Journal of Petrology Advance Access originally published online on February 25, 2005
Journal of Petrology 2005 46(6):1203-1241; doi:10.1093/petrology/egi014
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Extrusional Tectonics in the Core of a Transpressional Orogen; the Kaoko Belt, Namibia
1 SCHOOL OF EARTH SCIENCES, UNIVERSITY OF MELBOURNE, PARKVILLE, VIC. 3010, AUSTRALIA
2 CONTINENTAL EVOLUTION RESEARCH GROUP, SCHOOL OF EARTH AND ENVIRONMENTAL SCIENCES, UNIVERSITY OF ADELAIDE, S.A. 5005, AUSTRALIA
RECEIVED APRIL 19, 2004; ACCEPTED JANUARY 14, 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
The Pan-African Kaoko Belt of NW Namibia provides a well-exposed example of material flow in the evolving middle and lower crust during sinistral transpressional orogenesis. The Kaoko Belt is a composite metamorphic belt with shear zone bounded zones of contrasting metamorphic style that were metamorphosed at approximately the same time (575–550 Ma). It is divided into a Barrovian-style Escape Zone with inclined nappes and thrusts that verge outward towards the foreland, a steep high-grade Orogen Core that experienced intense wrench-style deformation and, in the hinterland, an older (650 Ma) exotic Coastal Terrane that was reworked at moderate-T/low-P conditions during transpressional orogenesis. The kinematic-lineation array (defined by stretching lineations and shear-sense indicators) shows variation in degree and polarity of lineation obliquity to the grain of the belt, lineation plunge and different polarity of vergence along the plunge, with respect to the overall sinistral shear sense. The gross regional lineation pattern defines an arcuate array from near orogen-parallel in the Orogen Core to higher-angle obliquity across the Escape Zone, reflecting lateral escape towards the orogen margin. Two high-grade lobes within the Orogen Core are coincident with oblique-inclined, upward flow trajectories based on the kinematic-lineation array, and represent oblique extrusion of lower-crustal material into middle-crustal levels. An intervening low-grade metamorphic ‘trough’ within the Orogen Core and the Coastal Terrane coincides with acute oblique downward and outboard-directed apparent flow trajectories. These kinematic patterns are consistent with barometric constraints for peak metamorphism of 5·2 kbar and 4·5 kbar in the low-grade ‘trough’ and Coastal Terrane and 8·0 kbar in the two high-grade lobes. Integration of the kinematic and metamorphic datasets shows that oblique, shallowly inclined extrusion trajectories occurred within the Orogen Core of this classic transpressional orogen, and that stretching lineations approximately reflect the particle paths experienced. Across-orogen metamorphic gradients indicate a marked contrast between the Escape Zone and Orogen Core. The Escape Zone experienced an inverted Barrovian-style high-P/moderate-T metamorphism and low thermal gradients of 20°C/km. The Orogen Core experienced moderate-P/high-T metamorphism and high thermal gradients of 30–40°C/km. Clockwise P–T paths were experienced in all parts of the Kaoko Belt. Tight and shallow conduction-dominated P–T paths were experienced in all Orogen Core domains, involving heating-dominated prograde paths followed by decompression, consistent with extrusional tectonics. Steeper and more open advection-dominated P–T paths were experienced in the crustal over-thrust Escape Zone, which experienced low thermal gradients.
KEY WORDS: Pan-African orogeny; transpression; extrusion; kinematic-lineation array; metamorphism; lower-crustal processes; metamorphic gradient
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Processes operating within the ductile middle to lower crust of transpressional systems have only recently been investigated (Hansen, 1989
; Holdsworth & Strachan, 1991; Comacho & McDougall, 2000; Martelat et al., 2000; Stowell & Crawford, 2000; Johnson & Kattan, 2001; Whitney et al., 2001; Little et al., 2002; Vassallo & Wilson, 2002; Goscombe et al., 2003a, 2003b). Integrated structural and metamorphic studies of middle- to lower-crustal exposures are critical to the understanding of material flow in these systems.
Transpressional orogens are an important part of many collisional orogenic systems (Jones et al., 1977; Tapponnier et al., 1982; Woodcock, 1986; Holdsworth & Strachan, 1991; Jones & Strachan, 2000). They are in effect large-scale triclinic shear zones (Lin et al., 1998; Jiang et al., 2001) involving a wrench or shear component parallel to, and a non-coaxial dip-slip shortening component normal to, the zone of transpression (Robin & Cruden, 1994; Fossen & Tikoff, 1998). Transpression operates along many plate margins of the Earth today as a result of oblique plate convergence. Low degrees of oblique convergence are involved, and Patchett & Chase (2002) estimated that at any given time >60% of convergent plate margin lengths worldwide will have ß
<30° (where ß is the angle between the convergence vector and the strike of the orogen). Consequently, transpressional orogens are expected to be a common part of the plate tectonic paradigm (Veevers, 2003). Despite this, it is no simple matter to recognize oblique convergence, let alone to quantify the degree of obliquity, in old orogenic belts.
A large body of work now exists that describes case studies and mechanical models that predict the way in which transpressional systems operate (Tikoff & Teyssier, 1994; Fossen & Tikoff, 1998). Most of what is known about transpressional systems has been derived from structural investigations, and the majority of these relate to the upper crust (Harland, 1971; Lowell, 1972; Wilcox et al., 1973; Sylvester & Smith, 1976; Sylvester, 1988; Oldow et al., 1990; Kirkwood et al., 1995; Teyssier & Tikoff, 1998; Holdsworth & Pinheiro, 2000; Jones et al., 2004). However, apart from the work of Schaller et al. (1999), Jones & Strachan (2000), Whitney et al. (2001), Jones & Escher (2002), Goscombe et al. (2003b) and Johnson et al. (2003), metamorphic studies of middle- to lower-crustal exposures of transpressional systems of orogen scale are few.
To a large extent the thermal evolution of orogenic belts is governed by the competition between heat conduction and heat advection as material is moved through the crust at different rates. This is quantified by the thermal Peclet number (PeT; Sandiford, 2002), where advection-dominated systems such as subduction have PeT >10, whereas conduction dominates at PeT <1. The rate of advection of heat is governed by the rate of deformation as well as magma transport; thus, metamorphism must be intimately linked to the kinematics of the orogen. Large-scale domains of contrasting kinematic style are characteristic of transpressional orogens, giving regions of differing rates of topographic development and erosion, which lead to spatially partitioned exhumation rates (Koons, 1990; Koons et al., 2003). The linked deformation–topography–erosion behaviour influences the rate of material advection and thus the thermal properties of the orogen, creating a feedback relation that is reflected by the metamorphic patterns. Consequently, a domainal variation in the metamorphic character of transpressional orogens is expected that reflects the patterns of kinematic partitioning in the orogen. The metamorphic gradient [variation in metamorphic parameters such as temperature (T), pressure (P) and average thermal gradient (G, T/depth) along a horizontal length scale] should reflect spatial variations in erosion rate, thermal parameters and the path that a rock takes as it moves through an orogen (Spear et al., 1984; Spear, 1993; Jamieson et al., 1996; Thompson et al., 1997a). Therefore, metamorphic gradients have the potential to provide insights into the way in which transpressional systems operate.
Although there are a large number of studies concerning the thermal evolution of transpressional systems (Koons, 1987, 1990; Batt & Braun, 1997; Thompson et al., 1997a), relatively few have specifically explored the consequences of variations in the degree of obliquity. Both Batt & Braun (1997) and Thompson et al. (1997a) showed that low angles of obliquity favour systems that evolve towards the conductive end-member (PeT < 1), as a result of the development of comparatively subdued topography, which inhibits rapid erosion, or the development of a threshold topography that strongly concentrates orographic effects. The outcomes of the Thompson et al. (1997a) model, in particular, suggest that systems dominated by low angles of convergence should develop comparatively high average thermal gradient metamorphic systems as a consequence of long residence times within the orogen. In contrast, domains with higher angles of convergence and considerable topographic development will be zones of intense erosion, advection-dominated metamorphism and rapid cooling histories, such as the New Zealand Alps (Grapes & Watanabe, 1992; Braun & Beaumont, 1995; Little et al., 2002; Koons et al., 2003). At higher angles of convergence, lower thermal gradients are experienced and the rocks undergo Barrovian-style metamorphism. Thus exhumation rate is proportional, and heating is inversely proportional, to the angle of convergence obliquity. Consequently, granulites with high average thermal gradients are expected in the lower-angle convergent domains in orogen cores and lower average thermal gradient Barrovian-style metamorphic conditions in the higher-angle convergent margins of these orogens.
This general picture is largely based on theoretical models for transpression (Thompson et al., 1997a) that have received only limited testing in deeply eroded orogens. Here we investigate how deformation and metamorphism are both temporally and spatially interlinked within a classic transpressional orogen, the Kaoko Belt in Namibia (Dürr & Dingeldey, 1996; Passchier et al., 2002; Goscombe et al., 2003a, 2003b), and present evidence for transpression-induced exhumation of the lower crust. The aim is to determine in what way particle paths are manifested in transpressional orogens, and, in particular, to constrain the vertical component using metamorphic information such as metamorphic gradients, P–T paths and relative barometric evolution of domains. Irrespective of the type of transpression (Fossen & Tikoff, 1998), material lines rotate towards vertical if there is a significant contractionally driven component of flow. Thus particle paths in the core of transpressional orogens are expected to have a vertical component of flow and some degree of lower-crust extrusion. This paper investigates how linear and planar structural fabrics relate to information about particle paths recorded by the P–T–t paths experienced and 
T/x, P/x, G/x metamorphic gradients (Sonder & Chamberlain, 1992), and how the structural and metamorphic datasets relate to the inferred particle paths. The approach taken was to synthesize new (this study) and existing results (Goscombe et al., 2003a, 2003b) of a seven year research programme covering the entire Kaoko Belt. Mapping, kinematic, metamorphic and geochronological datasets have been integrated and used to constrain orogen architecture, evolution and mechanisms of transpressional deformation and the resultant metamorphism. We present here the full metamorphic dataset for the entire Kaoko Belt; this is interpreted with reference to the kinematic-lineation array and orogen architecture, which have been presented elsewhere (Goscombe et al., 2003a; Goscombe & Gray, in preparation). This analysis has provided important constraints on the particle paths experienced and the nature of extrusional tectonics in the core of the orogen.
|
STRUCTURAL EVOLUTION AND ARCHITECTURE OF THE KAOKO BELT |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
The Kaoko Belt is a sinistral transpressional component of the late Neoproterozic to Cambrian Pan-African Orogenic System (Goscombe et al., 2000
), called the Damara Orogen in Namibia (Miller, 1983; Prave, 1996). The Damara Orogen has triple-junction geometry and the Kaoko Belt is the 700 km long, NNW-trending northern coastal arm (Fig. 1). Within the belt is an intensely deformed Damara Sequence of Neoproterozoic age and a mosaic of Archaean, Palaeoproterozoic and Mesoproterozoic basement units (Fig. 1) of the Congo Craton (Seth et al., 1998). Collision followed by transpressional orogenesis occurred between the Congo and Rio De La Plata Cratons at 580–550 Ma (Goscombe et al., 2003a, 2003b; Goscombe & Gray, in preparation) and was terminated by high-angle collisional orgenesis between the Congo and Kalahari Cratons at 540–510 Ma (Miller, 1983; Goscombe et al., 2003a, 2003b; Goscombe & Gray, in preparation).
|
Subdivision of the Kaoko Belt
The Kaoko Belt is subdivided into the following, laterally continuous, and shear zone bounded, components (Goscombe et al., 2003a
, 2003b; Goscombe & Gray, in preparation).
(1) An Eastern Kaoko Zone (Miller, 1983), in the foreland, consists of low-grade, low-strain, upright folded Damara Sequence platform carbonates (Prave, 1996; Hoffman et al., 1998) overlying Palaeoproterozoic basement of the Congo Craton (Fig. 1).
(2) A Barrovian-style Escape Zone contains inclined large-scale nappes and over-folds that verge outward towards the foreland margin and these typically have cores of Archaean and Palaeoproterozoic basement. Damara sediments in the Escape Zone are deep basin and slope facies that experienced intense transpressional deformation and S–L fabric development at greenschist- to upper amphibolite-facies metamorphic grade.
(3) The 20–40 km wide Orogen Core contains high-grade Damara Sequence, antiformal slivers of Mesoproterozoic basement and Pan-African granitoids that all experienced intense wrench-style deformation (Goscombe et al., 2003a). The Orogen Core is a composite of three domains with distinct lithostratigraphy and structural and metamorphic style (Goscombe & Gray, in preparation; Fig. 1). The Hoarusib Domain in the south and Hartmann Domain in the north are of upper amphibolite to granulite grade and were intensely deformed by isoclinal folding, steep penetrative foliations and shear zones with sub-horizontal stretching lineations (Goscombe et al., 2003a). Between the two high-grade domains, on the east side of the Orogen Core is the Khumib Domain, a chevron folded turbiditic Damara sequence of lower amphibolite-facies grade.
(4) The outboard Coastal Terrane may be an exotic terrane, with distinct lithostratigraphy and an early (650 Ma) high-grade metamorphic cycle that is otherwise absent from the rest of the Kaoko Belt (Goscombe & Gray, in preparation). The Coastal Terrane consists of coarse-grained migmatized meta-greywackes and meta-feldspathic psammites, and a high proportion of mafic, intermediate and granitic meta-intrusive rocks. This terrane experienced upper amphibolite-facies metamorphism at 650 Ma and was subsequently pervasively reworked by ductile grain refinement at lower amphibolite-facies conditions during transpressional orogenesis of the entire belt at 580–550 Ma (Goscombe et al., 2003a, 2003b; Goscombe & Gray, in preparation).
The structural architecture of the central portion of the Kaoko Belt has been described in detail by Guj (1970), Dingeldey et al. (1994), Dürr & Dingeldey (1996), Dingeldey (1997) and Goscombe et al. (2003a, 2003b).
Evolution of the Kaoko Belt
Three successive, but temporally unrelated Pan-African tectono-metamorphic cycles that formed in entirely distinct crustal stress regimes, have been recognized and defined by Goscombe et al. (2003a, 2003b) and Goscombe & Gray (in preparation). M1 has been labelled the Thermal Phase (660–640 Ma), M2 the Transpressional Phase (580–550 Ma) and M3 the Shortening Phase (540–510 Ma).
