University of Southern California Dissertations and Theses

Direct observation of fault zone structure and mechanics in three dimensions: a study of the SEMP fault system, Austria

(USC Thesis Other)

Direct observation of fault zone structure and mechanics in three dimensions: a study of the SEMP fault system, Austria

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DIRECT OBSERVATION OF FAULT ZONE STRUCTURE AND MECHANICS IN THREE-DIMENSIONS: A STUDY OF THE SEMP FAULT SYSTEM, AUSTRIA by Erik Karl Frost A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (GEOLOGICAL SCIENCES) August 2010 Copyright 2010 Erik Karl Frost ACKNOWLEDGEMENTS ii This work largely reflects the vision of James Dolan, my advisor. While I was an enthusiastic participant, I entered the Ph.D. program at USC with a project already set up and waiting for me. James’ ability to see the big picture and fund projects made my Ph.D. much easier than it might have been elsewhere, and for that I am thankful. I could never have completed this project without the help of my committee. James has helped keep me on track throughout the course of my PhD., making sure I never lost sight of the forest for the trees. Charles Sammis spent countless hours explaining brittle fracture processes with me, helping me produce results that I’m quite proud of. John Platt put up with my rudimentary knowledge of metamorphic petrology and helped me figure out just what the heck I was looking at sometimes. Brad Hacker and Lothar Ratschbacher helped me understand what we were seeing in the field, as well as how to analyze it back in the lab. Gareth Seward patiently taught me how to use the EBSD setup at Santa Barbara over and over again. Lastly, many thanks go out to my officemates over the years, Kurt Frankel, Ozgur Kozaci, Lorraine Leon, Plamen Ganev, Ben Haravitch, and Lee McAuliffe., who iii helped keep me sane with ping-‐pong breaks, viral videos, and other necessary diversions. TABLE OF CONTENTS iv Acknowledgements ii List of Figures vi Abstract viii Chapter 1: Introduction 1 Chapter 2: Progressive Strain Localization in a Major Strike-Slip Fault 5 Exhumed from Mid-Seismogenic Depths: Structural Observations from the SEMP Fault System, Austria Chapter 2 Abstract 2.1 Introduction 6 2.2 Geologic Setting 9 2.2.1 Study Site 11 2.3 Methods 15 2.4 Results 19 2.4.1 Structural Data 21 2.4.2 Relative Strain Distribution 25 2.5 Discussion 31 2.6 Conclusions 39 Chapter 3: Direct Observation of Fault Zone Structure at the Brittle- 41 Ductile Transition Along the SEMP Fault System, Austria. Chapter 3 Abstract 3.1 Introduction 42 3.2 Geologic Setting 45 3.3 Outcrop Description 45 3.3.1 Kitzlochklamm 46 3.3.2 Lichtensteinklamm 52 3.4 Optical Microstructures 55 3.4.1 Kitzlochklamm Klammkalk 56 3.4.2 Kitzlochklamm Grauwacken graphite-schist 58 3.4.3 Kitzlochklamm Grauwacken non-graphite schist 58 3.4.4 Lichtensteinklamm Klammkalk 59 3.5 Electron Backscatter Diffraction 63 3.5.1 Kitzlochklamm Klammkalk 64 3.5.2 Kitzlochklamm Grauwacken Zone 66 3.5.3 Lichtensteinklamm Klammkalk 69 3.6 Exhumation Depth 71 3.7 Discussion 73 v 3.8 Conclusions 77 Chapter 4: A comparison of fault zone structures throughout the crust 78 Chapter 4 Abstract 4.1 Introduction 79 4.2 Geologic Setting 80 4.3 Results 83 4.3.1 Results from the Brittle Crust 83 4.3.2 Results from the Brittle-Ductile Transition 86 4.3.3 Results from the Ductile/Brittle-Ductile Crust 90 4.4 Discussion 103 4.4.1 Brittle Fault Zone Structures 103 4.4.2 Brittle-Ductile Transition 105 4.5 Conclusions 109 Chapter 5: Conclusions 110 References 114 LIST OF FIGURES vi Figure 1: Regional Geologic Map 10 Figure 2: Gstatterboden Geologic Map 13 Figure 3: Gstatterboden Sampling Transect 14 Figure 4: Gstatterboden Dolomite Thin Section 18 Figure 5: Gstatterboden Outcrop Photos 20 Figure 6: Structural Data 22 Figure 7: Outcrop-Scale Damage Intensity 26 Figure 8: Dolomite Grain Size Distributions 28 Figure 9: Grain Size Summary Plot 30 Figure 10: Strain Localization at Gstatterboden 36 Figure 11: Schematic Reconstruction of SEMP 38 Figure 12: Regional Geologic Map 2 44 Figure 13: Geologic Map of Klamms 47 Figure 14: Kitzlochklamm Grauwacken Photos 48 Figure 15: Structural Data from Klamms 49 Figure 16: Kitzlochklamm Klammkalk Photos 51 Figure 17: Lichtensteinklamm Photos 53 Figure 18: Kitzlochklamm Klammkalk Photomicrographs 57 Figure 19: Kitzlochklamm Graphite Schist Photomicrograph 60 Figure 20: Kitzlochklamm Mica Schist Photomicrographs 61 Figure 21: Lichtensteinklamm Klammkalk Photomicrograph 62 vii Figure 22: Kitzlochklamm Klammkalk EBSD Analysis 65 Figure 23: Kitzlochklamm Mica Schist EBSD Analysis 67 Figure 24: Kitzlochklamm Carbonate Schist EBSD Analysis 68 Figure 25: Schematic Reconstruction of SEMP 2 70 Figure 26: Strain Localization at Gstatterboden 76 Figure 27: Regional Geologic Map and Rinderkarsee Location Map 91 Figure 28: Rinderkarsee Shear Zone Photos 93 Figure 29: Rinderkarsee Shear Zone Photomicrographs 95 Figure 30: Southern Shear Zone EBSD Analysis 96 Figure 31: Central Shear Zone EBSD Analysis 98 Figure 32: Incipient Shear Zone EBSD Analysis 100 ABSTRACT viii Outcrops of the Salzach-Ennstal-Mariazell-Puchberg (SEMP) fault system exhumed from depths of ~4-17 km allow for the direct observation of fault zone structures throughout the crust, and provide insights into the way this fault, and perhaps others, distributes strain in three dimensions. At Gstatterboden, exhumed from ~4-8 km, grain size distributions and small fault data reveal the presence of a 10-m-wide high-strain core towards which strain localized during fault evolution. Brittle fracture was accommodated via constrained comminution, which only occurs in strain-weakening rheologies and favors localization. Exposures of the SEMP at Lichtensteinklamm and Kitzlochklamm, exhumed from ~12 km depth, bracket the brittle ductile transition. At these outcrops, the SEMP is characterized by a ~70-m-wide, cataclastic fault core that has been altered to clays that transitions downward into a wide, ductile shear zone that has accommodated only minor amounts of strain, placing the majority of displacement on the razor-sharp fault contact. Deformation mechanisms transition from cataclasis and minor amounts of dislocation creep in calcite, to dislocation creep in quartz and calcite occurring against a background of fault-normal solution mass transfer. The ductile/ductile-brittle Rinderkarsee shear zone, exhumed from ~17 km, marks the SEMP’s continuation into the Tauern Window and is composed of three distinct shear zones. The southern, 100-m-wide shear zone has accommodated the most strain, and shows evidence for creep-accommodated grain boundary sliding in feldspar and quartz, while incipient shear zones contain ductile quartz and brittle-feldspars that undergo dislocation creep as fluids alter Kspar to muscovite, which localizes strain ix along felspar grain boundaries, encouraging ductility. These findings are compared to results from other faults exhumed from similar depth ranges, highlighting fundamental fault zone structures and characteristics. CHAPTER 1: Introduction 1 The following chapters present analysis and interpretation of the Salzach-Ennstal- Mariazell-Puchberg fault system (SEMP). This work is the result of a collaborative effort between researchers at the University of Southern California (USC), the University of California at Santa Barbara (UCSB), and the University of Freiberg. With the guidance of my advisors at USC, Dr. James Dolan and Dr. Charles Sammis, I examined outcrops of the SEMP exhumed from the brittle and brittle-ductile crust. Examination of the SEMP at ductile/ductile-brittle exhumation levels was led by Joshua Cole and his Master’s advisor at UCSB, Dr. Brad Hacker. Results from these deeper exhumation levels are combined with my work and then compared to other exhumed faults to present a crustal-scale summary of the variability of fault zone structures. Dr. Lothar Ratschbacher from the University of Freiberg provided valuable insight based on his extensive experience with Alpine geology. The goal of this study was to create a three-dimensional summary of fault zone structure and mechanics by directly observing the characteristics of a single fault. Specifically, we wanted to address the following questions: (1) How does fault zone thickness vary with depth? Can we observe a gradual widening or narrowing, or does thickness vary nonsystematically, reflecting the importance of variables such as host rock and deformation mechanism? 2 (2) Are faults characterized by a narrow principal slip surface and/or gouge zone throughout the seismogenic crust, or is the development of such features dependent upon host rock rheology? (3) Is the damage zone surrounding the fault core a relict feature formed during the early stages of fault evolution, or does damage occur continually with each seismic rupture? (4) How does fault zone geometry change across the brittle-ductile transition, and how abrupt is this transition in deformation mechanisms? While much has already been learned from seismic studies of fault zones, structural studies of exhumed faults, and laboratory studies of earthquake mechanics, the aforementioned issues have proven extremely difficult to address in a systematic fashion on a single fault. Three-dimensional models of fault zones have therefore been constructed by combining multiple datasets, each the product of a distinct set of variables. We sought to eliminate such complexities by studying the SEMP fault system, which has been differentially exhumed such that it exposes a continuum of structural levels along strike, ranging from near-surface conditions to ductile lower crustal outcrops. We focused our study on four key outcrops: Gstatterboden, where the brittle, seismogenic portion of the SEMP is exposed in the Northern Calcareous Alps; Lichtensteinklamm and Kitzlochklamm, which bracket the brittle ductile transition at 3 the northeast corner of the Tauern Window, and Rinderkarsee, which exposes the SEMP’s ductile/ductile-brittle continuation into the middle crust. Chapter 2 presents our results and interpretation of the brittle portion of the SEMP exposed at Gstatterboden. We characterized fault structure at this exhumation depth by measuring the orientation and distribution of damage elements across the fault. These data were then combined with calculations of damage intensity at the macro- and microscales to infer the history of strain accumulation within the fault zone. We find that at this exhumation depth, the SEMP evolved into a relatively narrow (~10m) fault zone by progressively localizing strain into an eventual fault core, although at the outcrop scale features such as a principal slip surface or gouge zone are not evident. Chapter 3 presents our results and interpretation of the brittle-ductile transition within the SEMP as seen at Lichtensteinklamm and Kitzlochklamm. Our analysis at these outcrops followed the same format as at Gstatterboden. We measured the orientation and distribution of damage elements across each fault outcrop, and combined those data with microstructural observations of samples taken along fault-normal transects. Comparison of the results from each klamm shows a gradual downward transition from cataclastic deformation at Lichtensteinklamm to dislocation creep at Kitzlochklamm (possibly in conjunction with localized aseismic slip), all occurring against a background of solution mass-transfer creep. At both outcrops, we would define the high-strain core of the fault as tens of meters wide, at the most. 4 In chapter 4, we combine our results in chapters 2 and 3 with results from the ductile/ductile-brittle outcrop of the SEMP at Rinderkarsee. The Rinderkarsee shear zone is composed of three distinct 1-100 m shear zones separated by low strain panels up to 500 m wide. The southernmost, 100-m-wide shear zone has accumulated the most strain and is interpreted to be the main continuation of the SEMP into the Tauern Window. Using optical microscopy and electron backscatter diffraction, we also find that shear zone nucleation at the grain scale involved dislocation creep and the transformation of plagioclase to muscovite, creating localized strain in what could otherwise be a regime of distributed ductile shear. We conclude chapter 4 by comparing our three-dimensional summary of fault zone structures along the SEMP against other exhumed faults covering a similar range of depth conditions. These comparisons allow us to determine the relative importance of variables such as cumulative displacement, host rock, and geometry, and better understand how faults operate throughout the crust. CHAPTER 2: Progressive Strain Localization in a Major Strike-Slip Fault Exhumed from Mid- Seismogenic Depths: Structural Observations from the SEMP Fault System, Austria. 5 Chapter 2 Abstract Analysis of a strike-slip fault exhumed from mid-seismogenic depths reveals that the fault experienced progressive strain localization toward a high-strain fault core. We focus on the Ennstal segment of the 400-km-long Salzach-Ennstal-Mariazell-Puchberg strike-slip fault system in the Eastern Alps, which accommodated ~60 km of left- lateral displacement during Oligo-Miocene time. Macroscopic and microscopic observations reveal a zoned fault featuring a high-strain core at least 10 m wide within a fault zone that is at least 150 m wide. Grain-size distribution analysis shows how the Ennstal segment of the SEMP evolved. Our data reveal a 10-m-wide high-strain fault core (characterized by a power-law relationship of grain sizes, D 2 ≈ 2.0) bordered by a 54-m-wide “transition zone” where the largest and smallest grains are characterized by two power-law relationships (D 2 ≈ 2.0 and 1.6, respectively). This zone is in turn bordered by a region with grain sizes that show a single low-strain power-law relationship of D 2 ≈ 1.6. We interpret these relationships to be the result of concentrated shear overprinting an initial low-strain, power-law grain-size distribution before strain localized to the core. This is consistent with the theory that faults mature by smoothing geometrical complexities, forming a highly localized, high-strain fault core. The data do not support the idea that damage forms primarily in response to 6 dynamic stresses during seismic rupture, although they do suggest that this mechanism may operate within tens of meters of the fault once it has developed its zoned structure. 2.1 Introduction One of the most important frontiers in earthquake science is the linkage between the structure and mechanical behavior of fault zones. In particular, little is known about how fault-zone structure varies as a function of depth, from near the surface down through the seismogenic crust and into the ductile lower crust. Studies of exhumed faults have delineated the presence in many outcrops of a meters-thick core that has accommodated the bulk of fault displacement, often localized into several centimeters to some decimeters of ultracataclasite (Anderson et al., 1983; Chester and Chester, 1998; Chester et al., 1993; Chester and Logan, 1986; Caine et al., 1996; Evans et al., 2000, Faulkner et al., 2003; Wibberly et al., 2003). The fault core may exhibit strong foliation, as well as distinct, millimeters-thick planar structures that are interpreted to have served as principal slip surfaces. This area is bordered by a gouge and breccia zone that is generally non-cohesive, coarser-grained, and lacks the shear structures characteristic of the fault core. The verge of the fault zone is marked by fractured wall rock, where fracture density gradually decreases away from the fault to the regional background level. 7 The actual process by which this zoned fault structure develops remains unresolved. Quasi-static models suggest that zoning occurs in structurally mature, large- displacement faults in which initial geometrical complexity has been reduced, at scales ranging from millimeters to kilometers. Power and Tullis (1987), for example, proposed that fractal surface roughness on a fault would lead to the development of a high-strain fault core that widens with increasing displacement, as larger and larger asperities are eliminated. This idea can be extended to the scale of step-overs, as the discontinuities between small faults are eliminated to create a smooth, through-going fault plane (Segall and Pollard, 1983; Wesnousky, 1988). In quasi-static models, the gouge and breccia zone outside the fault core is viewed as a relict feature that accommodated shear strain early in the fault’s history, before strain localized to a narrow fault core. An alternative hypothesis is that fault-core gouge forms dynamically during seismic ruptures (Rice et al., 2005). In this model, stress concentrated at the tip of each passing earthquake rupture cracks and shears the rocks near (within ~40 m) the rupture plane. Each subsequent earthquake reworks the gouge, resulting in a zoned fault. In this model, the fault core is again understood to be a high-strain feature, but the gouged and brecciated rocks outside the core are thought to have shattered in place by seismic waves and experienced only minimal strain. 8 These hypotheses are not mutually exclusive, and in fact may operate simultaneously. Both predict the presence of a high-strain fault core, with perhaps a greater width expected in the dynamic model. Outside the fault core, the quasi-static model predicts that the gouge and breccia zone has experienced more shear strain than in the dynamic model. In order to better understand the relative contribution of each process to the evolution of a fault zone, it is therefore necessary to examine how strain is distributed throughout the entire fault zone, and whether the mechanics of fragmentation are the same inside and outside the fault core. In this study, we examine the strain gradient and fragmentation mechanisms at an outcrop of the Salzach-Ennstal-Mariazell-Puchberg fault system (SEMP), a 400-km long strike-slip fault system with ~60 km of displacement in the Eastern Alps that has been differentially exhumed such that it exposes a continuum of structural levels along strike, from the near-surface downward across the seismic zone, through the brittle- ductile transition, and into the ductile lower crust. The results of this study allow us to assess the factors governing fault evolution in the brittle upper crust, and, combined with future studies at deeper outcrops, provide a comprehensive picture of the mechanical behavior of a major fault zone. 9 2.2 Geologic setting The Salzach–Ennstal–Mariazell–Puchberg (SEMP) fault is a primarily sinistral strike- slip fault zone that extends 400 km across the Eastern Alps (Fig. 1) (Ratschbacher et al., 1991a,b; Linzer et al., 2002). From west to east, the Salzach fault forms the northern boundary of the Tauern window (exposing European (Penninic) units beneath African (Austroalpine) units), the Ennstal fault forms the boundary between the central part of the Northern Calcareous Alps and the basement of the Austroalpine unit, and the Mariazell-Puchberg fault cuts the southern margin of the eastern part of the Northern Calcareous Alps. The Northern Calcareous Alps constitute the Late Permian to Cretaceous cover sequence of the Austroalpine basement unit and consist of mostly massive limestone and dolomite and interbedded shale and evaporite. The Northern Calcareous Alps were transformed into a thin-skinned thrust belt during the Cretaceous and dissected by mostly strike-slip faults during the Oligocene-Recent eastward displacement of crustal wedges in the eastern Alpine–western Carpathian orogen (e.g., Linzer et al., 1995; 2002; Ratschbacher et al., 1991a,b). Structural and kinematic analysis of the southernmost nappe boundaries of the Northern Calcareous Alps (Ratschbacher et al., 1991a,b; Decker et al., 1994; Linzer et al., 1995) demonstrates that the individual segments of the SEMP constitute a through-going fault that accommodated orogen-parallel extension. Models have explained this extension as a result of lateral extrusion (Ratschbacher et al., 1991a,b) or oblique indentation of the South Alpine block (Rosenberg et al., 2004). Geochronology within 10 Figure 1. The Salzach-Enntal-Mariazell-Puchberg (SEMP) fault zone and major structural features of the Eastern Alps (after Linzer et al., 2002 and Cole et al., 2007). Differential exhumation of the fault exposes a range of exhumation depths along strike, from ductile shear zones in the Tauern Window, to brittle-ductile outcrops near Taxenbach, and finally brittle fault zones of decreasing exhumation depth to the east. Our study site just north of Gstatterboden is at the eastern end of the Ennstal fault segment, exposing the upper seismogenic crust. A detailed geologic map of the Gstatterboden study area is shown in figure 2. 11 the Tauern Window and the ages of intraorogenic basins constrain the age of this deformation as Oligocene and Miocene (e.g., Linzer et al., 2002).The differential exhumation of the SEMP fault zone along strike resulted from greater shortening in the western and central Eastern Alps than in the eastern Eastern Alps, creating an overall higher amount of exhumation in the western part of the Eastern Alps. Combined with subsidence in the easternmost Eastern Alps and Pannonian basin, the entire SEMP fault zone was tilted eastward, resulting in exhumation depths of >20 km in the Tauern Window, decreasing eastward along strike to zero exhumation in the Vienna basin (Ratschbacher et al., 1991a,b). Deformation structures commensurately change from dominantly ductile where the SEMP enters the Tauern Window at Rinderkarsee (Cole et al., 2007) to ductile–brittle along the eastern end of the Salzachtal segment. Farther east, deformation changes to dominantly brittle along the Ennstal segment before shallowly exhumed levels of the SEMP are revealed along the Mariazell–Puchberg segment. 2.2.1 Study site The Ennstal segment represents the upper crustal portion of the SEMP (i.e., the “seismogenic zone”) along the Northern Calcareous Alps. In this region, the fault also locally branches northward to form NNE-striking thrust faults and duplex structures within the Northern Calcareous Alps; post-Middle-Miocene deformation is evident in the SW-directed thrust faults within the sedimentary basins near Wagrain and Hieflau (Peresson and Decker, 1997a; Wang and Neubauer, 1998). Total displacement along the Ennstal fault has been estimated at 60 km based on the offset between quartz 12 phyllite units and the Greywacke zone (Linzer et al., 1997; 2002). Our study site is 1 km north of the town of Gstatterboden at the eastern end of a subtle releasing bend in the Ennstal segment, in the Gesäuse National Park (Fig. 1). Based on the well-known stratigraphy of the Northern Calcareous Alps, the Rauwacken dolomite at this site should once have been buried by ~ 4 km of overlying sedimentary rocks (Geologische Bundesanstalt, 1980). However, thrusting across the NCA prior to Miocene strike-slip tectonics (Linzer et al., 1995) is thought to have at least doubled the section. Based on these arguments, the Gesäuse outcrop has been exhumed from 4–8 km depth. At the Gesäuse outcrop, the Ennstal fault zone comprises two strands (Figs. 2, 3). The southern strand is poorly exposed along the southern edge of a narrow (100–400 m wide) lens of resistant, cliff-forming limestone of the Dachsteinkalk that appears to have been slivered off a larger body of Dachsteinkalk ~4 km to the east. The northern (main) strand juxtaposes the Dachsteinkalk sliver against Rauwacken dolomite to the north; this strand, which is the focus of this study, has accommodated most of the 60 km of cumulative displacement attributed to this part of the SEMP (Linzer et al., 2002). At our study site along the northern (main) strand, the fault zone is exposed in a streambed and roadcut. Both exposures provide continuous outcrop of the Rauwacken dolomite for a fault normal distance of ~200 m. Following the stream toward the limestone to the south, however, colluvium and vegetation progressively obscure the 13 2 km 1 0 1000m N 47° 35’ 31.31” N 14° 38’ 07.88” E Gstatterboden Fig. 3 Figure 2. Geologic map of the Gstatterboden study area (from Ampferer, 1935). The Ennstal segment of the SEMP has been mapped on the northern and southern boundaries of the Dachsteinkalk sliver. Displacement along the southern fault strand is ~ 2 km, placing the majority of the ~ 60 km of fault slip on the northern strand. 