Elsevier

Journal of Structural Geology

Volume 23, Issue 9, September 2001, Pages 1359-1372
Journal of Structural Geology

Deformation partitioning inside a fissure swarm of the northern Icelandic rift

https://doi.org/10.1016/S0191-8141(01)00002-5Get rights and content

Abstract

Extension in an oceanic rift domain is accommodated by fissuring and faulting processes. The fissuring, including dike injection, involves extension and an increase in crustal volume, consistent with magma supply, as in oceanic ridge and hot spot contexts. The faulting mainly involves thinning and extension. The relationship between these two deformation modes is analysed in the northern rift of Iceland. In the active rift zone, deformation and volcanism occur within 2–3-km-wide fissure swarms separated by 10-km-wide zones where deformation is minor. Based on topographic and geological mapping, the local structure in a major fissure swarm was analysed. The fracture density increases near the faults, but also beyond their tips. The geometrical analysis of structures revealed that for the last 10,000 years, block faulting and tilting accommodated about one half of the extension, the other half resulting from fissuring. This distribution shows that even in an active volcanic zone with widespread magmatism, the stretching resulting from normal shear can play an important role. As estimated from the ages of eroded surfaces and lava flows, the deformation rates range between 1.5 and 15 cm/year. These values are compared with other rates estimated for different time intervals in the same area.

Introduction

In oceanic rifts, tectonic and volcanic processes display important variations in both space and time. Within the inner rift of an oceanic ridge, there is a close interaction between tectonics, plutonism and volcanism that occur simultaneously. In the shoulder areas, the tectonic processes dominate because the off-axis volcanism is scarce and not directly related to rift extension (Batiza et al., 1990).

The results from the FAMOUS program showed that the inner active zone of an oceanic rift is formed by a central part where the volcanic activity prevails, surrounded by two elongated domains where tectonic activity dominates, especially fissuring (Ballard and Van Andel, 1977). Observations on the seafloor during submersible dives, such as in the AMAR rift (Stakes et al., 1984, Kappel and Ryan, 1987) or the East-Pacific Rise (Macdonald et al., 1996), as well as field surveys in the rift of Iceland (Gudmundsson, 1987a, Gudmundsson, 1987b, Gudmundsson et al., 1992), revealed the presence of narrow fissure swarms separated by wide domains with minor deformation. This rather inhomogeneous distribution of the rift deformation can be regarded as a kind of structural partitioning. It was interpreted as a segmentation process occurring at different scales (Macdonald et al., 1988, Fox et al., 1991).

In a rift zone, the number and size of the fissure swarms clearly depend on the type of oceanic rift. In a typical undersea rift, a fissure swarm usually ranges from some hundreds of metres to a few kilometres in width. The usual length of the fissure swarm cannot be accurately determined because of the obvious limitations in submarine observation; however, the few data available indicate that this length may reach a few hundreds of meters. In more detail, the width of the fissure swarms observed in the slow-spreading rift systems ranges from 10 to 5000 m, whereas it may reach 10 km in the moderate to fast-spreading rift systems (Macdonald et al., 1996).

This lack of homogeneity in the distribution of the extensional deformation across an oceanic rift zone raises a major tectonic problem: how is the strain distributed in space and time inside the fissure swarms, and what is the nature of the deformation? In more detail, what is the relative importance of the increase in crustal volume related to magmatic supply on one hand and the stretching related to normal shear deformation on the other hand? An answer to these questions would help to constrain better the nature of the tectonic process contemporaneous with magma supply in oceanic rifts.

