Description:

_Journal of Structural Geology 32 (2010) 377–391_ Contents lists available at ScienceDirect Journal of Structural Geology journal homepage: www.elsevier.com/locate/jsg Linear to non-linear relationship between vein spacing and layer thickness in centimetre-to-decimetre-scale siliciclastic multilayers from the High-Ardenne slate belt (Belgium, Germany) Koen Van Noten*, Manuel Sintubin Geodynamics and Geo?uids Research Group, Department of Earth and Environmental Sciences, Katholieke Universiteit Leuven, Celestijnenlaan 200E, B-3001 Leuven, Belgium Article info Article history: Received 7 May 2009; Received in revised form 22 January 2010; Accepted 26 January 2010; Available online 4 February 2010 Keywords: Quartz vein Fracturing Spacing Lower Devonian Early Variscan High-Ardenne slate belt Abstract Typical spacing distributions have frequently been described for fractures in thin sedimentary layers (<1 m). Regularly spaced fractures often result from saturation during fracture development. Spatial distribution of veins is less commonly studied although it also can show regular patterns. This study focuses on the spatial distribution of quartz veins in Lower Devonian siliciclastic multilayer sequences from the High-Ardenne slate belt (Belgium, Germany) and compares the observed vein spacing with published fracture spacing in order to investigate (i) the effect of layer thickness on vein spacing and (ii) the processes of early vein development during late stages of burial in a sedimentary basin at the onset of orogeny. The results show that a quasi-linear relationship between vein spacing and layer thickness exists in thin (<40 cm) competent sandstone layers embedded in a pelitic matrix, but becomes non-linear in thicker sandstone layers (>40 cm). Vein spacing tends to increase to a maximum value becoming more or less independent of layer thickness. The resemblance with fracture spacing suggests that in an unfractured rock vein saturation can occur. High ?uid pressures are responsible for vein nucleation but the stress state around the initial veins controls the spacing pattern. Subsequently, in a vein-saturated rock, existing veins will thicken by crack-sealing or new crosscutting veins will develop if the regional stress field changes relatively with respect to the existing veins. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction Knowledge of the geometry of fracture and vein networks is essential in subsurface research (e.g., oil reservoir research, mining geology). Discontinuities such as fractures, joints, and veins are potential sites for fluid transport and have important implications for the hydraulic properties of rock (Clark et al., 1995; André? -Mayer and Sausse, 2007). Fracture spatial distribution has been studied in numerous papers commonly using statistical methods to find the best fit between actual or numerical datasets and theoretical distribution laws. This theoretical approach not only provides a simplified image of the fracture set but also enables a numerical definition (Huang and Angelier, 1989). Most fracture distributions, however, are not applicable to the spatial distribution of veins (Gillespie et al., 2001; Peacock, 2004). Although veins and joints initially are similar * Corresponding author. Tel.: +32 (0)16 32 64 48; fax: +32 (0)16 32 29 80. E-mail addresses: koen.vannoten@ees.kuleuven.be (K. Van Noten), manuel.sintubin@ees.kuleuven.be (M. Sintubin). 0191-8141 $ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsg.2010.01.011 mechanically (e.g., Thomas and Pollard, 1993), there are important geometric, scaling, and genetic differences and they should be considered separately (Gillespie et al., 2001). In this paper we refer to ‘joints’ or ‘fractures’ as unmineralized extension or opening (mode I) cracks across which host-rock minerals have been broken without any measurable displacement. Genetically, the term ‘‘joint’’ does not necessarily refer to fracture networks formed during uplift. We refer to ‘veins’ as mineral-filled fractures which usually form in vein arrays (cf. Twiss and Moores, 1992; Bai and Pollard, 2000b; Peacock, 2004). In this field-based study we want to investigate if repetitive spacing of late burial quartz veins in the High-Ardenne slate belt (Belgium, Germany) is proportional to the layer thickness and if the observed spacing can be explained by one of the numerous published geomechanical models which are developed to predict fracture patterns in naturally fractured reservoirs. In order to do so, we first discuss some representative geomechanical models of fracture and vein spacing and show why there is still a need for new field data. Secondly, we compare the spacing of planar veins and the aspect ratios of fractured blocks in thin (<1 m) siliciclastic multilayer sequences with (i) field examples of vein spacing in boudinaged layers; (ii) published fracture distributions for a range of lithologies; (iii) the spacing of lensoid veins in shortened sequences (i.e., mullions) in the central part of the slate belt to show the regional occurrence of vein spacing. It is interesting to examine the distribution of these stratabound veins to gain insight into a time-integrated history of compartmentalized fluid redistribution during low-grade metamorphism and to understand the initial development of veins (cf. Manning, 1994; Stowell et al., 1999). Working on veins rather than on fractures has moreover the advantage that the role of fluid pressure can be evaluated. This study could therefore serve as an analog for fracture or vein patterns in the subsurface. 