Author:

Keywords:

tectonic stress inversion, fluid redistribution, late burial, early tectonic, overpressured basin, fractured reservoir, Variscan orogeny, Ardenne-Eifel basin, quartz veins

Abstract:

In the recent decade, there is an increased interest in the processes active at the base of the seismogenic crust, with a particular focus on the evolution and characterisation of elevated fluid pressures that could possibly drive fault rupturing and initiate major earthquakes. The base of the seismogenic zone, however, is not directly accessible. One of the possibilities to overcome this research problem is the study of natural fossil analogues to pursue a better understanding of the formation and evolution of these overpressured fluids at relatively great depths. Within the broad framework of natural analogues, vein studies provide an important insight in the characterisation of fluids and the role of fluid pressure during a complete deformation cycle. In vein analysis, an extensive interdisciplinary approach is required in order to determine a fluid-assisted deformation during orogeny. This research project focuses on overpressured fluids that are generated in a sedimentary basin during late burial that is affected by an incipient tectonic shortening at the onset of orogeny. Quartz veins that result from the overpressures in sedimentary basins preserve valuable information on the metamorphic fluids which were present during burial and subsequent orogenic deformation. A detailed geometric and kinematic structural analysis of several successive quartz vein types has been performed to allow constraining a geometric evolutionary model of the emplacement of these different, successive veins with respect to structural features such as bedding, cleavage, folds and faults. This model is refined by means of a pervasive microstructural and microthermometric analysis of the vein infill in order to implement physico-chemical parameters such as formation temperature and pore-fluid pressure. Ultimately, this study aims at understanding why, how and when fluids become overpressured and how they subsequently evolve at the onset of orogeny. More specifically, the study focuses on Lower Devonian multilayer siliciclastic metasediments that were deposited in the Ardenne-Eifel basin. This basin formed part of the northern passive margin of the short-lived Rhenohercynian Ocean which was closed during the Variscan orogeny. These metasediments are currently exposed in the High-Ardenne slate belt (Belgium, France, Luxemburg & Germany), which belongs to the central part of the Rhenohercynian foreland fold-and-thrust belt in the northern extremity of the Central European Variscan belt. The rocks, in which quartz veins frequently occur, are studied in the north-eastern part of the High-Ardenne slate belt (Rursee, Urftsee; North Eifel; Germany) and are affected by a very low-grade metamorphism, which is considered to have a burial origin, pre- to early synkinematic with the prograding Variscan deformation. This metamorphism documents the peak of subsidence and sediment accumulation (at about 7 km) prior to the main Variscan contraction of the basin. The analysis focuses on two successive quartz vein types, oriented normal and parallel to bedding respectively, in which the very low-grade metamorphism is reflected. A first vein type consists of several generations of bedding-normal veins that remain perpendicular to bedding around the characteristic, NW-verging, upright to overturned folds of the North Eifel, thereby clearly predating the main Variscan contraction and fold-and-cleavage development of the slate belt. The planar to lensoid veins are mostly restricted to competent sandstone layers, although they sometimes continue into the adjacent incompetent layers, refracting at the competent-incompetent interface, similar to cleavage. The fabric of the bedding-normal veins predominantly shows a fibrous to elongate-blocky vein infill in which the fibres or the elongated crystals are oriented at high angle to the vein walls. Both syntaxial and ataxial growth morphologies are recognised. Repetition of host-rock inclusion bands internally in the vein quartz reflects episodic opening of the vein by the crack-seal mechanism. Pseudosecondary fluid-inclusion planes, oriented at high angle to the crystal fibre walls and reflecting intracrystal healed microcracks, confirm that crystal growth occurred by incremental crack-seal steps. The specific orientation of fluid-inclusion planes and host-rock inclusion bands indicate that the veins are extension veins which originally grew in Mode I fractures in a rock that was already significantly reduced in porosity during burial. The fluid-inclusion planes have been particularly useful as a microstructural marker to reconstruct the stress state