14 Figure 3. Map of sampling area north of Gstatterboden. Samples of dolomite were collected along the roadcut at 5 meter intervals, outcrop permitting, for the first 50 meters and then every 50 meters beyond that. 15 outcrop. The Dachsteinkalk unit is better exposed along the roadcut, which provides another 50 m of continuous outcrop south of the fault. Unfortunately, the fault surface at the limestone/dolomite contact along both the road cut and stream exposure is buried beneath a 3–5 m wide section of thick colluvium composed of meter-scale limestone boulders. Notwithstanding the lack of exposure at the fault contact, this study site still allows fundamental observations about the structural evolution of this major strike-slip fault zone. 2.3 Methods We documented the major structural properties of this outcrop by measuring the orientation and, where possible, the slip direction and slip sense, of faults and fractures throughout the fault zone. From these measurements, we calculated the reduced stress tensor (Angelier, 1984) and associated stress ratio R by the numerical-dynamic analysis (NDA) of Spang (1972) implemented by Sperner and Ratshbacher (1994). The NDA method assumes an arbitrary shear-stress magnitude of 1 acting along each fault in the direction of the striae. Summation of the stress tensors of all fault-striae sets and division by the number of faults yield the bulk stress tensor. The orientation and relative values of the principal stresses are derived from the eigenvalues and eigenvectors of the bulk stress tensor. The stress ratio R, defined as (σ2-σ3)/(σ1-σ3), expresses the relationship between the magnitude of the principal stresses. Extreme 16 values of R correspond to stress ellipsoids with σ2=σ3 (R=0; oblate ellipsoid) or σ1=σ2 (R=1; prolate ellipsoid). We also measured the degree of damage throughout the fault zone at both the macro- and microscales. At the macroscopic scale, the degree of brittle deformation was quantified using damage intensity (Chester and Logan, 1986; Schulz and Evans, 2000). Damage elements such as fractures, faults, and shear bands were counted along two perpendicular, 30-cm-long lines at multiple stations along a fault-normal transect, and the resulting damage intensities were plotted against perpendicular distance from the fault. We assessed damage at the microscopic scale by calculating the grain-size distribution of samples taken along the same fault-normal transect; the grain-size distribution is considered to indicate the degree of comminution a sample has experienced (Blenkinsop, 1991). Thin sections were cut parallel to the strike of the fault and photographed at magnifications of 25x and 200x with an optical microscope, with the respective photomosaics imaging 4 x 2 cm and 1.9 x 0.9 mm, on average. Areas analyzed at 200x were chosen so as to avoid particles larger than the largest size being counted. Grain sizes were calculated using ImageJ, an open-source image-processing program developed by the National Institutes of Health (http://rsb.info.nih.gov/ij/). Grayscale differences were used to distinguish individual dolomite grains (Fig. 4). The grains were then sorted into size classes from 4 mm to 5 µm in diameter, with each 17 class differing from the next by a factor of 2. All graphs are log–log plots of grain size versus grain density, and the power-law line of best fit connecting datapoints is the two-dimensional grain-size distribution, or mass dimension “D 2 ”, of the rock. The standard deviation of each grain size class is computed by analyzing each image at four threshold values above and below the optimal Otsu (1979) threshold (Fig. 4). Calculating the mass dimension of each sample allows us to evaluate proposed fragmentation mechanisms. A number of investigations of natural fault rocks have found that the gouged and brecciated zone outside the fault core produces D 2 values around 1.6 (e.g., Anderson, 1983; Sammis et al., 1987; Chester et al., 1993, 2005; Billi and Storti, 2004), whereas the fault-core rocks are characterized by D 2 values of 2.0 (Billi and Storti, 2004, Chester et al., 2005). These findings are consistent with the “constrained comminution” model for fragmentation (Sammis et al., 1987; Sammis and King, 2007). However, studies of carbonate cataclastic rocks (primarily from the Mattinata fault), have found that D 2 gradually increases from 0.88 at the periphery of the fault zone to 2.5 at the core (Billi et al., 2003; Billi, 2007; Storti et al., 2003). While the basic observation of these papers, that D increases toward localized zones of shear, agrees with findings from other faults zones, the variability of D values reported at carbonate fault zones has led these authors to propose an alternative explanation. Storti et al. (2003) suggested that damage in these fault zones is the result of a gradual transition from fragmentation to abrasion as the degree of comminution increases toward the fault core (although it is worth noting that this interpretation does not offer 18 (a) (b) (c) Figure 4. (a) A thin section of dolomite taken ~5 m into our sampling transect, shown at 25x magnification, with a scale bar of 4mm. (b) Magnifying the box highlighted in part (a), we see evidence that grains are composed of distinct smaller particles. This suggests the dolomite may have experienced episodes of fracture and healing, in keeping with predictions from the dynamic model of fault zone evolution. (c) Grains are resolved and analyzed by calculating the optimal grayscale threshold value via Otsu’s (1979) method in ImageJ. The Otsu threshold is the value that minimizes the grayscale variance within the foreground and background. 19 a mechanical explanation of power-law grain-size distributions at any D value). It remains unclear whether these differences in mass dimension truly reflect differences in mechanical behavior, or whether they are instead influenced by other factors such as lithology or structural maturity. Our study of cataclastic carbonates from the large- displacement Ennstal segment of the SEMP therefore provides additional insight into the controls governing the mechanics of fragmentation. 2.4 Results At the outcrop scale, fault damage along the northern fault strand is asymmetric. South of the fault, the limestone is cut by subsidiary faults and fractures, including meter- scale Riedel shears, but is otherwise largely intact without brecciation (Fig. 5). In contrast, the dolomite to the north has been pervasively reduced to millimeter- to centimeter-scale fragments throughout the length of available outcrop (~200 m fault- normal distance). The dolomite is also cut by numerous faults and through-going fractures, ranging from cm-scale fractures with no displacement, to multi-meter-long faults with gouge zones up to several centimeters thick. These observations lead us to assume that the fault here was seismogenic because: (1) slow, aseismic fault creep would not generate the brittle features observed in the dolomite; (2) the fault zone at Gstatterboden contains neither evaporite minerals (e.g. salt, gypsum), nor serpentinite, which typically characterize creeping continental faults. 20 Figure 5. (a) The Dachsteinkalk to the south of the Ennstal fault is cut by subsidiary faults and fractures, but does not show any evidence of brecciation, in contrast to (b) the Rauwacken dolomite to the north. (c and d) The dolomite is pervasively shattered and sheared throughout the width of available outcrop (~ 200 m), presenting what appears at first glance to be a very consistent degree of damage throughout the outcrop. 21 2.4.1 Structural data We measured the orientation, and where possible, the slip direction and slip sense, of >600 faults and fractures in the dolomite along a 160-m-long fault-normal transect. When binned in 20-m-wide intervals measured perpendicular to the fault trace, the small-fault data display a distinct pattern (Fig. 6). Closest to the limestone–dolomite fault contact, most of the small faults are subparallel to the master fault and subvertical. For the intervals between 20 and 100 m, the dominant small fault set is conjugate to the master fault with N and NNE strikes and dips that shallow farther from the main fault. Beyond 100 m, subsidiary fault orientations appear to be more evenly distributed, with little evidence for a preferred fault set. These data show that, at least within the first 100 m of the master fault, subsidiary faults occur in two key orientations. The fault-parallel orientation is most clearly expressed within the 20 m closest to the master fault, and diminishes with distance as the N and NNE conjugate fault orientation becomes more dominant. This transition coincides with the increasing activity of normal and thrust faulting with distance to the master fault, also reflected in the general increase in the stress ratio R (Fig. 6). A more detailed kinematic analysis of the small fault data is unwarranted because outcrop conditions greatly limited our ability to measure lineations (and therefore constrain R). 22 Figure 6. (a) Lower hemisphere, equal area fault-striae diagrams for each 20-m fault- trend-normal bin with reduced stress tensor. Parameters shown: σ1≥σ2≥σ3, the principal stresses; θ, the dihedral angle between σ1 and the fault measured in the plane that contains the pole to the fault and the associated striae; R, the stress ratio; n, the number of fault-striae measurements plotted; n-calc, the number of measurements used for calculation. Faults are drawn as great circles and striae are drawn as arrows pointing in the direction of hanging wall transport. Arrows around the plots give calculated local orientation of subhorizontal principal compression and extension. (b) Histogram of fault strike in 20° azimuth intervals and fault dip in 10° intervals for each 20-m bin. (c) As the great circle diagrams in (a) get illegible with large data sets, the fault-striae data of (a) are also plotted as poles to fault planes and with corresponding lineations drawn in the pole points as tangents to the common great circle of the fault pole and lineation. 23 Figure 6: Continued 24 Figure 6: Continued 25 We interpret this pattern to indicate the presence of a high-strain fault core within the 20 m closest to the master fault, possibly extending out as far as ~40 m. Small faults in this interval are almost exclusively fault parallel, with dips noticeably steeper than in any other interval. This is consistent with observations from other exhumed fault cores, which tend to exhibit fault-parallel foliations and shear structures (e.g., Chester et al., 1993; Schulz and Evans, 2000). 2.4.2 Relative strain distribution To further estimate the fault-perpendicular strain gradient within the fault zone, we quantified both macro- and microscopic damage. Calculation of the macroscopic damage intensity taken along a fault-perpendicular transect clearly reflects the asymmetric damage pattern, with almost all the deformation accommodated in the dolomite (Fig. 7). Within the dolomite, damage intensity is greatest close to the fault and decreases by ~25% at a distance of 150 m. These results could be due to a very high background level of deformation within the dolomite, which would blur the distinction between fractured wall rocks outside the fault zone and damage within the gouge and breccia zone, or could instead reflect the fact that at this depth, the brittle fault zone is at least 150 m wide. To determine whether our results are complicated by high levels of background deformation, we also examined microscopic grain-size distributions from samples along the same fault-perpendicular transect (Fig. 3). Samples that are characterized by a power-law relationship of grain sizes would 26 Figure 7. Measurements of macroscopic damage intensity along the Ennstal segment of the SEMP at Gstatterboden clearly reflect the asymmetric damage pattern observed between the limestone and dolomite. Damage intensity decreases by ~25% at a distance of 150 m in the dolomite. This could reflect either a very high background level of damage, or a fault zone that has distributed strain throughout at least 150 m during its lifetime. 27 indicate the activity of one of several mechanical processes thought to be active within the fault core and gouge and breccia zones. The lack of a power-law relationship, however, would suggest that the given sample is outside the fault zone, in the fractured wall rocks. In other words, the presence of a power-law distribution would indicate the creation of damage elements through fault-related shear strain, whereas the absence of a power-law distribution would imply comminution without significant shear strain (e.g., in-situ shattering near the lithologic contact, or unrelated background deformation at the margin of the fault zone). Samples of dolomite closest to the fault (within the 10 m of dolomite adjacent to the covered interval) yielded mass dimensions of D 2 ≈ 2.0 (Figs. 8, 9). At a distance of 109 m from the fault, the mass dimension of the dolomite has fallen to a value of D 2 ≈ 1.6. Between these samples, however, the distribution of grain sizes is best characterized by two power-law relationships. For samples taken 10 m to 64 m from the fault, the mass dimension at 25x magnification is ~2.0, whereas the mass dimension at 200x magnification is always closer to 1.6 (Fig. 9). The largest grains, in effect, share the characteristics of a high-strain gouge typically found near shear localizations, whereas the smallest grains more closely resemble what is typically found in a gouge and breccia zone. Such a pattern has not been reported from any other exhumed fault as far as we are aware. 28 Figure 8. Representative grain size distribution plots from the three different zones observed. For each sample, grains were sorted into bins based on mean diameter in pixels (534 pixel/mm at 25x, 7 pixel/µm at 200x), with each bin differing from the next by a factor of 2. Grain size was then plotted against relative area fraction for each size bin. (a) Samples within 10 meters of the fault display mass dimensions of D2 ≈ 2.0 for grains sized 4 mm to 5 µm. (b) Samples 10 to 64 m from the fault display consistently variable mass dimensions. The largest grains (2 mm to 62.5 µm) produce the typical high-strain fault-core mass dimension of D2 ≈ 2.0, while the smallest grains (125 to 5 µm) display more closely resemble damage zone gouge and breccias with a mass dimensions of D2 ≈ 1.6. (c) By a distance of 100 m from the fault, mass dimensions are nearer to the typical damage zone value of D2 ≈ 1.6 for all grain sizes. 29 Figure 8: Continued 30 Figure 9. Mass dimensions for the largest (black circles) and smallest (white circles) grains in each sample. Samples closest to the fault produce results typically found in high-strain fault-cores. For the 5 samples from 10 m to 64 m in the “transition zone,” we find two power-law relationships. Understood in the framework of constrained comminution, the largest grains in this zone must have been subject to a high-strain concentrated flow shear, yet the smallest grains show little evidence of this high-strain event. They instead more closely mirror what is seen in a low-strain damage zone. 31 2.5 Discussion These observations of fault zone structure can be used to evaluate the evolutionary processes that shaped the Ennstal fault. End-member models of fault-zone evolution propose that the zoned structure commonly found in exhumed faults is created by either dynamic or quasi-static stresses. The dynamic model predicts that each earthquake rupture cracks and shears the rocks close to (within ~40 m) the rupture plane, while the gouge and breccia zone rocks are shattered in place and experience almost no strain (Rice et al., 2005). Quasi-static models, on the other hand, require the entire width of the fault zone to experience strain as increasing displacement eliminates geometrical complexities, bringing larger and larger asperities into contact. Although we do not observe a clear structural zonation at the outcrop scale such as a fault core characterized by ultracataclasites and principal slip surfaces bordered by a less-cohesive zone of gouge and breccia, our measurements of small-fault data and grain-size distributions indicate a clear zonation of strain. The presence of numerous fault-parallel, sub-vertical subsidiary strike-slip faults in the 20-40 m of dolomite closest to the fault contact suggests that displacement was preferentially accommodated in this region. Furthermore, samples taken within 10 m of the fault contact display mass dimensions of D 2 ≈ 2.0 at all grain sizes, a feature unique to high-strain fault cores (Billi et al., 2004; Chester et al., 2005). These data indicate that the fault core at this outcrop of the Ennstal fault is at least 10 m wide, possibly 32 extending out to 40 m. The decrease in mass dimensions beyond 10 m, first seen in the smallest grains and eventually in all grains, combined with an increase in the variability of subsidiary fault orientations beyond 40 m, suggests that the fault core is bordered by a zone of lower-strain brecciation. Additionally, the bimodal distribution of power-law grain-size distributions observed out to a distance of 109 m is consistent with the fragmentation mechanism of constrained comminution operating throughout the dolomite (Sammis et al., 1987; Sammis and King, 2007), indicating that the relatively strong damage intensity observed as far as 100-150 m from the fault contact represents deformation within the fault zone. This is perhaps not surprising, given that other studies of carbonate fault zones document widths of 200-300 m (e.g., Billi et al., 2003; Billi and Storti, 2004). Our first-order observation of a high-strain fault core surrounded by a low-strain zone of brecciation and gouge is consistent with findings from other exhumed faults that have shown evidence of constrained comminution (Sammis et al., 1987; Chester et al., 1993, 2005; Evans et al., 2000; Billi and Storti, 2004). The fact that we do not find a variety of D values, as reported for carbonates of the Mattinata fault zone (e.g., Storti et al., 2003), leads us to suggest that these differences are not caused by differences in lithology. The major difference between the Mattinata and Ennstal faults is structural maturity: the Mattinata fault is estimated to have accommodated <2 km slip, whereas the Ennstal segment of the SEMP exhibits ~60 km of cumulative displacement (Linzer et al., 2002; Billi, 2003). One effect of lithology, however, may be related to the width 33 of the fault core. In contrast to the Punchbowl and San Gabriel faults, which cut through crystalline and siliciclastic rocks, the high-strain core of the Gstatterboden dolomites, as defined by mass dimensions, is at least an order of magnitude wider. While both quartzo-feldspathic and calcite gouges have demonstrated strain- weakening behavior in laboratory experiments, (e.g., Logan et al., 1992; Gu and Wong, 1994; Beeler et al., 1996), the added presence of mica at the Punchbowl and San Gabriel faults may favor extreme localization. Carbonate fault zones, on the other hand, have been shown to produce fault cores on the order of 20 m wide (Billi and Storti, 2004). Microscopic observations of the Gstatterboden outcrop suggest that this zoned fault structure is the result of progressive localization of strain within the dolomite. As noted above, all samples between 10 m and 64 m are characterized by two power- laws, with mass dimensions of ~ 2.0 in the largest grains and ~ 1.6 in the smallest grains (Fig. 9). We propose that this “transition zone” has recorded the history of strain localization within the dolomite during increasing displacement along the SEMP. The development of a gouge with a mass dimension of D 2 ≈ 1.6 has been demonstrated by experiments to coincide with the transition from velocity- strengthening to velocity-weakening, which favors strain localization (Biegel et al., 1989). In contrast, a mass dimension of D 2 ≈ 2.0 is thought to be the product of concentrated flow shear once strain has localized (Sammis and King, 2007). This second phase of fragmentation reworks the initial, geometrically stable arrangement of 34 grains, placing once-isolated grains back into contact with each other, causing one to fracture. Fragmentation starts at the largest grain size because these grains have the highest probability of coming into contact with each other in the low-strain gouge. Our observations from the transition zone lead us to propose that this zone only partially experienced the second phase of fragmentation as strain localized toward the fault core. We postulate that the initial phase of localization focused strain into a zone at least 64 meters wide (the distance to which a power-law relationship of 2.0 can be observed), inducing a concentrated flow shear that reworked the mass dimension of these rocks starting at the largest grain size (Fig. 8). Before this second phase of fragmentation could work its way down to the smallest grains, however, strain localized to an even narrower zone, forming the fault core characterized by D 2 ≈ 2.0 for all grain sizes, and leaving behind a volume of fault rock that records the history of strain localization (Fig. 10). We therefore interpret this as a “Type II” fault zone according to the classification of Means (1995). A “Type I” fault, which shows the same strain profile we document here but thickens with time, requires zoning to develop as a result of strain hardening. However, that interpretation is inconsistent with evidence presented here for the activity of constrained comminution. We can think of no mechanical process that would explain a Means (1995) Type III fault zone, in which strain accumulates in all parts of the fault at the same time, but at a slower rate at the periphery of the fault. 35 Taken together, these data and interpretations are consistent with the quasi-static model of fault-zone evolution, which suggests that our macroscopic observation of pervasively damaged rock at a fault-perpendicular distance of ~ 150 meters from the fault is largely a relic of the early stages of fault formation, reflecting the scale of initial geometrical complexity. However, while our data suggest that quasi-static mechanisms were largely responsible for the formation of the zoned Ennstal fault, it is certainly possible that zoning has been reinforced by dynamic forces. Grains analyzed in thin section are composed of smaller particles that have been cemented together, which could indicate repeated stages of shattering due to dynamic rupture followed by healing (Fig. 4). Our study at Gstatterboden is only one part of a larger investigation of fault-zone structure at multiple exhumation depths along the SEMP fault system (Fig 11). By comparing our observations from the seismogenic zone at Gstatterboden with future studies at the brittle-ductile transition, near Taxenbach, and down into the ductile middle crust (Cole et al., 2007; Rosenberg and Schneider, 2008), we can reconstruct the three-dimensional structure of the SEMP and evaluate the manner in which the structure of the fault has evolved. Such understanding is vital if we are to understand the mechanical instabilities that control the nucleation and propagation of seismic ruptures. 36 Figure 10. Schematic depiction of strain localization. White regions are the active zones of deformation. (a) During the early stages of fault formation, strain is distributed throughout a wide region. Deformation is accommodated at the grain-scale by constrained comminution, which produces a mass dimension of D 2 ≈ 1.6 everywhere. (b) Since constrained comminution favors shear localization, the next phase of deformation is focused into a narrower region. This concentrated flow shear disturbs the previously-stable geometrical arrangement of grains and shifts the mass dimension of the rock toward D 2 ≈ 2.0, starting at the largest grains. (c) Before the second phase of deformation is completed down to the smallest grains, strain again localizes to what will become the fault-core. This leaves behind a volume of rock that records the history of strain localization within the fault zone. 37 Figure 10: Continued 38 Figure 11. Schematic reconstruction of the SEMP fault system during Oligo-Miocene time (after Cole et al., 2007). At our study site north of Gstatterboden, we observe an asymmetric fault zone at least 150 m wide with a high strain core ~10 m wide (based on grain-size distributions), and up to 40 m wide (based on small fault data). At greater exhumation depths (15-25 km), the SEMP is characterized by a series of brittle-ductile and ductile shear zones that span less than 1 km, with the majority of strain accommodated in a single 100-m-wide shear zone (Cole et al., 2007). At the deepest (> 20 km) studied exposures of the SEMP, in the Ahorn shear zone, strain is distributed within a 2-km-wide mylonite belt (Rosenberg and Schneider, 2008). Future work at brittle-ductile exposures of the SEMP near Taxenbach will further constrain the three-dimensional structure of the fault. 39 2.6 Conclusions Previous investigations of the structure of exhumed, large-displacement faults have led to the formulation of a zoned fault model in which the majority of slip has been accommodated in a narrow fault core. Whether such zoning is the result of strain localizing toward the fault core during progressive slip along the fault (quasi-static model), dynamic stresses shattering and shearing rocks near the fault plane during successive earthquake ruptures (dynamic model), or a combination of the two, has been a matter of debate. While we find evidence of both processes, we conclude that the Ennstal segment of the SEMP fault zone in Austria evolved primarily by progressive localization of a wide zone of deformation toward a high-strain fault core. Specifically, grain-size distributions support the idea that the bulk of the slip on the SEMP has been accommodated in a zone at least 10 meters wide. Additionally, at least 54 m of dolomite adjacent to the high-strain zone contains structural features typical of both the high-strain core and low-strain gouge and breccia zone. We propose that this transition zone formed as a result of concentrated shear flow that partially overprinted the initial low-strain mass dimension in these rocks, creating a record of strain localization along this part of the exhumed SEMP fault system. The possible existence of an even-higher strain fault core within the 3-to-5-m-wide unexposed interval at the fault contact leaves open the possibility that slip in that interval is even more localized. This highlights the fact that evidence for a broad zone of deformation across a fault 40 may only be taken to represent the finite strain that has accumulated over the lifetime of the fault, and that the most recent increment of infinitesimal strain may be confined to a much more discrete zone. CHAPTER 3: Direct Observation of Fault Zone Structure at the Brittle-Ductile Transition along the SEMP Fault System, Austria 41 Chapter 3 Abstract Analysis of a strike-slip fault exhumed from brittle-ductile depths reveals a high degree of strain localization throughout this transition zone. We focus on the Salzachtal segment of the 400-km-long Salzach-Ennstal-Mariazell-Puchberg (SEMP) strike-slip fault system in the Eastern Alps, which accommodated ~60 km of left lateral displacement during Oligo-Miocene time. Outcrops of the Salzachtal fault segment at Lichtensteinklamm and Kitzlochklamm expose dominantly brittle and ductile fault zones, respectively, within 18 km of each other along strike of the SEMP. Observations at the outcrop- and grain-scale reveal a transition from dominant cataclasis to dislocation creep across the brittle-ductile transition, all occurring against a background of fault-normal shortening accommodated via solution mass transfer. The increasing importance of dislocation creep at Kitzlochklamm does not result in a wider fault zone, rather, this outcrop of the Salzachtal fault segment is characterized by a razor-sharp contact that we interpret to have accommodated at least tens of kilometers of displacement. This is supported by electron backscatter diffraction analysis of the marble tectonite and phyllite host rocks, which record only weakly developed crystallographic fabrics across the fault zone. 42 3.1 Introduction Knowledge of the structure and mechanics of fault zones at the brittle-ductile transition is vital to our understanding of the rheological characteristics of the base of the seismogenic zone, where many large earthquakes nucleate. Field and experimental studies have shown that, rather than being a simple transition from brittle faulting to ductile creep, deformation in this part of the crust can encompass a wide range of deformation mechanisms dependent upon pressure, temperature, strain rate, grain size, fluid activity, mineralogy, phase transformations, and microstructure (e.g., Tullis and Yund, 1977; Carter and Kirby, 1978; Sibson, 1980; Passchier, 1982; Sibson, 1982; Hobbs et al., 1986; Rutter, 1986; Janecke and Evans, 1988; Scholz, 1988; Shimamoto, 1989; Hacker and Christie, 1990; Tullis and Yund, 1992; Chester, 1995; White, 1996; Hacker, 1997; Montesi and Hirth, 2003; Shigematsu et al., 2004; Lin et al., 2005). In one widely accepted model of the structure and mechanics of faulting in the crust (Scholz, 2002, deformation mechanisms change across the brittle-ductile transition from cataclastic flow to crystal-plastic flow concurrent with the onset of interseismic quartz plasticity at ~300˚C. The resulting interpretation is that the seismogenic crust deforms via frictional sliding, with strength increasing into the middle of the brittle- ductile transition (BDT). In the mid- to lower-BDT, strength begins to decrease as the crust deforms via temperature-controlled flow. Assuming a constant lithology, this model also suggests that faults narrow downward toward the brittle-ductile transition, 43 and then gradually widen in the lower brittle-ductile transition as discrete slip surfaces disappear. Solution creep is expected to operate at all depths (Brace and Kohlstedt, 1980; Sibson, 1984; Chester, 1995), reducing fault strength compared to the traditional two-mechanism model. This study presents direct observations of a strike-slip fault exhumed from brittle- ductile conditions which, together with companion studies at both shallower and deeper exhumation levels, allows us to provide insights that further refine models of fault behavior throughout the crust (Cole et al., 2007, Frost et al., 2009). The unique opportunity to directly observe depth dependent changes on a single fault is afforded to us by the Salzach-Ennstal-Mariazell-Puchberg (SEMP) fault zone in central Austria. The SEMP fault zone has been differentially exhumed along strike, exposing a range of exhumation depths from near-surface conditions in the Vienna basin in the east to fully ductile middle crust at the western end of the Tauern Window (Ratschbacher et al., 1991a,b). The differential exhumation along strike resulted from westward- increasing north-south shortening across the Eastern Alps; combined with subsidence in the easternmost Eastern Alps and Pannonian basin, the entire SEMP fault zone was tilted eastward. Fault structures transition eastward from dominantly ductile to dominantly brittle around the northeast corner of the Tauern Window, providing a natural laboratory in which to evaluate models of fault behavior around the brittle- ductile transition. 