Description:

Journal of Structural Geology 32 (2010) 222–234 Contents lists available at ScienceDirect Journal of Structural Geology journal homepage: www.elsevier.com/locate/jsg Geometrically coherent continuous deformation in the volume surrounding a seismically imaged normal fault-array J.J. Long*, J. Imber Department of Earth Sciences, Durham University, Science Labs, Durham DH1 3LE, UK Article info Article history: Received 24 April 2009 Received in revised form 26 November 2009 Accepted 29 November 2009 Available online 7 December 2009 Keywords: Fault-propagation folds Normal fault Geometric coherence Seismic resolution Sub-seismic faults Abstract We calculated an apparent dip attribute, which was used to ascertain the spatial distribution of fault-related continuous deformation. The vertical component of displacement calculated from the continuous deformation acts to ‘fill-in’ missing displacement in the fault-throw profile. This result shows that apparently complex 3D patterns of continuous strain in the volumes surrounding the fault-array developed as part of a single, geometrically coherent fault-array. However, if this component of continuous deformation was not added to the throw profile, the fault-array could have been misinterpreted as a series of isolated fault segments with coincidental overlaps. This technique permits the analysis of continuous deformation structures, which are up to an order of magnitude smaller than previously described. In the study area, these structures are interpreted as small fault-propagation folds, forming in a shale-dominated cover sequence. The fault-propagation folds above the upper tip line of the mapped fault-array bifurcate upwards from the fault surface into three coherent lobes and resemble secondary fault segments. The near-constant along-strike length of the region of continuous deformation throughout the syn-rift sequence implies that the length of the fault-array was established at an early stage in its growth, prior to the establishment of a seismically-visible fault surface. © 2009 Elsevier Ltd. All rights reserved. 1. Introduction Fault-arrays comprise multiple fault segments (Peacock and Sanderson, 1991; Childs et al., 1995, 1996a; Willemse, 1997; Crider and Pollard, 1998; Peacock, 2002) that typically grow as geometrically coherent structures (Walsh and Watterson, 1991; Childs et al., 1995; Walsh et al., 2003b). Fault segments within these arrays can be hard-linked by discrete faults or soft-linked by zones of continuous deformation (Peacock and Sanderson, 1991; Trudgill and Cartwright, 1994; Childs et al., 1995; Walsh et al., 2003b). In seismic reflection profiles, continuous deformation is commonly expressed as the rotation, thickening or thinning of strata within the deformed volume between soft-linked faults. Continuous strains result from any combination of plastic deformation and/or small-scale faults or fractures below the resolvable limits of seismic data (e.g., Fig. 1, Steen et al., 1998; Townsend et al., 1998). The specific limits at which structures can be resolved depend on the depth of the feature and the quality of the seismic data. Geometric coherence is the concept that faults and fault-related strain maintain regular and systematic geometries and relationships. * Corresponding author. Fax: +44 0191 3342301. E-mail address: jonathan.long@durham.ac.uk (J.J. Long). 0191-8141 $ – see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsg.2009.11.009 Through the evolution of a fault-array, throws due to faults and associated continuous deformation should together produce smoothly varying displacement profiles, which resemble that of a single fault (Walsh et al., 2003b). If a fault-array has maintained geometric coherence this must suggest kinematic coherence, which is the systematic and linked accumulation of displacement across the fault-array (Walsh and Watterson, 1991). The fault-propagation model of Marchal et al. (1998) predicts that fault-arrays evolve by the coherent growth and linkage of secondary faults, which are small faults that form at the propagating tips of a primary fault segment. Secondary faults can form as separate fault segments soft-linked to the primary fault via relay zones, or as hard-linked structures that bifurcate from the main fault surface. This fault growth model can be applied to both horizontal and vertical tip lines (Marchal et al., 1998, 2003). Fault propagation will ultimately result in segmented fault tip lines, as shown by observations of naturally occurring faults (McGrath and Davison, 1995; Childs et al., 1996b; Marchal et al., 2003; Kristensen et al., 2008). Fault-propagation folds are manifestations of fault-related continuous deformation that develop ahead of a