The Thermal Phase has only been recognized in the Coastal Terrane and is apparently absent from all other parts of the Kaoko Belt (Goscombe et al., 2003b; Goscombe & Gray, in preparation). The Thermal Phase is manifested by high-T/low-P granulite- to upper amphibolite-facies metamorphism dated by metamorphic zircon and monazite at 645 ± 3·5 Ma (Franz et al., 1999), diorite and granodiorite with zircon crystallization ages of 656 ± 8 Ma (Seth et al., 1998) and granitic partial melt segregations of c. 650 Ma age (Goscombe & Gray, in preparation). The Coastal Terrane was metamorphosed prior to cessation of deposition of Damara sediments in the adjacent Orogen Core. Thus the Coastal Terrane must be an exotic terrane docked after 600 Ma, and the bounding Three Palms Mylonite Zone (Fig. 1) must be a suture. The Coastal Terrane contains lenticular bodies of more typical Pan-African granitoid bodies with 576–565 Ma zircon ages (Seth et al., 1998), identical to those pervading the Orogen Core. Consequently, the Coastal Terrane was docked with the rest of the Kaoko Belt subsequent to M1 metamorphism of the Coastal Terrane at 655–645 Ma and prior to granite emplacement at 580 Ma.
The sinistral Transpressional Phase produced almost all the deformation structures, pervasive fabrics, mineral parageneses and architecture of the Kaoko Belt during oblique collision between the Congo and Rio De La Plata Cratons. The deformation history spanned 580–550 Ma (Seth et al., 1998; Franz et al., 1999; Goscombe et al., 2003b; Goscombe & Gray, in preparation) and deformation style changed progressively both with time and with spatial partitioning across the belt (Dingeldey et al., 1994; Dürr & Dingeldey, 1996; Dingeldey, 1997; Goscombe et al., 2003a, 2003b; Goscombe & Gray, in preparation). Peak metamorphic, coarse-grained matrix and porphyroblastic early M2 mineral parageneses formed early in the Transpressional Phase. These were variably reworked by late M2 mineral parageneses developed in the pervasive S–L fabric and shear zones within the Orogen Core and grain-refinement fabrics in the Coastal Terrane (Goscombe et al., 2003b). In contrast to the remainder of the Kaoko Belt, which experienced a prograde M2 metamorphic cycle during transpressional orogenesis, the Coastal Terrane was extensively reworked and downgraded by ductile grain refinement of earlier coarse-grained parageneses during transpressional orogenesis.
Peak M2 metamorphism occurred at approximately 580–560 Ma and was approximately coeval in all parts of the Kaoko Belt (Goscombe et al., 2003b). Sm–Nd dating indicates peak metamorphic garnet growth at 575 ± 10 Ma (all errors quoted are 2 analytical uncertainties) in the Escape Zone, and both high-grade (Hoarusib) and low-grade (Khumib) domains within the Orogen Core (Goscombe et al., 2003b). Furthermore, concordant U–Pb and Pb–Pb zircon age determinations by different methods range between 580 ± 3 Ma and 564 ± 1·3 Ma for sheared granitoids in the Coastal Terrane and Orogen Core (Miller & Burger, 1983; Seth et al., 1998; Franz et al., 1999; Goscombe & Gray, in preparation). Concordant U–Pb monazite ages of 553·6 ± 1·4 Ma from late kinematic granites (Franz et al., 1999) and U–Pb zircon sensitive high-resolution ion microprobe (SHRIMP) dates of 549·2 ± 1·9 Ma and c. 550 Ma from late M2 pegmatites in the Hoarusib and Hartmann Domains (Goscombe & Gray, in preparation), suggest that the Transpressional Phase of deformation persisted to 550 Ma.
For convenience, the progressive Transpressional Phase has been broken into three temporally overlapping, but distinct deformation styles that reflect partitioning of the strain components (Goscombe et al., 2003a; Goscombe & Gray, in preparation).
(1) Pervasive main phase deformation fabrics were preceded by flat-lying L–S fabrics that are uncommonly preserved in low-strain shear lenses within the Orogen Core (Goscombe & Gray, in preparation).
(2) The Wrench Stage style of deformation was initiated early and generated steep, penetrative and widespread S–L fabrics. These fabrics became progressively more partitioned into crustal-scale, strike-slip shear zones within and bounding the Orogen Core. These shear zones have not been folded by convergence stage folding and are interpreted to have remained active to the end of all Transpressional Phase deformation (Goscombe et al., 2003a, 2003b; Goscombe & Gray, in preparation).
(3) Convergence Stage style of structures accommodated the contractional component of deformation, giving widespead mesoscopic to macroscopic, tight to isoclinal folding in the panels between crustal-scale shear zones, and large-scale east-vergent nappes, overfolds and thrusts. Convergence stage structures invariably rework main phase S–L fabrics by transposition cycling. They were initiated slightly later in the Transpressional Phase and with time became progressively more dominant than continuing development of the pervasive S–L fabric. Convergence stage folds, nappes and thrusts dominate deformation of the Escape Zone and Eastern Kaoko Zone (Goscombe et al., 2003a).
The Shortening Phase of deformation involved a change of crustal stress regime and accompanied coaxial NNE–SSW shortening during high-angle convergence between the Congo and Kalahari Cratons. Structural overprinting relationships in the southern Kaoko Belt, or Ugab Zone, indicate that transpressional orogenesis of the Kaoko Belt was terminated before Shortening Phase deformation (Coward, 1983; Freyer & Halbich, 1994; Maloof, 2000; Passchier et al., 2002; Goscombe et al., 2004). Shortening Phase strain in the Kaoko Belt proper was almost insignificant, resulting in upright and large-scale east–west-trending warps. There is no record of an M3 thermal peak and the Kaoko Belt cooled through 500°C at c. 565 Ma and through 400–300°C at 531–520 Ma (Goscombe et al., 2003a; Goscombe & Gray, in preparation). Shortening Phase deformation accompanied the emplacement of a late kinematic syenogranite of 530 ± 3 Ma age in the Ugab Zone (Seth et al., 2000) and late kinematic pegmatite of 530 Ma age in the southern Coastal Terrane (Goscombe & Gray, in preparation). The main tectono-metamorphic phase in the Inland Branch is widely regarded to have occurred over the interval 530–510 Ma (Miller, 1983; Jung et al., 2000).
Crustal architecture
The Kaoko Belt (Fig. 1) shows all the typical features of a deeply eroded transpressional orogen, such as strike-slip shearing, a major median shear zone in the internal part and outward to the margin, nappes and thrusts showing progressively more oblique to high-angle transport onto the foreland (Shackleton & Ries, 1984; Holdsworth & Strachan, 1991; Vassallo & Wilson, 2002; Goscombe et al., 2003a). On a gross scale, the Kaoko Belt is a co-temporal composite metamorphic belt, with two coeval halves of contrasting metamorphic styles (Goscombe et al., 2003b) and also kinematic partitioning that reflects this metamorphic pattern (Goscombe & Gray, in preparation). The Escape Zone represents an inverted sequence of Barrovian-style, high-P/moderate-T metamorphic rocks ranging progressively from sub-greenschist- in the Eastern Kaoko Zone to upper amphibolite-facies grades immediately adjacent to the Orogen Core. In contrast, the Orogen Core shows a complex pattern of metamorphism, ranging from lower amphibolite to granulite facies, but throughout with a consistently high average thermal gradient metamorphic style (Goscombe et al., 2003b).
Within the Orogen Core, the penetrative S–L fabric is continuous with, and dragged into, the steep, broad (1–5 km) high-strain zones with mylonitic to ultramylonitic fabrics (Fig. 1). These shear zones controlled the development of the gross orogen architecture (Goscombe et al., 2003a, 2003b; Goscombe & Gray, in preparation). Transpressional strain was partitioned into two crustal-scale sinistral shear zones either side of the Orogen Core: the Purros Mylonite Zone on the east and Three Palms Mylonite Zone on the west. The gross geometry of the Orogen Core is a steeply inclined divergent flower form in the north and moderately inclined half-flower form in the south (Goscombe & Gray, in preparation). There is a monotonous southward decrease in shear zone inclinations from sub-vertical in the north to westerly dips as shallow as 30° in the south, which suggests an overall listric form at depth with a westerly dip (Goscombe & Gray, in preparation). These major bounding shear zones are kinematically contrasted; the Purros Mylonite Zone experienced oblique, reverse transport along shallow north-plunging lineations and the Three Palms Mylonite Zone experienced extensional, strike-slip transport along shallow south-plunging lineations (Goscombe & Gray, in preparation). This pair of kinematically contrasted shear zones accommodated lateral extrusion of the Orogen Core upwards and southwards along shallowly inclined trajectories (Goscombe & Gray, in preparation). The kinematic and streatching lineation array for the entire Kaoko Belt, and its correlation with the pattern of metamorphism is described below.
|
METAMORPHIC PATTERNS |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Methods
The distribution of metamorphic mineral parageneses, P–T paths and peak metamorphic conditions in the Hoarusib corridor across the Kaoko Belt have been reported by Goscombe et al. (2003b)
. We have extended this work to cover the entire Kaoko Belt, in particular the Coastal Terrane, northern Khumib and Hartmann Domains in the north and the Hoanib corridor in the south (Fig. 1). The methods we have applied to establish this extension of our dataset are the same as those described in detail by Goscombe et al. (2003b). The petrography of the central Kaoko Belt samples (Hoarusib corridor) have been described in detail by Goscombe et al. (2003b) and the relationships between mineral parageneses and structural fabrics for the Kaoko Belt, in general, have been described by Goscombe et al. (2003a, 2003b) and Goscombe & Gray (in preparation). Samples from the newly investigated regions are described below (Fig. 2); their petrography is summarized in Table 1 and mineral chemistry in Table 2. New peak metamorphic conditions (Figs 3–6) have been established using the average P–T calculation method of Powell & Holland (1994), and THERMOCALC v3.1 (Powell & Holland, 1988) with the 1998 thermodynamic dataset (Powell et al., 1998). The results are listed in Electronic Appendix A (which may be downloaded from the Journal of Petrology website at http://www.petrology.oupjournals.org) and summarized in Table 3. These calculations use mineral end-member activities defined by Holland & Powell (1990) and calculated using the program AX. [AX is an independent program (Powell et al., 1998) which we used to calculate the non-ideal mineral activities (using the models of Holland & Powell, 1990). These mineral activity values then constitute the input data for the program THERMOCALC.]
|
|
|
|
|
|
|
|
The rationale behind interpretation of the textural relationships between mineral phases used for calculation of metamorphic conditions is the same as that outlined by Goscombe et al. (2003b)
. Garnet compositional profiles show (Fig. 7) that Barrovian-style mid-amphibolite samples in the Khumib Domain, Escape Zone and eastern Hartmann Domain all have typical, prograde growth zoning (Spear, 1993) and rarely also develop thin Mn-enriched rims as a result of garnet resorption (Kohn & Spear, 2000). Consequently, peak metamorphic average P–T calculations utilized the outermost rim of prograde garnet growth, being careful to avoid the thin resorption rim, in conjunction with matrix mineral cores (Spear, 1993). Garnets from the high-grade Hoarusib and Hartmann Domains and M1 mineral parageneses in the Coastal Terrane are compositionally uniform and typical of garnets that have been homogenized at high metamorphic grade (Tuccillo et al., 1990; Spear, 1993). We assume that adjacent grains approached equilibration during this homogenization stage and that this corresponds closely to the peak of metamorphism, or alternatively re-equilibration immediately after the metamorphic peak. Metamorphic conditions during reworking in the Coastal Terrane and in shear zones in the Orogen Core utilized the thin Mn-enriched rims of resorbed garnets in conjunction with cores of new minerals that grew in the reworking foliations.
|
, Mg cations; 
, Mn cations; •, Fe2+ cations; 
, Ca cations.The results of new P–T calculations (Table 3) from each region are presented in Figs 3–6. Within the errors of the THERMOCALC calculations, all of these results are consistent with the phase stability field of the developed matrix assemblage and thus considered plausible estimates of the equilibration conditions. Phase stability constraints on peak metamorphic conditions and P–T paths are summarized in Figs 8–10 and the critical constraints on P–T paths from all parts of the Kaoko Belt are presented in Table 4. The prograde portion of P–T paths is constrained by pre-peak mineral assemblages preserved as inclusions and also by garnet compositional zoning (Figs 7 and 8). The peak metamorphic portions of P–T paths are defined by both the THERMOCALC results and the stability field of matrix assemblages within published general pseudosections for average metapelite (White et al., 2001
; Johnson et al., 2003) and mafic rocks (Pattison, 2003). The sequence of mineral growth from inclusion mineral parageneses to matrix assemblages and to post-peak reaction textures and late-stage retrograde parageneses is used to constrain the P–T paths within these general pseudosections (Figs 9 and 10). The spatial distribution of the entire peak metamorphic dataset and summary P–T paths in each of five corridors across the belt are presented in Fig. 11. Only one thermobarometric method and thermodynamic dataset have been employed in all P–T calculations. Thus metamorphic gradients, both across the Kaoko Belt (Fig. 12) and along the length of the belt (Fig. 13) define an internally consistent spatial distribution of metamorphic parameters throughout the entire belt.
|
|
|
|
|
|
|
Petrography
The central portion of the Hartmann Domain is of upper amphibolite grade and shows a gradual decrease in grade both towards Barrovian-style rocks at the eastern margin and towards the western margin adjacent to the reworked Coastal Terrane (Fig. 11). Metapelite samples on the western margin of the Hartmann Domain have matrix metapelite assemblages of garnet–phlogopite–quartz–oligoclase ± K-feldspar ± muscovite ± ilmenite; some also have quartzo-feldspathic leucosome segregations. A garnet-free sample contains muscovite porphyroblasts with aligned sillimanite and biotite inclusions and an enveloping matrix foliation of quartz–oligoclase–muscovite–biotite that is devoid of sillimanite (Table 1). Mafic schists consist of hornblende–plagioclase–ilmenite–titanite assemblages.
