CHARACTERIZATION OF HIGH-ANGLE FAULTS ON THE ISLAND OF SYROS, GREECE
by
Eli D. Lazarus
A thesis submitted in partial fulfillm of the requirement for the Degree of Bachelor of the Arts with Honors in Geosciences
WILLIAMS COLLEGE
Williamstown, Massachusetts
May 2004 Acknowledgments
My deepest thanks to Paul Karabinos, my principal advisor, and Tekla Harms (Department of Geology, Amherst College) for overseeing this work; Larry Meinert at Smith College for his help and making available his fluid-inclusion laboratory; Nancy Piatczyc for her technical guidance on the SEM; David Dethier for his critical review; the Greece 2003 project leaders and my undergraduate colleagues; the Williams Department of Geosciences; the Keck Geology Consortium, for the generous grant that made this study possible; my loving parents; to Katherine Ackerly; and to Paige McClanahan, whose sharp rationale convinced me to pack my running shoes and go to Syros. Table of Contents ......
Acknowledgments ...... i . . Table of Contents ...... 11 List of Figures ...... ill Abstract ...... iv
Chapter 1: Introduction ...... 1 The Problem ...... 1 Regional Geologic History ...... 3 Geology of Syros ...... 15 Normal Faulting and Exhumation ...... 19
Chapter 2: Field Observations ...... 22 Location Descriptions ...... 22 Slip Indicators ...... 43 Discussion of Location and Slip Data ...... 49
Chapter 3: Petrographic and Fluid-Inclusion Analysis ...... 58 Petrography ...... 58 Fluid Inclusions ...... 69
Chapter 4: Discussion and Farther Work ...... 81 Discussion ...... 81 Further Work ...... 84
References ...... 87 List of Figures ......
Chapter 1: Introduction
Map of Syros Island ...... 2 Regional map of Aegean and Cycladic Islands ...... 4 Paleogeographic cartoon of central Aegean in late Cretaceous ...... 5 Schematic geometric model for exhumation of blueschist terranes ...... 9 Kinematic interpretation of regional extension in Aegean ...... 10 Simplified local kinematics of central Aegean ...... 11 Summarization of kinematic evolution of the eastern Mediterranean ...... 13 Regional structural features of the central Aegean ...... 14 Primary effects on modern deformation in Aegean region ...... 14 Simplified geologic map of Syros ...... 16 Example of antithetic faults in hanging wall of listric normal fault ...... 18 High-angle normal faults in the Halara highland ...... 18 Schematic geometry of imbricate listric normal faults ...... 20 Unroofing of a metamorphic core complex ...... 20
Chapter 2: Field Observations
Equal-area stereoplot of dominant fault planes ...... 23 Sample location map of study area at Diapori ...... 25 Oblique view and details of fault zone at Diapori ...... 26 Sharp contact between host marble and fault gouge at Diapori ...... 27 Detail of marble and schist blocks at Diapori ...... 28 Pod of crystalline quartz at Diapori ...... 28 Detail of gouge material in breccia wall at Diapori ...... 30 Southern terminus of main breccia wall at Diapori ...... 31 View toward schist saddle and calcite vein cascade at Diapori ...... 32 Sample location map of study area at Katakefalos ...... 34 View to NW toward Cape Katakefalos ...... 35 View to E of fault zone at Katakefalos ...... 35 Detail of gouge material at Katakefalos showing primary matrix ...... 36 Detail of gouge material at Katakefalos showing secondary seal ...... 36 Perspective and detail of breccia face at Katakefalos ...... 38 Examples of clasts of earlier breccia set in later matrix ...... 39 Sample location map of study area at Charasonas ...... 40 Oblique view and details of fault exposure at Charasonas ...... 41 Half-Rose diagram of vein trends at Charasonas ...... 42 Left-step at western contact with hematitic marble at Charasonas ...... 44 Equal-area stereoplot of groove rakes and related fault planes ...... 45 Grooves in footwall at northern end of fault exposure at Diapori ...... 47 Riedel shear zone at Diapori ...... 48 Polished face in breccia at Katakefalos ...... 50 2.25 Grooves in eastern breccia wall at Charasonas ...... 51 2.26 Grooves in western breccia wall at Charasonas ...... 52 2.27 Riedel shear zone at Charasonas ...... 53 2.28 Oblique view and detail of fault exposure at Mirties highland ...... 55 2.29 View along strike of fault exposure at Mirties ...... 56 2.30 View to NW over interior bays ...... 56 2.3 1 Structural interpretation of compiled slip data ...... 57
Chapter 3: Petrographic and Fluid-Inclusion Analysis
Microstructures in thin section from Diapori sample 27i ...... 59 Matrix and microstructures in thin section from Charasonas sample 42g ..... 59 Microstructures in thin section from Diapori sample 27b ...... 60 Mechanical twinning of calcite in marble clast (Diapori sample 2%) ...... 60 Fault gouge in thin section from Katakefalos sample 26e ...... 62 SEM image and spectral analysis from Diapori sample 27i ...... 64 SEM image and spectral analysis from Katakefalos sample 26e ...... 65 SEM image of subhedral calcite in cohesive matrix (Diapori sample 27i) ... 67 SEM image and spectral analysis from Charasonas sample 42h ...... 68 Depiction of fluid inclusions during analysis ...... 72 Phase diagram for pure water ...... 73 Fluid-inclusion data from Diapori sample 27i ...... 76 Fluid-inclusion data from Charasonas sample 42c ...... 77 Abstract ......
Faults on the Aegean island of Syros, part of the Cycladic archipelago southeast of mainland Greece, record the complex tectonic history of the Mediterranean region following blueschist metamorphism. The Cyclades are in the Attic-Cycladic crystalline belt, a series of high-pressure, low-temperature metamorphic core complexes derived from subduction between microplates during the Alpine orogeny. Beginning approximately 30-25 Ma, collision in the Mediterranean gave way to post-orogenic extension (Jolivet and Faccenna, 2000) and exhumation of the Aegean blueschist terranes by low-angle normal faults and detachments in a back-arc setting (Okrush and Brocker, 1990). Although Syros is famous for its exposures of high-grade metamorphic rocks, the island's high-angle normal faults with well-developed breccia zones have not been studied in detail. It is uncertain whether the high-angle normal faults are synchronous with or younger than the large-scale normal faults responsible for Aegean blueschist exhumation. Three high-angle faults are located along Syros' western coast at Cape Diapori, the northernmost tip of the island; the western end of Cape Katakefalos; and on the Charasonas headland near the Grotto of St. Stephanos. The faults are arranged en echelon, strike NW-SE, dip >60•‹,and have throws no greater than 30 m. All three breccia zones contain similar fault products, primarily equidimensional angular marble fragments in a cohesive, finely-ground carbonate matrix. No schist clasts are evident in the breccia. Field observations and petrographic descriptions provide the context for fluid-inclusion analyses of fault material from each of the breccia zones, which help constrain ambient conditions during faulting. Fluid inclusions in crystalline quartz from a fault-related vein at Diapori show homogenization temperatures of 140-150•‹C and freezing temperatures of -2.9 to -0.9"C; the latter suggest a fluid composition of 1.5-4.7 wt% NaC1. Fluid inclusions in crystalline quartz from breccia matrix at Charasonas contain at least three fluids with homogenization temperatures of 140-160, 200-210, and 220-240•‹C. Freezing temperatures of -5.9 to -5.4"C suggest a fluid composition of 8.5-9.0 wt% NaCl. Petrographic evidence suggests brittle deformation during faulting; brittle deformation tends to occur in low-temperature environments at shallow crustal levels. The three faults occur at or near the tips of the western peninsulas, and may be isolated parts of a NW-SE-trending graben with a W-dipping plane on its NE side and an E-dipping plane to the SW. Chapter 1
Introduction ......
I. The Problem
Although the framework for the metamorphic history of the Aegean Sea is well
constrained, the tectonic evolution of the region remains the subject of intense debate.
The Cycladic island of Syros is famous for its spectacular exposures of high-pressure
metamorphic rocks, and indeed, the island is a metamorphic petrologist's Elysium-its
eclogites are some of the most beautiful in the world, and Cape Pouli, just west of the
town of Kini, is the type locality of the blueschist-facies mineral glaucophane.
Exhumation of the Aegean blueschist terranes has been attributed to large, low-angle
normal faults and detachments in a back-arc setting (Okrush and Brocker, 1990). On
Syros, Kidley (1984a) described a network of post-metamorphic listric normal faults (Fig.
1.I) that he related to regional extension. Ridley (1984a) used evidence from neighboring
islands to bracket the late fault activity between the late Miocene to Pleistocene. The
normal faults on Syros have not been studied in detail, and they may be synchronous with
or younger than the large-scale normal faults responsible for Aegean blueschist
exhumation.
In this thesis I characterize three high-angle faults with well-developed breccia
zones along Syros' western coast: Cape Diapori, the northernmost tip of the island; the
western end of Cape Katakefalos; and on the Charasonas headland, near the Grotto of St.
Stephanos (Fig. I.I). A better understanding of the environment in which these faults
formed is necessary to determine their relative importance in the exhumation of the high- pressure rocks. Field observations and petrographic descriptions provide the context for Grammata Beach
See Fig. 2.2
Galissas Bay
Cape Katakefalos
7Grotto of St. Stephanos x Legend High-angle fauns (sludy focus) Low-angle normal faults' p/ (dip c30•‹) Moderate-angle normal faulls' (dip 30-80') Fregata \ High-angle normal fauns' $Beach . \ (dip >60•‹) ee Fig. 2.17 '... Inferred fault traces'
Figure 1.1. Map of Syros Island. Ridley (1984a) documented a network of faults on the island, including several low-angle listric normal structures. Insets show locations of the three study areas of this project-Diapori, Katakefalos, and Charasonas. At all three sites, high-angle faults with well-developed breccia zones strike NW-SE and demonstrate throws no greater than 30 m. The faults occur at or near the tips of the peninsulas, and may be part of a larger fault zone that controls the outer western coastline of the island. For detailed maps of the study areas at Diapori, Katakefalos, and Charasonas, see Figures 2.2,2.10, and 2.17, respectively. fluid-inclusion analyses of fault matrix material from each of the breccia zones. Quartz and calcite crystals in breccia matrix contain fluid inclusions that can be analyzed with a fluid-inclusion stage to provide temperature and composition data for some fault-related fluids, and help constrain ambient conditions during faulting.
11. Regional Geologic History
Fault structures on Syros derive from the complex tectonic history of the
Mediterranean region. Syros is in the Cyclades, an archipelago southeast of mainland
Greece in the Aegean Sea (Fig. 1.2). The Cycladic islands are part of the Attic-Cycladic crystalline belt and represent a high-pressure, low-temperature metamorphic terrane derived from subduction of several Mediterranean microplates during the Alpine
Orogeny (Fig. 1.3). Keay and Lister (2002) suggested that the continental blocks of the
Greek peninsula, including the Cyclades, once belonged to the Gondwanan margin. As the Tethys Ocean closed during the Triassic and early Jurassic, the Pelagonian microplate collided with the European continent (Fig. 1.3), marking the onset of the Alpine Orogeny at approximately 120 Ma (Bonneau, 1984; Turnell, 1987; Walcott and White, 1998). At one time situated between the Hellenide platform of Pelagonia and the Pindos Ocean
(Fig. 1.3), the Cycladic complex is a group of units that were subducted under the
Pelagonian margin and accreted below a pile of nappes caught between Apulia and the
European continent (Papanikolaou, 1984; Bonneau, 1984).
