Influence of carbonate facies on fault zone architecture

Journal of Structural Geology 65 (2014) 82e99

Contents lists available at ScienceDirect

Journal of Structural Geology journal homepage: www.elsevier.com/locate/jsg

Influence of carbonate facies on fault zone architecture E.A.H. Michie a, *,1, T.J. Haines a, D. Healy a, J.E. Neilson a, N.E. Timms b, C.A.J. Wibberley c a

School of Geosciences, King’s College, University of Aberdeen, Aberdeen AB24 3UE, UK Department of Applied Geology, Curtin University, Bentley Campus, Perth, Western Australia 6845, Australia c TOTAL, CSTJF, Avenue Larribau, 64108 Pau Cedex, France b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 4 December 2013 Received in revised form 4 April 2014 Accepted 23 April 2014 Available online 4 May 2014

Normal faults on Malta were studied to analyse fault propagation and evolution in different carbonate facies. Deformation of carbonate facies is controlled by strength, particle size and pore structure. Different deformation styles influence the damage characteristics surrounding faults, and therefore the fault zone architecture. The carbonates were divided into grain- and micrite-dominated carbonate lithofacies. Stronger grain-dominated carbonates show localised deformation, whereas weaker micritedominated carbonates show distributed deformation. The weaker micrite-dominated carbonates overlie stronger grain-dominated carbonates, creating a mechanical stratigraphy. A different architecture of damage, the ‘Fracture Splay Zone’ (FSZ), is produced within micrite-dominated carbonates due to this mechanical stratigraphy. Strain accumulates at the point of juxtaposition between the stronger graindominated carbonates in the footwall block and the weaker micrite-dominated carbonates in the hanging wall block. New slip surfaces nucleate and grow from these points, developing an asymmetric fault damage zone segment. The development of more slip surfaces within a single fault zone forms a zone of intense deformation, bound between two slip surfaces within the micrite-dominated carbonate lithofacies (i.e., the FSZ). Rather than localisation onto a single slip surface, allowing formation of a continuous fault core, the deformation will be dispersed along several slip surfaces. The dispersed deformation can create a highly permeable zone, rather than a baffle/seal, in the micrite-dominated carbonate lithofacies. The formation of a Fracture Splay Zone will therefore affect the sealing potential of the fault zone. The FSZ, by contrast, is not observed in the majority of the grain-dominated carbonates. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Fault zone architecture Carbonates Fault damage prediction Scaling attributes Normal faults Fluid flow

1. Introduction The details of fault zone architecture are important when considering fluid flow in the subsurface because faults can act as barriers, conduits or a combined conduit/barrier (Chester et al., 1993; Knipe, 1993; Antonellini and Aydin, 1994; Bruhn et al., 1994; Caine et al., 1996; Evans et al., 1997; Lockner et al., 2000; Billi et al., 2003; Agosta et al., 2007; Molli et al., 2010). Characterisation of the fault zone structure and analysis of fault propagation can help to provide a better understanding of the petrophysical properties of a fault zone, and therefore its influence on fluid flow.

* Corresponding author. Tel.: þ44 (0) 1790 753 472. E-mail addresses: [email protected], [email protected], e. [email protected] (E.A.H. Michie), [email protected] (T.J. Haines), d.healy@ abdn.ac.uk (D. Healy), [email protected] (J.E. Neilson), [email protected] (N.E. Timms), [email protected] (C.A.J. Wibberley). 1 Presently address: Badley Geoscience Ltd, North Beck House, North Beck Lane, Hundleby, Spilsby, Lincolnshire PE23 5NB, UK. http://dx.doi.org/10.1016/j.jsg.2014.04.007 0191-8141/Ó 2014 Elsevier Ltd. All rights reserved.

The established fault zone architectural model was developed from observations of fault zones in siliciclastic and basement crystalline rocks, and describes a ‘fault core’ surrounded by a ‘damage zone’, with an exponential decrease in damage into the protolith (Chester and Logan, 1986; Caine et al., 1996; Vermilye and Scholz, 1998; Mitchell and Faulkner, 2009; Savage and Brodsky, 2011). The fault core is a zone (continuous or patchy) of intense deformation, where most of the fault displacement is accommodated. It is composed of fault rock, such as breccias, cataclasites and gouges, which commonly show little evidence of the primary fabric of the protolith (Engelder, 1974; Sibson, 1977; Groshong, 1988; Caine et al., 1996; Evans et al., 1997; Agosta and Kirschner, 2003; Chester et al., 2004; Berg and Skar, 2005; Agosta and Aydin, 2006; Tondi, 2007; Mitchell and Faulkner, 2009; Faulkner et al., 2010). The damage zone is an approximately tabular halo of fractured rock that accommodates smaller deformation by micro and macrofractures, tension gashes and subsidiary faulting, which are related to the fault growth (Caine et al., 1996; Chester et al., 2004; Agosta et al., 2007; Gaviglio et al., 2009; Mitchell and Faulkner,

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

2009; Faulkner et al., 2010; Hausegger et al., 2010). It has been demonstrated that fault zones within carbonate facies can also conform to this architecture (e.g. Agosta and Kirschner, 2003; Storti et al., 2003). Conversely, other fault zones within carbonates are documented to have added architectural components, such as intensely and a weakly deformed damage zones (e.g. Micarelli et al., 2006; Ferrill et al., 2011). Variations to this established fault zone architectural model are documented for all lithologies, including added and/or modified architectural components (e.g. Childs et al., 1997; Micarelli et al., 2006; Faulkner et al., 2010). Fault zone architecture also varies along fault strike due to interactions of faults, which creates segment linkage/relay zones (Peacock and Sanderson, 1991; Huggins et al., 1995). For example, dual fault systems bounding an intensely deformed zone may have a relay ramp (Larsen, 1988; Childs et al., 2009). The different fault zone structures influence petrophysical properties, such as porosity and permeability, of the faults (e.g. Faulkner et al., 2010). The architecture of fault zones is controlled by how different lithologies and lithofacies deform (Aydin, 2000; Agosta and Aydin, 2006; Shipton et al., 2006; Riley et al., 2010; Loveless et al., 2011; Jeanne et al., 2012). In sedimentary rocks, the strength and texture (e.g. grain size, matrix, porosity) of the primary facies controls the deformation style (Hugman and Friedman, 1979; Shipton et al., 2006; Riley et al., 2010). The varied deformation styles causes the deformation mechanisms that accommodate the stress to differ in each lithology/facies. A mechanical stratigraphy also has a large influence over the evolution of the fault zone architecture, as the angle of dip will vary in layers of differing tensile strength, and the variable dip angles can create different architectures (Peacock and Zhang, 1994; Sibson, 1996; van Gent et al., 2010). These factors can create areas such as pull-aparts, dilatant jogs and

