Field trip: Some aspects of the structure of Reykjanes fissure swarm

Introduction
The Mid-Atlantic Ridge (MAR) defines the boundary between the North American and European plates. Its topography is usually quite rough, with high rift mountains bordering a down-faulted median rift valley (e.g., Vogt, 1986). However, the rift valley gradually disappears as the ridge approaches the major hot spots of the Azores and Iceland. Global studies of mid-ocean ridges have revealed that major differences in ridge morphology can often be attributed to spreading rate. However, overall magma budget (Sempere et al., 1990) and segment orientation with respect to the direction of plate motion (Taylor et al., 1994) also play critical roles in determining the structural architecture of a particular ridge segment.
The segment of the MAR north of the Charlie Gibbs Fracture Zone is known as the Reykjanes Ridge. North of latitude 57°N, the ridge bends and takes on a trend of 035° so that it is oblique to the direction of plate motion which varies along the length of the ridge from 096° to 100° (DeMets et al., 1994). The acute angle, a , between the rift trend and the direction of plate motion is approximately 60°.  At latitude 58°30’N, the topography of the ridge becomes smoother and its median valley disappears, due to a change in lithospheric rheology and an increased magma supply as the ridge approaches the Iceland mantle plume (Malinverno, 1993). At latitude 63°48’N, the ridge comes onshore at Reykjanes Peninsula, Iceland. Here it bends sharply to the east so that it becomes approximately 30° oblique to the direction of plate motion.
Our understanding of how Reykjanes Peninsula fits into the plate tectonic model has evolved considerably, as the model itself has evolved, over the past thirty years. Because of its geometry with respect to adjacent portions of the MAR, early researchers (e.g.Ward, 1971, Courtillot et al., 1974) believed the peninsula to be a transform fault, but problems arose with that model when no through-going strike-slip fault could be found.  Nakamura (1970) was one of the first researchers to suggest that it is an oblique rift zone. Based on the geometry of en-echelon extensional structures examined both in the field and with air photos, he proposed that the rift zone undergoes equal amounts of extension and shear, perhaps discontinuously in the short term, but accomplishing the “average” plate motion in the long term. The tectonic setting and state of stress are now fairly well established and characteristics of both transform and ridge segment are evident. Still unclear are some of the dynamics of this zone, how faulting and magmatism are related in time and space on the Peninsula, how the structures we see at the surface are related to the processes that formed them and what role ridge geometry (i.e., the angle of rift obliquity) plays in the tectonic development of a spreading segment.