The Attic-Cycladic belt is comprised of two main tectonic units. The poorly exposed upper unit contains remnants of ophiolite sequences and mklanges, Permian to Turkey
Andros Cycladic Islands
Santorini Aegean Sea b
Figure 1.2. Syros is an island in the Cyclades, an archipelago SE of mainland Greece in the Aegean Sea. Syros resides in the northwestern section of the group, just west of the sacred island of Delos, around which the Cyclades are said to form a ring. Pigure 13. Pahgeographic cartoon of the central Aegean in the Upper Creta~eous.From Bonneau (1984). p. 524. Cretaceous sedimentary rocks, and some low-pressure, high-temperature metamorphic rocks and granitoid intrusions dated at approximately 70 Ma. The lower unit is polymetamorphic and consists of a stacked series of thrust sheets that include pre-Alpine basement rocks, Mesozoic marble, metavolcanics, and metapelites (Okrusch and Brocker,
1990).
Polymetamorphism in the lower Cycladic unit includes three events: Eocene
(approximately 45 Ma) high-pressure blueschist-facies metamorphism; Oligocene-
Miocene (32-1 5 Ma) medium-pressure Barsovian metamorphism; and Miocene (18-1 2
Ma) contact metamorphism associated with the intrusion of I-type granites (Keay and
Lister, 2002; Oksusch and Brocker, 1990). White mica K-Ar, A"-A3', and Rb-Sr dates from Cycladic blueschist give ages of 53-40 Ma, but a more recent U-Pb metamorphic zircon date from blueschist on Syros records 7821 Ma as the minimum age for eclogite- facies metamorphism (Brocker and Enders, 1998). The age difference between the zircon and mica suggests a complicated metamorphic history extending beyond the Teritary into the late Cretaceous, and raises the question of whether Aegean eclogites and blueschists developed during a single metamorphic cycle or several separate episodes (Brocker and
Enders, 1998).
Blueschist metamorphism occurred during the Alpine orogeny, which culminated when the African plate collided with Eurasia in the eastern Mediterranean at approximately 35 Ma. By 30-25 Ma, crustal shortening gave way to extension (Jolivet and Faccenna, 2000). It is unclear, however, whether younger greenschist metamorphism overprinted blueschist rocks during isothermal decompression or during a separate thermal pulse (Avigad, 1993; Avigad and Garfunkel, 1991; Boronkay and Doutsous, 1994; Avigad and others, 1997). The best-constrained dates of extension are limited to the Miocene, and derive from extensional structures evident in granitoid intrusions emplaced in the upper crust between 20-1 1 Ma (Avigad and Garfunkel, 199 1; Altherr and others, 1982, as cited in Boronkay and Doutsos, 1994; Keay and Lister, 2002).
Models for Neogene tectonic evolution in the Aegean stem from a single fundamental question: How do high-pressure, low-temperature blueschist-facies rocks move from depths of 50-75 km to the Earth's surface? Preservation of blueschists during uplift calls for rapid exhumation, either by erosion or tectonism. If the cold oceanic slab stops subducting, or if exhumation of the blueschists is slow, conditions in the upper mantle and crust cook the rare facies into a Barrovian regime. In the Aegean, erosion is an unlikely unroofing mechanism. Workers have shown that clastic sequences are too thin to account for the tens of kilometers worth of missing crust. Furthermore, the clasts were not derived from lower-middle crustal rocks (Avigad and Gasfunkel, 1991).
The unroofing is, therefore, a tectonic consequence of extension. Aegean blueschists, which derive primarily from subducted oceanic and accretionary-prism sediments, likely delaminated from overlying units during the early stages of exhumation
(Avigad and Garfunkel, 1991). During lithospheric subduction, rock layers in the upper crust of the subducting unit can shear and delaminate along rheologically weak zones.
Subduction of lithosphere continues until buoyancy overcomes rock strength. While the denser lower crust continues to subduct, the lighter upper crust reverses movement, experiencing uplift and rapid exhumation along the latent shear zones formed during collision. Such a style of exhumation can account for local greenschist overprinting of blueschist metamorphism, and explicitly links contractional tectonics and Neogene
extension in a back-arc basin (Wijbrans and others, 1993; Avigad Garfunkel, 1991).
Early regional models generated from scattered local data included a single low-
angle detachment that encompassed much of the Aegean (Fig. 1.4) (Lister, 1984, as cited
in Gautier and Brun, 1994). The wide range of shear-lineation trends throughout the
Cyclades has long suggested a scenario of pervasive crustal fragmentation and wrenching
(Keay and Lister, 2002; Ridley, 1984). Mediterranean reconstructions now combine evidence from a mosaic of detachments and metamorphic core complexes to describe
small-scale tectonic block rotations within a back-arc setting (Fig. 1.5) (Gautier and
Brun, 1994; Walcott and White, 1998).
The Cycladic archipelago exposes a varied pattern of normal fault trends (Fig.
1.6). Major faults appear to follow the curve of the Hellenic subduction zone, perhaps reflecting a pattern of radial extension or the progressive flexing and breaking of a previously continuous horst-and-graben system (Gautier and Brun, 1994). One of the principal structures is the Mid-Cycladic Lineament, a NE-SW-trending fault zone running from Ikaria to Paros. The lineament, a variable-style fault zone that is no longer active, formed prior to 14 Ma and at one time split the Aegean into eastern and western blocks (Gautier and Brun, 1994; Walcott and White, 1998). The Scutari-Pec Line, the mainland counterpart to the Mid-Cycladic Lineament, defined the western block's northern boundary (Walcott and White, 1998).
Tertiary kinematic evolution in the Aegean involved approximately three stages of microplate interaction (Fig. 1.5) (Walcott and White, 1998). First, NE-directed shear affected the entire region during the compressional phase of the Alpine Orogeny. Second, M 0
50
Lower plofe
Figure 1.4. A schematic geometric model for the exhumation of high-pressure, low-temperature terranes along a single large displacement, low-angle normal fault, without (a) and with (b) the influence of lithospheric isostatic rebound. From Avigad and Garfunkel (1991), p. 365. Figure 1.5. Kinematic interpretation of regional extension and important geologic features within the Aegean. From Walcott and White (1998), p. 170. Figure 1.6. Simplified local kinematics of the central Aegean. Gray arrows indicate sense of shear in various lithologies, including late Miocene plutons. Bars indicate stretching lineations and/or stretching directions. Ticks along main brittle fault traces point to the downthrown side. Areas shown in black represent Tertiary volcanics, and those in dark gray represent syn-extension intrusives. The black dotted line follows the Mid-Cycladic Lineament, marking the boundary between NCSW and N-S lineation orientations in the western and eastern Aegean blocks, respectively. From Walcott and White (1998), p. 164. the Aegean parted into two blocks along the Mid-Cycladic Lineament (Walcott and
White, 1998), approximately coeval with extension in the Cyclades at approximately 21
Ma (Brocker and Franz, 1998, as cited in Gautier and others, 1999). Gravitational
collapse of thickened crust (Fig. 1.7), which spread toward the eastern Mediterranean,
drove post-orogenic extension from before 21 Ma until the collision of Arabia into
Eurasia at approximately 16 Ma (Fig. 1.8) (Gautier and others, 1999; Boronkay and
Doutsos, 1994; Avigad and others, 1997). Extension in the central Aegean and
simultaneous shortening to the NW forced the western block to rotate clockwise (Walcott
and White, 1998; Gautier and Brun, 1994), while the eastern Aegean block, which was
comparatively less coherent, rotated counterclockwise and opened E-W-trending grabens
(Walcott and White, 1998). Finally, in the Pliocene (approximately 3 Ma), the western
Aegean block broke into two sections. Mainland Greece continued to rotate clockwise
until motion stopped along the Mid-Cycladic Lineament. The central Aegean then fused
into a single coherent block and began migrating SW toward the Hellenic subduction
zone. Cessation of motion along the Mid-Cycladic Lineament in the latest Miocene-early
Pliocene was approximately contemporary with the development of the left-lateral, transverse North Anatolian Fault in northern Turkey and the Dardanelles (Walcott and
White, 1998).
Two effects dominate modern deformation in the Aegean region: the
southwestward motion of the southern Aegean, and the westward motion of Turkey, both relative to Eurasia (Fig. 1.9) (Jackson, 1994). Present extension in the Aegean is related to the southward retreat of the Hellenic trench. Right-lateral motion along the North
Anatolian Fault accommodates the extrusion of Anatolia away from the Arabian-Eurasian I Gravity spreading of the I ARABIA Aegsan thickened lithosphere
j Arabia Indenthflon and i lateral extrusmn of Anatolia
Figure 1.7. Two-step summarization of kinematic evolution of the eastern Mediterranean. Extension induced by gravitational collapse of overthickened crust predates the indentation of Arabia into Eurasia and the lateral extrusion of Anatolia along the North Anatolian Fault. From Gautier and others (1999), p. 41. Figure 1.8. Regional structural features of the central Aegean. 1: northern Pelagonian zone; 2: Chalkidiki; 3: Rhodope; 4: northwestern Turkey; 5: Menderes Massif; 6: Crete; 7: Peloponnesos; 8: Attic-Cycladic crystalline belt. Dashed arrows show senses of rotation based on paleomagnetic data. Note Scutari-Pec Line on the Greek mainland, the northwestern counterpart to the Mid-Cycladic Lineament (Fig. 1.6). The thrust sf the Mediterranean Rise just south of Crete (6) is the Hellenic trench. From Walcott and White (1998), p. 156.
Figure 1.9. Simplified map of modern deformation in the Aegean region. Note the northward collision of Arabia, the westward extrusion of Turkey along the North Anatolian Fault (NAF), and the southwestward motion of the Aegean Sea toward the Hellenic Trench (PIT). From Jackson (1994). collision, which continues to force the Aegean block to the southwest (Gautier and
others, 1999; Mejier and Wortel, 1997). Though the margins of the Aegean Sea endure
frequent and intense seismic activity, the interior of the block is tectonically quiet
(Jackson, 1994).
111. Geology of Syros
Much of Syros is comprised of a continuous, alternating sequence of north-
dipping schists, marbles, and gneisses (Fig. 1.10). The island is pelitic to the south, but
thick calcite and dolomite marble strata with interbanded pelites dominate the northern
end (Ridley, 1984a; Brocker and Enders, 1998; Schumacher and Helffrich, 2003). At the top of the structural pile, near the island's northern extreme, variegated meta-igneous blocks sit within a serpentinite matrix. Field evidence suggests that the unit represents a tectonic mklange-a chaotic arrangement of exotic blocks from various crustal levels locked in an altered matrix. Metamorphosed breccia crops out near Kastri on the northeastern coast, and preserves unsorted blocks of meta-igneous rocks within a glaucophane matrix (Schumacher and Helffrich, 2003).
