Age Constraints for the Thermal Evolution and Erosional History of the Central European Variscan Belt

By Mazur, Stanislaw; Dunlap, W James; Turniak, Krzysztof; Oberc- Dziedzic, Teresa

Abstract:

Three Carboniferous-age detrital muscovites from the Variscan foreland basin of SW Poland and two muscovites from phyllites underlying the basement have been dated by the ^sup 40^Ar/^sup 39^Ar step-heating and single-crystal laser fusion method. ^sup 40^Ar/ ^sup 39^Ar analysis of the detrital micas defines step-heating preferred ages of 370.7 1.4, 363.0 1.9 and 355.0 1.3 Ma. Single- crystal laser fusion data indicate little dispersion for the first of three samples, with an integrated age that closely matches the step-heating data, but the latter two describe inhomogeneous populations. The white mica concentrate from one phyllite yields a step-heating preferred age of 358.6 1.8 Ma. The second phyllite sample displays an incremental discordant apparent age spectrum representing a mixture of white mica grains of varying ages. Our most important finding is that the Variscan foreland basin was supplied by source rocks that were exhumed and cooled in the Late Devonian, probably as a result of an early Variscan collisional event, previously largely undocumented. Although accessible exposures of the Variscan basement in SW Poland exhibit only a minor component of rocks exhumed before the Carboniferous, our work suggests that large tracts of rocks with a Devonian cooling signature are preserved at depth beneath the foreland basin.

Sedimentary foreland basins adjacent to erogenic belts commonly contain erosional products that can be used to elucidate the structural and thermal evolution of the palaeohinterlands (White et al. 2002). Information derived from foreland basin sediments can provide important constraints on orogenic evolution, particularly because the metamorphic rocks that make up the source areas in the cores of mountain belts, along with the earliest stratigraphie records on the margin of the mountains, tend to be destroyed during orogenic development (Ziegler 1989).

The Variscan north foreland basin extends from Belgium through northern Germany to western Poland (Franke 1995). This basin contains Carboniferous flysch and molasse successions that fringe the NW flank of the Variscan belt. The easternmost part of the Carboniferous basin extends in western Poland between the Variscan crystalline basement of the Bohemian Massif and the SW margin of the East European Craton (Fig. 1). The basin overlies a complex area of Palaeozoic terrane accretion referred to as the Trans-European Suture Zone.

According to the conventional interpretation, the formation of the Variscan foreland basin was preceded by southward subduction of the Old Red Continent passive margin beneath the leading edge of a composite terrane assembled within the Variscan belt (e.g. Franke et al. 1995; Franke 2000) and collectively referred to as the Armorican terrane assemblage (Tait et al. 2000). The following continental collision between the two is thought to have commenced not earlier than the Tournaisian (Franke & Engel 1986; Franke 2000) and caused the emplacement of the Variscan orogenic wedge onto the underthrust passive margin during Visan-Westphalian times (335-305 Ma, e.g. Oncken 1997; Oncken et al. 2000). The resultant thrust loading on the margin of the Old Red Continent induced a regional subsidence related to flexure of the northern Variscan foreland and, thus, the initiation of the foreland basin (Ziegler 1990). The syntectonic turbidite sediments deposited in the foreland basin of Germany were sourced from the Mid-German Crystalline High (e.g. Engel & Franke 1983), the eastern part of which experienced major uplift and cooling at c. 337 Ma (Zen et al. 2000; Schubert et al. 2001). Although the Rhenohercynian flysch basin was supplied abundantly with detritus showing an older K-Ar cooling signature of 380-370 Ma (e.g. Huckriede et al. 1998, 2004), this fact has been thus far more readily interpreted in terms of subduction erosion (Oncken 1998; Franke 2000) than of an early Variscan collisional event (Huckriede et al. 2004). Similarly to the German section of the Variscan belt, the vast majority of Ar/Ar cooling ages obtained in the Polish Sudetes, currently located immediately SW of the Variscan foreland basin, also cluster within the range of 345-330 Ma (Marheine et al. 2002). Although the source area of the Carboniferous flysch of SW Poland has not been reliably determined it is consistently interpreted as derived from the adjacent Sudetic area (e.g. Cwojdzinski & Grocholski 1995). The lack of geochronological data has prevented, however, any conclusive inferences concerning the thermal evolution of the potential source rocks that were exposed in the Sudetes through Carboniferous times.

In this paper, we apply ^sup 40^Ar/^sup 39^Ar total fusion and incremental heating chronology of white mica single crystals and concentrates to palynologically dated Carboniferous sediments and to low-grade metamorphosed basement of the Variscan foreland basin in western Poland. The results constrain the exhumation history of the Variscan hinterland supplying material to the Carboniferous basin and facilitate comparison with the thermal evolution of the now exposed Variscan basement rocks. The data obtained for the basin floor shed new light on its provenance and relationship to the adjacent Variscan belt.