83

asperity bifurcations (Childs et al., 1995; Peacock and Sanderson, 1995; Childs et al., 1996, 1997; Schöpfer et al., 2007a). Detailed descriptions of different fault zone architectures, along with quantification of their petrophysical properties helps to predict fluid flow in fault zones. For example, a widely-held assumption is that the permeability in a damage zone increases into the fault zone with increased fracture-related porosity towards the principal slip surface (Caine et al., 1996; Evans et al., 1997; Lockner et al., 2000; Billi et al., 2003; Micarelli et al., 2006; Agosta et al., 2007; Agosta, 2008; Balsamo et al., 2010). However, fault zones may not conform to this traditional petrophysical characterisation of fault zones, because the spatial distribution of deformation will differ (e.g. Childs et al., 1997; Faulkner et al., 2003, 2010). Not only can microstructures help to assess the hydraulic behaviour of fault zones, they can also be used to analyse the seismic versus aseismic behaviour of faults. For example, the occurrence of mirror-like slip surfaces, open or closed fractures, breccias/pulverised rock and/or pressure solution seams, and the cross-cutting relationships of these microstructures are used to interpret the seismic cycle of the fault (Agosta and Aydin, 2006; Billi and Di Toro, 2008; Fondriest et al., 2013; Gratier et al., 2013). In this paper, we examine several normal fault zones on Malta, with varying displacement, from <0.1 m up to 90 m, so as to understand the propagation and evolution of faults in different carbonate lithofacies. Furthermore, we describe in detail the fault zone architecture and discuss the importance of fault zone architecture for fluid flow prediction. This analysis is achieved by using detailed field mapping, as well as spatial fracture analysis using circular scanlines. We will use these data to introduce a new architecture for fault zones. We also attempt to predict the fault zone architecture by understanding the influence of mechanical stratigraphy on deformation style.

Fig. 1. A: Map of Malta showing the main faults (after Pedley et al., 1976), particularly, the ENE-WSW trending Victoria Lines fault (VLF) and the NWeSE Il Maghlaq fault (IMF). The two main localities on Malta used for examination of several faults are highlighted in boxes. Detailed maps of the two main localities are also shown; B: Ras ir Raheb exposing several normal faults and C: Madliena Tower exposing a graben bounding fault (the VLF). The fault zones studied in detail at Ras ir Raheb are shown in bold, and have 0.52 m, 5.1 m, 7 m, 11.7 m and 25 m displacement fault zones.

84

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 2. A summary sedimentary column through the Oligo-Miocene succession from Ras ir Raheb, where the Xlendi Member is absent. Stratigraphic names from Pedley et al. (1976) and Pratt (1990). The log is shown with numbered microphotographs of microscale textures for units in sequence, a vertical strength profile for carbonate lithologies and carbonate lithofacies.

2. Geological setting The Maltese Islands are composed of different limestone facies that were deposited on the shallow Mediterranean Pelagian Platform (Pedley et al., 1976; Dart et al., 1993). Graben structures, such as the Maltese graben system formed during the Miocene to Quaternary, due to rifting in response to an NeS extension, creating many extensional basins in the foreland of the Sicilian ApennineMaghrebian fold and thrust belt (Dart et al., 1993). A conjugate normal fault system was produced due to the NWeSE trending Pantelleria Rift system SW of Malta and the ENE-WSW horsts and

grabens in the Malta Platform (Pedley et al., 1976; Dart et al., 1993; Bonson et al., 2007; Putz-Perrier and Sanderson, 2010). Several smaller scale rift systems dissect the Maltese stratigraphy, producing the most dominant extensional faults seen on the Maltese Islands, such as the Victoria Lines fault, a large graben-bounding fault with 90 m of displacement (Dart et al., 1993; Putz-Perrier and Sanderson, 2010, Fig. 1). NWeSE trending faults are also observed on Malta, including the Il Maghlaq fault, but are not discussed within this paper (Dart et al., 1993; Bonson et al., 2007, Fig. 1). The stratigraphic layers in the study area range from the Lower Coralline Formation up to the Middle Globigerina Member of the

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

85

Fig. 3. Examples of micrite-dominated and grain-dominated carbonate textures. A and B show two different micrite-dominated carbonates. A is a coarse grained but micritesupported rock, whereas B is very fine grained. C and D are coarse-grained, grain-dominated carbonates. 1: Lepidocyclina, 2: large broken fossil clast typically echinoids or bryozoans, 3: Geniculate coralline algae, 4: miliolid foraminifera, and 5: nummulitic foram.

Globigerina Formation (Fig. 2), of Oligocene to Miocene age (Pedley et al., 1976; Dart et al., 1993). The carbonates are limestone with very little dolomite content. These rocks were deposited either prior to rifting (Lower Coralline Formation and Lower Globigerina Limestone Member) or during rifting (Middle Globigerina Limestone Member onwards) (Dart et al., 1993). The Lower Coralline Limestone formation (LCL) is a bioclastdominated limestone containing benthic foraminifera and coralline red algae. It consists of packstones and rudstones, with occasional grainstones and boundstones. Up to 140 m in stratigraphic thickness, it is capped by a hardground (Pedley et al., 1976; Dart et al., 1993; Bonson et al., 2007). We studied part of the Xlendi Member (Fig. 3C), the uppermost strata of the Attard Member (Fig. 3D) and the Il Mara Member (Fig. 3A). The Il Mara Member directly overlies the Attard Member at Ras ir Raheb, with no Xlendi Member present (Fig. 2). The Globigerina Limestone Formation (GL) is composed of finegrained carbonates, dominated by Globigerina foraminifera (Pedley et al., 1976; Dart et al., 1993). We studied the Lower and Middle Globigerina Limestone Members (LGL and MGL respectively) of the GL (Figs. 2 and 3B).

Detailed maps and cross sections (Figs. 4e8) were constructed for six fault zone to illustrate zone architecture, secondary fracture patterns and fault rock distributions. To provide a consistent reference frame, where absent, fault zones were projected into horizontal map views. The fault zones at Ras ir Raheb are well-exposed cross-sectionally in cliff view but have narrow platforms for map view (Figs. 4e7). The fault zone at Madliena Tower has an exceptional plan view of the fault zone, but no cross sectional view (Fig. 8). 3.1.1. Damage zone and fault core width measurements The width of damage zones and fault cores were measured. The total width of a damage zone was defined as the distance between the first macrofractures observed in the hanging wall and footwall. Macrofractures are defined as easily visible to the naked eye in exposure. The width of a fault core was more difficult to define. For example, a bound zone containing fault rock, deformed rock and intact rock can be considered as a fault core (e.g. Faulkner et al., 2003; Mitchell and Faulkner, 2009). Alternatively, the fault core may be simply the thickness of fault rock surrounding each slip surface. That definition was used for our study. 3.2. Characterisation of fracture distribution in fault damage zones

3. Methods 3.1. Mapping and section construction The Ras ir Raheb west coast section and Madliena Tower on the east coast were studied (Fig. 1). Ras ir Raheb has several fault zones ranging from 0.52 m to 25 m dip-slip displacement. Madliena Tower exposes the Victoria Lines fault (VLF) with a c.90 m dip-slip displacement fault zone, including c.60 m accommodated on the principal slip surface (90 m value from House et al., 1961). These fault zones formed during a Miocene to Quaternary tectonic event and strike ENE-WSW (Dart et al., 1993, Fig. 1).