1.1. Fracture development It is widely accepted that extensional fracture spacing is mainly controlled by lithological parameters of the host-rock and layer thickness. There is a large literature on the relationship between fracture spacing (S) and layer thickness (T) in which the mechanics of fracture spacing has been described by both field observations (Hobbs, 1967; Ladeira and Price, 1981; Pollard and Aydin, 1988; Huang and Angelier, 1989; Narr and Suppe, 1991; Engelder et al., 1997; Gillespie et al., 2001) and geomechanical or numerical modelling (Mandal et al., 2000; Fischer et al., 1995; Wu and Pollard, 1995; Ji and Saruwatari, 1998; Ji et al., 1998; Bai and Pollard, 2000a,b; Olson, 2004; Li and Yang, 2007). Both linear and nonlinear relationships between fracture spacing and layer thickness have been described. Throughout these studies, spacing is often explained from a mechanical point of view without considering the role of pore-fluid pressure, although it is commonly accepted that high fluid pressures in combination with tectonic extension can induce opening fractures (Secor, 1965; Ladeira and Price, 1981; Simpson, 2000). Recent experimental results show that fracture spacing decreases progressively with increasing strain perpendicular to the fracture such that new fractures form between earlier fractures, with increasing bedding-plane friction and with increasing overburden pressure (cf. Arslan et al., 2010). The fractures are flanked by ‘stress shadows’ (Lachenbruch, 1961; Hobbs, 1967) in which new fractures are unlikely to form and the size of the stress shadow is directly proportional to the length of the fracture, i.e., the layer thickness (Pollard and Segall, 1987). In layered sequences, these fractures are usually prevented from further growth at the interface between competent and incompetent beds (Wu and Pollard, 1995; Engelder et al., 1997; Cooke et al., 2006). The final stage of fracture development is reached when no more ‘freedom’ exists for nucleating new fractures in between preexisting ones and this stage is referred to as the ‘saturation level’ by Narr and Suppe (1991). Recent modelling showed that the crack-normal stress between closely spaced fractures actually becomes compressive at certain fracture spacing so no more tensile fractures can form (Bai and Pollard, 2000b). This result dictates that at the saturation level, the opening of pre-existing fractures is preferred over nucleation of new additional fractures with increasing extensional strain (Olson, 2004; André? -Mayer and Sausse, 2007). Bai and Pollard (2000a) defined a critical S/T ratio of 0.976 which gives a lower limit for fractures driven by extension in a material without significant flaws and defines the conditions of fracture saturation. This critical value increases nonlinearly with increasing ratio of the Young’s modulus of the fractured layer with respect to that of the neighboring layers, with increasing Poisson’s ratio of the fractured layer, and with increasing magnitude of the overburden stress (Bai and Pollard, 2000b). However, the effects of the elastic constants on the critical value are minor. Spacing-to-thickness ratios less than the critical value have often been seen in outcrop. The concept of fracture saturation may break down if a local stress field exists due to the presence of flaws between adjacent fractures or due to a local mechanism such as high fluid pressures, which can overcome the compressive stress at the stage of fracture saturation in such way that new fractures are nucleated in the vicinity of pre-existing fractures (Simpson, 2000). This contrasts with a large body of work in which the nucleation of nearby fractures is prohibited at the stage of fracture saturation and the role of high fluid pressures has therefore been underexplored. These closely spaced infilling fractures are moreover more likely to initiate near the interfaces than in the middle of the layer. Besides lithological parameters, fracture spacing is also dependent on the degree of pre-existing rock deformation (Harris et al., 1960) and the interference of adjacent competent layers (Ladeira and Price, 1981; Bai and Pollard, 1999). The effect of layer thickness variability' _ Ключевые слова: quartz, high-ardenne, fracture, joint spacing, foxford, rursee wildenhof, stowell, cv, saturated fracture, wildenhof, formation, fisher, sandstone, gillespie, spacing distribution, jmicrovision, phasb, veining, noten sintubin, ardenne, hubertus, unfractured rock, special, fracture saturation, mechanics, variation, early, variscan, eds, geometry, quartz vein, noten, walsh, jointing, tectonophysics, sedimentary rock, thicker unit, earth, vein, cv cv, andre-mayer, grain, experimental study, eld, high, folding convention, sintubin journal, pressure, data, initiation propagation, lithology, scale, belt, strain, shortening, brabant parautochthon, middle crust, vein formation, boeur, hs sr, fracture spacing, spatial distribution, boudinage, olson, uid pressure, layer thickness, high-ardenne slate, simpson, ?eld, van, outcrop, bertrix, table, ardenne allochthon, belgium, study area, geology, van noten, le, peacock, ?uid, ardennes bastogne, mandal, von winterfeld, thicker bed, structural geology, order, ha, adjacent vein, bai pollard, roduit, gross, sequence, rijken, generation, metamorphic grade, geophysical, deformation, pore-uid pressure, journal structural, eifel, layer, competent, joint, result, adjacent, press, sintubin, structure, spaced, urftsee, vein wall, host rock, standard deviation, cleavage development, engelder, aspect, slate, ardennes, pollard, ratio, atkinson, crack, chasb, original, wa, earliest manifestation, cm, bai, competent bed, france, stage, fractured layer, pattern, germany, und, analysis, crack-seal microstructures, journal structural geology, der, increase non-linearly, competentincompetent interface, vein spacing, slate belt, table summary, rock, in?uence, host-rock lithology, aspect ratio, intermullion vein, compressive stress, bastogne, unit, discussion, cross-cutting generation, mullion shortening, geological, spacing, structural, initial vein, quartz grain, saturation, vein array, undeformed segment, layer-parallel shortening, boudin, crystal, mullion, sedimentary, spacing pattern, stress distribution, price, case study, development, competent layer, regional phenomenon, kenis, journal, relationship, planar vein, bed, rursee, elsevier, arslan, subcritical crack, variscan orogeny, initial, hydraulic fracturing, rectangular boudins, burial, vein nucleation, host, outcrop studied, thickness, host-rock, study, wildenhof eifel, stress, wider spacing, journal geophysical, vein gillespie, distribution, initial development, initial composition, eifel rursee, area