Metapelite schists in the eastern Hartmann Domain have coarse-grained matrix assemblages of garnet–biotite–muscovite–oligoclase–quartz ± kyanite ± rutile ± ilmenite–haematite solid solution (Table 1). Inclusion phases are staurolite and graphite within kyanite and ilmenite–biotite–tourmaline within garnet. Kyanite is typically corroded and enclosed by coarse-grained muscovite moats (Fig. 2f), with either oligoclase (Fig. 2e) or sillimanite, and also by monomineralic moats of oligoclase (Table 1). Primary rutile porphyroblasts in contact with quartz have an inner corona of kyanite and an outer corona of biotite. Garnet is post-kinematic and oligoclase rarely develops sub-grain margins of secondary oligoclase. There are at least two generations of muscovite: coarse matrix grains and late muscovite replacing kyanite or cutting across the biotite foliation (Table 1). Mafic gneiss contains actinolitic hornblende–diopside–quartz–oligoclase–K-feldspar–biotite–magnetite–titanite matrix assemblages. Calc-silicate samples have matrix assemblages of quartz–plagioclase–garnet–epidote–hornblende–ilmenite–titanite ± clinopyroxene ± scapolite (Table 1). Epidote–hornblende aggregates partially replace primary garnet and clinopyroxene and secondary garnet occurs as coronas around ilmenite (Table 1).
The highest-grade metapelite and quartz-rich metapelite samples of the central Hartmann Domain have matrix assemblages of garnet–phlogopite–quartz–andesine–K-feldspar–ilmenite and are devoid of muscovite, and also contain sillimanite, pseudomorphed cordierite and quartzo-feldspathic leucosome segregations (Table 1). Aggregates of chlorite–sillimanite ± biotite and sillimanite–quartz–orthoamphibole are interpreted to be pseudomorphs after cordierite (Fig. 2b). These pseudomorphs occur as lensoidal aggregates and also as partial moats at the margin of garnet porphyroblasts. Garnet porphyroblasts are pre-kinematic to syn-kinematic and two stages of garnet growth is uncommon. Garnets are commonly rounded (Fig. 2g) and embayed by significant resorption (Fig. 2h), often with accompanying growth of oligoclase moats (Fig. 2h) and sillimanite ± biotite (Fig. 2i) or oligoclase–biotite aggregates. Garnet is rounded and enveloped by foliation assemblages consisting of biotite–oligoclase–quartz–ilmenite ± rutile ± sillimanite (Fig. 2g). Andesine matrix plagioclase rarely develops sub-grain margins of secondary oligoclase plagioclase and is retrogressed to muscovite. Matrix biotite and garnet are partially retrogressed to chlorite.
Calc-silicate and calc-pelite samples in the central Hartmann Domain contain garnet–quartz–plagioclase–magnetite–clinopyroxene ± biotite ± K-feldspar ± calcite matrix assemblages (Table 1). Epidote occurs as a matrix phase in calc-pelite and as coronas around magnetite, calcite and clinopyroxene in calc-silicates. Mafic gneisses have either hornblende–plagioclase–clinopyroxene–ilmenite–titanite or garnet–hornblende–clinopyroxene–quartz–plagioclase–biotite–ilmenite–titanite matrix assemblages (Table 1). Primary garnet is extensively embayed and replaced by growth of plagioclase and epidote (Fig. 2c). Hornblende is ferroan pargasitic hornblende and plagioclase has labradorite–bytownite compositions. Clinopyroxene is salite and is partially replaced by either hornblende or epidote–clinozoisite solid solution. Hornblende is replaced by epidote or actinolite and ilmenite is enclosed by titanite coronas (Table 1).
All new metapelite samples from the Nadas corridor, across the Khumib Domain (Fig. 11), have matrix assemblages of garnet–biotite–quartz–plagioclase–ilmenite ± muscovite and are devoid of matrix K-feldspar. Metamorphic grade increases from east to west. The lowest grade samples contain inclusion assemblages of staurolite–plagioclase–biotite–muscovite–ilmenite–quartz (Fig. 2a) and two stages of garnet growth are evident by porphyroblasts with garnet overgrowths. To the west, quartzo-feldspathic leucosome segregations become common and sillimanite occurs within the foliation (Table 1). Sillimanite overprints both garnet porphyroblasts and biotite and is also aligned with the foliation. Most muscovite and all chlorite is late stage and overprints and partially replaces garnet, plagioclase and biotite (Table 1). Matrix plagioclase is oligoclase–andesine and rarely develops sub-grain margins of secondary oligoclase.
Coastal Terrane metapelites and meta-greywackes had pre-existing coarse-grained and high-grade matrix assemblages that have been reworked to varying degrees by ductile grain refinement and formation of a fine-grained foliation. M1 mineral parageneses are preserved as relic porphyroclasts; these are now isolated by the enveloping foliation and also by coarse-grained polygonal granoblastic assemblages in low-strain domains such as boudins and quartzo-feldspathic leucosome segregations. Relic M1 phases include quartz, garnet (Fig. 2j and l), oligoclase–andesine plagioclase (Fig. 2k), sillimanite, K-feldspar, biotite, haematite, ilmenite and migmatitic segregations (Table 1; Goscombe et al., 2003b). The M2 proto-mylonitic to mylonitic foliation has assemblages of biotite–muscovite–quartz–oligoclase ± K-feldspar and is devoid of sillimanite and garnet. Muscovite–biotite aggregates partially replace primary garnet (Fig. 2l) and the growth of new muscovite and biotite indicates that reworking involved hydration of the Coastal Terrane. Within the Three Palms Mylonite Zone the ductile foliation is in part, further reactivated by semi-ductile grain refinement of quartz, feldspar and mica to fine sub-grains in discrete foliation seams.
The Hoanib corridor in the south shows a similar metamorphic pattern to the Hoarusib corridor described in detail by Goscombe et al. (2003b). In both corridors the Escape Zone to the east of the Purros Mylonite Zone is an inverted Barrovian metamorphic series and to the west in the Orogen Core are high-grade migmatitic gneisses (Dingeldey, 1997). Metapelite gneisses in the Orogen Core have matrix assemblages of garnet–quartz–oligoclase ± K-feldspar ± cordierite ± sillimanite with migmatitic leucosome segregations (Table 1). All cordierite has been pseudomorphed to chlorite–sillimanite–magnetite or chlorite–muscovite–biotite aggregates. Sillimanite is both a primary phase and occurs in sillimanite–biotite aggregates corroding garnet and sillimanite–muscovite aggregates after plagioclase (Table 1). Garnet and biotite are retrograded by muscovite and chlorite. Metapelite gneiss in the highest-grade western Escape Zone has a matrix assemblage of garnet–biotite–quartz–oligoclase–K-feldspar. Metapelites in the eastern Escape Zone have garnet–biotite–muscovite–quartz–oligoclase matrix assemblages. These lower-grade samples have two-stage garnet with idioblastic overgrowths and garnet and biotite is partially replaced by chlorite (Table 1). The petrography of further samples from the Escape Zone within the Hoanib corridor has been described in detail by Dingeldey (1997).
Mineral chemistry
Methods
Mineral analyses from the Kaoko Belt were performed using a Cameca SX51 electron microprobe at Adelaide University. An operating voltage of 15 kV and beam current of 20 nA were used for all phases except micas (10 nA) and feldspar (15 nA) and a beam radius of 2 µm for most phases and 4 µm for micas and feldspars. Natural silicates were used as standards and checked periodically. Table 2 summarizes the full compositional range displayed by metapelite and metabasic samples from each metamorphic domain. Previous mineral analyses, largely from the Hoarusib corridor, have been described by Goscombe et al. (2003b).
Garnet. Garnets from Barrovian metapelites in the east Hartmann Domain and Khumib Domain show different growth zoning patterns with no significant resorption (Table 2; Fig. 7). Mole fractions in the octahedral site of east Hartmann Domain garnets have a total compositional range of XFe2+ 0·71–0·81, XMg 0·08–0·15, XCa 0·01–0·06 and XMn 0·06–0·15, with Fe2+ and Mn-rich rims and Mg-poor rims. Khumib Domain garnets vary greatly; higher-grade samples have rims poor in Fe2+ and Mg and rich in Mn, whereas low-grade samples show the opposite trends (Table 2). Higher-grade Khumib Domain garnets have a total compositional range of XFe2+ 0·69–0·74, XMg 0·06–0·18, XCa 0·01–0·02 and XMn 0·07–0·24. Low-grade Khumib Domain garnets have a total compositional range of XFe2+ 0·54–0·66, XMg 0·10–0·18, XCa 0·02–0·09 and XMn 0·15–0·28. Garnets from high-grade domains in the Orogen Core and Coastal Terrane have homogeneous compositions with thin Mn-rich and Mg-poor rims. The rims have low Fe2+ in the Coastal Terrane and high Fe2+ in the Hoarusib and central Hartmann Domains (Table 2; Fig. 7). Hoarusib and central Hartmann Domain garnets have a total compositional range of XFe2+ 0·58–0·83, XMg 0·06–0·39, XCa 0·01–0·03 and XMn 0·01–0·16. Relic garnets from the Coastal Terrane have a total compositional range of XFe2+ 0·69–0·79, XMg 0·05–0·24, XCa 0·01–0·04 and XMn 0·03–0·21.
Biotite. Biotite in Barrovian metapelite in the east Hartmann Domain is weakly zoned (Table 2), with Mg/(Fe + Mg) ratios ranging from 0·46 to 0·50 and 1·88–2·74 wt % TiO2. Higher-grade samples in the central and west Hartmann Domain have phlogopite that is weakly zoned with Mg/(Fe + Mg) ratios in the range from 0·47 to 0·75 and wt % TiO2 from 1·61 to 4·44. Khumib Domain biotite is unzoned; Mg/(Fe + Mg) ratios range from 0·41 to 0·62 and wt % TiO2 from 0·64 to 2·03. Fine-grained biotite in the low-grade foliation in the Coastal Terrane has 1·05–3·18 wt % TiO2 and Mg/(Fe + Mg) ratios of 0·32–0·56. Coarser early biotites in low-strain domains have 2·22–3·06 wt % TiO2 and 0·32–0·43 Mg/(Mg + Fe). Biotites from the high-grade Hoarusib Domain are essentially unzoned, with 1·18–3·00 wt % TiO2 and Mg/(Fe + Mg) ratios of 0·43–0·54.
Muscovite. All metapelite muscovites are weakly zoned and Na-poor, with a total range of 0·84–0·97 for the K/(K + Na + Ca) ratio. Muscovites from the Barrovian eastern Hartmann Domain have 3·11–3·20 Si-cations, K/(K + Na + Ca) ratios of 0·84–0·87 and Mg/(Fe + Mg) ratios of 0·56–0·74. The Coastal Terrane has the largest range in muscovite composition, with Si-cations ranging from 2·89 to 3·22, Mg/(Fe + Mg) ratios ranging from 0·50 to 0·78 and K/(K + Na + Ca) ratio ranging from 0·86 to 0·97. Khumib Domain muscovites have 3·07–3·12 Si-cations, K/(K + Na + Ca) ratios of 0·87–0·95 and Mg/(Fe + Mg) ratios of 0·70–0·77.
Chlorite. Retrograde ripidolite chlorites in both the low-grade Khumib and high-grade Hoarusib Domains have similar Mg/(Fe + Mg) ratios of 0·54–0·57 and Al/(Al + Si) ratios of 0·39–0·50.
Feldspar. Matrix oligoclase to andesine plagioclase in metapelites from all parts of the Hartmann and Khumib Domains show zoning with increasing anorthite component towards the rims. Xan in plagioclase ranges from 0·12 to 0·26 in the east and west margins of the Hartmann Domain and from 0·25 to 0·44 in the high-grade central Hartmann Domain. Plagioclase in the Khumib Domain has Xan ranging from 0·15 to 0·26 in the east and from 0·19 to 0·34 in higher-grade metapelites in the west. Inclusion plagioclase in the Khumib Domain has a higher Xan content and plagioclase sub-grains within foliation parageneses have lower Xan (Table 2). The Xor content in K-feldspar from all domains falls in the narrow range 0·85–0·98, except for high-grade samples in the central Hartmann Domain, which have Xor values of 0·71–0·84. Plagioclase in the high-grade Hoarusib Domain is homogeneous, with Xan ranging from 0·28 to 0·29; in the reworking foliation Xan ranges from 0·20 to 0·29. Primary, coarse-grained plagioclase in the Coastal Terrane is weakly zoned and the total range in Xan for all samples is 0·18–0·40. Plagioclase sub-grains in the reworking foliation have a more restricted Xan range of 0·27–0·36.
Peak metamorphic conditions
The Barrovian-style metamorphic rocks of the Escape Zone show a progressive increase in grade westward to higher structural levels, defining an inverted metamorphic sequence (Figs 11 and 12). Average P–T calculations quantify this progressive increase in metamorphic grade and also show that metamorphic conditions do not vary substantially along strike within the Escape Zone (Figs 11 and 13; Table 3). Escape Zone garnets within metapelites have typical prograde growth zoning and thin Mn-rich resorption rims. Peak metamorphic P–T calculations utilized the outermost preserved prograde garnet in conjunction with matrix mineral cores (Spear, 1993; Kohn & Spear, 2000) and assumed XH2O = 1·0. From east to west, peak metamorphic results give pooled averages of 545 ± 40°C and 8·9 ± 1·2 kbar (n = 7) in the garnet zone, 660 ± 45°C and 8·5 ± 1·6 kbar (n = 5) in the staurolite zone and 690 ± 50°C and 8·4 ± 1·3 kbar (n = 6) in the kyanite zone (Table 3; Fig. 11) (Goscombe et al., 2003b). Given the large range in P–T calculation errors and low number of results, the pooled averages presented throughout this paper are simply the average of P–T results and average of the errors (Table 3). This simple pooling method is used only where there is overlap between error ellipses and gives a realistic P–T field encompassing all calculations. The pooled results for parts of the Escape Zone show a progressive rise in temperature in parallel with the mineral isograds and are consistent with the phase stability fields (Ghent et al., 1979; Powell & Holland, 1990; Johnson et al., 2003) of the matrix mineral parageneses (Goscombe et al., 2003b). Across the Escape Zone pressures are consistently 8–9 kbar. Consequently, there is a progressive increase in average thermal gradient (G) from around 17°C/km in the east to around 24°C/km in the west (Table 3). The highest-grade rocks in the westernmost Escape Zone give P–T results from resorbed garnet rim assemblages with a pooled average of 625 ± 40°C and 6·3 ± 1·5 kbar (n = 2) (Table 3), suggesting decompression and cooling after the peak of M2 metamorphism.