propagating tip line and which deform the free surface (Withjack et al., 1990; Corfield and Sharp, 2000; Sharp et al., 2000; Gawthorpe et al., 2003; Finch et al., 2004; Jackson et al., 2006; White and Crider, 2006; Ford et al., 2007). In the case of synsedimentary normal faults, fault-propagation folds are expressed as monoclines whose axes lie parallel to the strike of the fault-array. The development of a synsedimentary monocline results in the main depocentre being offset into the hanging wall, in comparison with emergent synsedimentary normal faults where the depocentre is located in the immediate hanging wall of the fault (Sharp et al., 2000; Gawthorpe et al., 2003). Scaled analogue and numerical models of extensional fault-propagation folds above rigid basement fault blocks have shown that the amplitudes and wavelengths of monoclines are controlled by the dip of the basement fault and by the rheology of the overlying strata. These models also show that the mechanical stratigraphy controls whether fault-arrays within the cover are isolated or hard-linked to the basement fault (Withjack and Callaway, 2000; Finch et al., 2004). The aim of this paper is to describe the three-dimensional (3D) geometry of the brittle and continuous deformation at and beyond the upper tips of a synsedimentary normal fault-array in the Inner Moray Firth basin (IMF). However, the method and applications are not limited to synsedimentary settings, or to the IMF. We use interpretations of 3D seismic reflection data to test the idea that deformation at seismically imaged fault tips, including continuous deformation, is geometrically coherent. The methodology described here allows us to make inferences about the complex geometric arrangement of secondary faults, on which the offsets are below the resolution of the seismic data and are therefore manifest, at least in part, as continuous deformation at the scale of observation. The scale of structures studied in this paper (maximum throw ca. 115 ms) is greater than those described by Kristensen et al. (2008), but less than those of Corfield and Sharp (2000). 2. Geological setting The study area is located in the Inner Moray Firth basin (Fig. 2). The main phase of NW–SE extension occurred during the Late Jurassic to Early Cretaceous (represented by the mapped H5–BCU interval; Figs. 3 and 4). There is little evidence for active extension during the Triassic to Mid-Oxfordian, which is represented by the mapped Triassic chert to Horizon H5 interval (Figs. 3 and 4). This extension produced the regional NE–SW trending normal fault set (Fig. 2a) (Underhill, 1991b; Thomson and Underhill, 1993) and associated half-graben basin fill (Fig. 2) (Underhill, 1991b). Sediment packages thicken towards the north–west along the Helmsdale–Wick boundary fault systems (Fig. 2). Subsequent Cretaceous sedimentation records gentle regional subsidence. Post-Cretaceous reactivation of some large-offset faults has occurred in the IMF. Faults that offset the BCU are recognised as being reactivated, but no evidence is found for post-Cretaceous deformation having reactivated faults in the immediate study area. The study focuses on deformation within the Middle to Upper Jurassic succession, which encompasses the uppermost part of the pre-rift and lowermost syn-rift sequences (Figs. 3 and 4). Regionally, the onset of syn-rift sedimentation was marked by deposition of the H5 horizon (Intra Oxfordian reflector). Correlation of seismic reflectors with nearby wells shows that the mapped syn-rift sequence (H5–H1) is shale-dominated and overlies a sandstone-dominated pre-rift sequence, which includes Horizon H6 (Fig. 4). The mapped fault-array consists of three NE–SW trending enechelon segments (F1–F3; Fig. 3) separated by two relay zones. Aggregate displacement on the array decreases southwest towards the mapped lateral tips. The studied faults dip towards the NW, antithetic to nearby large-offset faults that dip towards the SE (Fig. 3). The mapped H1–H5 sequence thickens from footwall to hanging wall across F1, F2 and F3 (Fig. 3). Fault scraps in the IMF show no evidence for footwall erosion, which suggests that F1, F2 and F3 were either blind faults, or were synsedimentary faults that were blanketed with sediments during deposition of the Late Jurassic to Early Cretaceous syn-rift sequence (Underhill, 1991a,b; Nicol et al., 1997; Childs et al., 2003). Analysis of throws on the mapped faults shows that F1, F2 and F3 have vertical displacement gradients greater than ca. 0.16 (Fig. 5). This value is consistent with vertical displacement gradients calculated for synsedimentary faults in other areas (Nicol et al., 1996, 1997; Cartwright et al., 1998; Walsh et al., 2003a). Ключевые слова: e, r, o