44 Figure 12. The Salzach-Ennstal-Mariazell-Puchberg (SEMP) fault zone and major structural features of the Eastern Alps (after Linzer et al., 2002 and Cole et al., 2007). Differential exhumation of the fault exposes a range of exhumation depths along strike, from ductile shear zones in the Tauern Window, to brittle-ductile outcrops near Taxenbach, and finally brittle fault zones of decreasing exhumation depth to the east. Our study sites are located near Taxenbach, exposing the eastern end of the Salzachtal fault segment. 45 3.2 Geologic Setting The Salzach–Ennstal–Mariazell–Puchberg (SEMP) fault is primarily a sinistral strike- slip fault zone that extends for 400 km across the eastern Alps (Fig. 12) (Ratschbacher et al., 1991a,b; Linzer et al., 2002). From west to east, the Salzachtal fault forms the northern boundary of the Tauern window (exposing European and oceanic (Penninic) units beneath African (Austroalpine) units), the Ennstal fault forms the boundary between the central part of the Northern Calcareous Alps and the basement of the Austroalpine unit, and the Mariazell-Puchberg fault cuts across the southern margin of the eastern Northern Calcareous Alps, part of the Austroalpine cover. Structural and kinematic analysis of the southernmost nappe boundaries of the Northern Calcareous Alps (Ratschbacher et al., 1991a,b; Decker et al., 1994; Linzer et al., 1995) demonstrates that the individual segments of the SEMP constitute a through-going sinistral fault that accommodated north-south shortening and orogen-parallel extension. Geochronology within the Tauern Window and the ages of intraorogenic basins constrain the age of this deformation as Oligocene and Miocene (e.g., Linzer et al., 2002). 3.3 Outcrop description The Salzachtal fault forms the northeastern border of the Tauern Window. From west to east, this segment of the SEMP represents the eastward transition from dominantly 46 ductile deformation in the Ahorn shear zone within the Tauern Window (Rosenberg and Schneider, 2008) and ductile/ductile-brittle deformation in the Rinderkarsee shear zone at the edge of the Tauern Window (Cole et al., 2007) to dominantly brittle deformation in the Ennstal segment (along the northern edge and to the east of the Tauern Window). Our study area is at the eastern end of the Salzachtal fault, where the fault zone is exposed in a pair of deeply incised, narrow gorges, Kitzlochklamm and Lichtensteinklamm, separated from one another by ~18 km (Fig. 13). Along this part of the SEMP, the fault juxtaposes slates and phyllites of the Grauwacken zone of the Upper Austroalpine unit to the north against graphite-rich marbles tectonites (Klammkalk) of the Lower Austroalpine Nordrahmenzone (Wang and Neubauer, 1998). Total displacement along this section of the fault has been estimated at 60 km, based on offset of the Grauwacken zone units to the east (Linzer et al. 1997, 2002). 3.3.1 Kitzlochklamm The outcrop at Kitzlochklamm provides 650 m of continuous, fault-normal exposure. The Grauwacken zone north of the fault, which is well exposed for a fault-normal distance of 200 m in the wide bed of the Rauriser River, comprises intercalated centimeter- to meter-thick intervals of graphite schist, carbonate schist, mica schist, and chlorite schist in order of decreasing importance (Fig. 14a). The Grauwacken zone is characterized by a well-developed S-C fabric adjacent to the fault (Fig. 15a,b), a feature common to metapelites in this area (Wang and Neubauer, 1998). The S- 47 Figure 13. Simplified geologic map of study areas, redrafted from “Geologische Karte von Salzburg,” mapped by the Austrian Geologische Bundesanstalt (2005). Grid markings (10x10 km) show Universal Transverse Mercator coordinates (WGS84, Zone 33). The SEMP juxtaposes Grauwacken Zone rocks to the north against Klammkalk of the Tauern Window to the south. White areas on the map indicate Quaternary units, glacial cover, and vegetated land lacking outcrop. Lichtensteinklamm (LK) and Kitzlochklamm (KK) cut through units on both the north and south of the SEMP, providing natural fault-normal transects in which to study fault structure at this exhumation level. 48 Figure 14. (a) The Grauwacken zone at Kitzlochklamm, which is composed of alternating units of carbonate schist, mica schist, chlorite schist, and graphite schist, is well exposed for a fault-normal distance of 200 m in the bed of the Rauriser river. C- surfaces range from (b) mm-scale seams to (c) centimeter-scale shear zones. (a) (a) (b) (c) 49 Figure 15. Grauwacken zone rocks at Kitzlochklamm display a well-developed S-C fabric common to metapelites in this area (Wang and Neubauer, 1998). Red line shows the average strike of the SEMP in this region, as well as the strike of the gouge zone at Lichtensteinklamm. (a) The dominant set of S-surfaces dip steeply to moderately NNE, whereas (b) the C-surfaces tend to dip steeply N or S, with a second set of planes dipping moderately NW. (c) The Klammkalk unit lacks platy minerals but still contains a number of C-surfaces, with one set dipping steeply N and S and the other set dipping NW and SE. (d) C-surfaces measured in the Klammkalk at Lichtensteinklamm are broadly similar to those found at Kitzlochklamm, with a number of shear zones paralleling the SEMP (i.e., dipping steeply N to NNW/S and dipping SSE). Equal Area Great Circle: N = 163 ; first plane = 1 ; last plane = 163 Pattern = solid Equal area lower hemisphere N=163 Kitzlochklamm Grauwacken zone S-surfaces Equal Area Great Circle: N = 127 ; first plane = 1 ; last plane = 127 Pattern = solid Equal area lower hemisphere N=127 Kitzlochklamm Grauwacken zone C-surfaces (a) (b) Equal Area Great Circle: N = 37 ; first plane = 1 ; last plane = 37 Pattern = solid (c) Equal area lower hemisphere N=37 Kitzlochklamm Klammkalk C-surfaces Equal Area Great Circle: N = 43 ; first plane = 1 ; last plane = 43 Pattern = solid Equal area lower hemisphere N=43 Liechtensteinklamm Klammkalk C-surfaces (d) 50 surfaces result from the preferential alignment of platy minerals and dip steeply to moderately north-northeast, paralleling the original schistosity formed during earlier deformation within the Tauern Window (Bickle and Hawkesworth, 1978; Kurz et al., 1996). The C-surfaces range from millimeter- to centimeter-scale shear zones (Fig 14b,c), with the dominant set dipping steeply to the north and south. A small subset of the C-surfaces dips moderately northwest (Fig. 15). Carbonate veins occur throughout the outcrop, and are commonly sheared and boudinaged (Fig. 14b). Additional exposures to the north at a fault-normal distance of ~600 m lack well-developed shear zones and contain fewer veins. This provides an upper bound on the width of fault- related deformation in the Grauwacken zone north of the fault. The Klammkalk unit, exposed south of the fault, is composed primarily of a fine- grained, graphite-rich marble tectonite with local meter-scale lenses of chlorite schist. This unit is much more resistant than the Grauwacken zone, and the river creates a narrow, deeply incised gorge in the Klammkalk (Fig. 16a). An S-C fabric is only weakly developed in the Klammkalk, which lacks the platy minerals found in the Grauwacken zone rocks (Fig. 15c). As a result, S-surfaces are weaker and less abundant than in the Grauwacken zone rocks. C-surfaces are present in two main sets, with one set dipping steeply north and south, the other dipping moderately northwest and southeast. These surfaces are found throughout the 450 m of exposed Klammkalk, albeit at a lower frequency than in the Grauwacken zone. Carbonate veins are ubiquitous, and most are sheared and boudinaged (Fig. 16b). No change in 51 Figure 16. (a) The resistant Klammkalk at Kitzlochklamm forms a narrow, sub- vertical-walled gorge with deep, fast-moving water flowing through it. (b) Carbonate veins are found throughout the Klammkalk, and are commonly sheared and boundinaged. (c) The fault contact juxtaposes Klammkalk (left) against Grauwacken zone (right) along a razor-sharp contact. 52 macroscopic deformation style can be observed throughout the 450m fault-normal width of the outcrop. With the exception of a fault-bounded, two-meter-thick layer of Klammkalk that has been slivered off 30 m into the Grauwacken zone, the two units are sharply juxtaposed at the fault contact. This fault contact, which is well exposed, is a razor-sharp zone devoid of gouge in which the Klammkalk is juxtaposed directly against graphite-schist of the Grauwacken zone (Fig. 16c). In contrast to the rest of the Grauwacken zone schists, which are mostly continuous along strike, the graphite schist within 10’s of cm of the Klammkalk is heavily damaged, consisting of phacoids of schist up to 10 cm long with polished facets embedded in an anastomozing network of shear zones. 3.3.2 Lichtensteinklamm Lichtensteinklamm exposes the Salzachtal fault in the same lithologic units as at Kitzlochklamm. High water restricts access into the Klammkalk unit to a 150-m-long, fault-normal transect. Grauwacken-zone rocks crop out in riverbed exposures for 20 m adjacent to the fault, with additional outcrops 50 m north of the fault contact. As at Kitzlochklamm, shear zones and veins are apparent throughout the Klammkalk outcrop (Fig. 17a). Outcrops of the Grauwacken zone beyond 50 m fault-perpendicular distance lack ductile shear zones, and brittle fault intensity decreases away from the 53 Figure 17. (a) The gorge at Lichtensteinklamm is much narrower than at Kitzlochklamm, and the Klammkalk is pervasively veined. (b) The Grauwacken zone within 50 m of the fault contact has been gouged into a fine-grained clay. In-situ exposures exhibit left-lateral Riedel shears, rather than the S-C fabric found at Kitzlochklamm. Pebbles and cobbles evidence in photo are surficial stream deposits unrelated to clay gouge. (a) (b) 54 Salzachtal fault until a block and matrix mélange structure is attained at a distance of ~300 m. Although exposure at Lichtensteinklamm is more limited, a major difference in structure between the two klamms is readily apparent. The Grauwacken zone in the interval 20-50 m from the fault at Lichtensteinklamm has been gouged to a fine- grained clay (Fig. 17b). In situ exposures of the clay are only locally exposed; this mud gouge has clearly been remobilized by water and flowed downslope, such that the width of the gouge zone there is a maximum. In situ exposures of gouge display a vertical foliation but do not have a well-developed S-C fabric. Rather, the gouge exhibits distinct Riedel shears, most clearly visible in the chlorite-rich layers, consistent with left-lateral slip. On the other side of the main fault contact, Klammkalk within ~20 m of the gouged Grauwacken zone contains a strong sub-vertical, fault- parallel foliation with fault-parallel veins. This suite of rocks is much more easily eroded than the rest of the Klammkalk, and foliation surfaces exposed by erosion are polished. The contact between the well-foliated Klammkalk and gouged Grauwacken zone is not exposed. The Klammkalk unit at Lichtensteinklamm is pervasively veined, although the inaccessibility of riverbed exposures at Lichtensteinklamm makes it difficult to compare deformation between the two Klamms. However, from the outcrops to which we could gain access, which were located along a walkway carved into the gorge wall, 55 we observed a weakly developed S-C fabric, as at Kitzlochklamm. We infer the weakness of this fabric is due to the lack of platy minerals within the Klammkalk. C- surfaces are broadly parallel to the Salzachtal fault, with one set dipping steeply north to northwest, and a smaller population dipping steeply north-northeast and south- southwest (Fig. 15d). 3.4 Optical microstructures The exposure of the SEMP at Kitzlochklamm allows us to directly observe how strain is partitioned just below, or within the lower part of, the brittle-ductile transition. Although the first-order observation at both the map and outcrop scales is that the majority of strain is localized along the pronounced, very sharp contact between the Klammkalk and Grauwacken zone, indicating extreme strain localization, rocks throughout Kitzlochklamm also exhibit ductile deformation fabrics consistent with the transport direction of the fault. To constrain the magnitude of this secondary, off-fault deformation, we evaluated grain-scale deformation throughout the outcrop by analyzing representative samples of the three major units present at Kitzlochklamm: Klammkalk, graphite-rich Grauwacken zone, and non-graphitic Grauwacken zone. We also compared grain-scale deformation in the Klammkalk at Kitzlochklamm and Lichtensteinklamm to see how deformation mechanisms changed below the depth of gouge formation. We did not conduct a similar analysis of the Grauwacken zone 56 rocks, as those rocks closest to the fault at Lichtensteinklamm consist of clay gouge, which is unsuitable for this type of microstructural analysis. 3.4.1 Kitzlochklamm Klammkalk The Klammkalk at Kitzlochklamm is a fine-grained marble tectonite, composed primarily of calcite grains 10-40 microns in diameter with axial ratios of 2:1, embedded in a matrix of equant calcite grains < 5 microns in diameter (Fig. 18a). Vein calcite is larger, up to 1 mm in diameter. Most samples also contain local, thin (1-2 micron) solution seams parallel to the macroscopic foliation. The coarse-grained vein calcite displays twin intensities in the range 50-60 twins/mm. Twin widths are typically ~2-4 microns, although some grains contain twin sets that are too thin to measure. The thick twins are commonly curved, tapered, and/or dynamically recrystallized (Fig. 18b), typical of type 3 and 4 twinning (Ferrill et al., 2004). Twinning is less common in grains 10-40 microns in diameter; twins in such grains tend to be thick and tabular (type 4). The ultra-fine matrix calcite appears to have been recrystallized via grain-boundary bulging. Quartz grains compose up to 20% of the samples. Grain size varies greatly, from 10 to 50 microns (Fig 18c, d). They exhibit inhomogeneous flattening, irregular undulatory 57 Figure 18. (a) Representative section of Kitzlochklamm Klammkalk, showing range of calcite grain sizes (sample 8216B1). (b) Vein calcite twinning, type 3-4, sample 8216B1, Kitzlochklamm Klammkalk. (c) Quartz grains in sample 8216B1, Kitzlochklamm Klammkalk. (d) Quartz grains in sample 8286D1, Kitzlochklamm Klammkalk. 58 extinction, lobate grain boundaries, and large subgrains, all indicative of regime 1 bulging recrystallization (Hirth and Tullis, 1992). 3.4.2 Kitzlochklamm Grauwacken graphite-schist The graphite schist is primarily composed of quartz, calcite, and feldspar layered between numerous solution seams (Fig. 19). Calcite veins are heavily twinned. Rare quartz porphyroclasts up to 100 microns in diameter exhibit sweeping undulatory extinction. However, the majority of the quartz grains in these samples are 10-50 microns in diameter, with axial ratios of up to 4:1. Of all the lithologies that comprise the Grauwacken zone at Kitzlochklamm, the graphite schist shows the most evidence for diffusion creep. Graphite, micas, and other insoluble residues are concentrated along the numerous (~20-30/mm), thick (~50 micron) solution seams parallel to the foliation. Further evidence of shortening includes vein fragments that have been truncated and rotated into the foliation. 3.4.3 Kitzlochklamm Grauwacken non-graphite schist The graphite-poor units of the Grauwacken zone, which include calcite schist, chlorite schist, and mica schist, all display similar microstructures (Fig. 20). Quartz grains vary in diameter from 10-100 microns, with irregular to sweeping undulatory extinction. 59 The most heavily recrystallized samples (from the mica schist) also exhibit large quartz subgrains (on par with the size of recrystallized quartz grains) and lobate grain boundaries (Fig. 20b, c). Calcite is mostly 10-50 microns in diameter, although vein calcite grains are up to 200 microns in diameter. Whereas twinning is generally difficult to observe in the fine-grained matrix calcite, some grains do contain type 1 and 2 twins (Ferrill et al., 2004). The coarser vein calcite contains thick (2-4 microns), bent twins, typical of type 3 twinning (Ferrill et al., 2004). Feldspar is fractured in all samples, with some twins in the mica schist. All rocks have experienced solution mass transfer to varying degrees, as shown by offset and shortened veins, solution seams, and in the case of the mica schist, mica neoblasts in quartz strain shadows (Fig. 20d). 3.4.4 Lichtenstein Klammkalk samples The Klammkalk at Lichtensteinklamm shares many microstructures with the Kitzlochklamm Klammkalk. Specifically, at Lichtensteinklamm, most of the calcite grains vary from 10-40 microns in diameter with axial ratios of 2:1. Local lenses of ultra-fine-grained calcite contain grains < 5 microns in diameter, whereas the largest vein calcite grains are up to 500 microns in diameter (Fig. 21). Twins are most apparent in the coarse-grained vein calcite. Twins are typically thinner than at Kitzlochklamm, reaching a maximum of 2 microns. The thicker twins are generally tapered and curved, but lack recrystallized grains, and are therefore typical of type 2 60 Figure 19. Representative section of sample 5812G35, Grauwacken Zone graphite-rich schist from Kitzlochklamm. Veins are shortened and rotated into parallelism with the foliation, and pressure solution seams are abundant. 61 Figure 20. (a) The Grauwacken zone mica-schist at Kitzlochklamm (sample 589G3) is composed of varying amounts of quartz, mica, feldspar, calcite, and chlorite. (b) Quartz in sample 589G3 displays lobate grain-boundaries indicative of regime 1 bulge recrystallization. (c) Quartz in sample NOFG3, also from the Grauwacken zone mica- schist at Kitzlochklamm, displays lobate grain boundaries and large subgrains. (d) Mica neoblasts in quartz strain shadow, sample 589G3. 62 Figure 21. Sample 9106A1, Lichtensteinklamm Klammkalk, shows many microstructural similarities to the Klammkalk at Kitzlochklamm. Differences are primarily seen in calcite twin morphology: samples taken from Lichtensteinklamm tend to contain thinner twins at higher twin intensities, and type 4 recrystallized twins are observed much less often. 63 and 3 twinning (Ferrill et al., 2004). Twin intensities are in the range of 100-300 twins/mm. Quartz composes up to 25% of the Lichtensteinklamm Klammkalk samples, with most quartz grains between 10-100 microns in diameter. These grains exhibit irregular to sweeping undulatory extinction; smaller grains commonly have lobate grain boundaries. Solution seams are more common and are thicker than in Kitzlochklamm. 3.5 Electron Backscatter Diffraction In addition to analyzing the optical microstructures of samples taken from each klamm, we used electron backscatter diffraction (EBSD) to provide further information on the deformation mechanisms operating in these outcrops. Low- resolution maps made of points spaced at about the grain scale were collected and indexed using CHANNEL 5 HKL software, yielding lattice-preferred orientations (LPO) of specific mineral phases in each slide. These data allow us to determine the relative importance of dislocation creep versus diffusion creep in the various units of the Klamms. 64 3.5.1 Kitzlochklamm Klammkalk We analyzed five samples spanning a 450 m fault-perpendicular distance within the Klammkalk. The goal of these analyses was to determine whether significant strain had been distributed within the wall rocks on the southern side of the Salzachtal fault. Previous studies of ductile calcite from similar metamorphic conditions in the Morcles nappe have documented a marked increase in LPO strength within meters of the fault (Ebert et al., 2007; Austin et al., 2008), reflecting the width over which strain was distributed within that ductile shear zone. All Klammkalk samples along our transect (Fig. 22) yielded the ‘D3’ fabric typical of shear zones bordering the Tauern Window (Kurz et al., 2000). Calcite <c> axes [001] are clustered near Z, whereas <a> axes [110] are distributed along a girdle within the X-Y plane, representing fault-normal shortening (e.g., Wang and Neubauer, 1998). This kind of fabric is known as a “deformation LPO,” representing deformation at low temperatures prior to recrystallization (Behrmann, 1983; Dietrich and Song, 1984; Schmid, 1987; Ratschbacher et al., 1991b; Pieri et al., 2001a,b; Barnhoorn et al., 2004). The strength of this fabric, represented by both the texture index J (calculated with PFch5) and MUD (calculated in CHANNEL 5 HKL), is nearly constant throughout our 450-m-long transect (Mainprice, 2003). Most samples exhibited a MUD between 3 and 5, with a texture index J of ~1.5, which is relatively weak for rocks within a ductile shear zone associated with ~ 60km of displacement. 65 Figure 22. Lattice preferred orientations (LPOs) of calcite grains (point per grain) in samples of Kitzlochklamm Klammkalk taken along a fault-normal transect. All samples show the D3 fabric typical of shear zones bordering the Tauern Window (Kurz et al., 2000). The strength of this fabric, represented by both J and MUD, is nearly constant along our 450-m-long transect. 1 2 3 4 (a) Sample 8216B1, 0m from fault. n=7695 J=1.57 4.92 <0001> Z X Y <1014> <1012> <1120> <0001> Z X Y <1014> <1012> <1120> <0001> Z X Y <1014> <1012> <1120> <0001> Z X Y <1014> <1012> <1120> <0001> Z X Y <1014> <1012> <1120> 0.02 1 2 3 3.85 0.06 (b) Sample 9197A5, 60m from fault. n=5067 (c) Sample 8286D1, 80m from fault. n=3267 1 2 3 3.23 0.15 1 2 3 5.34 (e) Sample 9197A1, 440m from fault. n=5159 (d) Sample 8306E1, 340m from fault. n=3199 0.03 1 2 3 4 5 0.14 3.38 J=1.52 J=1.30 J=1.49 J=1.58 66 3.5.2 Kitzlochklamm Grauwacken zone Although solution transfer creep was clearly the dominant deformation mechanism operating in the graphite-rich schist of the Grauwacken zone, as shown by the extensive concentration of solution seams at all scales, the relative importance of dislocation and diffusion creep in the other Grauwacken units is difficult to determine from optical microstructures. To address this question, we analyzed representative samples of the mica schist and carbonate schist. Well-developed crystallographic fabrics would indicate dislocation creep, whereas weak to random fabrics would point to diffusion creep as the dominant deformation mechanism (Fliervoet et al., 1997). In both the mica schist (sample NOFG3) and carbonate schist (sample G19b), we analyzed quartz and calcite grains. Feldspar clearly deformed via microcracking, and the ease with which muscovite develops a preferred orientation makes it a poor indicator of the strength of dislocation creep. Lattice preferred orientations of quartz grains in the Kitzlochklamm Grauwacken zone mica schist (Fig. 23) produce a weak girdle of <c> axes in the X-Z plane, with <a> axes producing a few faint poles in the X-Y plane. However, the strength of this fabric (MUD = 1.86) is weak to random. Calcite grains from the same unit produce a random LPO (MUD = 1.68). In the Kitzlochklamm Grauwacken zone carbonate schist (Fig. 24), the quartz grains produce a slightly stronger fabric (MUD = 2.28), with a better- 67 Figure 23. (a) LPOs of quartz grains from the Grauwacken zone mica schist (sample NOFG3) collected via point-per-grain analysis. We see a girdle of <c> axes in the X-Z plane with a few <a> axes in the X-Y plane, possibly indicating basal slip. The strength of this LPO, however, is weak to random, suggesting only a small strain magnitude was accommodated by dislocation creep in the mica schist quartz grains. (b) Random LPOs in calcite. (a) Mica schist quartz grains (upper hemisphere) (b) Mica schist calcite grains (upper hemisphere) n=29644 0.37 1.86 1 <0001> <1120> Z X Y Z Y X <0001> <1014> <1012> <1120> 1 0.48 1.68 n=4661 68 Figure 24. (a) LPOs of quartz grains from the Grauwacken zone carbonate schist form a girdle of <c> axes in the X-Z plane with weakly developed <a> axis poles, possibly indicating a weak component of basal slip. (b) Calcite grains exhibit a random LPO. (a) Carbonate schist quartz grains (upper hemisphere) (b) Carbonate schist calcite grains (upper hemisphere) n=674 0.03 2.28 1 <0001> <1120> Z X Y Z Y X <0001> <1014> <1012> <1120> 1 0.30 1.80 n=2644 2 69 developed girdle of <c> axes in the X-Z plane. The <a> axes are more weakly developed, with at least one pole in the X-Y plane. Calcite grains from the carbonate schist produce a random LPO (MUD = 1.8), as seen in the mica schist. Collectively, the weak to random LPOs observed in the mica and carbonate schist therefore provide further evidence that the Grauwacken zone at Kitzlochklamm deformed primarily via solution transfer creep, accommodating fault-normal shortening. 