| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 42 Number 11 Pages 2015-2032 2001
© Oxford University Press 2001
P–T–t Evolution of Ultrahigh-Temperature Granulites from the Saxon Granulite Massif, Germany. Part II: Geochronology
1GEOFORSCHUNGSZENTRUM POTSDAM, TELEGRAFENBERG, D-14473 POTSDAM, GERMANY
2INSTITUT FÜR GEOWISSENSCHAFTEN, UNIVERSITÄT POTSDAM, POSTFACH 60 15 53, 14415 POTSDAM, GERMANY
Received January 28, 2000; Revised typescript accepted April 16, 2001
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Granulite-facies metamorphism in the structurally lower part of the Saxon Granulite Massif culminated at 1010–1060°C and
22 kbar. Ultrahigh-temperature conditions persisted even after a significant drop in pressure. We dated monazite, zircon, rutile, garnet and apatite from felsic granulite by U–Pb, biotite from retrogressed felsic granulite by Rb–Sr, and titanite from mafic granulite by U–Pb. Zircon and titanite give within analytical uncertainties the same age at 341·5 ± 0·8 Ma (2
) and 342·0 ± 0·8 Ma (2), respectively, demonstrating (1) a similar closure temperature for both minerals in dry systems and (2) a closure temperature for titanite considerably higher than 550°C. Monazite plots discordantly and yields a 207Pb/206Pb age at 338·0 ± 0·5 Ma (2), which represents a minimum age because of the possibility of excess 206Pb. Rutile, garnet and apatite have little radiogenic lead and show a wide range of apparent 206Pb/238U ages, which reflects initial isotopic heterogeneities originating from the reaction history rather than later disturbances. Biotite yields an Rb–Sr age at 323·0 ± 2·3 Ma (2). The age data in combination with the P–T path demonstrate that exhumation of the Saxon Granulite Massif to a middle- to upper-crustal level proceeded at a fast average rate (>9–18 mm/yr) and subsequently slowed down significantly (<2 mm/yr). KEY WORDS: exhumation; geochronology; Saxony; UHT granulites
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Ionic diffusion in crystals depends on a variety of factors such as distribution of crystal defects and impurities, dissolution and recrystallization reactions, activity of the involved chemical elements, and temperature and pressure (e.g. Lasaga, 1981
; Chakraborty, 1997). The rate of ionic diffusion in a crystal structure increases significantly with rise in temperature (e.g. Lasaga, 1981; Chakraborty & Ganguly, 1990; Chakraborty, 1997; O’Brien, 1997). Fast diffusion implies that a mineral easily re-equilibrates with its surroundings and gets rid of ions that do not fit well into its crystal structure. At high temperature, the increased overall diffusivity of cations may, to an even larger extent, apply to radiogenic daughter isotopes, as these ions have been displaced from their original structural site by recoil and now occur in structural sites that have a damaged surrounding (e.g. Nasdala et al., 2001) and these ions may have a different ion radius and charge than their parents (e.g. Hofmann & Giletti, 1970; Watson et al., 1985; Cherniak et al., 1991; Smith & Giletti, 1997). Therefore, the daughter isotope tends to be lost from the crystal structure. For a geochronological system, this implies that the mineral is resetting its age continuously. This concept is well known for minerals that reset their radiogenic clocks at rather low temperatures, as for instance the Rb–Sr and K–Ar systems of micas (e.g. Dodson, 1973; Jäger, 1973; Dahl, 1997). For rocks that experienced very high metamorphic temperatures, the problem arises of whether any mineral yields its crystallization age or they all provide cooling ages only.