Three types of structures accommodate stretching: joints, fissures and normal faults, the latter often being related to tilted block systems. Although joints play an important role in the development and evolution of series of faulted-tilted blocks, especially where the amount of tilting becomes large enough to induce their reactivation as normal faults (Angelier and Colletta, 1983), their contribution to the total extension is generally small compared with that of other brittle structures. Furthermore, in areas of volcanic rocks, the amount of strain accommodated by tectonic jointing is almost impossible to estimate because of the importance of thermal contraction resulting from cooling. Cooling joints predate the joints of tectonic origin, partly inhibiting their development and influencing their orientation. The total amount of extension can be approximated by consideration of other brittle structures, such as normal faults, dikes, fissures and veins. Within submarine rift areas, the difficult survey conditions result in a major limitation when one attempts to quantify the amount of extension associated with the different structures of a fissure swarm. Cowie et al. (1993), however, have estimated that faulting accounts for 5–10% of the total deformation in the East-Pacific Rise. Similar results have been obtained in the Mid-Atlantic Ridge (Cowie et al., 1993). Such studies are scarce and the accuracy is poor, however, mainly because of the limited possibilities of undersea observation. Better accuracy can be obtained on land, based on extensive field surveys. De Chabalier and Avouac (1994) carried out a structural analysis in the Asal rift where the faulting accommodates 5–15% of the total extension.

Despite the limited number of data, all the results mentioned above concur to suggest that in an oceanic rift context, faulting plays a limited role (about 5–15% of the total extension), and hence that the largest contribution comes from fissuring and volume change related to magmatic supply. In such cases, the extension is focused in narrow deformed areas, with a typical width of a few kilometres. It should be noticed, however, that most observations have been made in narrow rift zones. The case of a wide oceanic rift zone, like the rift of northern Iceland, is certainly more complex. In counterpart, accurate observation can be carried out on land in the Icelandic rift, where most of the deformation and volcanism occurs inside sets of elongated fissure swarms. These parallel fissure swarms are several kilometres wide and trend NEE–SSW within a wide axial zone that trends N–S in northeastern Iceland (Saemundsson, 1979, Saemundsson, 1986, Gudmundsson et al., 1992). The width of this axial zone reaches 50 km (Fig. 1).

In this paper, we aim to provide not only an estimate of the amount and rate of extension in a typical fissure swarm inside an oceanic rift zone, but also an evaluation of the deformation partitioning between faulting and fissuring. For this purpose, we consider one of the fissure swarms of the rift of northern Iceland, and we focus on the deformation that has occurred in the last 10,000 years, according to the age of the deformed lava flows and erosion surfaces. To be as complete as possible in this structural analysis, we took into account not only the faulting and fissuring but also the block tilting in our analysis, which has never been done in such a context. Two independent and complementary methods have been adopted. The first method involves strain analysis based on geometrical observations of faults and open fissures in the field. The second method consists of determining the stretching rate produced by block faulting and tilting related to normal displacement along curved faults. This tilting was quantified by using a numerical restoration of the topography, which was initially flat prior to deformation (i.e. the top of fluid basalt flows). Both these methods are needed in order to obtain a reliable estimate of the total deformation.

A direct output of this study is the determination of the average strain rate during the last 10,000 years, which can be compared with other determinations. At the scale of the entire rift of Iceland, the kinematic reconstruction provided an extension rate of 1.9 cm/year for periods longer than 1 Ma (De Mets et al., 1994). For very short time periods (7 years, from 1986 to 1992) higher values were obtained, up to 5 cm/year (Hofton and Foulger, 1996), which result from the inhomogeneity of deformation in space and time as a consequence of the elastic response to plate separation. These authors suggested that the difference between the long- and short-term strain rates results either from a stress redistribution after a recent tectono–volcanic event in a viscoelastic-layered model, or from continued opening of dikes at depth in an elastic half-space. In this respect, our analysis of the deformation that occurred during the last 10,000 years in a fissure swarm of the rift of northern Iceland provides a velocity estimate at an intermediate time scale, and hence provides some constraints about the mechanical behaviour of the lithosphere.

Section snippets

Geological setting of the study area

Numerous fissure swarms are present in the active rift zone of Iceland, where they are more or less arranged in ‘en échelon’ patterns (Fig. 1) These fissure swarms affect large domains of basaltic flows younger than 800,000 years. This ‘en échelon’ pattern results from the slight obliquity (about 10°) of the plate separation relative to the trend of the rift zone in northeastern Iceland. The spreading direction is N104°E (De Mets et al., 1994), whereas the rift zone trends N–S. Note that such an

Structural analysis

Five main faulted zones, labelled F1–F5 from E to W affect this area (Fig. 3, Fig. 4). They are formed by numerous faults trending from N005° to N030° (Fig. 5). The faults display the classical aspect of the Icelandic-type open fault, previously described by Gudmundsson, 1995, Angelier et al., 1997. That is, the fault surface is nearly vertical and widely open. As demonstrated in the neighbouring Krafla area based on statistical measurements of the transverse horizontal opening and the vertical

Topography analysis

Part of the stretching is accommodated by slip along listric faults that tilt blocks. Such a tilted structure was described by Angelier et al. (1997) inside a small graben, south of the Krafla caldera. The amount of tilting is small (less than 5°) but can be accurately based on levelling measurements and topography restoration.