3.5.3 Lichtensteinklamm Klammkalk Given the relative consistency of the fabrics at Kitzlochklamm, we analyzed only two samples of the Klammkalk at Lichtensteinklamm: one at a distance of 20 m to the fault, and the other at 50 m (Fig. 25). As with the analysis for Kitzlochklamm, these were “coarse” one-point-per-grain analyses designed to determine bulk calcite LPO. Both samples lack the D3 fabric that was observed in all Kitzlochklamm samples. In addition, the fabric strengths of MUD = 2.18 at 20m and 1.51 at 50m are weak to random. 70 Figure 25. Samples of Klammkalk taken at distances of 20 and 50m from the SEMP at Lichtensteinklamm lack the “deformation LPO” we observe at Kitzlochklamm, and that Kurz et al. (2000) documented in ductile shear zones bordering the Tauern Window. 1 2 (a) Sample 9106A1, 20m from fault. n=8953 2.18 <0001> Z X Y <1014> <1012> <1120> <0001> Z X Y <1014> <1012> <1120> 0.17 1 1.51 0.55 (b) Sample 9106A13, 50m from fault. n=13054 71 3.6 Exhumation Depth The Penninic units along the margins of the Tauern Window show evidence for greenschist-facies metamorphism (Genser et al., 1996), and our observations of ductile carbonate veins and shear bands in the Klamms, as well as ductile quartz and brittle feldspars, are consistent with this. The exhumation depth of the SEMP in this region has not been quantified, as the marbles and schists at these outcrops are unsuitable for thermobarometry. However, the large amount of carbonaceous material (CM) in these rocks allowed us to constrain the peak metamorphic temperature using Raman spectroscopy (Beyssac et al., 2002a). We conducted laser Raman measurements of CM in four polished thin sections, one from each side of the fault at each Klamm. Approximately 10 spot analyses were conducted for each thin section, depending on the prevalence of CM. The peak positions, band areas, and band widths of the D1, D2, and G bands for each analysis were determined with the program PeakFit 4.12 (Systat Software Inc.). These band areas and widths were then used to calculate peak metamorphic temperature for each spot analysis according to Beyssac et al. (2002a). T(°C) = -445 R2 + 641 Where R2 represents the relative areas of the D1, D2, and G bands: 72 R2 = D1 / (G + D1 + D2) The samples from Kitzlochklamm yielded average temperatures of 380 ± 12˚C (1s measurement precision) in the Grauwacken zone and 406 ± 13˚C in the Klammkalk. Samples from Lichtensteinklamm yielded temperatures of 435 ± 15˚C in the Grauwacken zone and 406 ± 15˚C in the Klammkalk. Beyssac et al. (2002a) did not present a statistical treatment of the errors in this method; simply noting that the maximum error observed during calibration was 50˚C. Their work suggested that the limiting factor in the precision of this measurement is within-sample heterogeneity, reflected in the standard deviation. Based on these calculations, the rocks appear to have experienced peak temperatures around 400˚C. These data, combined with our microstructural observations of dislocation creep in quartz and fully-brittle deformation in feldspar, suggest the outcrops of the SEMP at Kitzlochklamm and Lichtensteinklamm would most likely fall near the middle of the brittle-ductile transition of Scholz et al. (2002), between 300-400C. 73 3.7 Discussion The observations from Lichtensteinklamm and Kitzlochklamm allow us to characterize the structural architecture of the brittle-ductile transition of the exhumed Salzachtal fault. Most basically, these data reveal highly localized deformation along a narrow fault interface throughout the brittle-ductile transition, with relatively minor distributed strain in the wall rocks. Specifically, the presence of gouge several tens of meters thick that preserves evidence of left-lateral shear at Lichtensteinklamm requires that much of the displacement there was accommodated via cataclasis, with minor amounts of distributed shear present in the 20-50 m of Klammkalk adjacent to the gouge. Conversely, evidence for ductile deformation (sheared and boudinaged veins, S-C fabrics) is present in the wall rocks of both Klamms. At Kitzlochklamm, both the Klammkalk marble tectonites and Grauwacken zone schists exhibit uniformly ductile deformation, with sub-vertical shear zones consistent with the orientation of the Salzachtal fault throughout the Grauwacken zone. Although the graphite schist within centimeters of the fault contact is heavily damaged, it is distinct from a typical fault gouge in that large phacoids of schist are preserved intact. The age of formation of this meter-thick damage zone is not clear; it could have formed within the upper part of the BDT, or at shallower depths during exhumation. Taken together, these observations suggest that Lichtensteinklamm has been exhumed from the upper part of the brittle- ductile transition, whereas Kitzlochklamm appears to have been exhumed from slightly deeper structural levels nearer the base of the brittle-ductile transition zone. 74 These inferences are supported by microstructural analysis of samples from each klamm. Twin morphologies from the Klammkalk are consistently indicative of higher temperature deformation at Kitzlochklamm than at Lichtensteinklamm. Additionally, calcite from the Kitzlochklamm Klammkalk shows evidence for recrystallization via grain-boundary bulging, which is not observed in Lichtensteinklamm. Comparison of quartz in the same Klammkalk samples displays a similar pattern, with regime 1 bulging recrystallization evident at Kitzlochklamm but not at Lichtensteinklamm. Furthermore, calcite LPOs from Kitzlochklamm imply crystal plastic deformation throughout the Klammkalk, albeit at low strains, whereas the same unit at Lichtensteinklamm shows no evidence for the D3 fabric that has been documented at shear zones bordering the Tauern Window (Kurz et al., 2000). Together, these microstructural data suggest increasing dislocation and diffusion creep at the expense of cataclasis at the more deeply exhumed Kitzlochklamm outcrop. These macroscopic and microscopic data indicate that the bulk of the 60 km of displacement the SEMP experienced was accommodated in a narrow zone at the fault contact (tens of meters wide at Lichtensteinklamm, less than a meter at Kitzlochklamm). This conclusion is further supported by examination of gross outcrop patterns used to infer offset along the SEMP. Several studies have used the offset of the contact between Grauwacken and Phyllite zone to define the offset along the Salzachtal and Ennstal segments of the SEMP, after correcting for extension and 75 contractions along strike-slip duplexes (Ratschbacher and Frisch, 1993; Linzer et al., 1997, 2002). Note, however, that the tie points used to constrain the total displacement along the SEMP are not mapped all the way to the fault. Where the Grauwacken/Phyllite zone approaches the fault from the south, the fault lies within a 2-km-wide river valley filled with recent sediments. This permits a zone of distributed deformation to extend as much as 1 kilometer either side of the fault. However, 6 km to the west, bedrock outcrops extend along both sides of a major bend in the river. If the fault is relatively linear along this reach, as it is elsewhere, the geometry of bedrock outcrops requires that the maximum possible width of any distributed faulting is no more 100-200 m. Our interpretation of data from Lichtensteinklamm supports the notion that fault strength is primarily governed by frictional strength even above temperatures of 300˚C, as suggested in the fault-‐zone model of Scholz (2002) (Fig. 26). At the more deeply exhumed Kitzlochklamm outcrop, where the Salzachtal fault is defined by a sharp lithologic contact lacking gouged material, our data are consistent with slip accommodated by localized aseismic creep, encouraged by increases in the loading velocity during seismic ruptures from above. This is in agreement with geodetic data from large earthquakes (Burgmann et al., 1997; Savage and Svarc, 1997; Reilinger et al., 2000; Bechor et al., 2001), and is again broadly in agreement with the fault-‐zone model of Scholz (2002). However, this 76 30 20 10 Oligo-Miocene land surface varying horizontal scale present land surface 0 paleodepth (km) Mayrhofen Vienna Krimml study area Gstatterboden ductile Ahorn shear zone Rinderkarsee shear zone (ductile/ductile-brittle) brittle-ductile SEMP Taxenbach brittle SEMP fault ? ? ? onset of interseismic quartz plasticity onset of interseismic feldspar plasticity fully plastic mylonite cataclasite + veins cataclasite + veins; mylonite w/plastic quartz + brittle feldspar 300°C 450°C (a) (b) Figure 26. (a) Schematic reconstruction of the SEMP fault system during Oligo-Miocene time (after Cole et al., 2007) compared against (b) a generalized shear zone model (after Scholz, 1988). At our study sites near the brittle-ductile transition at Taxenbach, we observe the SEMP to be highly localized at outcrops that are both dominantly brittle (Lichtensteinklamm) and dominantly ductile (Kitzlochklamm). Our study at the more shallowly exhumed Gstatterboden outcrop found the majority of strain to be localized within a 10 to 40-m-wide brittle core (Frost et al. 2009), while Cole et al. found the more deeply exhumed Rinderkarsee shear zone had concentrated SEMP-related strain within a 100-m-wide zone ductile/ductile-brittle shear zone (Cole et al., 2007). The SEMP reaches its greatest width in the Ahorn shear zone, where strain has been distributed within a 2-km- wide mylonite belt (Rosenberg and Schneider, 2008). Our results at the SEMP suggest that faults may not necessarily widen significantly concurrent with quartz ductility, as suggested by the model of Scholz et al. (2002). 77 is purely conjecture, as we cannot present any evidence of seismogenesis in these outcrops. On the basis of the weak fabrics observed in EBSD, we can conclude that only minor strain was distributed off the fault, accommodated via dislocation creep occurring against a background of fault-perpendicular shortening via solution creep, although we were unable to quantify the latter deformation mechanism. 3.8 Conclusions Structural analysis of two fault-perpendicular transects along the exhumed Oligo- Miocene Salzachtal fault reveal highly localized strain throughout the brittle-ductile transition. Although there is minor distributed ductile deformation in the wall rocks at both study sites in the Klammkalk to the south of the fault, these fabrics are weak at Kitzlochklamm and absent at more than a few decameters from the fault at Lichtensteinklamm. The Grauwacken zone rocks to the north of the fault preserve evidence for a wider zone of deformation, up to several hundred meters, but the ductile component of deformation in these rocks was dominated by solution creep that accommodated fault-normal shortening, rather than sinistral shear. Weak fabrics observed in EBSD indicate little off-fault strain was accommodated by dislocation creep in these rocks. These results demonstrate that as deformation along the SEMP transitioned downward from dominantly brittle to dominantly ductile, strain remained highly localized, rather than broadening into a wide zone of distributed deformation. CHAPTER 4: A Comparison of Fault Zone Structures Throughout the Crust 78 Chapter 4 Abstract We summarize and synthesize studies of the Salzach-Ennstal-Mariazell-Puchberg (SEMP) strike-slip fault at outcrops exhumed from 4 to ~20 km depth. Taken together, these results constitute the first set of direct observations evaluating how a complex, natural fault operates throughout the crust. We compare our findings to classic examples of individual faults exhumed from comparable depth levels in order to evaluate the importance of the unique history of each fault, such as cumulative slip, host rock rheology, deformation mechanism, and geometry. We find that host rock rheology plays a vital role in shaping the structure of fault zones at both brittle and ductile deformation conditions. Specifically, the brittle portions of faults may only form localized, high-‐strain cores in host rocks that strain-‐weaken, whereas strain-‐hardening host rocks favor a wider fault zone with multiple cores. At depths where deformation is dominantly brittle-‐ductile to ductile, transformational plasticity can produce localized, high strain zones in what might otherwise be a zone of diffuse, ductile deformation. However, this process depends upon cataclasis and the presence of fluids. 79 4.1 Introduction The mechanical processes governing fault zone behavior and evolution are hotly debated. Understanding the significance of fault zone width, the formation of fractured wall rocks, the mechanical significance of gouge is vital if we are to properly model earthquake nucleation and propagation, especially at the base of the seismogenic zone where many large event nucleate. The most direct means of answering these questions is to study exhumed fault zones, and such studies have so far provided a wealth of information about the internal structure of fault zones (e.g., Chester et al., 1993; Chester and Logan, 1986; Evans et al., 2000; Imber et al., 2001; Knipe, 1989; Sibson, 1983). However, analysis and synthesis of these results has been complicated by the fact that each study deals with a fault exhumed from a unique depth, with varying host rock lithologies, covering a range of structural maturities. We have sought to eliminate such complexity by studying a major strike-slip fault (the Salzach-Ennstal-Mariazell-Puchberg fault, or SEMP) that has been differentially exhumed, revealing outcrops ranging from near-surface conditions to exhumation depths of ~20 km. The SEMP affords us the unique opportunity to directly observe a fault in three dimensions, allowing us to constrain what variables might contribute to changes in fault zone structure. 80 4.2 Geologic Setting The Salzach-Ennstal-Mariazell-Puchberg (SEMP) fault is a dominantly sinistral strike- slip fault that extends for 400 km across the eastern Alps, from the Tauern window in the west to the Vienna Basin in the east (Linzer et al., 2002; Ratschbacher et al., 1991a; Ratschbacher et al., 1991b). The Salzach fault forms the northern boundary of the Tauern window (exposing Penninic units), the Ennstal fault forms the boundary between the central part of the Northern Calcareous Alps and the basement of the Austroalpine unit, and the Mariazell-Puchberg fault cuts across the southern margin of the eastern part of the Northern Calcareous Alps. North of the SEMP fault zone lies the thin-skinned thrust belt overlying the European craton; this thrust belt is composed of molasse, Helvetic nappes (European sedimentary cover), Penninic ophiolites and flysch, and Austroalpine sedimentary cover nappes; the latter include the weakly metamorphosed Innsbruck quartz phyllite and Grauwacken zone, and the unmetamorphosed carbonates and evaporites of the Northern Calcareous Alps (e.g. Linzer et al., 2002). South of the SEMP fault zone are the same Austroalpine cover units with their crystalline basement, and the Tauern window. The Tauern window consists of oceanic rocks (Upper Schieferhülle) overlying thinned European continental crust (Lower Schieferhülle and Zentralgneis) (e.g. Linzer et al., 2002). Structural and kinematic studies, aimed toward understanding the geodynamic process 81 of lateral extrusion (see summaries in Frisch et al., 1998; Linzer et al., 2002; Ratschbacher et al., 1991a; Ratschbacher et al., 1991b), showed that the SEMP fault zone forms the northern boundary of an extruded wedge along which the central Eastern Alps were transferred eastward towards the intra-Carpathian domain. In this model, the extrusion was driven by crustal thickening and resultant E-directed gravitational spreading in the western Eastern Alps (Tauern Window and Ötztal unit), and coeval subduction slab retreat (normal to the strike of the Eastern Alps) and subsidence in the easternmost Eastern Alps and Pannonian basin (Ratschbacher et al., 1995; Ratschbacher et al., 1991a; Ratschbacher et al., 1991b). The extruding wedge is bounded by strike-slip faults along its northern (SEMP) and southern (Periadriatic Lineament) boundaries and by an extensional detachment at its base (now exposed along the eastern and western margins of the TW (the Brenner and Katschberg detachment zones) and along the Penninic windows along the eastern edge of the Eastern Alps (Rechnitz Window units, Ratschbacher et al., 1990). Overwhelming evidence for the interpretation that the bounding faults acted as strike-slip faults comes from numerous detailed structural studies (see summary papers above), demonstrating subhorizontal slip along the main strand of the SEMP fault zone and its subsidiary fault zones. Crustal thickening and exhumation of the western Eastern Alps was principally achieved by crustal-scale folding (e.g., the Tauern brachi-anticline); the northern termination of this brachi-anticline coincides with the SEMP fault zone. During the late stages of extrusion, both the detachment and the SEMP fault progressively cut to deeper levels, possibly rooting in a detachment at the top of the 82 subducting European continental wedge imaged in seismic profiles (preliminary data of the TRANSALP project). The differential exhumation of the SEMP fault zone along strike resulted from the larger shortening in the western and central Eastern Alps than in the eastern Eastern Alps, creating extrusion-related structures (both strike-slip and contractional kinematics) north of the SEMP fault zone (e.g. Inntal fault zone) and an overall higher amount of isostatically driven exhumation in the western Eastern Alps. The timing of this deformation is constrained to be Oligocene and Miocene by radiochronology (chiefly within the Tauern window) and the ages of intraorogenic basins along and near the SEMP fault (Linzer et al., 2002). The differential exhumation centered around the central Eastern Alps (Ötztal unit and Tauern window) and subsidence along the easternmost Eastern Alps and western Pannonian basin tilted the entire SEMP fault zone eastward, resulting in a range of exposure depths, from ~15—20 km in the Tauern window to surface levels in the Vienna Basin (see structural analysis and analogue material modeling in Ratschbacher et al., 1991a; Ratschbacher et al., 1991b). Consequently, in a broad sense, deformation structures change eastward from dominantly ductile along the Salzachtal fault zone within and adjacent to the Tauern window to brittle-ductile along the Ennstal fault to brittle deformation with multiple splays along the Mariazell-Puchberg fault. 83 4.3.1 Results from the brittle crust We characterized the structure of the SEMP in the brittle, seismogenic crust where the Ennstal segment is exposed in Gesause National Park (refer to Chapter 2 for supporting figures). Exhumation depth in this area is 4-8 km based on stratigraphic relationships (Geologische Bundesanstalt, 1980; Linzer et al., 1995). The Ennstal fault segment comprises two strands in this area; our study site is along the northern (main) strand that juxtaposes Dachsteinkalk limestone to the south against Rauwacken dolomite to the north. Fault-normal exposure runs for ~ 200 m in the dolomite, however, colluvium and vegetation limit exposure to 50 m in the limestone. Unfortunately, the fault surface at the limestone/dolomite contact is buried beneath a 3-5 m wide section of thick colluvium. Notwithstanding the lack of exposure at the fault contact, this study site still provides fundamental insight into the evolution of this major strike-slip fault zone. Our study of the Gesause outcrop combined measurement of faults and fractures throughout the outcrop along with calculations of damage at the macro- and microscales. We quantified the macroscopic degree of brittle deformation along our fault-normal transect using damage intensity (Chester and Logan, 1986; Schulz and Evans, 2000), while microscopic damage was characterized by calculating the grain- size distribution of samples taken along the same transect. 84 Damage at the Gesause outcrop is extremely asymmetric. South of the fault, the limestone is cut by subsidiary faults and fractures, including meter-scale Riedel shears, but is otherwise largely intact without brecciation. In contrast, the dolomite to the north has been pervasively reduced to millimeter- to centimeter-scale fragments throughout the length of available outcrop. The dolomite is also cut by numerous faults and through-going fractures, ranging from cm-scale fractures with no displacement, to multi-meter-long faults with gouge zones up to several centimeters thick. Calculations of macroscopic damage intensity in the dolomite only show a 25% decrease in damage at a distance of 150 m from the fault contact, giving little evidence for a zoned structure at this exhumation depth of the SEMP. However, our outcrop- scale measurements in the dolomite show that small faults in the 20-40 m closest to the fault contact are almost exclusively fault parallel, with dips noticeably steeper than in any other section of our study transect. We interpreted this pattern to indicate the presence of a high-strain fault core within the 20 m closest to the master fault, possibly extending out as far as ~40 m. This is consistent with observations from other exhumed fault cores, which tend to exhibit fault-parallel foliations and shear structures (e.g., Chester et al., 1993; Schulz and Evans, 2000). The evolutionary process that led to this zoned fault structure is brought out in the microscopic data. Samples of dolomite closest to the fault (within the 10 m of 85 dolomite adjacent to the covered interval) yield mass dimensions of D 2 ≈ 2.0. At a distance of 109 m from the fault, the mass dimension of the dolomite has fallen to a value of D 2 ≈ 1.6. Between these samples, however, the distribution of grain sizes is best characterized by two power-law relationships. For samples taken 10 m to 64 m from the fault, the mass dimension at 25x magnification is ~2.0, whereas the mass dimension at 200x magnification is always closer to 1.6. The largest grains, in effect, share the characteristics of a high-strain gouge typically found near shear localizations, whereas the smallest grains more closely resemble what is typically found in a fault’s gouge and breccia zone. We proposed that samples taken 10-64 m from the fault contact revealed a “transition zone” recording progressive localization of strain within the dolomite. The development of a gouge with a mass dimension of D2 ≈ 1.6 has been demonstrated by experiments to coincide with the transition from velocity-strengthening to velocity- weakening, which favors strain localization (Biegel et al., 1989). In contrast, a mass dimension of D2 ≈ 2.0 is thought to be the product of concentrated flow shear once strain has localized (Sammis and King, 2007). This second phase of fragmentation reworks the initial, geometrically stable arrangement of grains, placing once-isolated grains back into contact with each other, causing one to fracture. Fragmentation starts at the largest grain size because these grains have the highest probability of coming into contact with each other in the low-strain gouge. 