Geological setting

The Variscan foreland basin of western Poland comprises an entirely concealed succession, at least c. 2500 m thick, of Carboniferous clastic marine sediments inverted before the Permian. The Carboniferous succession includes sandy-silty rocks with plant detritus interbedded with conglomerates and claystones. The sandstones are represented mainly by lithic wackes composed of rock, quartz and feldspar grains (Zelichowski 1995). The Carboniferous subcrop zone extends over a considerable area of central western Poland (Fig. 1) and consists of fairly monotonous series of turbidites consistently interpreted as flysch that was deposited during Tournaisian(?) to Westphalian times (all chronostratigraphic ages are after the British Geological Survey timechart; Gradstein & Ogg 1996) and was locally buried below rather thin molasse of Late Westphalian-Stephanian age (Wierzchowska-Kiculowa 1984). Basement rocks underlying the Carboniferous succession remain generally unknown, except for two minor WNW-ESE-trending crystalline highs in southern Wielkopolska (Obere 1972; Fig. 2). The larger of these, the Wolsztyn-Leszno High, consists of phyllites, possibly of Devonian protolith age (Haydukiewicz et al. 1999), which underwent low-grade metamorphism at c. 340 Ma (Zelazniewicz et al. 2003). The northeastern border of the Wolsztyn-Leszno High is defined by the Dolsk Fault, which on seismic refraction profiles corresponds to a major boundary between the low-velocity Variscan-type crust to the SW and the three-layer ‘transitional’ crust of suspected East Avalonian affinities to the NE (Grad et al. 2002). The Carboniferous succession is unconformably overlain by a Permo-Mesozoic sequence of the German-Polish Basin that represents post-Variscan sedimentary cover 1-4 km thick (Fig. 3). The occurrence of the Variscan foreland basin in western Poland is known from over 100 boreholes drilled since the early 1960s, which have penetrated the Carboniferous strata. Despite a considerable database, the sedimentary development and tectonic inversion of the Carboniferous basin are still far from being well understood (Mazur et al. 2003).

Field relationships and petrography of the studied rocks

From four boreholes (Marcinki IG-1, Wrzesnia IG-1, Swieciechowa 1 and Bielawy 1; Fig. 2) representative sandstone and phyllite samples were selected for ^sup 40^Ar/^sup 39^Ar analysis. Two of these wells, Marcinki IG-1 and Wrzesnia IG-1, cored only Carboniferous strata in the southern and northern part of the basin, respectively. The other two boreholes, Swieciechowa 1 and Bielawy 1, encountered a crystalline basement high beneath the Lower Permian and/or Carboniferous sediments, composed of low-grade metamorphosed metapelitic rocks (phyllites). The basement is expressed as two WNW- ESE-trending narrow ridges, which are collectively referred to in this paper as the Leszno-Wolsztyn High (Fig. 3). The presence of the basement ridge or high is supported by the morphology of the base of the Carboniferous succession, which is largely a product of Late Carboniferous inversion tectonics (Fig. 2).

Marcinki IG-1

The Marcinki IG-1 well penetrated almost 2500 m of the uniform clastic succession of the Variscan foreland basin (Fig. 4). The predominant facies components are weighted towards so-called ‘deep- water’ siliciclastic successions (Pickering et al. 1986). The main part of the Carboniferous sediments intersected in cored intervals represents deposits of various sedimentary gravity flows, from cohesive debris flows to diluted turbidity currents. In this diverse group, deposits of variable-concentration turbidity currents are the most common lithofacies varieties. Associated, but less abundant, fine-grained hemipelagites are sporadically punctuated by thin intervening turbidites an\d/or sandy layers. The lower part of the drilled profile experienced apparently weak tectonic deformation indicated by shallow dips of strata (<30) and no evidence for thrusts. In contrast, the upper part of the Carboniferous succession was subjected to relatively intense deformation concentrated in a limited depth interval c. 150 m thick (Fig. 4). The tectonically affected part of the Carboniferous sequence is characterized by steeply inclined bedding and inverted fold limbs. Palynological data (Grecka-Nowak 2004) reveal extensive repetition of stratigraphie intervals and point to an important role of thrust tectonics that involved the Carboniferous succession, ranging from the Visan to the Westphalian B units (Fig. 4).

From the Marcinki IG-I well a medium-grained sandstone sample was taken within the depth interval of 2700.7-2706.5 m, situated far below the deformed part of the sedimentary succession (Fig. 4). The sample comes from a 6 m thick sandstone layer composed of several beds each up to a few tens of decimetres thick. It represents a massive sandstone composed of quartz, muscovite and a wide range of lithic clasts mainly derived from metamorphic schists. Locally the sandstone displays horizontal lamination emphasized by thin streaks of darker silty material. In places it contains thin (centimetre- scale) intercalations of mudstones and shales or randomly distributed mudstone intraclasts. At the microscale the sandstone has a mild deformation fabric, mainly reflected in undulose extinction of quartz as a result of in situ deformation. The occurrence of fresh plagioclase clasts indicates the generally unaltered state of the primary rock. White mica (125-1000 m) is present only as detrital grains (Fig. 5a), with a low modal percentage of about 3-4%. Although a precise depositional age for the sampled interval remains unknown, it is generally constrained to the time span between the Visan and the Namurian A by the palynological data from the top and bottom of the undeformed part of the sedimentary succession (Fig. 4).

Wrzesnia IG-1

The Wrzesnia IG-1 well penetrated the c. 1000 m thick Carboniferous clastic succession below a depth of 4880 m (Fig. 4). The studied drill-core reveals no systematic differences in sedimentological characteristics when compared with that from the Marcinki borehole. The entire drilled sequence underwent only weak deformation, as indicated by shallowly dipping strata (<30) and a lack of duplicated stratigraphie intervals, mesoscale folds or cleavages (Fig. 4).

From the Wrzesnia IG-1 well a medium-grained sandstone was sampled from the depth interval of 5131-5135 m, situated in the upper part of the drilled succession (Fig. 4). The sample comes from the 4 m thick layer of a massive sandstone similar to that described from the Marcinki IG-1 well. The sandstone exhibits very mild deformation features (undulose extinction in quartz) indicative of temperatures not exceeding about 200 C. The sandstone contains a high percentage of detrital white mica (about 10%; 125-1000 m in diameter), with minor, almost imperceptible growth of neocrystalline mica. This sample is characterized by good preservation of white mica and plagioclase grains (Fig. 5b) and strong retrogression of biotite, mainly by alteration into clay minerals. Although the precise depositional age of the sandstone remains unknown, it belongs to a depth interval (Fig. 4) with an age that is palynologically constrained to early Namurian A (Grecka-Nowak 2004).