Problems related to the isotopic dating of ultrahigh-temperature (UHT) metamorphism are illustrated on a sample suite from the Saxon Granulite Massif (SGM), which is part of the Bohemian Massif. The samples have experienced metamorphic temperatures that are much higher than the suggested temperatures of isotopic closure of most minerals. Thus, UHT metamorphism should have wiped out the chronological memory of all phases. We dated minerals, which supposedly encompass a wide range of temperature of isotopic closure and should yield a range of cooling ages, from three samples that experienced the same metamorphic history (Rötzler & Romer, 2001) to constrain both the age of the metamorphic peak and the exhumation history of the Saxon Granulite Massif. Combined with our P–T data (Rötzler & Romer, 2001), the isotopic ages demonstrate that exhumation of the Saxon Granulite Massif to a middle- to upper-crustal level proceeded at a very high rate, whereas subsequent exhumation to the surface occurred at a much lower rate. Our data demonstrate, furthermore, that literature values for temperature ranges of isotopic closure for titanite, and possibly monazite, are in conflict with our observations, and that reaction histories in minerals with low parent/daughter ratio may have a significant influence on the geochronological system of these minerals.
|
GEOLOGICAL SETTING |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
The Saxon Granulite Complex and overlying structural units of lower metamorphic grade constitute the Saxon Granulite Massif, an oval-shaped dome structure within the Variscan basement at the northwestern margin of the Bohemian Massif (Fig. 1). The internal structure of the SGM shows an excision of large parts of the metamorphic zonation between the Granulite Complex and the overlying units, but also between the overlying units (e.g. Rötzler, 1992
, and therein). The granulite-facies metamorphism has culminated at ultrahigh-temperature high-pressure conditions (Rötzler et al., 1994; Hagen et al., 1995; Rötzler & Romer, 2001). Geochemical and isotopic data demonstrate that the Granulite Complex, as the core of the dome structure, is almost exclusively composed of meta-igneous rocks (e.g. Rötzler, 1992, and therein; Kroner, 1995). The ages of the protoliths and the high-grade metamorphism are poorly constrained. Zircons from felsic granulites, analysed by SHRIMP ion microprobe, evaporation, and vapour digestion and conventional dissolution techniques, show (1) inherited Precambrian components, (2) cores with ages around 480 Ma that are interpreted as protolith ages, and (3) abundant metamorphic zircon with ages around 340 Ma that are thought to reflect near-peak granulite-facies conditions (von Quadt, 1993; Nasdala et al., 1996; Kröner et al., 1998). In addition, there are zircon data yielding ages around 405 Ma (conventional and evaporation) and 465–460 Ma, 380–365 Ma and 320–315 Ma (all SHRIMP ages), as well as monazite U–Pb ages at 315 Ma (see Nasdala et al., 1996; Baumann et al., 1997; Reinhardt & Vavra, 1997; Kröner et al., 1998; Vavra et al., 1998). Whole-rock and mineral isochron Sm–Nd ages from granulites and ultramafic rocks at 470 Ma, 380 Ma and 340 Ma were interpreted as the age of the protolith and two high-pressure metamorphic events, respectively, leaving open whether the latter reflect stages on a single P–T loop or polymetamorphism (von Quadt, 1993). Variably deformed monzogranites, which post-date the metamorphic peak, intruded the SGM at 333 Ma (zircon U–Pb SHRIMP, Nasdala et al., 1996; Kröner et al., 1997).
|
|
METAMORPHIC PETROLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Sample description
Our study is based on rock samples taken from two outcrops within
1 km distance in Waldheim. Two samples (G97-1 and G97-2) are from a railway cut at Pfaffenberg hill, whereas the third (G97-3) was collected NE of Waldheim station (Fig. 1, Tables 1 and 3). Sample G97-1 represents a new occurrence of sapphirine (Rötzler & Romer, 2001) different in mineralogy from already known localities for this mineral in the SGM (see Grew, 1986, 1989).
|
|
Sapphirine-bearing granulite (G97-1)
The fine-grained matrix of this rock shows a granoblastic lobate texture of mesoperthite, plagioclase and quartz, with accessory apatite, graphite, zircon, monazite, rutile, pyrrhotite and pyrite. Zircon and monazite are part of the peak assemblage, whereas apatite occurs in two generations. Elongate porphyroblasts and ribbons of quartz define a distinct foliation. Garnet is megacrystic, has inclusions of quartz, plagioclase, perthite, biotite, kyanite, pyrrhotite, rutile and zircon, and partially shows higher Ca contents in the core than at the rims (Rötzler & Romer, 2001). Kyanite is usually megacrystic and is either overgrown by sapphirine + spinel + plagioclase or pseudomorphed by sillimanite. Garnet in part shows replacement by sapphirine + spinel + plagioclase. Sporadically, sapphirine + spinel are cut or overgrown by minor biotite that is also found at rims and within fractures of garnet [for additional information, see Rötzler & Romer (2001)].