86 We therefore viewed the transition zone as the width of rock through which strain localized on its way towards forming the eventual high-strain fault core. These data highlight the fact that evidence for a broad zone of deformation across a fault may only be taken to represent the finite strain that has accumulated over the lifetime of the fault, and that the most recent increment of infinitesimal strain may be confined to a much more discrete zone. 4.3.2 Results from the brittle-ductile transition The Salzachtal segment of the SEMP forms the northern border of the Tauern Window, and marks the transition from dominantly ductile fault structures within the Tauern Window to dominantly brittle structures to the east (Cole et al., 2007; Frost et al., 2009; Ratschbacher et al., 1991a,b; Rosenberg and Schneider, 2008). We studied the Salzachtal segment where it is exposed in a pair of deeply incised, narrow gorges (Kitzlochklamm and Lichtensteinklamm) along the northeast edge of the Tauern Window (refer to Chapter 3 for supporting figures). Along this part of the SEMP, the fault juxtaposes slates and phyllites of the Grauwacken zone of the Upper Austroalpine unit to the north against graphite-rich marble tectonites (Klammkalk) of the Lower Austroalpine Nordrahmenzone (Wang and Neubauer, 1998). Peak metamorphic temperatures obtained via Raman spectroscopy are ~ 400C, further suggesting that these outcrops of the SEMP have been exhumed from the mid- to upper-BDT. 87 Outcrop at Kitzlochklamm extends ~200 m across the Grauwacken zone and ~450 m across the Klammkalk. The Grauwacken zone is characterized by a well-developed S- C fabric adjacent to the fault. S-surfaces result from the preferential alignment of platy minerals and dip steeply to moderately N-NE, paralleling the original schistosity formed during earlier deformation within the Tauern Window (Bickle and Hawkesworth, 1978; Kurz et al., 1996). The C-surfaces range from millimeter- to centimeter-scale seams, with the dominant set dipping steeply to the north and south. An S-C fabric is only weakly developed in the Klammkalk, which lacks the platy minerals in the Grauwacken zone rocks. As a result, S-surfaces are weak, however C- surfaces are still present in two main sets, with one set dipping steeply north and south, the other dipping moderately northwest and southeast. Carbonate veins are ubiquitous throughout both units, and most are sheared and boudinaged. The Grauwacken zone/Klammkalk contact is razor sharp, devoid of gouge, and characterized by phacoids of graphite schist embedded in an anastomozing network of shear zones. Lichtensteinklamm exposes the Salzachtal fault in the same units ~15 km farther east. Exposure in this outcrop extends for ~150m in the Klammkalk and 20m into the Grauwacken zone, with additional outcrop available at a distance of 50m from the fault. Although exposure at Lichtensteinklamm is more limited, a major difference between the two Klamms is readily apparent. The Grauwacken zone within ~50 m of 88 the fault at Lichtensteinklamm has been gouged to a fine-grained clay, although in some areas this mud gouge has clearly been remobilized by water and flowed downslope, such that the width of the gouge zone there is a maximum. In situ exposures of gouge do not display the well-developed S-C fabric observed at Kitzlochklamm. Rather, the gouge exhibits distinct Riedel shears, most clearly visible in the chlorite-rich layers, consistent with left-lateral slip. Furthermore, Klammkalk within ~20 m of the gouged Grauwacken zone contains a strong sub-vertical, fault- parallel foliation with fault-parallel veins. This suite of rocks is much more easily eroded than the rest of the Klammkalk, and surfaces exposed by erosion are polished. The contact between the well-foliated Klammkalk and gouged Grauwacken zone is not exposed. The Klammkalk unit at both outcrops is primarily composed of 10-40 micron calcite grains with axial ratios of ~2:1, embedded in a matrix of fine (<5 micron) equant calcite. Coarse grained (1mm) vein calcite contains numerous type 3-4 twins. Matrix calcite at Kitzlochklamm only infrequently displays type 4 twins, whereas matrix calcite from Lichtensteinklamm generally displays thinner type 2-3 twins. The ultrafine calcite at Kitzlochklamm appears to have recrystallized via grain boundary bulging. Quartz composes up to 20% of samples from the Kitzlochklamm Klammkalk and ranges from 10-50 microns. These quartz grains display inhomogeneous flattening, 89 irregular undulatory extinction, lobate grain boundaries, and large subgrains, all typical of regime 1 deformation (Hirth and Tullis, 1992), whereas quartz from Lichtensteinklamm only shows evidence for irregular to sweeping undulatory extinction and lobate grain boundaries. Solution seams are more common and thicker in samples from Lichtensteinklamm, oriented parallel to the macroscopic foliation. Graphite schist in the Grauwacken zone is composed of quartz, calcite and feldspar layered between numerous pressure solution seams. Calcite veins are heavily twinned. Rare quartz porphyroclasts up to 100 microns in diameter exhibit sweeping undulatory extinction. However, the majority of the quartz grains in these samples are 10-50 microns in diameter, with axial ratios of up to 4:1. The graphite schist shows extensive evidence of solutions mass transfer, containing numerous (~20-30/mm), thick (~50 micron) solutions seams parallel to the foliation. Non-graphite schists from the Grauwacken zone (calcite schist, mica schist, and chlorite schist) all display similar microstructures. Quartz grains vary in diameter from 10-100 microns, with irregular to sweeping undulatory extinction. The mica schist shows the most evidence for recrystallization, exhibiting large quartz subgrains and lobate grain boundaries. Matrix calcite are between 10 and 50 microns in diameter, locally containing type 1 and 2 twins (Ferrill et al., 2004), whereas vein calcite can be up to 200 microns in diameter with thick (2-4 micron), bent, type 3 twins (Ferrill et al., 2004). Feldspar is fractured in all samples, and all samples show varying intensities of 90 solution mass transfer in the form of offset/shortened veins, solutions seams, and mica neoblasts in quartz strain shadows. Election backscatter diffraction (EBSD) data from the Klammkalk at both outcrops shows more strongly developed calcite LPOs at Kitzlochklamm, suggesting a greater intensity of dislocation creep at this outcrop. However, the strength of these pole figures is still weak in an absolute sense (M.U.D. varying between 3 and 5; J values cluster around 1.5), indicating that only a minor portion of SEMP displacement has been accommodated via ductile creep of the Klammkalk unit. EBSD data from the Grauwacken zone at Kitzlochklamm shows weak to random LPOs for quartz and calcite respectively, suggesting that the extensive evidence at both the outcrop- and grain-scale for solution mass transfer reflects the dominant deformation mechanism. 4.3.3 Results from the ductile/brittle-ductile crust As the SEMP enters the NE corner of the Tauern Window, the fault is exposed in a shear zone near Rinderkarsee, south of Krimml (Fig. 27, all figures from Cole et al., 2007). Deformation is ductile to brittle-ductile at this exhumation depth (~16-17 km), with strain partitioned into three main high-strain shear zones separated by low-strain panels. The southernmost shear zone has the typical “ductile” appearance of a mylonite, whereas the central and northern shear zones both feature brittle-ductile 91 Figure 27. (a) The SEMP fault zone (after Linzer et al., 2002) in the framework of the Tertiary tectonic structures of the central and eastern Eastern Alps and the location and structural and kinematic characterization of two (Greiner and Ahrntal) of the three major ductile shear zones of the interior of the central and western Tauern Window. The Rinderkarsee shear zone that connects to the SEMP is southeast of Krimml (geologic contacts after Bigi et al., 1990). Structural data are plotted in lower hemisphere, equal-area stereograms, s 1 , first foliation; str 1 , first stretching lineations; e 3 to e 1 , principal strain directions. 92 fabrics at the grain scale, with ductilely stretched quartz grains and brittley deformed feldspars. Shear zone width and magnitude indicate that the southernmost of these shear zones is the main strand of the Rinderkarsee shear zone. It straddles the contact between relatively homogeneous granitic Zentralgneis (Fig. 28a) and the much more diverse amphibolite-facies metasedimentary host rocks, the Habach Group. The Zentralgneis is phyllonitic over a 3-m-wide zone, whereas the less competent rocks of the Habach group are mylonitic over a zone ≥ 100 m wide (Fig 28d). The southern shear zone is composed of meter-scale NE striking subvertical shear zones. Strain, estimated from the relative amount of grain size reduction, is greatest in this southern shear zone, likely because of the competency contrast across the contact. The central shear zone is ~600 m to the north and hosted fully within the Zentralgneis. This shear zone is ~ 50 m wide and is composed of at least three generations of structures. The earliest structures are en echelon arrays of decimeter-long, cm- to mm- wide E-NE striking subvertical sinistral and dextral shear zones filled with muscovite, biotite, and chlorite. Quartz was ductile in these shear zones, whereas feldspar was brittle. These structures are truncated by m-scale brittle-ductile shear zones forming conjugate sinistral and dextral sets (Fig. 28c). The youngest structures are sheared quartz veins, tension-gash arrays, and minor brittle faults (Fig. 28a). 93 Figure 28. Macroscopic deformation within the Rinderkarsee shear zone. (a) Host Zentralgneis with minimal background ductile strain cut by brittle chlorite-filled tension gash array. (b) Moderate-strain zone in gneiss showing ductile quartz and brittle-ductile feldspar. (c) Brittle-ductile incipient shear zones. (d) Wholly ductile high-strain zone. 94 We used optical microscopy combined with electron backscatter diffraction to evaluate the grain-scale deformation mechanisms within the Rinderkarsee shear zone. We chose three primary microstructural types for analysis: (1) an early high-strain ductile mylonite from the southern shear zone (Fig 29a), (2) a moderate-strain brittle- ductile shear zone from the central shear zone (Fig. 29b), and (3) a late stage(?) tip of a mm-scale brittle-ductile shear zone from the central shear zone (Fig. 29c, d). Mylonitic quartzofeldspathic rocks from the southern shear zone are predominantly composed of subequal amounts of plagioclase, quartz, and biotite. Rare plagioclase porphyroclasts up to 1 mm in diameter have patchy UE, subgrains, deformation twins, core-and-mantle recrystallization microstructures, and well-developed strain shadows filled with recrystallized quartz; the bulk of the plagioclase are 20-50 microns, have axial ratios of ~ 2:1, sweeping UE, and straight to curved grain boundaries. Where not pinned by other phase, quartz reaches 200 microns and has sweeping UE, well-defined subgrains, and dentate grain boundaries; most of the quartz grains are pinned < 50 microns. EBSD analysis of the large (up to 1 mm) plagioclase grains show 5-8 degrees of misorientation from core to rim, whereas the more abundant 50-100 micron matrix plagioclase show 3-6 degrees of misorientation from core to rim (Fig. 30). The matrix feldspar grains have weak LPOs (MUD ≤ 2), with [001] subparallel to the X strain axis and poles to (010) subparallel to the Z strain axis. All quartz grains show a random distribution of grain orientations (MUD < 2). 95 Figure 29. Rinderkarsee shear zone microstructures (a) Mylonitic amphibole from southern shear zone showing crystal-plastic deformation of amphibole and feldspar. (b) Brittle-ductile fabric in southern shear zone Zentralgneis with brittle feldspar and ductilely deformed quartz. Feldspar fractures are filled with recrystallized quartz, muscovite, and calcite. (c) Alteration of K-feldspar to muscovite. (d) A more mature, millimeter-scale shear zone formed via the linkage of muscovite-rich grain boundaries. Cracking and fluid infiltration produced the reaction K-feldspar + fluid → muscovite + albite. This muscovite-producing reaction served to drive both strain localization and dynamic recrystallization in K-feldspar via transformational plasticity. 96 Figure 30. Textures and LPOs of southern shear zone mylonite (Habach Group metaconglomerate). (a) Texture map; a strong shape preferred orientation in all grains. (b) Plagioclase porphyroclast texture. (c) Misorientation of 3˚ between core and rim of a plagioclase grain, indicating that subgrain rotation was an active recovery mechanism. (d) LPO for quartz grains within the feldspar strain shadow, implying some basal-<a> glide. (e) Plagioclase LPO; weak, with [001] subparallel to the X strain axis and poles to (010) subparallel to the Z strain axis. (f) Misorientation-angle distributions for neighbor pairs and random pairs of quartz matrix grains; equivalent at the 80% confidence level (using Kolmogorov-Smirnov test, d = 1.08). (g) Misorientation angle distribution for neighbor pairs and random pairs of recrystallized quartz grains in the strain shadow of a plagioclase porphyroclast; stronger correlation between neighbor and random pairs (d = 0.75; confidence interval <80%). The 60˚ peak is a results of Dauphine twinning. 97 The moderate-strain central shear zone exhibits mixed brittle and ductile deformation. Mm-scale quartz grains in the host gneiss have been deformed into polycrystalline clots of ~200-micron-scale grains with embayed boundaries, pervasive UE, island grains, and numerous subgrains, all features typical of regime 3 of Hirth and Tullis (1992). K-feldspar and plagioclase porphyroclasts have brittle fractures filled with epidote, muscovite, and quartz, deformation lamellae, deformation twins, patchy UE, and local core-and-mantle microstructures. Strain is concentrated in quartz- and mica- rich zones that anastamoze around brittley deformed K-feldspar and plagioclase porphyroclasts. Individual quartz augen larger than ~50 microns show a strong “single crystal” fabric in EBSD with their c axes subparallel to the Z strain axis and outward increasing misorientations of up to 5 degrees from core to rim that show rotation about poles to prism planes (Fig. 31). Matrix quartz shows a weaker, but still significant LPO with c axes subparallel to Y and dispersed <a> axes. The feldspar porphyroclasts contain no optically visible subgrains and are transected by fractures filled with neoblastic quartz and muscovite. Late, mm-scale brittle-ductile shear zone tips in the central shear zone show unexpected features. The host rocks show only brittle-ductile deformation, with evidence for grain-boundary migration in quartz accompanied by feldspar with brittle fractures, mechanical twins, deformation lamellae, UE and some subgrains. However, feldspar cataclasis was accompanied by the growth of muscovite, which localized deformation along muscovite-rich grain boundaries, eventually forming through- 98 Figure 31. Microstructure and quartz LPO of a moderate-strain brittle-ductile quartz- muscovite schist from the central shear zone. In equal-area, lower hemisphere stereonets, heavy line shows trace of foliation, S; dot shows lineation, L. (a) Microstructure map, with quartz grains colored by c axis orientations shown in stereonets (ms, muscovite). (b) LPO of quartz porphyroclasts, dominated by grains with c axes parallel to the Z strain axis. (c) LPO of matrix quartz grains, suggesting prism <a> slip. 99 going, discrete shear zones in which all grains are pervasively recrystallized. The feldspar grains display a strong LPO in EBSD analysis, characterized by [010] subparallel to the X strain axis and poles to (010) subparallel to the Z strain axis (Fig. 32). Quartz grains have a modest LPO (MUD ≤ 6) with c subparallel to Z and distributed <a> axes. Individual grains of feldspar and quartz have aspect rations of ~ 2:1 and show core-rim misorientations of 4-8 degrees; the intragrain lattice rotations in quartz are compatible with rotation about the c axis. Quartz neighbor pairs show an abundance of low misorientation angles compared to random pairs of grains, suggesting subgrain rotation recrystallization. All quartz grains throughout the Rinderkarsee shear zone therefore display optical microstructures typical of regime 3 dynamic recrystallization. Moderate-strain samples from the central shear zone as well as samples from shear zone tips also display strong LPOs, with clear intragrain lattice rotations in individual quartz grains, all suggesting that quartz in the central shear zone deformed via subgrain rotation recrystallization. However, EBSD analysis of high-strain samples from the southern shear zone shows these quartz grains to have a weak to random LPO. Combined with the similarity in misorientation distributions of neighbor pairs and random pairs in the matrix quartz, we concluded that the dominant quartz deformation mechanism in the southern shear zone quartz was dislocation creep-accommodated grain boundary sliding. 100 Figure 32. Shear zone sample from the Zentralgneis. In contrast to minerals outside shear zones, dislocation creep was important within shear zones. (a) Shear zone texture; SZB, shear zone boundary. (b) Texture map, with quartz grains colored by c axis orientations shown in stereonet. (c) Texture map, with feldspar grains colored by c axis orientations shown in stereonet. (d, e) Strong feldspar and quartz LPOs (maximum MUD 5-15), implying that deformation in the shear zones occurred via a combination of basal <a> and prism <a> slip in quartz and [010](001) slip in feldspar. (f) Lattice misorientations within individual grains. The 3˚-4˚ gradual change suggests dynamic recrystallization via subgrain rotation. (g) Significantly different misorientation angle distribution for quartz shear zone grains for neighbor pairs and random pairs. High frequency of quartz neighbor pairs with small misorientation angles points to subgrain rotation recrystallization. 101 Plagioclase in mylonitic samples show optical microstructures and intragrain lattice orientation indicative of dislocation creep. Combined with the absence of microcracking and a fine equant grain size, we concluded that feldspars in the high- strain southern shear zone deformed via dislocation creep-accommodated grain boundary sliding. In the moderate-strain central shear zone, plagioclase from incipient shear zone tips show similar evidence for dislocation creep, however, microcracking in these grains implies deformation was accommodated by a combination of microcracking, dislocation creep, and grain boundary sliding. Outside the mm-scale shear zones, there is a weak to random LPO in all phases. In summary, the ~100-m-wide main mylonitic shear zone, which has accommodated the majority of SEMP displacement, has a weak LPO implying the bulk grain-scale deformation of quartz and feldspar was dislocation assisted-grain boundary sliding. The moderate strain central shear zone shows dislocation creep in quartz rich samples with plagioclase cataclasis, whereas quartzofeldspathic samples show dislocation accommodated-grain boundary sliding of quartz and cataclasis + dislocation creep of plagioclase. Incipient shear zones show strong LPOs and dynamic recrystallization of all phases even though the plagioclase in the rest of the central shear zone was undergoing cataclasis. This suggests feldspar plasticity is the result of fluid driven alteration of kspar to muscovite, which localized strain along muscovite-rich grain boundaries, developing into grain-scale shear zones. Such alteration of feldspars to phyllosilicates is a recognized form of transformational plasticity, weakening the 102 whole rock and promoting deformation in the feldspar and quartz. Furthermore, the reduction in grain size can promote diffusion accommodated grain boundary sliding or solution transfer creep in the shear zone. Rosenberg and Schneider (2008) mapped the Rinderkarsee shear zone 80 km farther west as part of the Ahorn shear zone. They report a westward reduction in strain partitioning as the 1-100 m wide shear zones separated by 500 m low-strain panels give way to a continuous, 2 km wide ductile shear zone. They also document a temporal transition from early amphibolite-facies deformation at ≥ 500 degrees C in which quartz and plagioclase were ductile, to late greenschist-facies deformation at ≤ 300 degrees C in which quartz was undergoing dislocation glide, but not creep. In summary, the SEMP shows evidence for strain localization from near-surface levels to the base of the BDT, below which point the fault distributes strain into a fully ductile shear zone of up to 2 km width. We find phyllosilicates in the fault core of LK (exhumed from the top of BDT), as well as in incipient shear zones in the ductile/brittle-ductile Rinderkarsee shear zone, though we find no such evidence between these exhumation depths at Kitzlochklamm. 103 4.4 Discussion In order to determine what the fundamental structural properties of mature continental strike-slip fault zones are, we compare our results to those from other large displacement, mature strike-slip fault zones exhumed from brittle and ductile crustal conditions. Features common to all exhumed faults at a particular exhumation depth may indicate a fundamental property of plate-boundary-scale transform faulting, whereas differences should point out the effects of variables, such as lithology, geometry, and cumulative displacement. 4.4.1 Brittle fault zone structures Our idea of what faults look like in the brittle, seismogenic crust has largely been shaped by studies of the Punchbowl/North Fork San Gabriel (NFSG) faults in southern California and the Carboneras fault zone in Spain. These faults represent two end- member models of fault zone evolution, with the differences between them primarily attributed to the mechanical properties of the host rock at each locale (Faulkner et al., 2003; Imber et al., 2008). The Punchbowl and NFSG faults present examples of extreme strain localization within quartzofeldspathic host rocks at depths of 2-4 km. These faults feature one or two narrow (~1 m) ultracataclasite cores embedded within a foliated cataclasite, 104 hosted within a wider (~100 m) damage zone consisting of fractured country rock. The observation of a mesoscale ‘prominent fracture surface’ and microscale shear bands within the ultracataclasite (Chester and Chester, 1998), combined with grain-size distributions of D2 = 2.0 within the ultracataclasite and D2 = 1.6 in the fractured damage zone (Chester et al., 1993; Chester et al., 2005) all indicate that the narrow fault core accommodated most of the tens of kilometers of right-lateral displacement along these faults. This interpretation is in agreement with experimental data that show quartzofeldspathic rocks to strain weaken, favoring localization. On the other end of the spectrum, the Carboneras fault zone showcases the conditions that may lead to a broad zone of faulting in the brittle crust. This fault has been exhumed from 1.5-4 km depth and has accommodated ~40 km of left-lateral displacement within phyllosilicate-rich host rocks. In contrast to the Punchbowl/NFSG example, the Carboneras fault zone is up to 1 km wide and contains numerous phyllosilicate-rich gouge zones bounding lenses of variably deformed dolomite. Each individual gouge zone appears to have uniformly accommodated strain across its width, which can range up to 5m. Faulkner et al. (2003) attributed the generation of multiple gouge zones lacking ‘prominent fracture surfaces’ to the mechanical response of phyllosilicate, which displays strain-hardening behavior in experiments. Our results from the brittle exposure of the SEMP (exhumed from 4-8 km depth) most closely resemble the Punchbowl/NFSG style of faulting. Although the dolomite at 105 Gesause lacks outcrop-scale zones of ‘foliated cataclasite’ and ‘ultracataclasite,’ subsidiary fault orientations combined with grain-size distributions both point to the development of a single high-strain core, with the grain-size data further supporting a progressive evolution of strain localization. The fact that both quartzofeldspathic and calcite gouges both demonstrate strain-weakening behavior in lab experiments further highlights the fundamental control that lithology exerts on fault zone structure and mechanics in the brittle crust (e.g., Logan et al., 1992; Gu and Wong, 1994; Beeler et al., 1996). There are however other carbonate fault zones that differ from the SEMP, displaying multiple high-strain shear zones. The Mattinata fault of Italy, for example, shows evidence for four high-strain shear zones (defined by grain-size distributions) within a ~30 m wide fault core (Billi and Storti, 2004). The primary difference between the SEMP and the Mattinata faults is structural maturity – the Mattinata fault has accommodated ~ 2 km of left-lateral displacement, while the SEMP has experienced > 60 km of left-lateral displacement. 4.4.2 Brittle-ductile transition Analysis of faults exhumed from deeper structural levels has often been complicated by reactivation, causing initial fabrics to be overprinted by younger, possibly unrelated structures that formed at different structural levels, greatly complicating structural 106 analysis. Detailed studies of the Median Tectonic Line (MTL) and the Outer Hebrides Fault Zone (OHFZ) have sought to untangle such complications, however, giving us something to compare our results from the SEMP against. The MTL is thought to have accommodated 100’s of kilometers of sinistral strike-slip displacement since the late Cretaceous (Ichikawa, 1980). Outcrop of the fault above Miyamae village reveals a ~50 m wide fault core composed of cataclasites and foliated cataclasites, which is bordered by a ~250 m wide damage zone containing fractures and faulted granitoid mylonites. The fault zone itself is hosted within a 5-km-wide mylonite zone. The mylonites developed during exhumation from 18 to 11 km depth, while the cataclastic fault core and damage zone are thought to have developed during continued deformation at shallower crustal levels (Takagi, 1986). A second outcrop near Miyamae, in the Fukaya river, shows the fault core to be composed of quartzofeldspathic ultramylonite and phyllonite in addition to the previously observed cataclasite and foliated cataclasite. Jefferies et al. (2006a) concluded that cataclasis in the fault core allowed for late fluid influx, progressively altering fault core rocks into foliated cataclasite and eventually phyllonite. This process weakened the fault core as strong phases such as feldspar and hornblende were replaced by weak phyllosilicates. The Outer Hebrides Fault Zone (OHFZ) is a basement fault exhumed along the east coast of the Outer Hebrides, Scotland. Borehole data combined with conventional and deep 2-D seismic reflection data point to the OHFZ being a crustal-scale structure, 107 possibly offsetting the Moho at 25 km depth (Binns et al., 1974; Smythe et al., 1982; Peddy, 1984; Strachan and Holdsworth, 2000). Imber et al. (2001) defined a maximum of five fault-related deformation events along the onshore trace of the OHFZ; for the purposes of our comparison we focus on the late Caledonian sinistral strike-slip phase that accommodated 85-95 km of displacement. This deformation event is characterized by a macroscopically ductile network of narrow (<500 m) phyllonitic shear zones formed under greenschist facies conditions (Butler et al., 1995). These sinistral fabrics overprint the earlier brittle phase of thrusting. Using field and microstructural evidence, Imber et al. (2001) concluded that the transition from brittle thrusting to macroscopically ductile sinistral shear resulted from the influx of aqueous fluids, promoting greenschist facies retrogression (300-350C) and phyllonitization within the previously “dry” fault zone. They also document a similar process of fluid influx, retrogression and phyllonitization active during a Proterozoic thrusting event that occurred within the viscous flow regime (> 500 C), demonstrating that transformational plasticity and associated weakening/localization can occur at a range of crustal conditions. Our results at the Klamms and Rinderkarsee (covering exhumation depths of ~10-17 km) further support the observations of weakening and localization as seen in the MTL and OHFZ. At Lichtensteinklamm, we observe a ~50 m-wide core in which host Grauwacken zone schists and Klammkalk marble tectonites have clearly deformed via cataclasis, and were later transformed into fine-grained clays. Evidence for 108 transformational plasticity is also observed at the more deeply exhumed Rinderkarsee outcrop. Though the majority of SEMP-related displacement there is within a 100-m- wide ductile shear zone, we see evidence for phyllonitization at the grain scale, weakening the fault zone and promoting localization even in ductile crustal conditions. Between these outcrops, at Kitzlochklamm, we do not observed a gouged core nor do we see grain-scale evidence of phyllonitization. However, weak LPOs in both the Klammkalk and Grauwacken zone combined with heavily damaged graphite schist adjacent to the sharp lithologic contact all suggest the majority of SEMP displacement was accommodated in a very narrow zone at this outcrop. The lack of gouge at Kitzlochklamm could be due to several factors. Given the scale of displacement accommodated by the SEMP, one possibility is that a gouge zone has simply been slivered or sheared out at this outcrop. Alternatively, a gouge zone may also never have had the opportunity to develop at Kitzlochklamm. Given our grain- scale observations of increased dislocation and diffusion creep at this outcrop, which has been exhumed from a slightly deeper depth than Lichtensteinklamm, we may simply be observing the importance of cataclasis in forming phyllosilicates at greenschist facies conditions. If Kitzlochklamm records deformation conditions from just below the brittle regime, the fault was never able to mechanically reduce grain size and create permeability pathways for aqueous fluids to initiate phyllonitization. 