Structure of the fissure swarm

The Krafla fissure swarm to the north of the 1984 lava flow shows an asymmetric faulting pattern: a large single fault on the western side accommodates a vertical throw of about 40 m, whereas four fault sets are present on the eastern side. An area heterogeneously affected by fissures separates the western and eastern fault sets. The faults trend approximately N020° and generated eastward and westward tilts of the faulted blocks. The northward tilts are local, near fault tips or in transition

Conclusion

In areas submitted simultaneously to extension and to volcanism such as oceanic rifts or hot spot rifts, fissuring and faulting accommodates the deformation. Previous studies indicate that the fissuring process dominates over faulting. Our work based on detailed structural mapping, gives an example where the roles of these two processes in the uppermost brittle layer are nearly equal for a period of 10,000 years. Two phenomena can be invoked to explain the high proportion of the faulting-related

Acknowledgements

The help from the cooperation program French Ministère de Affaires Etrangères—Icelandic Ministry of Culture and Education and the French Embassy in Reykjavik is gratefully acknowledged. The CNRS–INSU projects Tectoscope and PNTS have supported this work.

References (41)

  • A. Björnsson et al.

    Current rifting episode in North Iceland

    Nature

    (1977)
  • A. Björnsson et al.

    Rifting of the plate boundary in North Iceland

    Journal of Geophysical Research

    (1979)
  • P.A. Cowie et al.

    Fault strain and seismic coupling on Mid-Ocean ridges

    Journal of Geophysical Research

    (1993)
  • O. Dauteuil et al.

    Deformation partitioning in a slow-spreading ridge undergoing oblique extension (Mohns Ridge–Norwegian Sea)

    Tectonics

    (1996)
  • R. Davies et al.

    Geodetic strain of Greece in the interval 1892–1992

    Journal of Geophysical Research

    (1997)
  • J.-B. De Chabalier et al.

    Kinematics of the Asal rift (Djibouti) determined from the deformation of Fieale volcano

    Geology

    (1994)
  • C. De Mets et al.

    Location of the Africa–Australia–India Triple Junction and motion between the Australian and Indian plates: results from an aeromagnetic investigation of the Central Indian and Carsberg Ridges

    Geophysical Journal International

    (1994)
  • P. Einarsson et al.

    Seismological evidence for lateral magma intrusion during the July 1978 deflation of Krafla volcano in NE-Iceland

    Journal of Geophysics

    (1980)
  • G.R. Foulger et al.

    Post-rifting stress relaxation at divergent plate boundary in Northeast Iceland

    Nature

    (1992)
  • P.J. Fox et al.

    The mid-Atlantic ridge (31°S–34°30′): temporal and spatial variations of accretionary processes

    Marine Geophysical Research

    (1991)
  • Cited by (39)

    • Normal fault growth in layered basaltic rocks: The role of strain rate in fault evolution

      2018, Journal of Structural Geology
      Citation Excerpt :

      The rift zone extends 80–100 km along strike (Fig. 9A), with a width of 4–10 km (Bjornsson et al., 2007). Magma is stored beneath the central volcano, in a reservoir at approximately 2.5–3.0 km depth and supplied at a rate of ∼1.6 km3 per year (Tryggvason, 1986; Dauteuil et al., 2001). Records of ground deformation, dating back to 1976, highlight pronounced and repeated episodes of steady inflation followed by rapid deflation (and subsidence), associated with rift zone extension (Bjornsson et al., 1978; Tryggvason, 1984; Rubin, 1992).

    View all citing articles on Scopus
    View full text