Scanlines are used to characterise fracture patterns across fault zones. Measurements of the intensity, density and mean tracelength of fractures proximal to the mapped faults were made using circular scanlines (scanlines with circular windows) along transects normal to and parallel to the map strike of a fault plane (Mauldon et al., 2001; Rohrbaugh et al., 2002). The intensity and density were calculated using the intersections and end points of each fracture trace within the scanline transects (Mauldon et al., 2001; Rohrbaugh et al., 2002). The density (P20) is the mean number of fractures per unit area, and is calculated using the number of fracture endpoints in a circular window. The intensity

86

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 4. A: Field photograph and B: cross section of a 0.52 m displacement fault zone at Ras ir Raheb. Red lines on field photograph are slip surfaces. The key shows colours for stratigraphic units and fault zone material, plus symbology for different features in Fig. 4 to 8. (I) Lower hemisphere stereonet showing poles to both faults and fractures trending ENE-WSW. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(P21) is the mean total trace-length of fracture traces per unit area, and is calculated using the number of fracture intersections with the circular window (Mauldon and Dershowitz, 2000; Mauldon et al., 2001). 80 scanlines were used to characterise fractures across six fault zones ranging in dip-slip displacement from 0.52 m to 90 m. Fractures of all orientations, including perpendicular and parallel to the fault, have been measured in each lithofacies for individual fault zones. Fracture data from scanlines enabled the comparison of deformation in different carbonate lithofacies across faults of different displacement. 3.3. Strength tests Strength tests using a Schmidt Hammer were completed up the stratigraphic section, to analyse the strength of the different lithofacies (Fig. 2). A type L Schmidt Hammer was used to measure the in situ elastic rebound properties of the rock, following the method outlined by Morris et al. (2009) and Ferrill et al. (2011). Measurements were made on either vertical or horizontal surfaces, creating variations induced by gravity, which were corrected when calculating the Uniaxial Compressive Strength (UCS) of the rocks. In each location, 10 Schmidt Hammer measurements were taken, within an area of <10 cm2. From these values the highest and lowest anomalous readings were removed and an average calculated (Fig. 2). Schmidt Hammer rebound values (R) were converted into UCS (MPa) using the 25 kN/m3 curve on the correlation graph (cf.

Hendron, 1968; Brown, 1981). The 25 kN/m3 line was used, as the average density of the Maltese carbonates has been calculated as 2550 kg/m3. Although it is unlikely that present day UCS measurements from Schmidt Hammer values represent rock strength at the time of deformation (Morris et al., 2009), the values provide a good proxy for the relative difference in strengths for the different lithologies in this sequence. 3.4. Oriented sample collection and petrographic analysis Samples were systematically collected around the six fault zones to characterise all areas of each fault zone. These samples were taken from each lithofacies, at specific distances from the slip surfaces. 150 samples were used to make petrographic thin sections, representing different architectural components for different displacements in each lithofacies. Thin sections were used to document textural variations from the protolith into the fault zones as a function of pore/particle size, type and sorting, cementation and microstructures. 4. Results 4.1. Protolith lithofacies Detailed examination of the protoliths via graphical sedimentary logging and petrographic analyses were used to develop a

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

87

Fig. 5. A and B: 5 m displacement fault zone. C and D: 7 m displacement fault zone. A and C: Field photographs, showing cliff-face views of two fault zones with similar displacement at Ras ir Raheb. Red lines indicate slip surfaces. B and D: plan view maps of displacement fault zones. (I) and (II) Lower hemisphere stereonet showing poles to both faults and fractures striking ENE-WSW, dipping southward. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

lithofacies scheme that permitted comparisons of the response of similar rocks to faulting in different fault zones. The rocks were subdivided into two lithofacies, reflecting fundamentally different lithological properties: micrite-dominated; and grain-dominated

carbonates (Fig. 2). The micrite-dominated carbonate lithofacies has a homogeneous texture, consisting of a micrite matrix (<4 mm in size) with fine (<50 mm diameter) fossil particles, usually Globigerina forams (Fig. 3B). It can include dispersed, larger fossil clasts

88

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 6. A: Field photograph and B: plan view map of an 11.7 m displacement fault zone at Ras ir Raheb. (II): an inset of a cross section of this fault zone. Red lines on the photograph are slip surfaces within the fault zone. (I) Lower hemisphere stereonet showing poles to both faults and fractures striking ENE-WSW. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 7. A: Field photograph and B: plan view map of a 25 m displacement fault zone at Ras ir Raheb. Red lines on the photograph are slip surfaces within the fault zone. (I) Lower hemisphere stereonet showing poles to both faults and fractures striking ENE-WSW, and dipping south. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

89

carbonates next to the grain-dominated carbonates. In the graindominated carbonates, the fault zones form single slip surface (Figs. 4e7). Also, faults with <1 m displacement form single slip surface for both lithofacies (Fig. 4). However, faults with displacements >1 m consist of fault zones with more than one slip surface within the weaker micrite-dominated carbonates in the hanging wall (Figs. 5e8). The slip surfaces occur in relatively narrow zones, and are often connected along strike. The slip surfaces connect down dip, merging to a principal slip surface (Figs. 5e7). Displacement changes along the trace of every fault, including the principal slip surface for each fault zone. Locally, these variations may mean that a subordinate fault has more displacement than the adjacent principal slip surface, but that kinematic relationship is not common. Fault dips in micrite-dominated carbonates are on average less than in the grain-dominated carbonates (Fig. 9). The average angle of dip in grain-dominated carbonates is 80 , whereas the average dip of slip surfaces in the micrite-dominated carbonates is 67 (Fig. 9). Displacement gradient profiles of the principal and bounding slip surfaces were created for the fault zones at Ras ir Raheb (Fig. 10BeD, for Figs. 4e7). For each of the three fault zones, about 20 m of along strike measurements of displacement values were used to create the profiles. Typical principal and outer bounding slip surfaces are shown in Fig. 10E, and they are the two outermost slip surfaces for fault zones. A simple prediction for the displacement profile of a typical pair of overlapping faults (Larsen, 1988; Peacock and Sanderson, 1991; Childs et al., 1995; Huggins et al., 1995) is shown in Fig. 10A. However, the actual displacement profiles do not match this prediction (Fig. 10BeD). Some bounding slip surfaces have similar displacement gradients, where the increase/decrease displacement is in the same direction (e.g. Fig. 10D), while other bounding slip surfaces show part of a displacement ellipse associated with a fully exposed fault trace (e.g. Fig. 10B, C). Overall, the number of slip surfaces increases with displacement, but some variations from this simple relationship exists (e.g. Fig. 11A, B). The number of slip surfaces also varies along single faults, as they are observed to bifurcate and join with other slip surfaces (Figs. 5, 6, 11C, D). Where the number of slip surfaces is greater, the slip per surface is less, whereas it is greater per surface where the number is smaller. Fig. 8. Plan view map of a portion of 90 m displacement fault zone (the VLF) at Madliena Tower. A white box in the satellite image for this fault zone shows map area. The trace of the principal slip surface is also shown on the satellite image. Satellite image from Google Earth 7.1.2. 2013. Madliena Tower, 35 560 10.5600 N, 14 280 11.0300 E, elev 6 m, [Accessed 18 November 2013].