Tectonic Setting and Geologic Background
The center of the Iceland mantle plume lies under the Vatnajokull glacier (Wolfe, et al., 1997) and probably has a major influence on the geometry of the plate boundary in Iceland. The plate motion direction varies in Iceland from 102.74° at Reykjanes Peninsula to 106.21° along the Northern Volcanic Zone according to the NUVEL-1A model (DeMets et al., 1994). According to the classification of Macdonald (1998), the volcanic zones in Iceland are second-order ridge segments that are approximately 100 km in length and separated from each other by either a segment overlap or a change in strike direction  (“rift valley jog”).  Each segment has a different orientation with respect to the direction of plate motion and each has a unique character. Along with differences in their magma budget, their differing orientations can account for many of the structural differences between these ridge segments.
A major ridge jump approximately 6-7 million years ago initiated active spreading on Reykjanes Peninsula (Saemundsson, 1979; Johannesson, 1980). A narrow zone of seismicity having an average trend of 075° runs along the length of the peninsula and has been used to define the active plate boundary (Klein et al., 1977; Einarsson, 1991). This zone intersects fissure swarms near the zone of maximum volcanic production and seems to control the location of geothermal areas on the peninsula (Einarsson, 1991). The peninsula is characterized by arrays of eruptive fissures, spaced on average approximately 5 km apart, and having an average strike of 040°. These have been described in the literature as comprising four distinct volcanic systems (Einarsson and Saemundsson, 1987), each with their own magma supply and high temperature geothermal system.
Sub-glacial and sub-aerial (post-glacial) fissure eruptions have formed prominent NE-trending ridges and crater rows that dominate the topography of the peninsula.  Sub-glacial eruptions produce ridges of hyaloclastite that bear an unmistakable resemblance in volume, height and aspect ratio to the Axial Volcanic Ridges (AVRs) that have been described along the Reykjanes Ridge (Murton and Parson, 1993; Searle et al., 1998).  A number of table mountains and hyaloclastite cones, products of sub-glacial eruptions from isolated vents, are also present on the peninsula and closely resemble the small seamounts that have been mapped on the MAR. Early post-glacial basaltic (large volume) and picritic (small volume) shield volcanoes have also played a major role in surfacing this ridge segment with voluminous pahoehoe lava flows, which both cover and are covered by the products of fissure eruptions. Shield volcanoes and eruptive fissures have been active on the peninsula during the Holocene, but the last known eruption was in the fourteenth century (Saemundsson, 1995).
Deformation on the peninsula is accommodated by extensional features that include normal faults, opening mode fissures, and small graben structures (25 to 40 meters wide) that have been closely associated with the eruptive fissures and used to define the boundaries of volcanic systems (Saemundsson, 1979; Gudmundsson, 1980). The base of the seismogenic zone is between 5 and 11 km on the Peninsula (Einarsson, 1991) and most seismicity occurs at depths from 1 to 5 km. A narrow zone of seismicity 2 to 5 km wide, characterized by predominantly strike-slip focal mechanisms and extending the entire length of the peninsula, was identified by Einarsson (1991) as the currently active plate boundary. In the eastern part of the Reykjanes Peninsula, seismicity is characterized by focal mechanisms indicating right-lateral strike-slip faulting on N-S planes or left-lateral strike-slip faulting on E-W planes. However, seismicity on the western part of the peninsula is principally characterized by normal faulting on NE-striking planes (Einarsson, 1991). During a large earthquake swarm in 1972, focal mechanisms ranged from normal to oblique to strike-slip faulting (Klein et al., 1977). Geodetic measurements between 1986 and 1998 (Sturkell et al., 1994; Hreinsdottir et al., 1999) show that left-lateral shear is currently accumulating on the peninsula. Data from Satellite Radar Interferometry support this and indicate that below a depth of 5 km plate motion is accommodated by continuous ductile deformation (Vadon and Sigmundsson, 1997).
The Reykjanes fissure swarm (RFS) is located in westernmost Reykjanes Peninsula (fig.1). It includes three en-echelon sets of eruptive fissures spaced between 2 and 3 km apart. They are, from west to east, the Stampar, Eldvorp and Sundhnukur (after Jonsson, 1978) eruptive fissures, all of which have been active in historic time. Normal faults, hybrid fractures and opening-mode fissures are associated with the eruptive fissures, but most occur on the edges of the area of eruptive activity.

Geological Descriptions
1.Vogar
 The area referred to as the Vogar graben lies near the village of Vogar, along the  northern coast of western Reykjanes Peninsula (Fig. 2). It is on the northwestern edge of the Reykjanes fissure swarm, outside the area of eruptive fissures. It lies mostly within a single compound lava flow erupted from the Thrainsskjoldur shield about 12,500 years BP (Saemundsson, 1995).
The graben structure is only present in the northern part of the map area, where the fault known as Hrafnagja defines its northwest boundary. The graben axis trends northeasterly. From Hrafnagja to the southwest, the graben both deepens and widens to a maximum width of 2 km where it becomes covered by early Holocene lavas and historic lavas from the Eldvorp crater row.  Therefore, its southern extent cannot be defined. The graben structure narrows and shallows to the northeast and dies out approximately 5 km northeast of the turnoff to the village of Vogar along Road 41. The remainder of the map area consists of tilted fault blocks with steeply dipping faults facing to the northwest, away from the rift center.
All fractures examined in this area cut early post-glacial Thrainsskjoldur shield lavas.  Field evidence confirms that fractures in the southwestern part of this study area were present when the Thrainsskjoldur lavas were erupted, and that, at least in that part of the graben, little movement has occurred along those fractures since the eruption of those lavas. Here, pahoehoe flows drape long, smooth, curvilinear fault scarps along which any possible evidence of segment linkage has been either eroded or covered up (Fig. 2B). These faults bound a small inlier of interglacial shield lavas. The presence of pillow breccia and scoria at the base of the fault scarp indicate that lava flowed over the fault and into a lake. Several small, fault-bounded lakes are currently present here. Subsidence along faults in the northeastern part of the study area, on the other hand, has been significant since early post-glacial times. The longer fractures here are complex, highly segmented faults that change character frequently along strike and commonly display a zigzag geometry (Fig. 2C). Maximum throw measured in this study along a single normal fault is 18 m and average throw for all normal faults in the graben is 4.7 m (Gudmundsson, 1986).
 
