Description:

_Journal of Structural Geology 32 (2010) 1887-1898_ Contents lists available at ScienceDirect Journal of Structural Geology journal homepage: www.elsevier.com/locate/jsg Multiple causes of diagenetic fabric anisotropy in weakly consolidated mud, Nankai accretionary prism, IODP Expedition 316 Kitty L. Milliken*, Robert M. Reed Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin, 10100 Burnet Road, Austin, TX 78713, USA Article info Article history: Received 4 August 2009; Received in revised form 23 February 2010; Accepted 17 March 2010; Available online 2 June 2010 Keywords: Mud Mudrock Mudrock fabric Compaction Diagenesis Abstract In the Nankai accretionary prism and its associated slope sediments, early (pre-lithification) mechanical modification of mud induces preferred alignments of elongate or platy particles and the loss of intergranular porosity. Generic types of particle alignment include: 1. particles having long axes aligned in the plane of bedding, most likely as a consequence of burial compaction; 2. diverse bioturbation structures including alignments parallel to burrow walls, burrows filled with obliquely aligned phyllosilicates, and blotchy disruption of bedding; and, 3. planar deformation bands showing parallel alignments of both silt- and clay-size particles. Subtle compositional contrast between deformation bands and host rocks is consistent with loss of intergranular micropores within bands and supports the dominance of mechanical over chemical processes in their formation. Field-emission SEM imaging of Ar-ion-milled cross-sections shows that collapse of larger (>2 mm) pores, many localized at the margins of silt-size particles, reduces porosity within the bands by about 5 percent compared to the adjacent host rock. Despite the clear role of shear, evidence for particle comminution is equivocal. These observations on mechanical processes in early diagenesis provide useful context for interpretation of pore types and fabric anisotropies in mudrocks across a wide range of subsurface conditions. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction In accretionary prisms, gravity-driven mass transport and compressional tectonic stresses interact with burial compaction early in the post-depositional history to impart macro- and microscale fabric anisotropies in mud (Karig and Lundberg, 1990; Prior and Behrmann, 1990; Behrmann and Kopf, 1993; Maltman, 1998; Ujiie et al., 2004). The presence of deformational fabrics in the shallow regions of accretionary prisms is of great interest because, owing to low geothermal gradients and resultant low temperatures, these complex early diagenetic mechanical processes can be readily examined without a strong overprint of thermally-driven chemical processes. At the same time, the combination of structural loading and high sedimentation rates in the prism environment yields samples of relative compositional uniformity that can be sampled across a broad range of depths. Thus, the accretionary prism setting is an excellent natural laboratory for examining early mechanical processes in mud. * Corresponding author. Tel.: +1 512 471 6082; fax: +1 512 471 0140. E-mail address: kitty.milliken@beg.utexas.edu (K.L. Milliken). 0191-8141 $ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsg.2010.03.008 Diagenesis is defined as inclusive of chemical and mechanical processes that affect sediments in the subsurface following deposition and prior to the onset of metamorphism or weathering (Milliken, 2003). Historically, however, much of the literature on post-depositional sediment alteration has focused on chemical processes within a conceptual framework that assumes an absence of deformation. Recent technical innovations for micro-scale imaging have fostered a growing recognition of a wide array of pervasive deformation features in rocks having very simple structural histories (e.g., Milliken, 1994; Laubach, 1997; Milliken and Laubach, 2000). Structural diagenesis (see Laubach et al., 2010 this issue) focuses attention on chemical and mechanical processes that affect structures on all scales in sedimentary rocks prior to the onset of metamorphism, emphasizing a more rigorous application of mechanics principles for understanding rock fabrics, even in little-deformed rocks, while also emphasizing the impact of chemical processes on the evolution of rock mechanical properties. Very importantly, this conceptual framework highlights the important genetic linkages that apply in some cases amongst chemical and mechanical processes in sedimentary basins. This study focuses on a case in which the mechanical processes of diagenesis dominate over chemical processes in the modification of 1888 K.L. Milliken, R.M. Reed Journal of Structural Geology 32 (2010) 1887-1898 Fig. 1. A. Bathymetry and locations of Sites C0004, C0008, C0006, and C0007, part