(>100 mm) supported by micritic matrix (Fig. 3A). This lithofacies can be classified as wackestones and packstones, and often resembles chalk. Grain-dominated carbonates are grain-supported and contain coarse (>250 mm) fossil clasts (Fig. 3C, D). This lithofacies can be classified as packstones, rudstones and grainstones, with occasional boundstones. The Lower Coralline Limestone (LCL) is classified as ‘grain-dominated’, and the Globigerina Limestone (GL) is ‘micrite-dominated’, although a slight overlap exists in this division, as the Il Mara Member of the LCL contains both grain- and micrite-dominated carbonate beds (Fig. 2). Undeformed micritedominated and grain-dominated carbonate samples have similar porosity values and both lack cementation. 4.2. Field structural data 4.2.1. Geometry and location of slip surfaces Given slickenline pitches of about 80 , the faults are essentially normal dip-slip. These faults downthrow micrite-dominated

4.2.2. Fracture distribution Within fault zones with 1 m displacement, the fractures are in close proximity to the slip surface with only a few fractures observed within the damage zone, and typically in the hanging wall rather than the footwall (Fig. 4). At greater displacements, >1 m, more fractures are observed in the damage zones within the micrite-dominated carbonates, whereas the grain-dominated carbonates have secondary fractures only close to the principal slip surface. An intense zone of fractures, bound either side by slip surfaces, is observed for all fault zones with displacements >1 m in the micrite-dominated carbonates (Figs. 5e8). These fractures have the same orientation as the slip surfaces in most locations, but occasionally join the slip surfaces at c.45 orientation to the slip surfaces. Outside of this zone, fractures are more widely spaced and less common (Fig. 6). Below 1 m throw, only one main slip surface is observed (Fig. 12A), with a lower fracture density of c.200 m2, compared with >600 m2 at greater displacements (Fig. 12A vs. C). In faults with a throw greater than 1 m, an abrupt increase in both intensity and density of fractures between two bounding primary slip surfaces is observed (Fig. 12). The intensity of fractures within the bound zone ranges from 25 m1 up to 50 m1, but does not correlate to displacement magnitude because a fracture intensity of

90

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 9. Frequency plot of slip surfaces as a function of dip angle and lithofacies. The average dip for the LGL member is 67.7, 66.2 for the MGL member and 79.9 for the LCL formation.

Fig. 10. A: General prediction for the displacement distribution between principal slip surface and outer bounding slip surface. BeD: Displacement profiles for paired principal and outer bounding slip surfaces of three fault zones, taken from their entire length of exposure. Almost whole displacement profiles are seen for B and C, along with slip surfaces dying out in the same direction along D. E: Illustration of the principal slip surface and outer bounding slip surface.

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

91

Fig. 11. Graphs showing the number of slip surfaces as a function of displacement for all six faults, and the variation in the number of slip surfaces for particular faults. A: minimum number of slip surfaces observed along the fault zones. B: maximum number of slip surfaces observed along the fault zones. C: Example showing the variation in the number of slip surfaces along the fault strike for the 7 m displacement fault zone. D: Example showing the variation in the number of slip surfaces along the fault strike for the 90 m displacement fault zone.

50 m1 is observed for a fault zone of 5 m displacement, whereas a fracture intensity of 25 m1 is observed for a fault zone of 90 m displacement (Fig. 12B, E). This bound zone of intense fracturing with several slip surfaces is called the Fracture Splay Zone (FSZ) (Fig. 13). Examples of the location and geometry of the FSZ are shown in Figs. 5e8. Outside of the FSZ is a weakly deformed damage zone (WDDZ), which contains fewer fractures than the FSZ (Fig. 13). The fractures tend to be evenly distributed in the WDDZ (Fig. 12BeE) (WDDZ terminology from Micarelli et al., 2006). The WDDZs stops abruptly at undamaged protoliths (Fig. 13). 4.2.3. Fault rock distribution The location of fault rock along fault surfaces was identified using geological mapping. Fault rock is found on faults with greater than 0.5 m displacement and tends to occur in thin pods along the fault trace for displacements of 25 m (Figs. 5e7). A smaller number of slip surfaces for a given displacement correlates to a greater amount of fault rock along the fault traces, as for example more fault rock occurring with a fault having less displacement but fewer surfaces (e.g. 5 m vs. 7 m displacement faults, Fig. 5A, B). For the large displacement (c.60 m) VLF principal slip surface, fault rock is sufficiently abundant to create a continuous core along the entire fault trace with core width variation (Fig. 8). Subtle differences in the architecture, such as interaction of slip surfaces, can lead to variations in fault rock abundance. FSZs with greater displacement have more abundant fault rock (Fig. 8). 4.3. Block models 3D block models were built using map data to show the geometry of slip surfaces and fault zones, as a function of the change in displacement in both strike and dip directions of the

faults (Fig. 14). The FSZ is shown in three of the block models (Fig. 14BeD), identified by several slip surfaces confined to a narrow zone in the weaker micrite-dominated layer (Fig. 13). Down fault dip these slip surfaces are observed to attach to the principal slip surface at roughly 45 . Significant interaction of all slip surfaces occur along fault strike. FSZs are present in faults with >1 m displacement that cut micrite-dominated carbonates (Fig. 14A vs. BeD). At greater displacements (between 12 m and 25 m) FSZs have more slip surfaces that detach from other slip surfaces within the FSZ, rather than detaching from the principal slip surface (Fig. 14D). 4.4. Scaling of fault zone architectural components 4.4.1. Damage zone scaling The width of FSZs generally increase with greater displacement (Fig. 15A). However, this relationship varies as a function of the number of slip surfaces. An increase in the number of slip surfaces decreases the width of the FSZ, and vice-versa. The total width of the damage zone (FSZ plus WDDZ) increases with displacement in a roughly logelinear relationship (Fig. 15B). This trend varies with the number of slip surfaces, such that a greater number of slip surfaces within the FSZ correlates to a narrower surrounding WDDZ. In contrast, a smaller number of faults in the FSZ increases both the width of the FSZ and the WDDZ. Where FSZ width increases with displacement for micritedominated carbonates, FSZs are only present in grain-dominated carbonates for displacements 7 m. Further, the target graindominated carbonates are absent at Madliena Tower with the VLF, so insufficient data exist to establish a relationship between FSZ widths and displacement for grain-dominated carbonates. Finally, as would be expected, FSZ and WDDZ widths are generally

92

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 12. Fracture intensity and density plots gathered from circular scanlines, across 5 fault zones, in different lithofacies. A: density data for a 0.52 m displacement fault. B: intensity data across a 5 m displacement fault. C: intensity data across an 11.7 m displacement fault. D: density data across a 25 m displacement fault. E: intensity data across a 90 m displacement fault zone with LCL (grain-dominated) in the footwall (from 0 to 16 m distance) and LGL (micrite-dominated) in the FSZ (>16 m distance). Dark green lines are fracture profiles of the grain-dominated carbonates, and lighter blue lines are the fracture profiles of the micrite-dominated carbonates. The density (P20) is the mean number of fractures per unit area. The intensity (P21) is the mean total trace-length of fracture traces per unit area (Mauldon and Dershowitz, 2000; Mauldon et al., 2001). Thin red vertical lines indicate subsidiary faults in the FSZ and the thick red vertical lines indicate the bounding slip surfaces encompassing the FSZ. Dashed black vertical lines represent the extent of the fault zone; between the thick vertical lines and dashed vertical lines is the WDDZ. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

greater in the micrite-dominated rather than grain-dominated carbonates (Fig. 15A, B). 4.4.2. Fault core scaling The average width of all the fault rock increases with overall fault zone displacement (Fig. 15). However, using this relationship to state a certain fault core thickness occurs at specific displacements may not be accurate because there are several slip surfaces with no fault core observed. Therefore, a second average width of the fault rock has been calculated that takes into consideration the width of fault rock on every slip surface within the FSZ, including slip surfaces with no fault rock (Fig. 15C). This averaging technique results in lower fault core widths for the lower displacement faults because there are many slip surfaces with no fault rock observed. However, in the highest fault zone displacement case (c.90 m), all

the slip surfaces within the FSZ have associated fault rock (Fig. 8). Therefore, the average total thickness of observed fault rock and the average thickness of fault rock on every slip surface are the same (Fig. 15C). 5. Discussion 5.1. Fault zone architecture Fault zones in this study with displacements over 1 m typically have four architectural components: Protolith, Weakly Deformed Damage Zone (WDDZ), Fracture Splay Zone (FSZ) and Fault Core (Fig. 13). This fault zone architecture is almost solely confined to the micrite-dominated carbonates. Grain-dominated carbonates commonly have a symmetrical WDDZ surrounding a single principal