109 4.5 Conclusions In comparing our three-dimensional results from the SEMP to other exhumed faults, we are able to note important differences and infer their causes. When looking at brittle fault zones, we conclude from the existing data that fault zone structure is largely shaped by the rheological properties of the host rocks. Rocks that strain- weaken produce relatively narrow (10’s of meters) high-strain zones that may feature one or several fault cores, depending on cumulative displacement, whereas rocks that strain-harden will produce a wide (km-scale) zone of faulting with numerous fault cores, none of which have accommodated a significant portion of the total slip on the fault. At brittle-ductile conditions, we find transformational plasticity (and specifically, phyllonitization) to potentially be a major factor in shaping fault structure. Faults that undergo cataclasis at these depths allow for the influx of aqueous fluids, progressively altering and weakening the fault zone, and producing a localized high-strain core in what otherwise might be a regime of diffuse, ductile shear. CHAPTER 5: Conclusions 110 The goal of my Ph.D. has been to address specific unresolved questions regarding fault zone structure and mechanics by studying the differentially-exhumed SEMP fault system. To restate: (1) How does fault zone thickness vary with depth? Can we observe a gradual widening or narrowing, or does thickness vary nonsystematically, reflecting the importance of variables such as host rock and deformation mechanism? (2) Are faults characterized by a narrow principal slip surface and/or gouge zone throughout the seismogenic crust, or is the development of such features dependent upon host rock rheology? (3) Is the damage zone surrounding the fault core a relict feature formed during the early stages of fault evolution, or does damage occur continually with each seismic rupture? (4) How does fault zone geometry change across the brittle-ductile transition, and how abrupt is this transition in deformation mechanisms? With regard to question (1), our results from the SEMP show that for this particular fault, thickness gradually increases with depth. Starting at the shallowest outcrop and moving progressively downward, the high strain fault core is ~10m wide at Gstatterboden, <50m wide at Lichtensteinklamm, and ~100m wide at Rinderkarsee. 111 Rosenberg and Schneider (2008) did not specify a strain gradient in the Ahorn shear zone that might reveal a fault core at this depth level. The greatest increase in fault zone thickness occurs within the ductile crust, as all mineral phases transition to ductile creep and fault zone thickness increases by an order of magnitude. We find that the specific fault zone structures specified in question (2) reflect the unique mineralogical and rheological properties of the rocks a particular fault is hosted within. Extreme localization, as reflected in the principal slip surface of the Punchbowl fault, is most likely due to the presence of minerals like micas. We see a similar process occurring at the incipient shear zones at Rinderkarsee, where the breakdown of plagioclase to muscovite promotes extreme strain localization at the grain scale. Outcrops of the SEMP at Gstatterboden and the klamms lack high concentrations of micas, and as a results, we do not find principal slip surfaces (although we should restate that limited outcrop at Gstatterboden prevents us from providing a definitive answer). Likewise, the formation of a fault core with a distinct gouge zone in brittle fault zones likely reflects the strain-weakening properties of the host rock, rather than a necessary fault zone structure. Our only ability to assess question (3) comes from Gstatterboden, where we can make a clear argument, based on grain-size distribution, that the damage surrounding the high-strain core at this outcrop is relict, formed during the early stages of fault evolution, before strain eventually localized to form a fault core. This argument is 112 based on an interpretation of the mechanism responsible for producing grain size distributions of D 2 = 1.6 and 2.0, and could be applied to the analysis of any brittle fault zone. The deeper outcrops at Lichtensteinklamm, Kitzlochklamm, and Rinderkarsee don’t lend themselves to similar analysis, and we are therefore unable to independently conclude how these outcrops evolved. Our study of Lichtensteinklamm and Kitzlochklamm allows us to directly observe how fault structure and mechanics changes across the brittle-ductile transition (BDT) to address question (4). We see almost no change in fault zone geometry or overall thickness across the BDT. We have interpreted the absence of a gouge zone at Kitzlochklamm, which has been exhumed from ~1km greater depth than Lichtensteinklamm, to indicate that the transition from cataclasis to crystal-plasticity occurs over a relatively narrow depth range. However, evidence for solution mass transfer and phyllonitization at these outcrops highlights the importance of understanding that crustal deformation mechanisms are not simply “brittle” or “ductile;” that a variety of cataclastic and creep processes operate throughout the crust, and can in fact cycle back and forth dynamically. In summary, our study of the SEMP fault system has shown the importance that local conditions play on the evolution of a particular fault. Generic models of fault zone structure should be understood to only provide a rough framework, specific to certain rock types, assumed deformation mechanisms, and strain rates. The goal to create a 113 universal, three-dimensional model of faulting may in fact be counterproductive to truly understanding deformation in the crust. Instead, we might be best served by focusing research efforts on further understanding how unique local conditions can shape individual faults. REFERENCES 114 Ampferer, O. (1935), Geologische karte der Gesäuseberge. Geologische Bundesanstalt, Vienna, Austria. Anderson, L., R.H. Osbourne, and D.F. Palmer (1983), Cataclastic rocks of the San Gabriel fault: An expression of deformation at deeper crustal levels in the San Andreas fault zones, Tectonophysics, 98, 209-251. Angelier, J. (1984), Tectonic analysis of fault slip data sets, J. Geophys. Res., 89, 5835-5848. Austin, N., B. Evans, M. Herwegh, and A. Ebert (2008), Strain localization in the Morcles Nappe (Helvetic Alps, Switzerland), Swiss J. Geosci., 101, 341-360. Barnhoorn, A., M. Bystricky, L. Burlini, and K. Kunze (2004), The role of recrystallization on the deformation behavior of calcite rocks: large strain torsion experiments on Carrara marble, J. Struct. Geol., 26(5), 885-903. Bechor, N., P. Segall, Y. J. Hsu, and S. B. Yu, Time-depended inversion for post- seismic slip following the 1999 Chi-Chi Taiwan earthquake using GPS observations, EOS. Trans., AGU, 82(47), Nov. 2001. 115 Beeler, N.M., T.E. Tullis, M.L. Blanpied, and J.D. Weeks (1996), Frictional behavior of large displacement experimental faults, J. Geophys. Res., 101, 8697-8715. Behrmann, J.H. (1983), Microstructure and fabric transitions in calcite tectonites from the Sierra Alhamilla (Spain), Geol. Rundsch., 72(2), 605-618. Beyssac, O., B. Goff´e, C. Chopin, and J.N. Rouzaud (2002a), Raman spectrum of carbonaceous material in metasediments: a new geothermometer. J. Metamorphic Geol., 20, 859-871. Bickle, M.J. and C.J. Hawkesworth (1978), Deformation phases and the tectonic history of the eastern Alps, Geol. Soc. Am. Bull., 89(2), 293-306. Biegel, R.L., C.G. Sammis, and J.H. Dieterich (1989), The frictional properties of a simulated gouge having a fractal particle distribution, J. Struct. Geol., 11, 827-846. Bigi, G., A. Castellarin, M. Coli, G. V. Dal Piaz, R. Sartori, P. Scandone, and G.B. Vai (1990), Structural model of Italy, sheet 1, Cons. Naz. delle Ric., Progretto Finalizzato Geodin, SELCA, Florence, Italy. 116 Billi, A. (2007), On the extent of size range and power law scaling for particles of natural carbonate fault cores, J. Struct. Geol., 29, 1512-1521, doi:10.1016/j.jsg.2007.06.007 Billi, A., F. Salvini, and F. Storti (2003), The damage zone-fault core transition in carbonate rocks: implications for fault growth, structure and permeability, J. Struct. Geol., 25, 1779-1794, doi: 10.1016/S0191-8141(03)00037-3. Billi, A. and F. Sorti (2004), Fractal distribution of particle size in carbonate cataclastic rocks from the core of a regional strike-slip fault zone, Tectonophysics, 384, 115-128. Binns, P.E., R. McQuillin, and N. Kenolty (1974), The geology of the Sea of the Hebrides, Rep 73/14, Inst. Geol. Sci., HMSO, London, pp. 44. Blenkinsop, T.G. (1991), Cataclasis and processes of particle size reduction, Pure Appl. Geophys., 126, 59-86. Brace, W. F., and D. L. Kohlstedt (1980), Limits on lithospheric stress imposed by laboratory experiments, J. Geophys. Res., 85, 6248– 6252. 117 Burgmann, R., P. Segall, M. Lisowski, and J. Svarc (1997), Postseismic strain following the 1989 Loma Prieta earthquake from GPS and leveling measurements, J. Geophys. Res., 102(B3), 4933-4956. Caine, J.S., J.P. Evans, and C.B. Forster (1996), Fault zone architecture and permeability structure, Geology, 24, 1025-1028. Carter, N. L., and S. H. Kirby (1978), Transient creep and semibrittle behavior of crystalline rocks, Pure Appl. Geophys., 116, 807–839. Chester, F. M. (1995), A rheologic model for wet crust applied to strike-slip faults, J. Geophys. Res., 100, 13,033– 13,044. Chester, F.M. and J.M. Logan (1986), Implications for mechanical properties of brittle faults from observations of the Punchbowl fault zone, California, Pure Appl. Geophys., 124, 79-106. Chester, F.M., J.P. Evans, and R.L. Biegel (1993), Internal structure and weakening mechanisms of the San Andreas Fault, J. Geophys. Res., 98, 771-786. Chester, F.M. and J.S. Chester (1998), Ultracataclasite structure and friction processes of the Punchbowl fault, San Andreas system, California, Tectonics, 295, 199-221. 118 Chester, J.S., F.M. Chester, and A.K. Kronenberg (2005), Fracture surface energy of the Punchbowl fault, San Andreas system, California, Nature, 437, 133-136. Cole, J., B. Hacker, L. Ratschbacher, J. Dolan, G. Seward, E. Frost, and W. Frank (2007), Localized ductile shear below the seismogenic zone: Structural analysis of an exhumed strike-slip fault, Austrian Alps, J. Geophys. Res., 112, B12304, doi:10.1029/2007B004975. Decker, K., H. Peresson, and P. Faupl (1994), Miocene tectonics in the eastern Limestone Alps; kinematics, paleostress and distribution of deformation during the lateral extrusion of the Central Alps, Jahrbuch der Geologischen Bundesanstalt, 137, 5-18. Dietrich, D. and H. Song (1984), Calcite fabrics in a natural shear environment, the Helvetic nappes of western Switzerland, J. Struct. Geol., 6(1-2), 19-32. Ebert, E., M. Herwegh, and A. Pfiffner (2007), Cooling induced strain localization in carbonate mylonites within a large-scale shear zones (Glarus thrust, Switzerland), J. Struct. Geol., 29, 1164-1184. 119 Evans, J.P., Z.K. Shipton, M.A. Pachell, S.J. Lim, and K. Robeson (2000), The structure and composition of exhumed faults, and their implications for seismic processes, in Proceedings of the 3 rd Conference on Tectonic Problems of the San Andreas Fault System, 6-8 December 2000, edited by G. Bokelmann and R.L. Kovach, pp. 67-81, Stanford University. Faulkner, D.R., A.C. Lewis, and E.H. Rutter (2003), On the internal structure and mechanics of large strike-slip fault zones: field observations of the Carboneras fault in southeastern Spain, Tectonophysics, 367, 235-251. Ferrill, D.A., A.P. Morris, M.A. Evans, M. Burkhard, R.H. Groshong, and C.M. Onasch (2004), Calcite twin morphology: a low-temperature deformation geothermometer, J. Struct. Geol., 26, 1521-1529. Fliervoet , T.F., S.H. White, and M.R. Drury (1997), Evidence for dominant grain- boundary sliding deformation in greenschist- and amphibolite-grade polymineralic ultramylonites from the Redbank Deformed Zone, Central Australia, J. Struct. Geol., 19(12), 1495-1520. Frisch, W., A. Kuhlemann, I. Dunkl, and A. Brügel (1998), Palinspastic reconstruction and topographic evolution of the Eastern Alps during late Tertiary tectonic extrusion. Tectonophysics, 297, 1–15. 120 Frost, E., J. Dolan, C. Sammis, B. Hacker, J. Cole, and L. Ratschbacher (2009), Progressive strain localization in a major strike-slip fault exhumed from midseismogenic depths: Structural observations from the Salzach-Ennstal-Mariazell- Puchberg fault system, Austria, J. Geophys. Res., 114, B04406, doi:10.1029/2008JB005763. Genser, J., J. van Wees, S. Cloetingh, and F. Neubauer (1996), Eastern Alpine tectono-metamorphic evolution: Constraints from two-dimensional P-T-t modeling, Tectonics, 15(3), 584-604. Geologische Bundesanstalt (1980), Der geologische Aufbau Österreichs, 700 pp., Springer, Wien, New York. Gu, Y. and T.-F. Wong (1994), Development of shear localization in simulated quartz gouge: effect of cumulative slip and gouge particle size, Pure Appl. Geophys., 143, 387-423. Hacker, B. R., and J. M. Christie (1990), Brittle/ductile and plastic/cataclastic transitions in experimentally deformed and metamorphosed amphibolite, in The Brittle-Ductile Transition in Rocks, Geophys. Monogr. Ser., vol. 56, edited by A. G. Duba et al., p. 127– 147, AGU, Washington, D. C. 121 Hacker, B. R. (1997), Diagenesis and the fault-valve seismicity of crustal faults, J. Geophys. Res., 102, 24,459– 24,467. Heilbronner, R. and N. Keulen (2006), Grain size and grain shape analysis of fault rocks, Tectonophysics, 427, 199-216. Hirth, G., and J. Tullis (1992), Dislocation creep regimes in quartz aggregates, J. Struct. Geol., 14, 145– 159. Hobbs, B. E., A. Ord, and C. Teyssier (1986), Earthquakes in the ductile regime?, Pure Appl. Geophys., 124(1– 2), 309– 336. Ichikawa, K. (1980), Geohistory of the Median Tectonic Line of southwest Japan, in Median Tectonic Line of southwest Japan, Memoirs of the Geol. Soc. of Japan, 18, 5- 24. Imber, J., R.E. Holdsworth, C.A. Buter, R.A. Strachan (2001), A reappraisal of the Sibson-Scholz fault zone model: The nature of the frictional to viscous (“brittle- ductile”) transition along a long-lived, crustal scale fault, Outer Hebrides, Scotland, Tectonics, 20(5), 601-624. 122 Imber, J., R.E. Holdsworth, S.A.F. Smith, S.P. Jefferies, and C. Collettini (2008), Frictional-viscous flow, seismicity and the geology of weak faults: a review and future directions. In The Internal Structure of Fault Zones: Implications for Mechanical and Fluid-Flow Properties, edited by C. A. J. Wibberley, W. Kurz, J. Imber, R.E. Holdsworth, and C. Collettini, Geol. Soc. Spec. Publ., 299, 151–173. Janecke, S. U., and J. P. Evans (1988), Feldspar-influenced rock rheologies, Geology, 16, 1064– 1067. Jefferies, S.P., R.E. Holdsworth, C.A.J. Wibberley, T. Shimamoto, C.J. Spiers, A.R. Neimeijer, and G.E. Lloyd (2006a), The nature and importance of phyllonite development in crustal-scale fault cores: an example from the Median Tectonic Line, Japan, J. Struct. Geol., 28, 220-235. Kendall, K. (1978), The impossibility of comminuting small particles, Nature, 272, 710-711. Knipe, R. J. (1989), Deformation mechanisms – recognition from natural tectonites, J. Struct. Geol., 11, 127–146. Kurz, W., F. Neubauer, and J. Genser (1996) Kinematics of Penninic nappes (Glockner nappe and basement-cover nappes) in the Tauern Window (eastern Alps, 123 Austria) during subduction and Penninic-Austroalpine collision, Eclogae Geol Helv 89, pp. 573-605. Kurz, W., F. Neubauer, W. Unzog, J. Genser, and X. Wang (2000), Microstructural and textural development of calcite marbles during polyphase deformation of Penninic units within the Tauern Window (Eastern Alps), Tectonophysics, 316(3-4), 327-342. Lin, A., S. Maruyama, A. Stallard, K. Michibayashi, A. Camacho, and K. Kano (2005), Propagation of seismic slip from brittle to ductile crust: Evidence from pseudotachylyte of the Woodroffe thrust, central Australia, Tectonophysics, 402, 21– 35. Linzer, H.G, L. Ratschbacher, and W. Frisch (1995), Transpressional collision structures in the upper crust of the fold-thrust belt of the Northern Calcareous Alps, Tectonophysics, 242, 41-63. Linzer, H.G., F. Moser, R. Nemes, L. Ratschbacher, and B. Sperner (1997), Build-up and dismemberment of a classical fold-and-thrust belt: from non-cylindrical stacking to lateral extrusion in the eastern Northern Calcareous Alps, Tectonophysics, 272, 97- 124. 124 Linzer, H.G., K. Decker, H. Peresson, R. Dell’Mour, and W. Frisch (2002), Balancing orogenic float of the Eastern Alps, Tectonophysics, 354, 211-237. Logan, J.M., C.A. Dengo, N.G. Higgs, and Z.Z. Wang (1992), Fabrics of experimental fault zones: their development and relationship to mechanical behavior, in Fault Mechanics and Transport Properties of Rocks, edited by B. Evans and T.-F. Wong, pp. 33-69, Academic Press, New York. Mainprice, D., 2003. “PF2k”—general purpose program for Euler angle triplets. http://www.dstu.univ- montp2.fr/TECTONOPHY/petrophysics/software/petrophysics_software.html. Means, W.D. (1995), Shear zones and rock history, Tectonophysics, 247, 157-160. Montesi, L.G.J. and G. Hirth (2003), Grain size evolution and the rheology of ductile shear zones: from laboratory experiments to postseismic creep, Earth Planet Sci. Lett., 211(1-2), 97-110. Otsu, N. (1979), A threshold selection method from gray-level histograms, IEEE Trans. on System, Man Cybernetics, 9, 62-66. Passchier, C. W. (1982), Pseudotachylyte and the development of ultramylonite bands 125 in the Saint-Barthelemy Massif, French Pyrenees, J. Struct. Geol., 4(1), 69–79. Peddy, C.P. (1984), Displacement of the Moho by the Outer Isles Thrust shown by seismic modeling, Nature, 3(2), 628-630. Peresson, H. and K. Decker (1997a), Far-field effects of Late Miocene subduction in the Eastern Carpathians: E–W compression and inversion of structures in the Alpine– Carpathian–Pannonian region., Tectonics, 16, 38–56. Pieri, M., L. Burlini, K. Kunze, D.L. Olgaard, and I.C. Stretton (2001a), Rheological and microstructural evolution of Carrara marble with high shear strain: results from high temperature torsion experiments, J. Struct. Geol., 23, 1393-1413. Pieri, M., K. Kunze, L. Burlini, I.C. Stretton, D.L. Olgaard, J.-P. Burg, and H.R. Wenk (2001b), Texture development of calcite by deformation and dynamic recrystallization at 1000 K during torsion experiments of marble to large strains, Tectonophysics, 330, 119–140. Power, W.L., T.E. Tullis, S.R. Brown, G.N. Boinott, and C.H. Scholz (1987), Roughness of natural fault surfaces, Geophys. Res. Lett., 14, 29-32. 126 Ratschbacher, L., J.H. Behrmann, and A. Pahr (1990), Penninic windows at the eastern end of the Alps and their relation to the intra-Carpathian basins, Tectonophysics, 172, 91-105 Ratschbacher, L., W. Frisch, H.-G. Linzer, and O. Merle (1991a), Lateral extrusion in the Eastern Alps, Part 2: structural analysis, Tectonics, 10, 257–271. Ratschbacher, L., O. Merle, P. Davy, and P. Cobbold (1991b), Lateral extrusion in the Eastern Alps, Part 1: Boundary conditions and experiments scaled for gravity, Tectonics, 10, 245–256. Ratschbacher, L. and W. Frisch (1993), Palinspastic reconstruction of the pre-Triassic basement units in the Alps: the Eastern Alps, in Pre-Mesozoic geology in the Alps, edited by J.F. Raumer and F. Neubauer, pp. 41-51, Springer, Heidelberg. Ratschbacher, L., J. Elias, W. Frisch, H.-G. Linzer, and B. Sperner (1995), Lateral extrusion in the eastern Alps: driving forces, the Carpathian connection, and vertical particle paths (a view from classical geology, experimental modeling, and Ar/Ar and FT thermochronology), Second workshop on Alpine Geology, Institute of Geology, Basel University, Basel, 53-55. 127 Reilinger, R.E., S. Ergintav, R. Burgmann, S. McClusky, O. Lenk, A. Barka, O. Gurkan, L. Hearn, K.L. Feigl, R. Cakmak, B. Aktug, H. Ozener, and M.N. Toksoz (2000), Coseismic and postseismic fault slip for the 17 August 1999, M = 7.5, Izmit, Turkey earthquake, Science, 289(5484), 1519-1524. Rice, J.R., C.G. Sammis, and R. Parsons (2005), Off-axis secondary faulting induced by a dynamic slip-pulse, Bull. Seis. Soc. America, 95, 109-134. Rosenberg, C.L., J.P. Brun, and D. Gapais (2004), Indentation model of the Eastern Alps and the origin of the Tauern Window, Geology, 32, 997-1000. Rosenberg, C.L. and S. Schneider (2008), The western termination of the SEMP Fault (eastern Alps) and its bearing on the exhumation of the Tauern Window, in Tectonic Aspects of the Alpine-Carpathian-Dinaride System, Geol. Soc. Spec. Publ. 298, 197- 218. Rutter, E. H. (1986), On the nomenclature of mode of failure transitions in rocks, Tectonophysics, 122(3– 4), 381– 387. Sammis, C.G., G.C.P. King, and R. Biegel (1987), The kinematics of gouge deformation, Pure Appl. Geophys., 125, 777-812. 128 Sammis, C.G., and G.C.P. King (2007), Mechanical origin of power law scaling in fault zone rock, Geophys. Res. Lett., 34, L04312, doi:10.1029/2006GL028548. Savage, J.C. and J.L. Svarc (1997), Postseismic deformation associated with the 1992 M w = 7.3 Landers earthquake, southern California, J. Geophys. Res., 102(B4), 7565- 7577. Schmid, J. (1997) The genesis of white marbles: geological cause and archaeological applications, p. 156, Ph.D. Thesis, University Bern. Scholz, C. H. (1988), The brittle-plastic transition and the depth of seismic faulting, Geol. Rundsch., 77, 319– 328. Scholz, C. H. (2002), The Mechanics of Earthquakes and Faulting, 471 pp., Cambridge Univ. Press, New York. Schulz, S.E. and J.P. Evans (2000), Mesoscopic structure of the Punchbowl Fault, Southern California and the geologic and geophysical structure of active strike-slip faults, J. Struct. Geol., 22, 913-930. Segall, P. and D.D. Pollard (1983), Nucleation and growth of strike-slip faults in granite, J. Geophys. Res., 88, 555-568. 129 Shimamoto, T. (1989), The origin of S-C mylonites and a new fault-zone model, J. Struct. Geol., 11, 51– 64. Sibson, R. (1980), Transient discontinuities in ductile shear zones, J. Struct. Geol., 2, 165–171. Sibson, R. H. (1982), Fault zone models, heat flow, and the depth distribution of earthquakes in the continental crust of the United States, Bull. Seismol. Soc. Am., 72(1), 151– 163. Sibson, R. H. (1983), Continental fault structure and the shallow earthquake source, J. Geol. Soc (London, U.K.), 1983, 140(5), 741-767. Sibson, R.H. (1984), Roughness at the base of the seismogenic zone: contributing factors, J. Geophys. Res., 89(B7), 5781-5800. Smythe, D.K., A. Dobinson, R. McQuillen, J.A. Brewer, D.H. Matthews, D.J. Blundell, and B. Kelk (1982), Deep structure of the Scottish Caledonides revealed by the MOIST reflection profile, Nature, 299, 338-340. 130 Spang, J.H. (1972), Numerical method for dynamic analysis of calcite twin lamellae, Geol. Soc. Am. Bull., 83, 467-472. Sperner, B. and L. Ratschbacher (1994), A Turbo Pascal program package for graphical presentation and stress analysis of calcite deformation, Zeitschrift der Deutschen Geolgischen Gesellschaft, 145, 414-423. Storti, F., A. Billi, and F. Salvini (2003), Particle size distributions in natural carbonate fault rocks: insights for non-self-similar cataclasis, Earth Planet Sci. Lett., 206, 173-186. Strachan, R.A and R.E. Holdsworth (2000), Late Neoproterozoic (<750 Ma) to Early Ordovician passive margin sedimentation along the Laurentian margin of Iapetus, in The Geological History of Britain and Ireland, edited by N.G. Woodcock and R.A. Strachan, Blackwell Science Ltd., 73-87. Takagi, H. (1986), Implications of mylonitic microstructures for the geotectonic evolution of the Median Tectonic Line, central Japan, J. Struct. Geol., 8, 3-14. Tullis, J., and R. A. Yund (1977), Experimental deformation of dry Westerly Granite, J. Geophys. Res., 82(36), 5705– 5718. 131 Tullis, J., and R. A. Yund (1992), The brittle-ductile transition in feldspar aggregates: An experimental study, in Fault Mechanics and Transport Properties of Rocks, edited by B. Evans and T.-F. Wong, pp. 89– 117, Academic, San Diego. Wang, X. and F. Neubauer (1998), Orogen-parallel strike-slip faults bordering metamorphic core complexes: the Salzach-Enns fault zone in the Eastern Alps, Austria, J. Struct. Geol., 20, 799-818. Wesnousky, S.G. (1988), Seismological and structural evolution of strike-slip faults, Nature, 335, 340-342. White, J. C. (1996), Transient discontinuities revisited; pseudotachylyte, plastic instability and the influence of low pore fluid pressure on deformation processes in the mid-crust, J. Struct. Geol., 18(12), 1471– 1486. Wibberly, C.A.J. and T. Shimamoto (2003), Internal structure and permeability of major strike-slip fault zones: The Median Tectonic Line in Mie Prefecture, southwest Japan, J. Struct. Geol., 25, 59-78, doi: 10.1016/S0191-8141(02)00014-7.