of a transect of sites drilled off the Kii Peninsula during Stage 1 of the NanTroSEIZE project. Modified from Moore et al. (2009) and Screaton et al., (2009). B. Interpreted structure cross-section showing the locations of drill sites C0004 and C0008 in the vicinity of the megasplay and Sites C0006 and C0007 near the toe of the prism. Modified from Moore et al. (2009), their Fig. 6. rock fabric and properties. Direct observation using a variety of microscale imaging techniques supports suggestions of previous studies concerning the dominant roles of porosity loss and particle reorientation and the lesser roles of particle comminution and chemical modification in the evolution of early diagenetic fabric anisotropy in mud. Fabric anisotropies and deformation localization documented here are clearly affiliated with embryonic stages in the diagenetic history and provide useful context for interpretation of similar features that may be observed in mudrocks that have experienced a more protracted history of deep burial and, especially, heating that may induce chemical processes such as cementation. This preliminary study reports, in part, results of our experimentation with imaging and analysis protocols that are useful for quantitative characterization of mudrock fabric and pore systems at the micro-scale. 2. Sampling and methods Pleistocene and Pliocene mud samples were obtained at IODP Sites C0004, C0006, C0007, and C0008 during IODP Expedition 316 (Kinoshita et al., 2009; Screaton et al., 2009) (Fig. 1; Table 1). Site C0004 is slightly landward of the mega-splay fault whereas Site C0008 is located at a small slope basin seaward of the intersection of the mega-splay with the sea floor (Strasser et al., 2009). Sites C0006 and C0007 are near the toe of the accretionary prism. Hemipelagic muds for this study range in burial depth from near the modern sediment surface to a depth of 573 mbsf and were obtained routinely, in concert with sampling of turbidite sands for provenance study, one or two samples per core, without regard to specific associations with macroscopically visible deformational features, thus providing an essentially random sampling of mudrock fabrics-Intervals rich in nannofossils and ash layers were avoided; samples described here are dominated by extrabasinal siliciclastic debris. Texturally, samples are dominantly silt-bearing clay-rich mudstone (classification of Macquaker and Adams, 2003), but range into sand- and silt-bearing clay-rich mudstone, and clay-rich mudstone (Fig. 2). Considerable textural heterogeneity is present within some individual thin sections as seen in Fig. 2D. All of the samples this study are weakly consolidated and K.L. Milliken, R.M. Reed Journal of Structural Geology 32 (2010) 1887-1898 1889 can be readily disaggregated into their natural particles by placing them in distilled water and applying moderate sonication. Thin sections of 82 samples were cut perpendicular to bedding, but no attempt was made to orient the samples other than to preserve the bedding direction and in some cases, the up-direction. Thin sections were ground to standard thickness (30 µm). Vacuum-pressure epoxy impregnation is not effective in these very low-permeability muds so surface impregnation using a low-viscosity medium was employed to reduce plucking and to preserve rock fabric during the final polish of the sections. Thin sections were observed in transmitted and reflected polarized light under both plane-light and cross-polarization. A full-wave plate (gypsum plate) was used in transmitted cross-polar observation to accentuate the spatial variations in the orientation of the phyllosilicate components. This method highlights, through visible changes in the amplified birefringence during stage rotation, the overall orientation, in aggregate, of particles that are too small to see individually. Back-scattered electron (BSE) imaging and X-ray mapping by wavelength dispersive spectroscopy (WDS) was performed on a JEOL 8200 Super probe. Mapping runs used either 0.5 or 1 mm stage increments, a 40 ms count time, and a 50 nA sample current (as measured on brass), and a focused spot (~1 µm). Texture and fabric (silt content, particle size, particle shape, particle orientation) within the bands and adjacent host rocks were analyzed from BSE images (1000x original magnification) using the image analysis program JMicrovision (Roduit, 2008). Images were displayed on a tracing screen and all particles >2 µm (9 Phi units; silt clay boundary) were digitally traced and established as 2D objects for determination of equivalent circular diameter and other measures of particle size, shape, and orientation. For two samples data from 3 such images were combined for each determination; data for the third sample is based one image for the band and on one for the host rock. Ключевые слова: e, r, o