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

93

Fig. 13. Schematic diagram of faults in the study area that offset micrite-dominated carbonates: a fault-bounded, intensely deformed zone named the Fracture Splay Zone (FSZ), surrounded by a weakly deformed damage zone (WDDZ). Fault cores occur on one or more of the slip surfaces.

slip surface. The most comparable architectural models in literature are those proposed by Faulkner et al. (2003) and Childs et al. (1997). Faulkner et al. (2003) observed a zone of ‘fault core’ with fault rock, fractured rock and areas of country rock. Childs et al. (1997) documents a fault zone bound between two slip surfaces. However, these

models have a different fault propagation and evolution, therefore they cannot be applied to the faults within this study. Neither can these faults be described as dual slip surfaces with overlapping faults (Fig. 10). A new evolutionary model for the faults in this study is proposed.

Fig. 14. 3D block models of faults observed at Ras ir Raheb with different displacements: A: 0.52 m. B: 7 m. C: 11.7 m. D: 25 m. These block models show the along strike and down dip variability to the FSZ slip surfaces (B, C, D). Conglomerate layers C1 and C2 are marker horizons for displacement.

94

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 15. A: Fracture Splay Zone (FSZ) width versus displacement in the micrite-dominated and grain-dominated carbonates. B: Total damage zone width versus displacement in the micrite-dominated and grain-dominated carbonates. C: Fault core width versus displacement, average total fault rock abundance, and average total number of slip surfaces within the FSZ.

5.2. Scaling relationships of architectural components with fault throw 5.2.1. Damage zone scaling relationships The FSZ is created when the juxtaposition of micrite-against grain-dominated carbonates occurs, propagating more surfaces within the hanging wall, causing the width of the FSZ to increase with displacement (Fig. 15A). At lower displacements damage is accommodated mainly within the FSZ, creating a narrow WDDZ. With further displacement the WDDZ width increases to accommodate increased stress (Fig. 15B). Deformation is localised onto the principal slip surface in the grain-dominated carbonates, creating a narrow WDDZ, and a

narrow/absent FSZ. Due to lithofacies mixing at the mechanical contact (Fig. 2), at displacements 7 m the FSZ penetrates into the Il Mara Member, incorporating some grain-dominated carbonates in the FSZ (Fig. 15A). Scaling of damage zones has been suggested to be associated with lithological controls on deformation style (Shipton and Cowie, 2003; Shipton et al., 2006; Riley et al., 2010; Loveless et al., 2011), particularly the initial porosity (Groshong, 1988; Wong et al., 1997; Shipton and Cowie, 2003; Schultz and Siddharthan, 2005; Agosta et al., 2010; Faulkner et al., 2010; Agosta et al., 2012; Jeanne et al., 2012). This study shows that the strength of carbonate lithofacies has a large control on the scaling relationships; weaker micrite-dominated carbonates

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

95

Fig. 16. Proposed evolution of the fault zone architecture of the studied faults. Thick red lines represent bounding slip surfaces and thin red lines represent subsidiary faults. (I): Refraction of the principal slip surface is proposed across the layers due to a mechanical contrast. Fault dip is less in the micrite-dominated carbonates (e.g. Fig. 8) (II): This refraction in combination with the truncated mechanical layer boundary influences nucleation of future slip surfaces. (IIIeV): The slip surfaces propagate outwards then curve to follow the same trend as the primary slip surface. (VI): At greater displacements large bounding slip surfaces in the hanging wall propagate more slip surfaces. (VII): Slip surfaces converge with others along the fault strike. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

disperse the deformation, whilst the stronger grain-dominated carbonates localise the deformation. Riley et al. (2010) noted a similar relationship, within faulted ignimbrite rocks, indicating that rock mineralogy is less important than rock strength in determining fracture attributes. Hugman and Friedman (1979) and Prikryl (2001) noted that the rock’s texture, e.g. grain size

influences the strength of a rock. Although this is observed within this study, the coarser grained lithofacies are stronger and finer-grained lithofacies are weaker, opposing that documented by these authors. Other factors such as grain rigidity, sorting, cementation and pore types also combine to influence the strength of the rock.

96

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

Fig. 17. Predictive equation based on fault geometry to calculate the FSZ width as a function of displacement. This equation uses the throw, heave, displacement and the angles of dip in the micrite-dominated carbonates and grain-dominated carbonates, along with the height above the contact where the measurement is taken. Black points on the graph show measured data from 5 fault zones across Ras ir Raheb (0.52 m, Fig. 4; 7 m, Fig. 5C, D; 11.7 m, Fig. 6; 25 m, Fig. 7) and Madliena Tower (90 m, Fig. 8), which are comparable to the predicted/calculated FSZ width.

5.2.2. Fault core scaling relationships Fault rock thickness has been documented to increase with displacement, as further sliding accumulates crushed grains and sheared-off asperities (Engelder, 1974; Robertson, 1982; Hull, 1988). However, the geometry of a fault zone controls the variations in fault core scaling (Childs et al., 2009). Therefore, although fault core thickness is generally observed to increase with displacement in this study, this relationship is complicated in the micritedominated carbonates due to formation of the FSZ dispersing the deformation. Fault core production in grain-dominated carbonates is generally confined to a single principal slip surface, allowing fault core to accumulate progressively with displacement. It is generally thought that fault core production initiates after a certain threshold (e.g. between 1.5 m and 5 m in Micarelli et al., 2006), after which a continuous fault core is observed. Each slip surface within the FSZ may show an increase in fault core with displacement, and begin to form after such a displacement is reached. However, the threshold for a continuous fault core is substantially higher (c.60 m). 5.3. Evolution of the fault zone A new fault zone evolution model is proposed to discuss the formation of an FSZ. A mechanical stratigraphy is assumed, as the strength of the underlying LCL is c.80 MPa, while the GL has a compressive strength of c.40 MPa (Fig. 2). This mechanical stratigraphy causes the dip of slip surfaces to vary between the two lithofacies (Fig. 9). Further slip surfaces form from subsequent rupture episodes, associated with the fault kinking around layers of contrasting mechanical strengths. The juxtaposition of weaker micrite-dominated carbonates against stronger grain-dominated carbonates produces a concentration of stress and allows for another slip surface to be created. This slip surface propagates

outwards and upwards from this point of high strain (Fig. 16). Nucleation of slip surfaces often leads to folding of the beds close to where the site of propagation. The folding is due to lense rotation in this area, and because these slip surfaces initially dip the opposite way to the principal fault, therefore, essentially causing reverse faulting. The ‘new’ slip surface curves around to follow the same strike and dip trend as the principal slip surface (Fig. 16). 5.3.1. Model predictions for the width of the Fracture Splay Zone in mechanically layered systems It could be assumed that the width of the FSZ is associated with the thickness of the weaker micritic carbonates, because the progressive displacement of the weaker lithofacies next to the stronger lithofacies will continue creating more slip surfaces. Only when another strong lithofacies is displaced next to another strong lithofacies will the FSZ cease to form. Predictions of the width of the FSZ can be made by using trigonometric arguments, using the geometry of the FSZ (Fig. 17). The calculation also takes the height above the strongeweak transition (LCL-LGL) level into consideration, because the FSZ varies in width dependent on the location above this level. This approach successfully replicates the FSZ widths measured in the field for all those above 1 m (Fig. 17). 5.3.2. Predicted fault zone architectures with varying mechanical stratigraphy Predictions of where the FSZ could be formed (and the thickness of the FSZ), in varying mechanical stratigraphy scenarios are shown in Fig. 18. Predictions of the fault zone architectures can be made using assumptions regarding how slip surfaces respond in different mechanical layers, e.g. the dips of the slip surfaces. The thickness and geometry of the FSZ is dependent on the mechanical stratigraphy and the thickness of the mechanical layers. If a thick, stronger

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

97

Fig. 18. Predicted fault zone architectures for different geometries for the layers of the mechanical stratigraphy. The geometries use different numbers of stratigraphic layers and ratios of thickness of weak:strong layers. Each mechanical stratigraphy generates a different fault zone architecture.