in the basin at the time of vein formation. In this respect, the maximum principal stress (sigma 1) was still vertical, corresponding with load of the overburden (sigma V) and the veins opened perpendicular to the least principal stress (sigma 3), that roughly indicate the extension direction in the basin. The propagation of fractures during vein formation occured in the sigma1 - sigma2 plane. Secondary inclusion planes, corresponding to post-veining transcrystal microcracks oriented parallel to the vein walls, evidence that microcracks still developed after opening of the extension veins but still in a similar stress regime as the pseudosecondary inclusion planes. The consistent pre-folding orientation of bedding-normal veins in the North Eifel corresponds to the pre-folding orientation of intermullion quartz veins in the central part of the High-Ardenne slate belt (Ardennes, Belgium) highlighting that this veining event occurred regionally. The (micro)structural analysis of bedding-normal veins eventually shows that these veins developed at low differential stress during the lateststage of the extensional stress regime and reflect a regional NW-SE opening of the Ardenne-Eifel basin. A structural change of the pre-folding vein orientation from NW-SE (Ardennes) to NNW-SSE (Eifel) is probably related to a post-veining oroclinal bending of the slate belt during the main Variscan contraction due to the presence of the rigid Brabant basement in the north. Apart from the orientation analysis, the spatial distribution of these bedding-normal quartz veins in the High-Ardenne slate belt is investigated in order to determine the effect of the layer thickness to vein spacing. The results show a quasi linear relationship between vein spacing and layer thickness in thin (<40 cm) competent sandstone layers and a non-linear relationship 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 of vein spacing with regularly spaced fractures that result from saturation during fracture development, suggests that in an unfractured rock, the host rock can get saturated by the presence of initial veins in which the veins subsequently either grow by progressive extension or that new cross-cutting veins develop in case the regional stress field changes relatively with respect to the existing veins. In the North Eifel, bedding-parallel quartz veins cross-cut, truncate and offset the bedding-normal veins and are continuously present around fold hinges. Bedding-parallel veins occur interbedded between two contrasting lithologies, as well as intrabedded in the competent and incompetent sequences. Macroscopically, the veins show a composite internal fabric consisting of several distinct generations of quartz laminae, often marked by slickenlines, intercalated with thin pelitic wall-rock inclusion seams. Microscopically, the variety in microstructures is indicative of different combined mechanisms of vein growth and mineral infill, reflecting a complex vein formation. Repeated crack-seal inclusion bands that are oriented parallel to the vein wall, and thus parallel to bedding, reflect bedding-normal opening of crack-seal quartz laminae. This mechanism also is also evidenced by the alignment of pseudosecondary fluid-inclusion planes internally in the fibres. The presence of blocky laminae reflects crystal growth in an open cavity during bedding-normal uplift. Bedding-parallel stylolites in often occur in these blocky laminae and represent pressure-dissolution during bedding-normal collapse. The occurrence of shear laminae, in which small quartz crystals are dynamically recrystallised during the main phase of the Variscan contraction, has, on the other hand, been used as indicator of bedding-parallel shear events. The pronounced bedding-parallel fabric is interpreted to form during several alternating phases of bedding-normal uplift and bedding-parallel shearing taking place at the onset of folding. Based on the microstructural analysis, the bedding-parallel veins are classified as extension veins to extensional-shear veins that formed at low differential stress. This vein type is the brittle expression of the first phases of the compressional stress regime affecting the siliciclastic metasediments of the Ardenne-Eifel basin at the onset of Variscan shortening. In this configuration, the bedding-parallel veins demonstrate a particular stress-state in the basin at the time of vein formation in which the maximum principal stress (sigma1) was oriented parallel to a NW-SE-oriented tectonic compression (sigmaT) and the minimum principal stress (sigma3) corresponded to the vertical load of the overburden. The succession of bedding-normal to bedding-parallel extension quartz veins thus materialises the transition from extension to compression at the onset of Variscan orogeny and is