal wells onshore Iceland. The borehole results are combined with other stress indicators including earthquake focal mechanism solutions, geological informa- Keywords: Iceland tion and overcoring measurements resulting in a dataset with 495 data records for the SHmax orientation. The re- Stress ﬁeld liability of each indicator is assessed according to the quality criteria of the World Stress Map project. Stress pattern The majority of SHmax orientation data records in Iceland is derived from earthquake focal mechanism solutions Borehole image logs (35%) and geological fault slip inversions (26%). 20% of the data are borehole related stress indicators. In addition

minor shares of SHmax orientations are compiled, amongst others, from focal mechanism inversions and the align-

ment of ﬁssure eruptions. The results show that the SHmax orientations derived from different depths and stress indicators are consistent with each other.

The resulting pattern of the present-day stress in Iceland has four distinct subsets of SHmax orientations. The SHmax orientation is parallel to the rift axes in the vicinity of the active spreading regions. It changes from NE–SW in the South to approximately N–S in central Iceland and NNW–SSE in the North. In the Westfjords which is located far

1. Introduction Iceland is one of the few places on the Earth with an onshore divergent plate boundary (e.g. Ward, 1971; Sæmundsson, 1979; Einarsson, The regional stress pattern along divergent plate boundaries has not 1991; Einarsson, 2008; Bird, 2003). It is in a unique geological and tec- been studied extensively yet due to the inaccessibility of submerged tonic setting, where an oceanic ridge (the Mid-Atlantic Ridge) traverses Mid Oceanic Ridges. Few and scattered earthquake focal mechanism so- a (purported) mantle plume (e.g. Lawver and Müller, 1994; Wolfe et al., lutions are the only sources of stress orientation in these areas in the 1997; Allen et al., 2002). The rift zones in and around Iceland are dom- World Stress Map (WSM) database (Heidbach et al., 2008; Heidbach inated by various volcanic systems of different extents and activities et al., 2010). These indicators generally show a ridge parallel maximum (Thordarson and Larsen, 2007; Jóhannesson and Sæmundsson, 1998). horizontal stress (SHmax) orientation (Zoback et al., 1989; Zoback, 1992). Induced by the hotspot the plumbing of the volcanic systems is extend- In intraplate regions the orientation of SHmax is often parallel to the ab- ed compared to a usual divergent plate boundary (Allen et al., 2002). As solute plate motion in a ﬁrst order approximation and therefore gener- the plate boundary crosses the hotspot, it breaks up into a complex se- ally normal to the ridges and subduction zones (e.g. Richardson, 1992; ries of segments. Purely divergent segments are the Northern Volcanic Müller et al., 1992; Grünthal and Stromeyer, 1992; Zoback, 1992; Zone (NVZ) in North Iceland, and the sub-parallel Western and Eastern

Zoback et al., 1989). A systematic rotation of SHmax from ridge parallel Volcanic Zones (WVZ, EVZ) in South Iceland which are generally as- to ridge normal has been observed close to ridges in the Indian Ocean sumed to be the expression of a ridge jump (Sæmundsson, 1979; (Wiens and Stein, 1984) and at Mid Oceanic Ridges in general (Sykes, Einarsson, 1991; Einarsson, 2008). In the South, the South Iceland 1967; Sykes and Sbar, 1974). Seismic Zone (SISZ) is the connecting segment between the Reykjanes peninsula and the Eastern Volcanic Zone (Sæmundsson, 1974; ⁎ Corresponding author. Sæmundsson, 1979; Einarsson, 1991; Stefánsson et al., 2008). In the E-mail address: [email protected] (M. Ziegler). North the Tjörnes Fracture Zone (TFZ) connects the NVZ to the southern

http://dx.doi.org/10.1016/j.tecto.2016.02.008 0040-1951/© 2016 Elsevier B.V. All rights reserved. 102 M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 end of the submarine Kolbeinsey Ridge (Sæmundsson, 1974; onshore boreholes. Furthermore, extensive ﬁeld campaigns to collect Sæmundsson, 1979; Einarsson, 1991; Stefánsson et al., 2008). The geological fault slip data provide information on the current and WVZ and NVZ are joined by a transverse E–W zone across central palaeo-stress ﬁeld in Iceland as well as its temporal evolution Iceland. Outside of the immediate plate boundary, volcanism occurs in (Gudmundsson et al., 1996; Bergerat and Angelier, 1998; Garcia and the South Iceland Volcanic Zone, the Snæfellsnes Volcanic Zone and Dhont, 2005; Angelier et al., 2008; Plateaux et al., 2012). In total, the the Öræfajökull Volcanic Zone (e.g. Jakobsson, 1979; Sæmundsson, compilation of stress data records in the World Stress Map (WSM) data-

1978; Sæmundsson, 1986). base 2008 resulted in 38 data records of the contemporary SHmax orien- This volcano-tectonic setting has received a particular attention in tation and the stress regime (9 focal mechanism solutions, 5 hydro- the first compilation of the present-day crustal stress by Hast (1969). fracturing orientations, and 24 overcoring measurements, (Heidbach Since then, several researchers investigated the state of stress in differ- et al., 2008; Heidbach et al., 2010)). However, this small data set is not ent parts of Iceland. An extensive campaign of in-situ stress measure- sufficient to reveal the presumably high variability of the stress field ments from shallow overcorings was carried out by Haimson and pattern of Iceland. This is especially important since Iceland's peculiar Rummel (1982) conducted hydro-fracturing experiments in six location causes extensive interactions between tectonic and volcanic

SHmax: 19 ± 34 SHmax: 17 ± 39 Quality: A-C Quality: A-D 68° N = 188 N = 318

67°

66°

Akureyri

Egilsstaðir North American Plate 65°

Hreppar Plate

Reykjavík Method: focal mechanism breakouts drill. induced frac. Eurasian Plate borehole slotter overcoring Vík hydro. fractures geol. indicators Regime: NF SS TF U Quality: A B −25° 0 100 km C −20° D

Fig. 1. The ﬁrst comprehensive stress map of Iceland with 318 data records with A–D quality according to the World Stress Map quality criteria (Sperner et al., 2003; Heidbach et al., 2010).

Lines represent the orientation of maximum horizontal stress SHmax with the length proportional to quality. The symbols in the middle of the lines display the method used for stress determination. The colour coding is according to the stress regime with red indicating normal faulting, green indicating strike slip faulting, blue indicating thrust/reverse faulting, and black for unknown regimes. The plate boundaries according to Bird (2003) and Einarsson (2008) are indicated in grey. Two rose diagrams display the unweighted frequency distribution of the

A–CandA–D quality data respectively. Mean SHmax orientations and their standard deviations are calculated with the circular statistics of bi-polar data (Mardia, 1972). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 103 processes which influence the local stress field (e.g. Sæmundsson, 1979; Sv Gudmundsson, 2006; Andrew and Gudmundsson, 2008). In this paper we present a new comprehensive compilation of the contemporary SHmax orientation for Iceland with 495 data records (Fig. 1). In particular, we analysed 37 km of borehole acoustic image logs from 57 geothermal wells to interpret present-day stress indicators, i.e. borehole breakouts (BOs) and drilling induced fractures (DIFs). Furthermore, we revised the 38 data records from the WSM 2008 and conducted an extensive literature study to compile published DIF focal mechanism solutions and geological stress indicators, e.g. fault slip S inversions or the alignment of volcanic vents and fissures. All data re- Hmax cords are quality ranked according to the WSM quality ranking system (Zoback, 1992; Sperner et al., 2003; Heidbach et al., 2010). We identify the regional pattern of the SHmax orientation by four different stress provinces with different mean SHmax orientations on Iceland.

SHmax

2. Stress data compilation BO The ﬁrst comprehensive compilation of the contemporary SHmax orientation was made by Sbar and Sykes (1973) who mapped the stress S pattern in North America. This effort was later institutionalised by hmin Zoback et al. (1989)) in the framework of the WSM project (e.g. Müller et al., 1992; Heidbach et al., 2010). In the literature there are several methods to determine the orientation of SHmax in a rock volume (Ljunggren et al., 2003; Zoback et al., 1989; Zang and Stephansson, 2010). However, these different methods may result in different orientations due to the depth of the phenomena, different reliability, or su- perposition of different forces at different scales (Heidbach et al.,

2007). Hence, comparison between the SHmax from different indicators have received a particular attention to establish a quality ranking Fig. 2. A vertical borehole section with stress indicator pairs. Top: Drilling induced scheme for the WSM database (Zoback and Zoback, 1991; Zoback, fractures (DIFs) are narrow vertical fractures which indicate the orientation of SHmax. 1992; Zoback et al., 1989; Sperner et al., 2003; Heidbach et al., 2010). Bottom: Borehole breakouts (BOs) are broad vertical widened zones of the borehole which indicate the orientation of S . These two features occur diametrically on both Following this scheme each data record is assigned a quality from A (re- hmin sides of the borehole wall. liability of orientation ±15°), B (±15–20°), C (±20–25°), D (±25–40°) up to E (N±40°) (Heidbach et al., 2010). A detailed description of the WSM quality ranking scheme for individual stress indicators can be 2.1. Borehole data found in Zoback (1992), Sperner et al. (2003)), and Heidbach et al. (2010). The possibility to determine the in-situ stress orientation from fail- Our stress data compilation extends from 62° to 68° northern ure of borehole walls was ﬁrst recognised by Bell and Gough (1979) in latitude and from −11° to −26° longitude. The image log data from Alberta, Canada. They showed that if the stresses around a borehole ex- the 57 geothermal wells resulted in 36 new A–D stress data records. ceed the strength of the rock, some pieces of the borehole wall spall off

In addition, we estimated 17 SHmax orientations from crater rows of fis- and the borehole is elongated in one orientation. According to Kirsch sure eruptions of different volcanic systems. Furthermore, an extensive (1898) and Scheidegger (1962) the highest stresses around a circular literature review resulted in 374 new stress data records which are hole are encountered perpendicular to the orientation of maximum mainly from focal mechanism solutions of earthquakes. These new compression (SHmax). These resulting broad elongated zones of so called data records are from different earthquake catalogues such as the Global borehole breakouts (BO, see Fig. 2) indicate the orientation of minimum CMT (Ekström et al., 2012), (Dziewonski et al., 1981), Geofon Potsdam horizontal stress (Shmin) which is perpendicular to SHmax under the as- (Centre, 1993) and Zurich Moment Tensors. Furthermore data records sumption that the vertical stress (Sv) is one of the principal stresses were included from published papers by Angelier et al. (2004), (Bell and Gough, 1979). Batir (2011), Bergerat et al. (1990), Bergerat and Angelier (1998), Furthermore, if the minimum circumferential stress around a bore- Bergerat et al. (1998), Bergerat and Plateaux (2012), Bjarnason and hole wall is smaller than the tensile strength of the rock, drilling induced Einarsson (1991), Einarsson (1979) Einarsson (1987), Forslund fractures (DIF, see Fig. 2) occur (Aadnoy, 1990; Aadnoy and Bell, 1998). and Gudmundsson (1991), Garcia et al. (2002), Garcia (2003), Green Therefore drilling induced fractures are recognised as an indicator for et al. (2014), Gudmundsson et al. (1992), Gudmundsson (1995), the orientation of SHmax as well (Wiprut et al., 1997; Bell, 1996; Gudmundsson et al. (1996); Jakobsson (1979), Jefferis and Voight Sperner et al., 2003). (1981), Hagos et al. (2008), Haimson and Voight (1977), Keiding et al. Acoustic image logs provide a picture of the borehole wall based on (2009), Khodayar and Franzson (2007), Kristjánsdóttir (2013), Lund acoustic contrast of borehole wall and fluids. Borehole breakouts usually and Slunga (1999), Lund and Bödvarsson (2002), Nakamura (1977), appear as broad vertical zones of a low acoustic amplitude on opposite Plateaux et al. (2014), Rögnvaldsson and Slunga (1994); Roth et al. sides of the borehole wall (separated by 180°) while drilling induced (2000), Schäfer and Keil (1979), Sigmundsson et al. (2005) Sigurdsson fractures are indicated by narrow vertical zones of low amplitude (1970), Soosalu and Einarsson (1997), Stefánsson (1966), Tibaldi et al. (Fig. 3). A pair of DIFs or BOs on opposite sides of the borehole wall is (2013), and Villemin et al. (1994). The detailed dataset of the Iceland considered as a single feature. Since the shapes of BOs and DIFs depend stress map is provided in the supplementary material. In the following on rock strength and the elastic properties of rocks and these features sections we briefly describe each individual stress indicator used for are time dependent, incipient breakouts form at the initial stage of the the Iceland stress dataset. formation of borehole breakouts (Aadnoy and Bell, 1998). 104 M. Ziegler et al. / Tectonophysics 674 (2016) 101–113

Borehole Breakouts (BO) Drilling induced tensile fractures (DIF) 0° 90° 180° 270° 360° 0° 90° 180° 270° 360° 540.0 1085.0

541.0 1086.0 Depth [m] Depth [m]

542.0 1087.0

543.0 1088.0

lowAmplitude high

Fig. 3. Borehole related stress indicators in acoustic image logs. Left: Borehole breakouts (BOs) in well ST-16 Sigtún close to Akureyri. The inferred overall orientation of SHmax from BOs is

127° in this well. Right: Drilling induced fractures (DIFs) in well HJ-20 Hjalteyri close to Akureyri. The inferred overall orientation of SHmax from DIFs is 144° in this well. The location of the two wells is shown in Fig. 4.

Iceland's volcano-tectonic setting results in large geothermal re- vertical (b10°) which still allows the interpretation of stress related fea- sources which are extracted by various boreholes (Ragnarsson, 2015). tures in every stress regime (Mastin, 1988; Tingay et al., 2005; Peška In 2002 through 2015 the Iceland GeoSurvey (ÍSOR) ran borehole and Zoback, 1995). image logs in 57 geothermal and scientiﬁc boreholes mainly in the 27 boreholes contained at least one BO or one DIF (Table 1, Fig. 4). In South Iceland Lowlands and around Akureyri and Kraﬂa in the North the case that both BOs and DIFs are found in the same well, the indepen-

(see Fig. 4 for locations). From these data we collected and analysed dently inferred SHmax orientations are generally in good agreement with 37 km of acoustic image logs. Most of them are slightly deviated from each other (Table 1). In addition to the newly analysed borehole images 3 BOs and 1 DIF from published articles were included. In the analysed boreholes stress indicators are mainly found between the surface and 1 km depth. Some few BOs/DIFs are located in deeper sections of the boreholes with a maximum depth of 2.34 km in well RN-34 on the Reykjanes peninsula (Fig. 4). Thus borehole stress data bridge the gap HJ-20 between shallow stress indicators from geological data and focal mech- Akureyri anism solutions at greater depth. Table 1 shows the results of image log interpretation and observed ST-16 BOs/DIFs in the studied wells. 11 data records have an A–C quality and 25 data records have a D quality. The high number of low quality data records is partly related to the challenges of well-logging in a high temperature and igneous environment resulting in a partly poor image quality. Special tools adapted to high temperatures are required and Reykjavík can only remain in the well for a short time period (Ásmundsson et al., 2014). In addition, in some of the studied wells image tools RN-34 were not centralised and produced low quality images with numerous vertical artefacts which do not allow a reliable detection of BOs and DIFs.

2.2. Focal mechanism solutions

Fig. 4. The location of geothermal boreholes with acoustic image logs. The black and grey Focal mechanism solutions of earthquakes have been used to infer triangles denote the location of boreholes with and without stress indicators (based on our image log analysis) respectively. The white triangles show the location of borehole HJ-20 stress information, both orientation and relative magnitudes, in the Hjalteyri, ST-16 Sigtún (Fig. 3) as well as in RN-34 Reykjanes. deeper part of the earth's crust which is beyond common drilling M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 105

Table 1 Stress indicators from the analysed acoustic borehole images of A–D Quality. All the information required for the WSM quality ranking is included in the Table. Azimuth: Interpreted orientation of SHmax. Number: The amount of recognised feature pairs (BOs or DIFs) in a single well. S.D.: Standard deviation calculated according to the circular statistics of bi-polar data by (Mardia, 1972) with a weighting depending of the length (short: L) of the feature. Length: The added length of the fractured borehole sections. Top and Bottom: The depth of the uppermost and lowermost stress indicator found in the borehole. Depth: The mean between top and bottom. Date: Date of the tool run.

Borehole Latitude Longitude Azimuth Type Depth Quality Location Date Number S.D. Length Weighting Top Bottom ID [km] [m] [m] [m]

HH-08 63.425023 −20.25904 133 BO 1.05 C Vestmannaeyjar 20050415 11 13 22 L 789 1719 RN-34 63.83951 −22.660869 36 BO 1.95 C Reykjanes 20150328 15 12 25 L 1412 2628 RN-34 63.83951 −22.660869 47 DIF 2.45 B Reykjanes 20150328 20 9 40 L 2317 2612 KH-34 63.98881 −20.44006 67 BO 0.04 D Kaldárholt 20050322 1 0 2 L 38 40 KH-34 63.98881 −20.44006 109 DIF 0.2 D Kaldárholt 20050322 2 2 3 L 55 390 SO-01 63.995165 −21.13729 47 DIF 0.32 D Sogn/Ölfus 20050322 3 13 6 L 314 325 HE-21 64.008906 −21.3438 41 BO 1.67 D Hellisheiði 20060215 11 14 16 L 1608 1748 HE-21 64.008906 −21.3438 67 DIF 1.35 B Hellisheiði 20060215 53 14 123 L 912 1812 HE-58 64.033132 −21.376734 35 DIF 1.9 D Hellisheiði 20150830 3 15 5 L 1609 2200 HN-01 64.026124 −21.45102 45 BO 0.9 C Hellisheiði 20050405 20 22 26 L 866 977 HN-01 64.026124 −21.45102 44 DIF 0.85 D Hellisheiði 20050405 7 18 10 L 768 977 HK-15 64.041 −20.81377 8 BO 0.1 C Grímsnes 20060303 33 15 25 L 37 183 HN-12 64.044597 −21.38636 84 DIF 1.5 D Hellisheiði 20101021 7 21 11 L 1152 1878 HN-16 64.045106 −21.3862 86 DIF 2.06 D Hellisheiði 20101018 6 12 9 L 2021 2187 NJ-28 64.098521 −21.270345 107 DIF 1.05 D Nesjavellir 20150625 5 9 11 L 1029 1057 HF-01 64.391916 −15.34195 151 DIF 0.6 D Hoffell 20130221 10 11 17 L 424 805 ASK-29 64.393293 −15.343563 130 BO 0.11 D Hoffell 20120926 6 16 6 L 103 123 ASK-57 64.393898 −15.34267 4 BO 0.28 D Hoffell 20120926 1 0 1 L 283 284 ASK-122 64.393778 −15.33175 65 DIF 0.35 D Hoffell 20150924 7 14 13 L 338 375 HO-02 65.04501 −22.77176 60 BO 0.36 D Stykkishólmur 20070215 1 0 4 L 366 370 ST-16 65.5519 −18.07022 127 BO 0.35 C Sigtún/Eyjafjörður 20050126 28 9 37 L 111 671 ST-16 65.5519 −18.07022 140 DIF 0.4 D Sigtún/Eyjafjörður 20050126 5 7 16 L 329 508 BO-3 65.562966 −18.10464 107 DIF 0.07 D Botn 20130122 3 13 10 L 60 80 KV-01 65.692163 −16.81934 29 BO 1.43 D Krafla 20060803 1 0 1 L 1435 1437 KV-01 65.692163 −16.81934 164 DIF 1.43 D Krafla 20060803 2 8 2 L 1432 1435 K-18 65.702026 −16.73063 17 BO 0.74 D Krafla 20081118 2 4 6 L 733 750 HJ-17 65.855115 −18.2105 151 DIF 0.15 D Hjalteyri 20020221 2 11 2 L 122 170 HJ-13 65.855337 −18.21303 145 DIF 0.06 D Hjalteyri 20020220 1 0 3 L 62 65 HJ-20 65.856089 −18.21142 141 BO 1 D Hjalteyri 20050202 4 8 12 L 784 1176 HJ-20 65.856089 −18.21142 144 DIF 0.75 A Hjalteyri 20050202 60 11 136 L 352 1346 HJ-15 65.859457 −18.21754 154 DIF 0.2 D Hjalteyri 20020223 1 0 2 L 204 207 ARS-32 65.931479 −18.33783 163 BO 0.75 D Árskógsströnd 20060608 6 19 6 L 668 842 ARS-32 65.931479 −18.33783 173 DIF 0.55 C Árskógsströnd 20060608 17 14 36 L 206 713 SK-28 65.997822 −19.33668 143 BO 0.5 C Hrolleifsdalur 20051008 55 25 137 L 240 821 SD-01 66.127507 −18.96229 146 BO 0.45 D Skarðdalur/Tröllaskagi 20100925 2 3 3 L 430 537 SD-01 66.127507 −18.96229 140 DIF 0.5 B Skarðdalur/Tröllaskagi 20100925 20 11 69 L 319 687

plans (Sbar and Sykes, 1973; Gephart and Forsyth, 1984; Zoback, 1992; couple mechanism (Nettles and Ekström, 1998; Ekström, 1994). There-

Heidbach et al., 2010). The orientation of SHmax is estimated from the fore events which can be spatially and temporally attributed to a volca- principal strain axes of the double couple components of the focal nic eruption or rifting event are assigned E quality. However, seismic mechanism (McKenzie, 1969; Barth et al., 2008). However, these axes events which are only located at a volcano but cannot be linked to an are not necessarily reliable proxies for the stress axis orientation eruption remain with a quality C. In the Vatnajökull area several thrust (McKenzie, 1969; Célérier, 2010; Heidbach et al., 2010). Therefore, sin- faulting events were recorded during an inter-volcanic period. Nettles gle focal mechanism solutions are never eligible for a quality better than and Ekström (1998) and Einarsson (1991) suggest that these events C in the WSM database (Heidbach et al., 2010; Barth et al., 2008). A are a movement of the Barðabunga caldera rim. Hence they are not di- stress determination though the averaging of several focal mechanism's rectly temporarily related to a volcanic eruption and assigned the qual- P, B, and T axes (FMA) is less reliable and is hence assigned D quality. ity C. Between 1994 through 2007 250,000 seismic events were recorded Furthermore, the phenomenon of induced seismicity in geothermal by the Iceland Meteorological Office with 11 events of M N 5(Einarsson, reservoirs is reported in Iceland (Flóvenz et al., 2015). The stress 1991; Einarsson, 2008; Jakobsdóttir et al., 2002; Jakobsdóttir, 2008; field in geothermal or hydrocarbon reservoirs can change significantly Einarsson et al., 1977; Keiding et al., 2009). The detection threshold in due to depletion and/or reinjection (Segall and Fitzgerald, 1998; this time frame has been between M l =2 and M l =0 depending on Martnez-Garzón et al., 2013). Hence, focal mechanisms of seismicity lo- the region (Jakobsdóttir, 2008). Focal mechanism solutions were public- cated in the vicinity or within active reservoirs are prone to exhibit a ly available for only a fraction of the recorded seismic events. perturbed stress state compared to the virgin in-situ stress state. There- Presumably especially in Iceland many seismic events are related to fore seismic events which are in spatial and temporal proximity to e.g. volcanic eruptions or dyke intrusions and thus are potentially spatially dams or geothermal power plants are identified as potentially induced and temporally restricted manifestations of the stress field (e.g. and are assigned E quality as well. Roman et al., 2004; Sánchez et al., 2004; Einarsson, 1991; White et al., In addition to single focal mechanism solutions (FMS) or an average 2011). Hence they do not necessarily represent the long-term stress of FMS (FMA), inversions of focal mechanisms (FMF) can be performed field but only short-term fluctuations of a perturbed regional stress (e.g. Gephart and Forsyth, 1984; Angelier, 1984). Generally results from field. In addition, such events may have a low double-couple and high inversions provide high quality (A or B) stress data records (e.g. Keiding compensated linear vector dipole (CLVD) component (Nettles and et al., 2009; Kristjánsdóttir, 2013). However, the inversions of focal Ekström, 1998). That means the main strain component is due to an in- mechanism solutions performed by Bergerat et al. (1998), Garcia et al. flation or deflation above some pressure source in contrast to a double- (2002), Angelier et al. (2004) and Plateaux et al. (2014))showthe 106 M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 existence of two spatially or temporally different local stress fields. Due 2.4. Vent alignments to the high quality of the inversions they are included in the database anyway but assigned E quality since these two stress fields cannot be Nakamura (1977)) was one of the first to recognise the alignment of distinguished. volcanic vents, eruptive fissures, and dykes (GVAs) as stress indicators. GVAs are always related to volcanic eruptions which tend to be easier to date compared to fault slip, since often the age of volcanic eruptions are 2.3. Geological indicators known. The high volcanic activity in Iceland allows inclusion of young eruptive Geological indicators of past fault slip events can also provide infor- fissures, vent alignments and dykes from the Quaternary. We therefore mation on the stress state and the data reliability is equal in comparison included 17 GVAs produced by recent volcanic activities (even in historic to other methods (Sperner et al., 2003). However, to prevent a mix of times, Thordarson and Larsen, 2007). The data originates in geological palaeo-stress and contemporary stress data records, geological indica- mapping campaigns and is also displayed in the Geologic Map of tors are generally not allowed to be older than Quaternary, i.e. not Iceland — Bedrock (Hjartarson and Sæmundsson, 2014). Table 2 shows older than 2.85 Ma (Zoback, 1992). Sometimes the age of a fault slip the stress orientations inferred from eruptive fissures mainly deduced or dyke intrusion is measured (e.g. radiocarbon dating), the relative from geologic mapping also presented in the map by Hjartarson and age deduced by the stratigraphy (e.g. in Bergerat and Angelier, 1998), Sæmundsson (2014). They are quality ranked according to the WSM or the maximum age of the rock is otherwise known (e.g. in Bergerat criteria shown in Table 3. and Plateaux, 2012). If this is not the case geological maps can provide information of the age of the indicators. Note that the rule applies to 2.5. Further stress indicators the age of the fault slip and not the age of the rock, in case where they can be distinguished. In total 25 overcoring (OC) stress measurements are available In the new Iceland Stress Map a large amount of data records are throughout Iceland. Due to their shallow depth (0–30 m) the inferred provided by geological indicators, i.e. stress tensors inferred from fault stress state may be highly influenced by local topography or strength slip data. This is due to the extensive work on the stress inversions of contrasts. Therefore the data records are assigned to E quality. Previous fault data (GFI) by J. Angelier, F. Bergerat & A. Guðmundsson undertaken data records from the WSM 2008 which were assigned a different qual- in Iceland (e.g. Bergerat et al., 1990; Bergerat et al., 1998; Angelier et al., ity according to an outdated version of the ranking scheme were 2004). updated.

We assessed geological indicators (GFI) following strictly the WSM In addition 9 SHmax orientations are available from hydraulic- quality ranking scheme (Heidbach et al., 2010; Sperner et al., 2003). fracturing (HF). Previously listed HF data records were revisited and Zoback (1992) discusses the possible necessity to alter the age restriction assigned a quality according to the most recent quality ranking scheme. according to the tectonic setting. In case two or more different temporally successive stress states are inferred in the exact same location and 3. Stress map & pattern of Iceland both originate in the Quaternary only the youngest can be taken into account in this compilation (as is the case in e.g. Bergerat and Angelier, The new compilation of stress data for Iceland has 495 data records 1998). In several instances, stress indicators from fault slip data are in with 318 having A–Dand188A–C quality (Table 4, Figs. 1 & 5). Most of close proximity to similarly oriented stress indicators which are definite- the A–D quality data records are from focal mechanism solutions (35%) ly from the currently active stress field (e.g. a borehole breakout or focal and geological fault inversions (26%). Borehole related indicators (BOs, mechanism solution). Their similar orientation is at least an indicator DIFs, HFs) have a share of 20% while the alignments of volcanic vents, that the age restriction also applies in Iceland. Even though local stress fissures and craters contribute with 8%. The inversion of several focal perturbations do occur due to the presence of local structures (Rajabi mechanism solutions make up 7% of the dataset. et al., 2016; Heidbach et al., 2007) hence different SHmax orientations in 56% of the data records are from the depth range of 0 to 1.25 km close spatial proximity must not be judged as unreliable. (Fig. 5). These are mainly geological stress indicators which are either

Table 2 Newly included volcanic vent and fissure alignments (GVAs) which are also shown in Hjartarson and Sæmundsson (2014). The required information for the World Stress Map as well as the age of the most recent eruption of the associated (central) volcano is listed. Number: The amount of parallel vent/fissure alignments. Vents: The overall number of vents/fissures which are considered. In case of parallel alignments the standard deviation is calculated according to the circular statistics of bi-polar data by Mardia (1972).

Latitude Longitude Azimuth Quality Location Number S.D. Vents Type Last eruption/rifting event

63.43 −20.2 45 C Vestmannaeyjar 1 5 Vents 1973 A.D.a 63.82 −18.83 18 C Eldgjá (South) 1 6 Fissures 934–940 A.D.a 63.9 −21.8 56 B Reykjanes 4 5 21 Vents 1231 A.D.b 63.94 −18.65 43 C Eldgjá (Middle) 1 5 Fissures 934–940 A.D.a 64.1 −18.3 35 C Eldgjá (North) 1 9 Fissures 934–940 A.D.a 64.25 −18.6 33 B Veiðivötn 4 13 67 Fissures 1477 A.D.a 64.29 −20.84 43 C Þjófahraun 1 11 Fissures 3600 B.P.c 64.4 −20.5 47 C Langjökull 2 3 10 Vents 950 A.D.c 64.75 −16.6 30 C Kverkfjöll 1 7 Fissures 9000 B.P.b 64.8 −17.3 22 B Dyngjuháls 3 6 28 Fissures 1902–1903 A.D.e 65 −17.15 29 C Trölladyngja/Frambruni 1 8 Fissures 1300 A.D.f 65.15 −16.6 21 C Askja 1 14 Vents 1961 A.D.b 65.4 −16.8 9 C Fremrinámur 1 9 Vents 4000 B.P.d 65.5 −16.45 8 C Nýjahraun 2 6 16 Fissures 1874–75 A.D.d 65.6 −16.8 8 B Reykjahlíð 4 2 16 Fissures 1975–1984 A.D.a 65.7 −16.8 6 C Kraﬂa 1 10 Fissures 1975–1984 A.D.a 65.9 −16.35 11 C Hólssandur 1 7 Fissures Holoceneg aThordarson and Larsen (2007), bHaﬂidason et al. (2000), cSinton et al. (2005), dSigurdsson and Sparks (1978), eBjörnsson and Einarsson (1990), fHjartarson (2003),andgHjartarson and Sæmundsson (2014). M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 107

Table 3 The World Stress Map quality ranking scheme version 2008 for borehole breakouts and drilling induced fractures from image logs and volcanic vent alignments (Heidbach et al., 2010). S.D. = standard deviation.

Stress indicator A B C D E

SHmax ±15° SHmax ±15–20° SHmax ±20–25° SHmax ±25–40° SHmax N±40° Borehole ≥10 distinct breakout zones ≥6 distinct breakout zones and ≥4 distinct breakouts and b4 distinct breakouts or b20 m Wells without breakouts and combined length ≥ 100 m combined length ≥ 40 m in a combined length ≥ 20 m with combined length in a single reliable breakouts in a single well with S.D. ≤ 12° single well with S.D. ≤ 20° S.D. ≤ 25° well with S.D. ≤ 40° or S.D. N 40° Drilling induced ≥10 distinct fracture zones in a ≥6 distinct fracture zones in a ≥4 distinct fracture zones in a b4 distinct fracture zones in a Wells without fractures single well with a combined single well with a combined single well with a combined single well or a combined fracture zones or length ≥ 100 m and S.D. ≤ 12° length ≥ 40 m and S.D. ≤ 20° length ≥ 20 m and S.D. ≤ 25° length b 20 m and S.D. ≤ 40° S.D. N 40° Volcanic vent ≥5 Quaternary vent ≥3 Quaternary vent Single well-exposed Volcanic alignment inferred alignment alignments or “parallel” dikes alignments or “parallel” dikes Quaternary dike or single from b5 vents with S.D. ≤ 12° with S.D. ≤ 20° alignment with ≥5 vents

exhumed faults or surface manifestations of the stress field. Most bore- the standard deviation for A–D quality data is between 19° and 29° hole indicators are in the same depth range. Even some very shallow which is comparable to other regional stress investigations (e.g. focal mechanism solutions and inversions of several focal mechanisms Pierdominici and Heidbach, 2012; Reiter et al., 2014; Reinecker et al., are located in that depth range. Around 2.5 km depth stress indicators 2010). from deep boreholes and earthquake related indicators are equally Generally the standard deviation of the mean SHmax orientation of abundant. Below that depth, focal mechanism solutions of seismic stress data records with A–C quality is found to be within ±25° (see events are the only available stress indicators. The peak of events rose diagrams in Figs. 1, 6–9). If D quality data records are included around 10 km is artificial because many focal mechanisms of small mag- the mean SHmax orientation changes by ≤4°. The standard deviation in- nitude seismic events are assigned this depth as a default value if the creases by approximately 5° reflecting that D quality data introduces depth cannot be estimated otherwise. more noise to the dataset. Therefore D quality data should not be used Some stress indicators (e.g. focal mechanism solutions, fault inver- individually for a local stress field analysis. Surprisingly the standard de- sions) allow characterisation of the Andersonian faulting type of the viation decreases by 1° with the introduction of 11 D quality data re- stress field (Anderson, 1905; Anderson, 1951). The method to derive cords in North Iceland. 10 of these data records are from boreholes the type of faulting is described by (Zoback, 1992). Fig. 5 shows that and their quality depends on the short length of the feature and/or miss- normal faulting prevails at the surface. However, within the first ing information on the standard deviation. These circumstances show kilometre this changes. In the following topmost 10 km a strike slip re- that a well-picked distinct single feature in a borehole provides valuable gime is dominant. With a further increase in depth the normal faulting information on the orientation of SHmax. regime prevails. Indicators for a reverse faulting regime are observed in The types of available stress indicators varies in the different subsets. all depths in a relatively small abundance. Nevertheless, around 1 km While all types of indicators are represented close to the plate boundary, and 10 km depth they have a significant share. in the Westfjords and Eastfjords the stress state is mainly derived from

The prevailing orientation of SHmax in Iceland inferred from A–C geological indicators and boreholes. That means that in those regions quality ranked data records is determined according to circular statistics the information on the stress ﬁeld is based mainly on shallow data. of bipolar data (Mardia, 1972) which shows a mean SHmax orientation of Apart from lateral variations of the orientation of SHmax, the possibil- 18° ± 35° for the entire dataset. A closer look at Fig. 1 demonstrates four ity of a vertical layering exists (Cornet and Röckel, 2012; Gudmundsson, predominant regional orientations of SHmax. In the Southwest and the 2002; Heidbach et al., 2007). In some regions, mainly sedimentary ba- Southern Iceland Lowlands SHmax is oriented approximately NE–SW sins, moderate (Reiter et al., 2014; Reiter and Heidbach, 2014)orsignif- (Fig. 6). In the Northern Volcanic Zone (north of the Vatnajökull glacier) icant (Röckel and Lempp, 2003; Roth and Fleckenstein, 2001; Rajabi which is presently the active rift zone, SHmax has almost N–Sorientation et al., 2016) stress rotations occur with depth. For example, Rajabi (Fig. 7). SHmax is rotated by about 20° to NNE–SSW in the easternmost et al. (2016) reported signiﬁcant rotation of the SHmax orientation with part of Iceland (Fig. 7). In Northern Iceland SHmax is rotated from the depth in the Clarence-Moreton Basin of eastern Australia due to pres- N–S orientation in the Northern Volcanic Zone to a predominant ence of geological structures including intrusions of igneous rocks into

NNW–SSE orientation (Fig. 8). Finally in the Westfjords the SHmax sedimentary successions. trend is approximately NW–SE oriented (Fige. 9). For these four subsets It is indicated by the propagation of dykes, that such a layering also exists in Iceland on a local scale (Gudmundsson, 2002; Gudmundsson,

2003). To find regional-scale depth-dependent differences in the SHmax Table 4 orientation we compiled surface data (GFI, GVA) as well as intermediate An overview of the quality and type of all stress indicators in the designated area (N: 62°– (0.2–2 km) borehole indicators (BO, DIF) and deep (2–20 km) focal – 68°, W: 11° 26°). They include the revisited and re-ranked data from the WSM 2008 as mechanism solutions (Fig. 5). In all areas where more than one type of well as the newly analysed data from acoustic image logs, the alignments of volcanic craters and fissures, and data records from literature research. indicator is available, the orientation of SHmax remains consistent with depth which highlights the independence from the type of stress indica- Quality tor and the vertical homogeneity of SHmax throughout the crust. Thus a A B C D E Total potential regional-scale depth-dependency of SHmax is not observed. Type FMF 15 7 ––14 36 FMS ––63 22 90 175 4. Discussion FMA ––– 9 – 9 –– BO 6133049 This study presents the first comprehensive and systematic compila- DIF 1 3 1 15 1 21 HF – 12 6 – 9 tion of the present-day tectonic stress in Iceland where all results are OC ––– – 25 25 ranked based on a quality ranking scheme for the in-situ stress state. GFI 1 11 40 63 14 129 A high density of data records is achieved on the Reykjanes peninsula, GVA 1 11 25 2 3 42 in South Iceland, East Iceland, and the Akureyri area and Tjörnes Total 18 33 137 130 177 495 Fracture Zone in North Iceland (Fig. 1). Few or no data records are 108 M. Ziegler et al. / Tectonophysics 674 (2016) 101–113

Number of indicators A-D 0 50 100 150 200 250 0 0

2.5 2.5

5 5

7.5 7.5

10 10 Depth [km] Depth [km] 12.5 12.5

15 Method: Regime: 15 FMS Normal 17.5 FMF Strike-Slip 17.5 BO/DIF/HF GVA Reverse 20 GFI 20

100 0 %255075

Fig. 5. The depth distribution of the 318 stress indicators (A–D quality) is displayed here. The cluster of seismic events around a depth of 10 km is biased since many of the events with an uncertain depth were assigned this depth arbitrarily. The data is colour coded according to the type of the indicator. The width of the bar indicates the quality of the data from A (thick/left) to D (thin/right). Please note that the colour coding is independent of the width of the bar in this plot. In addition the variation of the stress regime with depth is shown on the right side. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) available around Hofsjökull, Langjökull in the western Highlands, on the East and the North American plate to the West (Einarsson, 2008; Bird, Snæfellsnes peninsula, and in the Westfjords (Fig. 1). Based on the avail- 2003). At the Reykjanes peninsula where the ridge makes landfall able data from this compilation the orientation of the maximum com- SHmax remains mostly ridge parallel (Figs. 1 & 6). This pattern of SHmax pressive stress (SHmax) in Iceland is organised in four subsets and is is consistent with observations by e.g. Sykes (1967) and Wiens and consistent with the main plate boundaries in the region. This highlights Stein (1984) who show ridge parallel SHmax close to the spreading cen- the role of different plate boundary forces in the stress pattern of Iceland tre along divergent plate boundaries in general and especially in the (Fig. 10). Furthermore the highly dynamic geological setting of Iceland Indian Ocean. is reflected in the stress field by effects of eruptions, geothermal activity, Ridge parallel stress is also indicated further to the North along the and rifting events. WVZ (Fig. 1). The western boundary of the Hreppar microplate is at the WVZ and its northern boundary is the quietest Central Iceland Vol- 4.1. Regional stress pattern canic Zone (CIVZ) which is not represented by stress indicators here (Einarsson, 2008).

In the South-West a ridge parallel SHmax orientation can be observed In the South the Hreppar microplate meets the Eurasian plate at the along the Reykjanes Ridge (Fig. 1) which has the Eurasian plate to the transform SISZ (Einarsson, 2008). This is one of the two areas with the

WVZ Iceland Reykjavík

RPR 64° SISZ Hella

SHmax: 36 ± 23 SHmax: 38 ± 29 Quality: A-C Quality: A-D N = 73 N = 126 63.5°

Vestmannaeyjar

−23° −20° km −19° 050

Fig. 6. The orientation of SHmax (A–D quality) on the Reykjanes peninsula ridge (RPR), the transform South Iceland Seismic Zone (SISZ), and parts of the Western Volcanic Zone (WVZ). Legend is the same as in Fig. 1. M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 109

−17˚ −16˚ −15˚ S : 9 ± 19 Hmax SHmax: 8 ± 25 Quality: A-C Quality: A-D N = 36 N = 78 66.5˚

66˚

65.5˚

Egilsstaðir

65˚

64.5˚

Iceland km 02040 Höfn

Fig. 7. The orientation of SHmax (A–D quality) in the Eastern Highlands/Northern Volcanic Zone and the East Fjords. Legend is the same as in Fig. 1. Note that mainly surface geological indicators are available in this area.

largest seismic events (M = 7.2) in Iceland (e.g. Stefánsson et al., 2000; In the North, the TFZ connects the NVZ with the Kolbeinsey Ridge

Bergerat and Angelier, 2001). In the SISZ the SHmax isNNEtoNE(Fig. 6) north of Iceland (Sæmundsson, 1974; Sæmundsson, 1979; Einarsson, which is consistent with the surface ruptures of large earthquakes (e.g. 1991; Garcia, 2003; Stefánsson et al., 2008). The spreading is distributed Árnadóttir et al., 2003; Einarsson, 2008). between the Dalvík Zone (DZ), the Húsavík-Flatey-Zone (HFZ) and the In the North-East of the SISZ the EVZ and the NVZ are the currently Grimsey Oblique Rift (GOR) (Sæmundsson, 1974). This is the second active rift zones (Einarsson, 2008). Most of the rifting events (Laki, area with large magnitude seismic events in Iceland (Jakobsdóttir,

Eldgjá, Kraﬂa ﬁres, Holuhraun) and volcanic eruptions (Grimsvötn, 2008) and shows a NNW–SSE trend for the SHmax orientation which is Gjálp, Askja, Hekla, Barðabunga) are in these two zones (Sigmundsson mainly inferred from focal mechanism solutions (Fig.8). et al., 2015; Thordarson and Larsen, 2007). This activity is related to In the Westfjords which are the oldest part of Iceland (10–16 Ma, the current location of the centre of the hotspot which is considered Moorbath et al., 1968; McDougall et al., 1984) and also partly on the to be beneath the Vatnajökull glacier at the transition from the EVZ to Snæfellsnes peninsula a rotation of SHmax from ridge parallel towards the NVZ (e.g. Wolfe et al., 1997; Ito et al., 2003), Fig. 10). The SHmax is ridge perpendicular is observed (Figs. 9 & 1). This rotation is interpreted found to follow the orientation of the EVZ and NVZ which are consid- as the transition from the ridge parallel stress orientation to the com- ered as the plate boundary from NE–SW in the South to N–Sinthe mon intraplate stress orientation (Wiens and Stein, 1984; Sykes, 1967; North (Fig. 1). This pattern is also observed at some distance along the Sykes and Sbar, 1974; Müller et al., 1992; Grünthal and Stromeyer, Icelandic east coast (Fig. 7). 1992; Gudmundsson et al., 1996). This rotation is expected in some 110 M. Ziegler et al. / Tectonophysics 674 (2016) 101–113

Iceland GOR 66.5˚

HFZ

DZ Húsavík 66˚

Sauðárkrókur Akureyri

SHmax: 155 ± 22 SHmax: 151 ± 21 Quality: A-C Quality: A-D 65.5˚ N = 28 N = 39

km 02040

−20˚ −19˚ −18˚ −17˚

Fig. 8. The orientation of SHmax (A–D quality) in Northern Iceland. Displayed is the Tjörnes Fracture Zone with the Grimsey oblique rift (GOR), the Húsavík-Flatey-Zone (HFZ), and the Dalvík Zone (DZ). Legend is the same as in Fig. 1.

SHmax: 136 ± 12 SHmax: 137 ± 17 Quality: A-C Quality: A-D N = 4 N = 18

Iceland

ður 66˚

Hólmavík ður

65.5˚ km 050 −24˚ −23˚ −22˚ −21˚ −20˚

Fig. 9. The orientation of SHmax (A–D quality) in the Westfjords. In the oldest area of Iceland (10–16 Ma, Moorbath1968, McDougall1984) SHmax is rotated from rift parallel to rift normal. Legend is the same as in Fig. 1. Note that only surface geological indicators are available in this area. M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 111

orientation of S is sub-parallel to the plate motion (Fig. 9 and S orientation KR Hmax Hmax Árnadóttir et al., 2009, Figs. 3 & 4). fi Plate motion 18.9 Extensive maps of surface ssure swarms are available for Iceland (e.g. Gudmundsson, 1987; Clifton and Kattenhorn, 2006; Hjartardóttir et al., 2009; Einarsson, 2010; Hjartardottir et al., 2015). Even though GOR eruptive fissures can be used as stress indicators, surface fissure swarms HFZ do not provide information on the stress field but on the deformation (Hjartardottir et al., 2015). The fissure swarms are very similarly orient- DZ ed to the orientation of the SHmax. Especially in the NVZ the orientation of the fissure swarms are well in agreement with eruptive fissures and NVZ North American Plate other stress indicators (e.g. Hjartardottir et al., 2015). 19.3 WVZ 5. Conclusion Hreppar Plate fi RPR EVZ In this paper we present the rst comprehensive and quality ranked compilation of the contemporary stress data in Iceland including the SISZ analysis of image logs from 57 geothermal boreholes. In total we com- 19.9 piled 495 SHmax orientations from different stress indicators. The main Eurasian Plate contributions to the newly compiled database are from 171 surface geo- RR 20.2 km logical information, 61 geothermal wells (intermediate-depth), and 175 indicators from focal mechanism solutions of earthquakes (deep). The 0 100 two key findings of this compilation are: (1) no significant depth-

dependent variation in the SHmax orientation (±25°) is observed while

Fig. 10. A simplified tectonic map of Iceland with the mean orientations of SHmax in the four the stress regime changes with depth. (2) four distinct contemporary stress provinces estimated from A–D quality data records (black lines). The standard devi- stress provinces are present in Iceland. The stress provinces are in agree- – ation from A C quality is shown by the large dark grey areas while the light grey areas ment with the large-scale regional tectonic setting. show the standard deviation from A–D quality. The plate boundaries are from Einarsson (2008) & Bird (2003)). The plate motion (mm/yr) is indicated by black arrows relative to the fixed North American plate (Geirsson et al., 2006). The continental plates and the Acknowledgement approximate location of the hotspot (grey circle, Wolfe et al., 1997) are indicated. Further- more the tectonic features are labelled as follows according to Einarsson (2008): RR: The authors would like to thank Romain Plateaux and two anony- Reykjanes Ridge, RPR: Reykjanes peninsula ridge, WVZ: Western Volcanic Zone, SISZ: mous reviewers for their comments which significantly improved the South Iceland Seismic Zone, EVZ: Eastern Volcanic Zone, NVZ: Northern Volcanic Zone, DZ: Dalvík Zone, HFZ: Húsavík-Flatey-Zone, GOR: Grimsey-Oblique-Ridge, and KR: manuscript. Furthermore the authors would like to thank Orkuveita Kolbeinsey Ridge. Reykjavíkur, RARIK, Norðurorka, Landsvirkjun, HS Orka, and Skagafjarðarveitur for the permission to publish the televiewer data. The research leading to these results has received funding from the distance from the spreading centre which depends mainly on the com- European Community's Seventh Framework Programme under grant position of the rock and only partly on the age and distance from the agreement No. 608553 (Project IMAGE). With respect to this we like ridge (Wiens and Stein, 1984). Such a rotation is also expected to to thank David Bruhn and Ólafur G. Flóvenz. The maps are generated occur off the Icelandic east coast to meet the overall trend of SHmax ob- with CASMI (Heidbach and Höhne, 2008)andGMT(Wessel et al., served in Europe (e.g. Grünthal and Stromeyer, 1992; Müller et al., 2013) with topographic data from ETOPO1 (Amante and Eakins, 1992; Heidbach et al., 2007). 2009) and bathymetric data from the GEBCO_2014 Grid, www.gebco. Many of the stress indicators recognised in the applied quality rank- net. We also would like to thank Kristján Sæmundsson and Maryam ing, e.g. focal mechanism solutions or borehole breakouts, are manifes- Khodayar who read the manuscript as well as John Reinecker and tations of a stress field which generally can be assumed as the currently Sebastian Specht for their support in technical issues. Mojtaba Rajabi’s active in-situ stress field. Still, seismic events and volcanic eruptions contribution forms TRaX record #344. may change the local stress field in a very short time interval (e.g. Reasenberg and Simpson, 1992; King et al., 1994; Dieterich et al., Appendix A. Supplementary data 2000). Albeit, these changes induced by seismic events are generally smaller than the regional stress magnitude (Hardebeck, 2010). Hence Supplementary data to this article can be found online at http://dx. they are assumed to be within the uncertainty of SHmax ±15° of even doi.org/10.1016/j.tecto.2016.02.008. the highest quality stress indicators. As well the isostatic rebound from deglaciation is not expected to have an immediate impact on stress References orientation (Plateaux et al., 2014). Aadnoy, B.S., 1990. Inversion technique to determine the in-situ stress field from fracturing data. J. Pet. Sci. Eng. 4, 127–141. Aadnoy, B., Bell, J., 1998. Classification of drilling-induced fractures and their relationship 4.2. Comparison with other observations to in-situ stress directions. Log. Anal. 27–42. Allen, R.M., Nolet, G., Morgan, W.J., Vogfjörd, K., Nettles, M., Ekström, G., Bergsson, B.H., A comparison of the orientation of S with the direction of plate Erlendsson, P., Foulger, G., Jakobsdóttir, S., Julian, B.R., Pritchard, M., Ragnarsson, S., Hmax Ragnar, S., 2002. Plume-driven plumbing and crustal formation in Iceland. Journal motion (Geirsson et al., 2006) shows that they are in quite large areas of Geophysical Research: Solid Earth (1978–2012) 107, ESE–4. perpendicular to each other (Fig. 10). In a more local study (Keiding Amante, C., Eakins, B., 2009. ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data et al., 2009) compared the stress and strain in the Reykjanes peninsula. Sources and Analysis. NOAA Technical Memorandum NESDIS NGDC-24. National Geophysical Data Center, NOAA. The stress is determined from the inversion of focal mechanism from Anderson, E., 1905. The dynamics of faulting. Trans. Edinb. Geol. Soc. 8, 387–402. earthquake swarms while the strain is derived from GPS data. Keiding Anderson, E., 1951. The dynamics of faulting. Oliver and Boyd, Edinburgh. 2 ed. et al. (2009) conclude that the minimum horizontal stress S is paral- Andrew, R.E., Gudmundsson, A., 2008. Volcanoes as elastic inclusions: their effects on the hmin fi ∈_ propagation of dykes, volcanic ssures, and volcanic zones in Iceland. J. Volcanol. lel to the maximum horizontal strain Hmax. This also holds for detailed Geotherm. Res. 177, 1045–1054. GPS data provided by Árnadóttir et al. (2009). In the Westfjords the Angelier, J., 1984. Tectonic analysis of fault slip data sets. J. Geophys. Res. 89, 5835. 112 M. Ziegler et al. / Tectonophysics 674 (2016) 101–113

Angelier, J., Slunga, R., Bergerat, F., Stefánsson, R., Homberg, C., 2004. Perturbation of stress Garcia, S., Angelier, J., Bergerat, F., Homberg, C., 2002. Tectonic analysis of an oceanic and oceanic rift extension across transform faults shown by earthquake focal mech- transform fault zone based on fault-slip data and earthquake focal mechanisms: the anisms in Iceland. Earth Planet. Sci. Lett. 219, 271–284. Húsavík-flatey fault zone, Iceland. Tectonophysics 344, 157–174. Angelier, J., Bergerat, F., Stefánsson, R., Bellou, M., 2008. Seismotectonics of a newly Geirsson, H., Árnadóttir, T., Völksen, C., Jiang, W., Sturkell, E., Villemin, T., Einarsson, P., formed transform zone near a hotspot: earthquake mechanisms and regional stress Sigmundsson, F., Stefánsson, R., 2006. Current plate movements across the mid- in the South Iceland seismic zone. Tectonophysics 447, 95–116. Atlantic Ridge determined from 5 years of continuous GPS measurements in Árnadóttir, T., Jónsson, S., Pedersen, R., Gudmundsson, G.B., 2003. Coulomb stress changes Iceland. J. Geophys. Res. 111, B09407. in the South Iceland seismic zone due to two large earthquakes in June 2000. Gephart, J.W., Forsyth, D.W., 1984. An improved method for determining the regional Geophys. Res. Lett. 30, 1205. stress tensor using earthquake focal mechanism data: application to the San Árnadóttir, T., Lund, B., Jiang, W., Geirsson, H., Björnsson, H., Einarsson, P., Sigurdsson, T., Fernando earthquake sequence. J. Geophys. Res. 89, 9305. 2009. Glacial rebound and plate spreading: results from the first countrywide GPS Green, R.G., White, R.S., Greenfield, T., 2014. Motion in the North Iceland volcanic rift zone observations in Iceland. Geophys. J. Int. 177, 691–716. accommodated by bookshelf faulting. Nat. Geosci. 7, 29–33. Ásmundsson, R., Pezard, P., Sanjuan, B., Henninges, J., Deltombe, J.L., Halladay, N., Lebert, Grünthal, G., Stromeyer, D., 1992. The recent crustal stress field in central Europe: F., Gadalia, A., Millot, R., Gibert, B., Violay, M., Reinsch, T., Naisse, J.M., Massiot, C., trajectories and finite element modeling. J. Geophys. Res. 97, 11805–11820. Azas, P., Mainprice, D., Karytsas, C., Johnston, C., 2014. High temperature instruments Gudmundsson, A., 1987. Tectonics of the Thingvellir fissure swarm, SW Iceland. J. Struct. and methods developed for supercritical geothermal reservoir characterisation and Geol. 9, 61–69. exploitation — the HiTI project. Geothermics 49, 90–98. Gudmundsson, A., 1995. Infrastructure and mechanics of volcanic systems in Iceland. Barth, A., Reinecker, J., Heidbach, O., 2008. Stress derivation from earthquake focal mech- J. Volcanol. Geotherm. Res. 64. anisms. Technical Report. World Stress Map Project. Gudmundsson, A., 2002. Emplacement and arrest of sheets and dykes in central Batir, J., 2011. Stress Field Characterization of the Hellisheiði Geothermal Field and Possi- volcanoes. J. Volcanol. Geotherm. Res. 116, 279–298. bilities to Improve Injection Capabilities Master thesis RES — The School for Renew- Gudmundsson, A., 2003. Surface stresses associated with arrested dykes in rift zones. Bull. able Energy Science. Volcanol. 65, 606–619. Bell, J., 1996. In situ stresses in sedimentary rocks (part 1): measurement techniques. Gudmundsson, A., 2006. How local stresses control magma-chamber ruptures, dyke in- Geosci. Can. 23. jections, and eruptions in composite volcanoes. Earth Sci. Rev. 79, 1–31. Bell, J., Gough, D., 1979. Northeast-southwest compressive stress in Alberta — evidence Gudmundsson, A., Bergerat, F., Angelier, J., Villemin, T., 1992. Extensional tectonics of from oil wells. Earth Planet. Sci. Lett. 45, 475–482. Southwest Iceland. Bull. Soc. Geol. Fr. 163, 561–570. Bergerat, F., Angelier, J., 1998. Fault systems and paleostresses in the vestfirdir peninsula. Gudmundsson, A., Bergerat, F., Angelier, J., 1996. Off-rift and rift-zone palaeostresses in relationships with the tertiary paleo-rifts of skagi and snaefells (Northwest Iceland). Northwest Iceland. Tectonophysics 255, 211–228. Geodin. Acta 11, 105–118. Haflidason, H., Eiriksson, J., Kreveld, S.V., 2000. The tephrochronology of Iceland and the Bergerat, F., Angelier, J., 2001. Mechanisms of the faults of 17 and 21 June 2000 earth- North Atlantic region during the middle and late quaternary: a review. J. Quat. Sci. quakes in the South Iceland seismic zone from the surface traces of the árnes and 15, 3–22. hestfjall faults. Comptes Rendus de l'Académie des Sciences - Series IIA - Earth and Hagos, L., Shomali, H., Lund, B., Böðvarsson, R., Roberts, R., 2008. An application of relative Planetary Science. 333, pp. 35–44. moment tensor inversion to aftershocks of the June 1998 hengill earthquake in Bergerat, F., Plateaux, R., 2012. Architecture and development of (Pliocene to Holocene) Southwest Iceland. Bull. Seismol. Soc. Am. 98, 636–650. faults and fissures in the East Volcanic zone of Iceland. Compt. Rendus Geosci. 344, Haimson, B., Rummel, F., 1982. Hydrofracturing stress measurements in the Iceland 191–204. research drilling project drill hole at Reyðarfjörður, Iceland. J. Geophys. Res. 87, Bergerat, F., Angelier, J., Villemin, T., 1990. Fault systems and stress patterns on emerged 6631–6649. oceanic ridges: a case study in Iceland. Tectonophysics 179, 183–197. Haimson, B., Voight, B., 1977. Crustal stress in Iceland. Pageoph 115, 153–190. Bergerat, F., Gudmundsson, A., Angelier, J., Rögnvaldsson, S., 1998. Seismotectonics of the Hardebeck, J., 2010. Aftershocks are well aligned with the background stress field, contra- central part of the South Iceland seismic zone. Tectonophysics 298, 319–335. dicting the hypothesis of highly heterogeneous crustal stress. J. Geophys. Res. 115, Bird, P., 2003. An updated digital model of plate boundaries. Geochem. Geophys. Geosyst. B12308. 4, 1–52. Hast, N., 1969. The state of stress in the upper part of the earth's crust. Tectonophysics 8, Bjarnason, I.T., Einarsson, P., 1991. Source mechanism of the 1987 Vatnafjöll earthquake 169–211. in South Iceland. J. Geophys. Res. 96, 4313–4324. Heidbach, O., Höhne, J., 2008. CASMI — A visualization tool for the World stress map da- Björnsson, H., Einarsson, P., 1990. Volcanoes beneath Vatnajökull, Iceland: evidence from tabase. Comput. Geosci. 34, 783–791. radio echo-sounding, earthquakes and jökulhlaups. Jökull 22. Heidbach, O., Reinecker, J., Tingay, M., Müller, B., Sperner, B., Fuchs, K., Wenzel, F., 2007. Célérier, B., 2010. Remarks on the relationship between the tectonic regime, the rake of Plate boundary forces are not enough: second- and third-order stress patterns the slip vectors, the dip of the nodal planes, and the plunges of the P, B, and T axes highlighted in the World stress map database. Tectonics 26, 1–19. of earthquake focal mechanisms. Tectonophysics 482, 42–49. Heidbach, O., Tingay, M., Barth, A., Reinecker, J., Kurfeß, D., Müller, B., 2008. The 2008 Re- Centre, G.D., 1993. GEOFON Seismic Network. Other/Seismic Network. lease of the World Stress Map. Clifton, A.E., Kattenhorn, S.A., 2006. Structural architecture of a highly oblique divergent Heidbach, O., Tingay, M., Barth, A., Reinecker, J., Kurfeß, D., Müller, B., 2010. Global crustal plate boundary segment. Tectonophysics 419, 27–40. stress pattern based on the World stress map database release 2008. Tectonophysics Cornet, F.H., Röckel, T., 2012. Vertical stress profiles and the significance of ”stress 482, 3–15. decoupling”. Tectonophysics 581, 193–205. Hjartardóttir, Á.R., Einarsson, P., Sigurdsson, H., 2009. The fissure swarm of the askja Dieterich, J., Cayol, V., Okubo, P., 2000. The use of earthquake rate changes as a stress volcanic system along the divergent plate boundary of N Iceland. Bull. Volcanol. 71, meter at Kilauea volcano. Nature 408, 457–460. 961–975. Dziewonski, A.M.M., Chou, T.A., Woodhouse, J.H., 1981. Determination of earthquake Hjartardottir, A.R., Einarsson, P., Magnusdottir, S., Bjornsdottir, T., Brandsdottir, B., 2015. source parameters from waveform data for studies of global and regional seismicity. Fracture Systems of the Northern Volcanic Rift Zone, Iceland: an onshore part of J. Geophys. Res. 86, 2825. the Mid-Atlantic Plate Boundary. Geological Society. Special Publications, London. Einarsson, P., 1979. Seismicity and earthquake focal mechanisms along the mid- Hjartarson, Á., 2003. Postglacial lava production in Iceland. The Skagafjörður unconformi- Atlantic plate boundary between Iceland and the Azores. Tectonophysics 55, ty, North Iceland, and its geological history. Geological Museum. University of Copen- 127–153. hagen, Copenhagen, pp. 95–108 PhD thesis. Einarsson, P., 1987. Compilation of earthquake fault plane solutions in the North Atlantic Hjartarson, Á., Sæmundsson, K., 2014. Geologic Map of Iceland. Bedrock. 1: 600 000. Tech- and Arctic Oceans. Recent Plate Movements and Deformation, pp. 47–62. nical Report. Iceland GeoSurvey. Einarsson, P., 1991. Earthquakes and present-day tectonism in Iceland. Tectonophysics Ito, G., Lin, J., Graham, D., 2003. Observational and theoretical studies of the dynamics of 189, 261–279. mantle plume-mid-ocean ridge interaction. Rev. Geophys. 41, 1017. Einarsson, P., 2008. Plate boundaries, rifts and transforms in Iceland. Jökull 35–58. Jakobsdóttir, S., 2008. Seismicity in Iceland: 1994–2007. Jökull (1994–2007). Einarsson, P., 2010. Mapping of Holocene surface ruptures in the South Iceland seismic Jakobsdóttir, S., Gudmundsson, G., Stefánsson, R., 2002. Seismicity in Iceland 1991–2000 zone. Jökull 60, 121–138. monitored by the SIL seismic system. Jökull 87–94. Einarsson, P., Klein, F., Björnsson, S., 1977. The Borgarfjördur earthquakes of 1974 in West Jakobsson, S., 1979. Petrology of recent basalts of the Eastern Volcanic zone. Iceland. Acta Iceland. Bull. Seismol. Soc. Am. 67, 187–208. Natiuralia Islandica 26. Ekström, G., 1994. Anomalous earthquakes on volcano ring-fault structures. Earth Planet. Jefferis, R., Voight, B., 1981. Fracture analysis near the mid-ocean plate boundary, Sci. Lett. 128, 707–712. Reykjavík-Hvalfjörður area, Iceland. Tectonophysics 76, 171–236. Ekström, G., Nettles, M., Dziewonski, A., 2012. The Global CMT Project 2004–2010: Jóhannesson, H., Sæmundsson, K., 1998. Geological map of Iceland, 1:500,000. Bedrock centroid-moment tensors for 13,017 earthquakes. Phys. Earth Planet. Inter. Geology. Technical Report. Icelandic Institute of Natural History and Icelandic Geo- 200-201, 1–9. detic Survey, Reykjavík. Flóvenz, Ó.G., Ágústsson, K., Guðnason, E.Á., Kristjánsdóttir, S., 2015. Reinjection and in- Keiding, M., Lund, B., Árnadóttir, T., 2009. Earthquakes, stress, and strain along an oblique- duced seismicity in geothermal fields in Iceland. Proceedings World Geothermal ly divergent plate boundary: Reykjanes peninsula, Southwest Iceland. J. Geophys. Res. Congress 2015, Melbourne, Australia, pp. 1–15. 114, B09306. Forslund, T., Gudmundsson, A., 1991. Crustal spreading due to dikes and faults in South- Khodayar, M., Franzson, H., 2007. Fracture pattern of Thjórsárdalur central volcano with west Iceland. J. Struct. Geol. 13, 443–457. respect to rift-jump and a migrating transform zone in South Iceland. J. Struct. Geol. Garcia, S., 2003. Implications d'un Saut de Rift et Du Fonctionnement d'une Zone 29, 898–912. Transformante Sur les Deformations Du Nord de l'Islande (Phd-thesis) Université King, G., Stein, R., Lin, J., 1994. Static stress changes and the triggering of earthquakes. Bull. Pierre et Marie Curie. Seismol. Soc. Am. 84, 935–953. Garcia, S., Dhont, D., 2005. Structural analysis of the Húsavík-flatey transform fault and its Kirsch, E.G., 1898. Die theorie der Elastizität und die Bedürfnisse der festigkeitslehre. relationships with the rift system in Northern Iceland. Geodin. Acta 18, 31–41. Z. Ver. Dtsch. Ing. 42, 797–807. M. Ziegler et al. / Tectonophysics 674 (2016) 101–113 113

Kristjánsdóttir, S., 2013. Microseismicity in the Krýsuvk Geothermal Field, SW Iceland, Schäfer, K., Keil, S., 1979. In situ gesteinsspannungsermittlungen in island. Messtechnische from May to October 2009 (Master thesis) University of Iceland. Briefe 15, 35–46. Lawver, L., Müller, R., 1994. Iceland hotspot track. Geology 22, 311–314. Scheidegger, A., 1962. Stresses in the earth's crust as determined from hydraulic fractur- Ljunggren, C., Chang, Y., Janson, T., Christiansson, R., 2003. An overview of rock stress ing data. Geologie und Bauwesen 27, 1. measurement methods. Int. J. Rock Mech. Min. Sci. 40, 975–989. Segall, P., Fitzgerald, S.D., 1998. A note on induced stress changes in hydrocarbon and geo- Lund, B., Bödvarsson, R., 2002. Correlation of microearthquake body-wave spectral ampli- thermal reservoirs. Tectonophysics 289, 117–128. tudes. Bull. Seismol. Soc. Am. 92, 2419–2433. Sigmundsson, F., Einarsson, P., Halldorson, P., Jakobsdottir, S., Vogfjörd, K., Sigbjörnsson, Lund, B., Slunga, R., 1999. Stress tensor inversion using detailed microearthquake infor- R., Snaebjörnsson, J.T., 2005. Earthquakes and Faults in the Kárahnjúkar Area. Techni- mation and stability constraints: application to ölfus in Southwest Iceland. cal Report March, Jardvisindastofnun Haskolans. J. Geophys. Res. 104, 14947. Sigmundsson, F., Hooper, A., Hreinsdóttir, S., Vogfjörd, K.S., Ófeigsson, B.G., Heimisson, Mardia, K., 1972. Statistics of Directional Data: Probability and Mathematical Statistics. E.R., Dumont, S., Parks, M., Spaans, K., Gudmundsson, G.B., Drouin, V., Árnadóttir, T., Academic Press, London. Jónsdóttir, K., Gudmundsson, M.T., Högnadóttir, T., Fridriksdóttir, H.M., Hensch, M., Martnez-Garzón, P., Bohnhoff, M., Kwiatek, G., Dresen, G., 2013. Stress tensor changes re- Einarsson, P., Magnússon, E., Samsonov, S., Brandsdóttir, B., White, R.S., lated to fluid injection at the geysers geothermal field. Calif. Geophys. Res. Lett. 40, Ágústsdóttir, T., Greenfield, T., Green, R.G., Hjartardóttir, Á.R., Pedersen, R., Bennett, 2596–2601. R.A., Geirsson, H., La Femina, P.C., Björnsson, H., Pálsson, F., Sturkell, E., Bean, C.J., Mastin, L., 1988. Effect of borehole deviation on breakout orientations. J. Geophys. Res. 93, Möllhoff, M., Braiden, A.K., Eibl, E.P.S., 2015. Segmented lateral dyke growth in a 9187. rifting event at Bardarbunga volcanic system, Iceland. Nature 517, 191–195. McDougall, I., Kristjánsson, L., Sæmundsson, K., 1984. Magnetostratigraphy and geochro- Sigurdsson, H., 1970. Structural origin and plate tectonics of the snaefellsnes volcanic nology of Northwest Iceland. J. Geophys. Res. 89, 7029–7060. zone. Western Iceland Earth Planet. Sci. Lett. 10, 129–135. McKenzie, D., 1969. The relation between fault plane solutions for earthquakes and the Sigurdsson, H., Sparks, R.S.J., 1978. Rifting episode in North Iceland in 1874–1875 and the directions of the principal stresses. Bull. Seismol. Soc. Am. 59, 591–601. eruptions of Askja and Sveinagja. Bull. Volcanol. 41, 149–167. Moorbath, S., Sigurdsson, H., Goodwin, R., 1968. K–Ar ages of the oldest exposed rocks in Sinton, J., Grönvold, K., Sæmundsson, K., 2005. Postglacial eruptive history of the Western Iceland. Earth Planet. Sci. Lett. 4, 197–205. Volcanic Zone, Iceland. Geochem. Geophys. Geosyst. 6 (n/a–n/a). Müller, B., Zoback, M., Fuchs, K., 1992. Regional patterns of tectonic stress in Europe. Soosalu, H., Einarsson, P., 1997. Seismicity around the Hekla and Torfajökull volcanoes, J. Geophys. Res. Solid Earth 97, 11783–11803. Iceland, during a volcanically quiet period, 1991–1995. Bull. Volcanol. 59, 36–48. Nakamura, K., 1977. Volcanoes as possible indicators of tectonic stress orientation — Sperner, B., Muller, B., Heidbach, O., Delvaux, D., Reinecker, J., Fuchs, K., 2003. Tectonic principle and proposal. J. Volcanol. Geotherm. Res. 2, 1–16. stress in the Earth's crust: advances in the World stress map project. Geol. Soc. Nettles, M., Ekström, G., 1998. Faulting mechanism of anomalous earthquakes near Lond. Spec. Publ. 212, 101–116. Barðabunga volcano, Iceland. J. Geophys. Res. 103, 17973–17983. Stefánsson, R., 1966. Methods of focal mechanism studies with application to two Atlantic Peška, P., Zoback, M.D., 1995. Compressive and tensile failure of inclined well bores and earthquakes. Tectonophysics 3, 209–243. determination of in situ stress and rock strength. J. Geophys. Res. 100, 12791. Stefánsson, R., Gudmundsson, G.B., Halldórsson, P., 2000. The two large earthquakes in Pierdominici, S., Heidbach, O., 2012. Stress fieldofItaly— mean stress orientation at dif- the South Iceland Seismic Zone on June 17 and 21, 2000. Technical Report July. ferent depths and wave-length of the stress pattern. Tectonophysics 532–535, Icelandic Meterological Organisation. Reykjavk. 301–311. Stefánsson, R., Gudmundsson, G.B., Halldorsson, P., 2008. Tjörnes fracture zone. New and Plateaux, R., Bergerat, F., Béthoux, N., Villemin, T., Gerbault, M., 2012. Implications of frac- old seismic evidences for the link between the North Iceland rift zone and the turing mechanisms and fluid pressure on earthquakes and fault slip data in the East Mid-Atlantic ridge. Tectonophysics 447, 117–126. Iceland rift zone. Tectonophysics 581, 19–34. Sykes, L.R., 1967. Mechanism of earthquakes and nature of faulting on the mid-oceanic Plateaux, R., Béthoux, N., Bergerat, F., Mercier de Lépinay, B., 2014. Volcano-tectonic inter- ridges. J. Geophys. Res. 72, 2131–2153. actions revealed by inversion of focal mechanisms: stress field insight around and be- Sykes, L., Sbar, M., 1974. Focal mechanism solutions of intraplate earthquakes and stresses neath the Vatnajökull ice cap in Iceland. Front. Earth Sci. 2, 1.–21 in the lithosphere. In: Kristjansson, L. (Ed.), Geodynamics of Iceland and the North Ragnarsson, Á., 2015. Geothermal development in Iceland 2010–2014. Proceedings World Atlantic Area. D. Reidel Publishing Company, Dordrecht-Holland, pp. 207–224. Geothermal Congress 2015, Melbourne, p. 15. Thordarson, T., Larsen, G., 2007. Volcanism in Iceland in historical time: volcano types, Rajabi, M., Tingay, M., King, R., Heidbach, O., 2016. Present-day stress orientation in the eruption styles and eruptive history. J. Geodyn. 43, 118–152. Clarence-Moreton Basin of New South Wales, Australia: a new high density dataset Tibaldi, A., Bonali, F.L., Pasquaré, F.A., Rust, D., Cavallo, A., D'Urso, A., 2013. Structure of reveals local stress rotations. Basin Res. regional dykes and local cone sheets in the midhyrna-lysuskard area, snaefellsnes Reasenberg, P.A., Simpson, R.W., 1992. Response of regional seismicity to the static stress peninsula (NW Iceland). Bull. Volcanol. 75, 764. change produced by the Ioma Prieta earthquake. Science 255, 1687–1690. Tingay, M., Müller, B., Reinecker, J., Heidbach, O., Wenzel, F., Fleckenstein, P., 2005. Under- Reinecker, J., Tingay, M., Müller, B., Heidbach, O., 2010. Present-day stress orientation in standing tectonic stress in the oil patch: the World Stress Map Project. Lead. Edge 24, the Molasse Basin. Tectonophysics 482, 129–138. 1276–1282. Reiter, K., Heidbach, O., 2014. 3-D geomechanical-numerical model of the contemporary Villemin, T., Bergerat, F., Angelier, J., Lacasse, C., 1994. Brittle deformation and fracture crustal stress state in the Alberta Basin (Canada). Solid Earth 5, 1123–1149. patterns on oceanic rift shoulders: the Esja Peninsula, SW Iceland. J. Struct. Geol. Reiter, K., Heidbach, O., Schmitt, D., Haug, K., Ziegler, M., Moeck, I., 2014. A revised crustal 16, 1641–1654. stress orientation database for Canada. Tectonophysics 636, 111–124. Ward, P., 1971. New interpretation of the geology of Iceland. Geol. Soc. Am. Bull. 82, Richardson, R.M., 1992. Ridge forces, absolute plate motions, and the intraplate stress 2991 –3012. field. J. Geophys. Res. 97, 11739. Wessel, P., Smith, W.H.F., Scharroo, R., Luis, J., Wobbe, F., 2013. Generic mapping tools: im- Röckel, T., Lempp, C., 2003. Der Spannungszustand im norddeutschen becken. Erdöl proved version released. EOS Trans. Am. Geophys. Union 94, 409–410. Erdgas Kohle 119, 73–80. White, R.S., Drew, J., Martens, H.R., Key, J., Soosalu, H., Jakobsdóttir, S.S., 2011. Dynamics of Rögnvaldsson, S., Slunga, R., 1994. Single and joint fault plane solutions for microearth- dyke intrusion in the mid-crust of Iceland. Earth Planet. Sci. Lett. 304, 300–312. quakes in South Iceland. Tectonophysics 237, 73–80. Wiens, D.A., Stein, S., 1984. Intraplate seismicity and stresses in young oceanic litho- Roman, D.C., Moran, S., Power, J., Cashman, K., 2004. Temporal and spatial variation of sphere. J. Geophys. Res. 89, 11442. local stress fields before and after the 1992 eruptions of crater peak vent, Mount Wiprut, D., Zoback, M., Hanssen, T., Peska, P., 1997. Constraining the full stress tensor from Spurr volcano. Alaska Bull. Seismol. Soc. Am. 94, 2366–2379. observations of drilling-induced tensile fractures and leak-off tests: application to Roth, F., Fleckenstein, P., 2001. Stress orientations found in North-East Germany differ borehole stability and sand production on the. Int. J. Rock Mech. Min. Sci. 34, 3–4. from the West European trend. Terra Nova 13, 289–296. Wolfe, C.J., Bjarnason, I.Th., VanDecar, J.C., Solomon, S.C., 1997. Seismic structure of the Roth, F., Henneberg, K., Fleckenstein, P., Palmer, J., Stefanssson, V., Gudlaugsson, S., 2000. Iceland mantle plume. Nature 385, 245–247. Ergebnisse von bohrloch-spannungsmessungen in der Südisländischen Zang, A., Stephansson, O., 2010. Stress Field of the Earth's Crust. Springer Netherlands, Seismizitätszone. Mitt./Nachr. Deut. Geol. Ges. 3, 49–50. Dordrecht. Sæmundsson, K., 1974. Evolution of the axial rifting zone in Northern Iceland and the Zoback, M.L., 1992. First- and second-order patterns of stress in the lithosphere: the Tjörnes fracture zone. Geol. Soc. Am. Bull. 85, 495–504. World Stress Map Project. J. Geophys. Res. 97, 11703. Sæmundsson, K., 1978. Fissure swarms and central volcanoes of the neovolcanic zones of Zoback, M., Zoback, M., 1991. Tectonic stress field of North America and relative plate mo- Iceland. Geol. J. Spec 10, 415–432. tions. In: Slemmons, D., Engdahl, E. (Eds.), Neotectonics of North America. Geological Sæmundsson, K., 1979. Outline of the geology of Iceland. Jökull 29, 7–28. Society of America, Boulder, Colorado, pp. 339–366. Sæmundsson, K., 1986. Subaerial volcanism in the western north Atlantic. Geol. N. Am. Zoback, M.L., Zoback, M.D., Adams, J., Assumpção, M., Bell, S., Bergman, E.A., Blümling, P., 1000, 69–86. Brereton, N.R., Denham, D., Ding, J., Fuchs, K., Gay, N., Gregersen, S., Gupta, H.K., Sánchez, J.J., Wyss, M., McNutt, S.R., 2004. Temporal–spatial variations of stress at redoubt Gvishiani, A., Jacob, K., Klein, R., Knoll, P., Magee, M., Mercier, J.L., Müller, B.C., volcano, Alaska, inferred from inversion of fault plane solutions. J. Volcanol. Paquin, C., Rajendran, K., Stephansson, O., Suarez, G., Suter, M., Udias, A., Xu, Z.H., Geotherm. Res. 130, 1–30. Zhizhin, M., 1989. Global patterns of tectonic stress. Nature 341, 291–298. Sbar, M.L., Sykes, L.R., 1973. Contemporary compressive stress and seismicity in eastern North America: an example of intra-plate tectonics. Geol. Soc. Am. Bull. 84, 1861–1882. −25° −24° −23° −22° −21° −20° −19° −18° −17° −16° −15° −14° −13° 67° Stress Map Iceland 2016

Editors: Moritz Ziegler1,2, Mojtaba Rajabi3, Gylfi Páll Hersir4, Kristján Ágústsson4, Sigurveig Árnardóttir4, Arno Zang1,2, David Bruhn1,5, Oliver Heidbach1 1Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Germany 2University of Potsdam, Institute of Earth and Environmental Science, Germany 3Australian School of Petroleum, The University of Adelaide, Adelaide, Australia 4Iceland GeoSurvey (ÍSOR), Grensásvegur 9, 108 Reykjavík, Iceland 5TU Delft, Department of Geoscience & Engineering, Delft, The Netherlands

Introduction Iceland is located on the Mid-Atlantic Ridge which is any further information as a point. Used stress data the plate boundary between the Eurasian and the are part of the World Stress Map (WSM) database North American plate. It is one of the few places release 2016 and freely available. Further informa- where an active spreading can be observed onsho- tion about the data, criteria, data analysis and re. In this map we present a comprehensive compi- quality ranking are located on the WSM webpage: lation of the orientation of maximum horizontal www.world-stress-map.org. stress (SHmax). In particular we interpret borehole breakouts and drilling induced fractures from bore- The majority of SHmax orientation data records in hole image logs in 57 geothermal wells onshore Iceland is derived from earthquake focal mecha- Iceland. The borehole results are combined with nism solutions (35 %) and geological fault slip other stress indicators including earthquake focal inversions (26 %). 20 % of the data are borehole mechanism solutions, geological information and related stress indicators. In addition minor overcoring measurements resulting in a dataset with shares of SHmax orientations are compiled, 495 data records for the SHmax orientation. The amongst others, from focal mechanism inversi- Ísafjörður reliability of each indicator is assessed according to ons and the alignment of fissure eruptions. The the quality criteria of the World Stress Map project. results show that the S orientations derived Hmax 66° from different depths and stress indicators are Húsavík The dataset is collected under the assumption, that consistent with each other. the vertical stress (Sv) is a principal stress, the orientation of the 3D stress tensor is defined by the The resulting pattern of the present-day stress in orientation oft the maximum horizontal stress Iceland has four distinct subsets of SHmax orien- (SHmax) only. The minimum horizontal stress tations. The SHmax orientation is parallel to the (Shmin) is perpendicular to SHmax. The orientation rift axes in the vicinity of the active spreading of SHmax is illustrated by lines with different length regions. It changes from NE-SW in the South to in the map. The length of each line is a measure for approximately N-S in central Iceland and the quality of the data, the symbol specifies the NNW-SSE in the North. In the Westfjords which method and the colour indicates the stress regime. is located far away from the ridge the regional Data with the lowest quality (E) are illustrated without SHmax rotates and is parallel to the plate motion. Sauðárkrókur Akureyri Method Quality Stress Regime

focal mechanism A normal faulting borehole breakouts B drilling-induced fractures strike-slip faulting C overcoring thrust faulting D borehole slotter E unknown regime hydraulic fractures geological indicators Egilsstaðir

Stress Regime

SV SV SV

SHmax Shmin S SHmax SHmax NF hmin Shmin SS TF 65°

normal faulting strike-slip faulting thrust faulting S > S > S V Hmax hmin SHmax > SV > Shmin SHmax > Shmin > SV

GOR HFZ

DZ North American NVZ Plate

Eurasian WVZ Akranes Hreppar Plate Plate Höfn RPR EVZ SISZ Reykjavík RR Keflavík 64° The Mid-Atlantic Ridge and its different segments in Zone, WVZ: Western Volcanic Zone, EVZ: Eastern Iceland according to Bird (2003) and Einarsson Volcanic Zone, NVZ: Northern Volcanic Zone,DZ: (2008). RR: Reykjanes Ridge, RPR: Reykjanes Dalvík Zone, HFZ: Húsavík Flatey Zone, GOR: Peninsula Ridge, SISZ: South Iceland Seismic Grímsey Oblique Rift, KR: Kolbeinsey Ridge. Selfoss

Citation of this map Ziegler, M., Rajabi, M., Hersir, G. P., Ágústsson, K., Árnardóttir, S., Zang, A., Bruhn, D., Heidbach, O., (2016): Stress Map Iceland 2016, doi:10.5880/WSM.Iceland2016.

Key references Ziegler, M., Rajabi, M., Heidbach, O., Hersir, G. P., Ágústsson, K., Árnadóttir, S. and Zang, A., (2016): The Stress Pattern of Iceland. Tectonophysics, 674, 101-113, doi: 10.1016/j.tecto.2016.02.008. Heidbach, O., Rajabi, M., Reiter, K., Ziegler, M. and the WSM Team, (2016): World Stress Map Database Release 2016. GFZ Data Services, doi:10.5880/WSM.2016.001. Heidbach, O., Tingay, M., Barth, A., Reinecker, J., Kurfeß, D., Müller, B., (2010): Global crustal stress pattern based on the World Stress Map database release 2008. Tectonophysics, 482, 3-15, doi:10.1016/j.tecto.2009.07.023.

References of used data and software Amante, C. & Eakins, B., (2009): ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis NOAA Technical Memorandum NESDIS NGDC-24. National Geophysical Data Center, NOAA, doi: 10.7289/V5C8276M. Bird, P., (2003): An updated digital model of plate boundaries. Geochemistry, Geophysics, Geosystems, Heimaey 4, 1-52, doi: 10.1029/2001GC000252 Vík í Mýrdal Einarsson, P., (2008): Plate boundaries, rifts and transforms in Iceland. Jökull, 5-58. Heidbach, O., Höhne, J., (2008): CASMI - a tool for the visualization of the World Stress Map data base. Computers and Geosciences, 34, 783-791, doi:1016/j.cageo.2007.06.004. Wessel, P., Smith, W.H.F., (1998): New, improved version of Generic Mapping Tools released, Eos Trans., 79 (47), 579, doi:10.1029/98EO00426. Bathymetric data from the GEBCO_2014 Grid, www.gebco.net.

The research leading to these results has received funding from the European Community‘s Seventh Framework Programme under grant agreement No. 608553 (Project IMAGE). The editors would like to thank Orkuveita Reykjavíkur, RARIK, Norðurorka, Landsvirkjun, HS Orka, and Skagafjarðarveitur for the permission to publish the televiewer data. 63° −24° −23° −22° −21° −20° −19° −18° −17° −16° −15° −14°