Abstract (if available)

Abstract Outcrops of the Salzach-Ennstal-Mariazell-Puchberg (SEMP) fault system exhumed from depths of ~4-17 km allow for the direct observation of fault zone structures throughout the crust, and provide insights into the way this fault, and perhaps others, distributes strain in three dimensions. At Gstatterboden, exhumed from ~4-8 km, grain size distributions and small fault data reveal the presence of a 10-m-wide high-strain core towards which strain localized during fault evolution. Brittle fracture was accommodated via constrained comminution, which only occurs in strain-weakening rheologies and favors localization. Exposures of the SEMP at Lichtensteinklamm and Kitzlochklamm, exhumed from ~12 km depth, bracket the brittle ductile transition. At these outcrops, the SEMP is characterized by a ~70-m-wide, cataclastic fault core that has been altered to clays that transitions downward into a wide, ductile shear zone that has accommodated only minor amounts of strain, placing the majority of displacement on the razor-sharp fault contact. Deformation mechanisms transition from cataclasis and minor amounts of dislocation creep in calcite, to dislocation creep in quartz and calcite occurring against a background of fault-normal solution mass transfer. The ductile/ductile-brittle Rinderkarsee shear zone, exhumed from ~17 km, marks the SEMP’s continuation into the Tauern Window and is composed of three distinct shear zones. The southern, 100-m-wide shear zone has accommodated the most strain, and shows evidence for creep-accommodated grain boundary sliding in feldspar and quartz, while incipient shear zones contain ductile quartz and brittle-feldspars that undergo dislocation creep as fluids alter Kspar to muscovite, which localizes strain along felspar grain boundaries, encouraging ductility. These findings are compared to results from other faults exhumed from similar depth ranges, highlighting fundamental fault zone structures and characteristics.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Crustal-scale ductile shear zones in four dimensions