Swieciechowa 1

The Swieciechowa 1 well was drilled into the crystalline basement of the foreland basin at a depth of 2652.5 m (2552.5 m below sea level). The overlying Carboniferous succession is much thinner than in the other wells (c. 430 m: Fig. 5) probably as a result of deep erosion of the Leszno-Wolsztyn High in the Early Permian. This is confirmed by the occurrence of the major erosional unconformity at the base (Rotliegend) and by the distribution of Lower Permian sedimentary facies (Karnkowski 1999). The cored section of the crystalline basement is almost 130 m thick (2652.5-2776.8 m) and comprises a fairly monotonous sequence of yellowish grey quartz- sericite phyllites (Kiapcinski et al. 1975). These rocks show a well- developed metamorphic foliation, S^sub 1^, of mostly steep inclination that parallels the bedding. Foliation is involved in tight asymmetric folds, the axial cleavage of which shows cross- cutting relationships with the S^sub 1^ planes. The macroscopically detectable cleavage S^sub 2^ has a shallow inclination (20-30) and is best manifested within the shorter limbs and hinges of the mesoscale folds. S^sub 2^ locally parallels millimetre-scale shear zones characterized by thrust-type kinematics consistent with the orientation of quartz stretching lineations developed on the S^sub 2^ planes. The deformation pattern found in the examined drill core is essentially compatible with earlier published structural data (Obere 1972; Grecka et al. 1977; Zelazniewicz et al. 2003).

A representative phyllite sample was collected from the Swieciechowa 1 well within the 2759.8-2768.5 m interval, and situated near the top of the metamorphic basement (Fig. 6). It is composed of quartz and white mica with subordinate siderite, feldspars, calcite and iron oxides. The rock has multiple fabrics that point to its deformation at a temperature of c. 350 C (e.g. Dunlap 1997). The quartz textures provide clear evidence of grain boundary migration recrystallization. Variations in quartz grain size indicate that fabrics were formed while differential stress was varying dramatically and suggest consequent variations in either temperature or strain rate (e.g. Hirth et al. 2001). Brittle processes were operating simultaneously with fabric development, as shown by carbonate veins truncating recrystallization fabrics and then acquiring their own ductile fabric. White mica plates of 65- 125 m in diameter commonly form the steep metamorphic foliation S^sub 1^, the development of which is associated with alternating quartz- and mica-rich layers (Fig. 5c). The shallowly inclined (c. 30) S^sub 2^ cleavage is locally developed in zones of S^sub 1^ transposition, and is best manifested where recrystallization and growth of white mica and quartz has occurred. In such zones there is evidence of sharp cross-cutting relationships between S^sub 1^ and S^sub 2^ planes (Fig. 5c), with consistent parallel orientation of S^sub 2^ white mica flakes (250-400 m) within cleavage planes and microscale shear zones.

Bielawy 1

The Bielawy 1 well encountered a crystalline basement at a depth of 2627.0 m (2557.0 m below sea level). The overlying Carboniferous succession is considerably reduced (only 106m: Fig. 6) as a result of the Early Permian erosion of the LesznoWolsztyn High, a situation similar to that described for the Swieciechowa 1 well. The penetrated section of the crystalline basement is almost 150 m thick (2627.5-2774.6 m), comprising a rather monotonous sequence of greenish grey quartz-sericite phyllites (Klapcinski et al. 1975). These rocks have a well-developed metamorphic foliation S^sub 1^ of highly variable inclination (10-60), which apparently parallels relics of sedimentary bedding. Foliation is involved in a number of asymmetric mesoscale folds.

From the Bielawy 1 well a representative very fine-grained phyllite sample was collected within the depth interval 2715.3- 2719.4 m (Fig. 5). It has a very strong planar deformation fabric and is composed of quartz and white mica with subordinate pyrite, feldspars and carbonates. Very fine-grained white mica flakes (90- 125 m) commonly form a metamorphic foliation, which is associated with alternating quartz- and mica-rich laminae (Fig. 5d). The layering is probably inherited from the sedimentary protolith, as the quartz laminae often contain what we interpret to be relics of graded bedding. Some white mica fish, up to 180 m across, represent possible relict detrital grains that may have been preserved from the protolith (Fig. 5d). Most of the 90-125 m diameter white mica appears, however, to have neocrystallized during the deformation. Feldspars are largely retrogressed, although some still maintain the primary composition. Quartz still exhibits clastic habits and provides evidence of pressure solution perpendicular to main foliation. In carbonaterich zones the neocrystallized white mica that forms the foliation has been torn apart by granular flow apparent in the carbonate. The characteristics of the deformation fabric generally point to the deformation of the phyllite under very low-temperature conditions (c. 200-250 C; Dunlap 1997; Hirth et al. 2001).

^sup 40^Ar/^sup 39^Ar analytical procedures

For a detailed description of the ^sup 40^Ar/^sup 39^Ar technique the reader is referred to McDougall & Harrison (1999). Ages of five samples of white mica were determined by this method. White micas were separated using standard crushing, sieving, heavy liquid and magnetic separation methods. All materials were checked for purity using oil immersion microscopy and handpicked if possible. Two of the samples (Bielawy and Swieciechowa) were purified by centrifugation in heavy liquids.

Small amounts of these separates were weighed and wrapped in aluminium foil. Samples were then sealed in an outer aluminium canister. The inner packaging components consisted of a pure silica glass tube with a cadmium liner (0.2 mm thick) between the silica tube and the outer canister. The fluence monitor GA1550 biotite (K/ Ar age of 98.79 0.9 Ma) was also packed in the canister at regular intervals. (The J factor was subsequently determined by interpolation, with flux monitors spaced about every 4 mm in a linear stack with 2 mm diameter. The J estimates and associated errors are given in Table 1.) The canister was then irradiated for 24 days in the HIFAR reactor at Lucas Heights, New South Wales. The canister was \inverted three times during the irradiation, to reduce the neutron fluence gradient along the container.

After irradiation and a cooling-off period samples for step- heating analysis were repacked in Sn foil and loaded onto an extraction line connected to a VG 3600 gas source mass spectrometer and were heated in a resistance furnace. Each sample was subjected to c. 14 heating steps, for a duration of 12min per step. The number of steps and size of temperature increments are given in Table 1. Standards were analysed by single-grain laser fusion, as were a number of grains from the Marcinki and Wrzesnia detritus (the phyllite samples were too fine-grained for single-grain laser fusion). Data were reduced using the Macintosh program ‘Noble’, developed at the Research School of Earth Sciences, Canberra. Correction factors used to correct for K- and Ca-derived Ar isotopes are (^sup 36^Ar/^sup 37^Ar)^sub Ca^ = 3.5 10^sup -4^, (^sup 39^Ar/ ^sup 37^Ar)^sub Ca^ = 7.86 10^sup -4^, and (^sup 40^Ar/^sup 39^Ar)^sub K^ = 2.72 10^sup -2^. Data were corrected for system background, blank, and mass discrimination, as well as for decay of isotopes such as ^sup 37^Ar and ^sup 39^Ar during and after irradiation. All uncertainties are quoted at 1σ level.

Results

A 250-500 m size fraction of white mica has been extracted from the Marcinki sandstone sample for ^sup 40^Ar/^sup 39^Ar analysis. Detrital white mica in this fraction has a chemical composition characterized by a low celadonite and high paragonite component (Fig. 7). Thus, it appears to be derived from a low-grade metamorphic source area. The white mica sample of Marcinki has yielded a preferred age of 370.7 1.4 Ma (1σ error, including error in J, Fig. 8) from a standard step-heating analysis. Seventeen laser fusions of single grains of the Marcinki muscovite were performed (Fig. 9), the results of which gave a peak probability of c. 370 Ma and a subsidiary peak at c. 386 Ma. The age of 370 Ma is considered as geologically significant and interpreted to approximate the timing of cooling through the appropriate closure temperature range for white mica (c. 350-400 C; Hames & Bowring 1994; Dunlap 1997) within the source area of the detrital grains. A uniform low Cl/K ratio suggests no contamination by fluids that might contain excess argon.

Two size fractions of white mica, 250-500 and 500-1000 urn, have been extracted from the Wrzesnia sandstone for electron probe analysis of chemical composition and ^sup 40^Ar/^sup 39^Ar analysis of age. Detrital white micas of both fractions are characterized by comparable chemical compositions enriched in paragonite and depleted in the celadonite component (Fig. 7). Only one mica plate from the coarser fraction (represented by two points in Fig. 7; core and rim) departs from this trend. Thus the vast majority of material in the analysed concentrates seems to be derived from a low-grade metamorphic source area (Fig. 7). The 500-1000 m white mica sample from Wrzesnia has yielded a preferred age of 363.0 1.9 Ma by step- heating analysis (1σ error, Fig. 8). The 250-500 m white mica sample from Wrzesnia has yielded a preferred age of 355.0 1.3 Ma (to error, Fig. 8) by step heating. The age spectrum for this sample suggests that the material is inhomogeneous, and the bulk age is a few per cent younger than that of the coarser white mica. Single- crystal laser fusion analyses on the coarser of the two Wrzesnia fractions (n = 20) confirms that the detrital mica population is inhomogeneous, and that a major component has come from rocks with a likely cooling age of c. 335 Ma (Fig. 9). A uniform low Cl/K ratio suggests no contamination by fluids possibly containing excess argon.

A size fraction of 60-90 m has been extracted from the Swieciechowa phyllite sample for ^sup 40^Ar/^sup 39^Ar analysis. The aim was to concentrate the fine mica grains that form the main metamorphic foliation S^sub 1^. Consequently, the size range of the analysed grains generally excluded larger mica plates that had grown parallel to the oblique superimposed cleavage 82, although some incorporation of larger grains crushed during separation cannot be excluded. Despite their microstructural distinction, mica plates forming foliations S^sub 1^ and S^sub 2 ^display a very similar chemical composition. Both mica generations reveal a low celadonite and paragonite component (Fig. 7) and very uniform composition, indicating similar metamorphic conditions during both deformation events. The white mica sample from Swieciechowa has yielded a preferred age of 358.6 1.8 Ma (1σ error, Fig. 8) from step heating. The preferred age is considered geologically significant and we interpret it to date cooling of the phyllites forming the floor of the sedimentary basin through the appropriate closure temperature (350-400 C). An alternative interpretation is that the age reflects crystallization of the white mica; however, the microstructure of the rock is consistent with temperatures too high for a crystallization age to be retained. A uniform low Cl/K ratio suggests no contamination by fluids possibly containing excess argon.

White micas <45 m have been extracted from the Bielawy phyllite sample for ^sup 40^Ar/^sup 39^Ar dating. White mica of this fraction is characterized by a highly differentiated composition (Fig. 7). Some analyses show a low celadonite component compatible with a very low-grade metamorphic overprint recorded by the phyllites (Fig. 7). Nevertheless, the majority of grains, analysed directly in a thin section, are rich in a celadonite component and have compositions suggesting that they grew under high-grade metamorphic conditions. Consequently, they can be interpreted as detrital grains preserved from a sedimentary protolith of the phyllite. It should be emphasized, however, that the concentrate of white mica used for dating (at <45 m) is likely to contain both fragments of the detrital ‘fish’ and also a large amount of neocrystallized grains. The white mica concentrate from Bielawy displays an incremental discordant apparent age spectrum with a total gas age of 400.1 3.5 Ma (Fig. 8). The discordant nature of the spectrum is evidence that a mixture of white mica grains with different ages has been dated. Unfortunately, the degassing of the grains is simultaneous, so that we cannot resolve the ages of the individual components of the mixture. However, as ages as young as 300 Ma and as old as almost 500 Ma have been produced, this suggests that some of the detrital grains could be 500 Ma or older and some of the neocrystallized grains could be younger than 300 Ma. Only further work on such samples can resolve these issues. The Cl content of the sample, relative to K, is the highest of all the samples analysed in this study. This is clearly the result of abundant fluid inclusions within the micaceous concentrate, a situation difficult to avoid with such finegrained material.

Discussion

The ^sup 40^Ar/^sup 39^Ar results for detrital white micas obtained from Carboniferous sedimentary rocks (Marcinki and Wrzesnia samples) indicate that the Variscan foreland basin was supplied by source rocks that were exhumed and cooled in the Late Devonian or Early Carboniferous. The essential uniformity of material delivered to the basin at Marcinki is suggested by the uniform age spectrum with a limited range of single-crystal ages (Figs 8 and 9) and the similar chemical compositions of all analysed micas (Fig. 7). The Wrzesnia micas, in contrast, form an inhomogeneous population, reflected partially in the age spectra, but more dramatically in the single-crystal data (Fig. 9). The c. 335 Ma detrital micas in this rock come from a source distinct from that of the Marcinki rock, suggesting a source region with a spatially variable cooling history.

The Variscan belt appears to be the only possible source area contributing detritus to the basin, because the other surrounding regions such as the East European Craton or North GermanPomeranian Caledonides were at that time an area of subsidence and continuing sedimentation. Nevertheless, the present-day exposure of the Variscides in Central Europe shows only a minor component of rocks exhumed before the Carboniferous (e.g. Winchester et al. 2002). In the Sudetes, directly adjoining the Carboniferous basin on the SW (Fig. 1), the outcrop of preCarboniferous exhumed rocks is very narrow, being limited to the Gory Sowie Gneiss Massif and the Kiodzko Metamorphic Complex (e.g. Aleksandrowski & Mazur 2002; Mazur et al. 2004). The proof of the pre-Carboniferous exhumation is provided by general geological evidence (for an overview, see Aleksandrowski et al. 2000) rather than by thermochronological data (Marheine et al. 2002). If the Visan and Namurian outcrop of rocks exhumed in the Late Devonian was as small as the currently exposed source area in the Sudetes, it seems unlikely that we would find pre- Carboniferous cooling ages in the detritus filling the foreland basin. Therefore, it is possible that by Carboniferous times a significant portion of the rocks exposed in the Sudetes had already experienced cooling and exhumation in the Late Devonian. Although the amount of erosion in the Sudetes since the Carboniferous remains unknown (4-8 km during the last 90 Ma, Aramowicz et al. 2006), rocks containing a Devonian cooling signature are likely to have been removed by now. Consequently, it is possible that such rocks once formed a relatively thin cover of early Variscan nappes (Fig. 10), as has already been postulated in the case of the Gry Sowie Massif (Znosko 1981) and the Klodzko Complex (Mazur et al. 2004). Although sedimentation ages of the two analysed samples are not precisely established, the sandstone acquired from the Marcinki well is probably older (Visan-Namurian A) than that from the Wrzesnia well (Namurian A). This time relationship would be in accord with theinversion of cooling ages in a hypothetical nappe system, as the Marcinki white mica concentrate yields older ^sup 40^Ar/^sup 39^Ar ages.

The cooling ages obtained for the detrital micas from the Carboniferous sediments are very similar to those that characterize the Swieciechowa phyllite. Furthermore, the chemical composition of detrital micas points to mostly low-grade metamorphic rocks being a source of detritus for the Variscan foreland basin. Consequently, the phyllites underlying the basin seem to represent rocks comparable with those supplying detritus to the Carboniferous sediments. Such rocks, which were almost entirely eroded within the nearby uplifted part of the Variscan belt, may characterize the basement below the southern margin of the basin (Fig. 10). On the other hand, the cooling age (c. 340 Ma) previously acquired for the Swieciechowa phyllite by Zelazniewicz et al. (2003) is apparently attributed to a superimposed late folding event, as indicated by the compound fabric of the rock. We may be seeing evidence of the same event in the single-crystal data for Wrzeenia. This age correlates well with the final phase of Variscan collision, which followed a welldocumented thermal re-equilibration of the now exposed orogen (Marheine et al. 2002). The same tectonothermal events, widespread in the Variscan belt at the Tournaisian-Visan boundary, might be responsible for subsidence of the foreland basin and the supply of detritus from the freshly exhumed orogen. In addition to a large- scale flexural bending induced by orogenic loading, the subsidence of the Variscan foreland basin in western Poland was probably controlled by a tectonic downthrow of its base along the Odra Fault Zone (Figs 2 and 3). This major, multiply reactivated, structure (Oberc-Dziedzic et al. 1999) forms the present boundary between the Variscan internides of the Bohemian Massif and their northern foreland.

The results obtained suggest that the continental collision between the Old Red Continent and the Armorican terrane assemblage must have commenced in the Late Devonian, earlier than previously considered (e.g. Franke & Engel 1986; Franke 2000; Oncken et al. 2000). Evidence for this early Variscan tectonic event is further highlighted by K-Ar geochronology of detrital muscovites from the Harz Mountains (Huckriede et al. 2004). On the other hand, the record of the latest Devonian uplift and cooling is mostly erased from the present-day outcrops of the Variscan crystalline basement (see Marheine et al. 2002) by the subsequent thermal overprint and erosion. Despite the early commencement of the collisional event, the Variscan foreland turbidite basin was initiated only in late Visan times as a result of crustal flexure imposed as a response to the peak phase of nappe stacking within the Variscan palaeohinterland (McCann 1999). Therefore, the southern part of the basin could have onlapped the Variscan units, while still retaining a memory of the latest Devonian uplift and cooling. At the same time, the sedimentation continued without major stratigraphie gaps from the Tournaisian onwards in the northernmost parts of the developing basin (e.g. McCann 1999). Nevertheless, in contrast to the area that our samples come from, these distal parts of the Variscan foredeep are thought to be supplied from the foreland bulge situated to the north of the growing basin (McCann 1999; Jaworowski 2002).

The tectonic affinities of the crystalline basement underlying the Carboniferous basin in the northern Variscan foreland have been the focus of intense interest in recent years (e.g. Winchester et al. 2002). The basin floor was compared with the Avalonian basement of the Rhenohercynian Zone (Grad et al. 2002), but it has also been considered a small separate terrane (Zelazniewicz et al. 2003). The present results show that this controversy cannot be reliably solved unless some U/Pb dating of zircons is carried out. Although the boreholes reach the Variscan nappe complex beneath the sediments, the affinity of the basement underlying these nappes remains unknown. Early Devonian ages obtained for the Bielawy phyllite may indicate some affinity with the Caledonian thrust belt of Pomerania, which was folded after Silurian time (e.g. Dadlez et al. 1994); however, only further work will illuminate such issues.

Conclusions

The present study demonstrates the significance of isotopic age information derived from foreland sediments for deciphering the evolution of orogenic belts. In the case of the central European Variscides, models based upon information provided by hinterland outcrop emphasize the Early Carboniferous tectonothermal activity (e.g. Franke & Zelazniewicz 2000; Marhaine et al. 2002). In contrast, the results of our study suggest an important influence of Late Devonian tectonism on the evolution of the Variscan belt. The significance of early Variscan tectonic activity was stressed by Faure et al. (1997) for the French section of the Variscides (their ‘eo-Variscan’ phase). In the Sudetes and the entire Bohemian Massif, the implication of pre-Carboniferous events has usually been minimized because of the minor exposure of rock complexes exhumed at that time. Our data provide an indication that the original extent of the Late Devonian nappe complexes could be much wider than now and, thus, the Late Devonian events played an important role in the accretion of the Variscides.

This research was supported by the Polish Committee for Scientific Research and Ministry of the Environment (project PCZ- 007-21 ‘Palaeozoic Accretion of Poland’). We are indebted to the Australian Institute for Nuclear Science and Engineering, and the Australian Nuclear Science Technology Organisation. T. McCann and P. Layer are thanked for constructive and helpful reviews.

References

ALEKSANDROWSKI, P. & MAZUR, S. 2002. Collage tectonics in the northeasternmost part of the Variscan Belt: the Sudetes, Bohemian Massif. In: WINCHESTER, J.A., PHARAOH. T.C. & VERNIERS, J. (eds) Palaeozoic Amalgamation of Central Europe. Geological Society, London, Special Publications, 201, 237-277.

ALEKSANDROWSKI, P., KRYZA, R., MAZUR, S., PIN, C. & ZALASIEWICZ. J.A. 2000. The Polish Sudetes: Caledonian or Variscan? Transactions of the Royal Society, Edinburgh, 90, 127-146.

ARAMOWICZ, A., ANCZKIEWICZ, A.A. & MAZUR, S. 2006. Fission-track dating of apatite from the Gory Sowie Massif, Polish Sudetes, NE Bohemian Massif: implications for post-Variscan denudation and uplift. Nettes Jahrbuch fr Mineralogie, 183, in press.

CWOJDZINSKI, S. & GROCHOLSKI, A. 1995. Sudetes and adjacent areas. In: ZDANOWSKI, A. & ZAKOWA, H. (eds) The Carboniferous System in Poland. Prace Panstwowego Instytutu Geologicznego, 148, 19-22.

DADLEZ, R., KOWALCZEWSKI, Z. & ZNOSKO, J. 1994. Some key problems of the pre-Permian tectonics of Poland. Geological Quarterly, 38, 169-189.

DUNLAP, WJ. 1997. Neocrystallization or cooling?: ^sup 40^Ar/ ^sup 39^Ar ages of white micas from low grade mylonites. Chemical Geology (Isotope Geoscience Section), 143, 181-203.

ENGEL, W. & FRANKE, W. 1983. Flysch sedimentation: its relation to tectonism in the European Variscides. In: MARTIN, H. & EDER, F.W. (eds) Intracontinental Fold Belts. Springer, Berlin, 289-321.

FAURE, M., LELOIX, CH. & ROIG, J.-Y. 1997. L’volution polycyclique de la chane hercynienne. Bulletin de la Societ Gologique de France, 168, 695-705.

FOSSEN, H. & DUNLAP, W.J. 1998. Timing and kinematics of Caledonian thrusting and extensional collapse, southern Norway: evidence from ^sup 40^Ar/^sup 39^Ar thermochronology. Journal of Structural Geology, 20, 765-781.

FRANKE, D. 1995. The north Variscan foreland. In: DALLMEYER, R.D., FRANKE, W. & WEBER, K. (eds) Pre-Permian Geology of Central and Eastern Europe. Springer, Berlin, 554-566.

FRANKE, W. 2000. The mid-European segment of the Variscides: tectonostratigraphic units, terrane boundaries and plate tectonic evolution. In: FRANKE, W., HAAK, V., ONCKEN, O. & TANNER, D. (eds) Orogenic Processes: Quantification and Modelling in the Variscan Belt. Geological Society, London, Special Publications, 179, 35-61.

FRANKE, W. & ENGEL. W. 1986. Synorogenic sedimentation in the Variscan Belt of Europe. Bulletin de la Socit Gologique de France, 8(II), 25-33.

FRANKE, W. & ZELAZNIEWICZ, A. 2000. The eastern termination of the Variscides: terrane correlation and kinematic evolution. In: FRANKE, W., HAAK, V., ONCKEN. O. & TANNER, D. (eds) Quantification and Modelling in the Variscan Belt. Geological Society, London, Special Publications, 179, 63-86.

FRANKE, W., DALLMEYER, R.D. & WEBER, K. 1995. Geodynamic evolution. In: DALLMEYER, R.D., FRANKE. W. & WEBER, K. (eds) Pre- Permian Geology of Central and Eastern Europe. Springer, Berlin, 579- 593.

GRECKA. T., JUROSZEK, C. & KARWOWSKI. L. HT AL. 1977. [The sub- Permian basement of the western Fore-Sudetic Homocline, Zary Pericline and the adjacent part of the Fore-Sudetic Block]. Prace Naukowe Instytutu Grnictwa Politechniki Wroclawskiej, 22, 3-92 [in Polish].

GRECKA-NOWAK, A. 2004. [Redeposition as an important factor affecting the results of biostratigraphic investigations-a case study from the Carboniferous sediments in the basement of the Fore- Sudetic Homocline]. In: MUSZER, J. (ed.) XIX Conference on Palaeobiology and Biostratigraphy, Polish Geological Society, Abstract Volume. University of Wroclaw 21-23 [in Polish].

GRAD, M., GUTERCH. A. & MAZUR, S. 2002. Seismic refraction evidence for crustal structure in the central part of the Trans- European Suture Zone in Poland. In: WINCHESTER, J.A., PHARAOH, T.C. & VERNIERS, J. (eds) Palaeozoic Amalgamation of Central Europe. Geological Society, London, Special Publications, 201, 295-309.

GRADSTEIN, P.M. & OGG, J. 1996. A Phanerozoic time scale. Episodes, 19, 3-5.

GROCHOLSKI. W. 1975. Variscides of southern Wielkopolska. Przeglad Geologiczny, 4, 171-174 [in Polish, English summary].

HAMES, W. & BOWRING, S. 1994\. An empirical evaluation of the argon diffusion geometry in muscovite. Earth and Planetary Science Letters, 124, 161-167.

HAYDUKIEWICZ, J., MUSZER, J. & KLAPCINSKI, J. 1999. [Palaeontological documentation of the sub-Permian sediments in the vicinity of Zbaszyn (Fore-Sudetic Monocline).] In: MUSZER. A. (ed.) Selected Problems of Stratigraphy. Tectonics and Ore Mineralization in Lomr Silesia. University of Wroclaw, 7-17 [in Polish].

HIRTH, G., TEYSSIER, C. & DUNLAP, WJ. 2001. An evaluation of quartzite flow laws based on comparisons between experimentally and naturally deformed rocks. International Journal of Earth Sciences, 90, 77-87.

HUCKRIEDE, H., AHRENDT, H., FRANKE, W., WEMMER, K. & MEISCHNER, D. 1998. Orogenic processes recorded in Early Carboniferous and Devonian clastic sediments of the Rhenohercynian Zone. Terra Nostra, 98(2), 77-79.

HUCKRIEDE, H., WEMMER, K. & AHRENDT, H. 2004. Palaeogeography and tectonic structure of allochthonous units in the German part of the Rheno-Hercynian Belt (Central European Variscides). International Journal of Earth Sciences, 93,414-431.

JAWOROWSKI, K. 2002. Geotectonic significance of Carboniferous deposits NW of the Holy Cross Mts. (central Poland). Geological Quarterly, 46(3), 267-280.

KARNKOWSKI, P.H. 1999. Origin and evolution of the Polish Rotligend Basin. Polish Geological Institute Special Papers, 3, 1- 93.

KLAPCINSKI. J., JUROSZEK, C. & SACHANBINSKI, M. 1975. [New data on geology of the crystalline basement in the Fore-Sudetic area.] Geologia Sudelica, 10(2), 7-49 [In Polish].

MARHEINE. D., KACHLIK. V., MALUSKI. H., PATOCKA. F. & ZELAZNIEWICZ, A. 2002. The Ar-Ar ages from the West Sudetes (Bohemian Massif): constraints on the Variscan polyphase tectonothermal development. In: WINCHESTER, J.A., PHARAOH, T.C. & VERNIERS, J. (eds) Palaeozoic Amalgamation of Central Europe. Geological Society, London, Special Publications, 201, 133-155.

MAZUR, S., KUROWSKI. L., ALEKSANDROWSKI, P. & ZELAZNIEWICZ, A. 2003. Variscan foreland fold-thrust belt of Wielkopolska (W Poland): new structural and sedimentological data. Geolines, 16, 71.

MAZUR, S., TURNIAK, K. & BROCKER, M. 2004. Timing of magmatism in the Ktodzko Metamorphic Complex (West Sudetes, Poland)-new insights from conventional U/Pb zircon dating. International Journal of Earth Sciences, 93, 758-772.

McCANN, T. 1999. The tectonosedimentary evolution of the northern margin of the Carboniferous foreland basin of NE Germany. Tectonophysics, 313, 119-144.