 

2. Stora-Sandvík
 Fractures at Stora-Sandvík lie at the westernmost edge of the fissure swarm and cut early post-glacial pahoehoe shield lavas.  Normal faults here have displacement of several meters, but true displacement is difficult or impossible to determine because of the presence of abundant windblown sand.  In fact, dune structures are common in this area and frequently seem to follow the same trend as many faults, perhaps obscuring fault scarps in places. Narrow graben structures up to 40 meters wide are common in this location. Most trend in a northeasterly direction and many interact with ENE-striking normal faults resulting in very complex, zigzag structures at Stora-Sandvík (Fig. 3).
 The Stampar eruptive fissure lies directly south of here  and strikes 035°. The Eldvorp crater row lies 5.6 km to the east and has a strike of 045°. Both were active during historic time, between 600 and 800 years BP (Saemundsson, 1995). Fractures near each of these eruptive fissures strike approximately parallel to the crater rows.

3. Right-lateral oblique-slip fault

The fault shown in Fig. 4 is only a single km-long segment that is parat of a complex fault. The one shown here is in turn segmented, as shown in the figure. In spite of the rarity of these structures, they appear to play an important role in strain accommodation on the peninsula, so it is worth describing one of them in detail. The fault trends 010° and is 927 meters in length where mapped. In general, it appears to be made up of a series of elliptical mounds or push-up structures consisting of blocks of broken lava, each trending in a NW direction, commonly separated by small gashes or normal fault segments trending NE. All normal fault segments exhibit down-to-the-east displacement. However, most normal displacement is minimal (< 1 m). The exception is at the northern end of the fault where the dominant mode of faulting is dip slip. The geometry of the push-ups and gashes is consistent with a right-lateral sense of motion.
Although no evidence of lateral offset could be determined along this particular fault, the pattern and trends of alternating mounds and gashes is similar to that described along historically active right-lateral strike-slip faults in the SISZ (e.g., Bjarnasson et al., 1993). In several locations in the eastern part of Reykjanes Peninsula, offset has been determined to be right-lateral strike-slip on similar structures (Eyolfsson, 1998; Einarsson et al., in prep).