98

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99

layer dominates the mechanical stratigraphy within a multilayered system, a discontinuous, thin FSZ will be produced (Fig. 18, ratio 1:3). In this case, the FSZ can look similar to a pullapart, and may not represent an FSZ anymore. However, this ‘pull apart’ architecture will be produced mainly within the weaker micrite-dominate carbonates rather than the stronger layers, as described by Peacock and Zhang (1994), and Peacock and Sanderson (1995). Conversely, if a thick, weaker layer dominates the mechanical stratigraphy, a wide FSZ will be produced, which may penetrate through thin stronger layers, preventing reattachment onto one slip surface, creating a very wide FSZ (Fig. 18, ratio 3:1). Modelling has shown that the contrast in dip of faults in different layers controls the architecture (Schöpfer et al., 2007a, 2007b, 2009). Within their model, a fault bound block is created from further rupture and the asperities are removed. This structure is similar to what is observed and predicted using information on the Maltese mechanical stratigraphy. However, in this study an increase in the proportion of weak layers creates a wider damage zone/FSZ, whereas Schöpfer et al. (2009) recorded the bound zone to be narrower with an increased proportion of weaker layers. This contrasting result can be related to different deformation styles in carbonate lithofacies. 5.4. Influence on petrophysical properties Fractures in the damage zone increase the porosity and permeability, and the production of a fault core decrease the porosity and permeability (Caine et al., 1996; Evans et al., 1997; Fisher and Knipe, 1998; Flodin et al., 2005; Wibberley et al., 2008; Agosta, 2008). Fault zones with typical fault core-damage zone architectures may, therefore, create a combined barrierconduit system (Chester et al., 1993; Antonellini and Aydin, 1994; Bruhn et al., 1994; Caine et al., 1996; Evans et al., 1997; Lockner et al., 2000; Billi et al., 2003; Agosta et al., 2007; Balsamo et al., 2010). However, different architectures can alter the petrophysical properties of the rock surrounding faults (e.g. Childs et al., 1997; Faulkner et al., 2010). The high intensity of fractures within the FSZ would potentially create a spatially distinct, permeable zone. The distributed nature of fault core controls the sealing potential. Detailed examination of deformation styles on the petrophysical properties will be investigated in subsequent contributions. 6. Summary  Fault zones on Malta show that within the weaker micritedominated carbonates a bound zone of intense fracturing is formed; an FSZ, with a WDDZ beyond this. Conversely, the majority of the grain-dominated carbonates conform to a more classical fault zone architecture with a single slip surface (core) surrounded by a damage zone.  The Fracture Splay Zone is created due to the mechanical contrast between the stronger grain-dominated carbonates and overlying weaker micrite-dominated carbonates.  The protolith textures control the strength of the rock, which influences the deformation style. Micrite-dominated carbonates have a low strength due to a fine particle sizes with a micritic matrix. Grain-dominated carbonates have a higher strength because they are composed of coarser grains that are more rigid. Micrite-dominated carbonates show a dispersed deformation with through-going fractures, whereas the grain-dominated carbonates localise the deformation.  Knowledge of the mechanical stratigraphy of a faulted carbonate system and how each lithofacies accommodates the stress can be used to predict the fault zone architecture.

 Understanding how different carbonates deform and the architecture that may be produced is key to predicting the hydraulic behaviour of faults. Acknowledgements We thank Claude Gout, Frances Abbots, Peter Ellis, Neil Hurst, Andrea Billi and an anonymous reviewer for their constructive comments and suggestions to improve the standard of this manuscript. We also thank Total E&P and BG Group for project funding and support, and the Industry Technology Facilitator for facilitating the collaborative development (grant number 3322PSD). Partial financial support was also provided by the Aberdeen Formation Evaluation Society. References Agosta, F., Kirschner, D.L., 2003. Fluid conduits in carbonate-hosted seismogenic normal faults of central Italy. J. Geophys. Res. 108 (B4), 2221. Agosta, F., Aydin, A., 2006. Architecture and deformation mechanism of a basinbounding normal fault in Mesozoic platform carbonates, central Italy. J. Struct. Geol. 28 (8), 1445e1467. Agosta, F., Prasad, M., Aydin, A., 2007. Physical properties of carbonate fault rocks, fucino basin (Central Italy): implications for fault seal in platform carbonates. Geofluids 7 (1), 19e32. Agosta, F., 2008. Fluid flow properties of basin-bounding normal faults in platform carbonates, Fucino Basin, Central Italy. Geol. Soc. London Spec. Publ. 299, 277e 291. Agosta, F., Alessandroni, M., Antonellini, M., Tondi, E., Giorgioni, M., 2010. From fractures to flow: a field-based quantitative analysis of an outcropping carbonate reservoir. Tectonophysics 490 (3e4), 197e213. Agosta, F., Ruano, P., Rustichelli, A., Tondi, E., Galindo-Zaldívar, J., Sanz de Galdeano, C., 2012. Inner structure and deformation mechanisms of normal faults in conglomerates and carbonate grainstones (Granada Basin, Betic Cordillera, Spain): inferences on fault permeability. J. Struct. Geol. 45, 4e20. Antonellini, M., Aydin, A., 1994. Effect of faulting on fluid flow in porous sandstones: petrophysical properties. AAPG Bull. 78 (3), 355e377. Aydin, A., 2000. Fractures, faults, and hydrocarbon entrapment, migration and flow. Mar. Petrol. Geol. 17 (7), 797e814. Balsamo, F., Storti, F., Salvini, F., Silva, A.T., Lima, C.C., 2010. Structural and petrophysical evolution of extensional fault zones in low-porosity, poorly lithified sandstones of the Barreiras Formation, NE Brazil. J. Struct. Geol. 32 (11), 1806e 1826. Berg, S.S., Skar, T., 2005. Controls on damage zone asymmetry of a normal fault zone: outcrop analyses of a segment of the Moab fault, SE Utah. J. Struct. Geol. 27 (10), 1803e1822. Billi, A., Salvini, F., Storti, F., 2003. The damage zone-fault core transition in carbonate rocks: implications for fault growth, structure and permeability. J. Struct. Geol. 25 (11), 1779e1794. Billi, A., Di Toro, G., 2008. Fault-related carbonate rocks and earthquake indicators: recent advances and future trends. In: Landowe, S.J., Hammler, G.M. (Eds.), Structural Geology: New Research. Nova Science Publishers, New York, pp. 63e 86. Bonson, C.G., Childs, C., Walsh, J.J., Schopfer, M.P.J., Carboni, V., 2007. Geometric and kinematic controls on the internal structure of a large normal fault in massive limestones: the Maghlaq Fault, Malta. J. Struct. Geol. 29 (2), 336e354. Brown, E.T., 1981. Rock Characterisation Testing and Monitoring, ISRM Suggested Methods. Pergamon Press, Oxford. Bruhn, R.L., Parry, W.T., Yonkee, W.A., Thompson, T., 1994. Fracturing and hydrothermal alteration in normal fault zones. Pure Appl. Geophys. 142 (3e4), 609e 644. Caine, J.S., Evans, J.P., Forster, C.B., 1996. Fault zone architecture and permeability structure. Geology 24 (11), 1025e1028. Chester, F.M., Logan, J.M., 1986. Implications for mechanical properties of brittle faults from observations of the Punchbowl Fault Zone, California. Pure Appl. Geophys. 124 (1e2), 79e106. Chester, F., Evans, J., Biegel, R., 1993. Internal structure and weakening mechanisms of the San Andreas Fault. J. Geophys. Res. 98 (B1), 771e786. Chester, F.M., Chester, J.S., Kirschner, D.L., Schulz, S.E., Evans, J.P., 2004. In: Karner, G.D., Taylor, B., Driscoll, N.W., Kohlstedt, D.L. (Eds.), Structure of LargeDisplacement, Strike-slip Fault Zones in the Brittle Continental Crust, Rheology and Deformation in the Lithosphere at Continental Margins, vol. 1. Columbia University Press, New York, pp. 223e260. Childs, C., Watterson, J., Walsh, J.J., 1995. Fault overlap zones within developing normal fault systems. J. Geol. Soc. 152, 535e549. Childs, C., Watterson, J., Walsh, J.J., 1996. A model for the structure and development of fault zones. J. Geol. Soc. 153 (3), 337e340. Childs, C., Walsh, J.J., Watterson, J., 1997. Complexity in fault zone structure and implications for fault seal prediction. In: Møller-Pedersen, P., Koestler, A.G.