PDF

Heterogeneity of earthquake stress drops, focal mechanisms and active fault zones

PDF

Detection and modeling of slow slip events as creep instabilities beneath major fault zones

PDF

Fluid and structural evolution of Late Cretaceous Cordilleran ductile-brittle shear zones in hot and cooling orogens: examples from the Sierra Nevada, CA and Cascades Crystalline Core, WA

PDF

Multi-scale imaging of major fault zones in Southern California

PDF

Implications of new fault slip rates and paleoseismologic data for constancy of seismic strain release and seismic clustering in the eastern California shear zone

PDF

Paleoseismologic and slip rate studies of three major faults in southern California: understanding the complex behavior of plate boundary fault systems over millenial timescales

PDF

Quantifying ground deformation of large magnitude earthquakes using optical imgaging systems

PDF

Fault slip rates, constancy of seismic strain release, and landscape evolution in the eastern California shear zone

PDF

Observations of temporal and spatial patterns of strain accommodation and earthquake occurrence along strike-slip faults of New Zealand and southern California, USA

PDF

Spatiotemporal variations of stress field in the San Jacinto Fault Zone and South Central Transverse Ranges

PDF

Atomistic modeling of the mechanical properties of metallic glasses, nanoglasses, and their composites

Asset Metadata

Creator Frost, Erik Karl (author)

Core Title Direct observation of fault zone structure and mechanics in three dimensions: a study of the SEMP fault system, Austria

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Publication Date 05/21/2010

Defense Date 05/05/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag fault structure,microstructures,OAI-PMH Harvest,rheology

Place Name Austria (countries), faults: Salzach-Ennstal-Mariazell-Puchberg (geographic subject)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Dolan, James F. (committee chair), Nutt, Steven R. (committee member), Sammis, Charles G. (committee member)

Creator Email efrost@usc.edu,erik.frost@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3092

Unique identifier UC1418199

Identifier etd-Frost-3763 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-337121 (legacy record id),usctheses-m3092 (legacy record id)

Legacy Identifier etd-Frost-3763.pdf

Dmrecord 337121

Document Type Dissertation

Rights Frost, Erik Karl

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu