Grenville-age pseudotachylite in the Lewisian: laserprobe 40Ar/39Ar ages from the Gairloch region of Scotland (UK)

Sarah C. Sherlock, Kevin A. Jones and R.Graham Park
Journal of the Geological Society, 165, 73-83, 14 January 2008, https://doi.org/10.1144/0016-76492006-134

Abstract

Infrared laserprobe 40Ar/39Ar dating has been used to date pseudotachylite and host-rock minerals from a crush belt in the Lewisian basement of Scotland. It has revealed complexity in the pseudotachylite data that is attributable to the presence of refractory host-rock clasts and mineral fragments in the pseudotachylite. In conjunction with the host-rock mineral laserprobe 40Ar/39Ar data it has been possible to simplify the pseudotachylite data for the samples, and the preferred ages for these are: 980 ± 39 Ma, 999 ± 31 Ma and 1024 ± 30 Ma (2σ). These ages are the first record of Grenville-aged brittle deformation in the Lewisian. Further, this study serves to illustrate the complexity of dating pseudotachylites, and the advantages and limitations of the IR laserprobe applied to such materials.

Dating pseudotachylites by 40Ar/39Ar is a useful means to establish the timing of brittle, seismic, deformation in crustal rocks in fault zones (Kelley et al. 1994; Karson et al. 1998; Magloughlin et al. 2001; Sherlock & Hetzel 2001; Müller et al. 2002; Davidson et al. 2003; Kohút & Sherlock 2003; Sherlock et al. 2004) and in meteorite impact craters (Reimold et al. 1990; Spray. 1995; Kelley & Spray 1997; Thompson et al. 1998). For major crustal fault zones this technique is important because it is the only method that can date the near-instantaneous seismic failure events, and has proven useful in establishing the causal mechanisms and timing of sedimentation (Kohút & Sherlock 2003). Pseudotachylites are friction melts that form and solidify rapidly, and are usually a mixture of glass, devitrified glass, fine-grained newly crystallized minerals and microlites, refractory host-rock clasts and mineral fragments, and patches of alteration (Sibson 1977). In many cases a margin-parallel zoning and flow-banding are preserved (e.g. Sherlock et al. 2004). The heterogeneity of such material means that the argon system and any 40Ar/39Ar age data are complex and difficult to interpret, but where host-rock minerals are dated in conjunction with pseudotachylite matrix it is possible to decipher the age of the pseudotachylite-forming event. There are two main assumptions made when applying the 40Ar/39Ar dating system to pseudotachylite: (1) that all the measured 40Ar* (radiogenic) in the pseudotachylite analyses accumulated since the pseudoachylite cooled through closure to argon diffusion; (2) that the pseudotachylite has not experienced heating and argon loss subsequent to formation. The closure to argon diffusion in pseudotachylite refers to the whole rock. Pseudotachylite matrix is analogous to ultra-fine-grained igneous rocks that have cooled rapidly, and so a pseudotachylite age represents the time at which the whole rock cooled rather than any one mineral within it. However, unlike an ultra-fine-grained igneous rock, pseudotachylite also contains refractory minerals because the high temperatures were too short lived to have caused bulk melting of the host-rock. Consequently, when discussing the closure temperature in the context of a pseudotachylite, and what the 40Ar/39Ar age actually represents, it is the ‘cooling age’ of the ultra-fine grained matrix. With regard to whether or not a pseudotachylite has lost argon during subsequent heating or deformation, given the ultra-fine-grained nature of pseudotachylite matrix it is assumed that it will lose 40Ar during subsequent thermal events by diffusion to the pseudotachylite margin. The three key parameters are temperature and duration of the thermal event, and diffusion length scale. The third parameter is the most difficult to quantify because pseudotachylite is not monomineralic, it is not a coherent lattice, and it contains physical ‘sinks’ for argon at the interfaces between entrained refractory clasts and matrix, and the matrix and host-rock. As yet there are no published studies, but to assess whether or not any given pseudotachylite has been ‘reset’ by subsequent thermal events it is necessary to also date the host-rock minerals by 40Ar/39Ar as a direct comparison, assuming that their effective closure temperature is always going to be higher than that of the pseudotachylite.

Provided that these two conditions are met, dating pseudotachylites can be successful, although, in practice, Ar-isotope heterogeneity is often problematic. This heterogeneity is introduced because of the nature of the friction melting process: physical breakdown and melting of mechanically less stable minerals, which occurs very rapidly and at the focus of the highest strain. In practice, it is predominantly biotite that melts and contributes to the pseudotachylite, and if a pseudotachylite formed within a rock that was 100% biotite, then it is likely to be homogeneous in terms of argon isotope distribution. In samples where biotite is 20% modal proportion and with 10 wt% potassium, and amphibole 10% modal proportion with just 2 wt% potassium, for example, the resulting pseudotachylite will be heterogeneous in terms of the distribution of argon. This is further exacerbated by the rupturing of fluid inclusions in host-rock quartz, during high strain, which contributes excess 40Ar to the pseudotachylite matrix. In this study we apply the IR laserprobe 40Ar/39Ar dating method to pseudotachylite-bearing crush breccias in the Gairloch region of the Lewisian basement of mainland Scotland, which we show to be of Grenvillian age.

According to continental reconstruction models the Lewisian Complex resided on the far eastern or southeastern seaboard of Meso- and Neo-Proterozoic Laurentia (Buchan et al. 2000; Dalziel & Soper 2001). The margin of the Mesoproterozoic Grenville belt probably crossed Scotland immediately SE of the mainland Lewisian outcrop (Fig. 1), and there is evidence for Grenville-aged high-temperature metamorphism in the Lewisian of mainland Scotland (Giletti et al. 1961; Moorbath & Park 1972; Sanders et al. 1984; Brewer et al. 2003), the Outer Hebrides archipelago (Cliff & Rex 1989), and the Inner Hebrides (Daly & Flowerdew 2005). The Lewisian of Scotland is also noted for the preservation of thick Neoproterozoic sedimentary sequences derived from the erosion of the Grenville-age mountains, which are divided into the Stoer, Sleat and Torridon Groups. Overlying the Lewisian basement is the Stoer Group, the base of which is a massive conglomeratic breccia facies deposited in a fan setting (Stewart 1991). The deposition age of the Stoer Group is derived from a Pb–Pb age on the calcite fraction of the Stoer Group Limestone, and is 1199 ± 70 Ma (Turnbull et al. 1996). From this it was suggested that Stoer Group deposition was contemporaneous with Lewisian uplift, during a compressional phase of the Grenville orogeny (Turnbull et al. 1996). A more recent palaeomagnetic study of authigenic hematite within the Stoer Group yielded an age of c. 1180 Ma for early diagenesis (Darabi & Piper 2004), and the study concluded that that Stoer Group was lithified and had undergone deformation prior to Torridon Group deposition (Darabi & Piper 2004). The Sleat Group occurs only within the Kishorn Nappe on Skye (Stewart 1991), and so the younger Torridon Group is unconformable on the Stoer Group and in most places rests on Lewisian basement. The Sleat and Torridon Groups were probably deposited shortly after the Grenville orogeny, and the stratigraphic thickness and areal extent of more than 200 km in length suggests that major rivers deposited these groups in a large-scale basin. The Torridon Group basin is likely to be extensional, with the Sleat Group representing early rifting. Whole-rock Rb–Sr regression ages from silts within the Applecross and Diabaig Formations of the Torridon Group are 994 ± 48 Ma and 977 ± 39 Ma, and are interpreted as reflecting the timing of early diagenesis (Turnbull et al. 1996).

Fig. 1.
Fig. 1.

 Late Proterozoic plate-scale reconstruction showing the location of the Grenville belt and the position of NW Scotland.

Geological setting

The study area (Fig. 2a) is in the centrally located Carnmore–Tollie block (Park 2005), alternatively termed the Gairloch Terrane by Kinny et al. (2005). Here, the Carnmore and Tollie Archaean gneisses are structurally beneath supracrustal rocks of the Palaeoproterozoic Loch Maree Group, and intruded by a suite of late Laxfordian c. 1695 Ma pegmatites (Park et al. 2001). The Loch Maree Group is a suite of supracrustal rocks consisting of metasediments interbanded with amphibolites of volcanic origin (Park 2002) resulting from the accretion of oceanic and volcanic arc components to the upper continental plate of a subduction zone. A number of narrow NNW–SSE- to NW–SE-trending pre-Torridonian zones of structural weakness traverse the rocks in the Gairloch area, which generally exhibit both ductile and brittle characteristics (Lei & Park 1993), and were termed ‘crush zones’ by Peach et al. (1907). These ductile–brittle shear or fault zones are located within broader ductile shear zones developed during earlier Laxfordian tectonometamorphic events. The localized structures of various styles and orientations associated with these zones, most of which contain crush breccias with pseudotachylite, were assigned to the ‘late phase’ of the Laxfordian by Park (1964).

Fig. 2.
Fig. 2.

 Location of the study area. (a) Location of the Gairloch area in the context of Scotland, and the terranes of the Scottish Lewisian (after Park 2005): N, Nis block; T, Tarbert block; SH, South Harris block; WU, West Uist block; OHFZ, Outer Hebrides Fault Zone; Rh, Rhiconich block; A, Assynt block; G, Gruinard block; C-T, Carnmore–Tollie block; RM, Ruadh Mheallan block; Ro, Rona block; GSZ, Gairloch Shear Zone; MT, Moine Thrust. (b) Leth-chriege crush belt located within the Tollie antiform area to the east of Gairloch: LMF, Loch Maree fault; TFF, Tollie Farm fault; L-C CB, Leth-chreige crush belt; TA, Tollie antiform; CBB, Creag Bhan belt; FF, Flowerdale fault.

The Leth-chreige crush breccia is one such ‘crush zone’ and is located on the NE side of the Tollie antiform (Fig. 2b), immediately west of Loch Maree. The fault zone extends from the shore of Loch Maree at Slattadale for 7 km in a NNW direction to Tollie Farm, where it is displaced sinistrally by the Tollie Farm fault, a splay of the Loch Maree fault (Fig. 2b). It can be traced for a further 3.5 km on the NW side of this fault. The fault zone dips at 55–65° E and varies in width from c. 100 m to over 500 m. It consists of several narrow strands of crush breccia and highly fractured rock separated by wider panels of less fractured or unfractured rock.

Both dextral and sinistral senses of movement indicators were recorded on the Leth-chreige belt by Lei & Park (1993) but on this, as on the other similarly oriented belts, the sinistral indicators are dominant in the later phase associated with the pseudotachylite formation (Beacom et al. 2001; Park 2002). The combination of sinistral strike-slip movement on the NNW–SSE Leth-chreige crush belt and dextral on the WNW–ESE Flowerdale fault (which is of similar age and also contains pseudotachylite-bearing crush breccias) cannot be explained by late Laxfordian north–south to NE–SW compression (Park 1970) but is consistent with NW–SE compression, approximately perpendicular to the likely trend of the Grenville front. The NW–SE pseudotachylite-bearing faults are cut by a younger set of wrench-faults (dextral north–south and sinistral east–west faults) and by NE–SW normal faults and extensional clastic dykes (Beacom et al. 1999; Park 2002). These younger faults are consistent with NW–SE extension and are considered to be associated with the formation of the Torridon Group basin.

Stoer Group sediments overlie Lewisian basement in the north of the Gairloch region, and on the east side of Loch Ewe, NE of the Loch Maree fault. They dip at c. 20° NW and are overstepped by Torridon Group beds dipping to the SW or south. Further south, around and within the main part of the inlier, the Lewisian basement is overlain directly by the Torridon Group beds.

Sample descriptions

The crush breccia itself is a dominant topographic feature, forming a series of upstanding ridges some 40 m high. The sample locality is [GR 862785] and can be reached by a footpath south of the A832, the upstanding ridge lying to the east of the path after c. 750 m distance (location 57°44′13″N, 5°39′56″W). The crush breccia occurs in discrete pods measuring c. 15 cm in width and height, and comprises a mix of randomly oriented pieces of basic and acid gneiss basement (Fig. 3). These are set in a matrix of pseudotachylite that ranges in colour from brown to grey; the freshest material is black and vitreous.

Fig. 3.
Fig. 3.

 Leth-chriege crush belt outcrop: (a) field sketch of the pseudotachylite breccia lens that hosts disrupted blocks of Lewisian gneiss host-rock; (b) photograph of the pseudotachylite breccia lens that is depicted in (a). The black arrows on the photograph highlight pooled areas of pseudotachylite.

Sample SCS0303Aiii

Sample SCS0303Aiii is a 5 mm wide (maximum) pseudotachylite vein hosted by coarse-grained acid gneiss (Fig. 4a). The host gneiss contains amphibole (hornblende), K-feldspar, plagioclase and quartz, which form a strong gneissose fabric. To the ‘north’ of the pseudotachylite vein the acid gneiss is highly altered. Reasons for the significant contrast in alteration state above and below the pseudotachylite vein are not clear; it might be that the pseudotachylite acts as a barrier to fluids. Where the host-gneiss itself is altered it is in association with fractures that cut across the gneissose fabric, rather than being a pervasive feature. The contact between the pseudotachylite vein and the host-rock gneiss appears sharp (Fig. 4a), but at the millimetre scale it is less regular and there are two end-members of contact type along the margin: in the first a band of fine-grained crushed quartz is present (Fig. 4b), whereas in the other this is absent (Fig. 4c). Quartz grains range in size from c. 10 μm to c. 200 μm in diameter, mainly in the 50–10 μm range, and the width of the band is dictated by the diameter of the largest quartz grain. Where there is no crushed quartz band, the pseudotachylite matrix is in contact with host-rock minerals (Fig. 4c). In both cases there is an abrupt change in pseudotachylite matrix composition at the margin (Fig. 4b and c), but similar to the crushed quartz band, this is not present in all areas of the margin. The width of this material never exceeds 300 μm. The innermost (darker) matrix is essentially alkali feldspar in composition, and the dominant (lighter) matrix is lower in Na2O and Al2O3 and significantly higher in FeO (Table 1). The dominant matrix in the pseudotachylite is an ultra-fine groundmass of radiating microlites interspersed with ultra-fine oxide ‘dust’ (Fig. 4d), with fragments of quartz and to a lesser extent of K-feldspar. Quartz fragments are c. 20–75 μm in diameter, and feldspar fragments rarely exceed 20 μm in diameter, whereas the groundmass microlites measure c. 1–10 μm in width (Fig. 4d). Single microlites are beyond the resolution of the electron microprobe.

View this table:
Table 1. 

Electron microprobe analyses of light and dark zones in SCS0303Aiii and SCS0303Dii pseudotachylite matrix

Fig. 4.
Fig. 4.

 (a) Photograph of the polished thick sections of pseudotachylite sample SCS0303Aiii; the white dashed box highlights the piece that was irradiated and the dot–dashed line is the location of the 40Ar/39Ar age traverse in Figure 7a; (b) SEM image of SCS0303Aiii pseudotachylite margin heavily laden with host-rock quartz clasts and with a marked contrast difference that coincides with the band of quartz clasts; (c) SCS0303Aiii pseudotachylite margin with a reduced concentration of host-rock quartz clasts but still a marked contrast difference; (d) SCS0303Aiii matrix with a ‘dusting’ of oxide and ultra-fine-grained radiating microlites.

Sample SCS0303Dii

Sample SCS0303Dii is a series of veinlets that reach a maximum width of 5 mm and is hosted by a medium-grained biotite gneiss (Fig. 5a). The host gneiss contains biotite, quartz and alkali feldspar. The contacts between the veinlets and host gneiss appear sharp (Fig. 5a), but are also less regular and with two contact types at the SEM scale: areas along the margin where there is a clear compositional difference in the pseudotachylite matrix (Fig. 5b), and areas where there is not (Fig. 5c). Where there is a strong compositional difference at the pseudotachylite margin, indicated by the change from dark to light grey, there is an associated abundance of quartz and feldspar mineral fragments (Fig. 5c). Quartz is the dominant mineral and fragments range from c. 20 to 350 μm in diameter; the less abundant feldspars never exceed 250 μm but are more commonly in the 20–50 μm size range. The width of the dark pseudotachylite matrix, in contact with the host-rock, rarely exceeds 300 μm. This darker material is essentially alkali feldspar in composition, and the dominant (lighter) matrix is lower in Na2O, CaO and Al2O3 and higher in K2O, MgO and FeO (Table 1). The dominant matrix is an ultra-fine groundmass, which, in contrast to sample SCS0303Aiii, does not contain microlites. It does contain refractory quartz and feldspar fragments, and also clasts of host-rock that have not been broken down into their constituent minerals. For example, in Figure 5d a 200 μm × 300 μm quartz–feldspar clast in the main pseudotachylite matrix is mantled on three sides by a c. 150 μm thick zone of the darker pseudotachylite matrix and an abundance of <20 μm quartz, and lesser feldspar, fragments.