4. Mölvík
Mölvík (Figure 5) is a small graben structure just south of the Eldvörp crater row. The flows here are however older than the historic Eldvörp lavas. These 2000 year old lavas are cut by a series of faults. South of the road, three normal faults can be seen propagating through the ponded flow. To the north of the road, a small graben structure is apparent. The presence of magma at shallow crustal levels in Iceland perturbs the magnitude of the stress field and allows dikes to form episodically, perpendicular to the direction of Ehmax. Studies conducted during the Krafla eruptions in northern Iceland during the 1970’s and 1980’s (e.g., Einarsson and Brandsdóttir, 1980; Rubin, 1990) showed the direct relationship between dike injection and faulting. Based on their morphology and spatial distribution, many of the structures we are seeing are believed to be a result of similar but smaller scale dike injection events.
At Mölvík, it is possible to determine a sequence of events that shows this relationship of faulting to volcanism in the center of the neo-volcanic zone. Detailed mapping by Whitlock (1999) indicates that most, and perhaps all, of the faulting in this graben occurred during a sequence of at least three eruptions from a single vent close to the Eldvörp eruptive fissure. Displacement along a given fault in the graben is no more than 1.5 m, and individual fault segments are on the order of 100 m long. Graben width is approximately 25 m.
Graben structures at Stora-Sandvík have similar dimensions, although some are as wide as 40 m and are several meters deep. Such narrow graben structures have been mapped along the Juan deFuca Ridge (Chadwick and Embley, 1998). Pollard et al. (1983) have shown that graben width is a function of the depth of the dike tip at the time faulting occurs. Chadwick and Embley (1998) suggest that narrow grabens result when the ambient tectonic stress field is not so close to failure. During the 1974 to 1984 events at Krafla, 250 years worth of tensional strain had accumulated since the last volcanic event and the crust was near to failure when dike injection occurred. Consequently, graben faults could form while the dike tip was still at depth, as only a small stress perturbation was necessary to allow faulting. Some grabens formed during diking events at Krafla were greater than 1 km wide (Rubin, 1992). After many diking events, compressive stress may increase around the dike faster than it can be accommodated by extension due to plate motion (Rubin, 1990). Chadwick and Embley (1998) propose that narrow grabens along the Juan deFuca Ridge are a result of the compressive stress that builds up there due to frequent diking events. This is a plausible explanation for the narrow grabens at Stora-Sandvik because they are numerous in this particular location and may be a result of a series of diking events that were closely spaced in time.
An alternative, or perhaps additional, explanation is afforded by the oblique orientation of the rift zone on Reykjanes Peninsula. If graben width is truly a reflection of the ambient state of stress before dike injection, then it implies that extensional faulting is difficult on Reykjanes. If it is harder for normal faults to form, it is likely that it is also harder for magma to rise. The same explanation may be invoked to account for the lack of true central volcanoes on Reykjanes Peninsula.