In common with the Escape Zone, the Khumib Domain within the Orogen Core has metapelite garnets with typical prograde growth zoning and thin Mn-rich resorption rims. Average P–T calculations for the metamorphic peak used the outermost prograde garnet inside the resorbed rim, in combination with matrix mineral cores (Spear, 1993; Kohn & Spear, 2000) and assumed XH2O = 1·0. Average P–T calculations span a wide array of temperatures and give consistently low pressures in both the Nadas and Khumib corridors across the Khumib Domain (Fig. 4). Most average P–T calculations centre on pooled conditions of 600 ± 24°C and 5·2 ± 1·2 kbar (n = 5) in both the Khumib and Nadas corridor (Fig. 4), corresponding to average thermal gradients of around 33°C/km (Table 3). Metamorphic grade increases steeply towards the western margin of the Khumib Domain, where temperatures of 705 ± 45°C and 660 ± 55°C are calculated from the Khumib and Nadas corridors, respectively. Pressures in the high-grade western margin are similar to elsewhere in the Khumib Domain, being 4·2 ± 1·4 kbar and 4·7 ± 2·0 kbar in the Khumib and Nadas corridors, respectively (Fig. 4), indicating very high average thermal gradients around 40–48°C/km (Table 3).
In the Orogen Core, the Hoarusib Domain garnets have uniform compositional profiles (Goscombe et al., 2003b) indicating homogenization at high metamorphic grades (Spear, 1993). We have assumed that matrix mineral compositions were homogenized at, or immediately after, the peak of high-grade metamorphism. Compositional variation in most minerals is restricted to thin outer rims that are interpreted to have formed during post-peak partial re-equilibration (Frost & Chacko, 1989). Consequently, in these high-grade rocks, garnet cores and the cores of adjacent matrix minerals were used for average P–T calculations to approximate peak metamorphic conditions and avoid the deleterious effects of post-peak re-equilibration of the mineral grain margins. All high-grade samples have leucosome segregations (Table 1), suggesting the presence of partial melting and absence of a free fluid phase at the peak of metamorphism. Consequently, an assumed value of aH2O is required for all average P–T calculations. Estimates of aH2O from anhydrous granulites in the literature have reported values of 0·1–0·3 (B. Goscombe, unpublished data, 2003). The high-grade metapelite and mafic gneisses in the Orogen Core of the Kaoko Belt are not anhydrous granulites but have hydrous matrix assemblages with evidence for partial melting. Consequently, aH2O values of between 0·3 and 0·7 have been assumed (Electronic Appendix A; Goscombe et al., 2003b). Pooled average conditions were 845 ± 65°C and 8·1 ± 1·6 kbar (n = 4) in the eastern Hoarusib Domain, 810 ± 60°C and 6·2 ± 0·7 kbar (n = 2) in the western Hoarusib Domain and approximately 780 ± 55°C and 6·7 ± 1·6 kbar (n = 2) in the Hoanib corridor to the south, indicating average thermal gradients of around 30–38°C/km (Table 3). These results are consistent with the stability fields of garnet–cordierite–K-feldspar–melt ± sillimanite and garnet–sillimanite–K-feldspar–melt metapelite assemblages (White et al., 2001; Fig. 9) and orthopyroxene–clinopyroxene–hornblende–garnet–plagioclase mafic assemblages (Kohn & Spear, 1990).
Like the Hoarusib Domain, compositional profiles across garnets within the high-grade core of the Hartmann Domain (Fig. 11), have flat to nearly flat cores with thin Fe2+-rich and Mg-poor rims (Fig. 7). Peak metamorphic average P–T calculations using matrix mineral cores and assumed aH2O values of 0·3–0·7 (see above) give results ranging from 800 ± 55°C and 6·8 ± 1·6 kbar to 710 ± 55°C and 9·3 ± 1·8 kbar (Figs 3 and 10), corresponding to average thermal gradients ranging from 22 to 33°C/km (Table 3). These conditions are consistent with the stability fields of the developed garnet–K-feldspar–biotite–sillimanite–melt metapelite matrix assemblages (White et al., 2001; Fig. 9) and hornblende–clinopyroxene–plagioclase–ilmenite ± garnet mafic assemblages (Spear, 1981; Pattison, 2003; Fig. 10).
The eastern margin of the Hartmann Domain, within the Purros Mylonite Zone, experienced Barrovian metamorphic conditions. Garnets preserve typical growth zoning compositional profiles of decreasing Ca and Mn and increasing Fe and Mg towards rims. Peak metamorphic average P–T calculations used the rims of prograde growth in garnets, inward from thin Mn-rich resorbed margins where present, in combination with matrix mineral cores (Spear, 1993; Kohn & Spear, 2000; Goscombe et al., 2003b). The average P–T results for an assumed XH2O = 1·0 range from 653 ± 14°C, 8·8 ± 0·7 kbar to 620 ± 44, 6·3 ± 1·5 kbar (Fig. 3; Table 3). The pooled averages of these P–T calculations are 640 ± 25°C and 7·4 ± 1·2 kbar (n = 4), giving an average thermal gradient around 25°C/km. These results are consistent with the stability field of the developed metapelite matrix assemblages of kyanite–garnet–plagioclase–biotite–muscovite–quartz (Powell et al., 1998; Johnson et al., 2003; Fig. 9).
A diverse group of mineral parageneses in the high-grade domains of the Orogen Core represent late M2 reworking and continued shearing in the Purros Mylonite Zone and other shear zones. These include conditions of 585 ± 85°C and 4·4 ± 1·2 kbar from a late-stage, silica-undersaturated spinel–corundum–plagioclase–biotite symplectite in a Hoarusib Domain semipelite sample (Table 3). Calculations from Purros Mylonite Zone samples, and reworking parageneses using matrix mineral rims and the outer resorbed, Mn-enriched garnet rims, give pooled averages of 627 ± 30°C and 5·2 ± 1·3 kbar (n = 4) in the east Hartmann Domain and 630 ± 70°C and 4·2 ± 1·8 kbar (n = 2), in the Hoarusib Domain (Table 3). Collectively, these late M2 average P–T results are centred on a pooled average of 615 ± 46°C and 4·7 ± 1·4 kbar (n = 8). These calculations are consistent with the stability field of garnet–biotite–sillimanite–plagioclase–muscovite assemblages (Powell & Holland, 1990; Johnson et al., 2003; Fig. 9), typical of the overprinting foliation seams in these domains (Goscombe et al., 2003b; Table 1). These results are mutually consistent over the length of the Orogen Core and represent late M2 conditions during progressive reworking of the Orogen Core, particularly in shear zones. These P–T conditions indicate that transpressional shearing continued after significant post-peak cooling and exhumation and are broadly similar to conditions of reworking of the Coastal Terrane during transpressional orogenesis (see below; Table 3; Figs 6 and 9).
Coastal Terrane garnets are highly resorbed relics (Fig. 2j and l) with compositionally homogeneous cores and thin Mn-rich rims, formed by intracrystalline diffusion during garnet resorption (Kohn & Spear, 2000; Goscombe et al., 2003b). The compositionally uniform cores are interpreted to be due to homogenization at high grades (Tuccillo et al., 1990; Spear, 1993). Coarse-grained primary biotite and feldspar in low-strain domains also preserve compositionally uniform cores. Consequently, the cores of relic M1 mineral grains are interpreted as preserving the composition of the M1 equilibration assemblage. Average P–T calculations utilizing the homogenized cores of relic garnet, feldspar and coarse-grained biotite preserved in low-strain domains represent conditions at, or immediately below, the peak of M1 metamorphism. The pooled average of M1 results is 725 ± 64°C and 6·9 ± 1·8 kbar (n = 2) and these results are consistent with the relic M1 meta-greywacke assemblage of garnet–biotite–plagioclase–K-feldspar–quartz–melt (Fig. 6). Pervasive reworking of the Coastal Terrane gave rise to ductile grain refinement and formation of foliation seams (Fig. 2j and k) and hydrous retrogression (Fig. 2l). The M2 equilibrium assemblage is represented by Mn-rich outer-rims of resorbed garnet, cores of biotite and muscovite in the enveloping foliation and cores of plagioclase sub-grains within aggregate ribbons. M2 results give a pooled average of 565 ± 50°C and 4·5 ± 1·2 kbar (n = 5), corresponding to an average thermal gradient around 37°C/km (Table 3; Fig. 6).
P–T evolution
Clockwise P–T paths have been documented from the Barrovian-style Escape Zone in the Hoarusib corridor (Goscombe et al., 2003b). Similar phase relationships have been recognized in the Hoanib corridor to the south (Table 4) and clockwise P–T paths are also documented (Fig. 5; Dingeldey, 1997). Escape Zone P–T paths are constrained by the sequence of mineral growth preserved in metapelites and interpreted within general P–T pseudosections for average metapelite compositions (Goscombe et al., 2003b). The diagnostic sequence of mineral growth in the staurolite zone is from early chlorite–ilmenite–quartz inclusion assemblages within garnet, prograde garnet–biotite–plagioclase assemblages and garnet–biotite–staurolite matrix assemblages. Peak garnet is resorbed by continued staurolite–biotite growth and complete overgrowth of garnet by staurolite, indicating decompression in KFMASH pseudosections for typical metapelites (Powell et al., 1998; Vance & Mahar, 1998; Spiess et al., 2000; White et al., 2000). Plagioclase coronas on garnet, late-stage fibrolite after garnet and staurolite and peak metamorphic kyanite overgrown by sillimanite (Goscombe et al., 2003b) all indicate a component of decompression subsequent to formation of the matrix mineral assemblages (Harley, 1992; Powell et al., 1998; Johnson et al., 2003). Plagioclase moats around resorbed garnets in metabasic rocks also indicate decompression (Kohn & Spear, 1990). These metapelite and mafic phase relationships document clockwise P–T paths followed by decompressive cooling and retrograde chlorite and muscovite growth (Dingeldey, 1997; Goscombe et al., 2003b). Prograde burial trajectories are documented by the early chlorite-bearing parageneses and by garnet compositional profiles (Goscombe et al., 2003b). Escape Zone garnets in metapelites show typical growth zoning of increasing Fe and Mg and decreasing Ca and Mn towards the rim (Spear, 1993). Prograde P–T trajectories of increasing P with rising T are indicated by garnet compositional isopleths in P–T pseudosections for typical metapelites (Vance & Mahar, 1998).
The Barrovian-style eastern margin of the Hartmann Domain straddles the Purros Mylonite Zone and has almost identical clockwise P–T evolution to the Escape Zone. Garnet compositional profiles show typical growth zoning of increasing Fe and Mg and decreasing Mn and Ca from cores to rims. At the mid-amphibolite conditions experienced, these document prograde paths with shallow positive P/T trajectories when compared with garnet compositional isopleths for typical metapelites (Fig. 8; Spear, 1993; Vance & Mahar, 1998). The sequence of mineral growth preserved in metapelite samples constrains a clockwise P–T path (Fig. 9; Table 4). Early staurolite-bearing parageneses (Fig. 2d) are overgrown by coarse-grained matrix assemblages of garnet–kyanite–plagioclase–biotite–muscovite–quartz (Table 4). Peak metamorphic kyanite is resorbed by plagioclase–muscovite (Fig. 2e and f) and sillimanite–plagioclase–muscovite parageneses, followed by retrogressive chlorite and muscovite growth (Table 4). This sequence must have involved a prograde component from the staurolite field to peak metamorphic parageneses and a component of decompression throughout (Figs 3 and 9; Powell et al., 1998; Vance & Mahar, 1998; Johnson et al., 2003). Average P–T calculations also document a decompressive cooling path from peak M2 assemblages to late M2 reworked assemblages, which equilibrated at significantly lower T and P (Fig. 3; Table 3).
The low-grade Khumib Domain within the Orogen Core shows some similarities to the Escape Zone. Most metapelite garnet compositional profiles show typical growth zoning of decreasing Mn and Ca and increasing Fe and Mg. At mid-amphibolite conditions these indicate steep P/T prograde burial trajectories when compared with garnet compositional isopleths for average metapelite (Spear, 1993; Vance & Mahar, 1998). A single sample shows flat Fe2+, increasing Mn and decreasing Mg as a result of intracrystalline diffusion in the garnet rim, indicating decompression during post-peak re-equilibration (Fig. 8). In contrast to the Escape Zone, peak metamorphic conditions were reached at significantly lower pressures of 5·2 ± 1·1 kbar, but at similar temperatures, indicating much tighter P–T paths with shallower P/T trajectories (Figs 4 and 9). Peak metamorphic conditions were in the low- to mid-amphibolite facies, typified by metapelite assemblages of garnet–biotite–muscovite–oligoclase–quartz (Fig. 9). The sequence of mineral parageneses and P–T calculations suggest that the P–T paths were clockwise (Table 4). Early staurolite–plagioclase–muscovite–biotite inclusion parageneses are overgrown by peak parageneses devoid of staurolite, indicating decompression (Figs 4 and 9; Tinkham et al., 2001; Johnson et al., 2003). Decompression and/or cooling is indicated by the resorption of garnet (Spear, 1993) and by replacement of garnet by sillimanite–biotite growth in higher-grade parts of the Khumib Domain. Plagioclase in metapelites develops overgrowths (Table 1) and rims are relatively Ca-rich (Table 2), indicating increasing modal plagioclase and Xan as a result of decompression and/or cooling (Spear, 1993).
The high-grade Hoarusib and Hartmann Domains within the Orogen Core share very similar phase relationships (Table 4) that constrain tight clockwise P–T paths with shallow P/T trajectories (Figs 3 and 9; Goscombe et al., 2003b). The diagnostic prograde sequence of mineral growth in Hoarusib Domain metapelites (Goscombe et al., 2003b) is from early kyanite inclusions within cordierite to peak metamorphic assemblages of garnet–cordierite–K-feldspar–melt–sillimanite ± biotite, indicating either isobaric heating or decompressional prograde paths (Fig. 9; White et al., 2001). Decompression through the peak of metamorphism is indicated by cordierite moats around peak metamorphic garnet, followed by near-isobaric cooling indicated by sillimanite–biotite aggregates replacing cordierite. Similarly, plagioclase coronas around garnet in mafic gneiss indicate decompression (Goscombe et al., 2003b). Post-peak decompressional cooling is supported by the development of silica-undersaturated corundum–spinel–biotite–plagioclase symplectites after garnet in semi-pelite gneiss, which formed well below peak metamorphic conditions, at 4·4 ± 1·2 kbar and 585 ± 85°C (Goscombe et al., 2003b). Furthermore, reworking of the Hoarusib Domain formed late-stage fabrics and shear zones, with assemblages that equilibrated at pressures centred on 4·2 ± 1·8 kbar and temperatures between 690 ± 45°C and 565 ± 95°C (Table 3), further constraining the clockwise P–T path (Fig. 5).
High-grade metapelites in the central Hartmann Domain also contain partial cordierite moats at garnet margins and both cordierite and garnet are resorbed by sillimanite and biotite growth, followed by retrograde chlorite and muscovite (Fig. 9). Some garnet cores preserve weak growth zoning compositional profiles (Fig. 7) of increasing Fe and decreasing Mg (Fig. 7). These indicate shallow negative P/T prograde trajectories based on general garnet compositional isopleths for average metapelite (Fig. 8; Spear, 1993; Vance & Mahar, 1998). Other Hartmann Domain garnets have high-grade homogenized cores, with distinct Fe2+-rich and Mg-poor re-equilibrated rims (Fig. 7), indicating significant decompression during post-peak re-equilibration (Fig. 8). Decompression is also documented by plagioclase–epidote corrosion of garnet in mafic gneiss (Fig. 10). Metapelite garnets are resorbed and enveloped by sillimanite–plagioclase–biotite–quartz parageneses (Fig. 2g), and are also corroded by plagioclase, plagioclase–biotite and sillimanite coronas and reaction textures (Table 4), all indicating a component of post-peak decompression (Figs 3 and 9; White et al., 2001). Modal garnet and plagioclase isopleths for average metapelite (Spear, 1993) indicate that garnet resorption and secondary plagioclase growth (Table 1) must involve decompression and/or cooling.
The Coastal Terrane was so extensively reworked during transpressional orogenesis that M1 P–T paths cannot be documented from the relic mineral parageneses preserved. Matrix mineral parageneses and P–T calculations constrain the respective M1 and M2 peak metamorphic conditions (see above; Table 3). M2 parageneses in meta-greywacke are defined by Mn-enriched garnet rims and biotite–muscovite–oligoclase–quartz in reworking foliations. These parageneses overprint migmatitic high-grade assemblages; P–T calculations (Table 3; Fig. 6) indicate that reworking was at significantly lower temperatures (T = 150°C) and pressures (P = 2·4 kbar). There are, broadly, two alternative theoretical trajectories between the M1 and M2 metamorphic peaks (Table 4), and unfortunately there is insufficient textural evidence preserved in these rocks confidently to reconstruct this P–T history. A possibility is a protracted period (t = 75–100 Myr) of near-isobaric to shallow decompressional cooling from the M1 peak at 650 Ma to M2 reworking at 575–550 Ma. Alternatively, the terrane cooled to a normal geotherm rapidly after M1 metamorphism and experienced a later, unrelated prograde metamorphic cycle during M2 (Fig. 6). Water introduced into the Coastal Terrane during M2 reworking was sourced from the structurally lower Orogen Core, which was undergoing a prograde metamorphic cycle (M2) at this time.
In all profiles across the Kaoko Belt, each terrane shows similar P–T paths. The Escape Zone experienced clockwise P–T paths with steep burial and steep decompression trajectories, reflecting advection-dominated metamorphism that is typical of crustal overriding systems. In contrast, both high-grade and low-grade domains in the Orogen Core experienced tighter clockwise P–T paths with shallower P/T trajectories, indicating more conduction-dominated metamorphism (Fig. 11). Clockwise P–T paths were experienced in all parts of the Kaoko Belt, regardless of the average thermal gradient and kinematic and metamorphic style. These paths document the vertical component history of what must be complex particle paths involving early burial and later decompression through the thermal peak of metamorphism.
Metamorphic gradients
We cast a metamorphic gradient ‘net’ across the Kaoko Belt, composed of three longitudinal profiles, one each along the length (y) of each major terrane (Fig. 13), and five corridors across (x) the belt (Fig. 12). Metamorphic gradients are the variation in metamorphic parameters with respect to a horizontal length scale (x and y in km). The metamorphic parameters investigated are temperature (T in °C), pressure (P in kbar) and average thermal gradient defined as the ratio of temperature over depth (G in °C/km). By applying a consistent technique for determining the peak metamorphic conditions, involving mineral end-member activities (Holland & Powell, 1990; Powell et al., 1998), average P–T calculation using equilibrium thermodynamics (Powell & Holland, 1994) and the same 1998 thermodynamic dataset throughout (Powell et al., 1998), the results are directly comparable across the entire Kaoko Belt (Fig. 5). Previously, detailed metamorphic gradients were presented for a single corridor across the grain of the Kaoko Belt (Goscombe et al., 2003b). We have now increased this dataset to five broad corridors across the belt (Fig. 11) and with this a clear picture of metamorphic variation along the length of the orogen has emerged (Fig. 13). These metamorphic gradients quantify the peak metamorphic thermal pattern represented by the metamorphic isograd map (Fig. 11). This dataset goes further to quantify the barometric patterns of peak metamorphism and thus average thermal gradients, information that would otherwise not be available without robust absolute determinations of peak metamorphic conditions over such a wide region. Along-orogen variation in metamorphic parameters is rarely investigated (Goscombe et al., 2004); most being restricted to across-strike metamorphic gradients. This is the first such metamorphic gradient ‘net’ presented in the literature for any transpressional belt.
The metamorphic isograd map of the Kaoko Belt (Fig. 11) illustrates the distribution of peak M2 metamorphic temperatures. This illustrates a simple inverted metamorphic pattern in the Escape Zone of temperatures rising smoothly from sub-greenschist to kyanite grades at higher structural levels in the west (Fig. 11), defining an inverted metamorphic sequence (Goscombe et al., 2003b). A temperature rise from 450 to 700°C is quantified across the Escape Zone with an almost flat pressure distribution decreasing from 9 to 8 kbar. This gives a systematic increase in average thermal gradient from 15 to 25°C/km towards the high-grade Orogen Core (Fig. 13). Along the length of the western Escape Zone, peak metamorphic conditions remain similar with only small variations (Fig. 13). From north to south there is an apparent systematic increase in both pressure from 7 to 9 kbar and temperature from 640 to 690°C, and possibly also a small decrease in average thermal gradient from 25 to 22°C/km (Figs 11 and 13). Metamorphism of the Escape Zone is contrasted with the Orogen Core (Fig. 11), which typically, but not in all profiles, has lower pressures and higher temperatures, and higher average thermal gradients (Figs 12 and 13).
In contrast, the Orogen Core has a complex peak metamorphic pattern with two high-grade lobes that are connected along the western margin of the Orogen Core, and a low-grade metamorphic ‘trough’ on the east margin, between the high-grade lobes (Fig. 11). The metamorphic gradients (Fig. 13) show a complex longitudinal variation with an extreme range in peak metamorphic temperature and pressure. Nevertheless, the average thermal gradient remains high (28–40°C/km) throughout most of the Orogen Core and is distinct from the Escape Zone (Fig. 13). The northern high-grade lobe or Hartmann Domain equilibrated at 780–800°C at pressures ranging between 7 and 9 kbar, indicating average thermal gradients of 24–34°C/km. Very similar conditions are recognized in the southern high-grade lobe, where peak conditions in the Hoarusib Domain are in the range of 800–850°C and 6·0–8·5 kbar, indicating average thermal gradients of 29–38°C/km. Latitudinal metamorphic gradients across these high-grade lobes are similar, with temperature rising to a thermal maximum in the core (Figs 11 and 13). Typically there is an increase in pressure immediately west of the Purros Mylonite Zone and gradual decrease towards the Three Palms Mylonite Zone, resulting in a systematic increase in average thermal gradient across the high-grade lobes (Fig. 12).
The Khumib Domain in the Orogen Core is a low-grade ‘trough’ with latitudinal metamorphic gradients that contrast strongly with the high-grade lobes. From east to west peak metamorphic temperature falls steeply across the Purros Mylonite Zone to a minimum of 520°C, rising steeply to 705°C in the Khumib Mylonite Zone on the western margin (Fig. 11). Peak metamorphic pressures are consistently between 4·4 and 5·9 kbar with no apparent systematic pattern, although lower than all adjacent domains in the Orogen Core as well as the Escape Zone (Figs 11 and 12). Consequently, the Khumib Domain constitutes a true thermal and barometric trough. Despite the marked variation in metamorphic grade along the Orogen Core, similar average thermal gradients of 26–37°C/km are also documented in the Khumib Domain. Consequently, a common thermal regime appears to have existed throughout the Orogen Core, contrasting with the Escape Zone. The low pressures in the Khumib Domain indicate that the high-grade lobes and this low-grade ‘trough’ represent different crustal levels (z = 8–12 km) brought into juxtaposition subsequent to peak metamorphism in the M2 metamorphic cycle. Peak metamorphism in the Khumib and Hoarusib Domains both occurred at c. 575 ± 10 Ma, although at different crustal levels. Continued transpressional orogenesis in the M2 metamorphic cycle brought the Hourusib Domain adjacent to the Khumib Domain, indicating lower-crustal extrusion into mid-crustal levels.
There is no discernible spatial variation in the grade of late M2 transpressional reworking of the Coastal Terrane (Fig. 11). All average P–T calculations indicate narrow ranges of 535–605°C and 4·3–4·8 kbar, corresponding to metamorphic gradients around 37°C/km (Table 3). These results show no systematic variation along the length of the Coastal Terrane (Fig. 13). Late M2 reworking of the Coastal Terrane occurred at low grades, in sharp contrast to peak metamorphic conditions in the adjacent Orogen Core, differing by 
P = 1·7–3·6 kbar and T = 205–280°C. Conditions of late M2 reworking are substantiated by similar metamorphic grades in shear zones and late-stage reworking fabrics within the Orogen Core (Table 3; Figs 12 and 13). Consequently, the Coastal Terrane as well as the Khumib Domain represents mid-crustal levels that were juxtaposed against lower-crustal high-grade rocks of the Orogen Core (Fig. 11) late in the M2 metamorphic cycle.
|
MATERIAL FLOW TRAJECTORIES |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
To investigate material flow in the evolving middle and lower crust during transpressional orogenesis in the Kaoko Belt, vectors representing particle paths in each metamorphic domain need to be established. Representative particle paths are composed of three mutually orthogonal components of movement: two horizontal components being the longitudinal shear or wrench component (
y) and latitudinal shortening or contractional component (x) and a vertical component of exhumation (negative z) or burial (positive z). Information pertaining to each component is sourced from different aspects of geological investigation. Consequently, we have documented Transpressional Phase particle paths by integrating our structural datasets, such as orogen architecture, bulk strain, stretching lineation array and shear-sense indicators (Goscombe & Gray, in preparation), with the metamorphic datasets, such as peak metamorphic conditions, metamorphic gradients and P–T paths (this study). Information on each direction component is to some degree decoupled; the vertical component is largely represented by metamorphic and specifically barometric information, and conversely metamorphic information cannot impart significant constraints on the horizontal components. Information on the horizontal components is derived almost exclusively from structural datasets.
Kinematic-lineation array
Stretching lineation trajectories define a kinematic-lineation array for the entire Kaoko Belt (Fig. 14), which also combines the direction of shear determined from a variety of independent shear-sense indicators (Goscombe et al., 2003a). Two types of information are contained in the kinematic-lineation array plotted in Fig. 14: the trace of the stretching lineation in the horizontal plane and the local shear sense defined by transport of the upper plate, which is not independent of main foliation dip direction (Figs 12 and 13). A third type of information, the plunge and plunge direction of stretching lineations, is plotted in Fig. 14. This may contain information of the vertical component (z) of particle paths and vergence of transport. Together, all these parts of the kinematic-lineation array define apparent flow trajectories resulting from the wrench component of deformation (y). A contractional component (x) during Transpressional Phase deformation is evident in all domains making up the Kaoko Belt, including all shear zones and intervening panels in the Orogen Core (Goscombe & Gray, in preparation). The contractional component is expressed as the widespread Convergence Stage folding of the main pervasive foliation; these folds have sub-horizontal axes and axial surfaces parallel to the belt. The vector of contractional strain is defined as being approximately normal to fold axial surfaces. This contractional component is hard to quantify because folds are non-cylindrical and commonly rootless, and the contractional strain and shear strain components of bulk strain cannot be separated.
|
The kinematic-lineation array for the Orogen Core and Coastal Terrane (Fig. 14) shows smooth curvilinear traces that range from sub-parallel to acutely oblique to the grain of the belt and major shear zones (i.e. lineation obliquity or ßL
30°). These are continuous with an arcuate stretching lineation trace in the Escape Zone, which makes systematically higher ßL angles to the trace of the belt towards the margin (i.e. ßL up to 70–80°; Dürr & Dingeldey, 1996; Goscombe et al., 2003a). The higher-angle lineation trajectories are consistent with lateral escape as a result of large-scale nappes, thrusts and over-folds towards the margin. This broad-scale pattern is typical of all classic transpressional orogens such as the Mozambique Belt (Shackleton & Ries, 1984) and the Caledonides in Greenland (Holdsworth & Strachan, 1991; Strachan et al., 1992). The finer details of the Orogen Core and Coastal Terrane show a dominant orogen-parallel stretching lineation array; however, different domains show contrasting obliquity (±ßL) to the belt. These oblique domains also show contrasting plunge directions and either upward-vergent or downward-vergent lineations with respect to the overall sinistral shear sense (Fig. 14). The southern half of the belt is an oblique domain with NNW-plunging lineations, indicating low-angle oblique, upward and SSE-directed convergent vergence (i.e. +ßL; Fig. 14). Within the Orogen Core adjacent to the Purros Mylonite Zone, lineations are upward vergent along shallow NNW-plunging oblique lineations for most of its length including all of the south and extreme north (Fig. 14). Where the Purros Mylonite Zone is steep, the lineations swing into steeper orientations towards the Purros Mylonite Zone, indicating up-ramping convergent trajectories. In contrast, the northern portion of the Khumib Domain in the Orogen Core has stretching lineation trajectories that are obliquely divergent (i.e. –ßL) with SSW plunges, indicating oblique, downward and SSW-directed extensional vergence (Fig. 14). Similarly, the northern sector of the Coastal Terrane also has shallow downward-vergent and obliquely extensional stretching lineation arrays within and adjacent to the Three Palms Mylonite Zone. Consequently, the kinematic-lineation arrays in the Orogen Core and Coastal Terrane suggest relative downward transport of the northern Coastal Terrane and Khumib Domain and upward transport of the southern Coastal Terrane and the two high-grade domains within the Orogen Core (Fig. 14).
Metamorphic constraints
The metamorphic gradient ‘net’ (Figs 12 and 13) constrains the vertical component of particle paths (z) and this information implies lower-crust extrusion into middle-crust levels during transpressional orogenesis. Metamorphic gradients indicate barometric (P) differences between the different metamorphic domains in the Orogen Core and Coastal Terrane (Figs 12 and 13). Figure 14 shows that the metamorphic pattern in the Kaoko Belt is linked with the kinematic-lineation array. Domains of oblique upward- and downward-vergent stretching lineations imply vertical components of relative transport between domains and terranes in the western parts of the Kaoko Belt. The two high-grade lobes within the Orogen Core are coincident with an oblique upward and SSE-directed convergent kinematic-lineation array (Fig. 14), consistent with oblique extrusion of lower-crustal material into middle-crustal levels. Similarly, both the low-grade ‘trough’ within the Orogen Core and the Coastal Terrane coincide with oblique downward and SSW-directed extensional kinematic-lineation arrays (Fig. 14). These apparent flow trajectories are entirely consistent with barometric constraints of peak metamorphic pressures of 5·2 kbar in the Khumib ‘trough’, 4·5 kbar in the Coastal Terrane and 6·2–8·1 kbar in the Hoarusib and Hartmann high-grade lobes. All domains in the Orogen Core experienced heating-dominated prograde metamorphic paths followed by decompression from near-coincident Pmax and Tmax conditions (Figs 3, 9 and 11) and are entirely consistent with extrusional tectonics (Thompson et al., 1997b).
The pressure differential between peak metamorphism in the high-grade Hoarusib Domain and low-grade Khumib Domain is 2·7–3·0 kbar, corresponding to a 9·5–10·5 km component of relative vertical exhumation of the high-grade domain. The cores of these two domains are separated by 80 km lateral distance along orogen strike. The documented 10 km of relative vertical exhumation may be accommodated by 80 km of lateral transport upward along a trajectory that plunges 7·5° to the NNW. This hypothetical upward trajectory is identical to the average NNW plunge of stretching lineations in this region (Fig. 14). Consequently, both the kinematic-lineation array and metamorphic gradients in the Orogen Core are consistent with a mechanism of oblique upward and SSE-directed extrusion of the high-grade domains, along the stretching lineation trajectories. Similarly, the Coastal Terrane was reworked at significantly lower pressures than the adjacent Orogen Core, corresponding to 6–12 km of relative vertical exhumation of the Orogen Core with respect to the Coastal Terrane. The Three Palms Mylonite Zone is an extensional strike-slip shear zone for most of its length and the kinematic-lineation array in the Coastal Terrane is consistent with shallow oblique downward and SSW-directed transport. Stretching lineations in the Three Palms Mylonite Zone and Coastal Terrane range from horizontal to an average plunge of 10° SSW. Consequently, the 6–12 km of relative downward transport of the Coastal Terrane could have been accommodated by a minimum of 34–68 km lateral transport southward.
Both the structural and metamorphic datasets are mutually consistent and suggest that the kinematic-lineation array does contain meaning for the particle paths experienced. It may be concluded that stretching lineations in the western parts of the Kaoko Belt do represent oblique particle paths (Fig. 14) and that longitudinal extrusion along shallow oblique trajectories occurred. Constraints on particle paths from the structural and metamorphic datasets show that extrusion of lower-crustal material into middle-crustal levels is possible in orogens with low convergence angle. Upward extrusion of the high-grade lobes of the Orogen Core suggests relatively higher rates of exhumation in these domains (Thompson et al., 1997b). The higher rates of material transport through the crustal column in these domains also imply greater development of topography and high rates of topography-induced erosion as described by Jamieson et al. (1996, 2002) and Beaumont et al. (1996, 2001). Maximum limiting pressures of 8–9 kbar in the Orogen Core are consistent with exhumation during transpressional orogenesis being dependent on topographic development and erosion (Stüwe & Barr, 1998). Consequently, the complex metamorphic pattern at middle-crustal levels (Fig. 5) indicates an irregular distribution of surface topography. The deepest exhumed parts of the Kaoko Belt, the western Escape Zone, were probably represented at the surface by a continuous mountain range extending the length of the belt and flanked by the Purros Mylonite Zone. The Orogen Core was possibly expressed at the surface by a series of en echelon ranges at low angles to the belt and verging in the same direction as overall sinistral tectonic transport. The Coastal Terrane developed comparatively subdued topography and at surface the Khumib Domain may have been a broad intermontane plain. A suggested modern analogue would be a broad orogen containing an en echelon set of multiple, small, New Zealand Southern Alp-like systems.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
The architecture of the Kaoko Belt strongly governs the metamorphic response of its component parts. On a gross scale the Orogen Core represents the high-grade extruded core of a metamorphic belt. This core is bounded on both sides by crustal-scale shear zones of opposing shear sense (Fig. 14). To the west the Three Palms Mylonite Zone is an obliquely extensional sinistral shear zone for most of its length, with the Coastal Terrane being transported downwards and to the SSW with respect to the Orogen Core. To the east the Purros Mylonite Zone is an obliquely reverse sinistral shear zone accommodating upward and SSE-directed extrusion of most parts of the Orogen Core. The Orogen Core experienced differential extrusion, with upward transport of high-grade lobes and relative downward transport of a low-grade ‘trough’. For the first time, integrated structural and metamorphic datasets from the Kaoko Belt show the existence of longitudinal extrusion of lower-crustal rocks into middle-crustal levels along oblique trajectories within a classic transpressional orogen (Figs 14 and 15).
|
The shear strain and lateral transport estimated for the wrench component alone (Goscombe & Gray, in preparation) is entirely compatible with the hypothetical amount of oblique transport along the stretching lineation arrays required to give rise to the metamorphic patterns observed in the Orogen Core (Goscombe & Gray, in preparation). These two independent datasets, metamorphic and strain, both give rise to similar estimates of the lateral displacement experienced across the Orogen Core. Furthermore, these estimates are entirely compatible with this transport being along shallowly inclined trajectories parallel to the developed stretching lineation array. Consequently, it is possible that the lower-crust extrusion indicated by the metamorphic dataset occurred by transport along shallowly inclined oblique trajectories and not by decoupled horizontal and vertical motions. Extrusional tectonics gave rise to a co-temporal, composite metamorphic belt that contains a high-T/moderate-P Orogen Core juxtaposed against a high-P/moderate-T Barrovian-style Escape Zone of contrasting metamorphic character. The Escape Zone has an inverted Barrovian metamorphic sequence, which is typical of high-angle convergence, crustal overriding orogenic systems, such as the Himalayas (Goscombe & Hand, 2000
) and Appalachians (Armstrong et al., 1992), and is consistent with thermal models of the foreland block in transpressional orogens (Thompson et al., 1997a).
The Kaoko Belt is a classic example of an orogen-scale transpressional system and shows all the typical salient features (Goscombe et al., 2003a, 2003b). However, we feel that transpressional systems are too variable and complex for our findings on the metamorphic response and paths of material through the orogen to be taken as universally applicable. Comparison with metamorphic investigations of other transpressional orogens (Hand et al., 1995; Martelat et al., 1999; Nedelec et al., 2000; Scrimgeour & Raith, 2001; Stipska et al., 2001; Carosi & Palmeri, 2002; Johnson et al., 2003) shows a wide variety of metamorphic responses, with contrasting metamorphic histories, P–T paths and metamorphic gradients, and a consistent picture does not emerge. This is because of the large number of potential first-order variables influencing this metamorphic response. For example, thermal variables include thermal structure and rheology of the crustal column (Stipska et al., 2001) and lithospheric mantle, thickness of basinal sequences and influence on thermal structure of the crust (Sandiford, 2002), magmatic activity (Druguet, 2001), and concentration and distribution of heat-producing radiogenic elements in the crust (Brown, 1999). Examples of structural variables include strain rate, orogen architecture, degree of strain partitioning (Dewey et al., 1998; Lin et al., 1998; Stipska et al., 2001), simple shear/pure shear ratio (Lin et al., 1998), degree of obliquity of convergence (Thompson et al., 1997a), inclination of orogen median shear system (Little et al., 2002) and rates of exhumation, which are linked with topographic development and erosion rates (Jamieson et al., 2002).
Specifically, extrusion of lower-crustal material into middle-crustal levels in the Kaoko Belt is driven by collisional upramping along listrically inclined deep-crustal structures (Goscombe & Gray, in preparation). Similar transpressional orogen architecture has been recognized elsewhere (Brown & Solar, 1999; Solar & Brown, 2001; Little et al., 2002; Thiel, 2002), but because examples of exposure or imaging of the deep crust are limited this architectural style may not be typical of all transpressional orogens. Furthermore, the identified differential extrusion of domains within the Orogen Core is contingent on orogen-specific parameters, such as arrangement of shear zones and erosion rates dependent on topographic development, all of which govern material advection through the crust (Beaumont et al., 2001; Jamieson et al., 2002). The oblique (ßL) and inclined (z) particle paths in the Kaoko Belt are consistent with modelled flow of material in transpressional systems assuming triclinic shear and continuous strain partitioning (Robin & Cruden, 1994; Tikoff & Teyssier, 1994). However, continuous strain partitioning is an end-member case and most transpressional orogens, including the Kaoko Belt, show some degree of discontinuous strain partitioning (Dewey et al., 1998; Solar & Brown, 2001; Stipska et al., 2001), resulting in contrasting domains dominated by either lateral transport or vertical transport.
Convergent orogens with low-angle (ß <30°) and high-angle (ß = 90°) tectonic obliquity (ß is the angle between the convergence vector and the strike of the orogen) show obvious differences in architecture such as steeper inclination of the crustal-scale shear system in transpressional systems (Fig. 15). Nevertheless, the Kaoko Belt and central Himalayas are examples of low-angle and high-angle convergence, respectively, but both show considerable similarities in metamorphic response and lower-crustal processes (Fig. 15). Kinematically, both examples are convergent systems and both are centred on an extruded high-grade core that is bounded by a crustal-scale shear pair with contrasting shear sense. Extrusion of the Orogen Core in the Kaoko Belt was accommodated by a pair of bounding oblique-slip shear zones with contrasting vertical component of movement sense (Fig. 14). The Three Palms Mylonite Zone is the margin between the upper-plate Coastal Terrane and extruded Orogen Core, and has a hanging-wall-down oblique movement sense implying an extensional component. In contrast, the Purros Mylonite Zone margin between the Orogen Core and lower plate Escape Zone has a reverse component of oblique slip (Fig. 15).
Similarly, the High Himalayan Gneisses are the high-grade extruded core of the central Himalayan metamorphic front, which is accommodated by bounding crustal-scale shear systems of contrasting relative movement (Fig. 15; Grujic et al., 1996; Vannay & Hodges, 1996; Beaumont et al., 2001). The upper-plate, low-grade Tethyan Zone is juxtaposed against the high-grade orogen core by extensional movements in the South Tibet Detachment System. The extruded orogen core overrides the lower-plate, Lesser Himalayan Sequence, along a broad crustal-scale reverse shear system composed of the High Himal Thrust and Main Central Thrust Zone (Goscombe & Hand, 2000; Goscombe et al., 2003c). In both examples, the relative extensional transport (on the Three Palms Mylonite Zone and South Tibet Detachment System) is an artefact of high relative rates of transport (i.e. extrusion) of the orogen cores (Grujic et al., 1996; Goscombe et al., 2003b, 2003c). Consequently, the Three Palms Mylonite Zone and South Tibet Detachment System are considered to be analogous extensional crustal-scale structures, which accommodate extrusion within convergent orogenic systems that nevertheless show contrasting tectonic obliquity and shear system inclination.
In addition to these kinematic similarities, the developed metamorphic response and metamorphic gradients are similar in these two examples of low- and high-obliquity convergent systems. In both, the extruded core of the metamorphic front is dominantly of high-T/moderate-P type, with high average thermal gradients (30–45°C/km) and shallow conduction-dominated clockwise P–T paths (Fig. 11; Goscombe & Hand, 2000). The extruded orogen cores override, and are juxtaposed against, inverted Barrovian-style metamorphic sequences within Escape Zones characterized by nappes, over-folds and thrusts directed outward toward the orogen margin (Fig. 15). Steep (P/T) burial and decompression paths and low average thermal gradients (15–20°C/km) indicate advection-dominated clockwise P–T paths in these Escape Zones (Fig. 11; Goscombe & Hand, 2000; Beaumont et al., 2001). In contrast, however, the peak metamorphic gradient in the Kaoko Belt is approximately time equivalent across the orogen (Goscombe et al., 2003b), but is diachronous across the central Himalayan metamorphic front (Harrison et al., 1997; Beaumont et al., 2001).
Both the transpressional Kaoko Belt and the high-obliquity convergent central Himalayas show processes in common for extrusion of lower-crustal material into mid-crustal levels. Extruded orogen cores are transported along apparently continuous inclined trajectories defined by stretching lineation arrays that are broadly sub-parallel to inferred tectonic transport vectors (i.e. ßl 
ß, Fig. 15). In the Kaoko Belt the kinematic-lineation array is entirely consistent with the metamorphic gradient ‘net’, suggesting that in this orogen the stretching lineations match real particle paths (Fig. 14). In the central Himalayas stretching lineations are sub-parallel to the tectonic transport vector defined by sea-floor spreading histories (i.e. ßL = ß; Shackleton & Ries, 1984). In both cases, the shallowly inclined particle paths indicate a significant component of horizontal transport in the lower crust, resulting in lateral extrusion directed at high angles to the Himalayan orogen and longitudinal extrusion at acute angles to the Kaoko Belt (Fig. 15). In both cases, the channelized flow of material accompanying high exhumation rates in the orogen cores is interpreted to be driven by focused, high rates of erosion in parts of the metamorphic front, leading to relatively high rates of advection of material through the crustal column (Beaumont et al., 2001; Jamieson et al., 2002).
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data for this paper are available on Journal of Petrology online.
|
ACKNOWLEDGEMENTS |
|---|
Support for the research was from an ARC Large Grant A00103456 awarded to D.R.G. Write-up was undertaken whilst B.G. was on leave from the WA Geological Survey. B.G. was partially supported as a Post-Doctoral Fellow by ARC Large Grant A00103456, and he acknowledges support from the University of Adelaide, Monash University and the University of Melbourne. We thank the director of the Namibian Geological Survey for support, Mimi Dunaeski and Thomas Bekker for logistical help in Namibia, Peter Weber (Bushveld Car Hire) for assistance with vehicles, and Gordon Holm (The University of Melbourne) for thin-section preparation. We acknowledge discussions on Namibian geology with Thomas Bekker, Charlie Hoffman, Roy Miller, Paul Hoffman, Cees Passchier and Rudolph Trouw. We are grateful for discussions with Paul Hamilton on data presentation. The very encouraging, constructive and patient reviews of Mike Brown, Bob Holdsworth and an anonymous reviewer and editorial contributions by Marjorie Wilson are gratefully acknowledged.
* Corresponding author. E-mail: ben.goscombe{at}nt.gov.au
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSTRUCTURAL EVOLUTION AND...METAMORPHIC PATTERNSMATERIAL FLOW TRAJECTORIESDISCUSSIONSUPPLEMENTARY DATAREFERENCES |
|---|
Armstrong, T. R., Tracy, R. J. & Hames, W. E. (1992). Contrasting styles of Taconian, Eastern Acadian and Western Acadian metamorphism, central and western New England. Journal of Metamorphic Geology 10, 415–426.[Web of Science]
Batt, G. E. & Braun, J. (1997). On the thermomechanical evolution of compressional orogens. Geophysical Journal International 128, 364–382.[Web of Science]
Beaumont, C., Kamp, P. J. J., Hamilton, J. & Fullsack, P. (1996). The continental collision zone, South Island, New Zealand: comparison of geodynamic models and observations. Journal of Geophysical Research 101, 3333–3359.[CrossRef]
Beaumont, C., Jamieson, R. A., Nguyen, M. H. & Lee, B. (2001). Himalayan tectonics explained by extrusion of a low-viscosity crustal channel coupled to focused surface denudation. Nature 414, 738–742.[CrossRef][Medline]
Braun, J. & Beaumont, C. (1995). Three-dimensional numerical experiments of strain partitioning at oblique plate boundaries: implications for contrasting tectonic styles in the southern Coast Ranges, California, and central South Island, New Zealand. Journal of Geophysical Research 100, 18059–18074.[CrossRef]
Brown, M. (1999). Thermal structure of the crust during obliquely convergent (transpressive) orogenesis involving high-temperature (HT) metamorphism; an example from the Northern Appalachians, USA. European Union of Geosciences Journal of Conference Abstracts 4, 462.
Brown, M. & Solar, G. S. (1999). The mechanism of ascent and emplacement of granite magma during transpression; a syntectonic granite paradigm. Tectonophysics 312, 1–33.[CrossRef][Web of Science]
Carosi, R. & Palmeri, R. (2002). Orogen-parallel tectonic transport in the Variscan Belt of northern Sardinia (Italy); implications for the exhumation of medium-pressure metamorphic rocks. Geological Magazine 139, 497–511.
Comacho, A. & McDougall, I. (2000). Intracratonic, strike-slip partitioned transpression and the formation and exhumation of eclogite facies rocks: an example from the Musgrave Block, central Australia. Tectonics 19, 978–996.[CrossRef][Web of Science]
Coward, M. P. (1983). The tectonic history of the Damara belt. In: Miller, R. McG. (ed.) Evolution of the Damara Orogen. Special Publication of the Geological Society of South Africa 11, 409–421.
Dewey, J. F., Holdsworth, R. E. & Strachan, R. A. (1998). Transpression and transtension zones. In: Holdsworth, R. E., Strachan, R. A. & Dewey, J. F. (eds) Continental Transpressional and Transtensional Tectonics. Geological Society, London, Special Publications 135, 1–14.
Dingeldey, P. (1997). Tectono-metamorphic evolution of the Pan-African Kaoko Belt, NW-Namibia. Ph.D. thesis, Julius-Maximilians-Universität Wurzburg, 246 pp.
Dingeldey, D. P., Dürr, S. B., Charlesworth, E. G., Franz, L., Okrusch, M. & Stanistreet, I. G. (1994). A geotraverse through the northern coastal branch of the Damaran Orogen west of Sesfontein, Namibia. Journal of African Earth Science 19, 315–329.
Druguet, E. (2001). Development of high thermal gradients by coeval transpression and magmatism during the Variscan orogeny; insights from the Cap de Creus (eastern Pyrenees). Tectonophysics 332, 275–293.[CrossRef][Web of Science]
Dürr, S. B. & Dingeldey, D. P. (1996). The Kaoko belt (Namibia): part of a late Neoproterozoic continental-scale strike-slip system. Geology 24, 503–506.
Fossen, H. & Tikoff, B. (1998). Extended models of transpression and transtension, and application to tectonic settings. In: Holdsworth, R. E., Strachan, R. A. & Dewey, J. F. (eds) Continental Transpressional and Transtensional Tectonics. Geological Society, London, Special Publications 135, 15–33.
Franz, L., Romer, R. L. & Dingeldey, D. P. (1999). Diachronous Pan-African granulite-facies metamorphism (650 Ma and 550 Ma) in the Kaoko belt, NW Namibia. European Journal of Mineralogy 11, 167–180.
Freyer, E. E. & Halbich, I. W. (1994). Deformation history of the lower Ugab Belt. In: Niall, M. & McManus, C. (eds) Proterozoic Crustal and Metallogenic Evolution. Abstracts of Geological Society and Geological Survey of Namibia 18.
Frost, B. R. & Chacko, T. (1989). The granulite uncertainty principle: limitations on thermobarometry in granulites. Journal of Geology 97, 435–450.[Web of Science]
Ghent, E. D., Robbins, D. B. & Stout, M.Z. (1979). Geothermometry, geobarometry, and fluid compositions of metamorphosed calc-silicates and pelites, Mica Creek, British Columbia. American Mineralogist 64, 874–885.[Abstract]
Goscombe, B. D. (1998). Geological map of the Sesfontein 1:250 000 sheet. Windhoek: Geological Survey of Namibia.
Goscombe, B. D. (1999a). Geological map of Tomakas 1:50 000 sheet. Windhoek: Geological Survey of Namibia.
Goscombe, B. D. (1999b). Geological map of Omapungwe 1:50 000 sheet. Windhoek: Geological Survey of Namibia.
Goscombe, B. D. (1999c). Geological map of Orumpembe 1:50 000 sheet. Windhoek: Geological Survey of Namibia.
Goscombe, B. D. & Hand, M. (2000). Contrasting P–T paths in the Eastern Himalaya, Nepal: inverted isograds in a paired metamorphic mountain belt. Journal of Petrology 41, 1673–1719.
Goscombe, B. D., Armstrong, R. & Barton, J. M. (2000). Geology of the Chewore Inliers: constraining the Mesoproterozoic to Palaeozoic evolution of the Zambezi Belt. Journal of African Earth Science 30, 589–627.
Goscombe, B. D., Hand, M. & Gray, D. (2003a). Structure of the Kaoko Belt, Namibia: progressive evolution of a classic transpressional orogen. Journal of Structural Geology 25, 1049–1081.[CrossRef][Web of Science]
Goscombe, B. D., Hand, M., Gray, D. & Mawby, J. (2003b). The metamorphic architecture of a transpressional orogen: the Kaoko Belt, Namibia. Journal of Petrology 44, 676–711.
Goscombe, B., Hand, M. & Gray, D. (2003c). Structural and metamorphic architecture of the east Nepal Himalayas. Geological Society of Australia, Specialist Group in Tectonics and Structural Geology, Conference Abstracts, Kalbarri, p. 135.
Goscombe, B. D., Gray, D. & Hand, M. (2004). Variation in metamorphic style along the northern margin of the Damara Orogen, Namibia. Journal of Petrology 45, 1261–1295.
Grapes, R. & Watanabe, T. (1992). Metamorphism and uplift of Alpine schist in the Franz Josef–Fox Glacier area of the Southern Alps, New Zealand. Journal of Metamorphic Geology 10, 171–180.[Web of Science]
Green, D. H. & Ringwood, A. E. (1967). An experimental investigation of the gabbro to eclogite transformation and its petrological applications. Geochimica et Cosmochimica Acta 31, 767–833.[Web of Science]
Grujic, D., Casey, M., Davidson, C., Hollister, L. S., Kundig, R., Pavlis, T. & Schmid, S. (1996). Ductile extrusion of the Higher Himalayan crystalline in Bhutan: evidence from quartz microfabrics. Tectonophysics 260, 21–43.[CrossRef][Web of Science]
Guj, P. (1970). The Damara Mobile Belt in the south-western Kaokoveld, South West Africa. Ph.D. thesis, University of Cape Town, 168 pp.
Hand, M., Bendall, B. R. & Sandiford, M. (1995). Metamorphic evidence for Palaeoproterozoic oblique convergence in the eastern Gawler Craton. Geological Society of Australia, Abstracts 40, 59.
Hansen, V. L. (1989). Structural and kinematic evolution of the Teslin suture zone, Yukon: record of an ancient transpressional margin. Journal of Structural Geology 11, 717–733.[CrossRef][Web of Science]
Harland, W. B. (1971). Tectonic transpression in Caledonian Spitsbergen. Geological Magazine 108, 27–42.[Abstract]
Harley, S. L. (1992). Proterozoic granulite terranes. In: Condie, K. C. (ed.) Proterozoic Crustal Evolution. Developments in Precambrian Geology 10, 301–359.
Harrison, T. M., Ryerson, F. J., LeFort, P., Yin, A., Lovera, O. M. & Catlos, E. J. (1997). A Late Miocene–Pliocene origin for the Central Himalayan inverted metamorphism. Earth and Planetary Science Letters 146, E1–E7.[CrossRef][Web of Science]
Hoffman, P. F., Kaufman, A. J., Halverson, G. P. & Schrag, D. P. (1998). A Neoproterozoic snowball Earth. Science 281, 1342–1346.
Holdsworth, R. E. & Pinheiro, R. V. L. (2000). The anatomy of shallow-crustal transpressional structures: insights from the Archaean Carajas fault zone, Amazon, Brazil. Journal of Structural Geology 22, 1105–1123.[CrossRef][Web of Science]
Holdsworth, R. E. & Strachan, R. A. (1991). Interlinked system of ductile strike slip and thrusting formed by Caledonian sinistral transpression in northeastern Greenland. Geology 19, 510–513.
Holland, T. J. B. & Powell, R. (1990). An enlarged and updated internally consistent thermodynamic dataset with uncertainties and correlations: the system K2O–Na2O–CaO–MgO–MnO–FeO–Fe2O3–Al2O3–TiO2–SiO2–C–H2–O2. Journal of Metamorphic Geology 8, 89–124.[Web of Science]
Jamieson, R. A., Beaumont, C., Hamilton, J. & Fullsack, P. (1996). Tectonic assembly of inverted metamorphic sequences. Geology 24, 839–842.
Jamieson, R. A., Beaumont, C., Nguyen, M. H. & Lee, B (2002). Interaction of metamorphism, deformation and exhumation in large convergent orogens. Journal of Metamorphic Geology 20, 1–16.[CrossRef][Web of Science]
Jiang, D., Lin, S. & Williams, P. F. (2001). Deformation path in high-strain zones, with reference to slip partitioning in transpressional plate-boundary regions. Journal of Structural Geology 23, 991–1005.[CrossRef][Web of Science]
Johnson, P. R. & Kattan, F. (2001). Oblique sinistral transpression in the Arabian shield: the timing and kinematics of a Neoproterozoic suture zone. Precambrian Research 107, 117–138.[CrossRef][Web of Science]
Johnson, T. E., Brown, M. & Solar, G. S. (2003). Low-pressure subsolidus and parasolidus phase equilibria in the MnNCKFMASH system: constraints on conditions of regional metamorphism in western Maine, northern Appalachians. American Mineralogist 88, 624–638.
Jones, D. G., Silberling, N. J. & Hillhouse, J. (1977). Wrangellia—a displaced terrane in northwestern North America. Canadian Journal of Earth Sciences 14, 2565–2577.[CrossRef][Web of Science]
Jones, K. A. & Escher, J. C. (2002). Near-isothermal decompression within a clockwise P–T evolution recorded in migmatitic mafic granulites from Clavering, NE Greenland: implications for the evolution of the Caledonides. Journal of Metamorphic Geology 20, 365–378.[CrossRef][Web of Science]
Jones, K. A. & Strachan, R. A. (2000). Crustal thickening and ductile extension in the NE Greenland Caledonides: a metamorphic record from anatectic pelites. Journal of Metamorphic Geology 18, 719–735.[CrossRef][Web of Science]
Jones, R. R., Holdsworth, R. E., Clegg, P., McCaffrey, K. & Tavarnelli, E. (2004). Inclined transpression. Journal of Structural Geology 26, 1531–1548.[CrossRef][Web of Science]
Jung, S., Hoernes, S. & Mezger, K. (2000). Geochronology and petrology of migmatites from the Proterozoic Damara Belt—importance of episodic fluid-present disequilibrium melting and consequences for granite petrology. Lithos 51, 153–179.[CrossRef][Web of Science]
Kirkwood, D., Malo, M., St-Julien, P. & Therrien, P. (1995). Vertical and fold-axis parallel extension within a slate belt in a transpressive setting, northern Appalachians. Journal of Structural Geology 17, 329–343.[CrossRef][Web of Science]
Kohn, M. J. & Spear, F. S. (1990). Two new geobarometers for garnet amphibolites, with applications to southeastern Vermont. American Mineralogist 75, 89–96.[Abstract]
Kohn, M. J. & Spear, F. (2000). Retrograde net transfer reaction insurance for pressure–temperature estimates. Geology 28, 1127–1130.
Koons, P. O. (1987). Some thermal and mechanical consequences of rapid uplift; an example from the Southern Alps, New Zealand. Earth and Planetary Science Letters 86, 307–319.[CrossRef][Web of Science]
Koons, P. O. (1990). The two sided wedge in orogeny; erosion and collision from the sand box to the Southern Alps, New Zealand. Geology 18, 679–682.
Koons, P. O, Norris, R. J., Craw, D. & Cooper, A. F. (2003). Influence of exhumation on the structural evolution of transpressional plate boundaries: an example from the Southern Alps, New Zealand. Geology 31, 3–6.
Kretz, R. (1983). Symbols for rock-forming minerals. American Mineralogist 68, 277–279.[Abstract]
Lin, S., Jiang, D. & Williams, P. F. (1998). Transpression (or transtension) zones of triclinic symmetry: natural example and theoretical modeling. In: Holdsworth, R. E., Strachan, R. A. & Dewey, J. F. (eds) Continental Transpressional and Transtensional Tectonics. Geological Society, London, Special Publications 135, 41–57.
Little, T. A., Holcombe, R. J. & Ilg, B. R. (2002). Kinematics of oblique collision and ramping inferred from microstructures and strain in middle crustal rocks, central Southern Alps, New Zealand. Journal of Structural Geology 24, 219–239.[CrossRef][Web of Science]
Lowell, J. D. (1972). Spitsbergen Tertiary orogenic belt and the Spitsbergen fracture zone. Geological Society of America Bulletin 83, 3091–3102.
Maloof, A. C. (2000). Superposed folding at the junction of the inland and coastal belts, Damara Orogen, NW Namibia. Communications of the Geological Survey of Namibia 12, 89–98.
Martelat, J. E., Lardeaux, J. M., Nicollet, C. & Rakotondrazafy, R. (1999). Exhumation of granulites within a transpressive regime; an example from southern Madagascar. Gondwana Research 2, 363–367.[CrossRef][Web of Science]
Martelat, J. E., Lardeaux, J. M., Nicollet, C. & Rakotondrazafy, R. (2000). Strain pattern and late Precambrian deformation history in southern Madagascar. Precambrian Research 102, 1–20.[CrossRef][Web of Science]
Miller, R. McG. (1983). The Pan-African Damara Orogen of Namibia. In: Miller, R. McG. (ed.) The Damara Orogen. Special Publication of the Geological Society of South Africa 11, 431–515.
Miller, R. McG. & Burger, A. J. (1983). U–Pb zircon age of the early Damaran Naauwpoort Formation. Special Publication of the Geological Society of South Africa 11, 267–272.
Nedelec, A., Ralison, B., Bouchez, J. L. & Gregoire, V. (2000). Structure and metamorphism of the granitic basement around Antananarivo; a key to the Pan-African history of central Madagascar and its Gondwana connections. Tectonics 19, 997–1020.[CrossRef][Web of Science]
Oldow, J. S., Bally, A. W. & Ave'Lallemant, H. G. (1990). Transpression, orogenic float, and lithospheric balance. Geology 18, 991–994.
Passchier, C. W., Trouw, R. A. J., Ribeiro, A. & Paciullo, F. V. P. (2002). Tectonic evolution of the southern Kaoko Belt, Namibia. Journal of African Earth Sciences 35, 61–75.[CrossRef]
Patchett, P. J. & Chase, C. G. (2002). Role of transform continental margins in major crustal growth episodes. Geology 30, 39–42.
Pattison, D. R. M. (2003). Petrogenetic significance of orthopyroxene-free garnet + clinopyroxene + plagioclase ± quartz-bearing metabasites with respect to the amphibolite and granulite facies. Journal of Metamorphic Geology 21, 21–34.[CrossRef][Web of Science]
Powell, R. & Holland, T. J. B. (1988). An internally consistent dataset with uncertainties and correlations: 3. Applications to geobarometry, worked examples and a computer program. Journal of Metamorphic Geology 6, 173–204.[Web of Science]
Powell, R. & Holland, T. (1990). Calculated mineral equilibria in the pelite system, KFMASH (K2O–FeO–MgO–Al2O3–SiO2–H2O). American Mineralogist 75, 367–380.[Abstract]
Powell, R. & Holland, T. J. B. (1994). Optimal geothermometry and geobarometry. American Mineralogist 79, 120–133.[Abstract]
Powell, R. Holland, T. B. J. & Worley, B. (1998). Calculating phase diagrams involving solid solutions via non-linear equations, with examples using THERMOCALC. Journal of Metamorphic Geology 16, 577–588.[CrossRef][Web of Science]
Prave, A. R. (1996). Tale of three cratons: tectonostratigraphic anatomy of the Damara Orogen in northwestern Namibia and the assembly of Gondwana. Geology 24, 1115–1118.
Robin, P. Y. F. & Cruden, A. R. (1994). Strain and vorticity patterns in ideally ductile transpression zones. Journal of Structural Geology 16, 447–466.[CrossRef][Web of Science]
Sandiford, M. (2002). Low thermal Peclet number intraplate orogeny in central Australia. Earth and Planetary Science Letters 201, 309–320.[CrossRef][Web of Science]
Schaller, M., Steiner, O., Studer, I. Holzer, L., Herwegh, M. & Kramers, J. D. (1999). Exhumation of Limpopo Central Zone granulites and dextral continent-scale transcurrent movement at 2·0 Ga along the Palala shear zone, Northern Province, South Africa. Precambrian Research 96, 263–288.[CrossRef][Web of Science]
Schreiber, U. M. (2002). Geological map of the Swartbooisdrif 1:250 000 sheet. Windhoek: Geological Survey of Namibia.
Scrimgeour, I. & Raith, J. G. (2001). High-grade reworking of Proterozoic granulites during Ordovician intraplate transpression, eastern Arunta Inlier, central Australia. In: Miller, J. A., Holdsworth, R. E., Buick, I. S. & Hand, M. (eds) Continental Reactivation and Reworking. Geological Society, London, Special Publications 184, 261–287.
Seth, B., Kroner, A., Mezger, K., Nemchin, A. A., Pidgeon, R. T. & Okrusch, M. (1998). Archaean to Neoproterozoic magmatic events in the Kaoko belt of NW Namibia and their geodynamic significance. Precambrian Research 92, 341–363.[CrossRef][Web of Science]
Seth, B., Okrusch, M., Wilde, M. & Hoffmann, K.-H. (2000). The Voetspoor Intrusion, Southern Kaoko Zone, Namibia: mineralogical, geochemical and isotopic constraints for the origin of a syenitic magma. Communications of the Geological Survey of Namibia 12, 125–137.
Shackleton, R. M. & Ries, A. C. (1984). The relation between regionally consistent stretching lineations and plate motions. Journal of Structural Geology 6, 111–117.[CrossRef][Web of Science]
Solar, G. S. & Brown, M. (2001). Deformation partitioning during transpression in response to Early Devonian oblique convergence, northern Appalachian orogen, USA. Journal of Structural Geology 23, 1043–1065.[CrossRef][Web of Science]
Sonder, L. J. & Chamberlain, C. P. (1992). Tectonic controls of metamorphic field gradients. Earth and Planetary Science Letters 111, 517–535.[CrossRef][Web of Science]
Spear, F. S. (1981). Amphibole–plagioclase equilibria: an empirical model for the relation albite + tremolite = edenite + 4 quartz. Contributions to Mineralogy and Petrology 77, 355–364.[CrossRef][Web of Science]
Spear, F. S. (1993). Metamorphic Phase Equilibria and Pressure–Temperature–Time Paths. Mineralogical Society of America Monograph 1.
Spear, F. S., Selverstone, J., Hickmott, D., Crowley, P. & Hodges, K. V. (1984). P–T paths from garnet zoning: a new technique for deciphering tectonic processes in crystalline terranes. Geology 12, 87–90.
Spiess, R., Bertolo, B., Borghi, A., Chinellato, M. & Tinor Centi, M. (2000). Microtextures of opaque inclusions: their use as indicators for hiatuses during garnet porphyroblast growth. Journal of Metamorphic Geology 18, 591–603.[CrossRef][Web of Science]
Stipska, P., Schulmann, K., Thompson, A. B., Jezek, J. & Kroener, A. (2001). Thermo-mechanical role of a Cambro-Ordovician paleorift during the Variscan collision; the NE margin of the Bohemian Massif. Tectonophysics 332, 239–253.[CrossRef][Web of Science]
Stowell, H. H. & Crawford, M. L. (2000). Metamorphic history of the Coast Mountains Orogen, western British Columbia and Southeastern Alaska. Geological Society of America, Special Papers 343, 257–283.
Strachan, R. A., Holdsworth, R. E., Friderichsen, J. D. & Jepsen, H. F. (1992). Regional Caledonian structure within an oblique convergence zone, Dronning Louise Land, NE Greenland. Journal of the Geological Society, London 149, 359–371.
Stüwe, K. & Barr, T. D. (1998). On uplift and exhumation during convergence. Tectonics 17, 80–88.[CrossRef][Web of Science]
Sylvester, A. G. (1988). Strike-slip faults. Geological Society of America Bulletin 100, 1666–1703.
Sylvester, A. G. & Smith, R. R. (1976). Tectonic transpression and basement-controlled deformation in the San Andreas fault zone, Salton Trough, California. AAPG Bulletin 60, 2081–2102.[Abstract]
Tapponnier, P., Peltzer, G., LeDain, A.Y. & Armijo, R. (1982). Propagating extrusion tectonics in Asia: new insights from simple experiments with plasticine. Geology 10, 611–616.
Teyssier, C. & Tikoff, B. (1998). Strike-slip partitioned transpression of the San Andreas fault system: a lithospheric-scale approach. In: Holdsworth, R. E., Strachan, R. A. & Dewey, J. F. (eds) Continental Transpressional and Transtensional Tectonics. Geological Society, London, Special Publications 135, 143–158.
Thiel, S. (2002). Electromagnetic induction studies of the Eyre Peninsula Anomaly and their relationship to the tectonics of the southern Gawler Craton, South Australia. B.Sc.(Hons) thesis, University of Adelaide.
Thompson, A. B., Schulmann, K. & Jezek, J. (1997a). Thermal evolution and exhumation in obliquely convergent (transpressive) orogens. Tectonophysics 280, 171–184.[CrossRef][Web of Science]
Thompson, A. B., Schulmann, K. & Jezek, J. (1997b). Extrusion tectonics and elevation of lower crustal metamorphic rocks in convergent orogens. Geology 25, 491–494.
Tikoff, B. & Teyssier, C. (1994). Strain modeling of displacement-field partitioning in transpressional orogens. Journal of Structural Geology 16, 1575–1588.[CrossRef][Web of Science]
Tinkham, D. K., Zuluaga, C. A. & Stowell, H. H. (2001). Metapelite phase equilibria modeling in MnNCKFMASH: the effect of variable Al2O3 and MgO/(MgO + FeO) on mineral stability. Geological Materials Research 3, 1–42.
Tuccillo, M. E., Essene, E. J. & van der Pluijm, B. A. (1990). Growth and retrograde zoning in garnets from high-grade metapelites: implications for pressure–temperature paths. Geology 18, 839–842.
Vance, D. & Mahar, E. (1998). Pressure–temperature paths from P–T pseudosections and zoned garnets: potential, limitations and examples from the Zanskar Himalaya, NW India. Contributions to Mineralogy and Petrology 132, 225–245.[CrossRef][Web of Science]
Vannay, J.-C. & Hodges, K. V. (1996). Tectonometamorphic evolution of the Himalayan metamorphic core between the Annapurna and Dhaulagiri, central Nepal. Journal of Metamorphic Geology 14, 635–656.[CrossRef][Web of Science]
Vassallo, J. J. & Wilson, C. J. L. (2002). Palaeoproterozoic regional-scale non-coaxial deformation: an example from eastern Eyre Peninsula, South Australia. Journal of Structural Geology 24, 1–24.[CrossRef][Web of Science]
Veevers, J. J. (2003). Pan-African is Pan-Gondwanaland: oblique convergence drives rotation during 650–500 Ma assembly. Geology 31, 501–504.
White, R. W., Powell, R., Holland, T. J. B. & Worley, B. A. (2000). The effect of TiO2 and Fe2O3 on metapelitic assemblages at greenschist and amphibolite facies conditions: mineral equilibria calculations in the system K2O–FeO–MgO–Al2O3–SiO2–H2O–TiO2–Fe2O3. Journal of Metamorphic Geology 18, 497–511.[CrossRef][Web of Science]
White, R. W., Powell, R. & Holland, T. J. B. (2001). Calculation of partial melting equilibria in the system Na2O–CaO–K2O–FeO–MgO–Al2O3–SiO2–H2O (NCKFMASH). Journal of Metamorphic Geology 19, 139–153.[CrossRef][Web of Science]
Whitney, D. L., Teyssier, C., Dilek, Y. & Fayon, A. K. (2001). Metamorphism of the Central Anatolian Crystalline Complex, Turkey: influence of orogen-normal collision vs. wrench-dominated tectonics on P–T–t paths. Journal of Metamorphic Geology 19, 411–432.[CrossRef][Web of Science]
Wilcox, R. E., Harding, T. P. & Seely, D. R. (1973). Basic wrench tectonics. AAPG Bulletin 57, 74–96.[Abstract]
Woodcock, N. H. (1986). The role of the strike-slip fault systems at plate boundaries. Philosophical Transactions of the Royal Society of London, Series A 317, 13–29.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Konopasek, J. Kosler, L. Tajcmanova, S. Ulrich, and S. L. Kitt Neoproterozoic igneous complex emplaced along major tectonic boundary in the Kaoko Belt (NW Namibia): ion probe and LA-ICP-MS dating of magmatic and metamorphic zircons Journal of the Geological Society, January 1, 2008; 165(1): 153 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Gray, D. A. Foster, J. G. Meert, B. D. Goscombe, R. Armstrong, R. A. J. Trouw, and C. W. Passchier A Damara orogen perspective on the assembly of southwestern Gondwana Geological Society, London, Special Publications, January 1, 2008; 294(1): 257 - 278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Gerbi and D. P. West Jr. Use of U-Pb geochronology to identify successive, spatially overlapping tectonic episodes during Silurian-Devonian orogenesis in south-central Maine, USA Geological Society of America Bulletin, September 1, 2007; 119(9-10): 1218 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Roeske, A. B. Till, D. A. Foster, and J. C. Sample Introduction Geological Society of America Special Papers, January 1, 2007; 434(0): vii - x. [Full Text] [PDF]
|
|
|
|
|
|
R. L. Brown and H. D. Gibson An argument for channel flow in the southern Canadian Cordillera and comparison with Himalayan tectonics Geological Society, London, Special Publications, January 1, 2006; 268(1): 543 - 559. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||