McDOUGALL, I. & HARRISON, T.M. 1999. Geochronology and Thermochronology by the ^sup 40^Ar/^sup 39^Ar Method. 2nd Edition. Oxford University Press, Oxford.

OBERC, J. 1972. [Geology of Poland IV. Tectonics 2. Sudetes and Adjacent Areas.] Wydawnictwa Geologiczne, Warsaw [in Polish].

OBERC-DZIEDZIC, T., ZELAZNIEWICZ, A. & CWOJDZINSKI, S. 1999. Granitoids of the Odra Fault Zone: late- to post-orogenic Variscan intrusions in the Saxothuringian Zone, SW Poland. Geologia Sudetica, 32, 55-71.

ONCKEN, O, 1997. Transformation of a magmatic arc and orogenic root during the oblique collision and its consequences for the evolution of the European Variscides (Mid-German Crystalline Rise). Geologische Rundschau, 86, 2-21.

ONCKEN, O. 1998. Tectonic mass transfer prior to collision- subduction erosion versus accretion in the Mid-European Variscides. Ada Universitatis Carolinae-Geologica, 42, 313-314.

ONCKEN, O., PLESCH. A., WEBER, J., RICKEN, W. & SCHRADER, S. 2000. Passive margin detachment during arc-continent collision (central European Variscides). In: FRANKE. W., HAAK, V., ONCKEN, O. & TANNER, D. (eds) Quantification and Modelling in the Variscan Belt. Geological Society, London, Special Publications, 179, 199- 216.

PICKERING, K.T., STOW, D.A.V., WATSON, M.P. & HISCOTT, R.N. 1986. Deep water facies, processes and models: a review and classification scheme for modem and ancient sediments. Earth-Science Review’s, 23, 75-174.

SCHUBERT. W., LIPPOLT, H.J. & SCHWARZ, W. 2001. Early to Middle Carboniferous ^sup 40^Ar/^sup 39^Ar ages of amphibolites and gabbros from the Bergstrsser Odenwald. Mineralogy and Petrology, 72, 113- 132.

TAIT, J., SCHTZ, M., BACHTADSE, V. & SOFFEL, H. 2000. Palaeomagnetism and Palaeozoic palaeogeography of Gondwana and European terranes. In: FRANKE, W., HAAK, V., ONCKEN, O. & TANNER, D. (eds) Orogenic Processes: Quantification and Modelling in the Variscan Belt. Geological Society, London, Special Publications, 179, 21-34.

WHITE, N.M., PRINGLE. M., GARZANTI, E., BICKLE, M., NAJMAN, Y., CHAPMAN, H. & FRIEND. P. 2002. Constraints on the exhumation and erosion of the High Himalayan Slab, NW India, from foreland basin deposits. Earth and Planetary Science Letters, 195, 29-44.

WIERZCHOWSKA-KICULOWA, K. 1984. Geology of the Pre-Permian series of the Fore-Sudetic Monocline. Geologia Sudetica, 19(1), 120-142 (in Polish, English summary).

WINCHESTER, J.A. & PACE TMR NETWORK TEAM, 2002. Palaeozoic amalgamation of Central Europe: new results from recent geological and geophysical investigations. Tectonophysics, 360, 5-21.

ZEH, A., COSCA, M.A., BRATZ, H., OKRUSH, M. & TICHOMIROWA, M. 2000. Simultaneous horst-basin formation and magmatism during Late Variscan transtension: evidence from ^sup 40^Ar/^sup 39^Ar and ^sup 207^Pb/^sup 206^Pb geochronology in the Ruhla Crystalline Complex. International Journal of Earth Sciences, 89, 52-71.

ZELAZNIEWICZ, A., MARHEINE, D. & OBERC-DZIEDZIC, T. 2003. A Late Tournaisian synmetamorphic folding and thrusting event in the eastern Variscan foreland: ^sup 40^Ar/^sup 39^Ar evidence from the phyllites of the WolsztynLeszno High, western Poland. International Journal of Earth Sciences, 92, 185-194.

ZELICHOWSKI, A.M. 1995. Wielkopolska region. In: ZDANOWSKI, A. & ZAKOWA, H. (eds) The Carboniferous System in Poland. Prace Panstwowego Instytutu Geologicznego, 148, 148-151.

ZIEGLER, P. 1989. Geodynamic model for Alpine intra-plate compressional deformation in Western and Central Europe. In: COOPER, M.A. & WILLIAMS, G.D. (eds) Inversion Tectonics. Geological Society, London, Special Publications, 44, 63-85.

ZIEGLER. P.A. 1990. Geological Atlas of Western and Central Europe, 2nd edn. Shell Internationale Petroleum, The Hague; Geological Society of London, London.

ZNOSKO, J. 1981. The problem of oceanic crust and of ophiolites in the Sudetes. Bulletin de l’Acadmie Polonaise des Sciences. Srie Sciences de la Terre, 29, 185-197.

Received 7 January 2005; revised typescript accepted 4 March 2006.

Scientific editing by Mike Villeneuve

STANISLAW MAZUR1, W. JAMES DUNLAP2, KRZYSZTOF TURNIAK1 & TERESA OBERC-DZIEDZIC1

1 Institute of Geological Sciences, University of Wroclaw, Pl. Maksa Borna 9, 50-204 Wroclaw, Poland

(e-mail: smazur@ing.uni.wroc.pl)

2 The Australian National University, Research School of Earth Sciences, Mills Road, Canberra, ACT 0200 Australia

Copyright Geological Society Publishing House Nov 2006

(c) 2006 Journal of the Geological Society. Provided by ProQuest Information and Learning. All rights Reserved.