Blueschist-facies metasedimentary and meta-igneous units are predominant on
Syros (Ridley, 1984a). A greenschist-facies overprint of preexisting blueschists is evident on the southern end of the island and more variably to the north. Both the mklange and breccia exhibit high-grade metamorphic assemblages, indicating that the structural rearrangements that formed the features occurred prior to the peak of metamorphism
(Schumacher and Helffrich, 2003). Generalized Geology of Syros by N. Hijpfer (1997)
Figure 1.10. Simplified geologic map of Syros based on map compiled by Hecht (1977), but re-interpreted using data collected by students of Schumacher at Freiburg (1996-1999). From Schumacher and Helffrich (2003). Ubiquitous within the marbles of Syros are oriented, acicular calcite textures that
appear to be calcite pseudomorphs after aragonite. The mineral aragonite is the high-
pressure polymorph of calcite, and is stable under blueschist-to-eclogite metamorphic
conditions. Aragonite itself is rare in high-pressure marbles because it transforms to
calcite during exhumation. But the calcite pseudomorphs on Syros display the acicular habit of aragonite, and the island's marbles experienced peak deformation of 12-20 kb and 450-500•‹C, well within aragonite's stability field (Brady and others, 2003).
In a brief survey of the island's fault structures and strata, Ridley (1984a) equated
Syros to a north-dipping horst block. A number of brittle, late, post-metamorphic listric normal and high-angle secondary faults cut the island, consistent with proposed regional extensional tectonics (Fig. 1.1) (Ridley, 1984a). Almost all faults within the marble units contain similar products, primarily an even-sized breccia of equidimensional marble fragments in what Ridley (1984a) characterized as "a milled carbonate matrix" (Ridley,
1984a, 755). The major, low-angle faults show hanging-wall displacement to the east and west, away from the spine of the island, while steeper normal faults with little apparent offset dip toward the spine, characteristic of secondary faults that often occur in listric hanging walls (Fig. 1.11) (Hamblin, 1965, as cited in Ridley, 1984a). The high-angle faults, which dip >60•‹,tend to strike to the north and northwest; the low-angle faults, which dip <30•‹,occur over a range of strike directions. The high-angle faults always truncate subhorizontal faults -presumably the basal detachments along which the primary listric blocks rotated (Fig. 1.12) (Ridley, 1984a). Figure 1.11. West-to-east cross-section through north Syros. Note listric normal faults with downthrow mvay from central spine, and steep, secondary, antithetic faults in the hanging wall with downthrow towardcentral spine. From Ridley (1984a), p. 758.
Figure 1.12. View W, toward high-angle normal faults in the Halara highland. Stratigraphic offset between schist (dark) and marble (light) units provides a clear sense of displacement at the faults, which are representative of the post- metamorphic, brittle, listric-normal and high-angle secondary faults that occur in the hanging walls of low-angle listric faults (Ridley, 1984a). The high-angle faults on Syros may have been active during the late stages of
Aegean blueschist exhumation. Alternatively, the high-angle faults may have formed well after blueschist exhumation, perhaps in response to complex microplate rotation.
IV. Normal Faulting and Exhumation
In their seminal paper entitled "Modes of Extensional Tectonics," Wernicke and
Burchfiel (1982) offer a compilation of characteristics and phenomena associated with extended terranes. Large, low-angle detachments tend to underlie highly extended terranes, and often contain listric and planar normal faults within their hanging walls.
Extensional structures are either rotational or non-rotational. Non-rotational faults and more complex variants have planar geometries in which beds translate along a stationary fault. In rotational faults, both the faults and blocks rotate, and their geometries can either be planar or listric. Listric geometries are defined by extension between two blocks that occurs along a curved surface (Wernicke and Burchfiel, 1982).
High-angle normal faulting typically occurs along non-rotational planar faults. In domino-style normal faulting, planar faults rotate to progressively shallower dips while the originally flat-lying strata rotate to increasingly steeper dips. Listric normal faulting requires a curved fault plane and tends to involve collapse and reverse-drag in the hanging wall (Fig. 1.13). Antithetic planar faults may also propagate in the listric hanging wall. Imbricate listric normal faulting is essentially a domino-style collapse along curved fault planes (Davis and Reynolds, 1996). Figure 1.13. Schematic geometry of imbricate listric normal faults, after Wernicke and Burchfiel(l982). From Ridley (1984a), p. 758.
Figure 1.14. Unroofing a metamorphic core complex along a low-angle normal detachment. Note formation and imbrication of subsequent high-angle faults in the listric headwall. From Davis and Reynolds (1996), p. 353. The distinguishing characteristic of low-angle normal faults is the juxtaposition of younger strata on older, but with the omission of rock units. (Thrust faulting, by comparison, results in unit repetition.) The kind of large-scale extension responsible for tectonic exhumation of metamorphic terranes is the result of paired modes of detachment: large displacement on low-angle non-rotational faults; and secondary displacement on planar and listric faults, along which fault-bounded blocks and faults have rotated
(Wernicke and Burchfiel, 1982). Figure 1.14 shows low-angle detachment faulting and the formation of a metamorphic core complex, such as the Cyclades, and the progression from a large low-angle basal detachment to an imbricated, extended range and ultimate exhumation of crustal basement (Davis and Reynolds, 1996). Chapter 2
Field Observations
I. Location Descriptions
High-angle faults at Diapori, Katakefalos, and Charasonas strike NW-SE (Fig.
2.1) and have well-developed breccia zones that contain primarily equidimensional,
angular marble fragments in a cohesive, finely-ground carbonate matrix. The clast-rich
compositions (10-90%) of the fault rocks fall into textural classifications spanning crush
breccia to cataclasite (Table I). The faults dip >60•‹and have throws no greater than 30 m.
Table I. Textural classifications of fault gouge. Modified from Sibson (1977).
Percent Cohesive Fragment Size Classification Crystalline Matrix (cm) crush breccia 0-10
fine crush breccia 0-10 1 crush microbreccia 0-10
protocatalasite 10-50
cataclasite 50-90
Fieldwork involved describing and collecting samples from each site,
documenting lithologies and contact relationships, measuring fault orientation, jointing, bedding planes, foliation, and groove rakes, and conducting a general survey of fault
material for clast type, size, roundness, and percent matrix. Figure 2.1. Equal-area stereoplot of dominant fault planes at Diapori (dashed), Matakefalos (bold), and Charasonas (fine). Note that all planes are steeply dipping and trend NW-SE or N-S. Cape Diapori
The fault exposure at Cape Diapori, the northernmost headland of the island's western coastal peninsulas (Fig. 1.I), is approximately 200 m long and strikes NW-SE along three en echelon steps (Fig. 2.2), forming a ragged, west-dipping, near-vertical scarp approximately 50 m high (Fig. 2.3).
At the north end of the fault zone, a sharp contact between marble and fault gouge
(Fig. 2.4) extends northwest for approximately 20 m before disappearing down a cliff of jointed, fractured marble. Along strike to the southeast, fault expression splays into highly fractured marble with isolated breccia lenses. The fault gouge is clast-rich (80% clasts, 20% matrix), with an orange-tan, calcareous, cohesive rock-flour matrix. Clast lithology is uniformly marble, with fragments typically 1-3 cm. The host marble at
Diapori is graphitic gray, and contains aragonite pseudomorphs-rods of acicular calcite with a shape-preferred orientation similar to the habit of aragonite (Brady and othcrs,
2003). Where aragonite pseudomorphs are evident in marble clasts in the breccia, the rods are randomly oriented from clast to clast.
The dominant feature at Diapori is a looming breccia face approximately 70 m long and 15 m high (Fig. 2.3). The northern end of the wall rises out of tan-colored quartz-glaucophane schist that includes cross-cutting veins of travertine, white crystalline calcite seal (Fig. 2.5), and pods of crystalline quartz (Fig 2.6). The schist has a foliated shear fabric but is not brecciated. Indeed, no schist clasts are evident anywhere in the breccia wall-clasts are exclusively graphitic marble and generally no larger than 6 cm in diameter. Like the northernmost echelon step, the main breccia face is clast-rich (55-80% clasts, 2045% matrix), with aragonite pseudomorphs randomly oriented among different shattered marblc eccia lenses rn ;ch and vein material (crystallir calcite, quartz)
Figure 2.2. Sample location map of study area at Diapori. Labeled black-rimmed circles indicate sample numbers and locations. The two ellipses mark the samples used in petrographic analysis (see Chapter 3). The northern echelon strikes 148" and dips 70" W; the main breccia face strikes 155" and dips 55" W; and the southern fault segment strikes 155" with a vertical dip. Island (schist point) /
4 schist saddle (partial hanging wall) ...., A- -. - IC *. d Figure 2.3. Oblique view and details of fault zone at Diapori. (A) View to NE of main breccia face and southern terminus of the fault zone. Arrow denotes schist cliff and crystalline calcite vein cascade (photo: T. Harms). (B) Detail of goat cave at southern terminus of main breccia face, where the dominant fault trace disappears. View to NE. Hammer for scale (photo: T. Harms). (C) Detail of main breccia wall, view to NE. The exposure is approximately 70 m long and 15 m high, with a strike and dip of 155" 55"W. Figure 2.4. View SE of sharp, near-vertical contact between host marble and fault gouge at the north end of Daipori fault zone. The contact, which strikes and dips 148" 7WW, extends NW for approximately 20 m before disappearing down a cliff of jointed, fractured marble. Upstrike to the SE, the fault expression splays into highly fractured marble with isolated breccia lenses (Fig. 2.2). Hammer for scale (photo: T. Harms). Figure 2.5. Detail of marble and schist blocks set in white crystalline calcite seal and travertine veining, located at .the northern end of the major breccia wall at Diapori (Fig. 2.2). Hammer for scale (photo: T. Harms).
Figure 2.6. Pod of crystalline quartz in tan-colored quartz-glaucophane schist at the northern end of the major breccia wall at Diapori, from which sample 27b was taken (Fig. 2.2). The quartz sample proved integral to the fluid- inclusion analyses described in Chapter 3, Part 11, and summarized in Fig. 3.12. Hammer for scale (photo: T. Harms). 29
clasts (Fig. 2.7). An indistinct vertical color band approximately 10 m wide appears in the
middle of the wall. The band is darker and more blue-tan than the surrounding fawn-
colored exposure, but color seems to be the only material difference. The dark section is
similarly clast-rich (50-80% clasts, 20-50% matrix) and has the same clast-size
distribution as the rest of the wall.
The southern end of the fault zone is perhaps the most complex. Four principal features meet at the fault terminus: the southern tail of the major breccia wall; the massive, highly fractured host-marble unit; a west-dipping schist saddle; and a brilliant, white cascade of crystalline vein calcite.
Midway along the breccia face, the orientation of the hanging-wall/footwall contact is obvious: The breccia is plastered against the massive marble unit of the footwall, and the schist unit underfoot is the down-dropped hanging wall. At its southernmost extent, however, the breccia face tapers away and disappears into tilted blocks of fractured marble. The vertically arcuate, diminishing exposure of the fault is visible in Figure 2.3 a. At the goat cave noted in Fig. 2.8 and shown in detail in Fig. 2.3b, host marble sits directly atop schist. Though the hanging-wall-to-footwall contact is plain a few meters to the north, the fault trace ends at the goat cave-one must look up and imagine it arcing overhead.
Where the breccia wall terminates, however, the fault jumps west toward the crystalline calcite vein cascade (Fig. 2.9), the third of its echelon steps. The calcite exposure has a ropey, fibrous texture, and shows layered bands of travertine-like crystallization. The cascade includes isolated blocks of host schist (10-15 cm) and open S'i78" goat cave (see Fig. 2.3b)
Figure 2.8. View to S toward southern terminus of main breccia wall at Diapori. The breccia wall tapers away and diappears into tilted blocks of fractured host marble. The gouge material is plastered to the massive marble unit of the footwall; the adjoining schist platform comprises the down-dropped hanging wall. Near the goat cave (Fig. 2.3), the fault steps W toward a cascade of white, crystalline calcite veining (Fig. 2.9). Author for scale (photo: T. Harms). -+wv '4 -&*p+' ,- cut ro..
Figure 2.9. View NW from goat cave (Fig. 2.8) toward the schist saddle and calcite vein cascade near the southern terminus of the main breccia face at Diapori (Fig. 2.2). The calcite exposure has a ropey, fibrous texture, and shows layered bands of travertine-like crystallization. The cascade extends NW for approximately 50 m along and down a schist cliff 10-15 m high. Pack and hammer for scale. spaces as large as 4-5 cm in diameter, and extends for approximately 50 m along and
down a schist cliff 10-15 m high.
Gray-blue, glaucophane-rich schist forms a saddle between the breccia wall and
calcite veining, rising to the south to meet the graphitic marble unit at the goat cave (Figs.
2.8 and 2.9). The schist ends in a cliff where it meets the calcite, but is visible in the bulk
of Diaporakia Island, a massive schist block that tilts out of the sea less than 100 m to the
west (Fig. 2.3a). Diaporakia may be a detached part of the hanging wall once connected
to the footwall of the dilatant calcite zone.
Cape fitcekefalos
Cape Katakefalos extends west from the town of Galissas and shelters Galissas
Bay to the south (Fig. 1.1). A steep, east-dipping, NW-SE-striking normal fault segment
approximately 200 m long (Fig. 2.10) cuts the promontory near its wcstern limit, forming the topographic hump that lends the peninsula a camel-backed profile (Fig. 2.11).
Because white-pink hematitic marble sits atop green quartz-mica schist along much of the cape, stratigraphic offset at the fault provides a clear sense of displacement.
The hanging wall, on the eastern side of the fault trace, has slipped such that the upper marble unit now lies adjacent to schist in the footwall (Fig. 2.12). The height of the contact suggests a throw of at least 20-30 m.
Three types of non-clast material are evident in the fault gouge: an orange-pink, finely ground, cohesive, calcareous rock-flour (Fig. 2.13); a coarse-grained, crystalline calcite vein seal that forms thin rinds around clasts (Fig. 2.14); and a white calcareous Figure 2.10. Sample location map of study area at Katakefalos. The fault trace splits near the northern terminus. The west splay strikes 335" with a vertical dip, the east splay 357" with a vertical dip. Labeled black-rimmed circles indicate sample numbers and locations. Ellipse marks the breccia sample used in petrographic analysis (see Chapter 3). Figure 2.1 1. View NW toward Cape Katakefalos. Arrow shows where the steep, east-dipping NWSE striking normal fault (Fig. 2.10) cuts the promontory and forms a topographic hump that lends the peninsula its camel-backed profile.
Figure 2.12. Fault zone at Katakefalos, view to E. Footwall schist of the W side of the fault is juxtaposed against marble in the hanging wall to the E, a stratigraphic offset that provides a clear sense of displacement. The height of the contact suggests a throw of at least 20-30 m. The fault strikes NWSE and dips 70"-90" east. Grooves in a polished face (Fig. 2.24) suggest oblique slip to the NE (Fig. 2.21). Figure 2.13. Detail of gouge material at Katakefalos. Primary, cohesive, cataclastic breccia matrix with orange-pink hematitic staining typifies 60-80% of the matrix in the fault exposure. Clasts entrained in the breccia are exclusively marble, angular, and typically a3cm in diameter, though larger cobbles (Acm) are locally present (photo: T. Harms).
Figure 2.14. Detail of gouge material at Katakefalos showing secondary, coarse- grained, crystalline calcite seal. The vein calcite tends to form thin rinds around clasts, and is one of three kinds of non-clast materials evident in the breccia wall (photo: T. Harms). 37 precipitate confined to joint fractures. The orange-pink rock-flour is the dominant matrix type, comprising approximately 60-80% of the matrix in the fault exposure.
Though the base of the breccia face is in direct contact with the schist unit of the hanging wall (Fig. 2.15), no schist clasts are evident in the gouge. Clasts are exclusively marble, angular, and typically <3 cm in diameter, though larger blocks (>6 cm) are locally present. Aragonite pseudomorphs are randomly oriented from clast to clast.
Among the fault products in the central section of the wall are clasts of earlier breccia, visible in hand sample and thin section, indicative of at least one fault-reactivation event
(Fig 2.16). The rebrecciated clasts typically contain dark fragments of marble within a white, chalky, calcareous matrix that contrasts against the orange-pink color of the primary gouge matrix. The weathered appearance of the older breccia clasts perhaps reflects some chemical alteration during subsequent faulting episodes.
Charasonas Headland
Two east-dipping, near-vertical subparallel breccia faces 10-15 m apart frame the fault zone at Charasonas (Fig. 1.I), which extends for approximately 60 m in a NW-SE direction (Fig. 2.17). Marble lies on the western side of the fault, schist on the eastern side (Fig. 2.18). The marble contains 1-2 cm hematite- and graphite-rich compositional bands, and bears aragonite pseudomorphs approximately 1 cm long. The schist bears a heavy greenschist overprint, and is crosscut by calcareous crack-and-seal veins, and pods of crystalline quartz, some of which are tens of cm in diameter. The veins are seal-rich, breccia poor, and tinted orange with oxidized iron. Sixteen veins measured in the greenschist belt strike between 345" and 30" (Fig. 2.19). Figure 2.15. Perspective and detail of the breccia face at Katakefalos. (A) View to NW, upstrike, of the fault exposure from the southern terminus. The breccia face (center) is plastered against a hematitic marble unit on the western side of the fault (right) comprising the footwall. To the E, glaucophane-bearing quartz-mica schist comprises the down-dropped hanging wall (photo: T. Harms). (B) Detail of breccia wall. Though the base of the breccia face rests in direct contact with the schist unit of the hanging wall, no schist clasts are evident in the gouge. Clasts are exclusively marble (Fig. 2.13). Figure 2.16. Examples of clasts of earlier breccia set in a later matrix, indicative of at least one fault reactivation event. The rebrecciated clasts typically contain dark fragments of marble within a white, chalky, calcareous matrix that contrasts with the orange-pink color of the later gouge material (photos: T. Harms). Figure 2.17. Sample location map of study area at Charasonas. Labeled black- rimmed circles indicate sample numbers and locations. The three ellipses mark the samples used in petrographic analysis (see Chapter 3). breccia wal
Figure 2.18. Oblique view and details of fault exposure at Charasonas. (A) View to W of fault zone, including the two subparallel breccia faces, medial section of hydrothermally altered matrix material, and eastern schist unit. Breccia faces at the eastern contact are 2-5 m high (photo: T. Harms). (B) View to W of eastern breccia face. Note the oxidized reds, oranges, and yellows visible in the cataclastic matrix. Arrow shows perspective of photo in Part C. (C) View to NW (along strike) of a section of the eastern breccia wall. The exposed face of the gouge is notably dark-in parts of the eastern breccia wall adjacent to the schist unit, the matrix has a distinct purple-green color not evident at the marble contact to the west. Vein Trends in Greenschist, St. Stephanos N= 16 Class Interval = 10 degrees Conf. Angle = 14.09 Maximum Percentage = 3 1.2 R Magnitude = 0.885 Mean Percentage = 16.67 Standard Deviation = 8.97 Rayleigh = 0.0000
Figure 2.19. Half-Ro diagram of calcite- and quartz-vein trends fro unit on the eastern s of the Charasonas fault zone. Sixteen veins showed measurements between 3 30•‹,with an average strike just east of n Between the two faces is a zone of hydrothermally altered matrix material and shattered host marble (Fig. 2.18). The predominant matrix is a fine-grained, oxidized, calcareous rock-flour that crops out in vivid reds, oranges, and purples. In parts of the eastern breccia wall adjacent to the schist unit, the matrix has a distinct purple-green color not evident at the marble contact to the west, and is likely the result of hydrothermal alteration occurring in the schist (Fig. 2.1%). Nodules of crystalline quartz occur throughout the gouge, commonly as subhedral clusters as wide as 10-15 cm. The fault gouge in both breccia faces is more matrix rich, breccia poor (50-90% matrix,
10-50% clasts), than the exposures at Diapori or Charasonas, but contains the same angular marble clasts <3 cm in diameter, with a scattering of larger fragments (>6 cm).
The green-colored sections of matrix apparently contain no clasts at all, but do include subhedral quartz crystals approximately 0.5 cm in size.
The contact between the fault zone and the marble jogs less <0.5 m in several locations. At each small-scale step, the marble was shattered into rectangular blocks that were later mortared together by calcareous seal (Fig. 2.20).
11. Slip Indicators
I found indications of slip at all three study areas. Polished faces in the breccia walls held slickensides, grooves scraped into the footwall as the hanging wall slid down the fault plane. Slickenside orientations suggest the direction of movement of one block relative to another (Fig. 2.21). Figure 2.20. Westward step along the western contact with hematitic marble at Charasonas (Fig. 2.17). Top of photo is oriented NW. The marble was shattered into rectangular blocks and later mortared together with oxidized calcareous cataclastic matrix-note the vibrant reds and oranges in the gouge material at right. The step shows no more than 30-50 cm of lateral displacement. Figure 2.21. Equal-area steroplot of groove rakes (dots) and related fault planes at Diapori (dashed), Katakefalos (bold), and Charasonas (fine). Groove axes occur parallel to the direction of displacement. Grooves in west-dipping planes at Diapori show oblique slip to the south, while grooves in east-dipping planes at Katakefalos and Charasonas show oblique slip to the north. 46
Riedel-shear structures were also evident at Diapori and Charasonas. Fault zones that consist of multiple faults tend to include smaller faults that either link the main breaks or branch into splays. When two subparallel faults within the same zone are simultaneously active, such as in an en echelon construction, conjugate shear fractures can form at an angle to the principal fault planes. Riedel shears are a specific type of subsidiary fractures that form along strike-slip boundaries. As two blocks slide past each other, the first failures to develop are short shear fractures inclined at an angle to the principle plane of movement. Such Riedel, or "R-shears," occur in discrete, conjugate pairs (R and R'). With progressive shear, a third set of fractures, known as P-shears, develops to link the R- and R'-shears and thus establish the throughgoing fault trace (Van der Pluijm and Marshak, 2004). Riedel shears themselves form en echelon and tend to occur in zones of larger-scale en echelon faulting (Davis and Reynolds, 1996).
Cape Diapori
Grooves in the west-dipping host marble and breccia in the footwall at the northern end of the fault exposure at Diapori (Fig. 2.22) plunge to the south (Fig. 2.21).
Nearby to the west, a small-scale set of Riedel shears fractured into the host marble suggest movement of the hanging wall, or at least a local section of it, to the southeast
(Fig. 2.23). The combination of grooves and Riedel shears indicate oblique slip of the hanging wall to the south. Figure 2.22. View NE showing grooves in the footwall at the northern end of the fault exposure at Diapori. Grooves trend 169" and plunge 30" S (photos: T. Harms). Figure 2.23. View SE (top of photo) of subhorizontal Riedel shear zone in marble at the northern end of the fault exposure at Diapori. Black lines and relative motion arrows superimposed on the photo illustrate the actual fracture pattern, with the simplified schematic diagram of Riedel shearing (at right) for comparison. Pattern suggests movement of the hanging wall (right side of photo) to the SE. R and R' denote conjugate Riedel shears, P marks progressive shears. Simplified pattern after Groshong (1988). Cape Katakefalos
No Riedel shears were evident at Katakefalos, but one polished face in the east-
dipping breccia wall (Fig. 2.24) yielded grooves that plunge steeply (63") to the north
(Fig. 2.21), suggesting a slightly oblique slip of the hanging wall to the north.
Charasonas Headland
Grooves were evident in both the east-dipping eastern and western breccia walls
at Charasonas. In the eastern wall, the grooves (approximately 30 cm long) are clearly
defined (Fig. 2.25) and plunge 40" to the north. Slickensides on the western wall occur
over a larger area but are comparatively less distinct (Fig. 2.26), and plunge 55" to the north. A few meters south of the western face, a set of Riedel shears fractured into the host marble suggests northward movement of the eastern block relative to the western unit (Fig 2.27). In combination, the grooves and Riedel shears indicate oblique slip of the hanging wall- the eastern unit- to the north.
111. Discussion of Location and Slip Data
The faults at Diapori, Katakefalos, and Charasonas occur at or near the tips of their respective peninsulas, and may be parts of a larger fault zone that controls the shape of the outer western coastline of the island. The three faults strike en echelon, and therefore do not represent a continuous, large-scale, high-angle fault along the length of Figure 2.24. View SW showing grooves in polished face of breccia wall at Katakefalos. Grooves trend 327" and plunge 63" N (photo: T. Harms). Figure 2.25. View W showing grooves at the base of the eastern breccia wall at Charasonas. Grooves trend 355" and plunge 40" N. Figure 2.26. View W showing grooves in western breccia wall at Charasonas. Grooves trend 355" and plunge 55" N. Figure 2.27. View NW (top of photo) of subhorizontal Riedel shear zone in marble on the western side of the fault exposure at Charasonas. Black lines and relative motion arrows superimposed on the photo illustrate the actual fracture pattern, with the simplified schematic diagram of Riedel shearing (at right) for comparison. Pattern suggests movement of the hanging wall (right side of photo) to the NW. R and R' denote conjugate Riedel shears, P marks progressive shears. Simplified pattern after Groshong (1988). the western coast. Given a fault at least 15 km long-the approximate distance from
Diapori to Charasonas-one would expect throw greater than 30 m, the extent of offset at
Katakefalos, where the maximum displacement is evident. Furthermore, fault materials are of local origin, pulled from the immediate wall units-I found no exotic clasts in the breccia that were brought up from great depth and thrown into the fault zone. Indeed, homogeneous clast lithology may be a function of the physical properties of the wall rocks. Schist, with its malleable fabric, is more likely to flex and warp to accommodate stress, whereas brittle marble is prone to snap and shatter. Regardless, movement along a
15 km plane would likely show more unconformable juxtapositions between the footwall and hanging wall.
Although the three exposures may not be separate expressions of the same fault trace, they might be isolated parts of a single structure. I examined a fourth high-angle fault exposure, with similar characteristics to the other three, located on the south side of the Mirties highland (Fig. 1.1 and Fig. 2.28). The fault on Mirties strikes NW-SE and dips west, much like the Diapori fault approximately 5 km to the northwest (Fig. 2.29).
The faults at Katakefalos and Charasonas, however, dip east. Grooves and Riedel shears at Diapori indicate oblique slip of the hanging wall to the south; similar features at
Katakefalos and Charasonas indicate oblique slip to the north. Additionally, the major promontories at Diapori and Katakefalos frame a linear series of inland bays (Fig. 2.30).
The pattern that emerges is that of a NW-SE-trending graben block with a west-dipping plane on its northeast side and an east-dipping plane to the southwest (Fig. 2.31). The faults examined in this study may be limited, peripheral manifestations of the larger structure, or the graben may be incompletely formed and therefore experienced little displacement anywhere along its active perimeter. Figure 2.28. (A) View N, toward S side of the Mirties highland. The fault extends for approximately 100 m on a NWSE strike, with a consistent 50" dip to the W. A unit of host marble preserves the breccia face, while the base of the exposure marks a sharp contact between schist and marble units. The orientation of the fault zone at Mirties is similar to that at Diapori, approximately 5 km to the NW. (B) Breccia exposure at Mirties. Fault material consists of primarily equidimensional, angular marble fragments in a primary, cohesive, milled carbonate matrix and an apparently secondary crystalline calcareous matrix. Figure 2.29. W-dipping fault exposure at Mirties, view SW along strike (approximately 155" 5O0W). Vegetation visible in the upper right marks the contact between the bare marble of the footwall and overgrown schist units of the hanging wall.
Figure 2.30. View NW over Lia Beach, Marmari Bay, and Cape Grammata (south to north, respectively). Interior bays may be part of a larger fault zone, such as a NW-trending keystone graben, that controls the outer and inner western coastline of the island. Legend
Possible grabben + Grabben walls (hypothetical) Low angle faults' \ (dip 40deg.) Moderate angle faults' \ (dip -0 deg.) High angle faults' \ (dip >6O deg.) 'adapled from Ridley (1984a), p. 756.
direction of t tensile stress
A A' (schematic)
Figure 2.31. One interpretation of the slip data from Diapori, Katakefalos, and Charasonas, and the Mirties highland (Figs 1.1,2.1, and 2.21). The four exposures may be isolated parts of a single structure, such as a NW-SE-trending graben block with a W-dipping plane on its NE side and an E-dipping plane to the SW. Red arrows exaggerate the left-lateral oblique movement that the rake data suggest. Chapter 3
Petrographic and Fluid-Inclusion Analysis ......
I. Petrography
Thin Sections
Thin-section analysis revealed microscopic stylolites in finely-ground, calcareous fault material from Charasonas (sample 27i), providing evidence for syndeformational dissolution and diffusion. The stylolites occur as ragged, dark-fringed seams within the matrix, and are commonly normal to veins of clear crystalline calcite (Fig. 3.1). Another sample (42g) shows extensional vugs in the cohesive matrix where the material has split and then filled with a clear, crystalline calcite (Fig. 3.2). Areas of matrix around the vugs include patches of optically continuous calcite, which together with the calcite crystals suggest pervasive fluid movement within the fault rocks.
Crystal plasticity records strain accumulation according to movement along lattice dislocations and mechanical twinning (Knipe, 1988). Fractured, crystalline quartz from
Diapori (sample 27b) and Charasonas (sample 42c) contain inclusion-rich bands of deformation lamellae (Fig. 3.3). At high temperatures and in the presence of a reactive pore fluid, microfractures in quartz tend to heal (Groshong, 1988). Open microfractures, therefore, suggest that the mineral is young and has remained at relatively low temperatures. Calcite fragments from breccia samples (27i, 42h) show complex, contorted twinning that was more likely produced during deformation than growth (Fig.
3.4).
Fault matrix from all three locations is all but opaque under a petrographic microscope. Held up to a light source, the slides are brilliantly colored in yellows, blue- Figure 3.1. Photo of doubly-polished thin section from sample 27i, fault gouge material from Diapori, showing stylolites, calcite veins, angular marble clasts, and cohesive cataclastic matrix.
Figure 3.2. Photo of doubly-polished thin section from sample 42g, fault gouge material from Charasonas, showing fine-milled, iron-oxide-stained matrix and extensional vugs filled with clear crystalline calcite. Figure 3.3. Photo of doubly-polished thin section from sample 27b, crystalline quartz from Diapori, showing deformation lamellae and open microfractures.
Figure 3.4. Photo of doubly-polished thin section from sample 27i, fault gouge material from Diapori, showing mechanical twinning of calcite in marble clast. reds, and pinks. But the finely-ground nature of the matrix makes the material impenetrable at a small scale. Under high magnification, angular, tinted fragments (<1 mm) are visible around some vein lenses-elsewhere, however, the fragments are stacked and intercalated, complicating closer observation.
The sample from Katakefalos (26e) includes visible platelets of hematite that coat tiny grains of calcite (
SEna Analysis
To make a closer examination of the matrix material in the fault gouge samples, I analyzed uncoated thin-section billets in a scanning electron microscope (SEM). The microscope -an FEI Quanta 480 -provided high-magnification, high-resolution imaging. Mounted on the scope platform, a Prism 2000 Princeton Gamma-Tech electron- dispersive spectrometer (EDS) offered a qualitative spectral analysis of the elements present in each sample.
* Cape Diapori: Sample 27i
Sample 27i came from the dark band of fault gouge in the major breccia face at
Diapori. Thin section analysis revealed stylolites and clear calcite veining, but little primary cataclastic matrix I
Figure 3.5. Photo of doubly-polished thin section from sample 26e, fault gouge material from Katakefalos, showing clasts of earlier breccia in a later (primary), hematite-stained cataclastic matrix. insight into the bright yellow matrix. High magnification in the SEM, however, showed the habit of the matrix to be cataclastic and cohesive, with rock fragments 1-10 ym in diameter set in a crystalline cement (Fig. 3.6a). The cement consists of subhedral calcite, exhibiting perfect cleavage and short prisms with prominent faces. Qualitative spectral analysis shows high calcium (Ca) a and 13 peaks and smaller peaks for iron (Fe), signature of marble with iron-oxidation staining (Fig. 3.6b).
Cape Katakefalos: Sample 26e
Sample 26e came from gouge with rebrecciated clasts at Katakefalos. Though I was not able to distinguish among primary cataclastic matrix (hematitic pink in hand sample), earlier breccia matrix (white), and marble fragments with the SEM, I discovered that the sample on the whole contains significantly more open space than was evident in the field. High magnification shows the sawn surface of the sample pockmarked with holes, most of which contain clusters of euhedral calcite crystals (Fig. 3.7a). The crystals are distinctly rhombohedra1 with the cleavage forms e{0112) and more uncommon r(1011) (Klein and Hurlbut, 1993). As in sample 27i, qualitative spectral analysis shows high Ca a and B peaks, with smaller peaks for Fe, the expected pattern for marble with iron-oxidation staining (Fig. 3.7b).
a Charasonas Headland: Samples 42g, 42h
Sample 42g, which came from an area of yellow matrix between the eastern and western fault faces at Charasonas, appeared dark to opaque in thin section with the exception of clear crystalline calcite that had grown in isolated vugs. Under the SEM, the
Figure 3.7. (A) SEM image of euhedral calcite crystals in sample 26e from fault gouge material at Katakefalos. Image captured using solid-state detector (SSD) at a high voltage (HV) of 25.0 kV. (B) Qualitative spectral analysis for sample 26e generated from area shown in Part A. matrix appeared highly crystalline, with a massive arrangement of subhedral calcite prisms and little evidence of cataclastic disruption (Fig. 3.8). Qualitative spectral analysis showed Ca a and 13 peaks, prominent Fe peaks, and a magnesium (Mg) spike, perhaps indicative of Mg-substitution for Ca or a more dolomitic host marble with iron-oxide staining .
In the field, sample 42h, taken from the eastern breccia wall near the schist contact, shattered under my hammer, suggesting the presence of quartz in the matrix.
Thin section analysis was inconclusive-the matrix, though colorful, is opaque. High- magnification investigation in the SEM, however, revealed a matrix consisting of subhedral to euhedral crystalline quartz (Fig. 3.9a). Hexagonal crystals 3 pm in diameter have grown around a finely-ground cataclastic matrix of quartz shards, larger quartz fragments, and occasional clasts of marble. Qualitative spectral analysis shows a high silica (Si) peak, low Ca a and 13 peaks, and two prominent but apparently anomalous peaks at sodium (Na) and chlorine (Cl) (Fig. 3.9b). I considered various explanations for the presence of halite in the sample, but concluded that the salt is likely a byproduct of the spray-on carbon coating used to prepare the samples for the SEM. I analyzed samples
27i and 26e with the carbon coating face-down on the stage, away from the electron beam, so that only uncoated rock surfaces were exposed. For samples 42h and 42g, however, (the latter also shows two minimal Na and Cl peaks) the carbon-spray treated sides were facing the beam. Salt contamination does not occur with standard carbon coating. The adhesive in the spray necessarily includes chemical components other than pure carbon.
Figure 3.9. (A) SEM image of euhedral quartz crystals in sample 42h fault gouge material at Charasons. Image captured using an Everhart-Thornley detector (ETD) at a high voltage (HV) of 25.0 kV. (B) Qualitative spectral analysis for sample 42h generated from area shown in Part A. Significance of Petrographic Data
The presence of euhedral crystals preserved in open spaces in the fault gouge at
Katakefalos and Charasonas -calcite and quartz, respectively -suggests that late fluids
moved through both fault zones after motion along the faults had stopped. Had the fluid
crystallized before the faults fell dormant, the euhedral crystals would have been ground
into cataclastic fragments like those in the gouge around them. In both localities, the
crystals represent at least one fluid event. If the crystals are syndeformational, they may
represent the last of the fluids to have flushed through the respective fault zones before
quiescence.
II. Fluid Inclusions
An Introduction to Fluid Inclusions
Quartz and calcite crystals in breccia matrix contain fluid inclusions that can be analyzed with a fluid-inclusion stage to provide estimates of depth and temperature of matrix formation. The analyzing apparatus is a heating and cooling stage, mounted on a petrographic microscope, used to measure the temperatures of phase transitions in individual fluid inclusions. The composition of the fluid trapped in the inclusions can also be determined from phase-transition temperature data (Hollister, 2003).
The primary assumptions in fluid inclusion work are that the composition and density of the fluid in an inclusion has not changed since the inclusion formed, and that the fluid is homogeneous (Hollister, 2003). Because petrologic fluids are typically trapped at higher than ambient temperatures, the fluid within an inclusion undergoes
phase separation with cooling. Two fluid phases will thus occur simultaneously inside a
simple inclusion, so that a free-moving vapor bubble is visible inside a fluid-filled
chamber. Fluid inclusions that form at ambient conditions undergo no phase separation,
and therefore contain a single-phase fluid.
Fluid inclusions can be described in terms of genetic classification. The inclusions
in this study are primary and secondary inclusions: primary inclusions are trapped during
the growth of the host mineral, and are common in hydrothermal veins; secondary
inclusions occur along healed fractures, and are therefore entrapped after the growth of
the host crystal. A generally accepted corollary holds that isolated inclusions tend to form
early in a trapping sequence, while inclusion trails that follow fracture planes typically represent later fluids (Hollister, 2003). An additional descriptive classification regards inclusion type. The samples examined in this study contained Type I and I1 inclusions: in
Type I inclusions, vapor bubbles are small, comprising <40% of the fluid volume, and salt daughter crystals are absent, indicating salinity of <26% NaCl (the saturation point for NaC1); Type 11 inclusions have larger vapor bubbles (>60% of the fluid volume), and typically low salinity with no daughter minerals (Nash, 1976). Daughter crystals occur in fluids supersaturated with a soluble salt, such as NaC1, or some other mineral that crystallizes from the fluid upon cooling (Roedder, 1984).
A fluid of known composition has an empirically derived boiling curve that represents the critical point of the fluid over a range of temperatures and pressures.
Salinity determination for a fluid without daughter minerals requires freezing the fluid and then reheating it. The freezing temperature (T,) is the temperature at which the last ice crystal melts. The weight-percent (wt%) NaCl equivalent is then given by the
function:
f(wt% NaCl) = 1.76958TF- (4.2384 x 10-'T:) + (5.2778 x 10-4T:) + 0.028
as shown in Hall and others (1988). Thus, if the composition of the trapped fluid can be
determined and there is a known liquid-vapor curve for that fluid, then the fluid density
can be derived from the temperature of homogenization of the vapor and liquid phases
into a single fluid phase. A fluid with known P-V-T properties therefore has an isochore,
an array of lines of constant density extending from the boiling curve and defined in P-T
space that represents the range of P-T conditions over which a fluid of that density could
have been trapped (Hollister, 2003).
Homogenization occurs when, as the inclusion is heated, pressure in the inclusion
drops below the vapor pressure of the trapped fluid, causing the vapor bubble to nucleate.
The temperature, or critical point, at which the bubble nucleates is known as the homogenization temperature (T,) (Fig. 3.10). Each T, on the boiling curve corresponds to an isodensity curve. Because the isodensity curves have slopes where m * x (slope is not vertical), to determine the original temperature at which the fluid was trapped (TT) requires T, and an independent estimate of pressure. Only if the inclusion formed on the boiling curve does TH= TT (Fig. 3.1 1).
In my analysis I used a Linkham THMSG 600 Fluid Inclusion Stage mounted on an Olympus BX51 scope, connected to LinkSys 2.38 software, and capable of producing temperature range of -190" to 600•‹C.Calibration of the stage (using the critical points for Figure 3.10. Depiction of fluid inclusions during analysis. (A) Two-phase fluid inclusions aligned along fracture planes in crystalline quartz. V denotes vapor bubble, F the fluid from which it separated. Photo from Roedder, 1984, p. 344. (B-D) Three-part schematic diagram of two-phase fluid inclusion behavior during heating on fluid-inclusion analysis stage. Temperature of homogenization (TH)ib the temperature (critical point) at whiih the bubble nucleates and restores the trapped fluid to a single phase (part D). Figure 3.11. Phase iagram for pure water, which saline solutions (< wt% NaCl). Space defined by ()spheric pressure (PAW) versus temperature PC). Isodensity curves extend from boiling curve for water. Critical point labeled CP. Figure from L. Meinert. water) determined heating accuracy of +2"C at 373OC and freezing accuracy of +O.l•‹Cat
0•‹C (Meinert, personal communication). In all methods of fluid inclusion analysis, the accuracy and precision of analysis is proportional to the size of the inclusions under investigation-small fluid inclusions are inherently difficult, and the best possible resolution in a given field of view, whether down the barrel of the scope or on a video monitor, is an additional limitation. I conducted non-destructive analysis, observing inclusion properties under the microscope without opening or separating the inclusions from their host. I focused on thermal expansion, phase changes upon heating and freezing, and salinity determinations (Meinert, personal communication).
Fluid Inclusion Analysis
The crux of my analysis centered on two crystalline quartz samples, one from
Diapori (27b), the other from Charasonas (42c). I examined three samples of breccia, as well (27i, 26e, and 42h), but found them less conducive to analysis than the quartz samples.
Cape Diapori: Sample 27b
Sample 27b came from a large nodule of crystalline quartz near the base of the main breccia face at Diapori (Figs. 1.1 and 2.6). I analyzed 41 fluid inclusions arranged in five fluid-inclusion assemblages (FIAs -groups of fluid inclusions that all exhibit similar homogenization temperatures) of primary, Type I inclusions. All five FIAs yielded homogenization temperatures between 140-150•‹C, which, given the scattered spatial distribution of the FIAs and uniform THrange, is representative of inclusion
behavior throughout the sample (Fig. 3.12).
To determine fluid composition, I analyzed five large fluid inclusions with well-
defined vapor bubbles. Upon freezing and reheating, the last ice crystals melted between
-0.9" and -2.9"C, suggesting salinity of 1.5-4.7 wt% NaCl (Fig. 3.12).
* Charasonas Headland: Sample 42c
Sample 42c came from a quartz cluster in the hydrothermally altered zone between the two breccia walls at Charasonas (Fig. 1.I). I analyzed 52 fluid inclusions arranged in multiple FIAs, but each assemblage included at least one isolated inclusion.
Thus, in some cases, the "assemblages" were several isolated inclusions occupying the same field of view in the microscope that could be observed simultaneously. I documented a range of homogenization temperatures between 140•‹Cand 235•‹Cthat group into at least four broad temperature windows: 140-160, 180-200,200-220, and
220-240•‹C. Such bin sizes can vary with statistical organization of data from isolated inclusions, but among the different temperatures one sees a predominance of those between 210-240•‹C (Fig. 3.13).
To determine fluid composition, I analyzed 12 fluid inclusions with well-defined vapor bubbles. Upon freezing and reheating, the last ice crystals melted between -5.5" and -5.7"C, suggesting salinity of 8.5-9.0 wt% Wac1 (Fig. 3.13). Histogram of Homogenization Temperatures for FlAs in Sample #27b
Sample Location (Diapori)
Homogenization Temperature (OC)
Histogram of Freezing Temperatures for FlAs in Sample #27b
-4 -3 -2 1 0 1
Freezing Temperature ("C)
Figure 3.12. Fluid inclusion data from sample 27b, fault-related crystalline quartz from Diapori. Sample contains a fluid with a homogenization temperature (T,) of 140-150•‹C, and freezing temperature (TF)of -2.9 to -0.9"C, which suggests a composition of 1.5-4.7 wt % NaCl. Histogram of Homogenization Temperatures for FlAs in Sample #42c
Homogenizat~onTemperature ('C)
Histogram of Freezing Temperatures for FlAs in Sample #42c I I I I I I I I i 1 I I !
Sample Location (Charasonas)
-5 R -5 7 5 6 5 5 -5 4 5 3 -5 2
Freezing Temperature ("C)
Wgure 3.13. Fluid inclusion data from sample 42c, fault-related crystalline quartz from Charasonas. Sample contains at least three fluids with homogenization temperatures (T,) of 140-160,200-210, and 220-24Q•‹C,and freezing temperatures (TF)of -5.9 to -5.4"C, which suggest a composition of 8.5-9.0 wt% NaCl. * Inclusions in Clasts
My intention in employing fluid-inclusion analysis was to look for similarities
and differences in fluids trapped in several kinds of fault products. For Diapori and
Charasonas, I had hoped to compare fluids from crystalline quartz samples to those in the
breccia matrix, marble clasts, and calcite veining. In the Katakefalos samples, I wanted to
compare fluids in the primary, latest matrix to those in the earlier brecciated matrix and
the entrained marble fragments.
The practicalities of inclusion work, however, redefined my ambitions. The
cataclastic matrix samples were too open and opaque to have or reveal fluid inclusions.
The crystalline calcite veins I suspected would be inclusion rich were in fact completely
clear, or held inclusions too small to investigate. Likewise, although the marble clasts
contained fluid inclusions-I found trains of tiny inclusions along multiple fracture planes- the bubbles were either too small to see or visual resolution was inadequate at high magnification.
Based on the temperature data from the Diapori and Charasonas quartz samples, however, I decided to test whether the fluid inclusions trapped in marble clasts from those two locations reflected original metamorphic conditions, or instead contained some later fluid related to fault activity. Marbles on Syros experienced metamorphic conditions of at least 15 kbar and 450-500•‹C (Schumacher and others, 2000; Schumacher and others, 2001; Brady and others, 2001). Inclusions in high-grade rocks tend to contain high concentrations of CO, and are commonly <10 pm in size. Though perhaps difficult to see, most of the inclusions in my samples were larger than 10 pm. High-grade metamorphic inclusions also typically contain recrystallized daughter minerals of various
kinds, which my samples lack entirely.
I found a particularly large, secondary, Type I1 inclusion in a marble clast that
homogenized at 140-160•‹C, well short of any THnecessary to reflect a trapping
temperature near 500•‹C.The structural characteristics of the breccia itself are further
evidence that the major fluid inclusions trapped in the marble are more likely related to faulting than a high-grade metamorphic event: the mechanical calcite twinning, angular shapes, and open fractures evident in the marble clasts are typical of low-temperature deformation; with high-temperature deformation, one could expect to see plastic deformation, recrystallization after partial melting, and rehealing.
Discussion of Fluid Inclusion Data
The fluid inclusions trapped in the marble clasts of the breccia samples contain fluids that postdate metamorphism and derive from at least one low-salinity fluid that moved through the faults at some time.
Sample 27b from Diapori reflects a single fluid event with a minimum trapping temperature of 145•‹Cand salinity lower than that of seawater (3.5% NaC1). Such a dilute fluid could be heated meteoric water (groundwater) that moved through the fault at some time.
Sample 42c from Charasonas reflects several fluid events with different homogenization temperatures, but at least one fluid with a minimum trapping temperature of 230-235•‹C and salinity more than twice that of seawater. Salinity of 8-9% is still dilute compared to the 26% saturation concentration for NaC1, but must have come from some source other than seawater. One possibility is that the fluid is formation water that
migrated from sedimentary source rocks. Formation water originates in the pore spaces of
wet sedimentary rocks, and commonly contains concentrations of various salts and trace
minerals in solution. Compressional or shear stress can force formation waters out of the
source rock, so that, like any petrologic fluid, the waters migrate toward a reservoir.
Another hypothesis is that the fluid in the inclusions originated near an evaporite deposit.
Though no evaporite units are immediately evident near the Charasonas fault, evaporites are common in the Aegean and could underlie some of the local stratrigraphy. In the case of a proximal evaporite unit, saline water could have mixed with other migrating fluids and diluted to its present concentration by the time it reached the fault zone. Chapter 4
Discussion and Further Work ......
I. Discussion
The relative timing of the faults on Syros remains unclear. The extensive open- fracture networks evident in the host rocks at each of the fault zones suggests that the high-angle faults formed long after Syros rotated to the south along its listric basal decoll6ment, which tipped the island's stratigraphy to the north. The horst rotation might have occurred approximately 30-25 Ma during the major period of large-displacement extension responsible for blueschist exhumation. Given their scale, high angle, and limited displacement, it is highly unlikely that the faults examined in this study were at all responsible for or contemporaneous with the exhumation of blueschist terranes. Their en echelon arrangement and NW-SE orientation, however, is consistent with some of the regional extensional stress fields described in recent Aegean tectonic models (see discussion of Walcott and White (1998) in Chapter 1.11, p. 8, and Figs. 1.5 and 1.6). The notion of a partially-formed, NW-SE-trending graben on the west side of the island (Fig.
2.3 1) is at least feasible, given the NE-SW-directed tensile stresses that Syros would have experienced during microblock rotation and late extension in the Miocene.
According to Walcott and White's (1998) kinematic model for the central
Aegean, because Syros lies west of the Mid-Cycladic Lineament, the island would have been part of the western Aegean block, which rotated clockwise in the middle Miocene.
My model for a late-forming graben on Syros shows an opposite kind of rotation:
Oblique-slip indicators suggest clocltwise rotation of the graben with respect to the island, and therefore countercloclcwise rotation of the island with respect to the graben.
The Mid-Cycladic Lineament, as detailed in Fig. 1.6, strikes NE-SW less than 40 km SE
of Syros. Major structural features never exist alone-large transform faults tend to splay,
incorporating smaller subsidiary blocks. Amid the grist of the Aegean tectonic mill, Syros
might have rotated more than once in opposite directions during late Miocene or more
recent tectonic episodes.
In the broadest regional view, it is interesting to note the overall trend of the
Cycladic islands as a group. The Attic-Cycladic crystalline belt extends from continental
Europe in a south-sweeping arc through the Aegean before bending north toward Turkey.
The islands of Andros, Tinos, and Myltonos define a NW-SE-trending line through the northern Cyclades; Kea, Kithnos, Serifos, Sifnos, and Sikinos form a similar chain on the southern Cycladic margin. (Syros is centered between the two island rows, which is one reason for its being the regional capitol.) Thus schematically divided, the Cyclades strike parallel to the NW-SE-trending jaw of the Hellenic Trench that lies just SW of the
Peloponese and Crete. The arrangement of the archipelago is perhaps the effect of Attic orogenic collapse toward the Hellenic Trench that began approximately 21-16 Ma years ago (Gautier and others, 1999), and has continued since the Miocene indentation of
Arabia and westward extrusion of the Anatolian block (Fig. 1.7). The high-angle faults on
Syros might be small-scale manifestations of that pervasive trenchward collapse to the
SW.
The fluid-inclusion data presented here describe homogenization rather than trapping temperatures. It is important to note that without independent constraints for pressure, the homogenization temperature nevertheless stands as a minimum bound for trapping temperature, and places the fault-related fluids within a preliminary context for
fault formation.
I found evidence for fault-related fluids at Diapori and Charasonas with original
trapping temperatures of at least 150 and 240•‹C,respectively. If one assumes, for the
sake of argument, that the temperatures of homogenization I recorded are equal to the
temperatures at which the fluids were trapped, and further assume a geothermal gradient
of 30•‹Cper lm-a reasonable figure for the highly attenuated lithosphere of the
Aegean-one can calculate the approximate crustal depths from which the fluids may have migrated. The 150•‹Cfluid at Diapori suggests a source depth of at least 4 lm; the
240•‹Cfluid at Charasonas suggests a source depth of at least 7 krn. Because Syros is a north-dipping horst block, the northern fault zone at Diapori cuts through host rocks that are stratigraphically higher than those at Charasonas. Peninsula to peninsula, Diapori and
Charasonas are approximately 10-12 lun apart. It is possible that faulting occurred at shallow depths and the warm fluids migrated from deeper sources to infiltrate the stsuctures, but with a liberal allowance for variations in strata thickness and an average stratagraphic dip of 20-50" N, the horizontal distance of 10-1 2 km might also account for the vertical difference of approximately 3 km that the temperature data suggest.
The fluid composition in the Diapori sample is dilute enough to derive from meteoric water-fluid that percolated into the rock from surface precipitation. The fluid composition in the Charasonas sample is significantly more saline than the Diapori fluid, and possibly derives from sedimentary formation waters or fluid escape from a buried halite deposit. If the lower-temperature, less-saline fluid at Diapori migrated from a shallower crustal level than the higher-temperature, more-saline fluid at Charasonas, then the thermal and compositional differences between the samples might indeed be a
function of relative crustal depth. Such a scenario agrees with petrographic evidence of
mechanical twinning, open microfractures, angular fragments, and cataclastic matrix
from the fault zones, features that derive from brittle deformation during faulting; brittle
deformation occurs in low-temperature environments at shallow crustal levels.
PI. Further Work
The goal of this study was to characterize three previously unexamined high- angle faults with well-developed breccia zones along Syros' western coast. I documented the scales of the faults, their orientations, products, and structural contexts in an effort to place them within the general tectonic framework for the island, and possibly the Aegean.
Ideally, I might have found evidence that would have enabled me to constrain the age of faulting-a cross-cut granitoid dike or pluton, for example, such as those on the neighboring Cycladic islands of 10s and Naxos (Boronkay and Lister, 1994; Vandenberg and Lister, 1996), or offset in a datable layer of volcanic ash. But we have no such definitive timing indicators, only implications of relative age from youthful, sharp features of brittle deformation.
Despite the pieces of the puzzle still missing, future workers on this project can do more with what is immediately at hand. Although I did not analyze them, I collected samples of host rock from each of the fault zones, and a complete suite of samples-fault products and wall rocks-from an additional fault zone on the Mirties highland (Figs.
2.28, 2.29). A second round of fluid-inclusion analysis might make more intensive comparisons between fluids in the host roclts and those trapped in fault products to
confirm that the fluids within the faults are indeed fault related, exhibiting a truly
different thermal and compositional signature than fluids in the wall roclts. Further
petrologic and fluid-inclusion analyses might more completely describe the crystalline
calcite vein-cascade material from Diapori. Do the samples contain analyzable fluid
inclusions? Field evidence shows that the veining is fault related (the necessary and
adequate context for data from any fluids trapped therein). Workers still will not how
when the fluids were trapped. For a given inclusion-rich sample, worlters will only be
able to state that a fluid of composition x and minimum temperature y infiltrated the fault
system at some time.
Were I to return to Syros to resample the same fault zones and sample from other
fault zones with similar characteristics, I would focus my sampling efforts on any nodules
of crystalline quartz and calcite I could find-the roclts that contain the best specimens of
fluid inclusions. Most of the breccia samples I collected are too cataclastic in texture, rife with networks of open space, to have trapped anything but the last fluids to have entered the fault system since seismic activity ended (Figs. 3.7, 3.9). Infrared petrographic analysis might allow a closer look at some of the fluid inclusions in the breccia, but the infrared cameras currently available for such work capture lower-resolution images than standard scope-mounted cameras. The precision necessary to better access the inclusions
I saw requires higher resolution, not grainier imagery.
Indeed, further study of faulting on Syros might take a different approach altogether and incorporate geographic information systems (GIs) and remote-sensing technologies. If Syros itself is a horst block, we may expect to find patterns of fault traces preserved in the Aegean Sea floor. Walcott and White (1998) present a figure with
schematic traces that splay from the Mid-Cycladic Lineament SE of Syros (Fig. 1.6).
Sensor arrays mounted on geographic-information satellites, however, are able to
generate incredibly accurate images of surface and benthic terrain. High-resolution
bathymetric data-digital elevation model (DEM) images with 10 m pixels-would
surely show major sea-floor structures around Syros and the entire central Aegean. The
data exist; the principal problem would be their acquisition and processing, which would
come at some price.
But the possibilities that such high-resolution bathymetry presents are too exciting to ignore. The images would be detailed enough to see topographic manifestations of fault displacement, and distributions and patterns of medium-to-large-scale lineaments. Is there evidence of a NW-SE-trending en echelon fault system that cuts into Syros from the west, as the model in Fig. 2.3 1 suggests? Is the Mid-Cycladic Lineament a distinct trench to the SE, or does it branch into interfering splays that wander among the Cycladic islands? Might we see multiple large-scale faults that suggest opposing stress-field orientations, the ground-up remnants of Miocene-Pliocene block rotations? Or will three million years of benthic sedimentation have obscured any details of interest beyond recognition?
Structural analysis of Aegean bathymetry would serve to better link the late faulting on Syros to the larger tectonic scheme of the Cyclades and eastern
Mediterranean, and refine aspects of existing models for Miocene microblock rotation.
We can only imagine what unexpected features and relationships might surface. References ......
Avigad, D., 1993, Tectonic juxtaposition of blueschists and greenschists in Sifnos Island (Aegean Sea) -implications for the structure of the Cycladic blueschist belt: Journal of Structural Geology, v. 15, no. 12, p. 1459-1469.
Avigad, D., Baer, G., and Heimann, A., 1998, Block rotations and continental extension in the central Aegean Sea: paleomagnetic and structural evidence from Tinos and Mykonos (Cyclades, Greece): Earth and Planetary Science Letters, v. 157, p. 2340.
Avigad, D., and Garfunkel, Z., 1991, Uplift and exhumation of high-pressure metamorphic terrains: the example of the Cycladic blueschist belt (Aegean Sea): Tectonophysics, v. 188, p. 357-372.
Avigad, D., Garfunkel, Z., Jolivet, L., and Azafion, J. M., 1997, Back arc extension and denudation of Mediterranean eclogites: Tectonics, v. 16, no. 6, p. 924-941.
Bsady, J. B., Able, L. M., Cheney, J. T., Sperry, A. J., and Schumacher, J. C., 2001, Prograde lawsonite pseudomorphs in blueschists from Syros, Greece: Geological Society of America Abstracts with Programs, v. 33, no. 6, p. 250-25 1.
Brady, J. B., Markley, M. J., Schumacher, J. C., Cheney, J. T., and Bianciardi, G. A., 2003, Aragonite pseudomorphs in high-pressure marbles of Syros, Greece: Journal of Structural Geology, p. 1-7 (in press).
Bonneau, M., Correlation of the Hellenide nappes in the south-east Aegean and their tectonic reconstruction: in Dixon, J. E., and Robertson, A. F., The Geological Evolution of the Eastern Mediterranean, The Geological Society of London Special Publication 17, p. 5 17-527.
Boronkay, K., and Doutsos, T., 1994, Transpression and transtension within different structural levels in the central Aegean region: Journal of Structural Geology, v. 16, no. 11, p. 1555-1573.
Borradaile, G. J., Baylay, M. B., Powell, C., eds., 1983, Atlas of deformational and metamorphic rock fabrics: New York, Springer-Verlag, Incorporated, 530 p.
Brocker, M., and Enders, M., 1999, U-Pb zircon geochronology of unusual eclogite- facies rocks from Syros and Tinos (Cyclades, Greece): Geological Magazine, v. 136, p. 111-1 18.
Buick, I. S., 1991, Mylonite fabric development on Naxos, Greece: Journal of Structural Geology, v. 13, no. 6, p. 643-655. Davis, G. H., and Reynolds, S. J., 1996, Structural geology of rocks and regions (second edition): John Wiley & Sons, Inc., New York, 776 p.
Engelder, J. T., 1974, Cataclasis and generation of fault gouge: Geological Society of America Bulletin, v. 85, no. 10, p. 1515-1522.
Gautier, P., and Brun, J. P., 1994, Crustal-scale geometry and kinematics of late-orogenic extension in the central Aegean (Cyclades and Evvia Island): Tectonophysics, v. 238, p. 399424.
Gautier, P., Brun, J. P., Moriceau, R., Sokoutis, D., Martinod, J., and Jolivet, L., 1999, Timing, kinematics, and cause of Aegean extension: a scenario based on a comparison with simple analogue experiments: Tectonophysics, v. 315, p. 31-72.
Gealy, W. K., 1998, Plate tectonic evolution of the Mediterranean-Middle East region: Tectonophysics, v. 155, p. 285-306.
Goldstein, J. I., Newbury, D. E., Echlin, P., Joy, D. C., Fiori, C., and Lifshin, E., 1981, Scanning electron microscopy and x-ray microanalysis: New York, Plenum Press, 688 p.
Groshong, R. H., Jr., 1988, Low-temperature deformation mechanisms and their interpretation: Geological Society of America Bulletin, v. 100, no. 9, p. 1329-1360.
Hall, D. L., Sterner, S. M., and Bodnas, R. J., 1988, Freezing point depression of NaC1- KCl-H,O solutions, Economic Geology, v. 83, p. 197-202.
Hollister, L. S., 2003, Information intrinsically available from fluid inclusions: in Samson, I., Anderson, A., and Masshall, D., eds., Fluid inclusions: analysis and interpretation, Mineralogical Association of Canada Short-Course Series, v. 32, p. 1-12.
Holloway, J. R., 2003, Compositions and volumes of supercritical fluids in the Earth's crust: in Samson, I., Anderson, A., and Marshall, D., eds., Fluid inclusions: analysis and interpretation, Mineralogical Association of Canada Short-Course Series, v. 32, p. 13-38.
Jolivet, L., and Faccenna, C., 2000, Mediterranean extension and the Africa-Eurasia collision: Tectonics, v. 19, no. 6, p. 1095-1 106.
Keay, S., and Lister, G., 2002, African provenance for the metasediments and metaigneous rocks of the Cyclades, Aegean Sea, Greece: Geology, v. 30, p. 235-238. Klein, C., and Hurlbut, C. S., Jr., 1993, Manual of Mineralogy (after Dana, J. D.): New York, John Wiley and Sons, Inc., 681 p.
Knipe, R. J., 1989, Deformation mechanisms: recognition from natural tectonites: Journal of Structural Geology, v. 11, p. 127-146.
Logan, J. M., and Rauenzahn, K. A., 1987, Frictional dependence of gouge mixtures of quartz and montmorillonite on velocity, composition, and fabric: Tectonophysics, V. 144, p. 87-108.
Meijer, P. T., and Wortel, M. J. R., 1997, Present-day dynamics of the Aegean region: A model analysis of the horizontal pattern of stress and deformation: Tectonics, v. 16, no. 6, p. 879-895.
Meinert, L., 2004, personal communication: Department of Geology, Smith College, Northampton, MA, 01063.
Nash, J. T., 1976, Fluid-inclusion petrology: data from porphyry copper deposits and application to exploration: U. S. Geological Survey Professional Paper, 907D, 16 P.
Okrush, M., and Brocker, M., 1990, Eclogites associated with high-grade blueschists in the Cyclades archipelago, Greece: a review: European Journal of Mineralogy, v. 2, p. 451-478.
Ridley, J., 1984 (a), Listric normal faulting and the reconstruction of the synmetamorphic structural pile of the Cyclades: in Dixon, J. E., and Robertson, A. F., The Geological Evolution of the Eastern Mediterranean, The Geological Society of London Special Publication 17, p. 755-761.
Ridley, J., 1984 (b), The significance of deformation associated with blueschist facies metamorphism on the Aegean island of Syros: in Dixon, J. E., and Robertson, A. F., The Geological Evolution of the Eastern Mediterranean, The Geological Society of London Special Publication 17, p. 545-550.
Roedder, E., 1984, Fluid inclusions: in Ribbe, P. H., series ed., Reviews in Mineralogy, Mineralogical Society of America, v. 12, 644 p.
Schumacher, J. C., Brady, J. B., Prinkey, D. R., Walton, A., Able, L. M., Sinitsin, A. G., and Cheney, J. T., 2001, Geochemistry and metamorphism of blueschist/eclogite facies rocks on the island of Syros, Cyclades, Greece: Geological Society of America Abstracts with Programs, v. 33, no. 6, p. 250.
Schumacher, J. C., and Helffrich, G., 2003, Bristol mapping course: Syros and the Cyclades: personal courtesy of J.C. Schumacher, informal field guide for Bristol University, England, mapping excursions, p. 1-25. lumacher, J. C., Tonnsen, R., Cheney, J. T., Brady, J. B., and Koontz, C., 2000, Glaucophane marbles and associated high-pressure rocks on the island of Syros, Cyclades, Greece: Geological Society of America Abstracts with Programs, v. 32, no. 7, p. 114.
Sibson, R. H., 1977, Fault rocks and fault mechanisms: Journal of the Geological Society, London, v. 133, p. 191-213.
Smith, A. G., and Woodcock, N. H., 1982, Tectonic synthesis of the Alpine- Mediterranean region: a review: in Berckhemer, H., and Hsu, K., eds., Alpine- Mediterranean Geodynamics, American Geophysical Union Geodynamic Series, V. 7, p. 343-366.
Turnell, H. B ., 1988, Mesozoic evolution of Greek microplates from paleomagnetic measurements: Tectonophysics, v. 155, p. 307-316.
Vandenberg, L. C., and Lister, G. S., 1996, Structural analysis of basement tectonites from the Aegean metamorphic core complex of Ios, Cyclades, Greece: Journal of Structural Geology, v. 18, no. 12, p. 1437-1454.
Van der Pluijm, B. A., and Marshak, S., 2004, Earth structure: New York, W. W. Norton and Company, Incorporated, 656 p.
Vidal, N. V., Joaquina, A. M, and Klaeschen, D., 2000, The structure of the African- Anatolia plate boundary in the eastern Mediterranean: Tectonics, v. 19, no. 4, p. 723-739.
Walcott, C. R., and White, S. H., 1998, Constraints on the kinematics of post-orogenic extension imposed by stretching lineations in the Aegean region: Tectonophysics, V. 298, p. 155-175.
Wenk, H. R., and Christie, J. M., 1991, Review paper: comments on the interpretation of deformation textures in rocks: Journal of Structural Geology, v. 13, p. 1091-1 110.
Wernicke, B., and Burchfiel, B. C., 1982, Modes of extensional tectonics: Journal of Structural Geology, v. 4, no. 2, p. 105-1 15.
Wijbrans, J. R., van Wees, J. D., Stephenson, R. A., and Cloetingh, S. A. P. L., 1993, Pressure-temperature-time evolution of the high-pressure metamorphic complex of Sifnos, Greece: Geology, v. 21, p. 443-446.