References cited:
Bjarnason, I. Th., Cowie, P.A., Anders, M.H., Seeber, L., and Scholz, C.H., 1993, The 1912 Iceland Earthquake rupture:growth and developmenet of a nascent transform system: Bulletin of the Seismological Society of America 83, 416-435.
Chadwick, W.W. Jr. and Embley, R.W.,1998, Graben formation associated with recent dike intrusions and volcanic eruptions on the mid-ocean ridge: Journal of Geophysical Research 103, 9807-9825.
Courtillot, V. Tapponier, P. and Varet, J., 1974, Surface features associated with transform faults: a comparison between observed examples and an experimental model: Tectonophysics 24, 317-329.
DeMets, C., Gordon, R., Argus, D. and Stein, S., 1994, Effect of recent revisions to the geomagnetic reversal time scalel on estimates of current plate motions: Geophysical Research Letters 21, 2191-2194.
Einarsson, P., 1991, Earthquakes and present-day tectonism in Iceland: Tectonophysics 189, 261-279.
Einarsson, P. and Brandsdottir, B., 1980, Seismological evidence for lateral magma intrusion during the July 1978 deflation of the Krafla volcano in NE Iceland: Journal of Geophysics 47, 160-165.
Einarsson, P. and Saemundsson, K., 1987, Earthquake epicenters 1982-1987 and volcanic systems in Iceland (map). In: Í hlutarins edli,  Thortsteinn I.Sígfusson (ed) Menningarsjódur, Reykjavik.
Eyolfsson, V., 1998, Kortlagning sprungna og nútíma eldvarpa í Fagradalsfjalli á vestanverðum Reykjanesskaga (Mapping of fractures and Holocene volcanic vents in Fagradalsfjall, Western Reykjanes Peninsula. In Icelandic). B.Sc. thesis, University of Iceland, Department of Geology and Geography, 70 pp, 1998.
Gudmundsson, A., 1980, The Vogar fissure swarm, Reykjanes Peninsula, SW-Iceland: Jökull 30, 43-63.
Gudmundsson, A., 1986, Mechanical aspects of postglacial volcanism and tectonics of the Reykjanes Peninsula, SW Iceland. Joural of Geophysical Research 91, 12,711-12,721.
Hreinsdottir, S., 1999, GPS geodetic measurements on the Reykjanes Peninsula, SW Iceland: crustal deformation from 1993 to 1998. M.Sc. Thesis, University of Iceland, 114 pp.
Johannesson, H., 1980, Jardlagaskipan og thróun rekbelta á Versturlandi (Evolution of rift zones in western Iceland, in Icelandic with English summary). Náttúrufraedingurinn 50, 13-31.
Jonsson, J.,1978, Jardfraedikort af Reykjanesskaga (Report on the geology of Reykjanes Peninsula) in Icelandic. Orkustofnun report OS-JHD-7831, Reykjavik.
Klein, F.W:, Einarsson, P. and Wyss, M., 1977, The Reykjanes Peninsula, Iceland , earthquake swarm of September 1972 and its tectonic significance: Journal of Geophysical Research 78, 5084-5099.
Macdonald, K., 1998, Linkages between faulting, volcanism, hydrothermal activity and segmentation on frast spreading centers. In: Buck, W.R., Delaney, P.T., Karson, J.A. and Lagabrielle, Y. (eds) Faulting and Magmatism at Mid-Ocean Ridges, Geophysical Monograph 106, American Geophysical Union.
Malinverno, A., 1993, Transition between a valley and a high at the axis of  mid-ocean  ridges: Geology, v.21, p.639-642.
Murton, B.J. and Parson, L.M., 1993, Segmentation, volcanism and deformation of oblique spreading centers: a quantitative study of the Reykjanes Ridge: Tectonophysics 222, 237-257.
Nakamura, K., 1970, En echelon features of Icelandic ground fissures: Acta Naturalia Islandica, Vol. II, no.8, Reykjavik.
Rubin, A.M. 1990, A comparison of rift zone tectonics in Iceland and Hawaii: Bulletin of Volcanology, v. 52, p.302-319.
Rubin, A.M., 1992, Dike-induced faulting and graben subsidence in volcanic rift zones: Journal of Geophysical Research 97, p.1839-1858.
Saemundsson, K., 1979, Outline of the geology of Iceland: Jökull 29, 7-28.
Saemundsson, K., 1995, Svartsengi geological map (bedrock) 1:25000. Orkustofnun, Hitaveita Sudurnesja and Landmaelingar Islands. Reykjavik.
Sempere, J-C., G.M. Purdy, and H. Schouten, 1990, Segmentation of the mid-Atlantic ridge between 24oN and 30o40’N: Nature 344, 427-431.
Sturkell, E., Sigmundsson, F., Einarsson, P., and Bilham, R., 1994, Strain accumulation 1986-1992 across the Reykjanes Peninsula plate coundary, Iceland, determined from GPS measurements: Geophysical Research Letters 21, 125-128.
Taylor, B., Crook, K. and Sinton, J., 1994, Extensional transform zones and oblique spreading centers: Journal of Geophysical Research 99, 19,707-19,718.
Vadon, H. and Sigmundsson, F., 1997, Crustal deformation from 1992 to 1995 at the Mid-Atlantic Ridge, Southwest Iceland, mapped by satellite radar interferometry: Science v 275, p.193-197.
Vogt, P.R., 1986, The present plate boundary configuration. In: Vogt., P.R., Tucholke, B.E., editors, The Geology of North America, Volume M, The Western North Atlantic Region.Geological Society of America, pp.189-204.
Ward, P.L., 1971, New interpretation of the geology of Iceland: Geological Society of America Bulletin 82, 2991-3012.
Whitlock, J.S., 1999, The tectonic geomorphology of the Molvik graben, Reykjanes Peninsula, Iceland. B.S. Thesis, Cook College, Rutgers University.