E.A.H. Michie et al. / Journal of Structural Geology 65 (2014) 82e99 (Eds.), Hydrocarbon Seals: Importance for Exploration and Production. Spec. Public. Nor. Petrol. Soc. 7, 61e72. Childs, C., Manzocchi, T., Walsh, J.J., Bonson, C.G., Nicol, A., Schöpfer, M.P.J., 2009. A geometric model of fault zone and fault rock thickness variations. J. Struct. Geol. 31 (2), 117e127. Dart, C.J., Bosence, D.W.J., McClay, K.R., 1993. Stratigraphy and structure of the Maltese graben system. J. Geol. Soc. 150 (6), 1153e1166. Engelder, J.T., 1974. Cataclasis and the generation of fault gouge. Geol. Soc. Am. Bull. 85 (10), 1515e1522. Evans, J.P., Forster, C.B., Goddard, J.V., 1997. Permeability of fault-related rocks, and implications for hydraulic structure of fault zones. J. Struct. Geol. 19 (11), 1393e 1404. Faulkner, D.R., Lewis, A.C., Rutter, E.H., 2003. On the internal structure and mechanics of large strike-slip fault zones: field observations of the Carboneras fault in southeastern Spain. Tectonophysics 367 (3e4), 235e251. Faulkner, D.R., Jackson, C.A.L., Lunn, R.J., Schlische, R.W., Shipton, Z.K., Wibberley, C.A.J., Withjack, M.O., 2010. A review of recent developments concerning the structure, mechanics and fluid flow properties of fault zones. J. Struct. Geol. 32 (11), 1557e1575. Ferrill, D.A., Morris, A.P., McGinnis, R.N., Smart, K.J., Ward, W.C., 2011. Fault zone deformation and displacement partitioning in mechanically layered carbonates: the Hidden Valley fault, central Texas. AAPG Bull. 95 (8), 1383e1397. Fisher, Q.J., Knipe, R.J., 1998. Fault sealing processes in siliciclastic sediments. In: Jones, G., Fisher, Q.J., Knipe, R.J. (Eds.), Faulting, Fault Sealing and Fluid Flow in Hydrocarbon Reservoirs, Geological Society, London, Special Publications, vol. 147, pp. 117e134. Flodin, E., Gerdes, M., Aydin, A., Wiggins, W., 2005. Petrophysical properties and sealing capacity of fault rock, Aztec Sandstone, Nevada. In: Sorkhabi, R., Tsuji, Y. (Eds.), Faults, Fluid Flow, and Petroleum Traps: AAPG Memoir, vol. 85, pp. 197e 218. Fondriest, M., Smith, S.A., Candela, T., Nielsen, S.B., Mair, K., Di Toro, G., 2013. Mirrorlike faults and power dissipation during earthquakes. Geology 41 (11), 1175e 1178. Gaviglio, P., Bekri, S., Vandycke, S., Adler, P.M., Schroeder, C., Bergerat, F., Darquennes, A., Coulon, M., 2009. Faulting and deformation in chalk. J. Struct. Geol. 31 (2), 194e207. Gratier, J.-P., Thouvenot, F., Jenatton, L., Tourette, A., Doan, M.-L., Renard, F., 2013. Geological control of the partitioning between seismic and aseismic sliding behaviour in active faults: evidence from the Western Alps, France. Tectonophysics 600, 226e242. Groshong, R., 1988. Low-temperature deformation mechanisms and their interpretation. Geol. Soc. Am. Bull. 100, 1329e1360. Hausegger, S., Kurz, W., Rabitsch, R., Kiechl, E., Brosch, F.J., 2010. Analysis of the internal structure of a carbonate damage zone: implications for the mechanisms of fault breccia formation and fluid flow. J. Struct. Geol. 32 (9), 1349e1362. Hendron, A.J., 1968. Mechanical properties of rock. In: Stagg, K.G., Zienkiewicz, O.C. (Eds.), Rock Mechanics in Engineering Practice. Wiley, New York. House, M.R., Dunham, K.C., Wigglesworth, J.C., 1961. Geology of the Maltese Islands. In: Bowen-Jones, H., Dewdney, J.C., Fisher, W.B. (Eds.), Malta, Background for Development. Univ. Durham, pp. 24e33. Huggins, P., Watterson, J., Walsh, J.J., Childs, C., 1995. Relay zone geometry and displacement transfer between normal faults recorded in coal-mine plans. J. Struct. Geol. 17 (12), 1741e1755. Hugman III, R., Friedman, M., 1979. Effects of texture and composition on mechanical behavior of experimentally deformed carbonate rocks. AAPG Bull. 63 (9), 1478e1489. Hull, J., 1988. Thickness-displacement relationships for deformation zones. J. Struct. Geol. 10 (4), 431e435. Jeanne, P., Guglielmi, Y., Lamarche, J., Cappa, F., Marié, L., 2012. Architectural characteristics and petrophysical properties evolution of a strike-slip fault zone in a fractured porous carbonate reservoir. J. Struct. Geol. 44, 93e109. Knipe, R., 1993. The influence of fault zone processes and diagenesis on fluid flow. In: Horbury, A.D., Robinson, A.G. (Eds.), Diagenesis and Basin Development, AAPG Studies in Geology, vol. 36, pp. 135e154. Larsen, P.H., 1988. Relay structures in a Lower Permian basement-involved extension system, East Greenland. J. Struct. Geol. 10 (1), 3e8. Lockner, D., Naka, H., Tanaka, H., Ito, H., Ikeda, R., 2000. Permeability and strength of core samples from the Nojima fault of the 1995 Kobe earthquake. In: Ito, H., Fujimoto, K., Tanaka, H., Lockner, D. (Eds.), Proceedings of the International Workshop on the Nojima Fault Core and Borehole Data Analysis, pp. 147e152. Tsukuba, Japan, Nov 22e23 (1999). U. S. Geol. Surv. Loveless, S., Bense, V., Turner, J., 2011. Fault architecture and deformation processes within poorly lithified rift sediments, Central Greece. J. Struct. Geol. 33 (11), 1554e1568. Mauldon, M., Dershowitz, W., 2000. A multi-dimensional system of fracture abundance measures. Geol. Soc. Am. Abstr. Prog. 32 (7). Mauldon, M., Dunne, W.M., Rohrbaugh Jr., M.B., 2001. Circular scanlines and circular windows: new tools for characterizing the geometry of fracture traces. J. Struct. Geol. 23 (2e3), 247e258. Micarelli, L., Benedicto, A., Wibberley, C.A.J., 2006. Structural evolution and permeability of normal fault zones in highly porous carbonate rocks. J. Struct. Geol. 28 (7), 1214e1227.