Fig. 5.
Fig. 5.

 (a) Photograph of the polished thick sections of pseudotachylite sample SCS0303Dii; dot–dashed lines (1) and (2) are the locations of the 40Ar/39Ar age traverses in Figure 7b and c; (b) SCS0303Dii irregular pseudotachylite margin with a high concentration of refractory host-rock quartz and a marked contrast change; (c) SCS0303Dii pseudotachylite margin with a much lower concentration of refractory host-rock quartz clasts and no marked change in contrast; (d) refractory quartz–feldspar host-rock clasts in SCS0303Dii, surrounded by pseudotachylite matrix and on three sides mantled by a zone of darker pseudotachylite matrix and finer-grained quartz, and subordinate feldspar, grains.

Sample SCS0303Biii

Sample SCS0303Biii is a piece of pseudotachylite matrix but was not analysed in conjunction with its host-rock, which has prevented the host-rock–pseudotachylite traverse approach adopted in samples SCS0303Aiii and SCS0303Dii. This is due to the complex nature of sample SCS0303Biii (Fig. 6a), where the pseudotachylite vein is hosted by mylonite (white box 2 in Fig. 6a) and altered biotite gneiss (white box 1 in Fig. 6a). The pseudotachylite matrix is an ultra-fine-grained groundmass with radiating microlites that are c. 10 μm (Fig. 6b) in width and beyond the resolution of the electron microprobe. Quartz fragments occur in the matrix (Fig. 6b) and these are commonly in the size range 20–200 μm.

Fig. 6.
Fig. 6.

 (a) Photograph of the polished thick sections of pseudotachylite sample SCS0303Biii; (1) and (2) correspond to pseudotachylite adjacent to altered biotite gneiss, and adjacent to mylonite; (b) SCS0303Biii pseudotachylite matrix, an ultra-fine-grained groundmass with radiating microlites.

Analytical methods

All uncertainties are presented at the 2σ level unless otherwise indicated. All 40Ar/39Ar data have been produced by S.C.S. in the Argon–Argon and Noble Gas Research Laboratory at The Open University, and electron microprobe data by S.C.S. in the Department of Earth and Environmental Sciences, The Open University. Polished thick sections (300 μm thick) were prepared, and 5 mm × 5mm sections selected for irradiation. Samples were cleaned ultrasonically in alternate de-ionized water and methanol prior to packaging in aluminium foil. Samples were irradiated in the McMaster (Canada) reactor for 50 h, neutron flux was monitored using GA1550 biotite standard (98.79 ± 0.96 Ma, Renne et al. 1998) yielding a J value of 0.01175 ± 0.000057. The K decay constant of Steiger & Jäger (1977) has been used. In situ laser spot analyses were performed using a Spectron Lasers Ltd SL902 CW Nd–YAG 1064 nm laser with a manual shutter. Analyses were carried out by firing the laser in 30 ms shots, commonly one shot per analysis, resulting in pits of c. 75 μm diameter. Unwanted gas species were removed from released gases by two SAES getters, one operating at room temperature and the other at 450 °C, for 5 min prior to automatic inlet into an MAP 215-50 noble gas mass spectrometer, where 41Ar to 35Ar were analysed. Analyses were corrected for blanks measured either side of two consecutive samples analyses, 37Ar decay and neutron-induced interference reactions using the correction factors: (39Ar/37Ar)Ca=0.00065±0.0000033, (36Ar/37Ar)Ca = 0.000264±0.0000013 and (40Ar/39Ar)K=0.0085± 0.0 based on the measurement of K and Ca salts, and the mass discrimination value used was 283 ± 1 measured with multiple glass analyses fused with an IR laserprobe. For samples SCS0303Aiii and SCS0303Dii, laser spots were positioned in host-rock amphibole or biotite, and within the pseudotachylite, at varying distances from the host-rock–pseudotachylite interface. For sample SCS0303Biii the laser spot was positioned within the pseudotachylite matrix with no distinction between distance to the host-rock–pseudotachylite margin. All weighted mean ages have been calculated using Isoplot 3 (Ludwig 2003).

A key problem that prevails in 40Ar/39Ar microanalysis by laserprobe is that the ultra-small 36Ar beam intensities in both samples and blanks approach detection limits. In addition, the measured 36Ar peaks in sample analyses commonly fall within analytical error of the 36Ar blank measurements. The small variations in 36Ar are magnified by the correction for atmospheric argon, and result in anomalously high analytical errors on the final 40Ar/39Ar age if fully applied. The actual atmospheric contents measured on larger samples, in the present study, are generally less than 5% and vary by as much as a few per cent, so a more precise age can be obtained by assuming that the atmospheric content is similar to the mean blank variation, a procedure that results in analytical errors of around double the error on the raw ratio, in comparison with much larger errors induced by error magnification resulting from the 36Ar atmospheric correction. Each datapoint is assessed in terms of measured 36Ar values v. blank 36Ar values and uncertainties. Where measured 36Ar is within error of blank 36Ar no atmospheric correction is applied but the final errors are doubled. In this study there were only three analyses that had measured 36Ar values that were greater and outside of the error of the blank 36Ar values: point 16 in SCS0303Aiii, points 2 and 3 in SCS0303Biii (1) and points 1 and 6 in SCS0303Biii (2). All 40Ar/39Ar laserprobe data are presented in Table 2. Electronprobe microanalyses were obtained with a Cameca SX100, using operating conditions of 20 kV accelerating voltage, 20 nA beam current and both 10 and 20 μm beam sizes. Elements were analysed with four wavelength-dispersive detectors.

View this table:
Table 2. 

Infrared laserprobe 40Ar/ 39Ar data for pseudotachylite and host-rock analyses (× 1010, cm3 STP)

Results

40Ar/39Ar data are available online at http://www.geolsoc.org.uk/SUP18283. The 40Ar/39Ar amphibole ages from sample SCS0303Aiii form a traverse from the host-rock into and across the pseudotachylite vein (Fig. 7a). The ages of the analyses within the host-rock amphiboles range from 1558 ± 24 to 1690 ± 23 Ma (Table 2). The pseudotachylite ages range from 895 ± 4 to 1119 ± 11 Ma. From sample SCS0303Dii there are two traverses (Fig. 7b and c). In traverse 1 the host-rock biotites range in age from 1036 ± 34 to 1242 ± 23 Ma (Table 2), and in traverse 2 the host-rock biotite ages range from 1128 ± 11 to 1305 ± 28 Ma (Table 2). The pseudotachylite ages in traverse 1 range from 823 ± 11 to 1282 ± 39 Ma; there is a cluster of ‘older’ ages in the centre of the vein (Fig. 7b). In traverse 2 the pseudotachylite ages range from 970 ± 13 to 1165 ± 10 Ma, with no systematic age variation with distance from the host-rock–pseudotachylite margin (Fig. 7c). From sample SCS0303Biii there are 18 laser spot ages distributed between the two pieces that were irradiated (Fig. 6a) and these range from 1003 ± 11 to 1177 ± 19 Ma.

Fig. 7.
Fig. 7.

 40Ar/39Ar laserprobe analyses plotted against distance for the three traverses: (a) SCS0303Aiii; (b) SCS0303Dii (1); (c) SCS0303Dii (2). The grey boxes highlight the data points that have been used to calculate the final ages.

Discussion

Host-rock 40Ar/39Ar ages

The 40Ar/39Ar laserprobe data are complex, but using these in conjunction with the host-rock mineral data and detailed sample observations it is possible to decipher geochronological information. The Leth-chriege crush zone and associated pseudotachylites contains basement pieces of different lithologies and they record slightly different isotopic information. In sample SCS0303Aiii the host-rock amphiboles range in age from 1558 ± 24 to 1690 ± 23 Ma. There are a number of physical processes that could be responsible for this range in age: more than one amphibole-bearing metamorphic fabric in the rock; heterogeneity in the age of single amphiboles that is beyond the resolution of the IR laserprobe; the resolution of the IR laserprobe being too low to avoid releasing argon from surrounding minerals during analysis. All these are inherent problems in dating metamorphic fabrics with an in situ IR laserprobe 40Ar/39Ar dating method (Reddy et al. 1997; Sherlock et al. 2004). The same is true of biotite, and in sample SCS0303Dii the host-rock biotites range in age from 1036 ± 34 to 1305 ± 23 Ma. This age range may also reflect more than one metamorphic fabric in the rock. It is also possible that because biotite is the least mechanically stable mineral, and has a low closure temperature for the Ar system, these minerals have been affected by the intense deformation and heating during pseudotachylite formation to a greater degree than have the amphiboles. The biotites are also much smaller than the amphiboles (Figs 4a and 5a⇑) and so dating them using the IR laserprobe method is more likely to bring about loss of argon from surrounding minerals.

The explanation of the different ages of amphibole and biotite host-rock minerals, both measured in this study and published elsewhere, is probably to be explained by their different closure temperatures for the Ar system. Thus the c. 1600 Ma amphibole ages record cooling from a thermal event prior to 1600 Ma (i.e. the c. 1700 Ma Late Laxfordian), whereas the c. 1100–1300 Ma biotite ages record cooling from a separate thermal event that did not achieve temperatures high enough to reset amphibole. These ages are consistent with the published record of Rb–Sr and K–Ar biotite ages for the Lewisian of the Scottish mainland, which fall in the range c. 1148–1169 Ma (Giletti et al. 1961; Moorbath & Park 1972), upon which the presence of a Grenville-aged thermal event has been based (e.g. Moorbath & Park 1972; Cliff & Rex 1989). The temperature of this event must have reached or exceeded c. 300 °C to reset or partially reset the biotites, but did not reach c. 500–550 °C or the amphiboles would also have experienced resetting.

Pseudotachylite 40Ar/39Ar ages

The pseudotachylite data are complex and in each sample there is a wide range of ages: those for sample SCS0303Aiii span c. 230 Ma, those for SCS303Dii span c. 459 Ma and c. 142 Ma (traverses 1 and 2, respectively), and there is a 174 Ma spread in ages for sample SCS0303Biii. Arriving at a definitive age for these pseudotachylites is not straightforward and several approaches might be taken. The first is to calculate a weighted mean for the pseudotachylite spot ages, which takes into account the weighting of the errors on each datapoint. In a second approach, it might be appropriate to cite the youngest pseudotachylite ages for each sample and describe them as the ‘minimum’ age for pseudotachylite formation. These approaches are too simplified for complex data arising from heterogeneous samples. It is known from textural analysis that the pseudotachylites are strongly heterogeneous and contain an abundance of entrained host-rock clasts. In these samples these are mainly located at the pseudotachylite margins (Figs 4 and 5⇑) and so it would be appropriate to discount datapoints from within the first 300 μm of each pseudotachylite vein, as these datapoints are most likely to be a mixture of pseudotachylite matrix and refractory clasts. Also, because the pseudotachylite must be younger than any of the host-rock minerals, knowledge of the thermal history of the host-rock from dating the host-rock minerals can be used to discount pseudotachylite datapoints that are older than the host-rock. These probably reflect the incorporation of excess 40Ar, or inherited 40Ar, that failed to outgas from melting host-rock minerals during the friction melting event. For sample SCS0303Biii there were no host-rock mineral analyses as the pseudotachylite vein was sampled in isolation. However, on the basis that the host-rocks from the same outcrop would have undergone the same thermal history, the ages from the host-rock minerals from samples SCS0303Aiii and SCS0303Dii can be used to remove the oldest data points from samples SCS0303Biii. Applying these criteria to the data the final ages are 980 ± 39 Ma for SCS0303Aiii, 999 ± 31 Ma for SCS0303Dii and 1024 ± 30 Ma for sample SCS0303Biii.

The two questions now are whether or not the data from the three samples should be treated separately or pooled as one dataset, and which of the different methods is the most appropriate for describing the ages of the pseudotachylites. It would seem logical to pool the data from the three samples on the basis that they are derived from the same outcrop, and formed during the same friction melting event. Excluding datapoints that are from within the margin areas that are laden with refractory mineral clasts, and datapoints that are within error of the youngest host-rock age, and excluding the youngest pseudotachylite datapoints from each sample, the weighted mean age is 967 ± 20 Ma (n&=14, MSWD=46). Given the very high MSWD value, which indicates the presence of multiple populations within this dataset, this is not an appropriate method of calculating a final age for these samples. It also demonstrates that the Ar-isotope heterogeneity in pseudotachylite, certainly in these three samples, is beyond the resolution of the IR laserprobe 40Ar/39Ar dating technique. There is clearly more than one ‘age’, or argon reservoir, represented in the dataset, and with no further reasonable criteria with which to distinguish or identify these different components that must be beyond the resolution of the IR laserprobe technique, the remaining option is to state the age range and associated errors (910 ± 19 to 1019 ± 14 Ma); thus the seismic faulting event giving rise to the pseudotachylite breccia has an age between 910 and 1019 Ma, or c. 900–1000 Ma.

Implications of the pseudotachylite ages

Seismic faulting and pseudotachylite formation at between 910 ± 19 and 1019 ± 14 Ma has implications for the timing of deformation, of basement and Stoer Group uplift and erosion, and of Torridon Group deposition.

The combination of sinistral displacement on the NNW–SSE Leth-chriege belt and dextral displacement on the WNW–ESE Flowerdale fault, together with other similarly oriented pseudotachylite-bearing structures, suggests that the seismic faulting responsible for these structures formed under NW–SE compression, perhaps in a zone of foreland thrusting near the contemporary Grenville front. The Grenvillian thrusting event in NE Labrador, which would have lain along-strike and to the SW of the present area, is dated at 989 ± 12 Ma, based on U–Pb zircon age data (Connelly & Heaman 1993). The zircons are from a mylonitized granite, deformed during the juxtaposition of two terranes (Lac Joseph and Molson Lake terranes), and the 989 ± 12 Ma is defined by the lower intercept, interpreted as recording the timing of shearing (Connelly & Heaman 1993).

The phase of uplift that exposed the present erosion surface and removed overlying Lewisian and Stoer Group cover is considered to have formed part of a regional late or post-Grenville extension that accompanied the formation of the Torridon Group depositional basins (e.g. Williams 2001). The effects of this NW–SE extension are represented by the north–south and east–west wrench faults and the NE–SW normal faults that cut the earlier pseudotachylite-bearing structures. In this extensional regime, the Gairloch area would represent a north-tilted extensional fault block bounded by east-dipping normal faults in the Minch to the west and along the line of the Moine thrust zone in the east. This is supported by evidence of a c. 1.08 Ga Grenvillian extension along an east-dipping fault zone in the Glenelg area, described by Temperley & Windley (1997). The Torridon Group depositional basins would form in the hanging wall of the western fault while the eastern part of the block, forming the footwall of the eastern bounding fault, would be uplifted and eroded. With time, the Torridon Group deposition would gradually encroach eastwards to lie directly over the whole of the Gairloch area. In this scenario, the timing of the uplift would correspond to the depositional date of the Torridon Group (i.e. 994 ± 48 Ma and 977 ± 39 Ma; Turnbull et al. 1996). If the Stoer Group beds originally extended over the whole of the Gairloch inlier, a considerable thickness of Lewisian rocks (up to 8 km if the 20° dip is projected to the southern edge of the inlier) could have been removed during this erosive period, along with up to 2 km of Stoer Group beds, prior to the formation and deposition of the Torridon Group basin.

The new Grenville age for pseudotachylite formation, given the current precision, is indistinguishable from the published ages for the extensional events responsible for basement uplift and erosion. If, as we suggest, the pseudotachylite was formed under Grenvillian compression, the time interval between this compressional event and the subsequent extension is beyond the resolution of this dating method. The possibility that the samples analysed and resulting 40Ar/39Ar age data reflect reactivation of a compressional structure by later extension was considered, but we have found no textural evidence within these pseudotachylites to support this.

Conclusions

The new 40Ar/39Ar laserprobe data provide the first record of Grenville-age deformation in the Lewisian foreland and confirm previous suggestions that pre-Torridonian faults cutting the Lewisian basement were related to the Grenville orogeny. The orientation and sense of movement of the pseudotachylite-bearing faults is consistent with NW–SE compression that could be attributed to Grenvillian foreland thrusting at c. 1.0 Ga; this post-dates Stoer Group deposition at c. 1180 Ma (Turnbull et al. 1996). The fact that the pseudotachylite dates are within error of the depositional dates for the Torridon Group suggests that the uplift of the Lewisian basement, which formed part of a late or immediately post-Grenvillian extensional event, closely followed the earlier compressional event.

Acknowledgements

S.C.S. acknowledges NERC Fellowship NER/I/S/2002/00692 and R. Strachan for valuable comments on an earlier version. All the authors gratefully acknowledge the detailed and constructive reviews of C. Davidson and M. Villeneuve.

References

  1. ↵
    Beacom, L.E., Anderson, T.B. & Holdsworth, R.E. 1999. Using basement-hosted clastic dykes as syn-rifting palaeostress indicators: An example from the basal Stoer Group, northwest Scotland. Geological Magazine, 136, 301–310.
    OpenUrlAbstract
  2. ↵
    Beacom, L.E., Holdsworth, R.E., McCaffrey, K.J.W. & Anderson, T.B. 2001. A quantitative study of the influence of pre-existing compositional and fabric heterogeneities upon fracture-zone development during basement reactivation. In: Holdsworth, R.E., Strachan, R.A., Magloughlin, J.A. & Knipe, R.J. (eds) The Nature and Tectonic Significance of Fault Zone Weakening. Geological Society, London, Special Publications, 186, 195–211.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Brewer, T.S., Storey, C.D., Parrish, R.R., Temperley, S. & Windley, B.F. 2003. Grenvillian age decompression of eclogites in the Glenelg–Attadale Inlier, NW Scotland. Journal of the Geological Society, London, 160, 565–574.
    OpenUrlCrossRefWeb of Science
  4. ↵
    Buchan, K.L., Mertanen, S., Park, R.G., Pesonen, L.J., Elming, S.-A., Abrahamsen, N. & Bylund, G. 2000. Comparing the drift of Laurentia and Baltica in the Proterozoic: the importance of key palaeomagnetic poles. Tectonophysics, 319, 167–198.
    OpenUrlCrossRefWeb of Science
  5. ↵
    Cliff, R.A. & Rex, D.C. 1989. Evidence for a ‘Grenville’ event in the Lewisian of the northern Outer Hebrides. Journal of the Geological Society, London, 146, 921–924.
    OpenUrlCrossRefWeb of Science
  6. ↵
    Connelly, J.N. & Heaman, L.M. 1993. U–Pb geochronological constraints on the tectonic evolution of the Grenville Province, western Labrador. Precambrian Research, 63, 123–142.
    OpenUrlCrossRefWeb of Science
  7. ↵
    Daly, J.S. & Flowerdew, M.J. 2005. Grampian and late Grenville events recorded by mineral geochronology near a basement–cover contact in north Mayo, Ireland. Journal of the Geological Society, London, 162, 163–174.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Dalziel, I.W.D. & Soper, N.J. 2001. Neoproterozoic extension on the Scottish promontory of Laurentia: palaeogeographic and tectonic implications. Journal of Geology, 109, 299–317.
    OpenUrlCrossRefWeb of Science
  9. ↵
    Darabi, M.H. & Piper, J.D.A. 2004. Palaeomagnetism of the (Late Mesoproterozoic) Stoer Group of northwest Scotland: implications for diagenesis, age and relationship to the Grenville Orogeny. Geological Magazine, 141, 15–39.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Davidson, C., Davis, K.J., Bailey, C.M., Tape, C.H., Singleton, J. & Singer, B. 2003. Age, origin, and significance of brittle faulting and pseudotachylyte along the coast shear zone, Prince Rupert, British Columbia. Geology, 31, 43–46.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Giletti, B., Moorbath, S. & Lambert, R. 1961. A geochronological study of the metamorphic complexes of the Scottish Highlands. Quarterly Journal of the Geological Society of London, 117, 233–264.
    OpenUrlCrossRef
  12. ↵
    Karson, J.A., Brooks, C.K., Storey, M. & Pringle, M.S. 1998. Tertiary faulting and pseudotachylytes in the East Greenland volcanic rifted margin: Seismogenic faulting during magmatic construction. Geology, 26, 39–42.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Kelley, S.P. & Spray, J.G. 1997. A late Triassic age for the Rochechouart impact structure, France. Meteoritics and Planetary Science, 32, 629–636.
    OpenUrl
  14. ↵
    Kelley, S.P., Reddy, S.M. & Maddock, R. 1994. Laser-probe 40Ar/39Ar investigation of a pseudotachylyte and its host rock from the Outer Isles thrust, Scotland. Geology, 22, 443–446.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Kinny, P.D., Friend, C.R.L. & Love, G.J. 2005. Proposal for a terrane-based nomenclature for the Lewisian Gneiss Complex of NW Scotland. Journal of the Geological Society, London, 162, 175–186.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Kohút, M. & Sherlock, S.C. 2003. Laser microprobe 40Ar/39Ar analysis of pseudotachylyte and host-rocks from the Tatra Mountains, Slovakia: Evidence for late Palaeogene seismic/tectonic activity. Terra Nova, 15, 417–424.
    OpenUrlCrossRefWeb of Science
  17. ↵
    Lei, S. & Park, R.G. 1993. Reversals of movement sense in Lewisian brittle–ductile shear zones at Gairloch, NW Scotland, in the context of Laxfordian kinematic history. Scottish Journal of Geology, 29, 9–19.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Ludwig, K.R., 2003. User&s Manual for Isoplot/Ex Version 3.0: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center Special Publications, 4..
  19. ↵
    Magloughlin, J.F., Hall, C.M., Van Der Pluijm, B.A. 2001. 40Ar/39Ar geochronometry of pseudotachylytes by vacuum encapsulation: North Cascade Mountains, Washington, USA. Geology, 29, 51–54.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Moorbath, S. & Park, R.G. 1972. The Lewisian chronology of the southern region of the Scottish Mainland. Scottish Journal of Geology, 8, 51–74.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Müller, W., Kelley, S.P. & Villa, I.M. 2002. Dating fault-generated pseudotachylytes: Comparison of 40Ar/39Ar stepwise-heating, laser-ablation and Rb–Sr microsampling analyses. Contributions to Mineralogy and Petrology, 144, 57–77.
    OpenUrlCrossRefWeb of Science
  22. ↵
    Park, R.G., 1964. The structural history of the Lewisian rocks of Gairloch, Wester Ross. Quarterly Journal of the Geological Society of London, 120, 397–434.
    OpenUrlCrossRef
  23. ↵
    Park, R.G., 1970. Observations on Lewisian chronology. Scottish Journal of Geology, 6, 379–399.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Park, R.G., 2002. The Lewisian Geology of Gairloch, NW Scotland. Geological Society, London, Memoirs, 26..
  25. ↵
    Park, R.G., 2005. The Lewisian terrane model: A review. Scottish Journal of Geology, 41, 105–118.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    Park, R.G., Tarney, J. & Connelly, J.N. 2001. The Loch Maree Group: Palaeoproterozoic subduction–accretion complex in the Lewisian of NW Scotland. Precambrian Research, 105, 205–226.
    OpenUrlCrossRefWeb of Science
  27. ↵
    Peach, B.N., Horne, J., Gunn, W., Clough, C.T., Hinxman, L.W. & Teall, J.J.H. 1907. The Geological Structure of the North-West Highlands of Scotland. Memoir of the Geological Survey..
  28. ↵
    Reddy, S.R., Kelley, S.P. & Magennis, L. 1997. A microstructural and argon laserprobe study of shear zone development at the western margin of the Nanga Parbat–Haramosh Massif, western Himalaya. Contributions to Mineralogy and Petrology, 128, 16–29.
    OpenUrlCrossRefWeb of Science
  29. ↵
    Reimold, W.U., Jessberger, E.K. & Stephan, T. 1990. 40Ar/39Ar dating of pseudotachylite from the Vredefort dome, South Africa: a progress report. Tectonophysics, 171, 139–152.
    OpenUrlCrossRefWeb of Science
  30. ↵
    Renne, P.R., Swisher, C.C., Deino, A.L., Karner, D.B., Owens, T.L. & DePaolo, D.J. 1998. Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating. Chemical Geology, 145, 117–152.
    OpenUrlCrossRefWeb of Science
  31. ↵
    Sanders, I.S., Van Calsteren, P.W.C. & Hawkesworth, C.J. 1984. A Grenville Sm–Nd age for the Glenelg eclogite in north-west Scotland. Nature, 312, 439–440.
    OpenUrlCrossRefWeb of Science
  32. ↵
    Sherlock, S.C. & Hetzel, R. 2001. A laser-probe 40Ar/39Ar study of pseudotachylite from the Tambach Fault Zone, Kenya: Direct isotopic dating of brittle faults. Journal of Structural Geology, 23, 33–44.
    OpenUrlCrossRefWeb of Science
  33. ↵
    Sherlock, S.C., Watts, L.M., Holdsworth, R.E. & Roberts, D. 2004. Dating fault reactivation by 40Ar/39Ar laserprobe: An alternative view of apparently cogenetic mylonite–pseudotachylite assemblages. Journal of the Geological Society, London, 161, 335–338.
    OpenUrlAbstract/FREE Full Text
  34. Sibson, R.H., 1977. Fault rocks and fault processes. Journal of the Geological Society, London, 133, 191–213.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Spray, J.G., 1995. Pseudotachylyte controversy: fact or friction? Geology, 23, 1119–1122.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Steiger, R.H., Jäger, E. 1977. IUGS Subcommission on Geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359–362.
    OpenUrlCrossRefWeb of Science
  37. ↵
    Stewart, A.D., 1991. Geochemistry, provenance and climate of the Upper Proterozoic Stoer Group in Scotland. Scottish Journal of Geology, 26, 89–97.
    OpenUrlWeb of Science
  38. ↵
    Temperley, S. & Windley, B.F. 1997. Grenvillian extensional tectonics in northwest Scotland. Geology, 25, 53–56.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Thompson, L.M., Spray, J.G. & Kelley, S.P. 1998. Laser probe 40Ar/39Ar dating of pseudotachylyte from the Sudbury structure: Evidence for postimpact thermal overprinting in the North Range. Meteoritics and Planetary Science, 33, 1259–1269.
    OpenUrlWeb of Science
  40. ↵
    Turnbull, M.J.M., Whitehouse, M.J. & Moorbath, S. 1996. New isotopic age determinations for the Torridonian, NW Scotland. Journal of the Geological Society, London, 153, 955–964.
    OpenUrlCrossRefWeb of Science
  41. ↵
    Williams, G.E., 2001. Neoproterozoic (Torridonian) alluvial fan succession northwest Scotland, and its tectonic setting and provenance. Geological Magazine, 138, 471–494.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

Journal of the Geological Society: 165 (1)
Journal of the Geological Society
Volume 165, Issue 1
January 2008
Alerts
Citation tools
Permissions
View PDF
Email to
Print
Download PPT