interpreted to have formed during early orogenic compressional tectonic inversion. A petrographic and micro-thermometric analysis of fluid inclusions within the vein quartz has further helped to constrain the kinematic and pressure-temperature trapping conditions of both vein types, allowing the reconstruction of the fluid pressure and stress-state evolution during tectonic inversion. The method used for defining the trapping conditions involves cross-cutting isochores, calculated from the salinity and homogenisation temperatures representative of both vein types, with an independent vitrinite reflectance geothermometer that defines the maximum temperature of 250degrees°C at the time of vein formation. The results demonstrate that the bedding-normal extension veins are trapped from a low salinity (3.5-8 eq.wt.% NaCl) H2O-NaCl fluid at near-lithostatic fluid pressures prior to inversion. After the tectonic inversion, bedding-parallel veins formed from a low salinity (4-7 eq.wt.% NaCl) H2O-NaCl fluid at lithostatic to supra-lithostatic fluid pressures during the early stages of the compressional stress regime. Subsequently, during progressive Variscan contraction both vein types were passively folded within characteristic, NW-verging, upright to overturned folds of the North Eifel. In contrast to both extensive veining events that characterise the tectonic inversion, quartz veining occurred rather occasionally during the main compression stage of orogeny. This kinematic history eventually shows a clear relationship between fluid-pressure evolution and the stress-state changes in the basin and exemplifies that (supra‑)lithostatic overpressures are easier to maintain during compressional tectonic inversion at the onset of orogeny, than during the main phase of compression. This intimate relationship during the stress-state evolution from extension to compression is illustrated by plotting the changing differential stress (sigma1-sigma3) against the vertical effective stress and the fluid-pressure evolution in a 2D brittle failure mode plot. The results show that the two vein types, which are induced at elevated to (supra‑)lithostatic fluid pressures, can only form at low differential stress closely related in time to the tectonic switch. The tectonic setting and the localised stress state in the basin are thus both crucial to determine whether the lithostatic fluid overpressure can be sustained by a rock prior to failure at depth. These 2D brittle failure mode plots are moreover very useful to visualisethe influence of rock parameters such as the tensile strength of rock on the maximum overpressures that can be built up during the transition between two stress regimes. This regional aspect of fluid redistribution within overpressured fluid reservoirs contrasts with the more localised fluid flow along fault systems caused by fault-valving. The 3D aspect of stress transitions during tectonic inversion of a crust is, however, much more complex than represented in a 2D brittle failure mode plot. The tectonic switch, illustrated in these brittle failure mode plots, occurs at a specific isotropic stress state in which the three principal stresses are equal and sigma1 - sigma3 = 0. The chance that a stress state in the Earth’s crust equals an isotropic pressure state is, however, highly improbable. The 2D stress-state evolution visualised in the brittle failure mode plots is therefore an oversimplification of the actual 3D stress-state evolution in the Earth’s crust. To discuss the 3D aspects of stress transitions and to illustrate the complexity of triaxial stress transitions during inversion of Andersonian stress regimes, possible 3D stress-state evolutions are reconstructed based on the bedding-normal and bedding-parallel veins reflecting the early Variscan tectonic inversion. From these 3D stress-state reconstructions is concluded that, no matter what orientation of basin geometry or shortening, a transitional wrench tectonic regime should always occur between extension and compression. This transitional stage should contribute to the permeability enhancement during tectonic inversion, although structures that are related to this transitional stage have not yet been reported in a shortened basin affected by tectonic inversion at low differential stress. Ideally, a transitional ‘wrench’ tectonic regime should be implemented in brittle mode plots at the tectonic inversion. It has been concluded from this research that the naturally fractured Ardenne-Eifel basin can serve a possible analogue to the present upper crust by its regional extent of overpressuring, but more importantly by demonstrating that a tectonic inversion from extension to compression at the onset of orogeny is the crucial timing during which maximum (lithostatic) overpressures can be sustained.