99

Mitchell, T.M., Faulkner, D.R., 2009. The nature and origin of off-fault damage surrounding strike-slip fault zones with a wide range of displacements: a field study from the Atacama fault system, northern Chile. J. Struct. Geol. 31 (8), 802e816. Molli, G., Cortecci, G., Vaselli, L., Ottria, G., Cortopassi, A., Dinelli, E., Mussi, M., Barbieri, M., 2010. Fault zone structure and fluiderock interaction of a high angle normal fault in Carrara marble (NW Tuscany, Italy). J. Struct. Geol. 32 (9), 1334e1348. Morris, A.P., Ferrill, D.A., McGinnis, R.N., 2009. Mechanical stratigraphy and faulting in Cretaceous carbonates. AAPG Bull. 93 (11), 1459e1470. Peacock, D.C.P., Sanderson, D.J., 1991. Displacements, segment linkage and relay ramps in normal fault zones. J. Struct. Geol. 13 (6), 721e733. Peacock, D.C.P., Zhang, X., 1994. Field examples and numerical modelling of oversteps and bends along normal faults in cross-section. Tectonophysics 234 (1e2), 147e167. Peacock, D.C.P., Sanderson, D.J., 1995. Pull-aparts, shear fractures and pressure solution. Tectonophysics 241 (1e2), 1e13. Pedley, H.M., House, M.R., Waugh, B., 1976. The geology of Malta and Gozo. Proc. Geol. Assoc. 87 (3), 325e341. Pratt, S.K., 1990. Hardground Genesis in Pelagic Carbonates from the Miocene of Malta and Cretaceous of Southern England. Ph.D. Thesis. University of London. Prikryl, R., 2001. Some microstructural aspects of strength variation in rocks. Int. J. Rock Mech. Min. Sci. 38 (5), 671e682. Putz-Perrier, M.W., Sanderson, D.J., 2010. Distribution of faults and extensional strain in fractured carbonates of the North Malta Graben. AAPG Bull. 94 (4), 435e456. Riley, P.R., Goodwin, L.B., Lewis, C.J., 2010. Controls on fault damage zone width, structure, and symmetry in the Bandelier Tuff, New Mexico. J. Struct. Geol. 32 (6), 766e780. Robertson, E.C., 1982. Continuous formation of gouge and breccia during fault displacement. In: Goodman, R.E., Hulse, F. (Eds.), Issues in Rock Mechanics, Proceeding Symposium 23rd of Rock Mechanics. American Institute of Mining and Engineering, New York, pp. 397e404. Rohrbaugh, M.B., Dunne, W.M., Mauldon, M., 2002. Estimating fracture trace intensity, density, and mean length using circular scan lines and windows. AAPG Bull. 86 (12), 2089e2104. Savage, H.M., Brodsky, E.E., 2011. Collateral damage: evolution with displacement of fracture distribution and secondary fault strands in fault damage zones. J. Geophys. Res. Solid Earth (1978e2012) 116 (B3). Schöpfer, M.P.J., Childs, C., Walsh, J.J., 2007a. Two-dimensional distinct element modeling of the structure and growth of normal faults in multilayer sequences: 1. Model calibration, boundary conditions, and selected results. J. Geophys. Res. Solid Earth (1978e2012) 112 (B10). Schöpfer, M.P.J., Childs, C., Walsh, J.J., Manzocchi, T., Koyi, H.A., 2007b. Geometrical analysis of the refraction and segmentation of normal faults in periodically layered sequences. J. Struct. Geol. 29 (2), 318e335. Schöpfer, M.P.J., Childs, C., Walsh, J.J., 2009. Two-dimensional Distinct Element Method (DEM) modeling of tectonic fault growth in mechanically layered sequences. In: Mechanics of Natural Solids. Springer Berlin, Heidelberg, pp. 127e 146. Schultz, R.A., Siddharthan, R., 2005. A general framework for the occurrence and faulting of deformation bands in porous granular rocks. Tectonophysics 411 (1), 1e18. Shipton, Z.K., Cowie, P.A., 2003. A conceptual model for the origin of fault damage zone structures in high-porosity sandstone. J. Struct. Geol. 25 (3), 333e344. Shipton, Z.K., Soden, A.M., Kirkpatrick, J.D., Bright, A.M., Lunn, R.J., 2006. How thick is a fault?. In: Fault Displacement-thickness Scaling Revisited Geophysical Monograph Series, vol. 170, pp. 193e198. Sibson, R.H., 1977. Fault rocks and fault mechanisms. J. Geol. Soc. 133, 191e213. Sibson, R.H., 1996. Structural permeability of fluid-driven fault-fracture meshes. J. Struct. Geol. 18 (8), 1031e1042. Storti, F., Billi, A., Salvini, F., 2003. Particle size distributions in natural carbonate fault rocks: insights for non-self-similar cataclasis. Earth Planet. Sci. Lett. 206 (1e2), 173e186. Tondi, E., 2007. Nucleation, development and petrophysical properties of faults in carbonate grainstones: evidence from the San Vito Lo Capo peninsula (Sicily, Italy). J. Struct. Geol. 29 (4), 614e628. van Gent, H.W., Holland, M., Urai, J.L., Loosveld, R., 2010. Evolution of fault zones in carbonates with mechanical stratigraphyeInsights from scale models using layered cohesive powder. J. Struct. Geol. 32 (9), 1375e1391. Vermilye, J., Scholz, C., 1998. The process zone: a microstructural view of fault growth. J. Geophys. Res. 103 (B6), 12223e12237. Wibberley, C.A.J., Yielding, G., Di Toro, G., 2008. Recent advances in the understanding of fault zone internal structure: a review. In: Wibberley, C.A.J., Kurz, W., Imber, J., Holdsworth, R.E., Collettini, C. (Eds.), The Internal Structure of Fault Zones: Implications for Mechanical and Fluid-flow Properties, Geological Society, London, Special Publications, vol. 299 (1), pp. 5e33. Wong, T., David, C., Zhu, W., 1997. The transition from brittle faulting to cataclastic flow in porous sandstones: mechanical deformation. J. Geophys. Res. 102 (B2), 3009e3025.