Retrogressed felsic granulite (G97-2)
This rock is a retrograde product of the typical felsic granulite. It has a granoblastic inequigranular texture and experienced the same metamorphic evolution as the sapphirine-bearing granulites, except for a distinct retrogression that led to extensive formation of biotite. Red–brown or greenish biotite replaces garnet, which locally occurs as relict crystals. Biotite is also invariably overgrown on sillimanite and commonly mantled by plagioclase, suggesting that biotite was formed by the reaction Grt + Sil + Kfsss + H2O 
Bt + Pl + Qtz. Late alterations include the overgrowth of haematite and muscovite on feldspar and of chlorite on biotite.
Garnet–clinopyroxene rock (G97-3)
This fine- to medium-grained rock contains the restricted peak assemblage garnet + clinopyroxene + titanite + apatite that forms a granoblastic polygonal texture. Apart from homogeneous dark green areas of clinopyroxene with scattered garnet, this rock contains millimetre-thick bands with concentrations of garnet. Titanite and apatite commonly occur enclosed in both clinopyroxene and garnet. Early plagioclase occurs in interstices between clinopyroxene, from which it may have exsolved. It also occurs as blebs within clinopyroxene. Late hydration must have involved the introduction of potassium by fluids, as plagioclase is largely overgrown by muscovite.
P–T path
The P–T path for the Saxon granulites (Fig. 2) has been presented in detail by Rötzler & Romer (2001). It is constrained by phase equilibria and phase compositions of the samples here dated, by recalculated estimates from Rötzler et al. (1994) and by data from Rötzler (1992). These P–T estimates demonstrate that metamorphism in the Saxon Granulite Complex culminated in UHT–HP conditions of 1010–1060°C and 22 kbar. The peak mineral assemblages and isothermal-decompression assemblages related to the exhumation of the Saxon granulites define a clockwise loop, where UHT conditions persisted even after a significant pressure drop (Fig. 2). Retrograde hydrous phases are characteristic for the P–T loop once the rocks have reached intermediate to low pressure. The granulites seem to have experienced pronounced cooling at little pressure change only after they had reached a middle- to upper-crustal level (Fig. 2).
|
|
GEOCHRONOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Sapphirine-bearing granulite (G97-1)
Zircon crystals in this rock are multifaceted short-prismatic to equant, as is typical for zircon from high-grade metamorphic rocks (e.g. Vavra et al., 1994
). There is no indication of inherited grains or partially resorbed older zircon. Six zircon samples are concordant (Fig. 3; Table 1). The best-constrained age estimate of these six samples is the weighted 206Pb/238U age at 341·5 ± 0·8 Ma (2; Fig. 3). This age is a minimum age for the crystallization of zircon. It represents either a cooling age or a crystallization age, depending on whether the temperature for isotopic closure of the U–Pb system of zircon is lower or higher than the temperature at which zircon crystallized.
|
The U–Pb system of monazite is disturbed and all six samples are discordant (Fig. 3; Table 1). The discordia intersects the concordia at 338·2 ± 1·4 Ma (2; MSWD = 0·382). The weighted 207Pb/206Pb age of all six samples is 338·0 ± 0·5 Ma (2; Fig. 3, inset). This is a minimum age for the crystallization of monazite, as (1) the monazite might have crystallized at a temperature considerably higher than its commonly assumed closure temperature for lead diffusion (710–750°C; see Copeland et al., 1988; Parrish, 1990; Dahl, 1997) and (2) there might have been excess 206Pb. Excess 206Pb originates from 230Th (e.g. Mattinson, 1973; Schärer, 1984), which is an intermediate member of the 238U decay series and eventually decays to 206Pb. Excess 206Pb can be accounted for by using the 232Th/238U of monazite and the rock (see Schärer, 1984), provided the observed Th/Urock is representative for the original crystallization environment of monazite. For Th/Urock 2·0, which falls in the range characteristic for felsic rocks, the weighted 207Pb/206Pb age of all six samples becomes 341·1 ± 1·1 Ma (2; Fig. 3, inset), which is within analytical uncertainties identical with the age obtained for zircon. It should be noted that the scatter of the data is increased by this correction procedure, which might hint that a Th/Urock of 2·0 already results in an overcorrection. Lower Th/Urock requires a larger correction, and thus results in even larger scatter. Thus, the best age estimate for the monazite is bracketed by the minimum age at 338·0 ± 0·5 Ma and the age at 341·1 ± 1·1 Ma (2). Actually, as any correction for excess 206Pb results in enhanced scatter of the data about the discordia line, it is likely that excess lead does not contribute significantly to the lead of these monazite samples, in which case Th/Urock was larger than two. For Th/Urock > 3·5, the U–Pb age of monazite is beyond analytical uncertainties lower than the U–Pb zircon age. Th/U of the whole-rock sample was not determined, as this ratio may be strongly affected by U mobility at the surface and—even for undisturbed Th/U values—it never is clear whether the value for the bulk sample is representative for the surrounding of monazite. Therefore, the extent to which 206Pb excess affects the intercept age of monazite remains unresolved. In the discussion, we use in a first step the monazite minimum age (no excess 206Pb) to illustrate the exhumation and cooling history of the Saxon granulites. In a second step, we focus on the effect of an older monazite age for the exhumation and cooling histories and for the temperature of isotopic closure of monazite in dry systems.
Garnet has a low and highly variable U content, but rather high common lead contents (0·3–18 ppm; Table 1). Therefore, its lead isotopic composition is rather unradiogenic. The U–Pb system of garnet yields concordant data, largely because of the large uncertainty of the 207Pb/235U ratio, which originates from the uncertainty of the estimated initial lead composition. The apparent 206Pb/238U age, however, differs between the various garnet samples and ranges between 337 and 460 Ma (Table 1), which implies that the U–Pb system of garnet carries inherited components or the various samples did not have a common isotopic composition of initial lead. During sample preparation, special care was taken to avoid any garnet fragments that might contain inclusions; therefore, we consider the presence of inclusions as unlikely. This is also corroborated by the following argument. The apparent U–Pb garnet ages in part are much older than the U–Pb age of monazite, zircon and titanite, despite textural evidence of a common crystallization history and age. Furthermore, electron microprobe BSE imaging did not reveal submicroscopic inclusions of inherited zircon or monazite in garnet. The 206Pb/238U age discrepancy among the various garnet fractions is also reflected in the 206Pb/204Pb–238U/204Pb and 206Pb/204Pb–207Pb/204Pb diagrams (Fig. 4). Two reference lines are shown in the 206Pb/204Pb–238U/204Pb diagram (Fig. 4a). They are fitted through the isotopic composition of the alkali feldspar and garnet samples to yield the steepest and the flattest slope, respectively. The flatter reference line corresponds to an age at 337 Ma, which by and large agrees with the zircon, titanite and monazite ages. The steeper reference line corresponds to an age at 460 Ma. If all garnet samples had formed during the same event, and thus, have the same age, the initial lead isotopic composition would have to vary among the various samples. The extreme values for the initial isotopic composition of garnet are given by the isotopic composition of alkali feldspar (206Pb/204Pb = 18·31) and 206Pb/204Pb 18·85 (i.e. the least radiogenic garnet sample; see Fig. 4b and c). In this case, all garnet data represent mixtures and the reference lines for 337 and 460 Ma probably represent mixing lines without age significance. Heterogeneous initial lead isotopic composition of garnet may also explain the anomalously high scatter of the data in the 206Pb/204Pb–207Pb/204Pb diagram. The anomalous behaviour of lead is paralleled by the Th/U signature of garnet as reflected in the 206Pb/204Pb–208Pb/204Pb diagram (Fig. 4b). The reference lines connecting the isotopic composition of garnet with that of alkali feldspar correspond to Th/U340 Ma values that range from 0·5 to 6·0 (Fig. 4b).
|
|
Rutile has little radiogenic lead, which is due to high contents of common lead (3·1–8·0 ppm) rather than low contents of uranium (18·2–22·5 ppm; Table 1). The 206Pb/238U age of the rutile samples, therefore, is sensitive to the correction of common lead. Using the lead isotopic composition of alkali feldspar (Table 2) for the common lead composition results in apparent 206Pb/238U ages ranging between 320 ± 4 Ma and 345 ± 4 Ma (Table 1). This variation in apparent age cannot be removed by choosing different values for the common lead isotopic composition or higher or lower blank levels. For instance, using a common lead composition with 206Pb/204Pb 18·0 results in apparent 206Pb/238U ages ranging from 329 to 350 Ma. In any case, the U–Pb age of rutile remains ambiguous, as a result of the unknown and potentially heterogeneous nature of the initial lead isotope composition.
Apatite is a rarely used phase in U–Pb geochronology, as this mineral has a rather low closure temperature for dry systems (400–450°C; Watson et al., 1985; Cherniak et al., 1991), which may be even lower in wet systems (see Romer, 1996), and tends to recrystallize and thereby lose lead even at low temperatures. The four apatite samples (Table 1) have contrasting 206Pb/238U ages ranging from 273 to 330 Ma, which implies either that there has been lead loss in some samples after the ambient temperature had fallen below the closure temperature of apatite or that the initial lead isotopic composition in apatite was heterogeneous. This second possibility is also compatible with the lead systematics in the 206Pb/204Pb–207Pb/204Pb and 206Pb/204Pb–208Pb/204Pb diagrams (excess scatter; Fig. 4b and c, Table 2), which reflects a broad range of apparent Th/U after crystallization.
Retrogressed felsic granulite (G97-2)
Fine-grained green and red–brown biotite and alkali feldspar were separated from this sample for Rb–Sr geochronology. Both green and red–brown biotite are retrograde and formed from the breakdown of garnet. Green and brown biotite formed at temperatures higher than the closure temperature for the Rb–Sr system of biotite, which is commonly thought to be 350°C (e.g. Armstrong et al., 1966), although there is an increasing number of arguments that this temperature estimate is too low (see Villa, 1998; Kühn et al., 2000). The two biotite samples had low initial strontium contents (4·96 and 4·71 ppm, respectively) and high Rb contents (1015 and 981 ppm, respectively). Therefore, they developed highly radiogenic Sr isotope ratios (Table 3). The two biotite samples and the alkali feldspar fall on a common isochron that corresponds to an age at 323·0 ± 2·3 Ma (2; MSWD = 0·066; Fig. 5). Because of the highly radiogenic nature of Sr in biotite, possible heterogeneities in the isotopic composition of alkali feldspar and biotite at the time the ambient temperature of the granulite had fallen below the closure temperature of biotite do not have an effect on the calculated Rb–Sr age. The initial composition of strontium at Sri = 0·7191 is highly radiogenic and indicates that the protoliths of the granulite had a long crustal residence before granulite-facies metamorphism.
|
Garnet–clinopyroxene rock (G97-3)
Four samples of titanite from this rock were analysed. All four samples had intermediate 206Pb/204Pb values that demand a good knowledge of the initial lead isotopic composition to define a precise titanite age. The initial lead isotopic composition was estimated using diopside from the same sample and alkali feldspar from sample G97-1. Two samples of diopside were analysed (Table 2). Their lead isotopic compositions are slightly radiogenic and differ between the two samples. Thus, the lead isotopic composition of diopside seems to have been affected by in situ lead growth, resulting in minimum values for the 206Pb/238U and 207Pb/235U ages of titanite. The diopside and titanite samples define a secondary lead isochron that corresponds to an age at 341·9 ± 3·8 Ma (2; MSWD = 1·15) in the 206Pb/204Pb–207Pb/204Pb diagram (Fig. 6). For an undisturbed system, the initial lead isotopic composition of titanite lies on the secondary lead line. Alkali feldspar from sample G97-1 falls on this line and possibly represents the best estimate for the initial lead isotope ratios of titanite. Data reduction using the lead isotopic composition of alkali feldspar yields concordant ages (Fig. 6; Table 1) that agree with the 207Pb/206Pb age of the secondary lead isochron (341·7 ± 4·8 Ma, 2, MSWD = 1·79). The age of the lead line agrees within error limits with the 206Pb/238U age of the individual titanite samples (see Fig. 6), whose ages range from 338·8 to 341·1 Ma if the diopside lead is used for the common lead correction and from 341·4 to 342·4 Ma if the alkali feldspar lead is used for the common lead correction. The weighted mean 206Pb/238U age of the titanite samples is 340·6 ± 0·8 Ma (2; MSWD = 1·49; diopside) and 342·0 ± 0·8 Ma (2; MSWD = 0·219, Fig. 6; alkali feldspar), respectively. It should be noted that 206Pb/238U ages are more precise than 207Pb/206Pb ages for Phanerozoic samples, and that the 206Pb/238U age at 340·6 ± 0·8 Ma represents a minimum age. The 206Pb/238U age at 342·0 ± 0·8 Ma is considered here to be the best age estimate for the titanite from sample G97-3 (see Fig. 6, inset).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Closure temperature of minerals
Ionic diffusion in the crystal structure may result in the loss of mobile ions from minerals. If the mobile ion is the daughter isotope of a geochronological system, its loss will cause the reduction of the apparent isotopic age of the mineral. The widely held assumption that this diffusion and loss from the crystal structure are dependent on the ambient temperature and the cooling rate of the mineral eventually resulted in the definition of the closing temperature of different geochronological systems for a wide variety of minerals and cooling rates (e.g. Dodson, 1973
; Jäger, 1973; Copeland et al., 1988; Mezger et al., 1991; Cherniak, 1995, 2000). Ionic mobility, however, is less dependent on temperature than on the distribution and density of crystal defects, availability and composition of fluids, deformational microstructures and metamorphic history (e.g. Lasaga, 1981; Chakraborty, 1997). Therefore, it has been argued that these factors may affect the geochronological system of a mineral in a much more profound way than temperature does and that many estimates of the closure temperature are too low (e.g. Dahl, 1997; Kamber et al., 1998; Villa, 1998).
Estimates of the closure temperature for the U–Pb system of zircon, monazite and titanite are based on field data, i.e. the presence of inherited zircon, monazite and titanite in magmatic systems (e.g. Copeland et al., 1988; Zhang & Schärer, 1996), comparison among several geochronometers in the same suite or region (e.g. Mezger et al., 1991) and limited experimental studies (e.g. Cherniak et al., 1991; Cherniak, 1995, 2000). Some of these studies, and to an even larger extent confirmations of earlier estimates, involve circular arguments, whereas the petrological basis of other estimates has become obsolete [for examples and a discussion, see Villa (1998)]. From these studies, one would expect that the closing temperature for the U–Pb system under dry conditions is zircon > monazite 
titanite. U–Pb age data of granulites from the SGM, however, give an age sequence titanite zircon > monazite (little or no excess of 206Pb in monazite) or titanite zircon monazite (significant 206Pb excess in monazite), which indicates that the closing temperature for the U–Pb system under dry conditions is in the sequence titanite zircon > monazite or titanite zircon monazite. Furthermore, the closure temperature of titanite, and probably monazite, is considerably higher than previously thought. In either case, the observed age distribution is not compatible with earlier estimates of closure temperatures for these minerals. It should be noted that coincident ages for zircon, titanite and monazite could be obtained even for highly contrasting temperatures of isotopic closure for these minerals, as known from the literature, if the Saxon granulites cooled extremely rapidly. This alternative, however, would imply initial cooling rates in excess of 250–350°C/my (depending on the chosen literature values), which appears unlikely for rocks that are transferred from lower- to middle-crustal levels (see also below).
The contrast of our results with earlier estimates of closure temperatures may be due to the type of sample used. Samples with low apparent closure temperatures may originate from deformed and dynamically recrystallized rocks or ‘wet’ systems, i.e. rocks with a high activity of H2O (aH2O). The granulite samples from Waldheim used for U–Pb dating represent ‘dry’ systems. Sample G97-3, however, shows locally minor secondary muscovite. With ‘dry’, we refer here to systems with low aH2O, rather than the absence of fluids. In this context, it is important to note that aH2O depends strongly on pressure (e.g. Aranovich & Newton, 1996, 1997). During decompression, the same fluid may thus pass through a wide range of aH2O and may change its character at lower pressure. The same fluid that at high pressure is unable to drive hydration reactions may at low pressure drive the formation of hydrous phases. Thus, secondary muscovite in sample G97-3 does not necessarily imply fluid influx as long as the other ions are ‘locally’ available. The formation of muscovite in G97-3, however, requires mobility of K over distances larger than the size of the investigated sample, i.e. decimetre scale or larger. The U–Pb age of titanite, which coincides with that of zircon and possibly is higher than that of monazite, implies that the closure temperature of the U–Pb system of titanite in systems with low aH2O is similar to or higher than that of monazite. The apparently unaffected U–Pb systematics of titanite may reflect the spatial separation of muscovite and titanite in the sample and not necessarily demonstrate its insensitivity to increases in aH2O.
Age of metamorphic peak
The minimum age for the granulite-facies metamorphism is given by the highest U–Pb age, i.e. the titanite age at 342·0 ± 0·8 Ma (2; Fig. 6) and the zircon age at 341·5 ± 0·8 Ma (2; Fig. 3). If all ages represent the time at which the ambient temperature of the rock had fallen below the closure temperature of titanite, zircon and monazite, respectively, then this age would represent a minimum age for both the pressure and temperature peak of metamorphism. For instance, if the temperature of isotopic closure for zircon is 800–900°C (e.g. Dahl, 1997; Mezger & Krogstad, 1997), the age data would contain no information about the high-temperature and high-pressure exhumation history of the SGM, because at such low temperatures the SGM already would have reached a middle-crustal level (
8 kbar; Fig. 2). In this example, both the titanite and zircon age would not be related to the phases used to constrain the P–T conditions. We consider such a low temperature for isotopic closure, however, as unlikely, especially because of the occurrence of inherited zircon crystals in mantle rocks and ultrahigh-temperature rocks. These inherited crystals suggest that the closure temperature of zircon in dry systems far exceeds 1000°C (e.g. Gebauer, 1990; Kröner et al., 1998). The coincidence of the age of zircon with the age of titanite indicates that the closure temperature of titanite should be in the same range as that of zircon under ‘dry’ conditions.
Alternatively, the U–Pb ages of titanite, zircon and monazite represent formation ages. All investigated zircon crystals are of the poly-faceted type commonly found in high-grade metamorphic rocks. No inherited zircon was observed in sample G97-1. Rare inclusions of metamorphic zircon in garnet indicate that zircon formed at the same time as or earlier than garnet, which is part of the peak-pressure assemblage. Thus, the U–Pb zircon age should be related to the pressure maximum. This conclusion is supported by the following argument. Zircon growth is controlled by the availability of zirconium, which essentially becomes available through the breakdown of major phases that contain significant amounts of zirconium (e.g. clinopyroxene, amphibole, biotite, Fe-oxides) and the dissolution of old zircon grains. For instance, ilmenite, magnetite and biotite contain 150 to >550 ppm, 20–320 ppm and 17–50 ppm zirconium, respectively (e.g. Brooks, 1969; McKay et al., 1986; Ewart & Griffin, 1994). The consumption of these phases during prograde metamorphism and close to metamorphic peak conditions should release zirconium and allow the crystallization of metamorphic zircon. Although the growth of zircon cannot be uniquely related to a distinct P–T window, it seems likely that it is related to major phases that have grown close to Pmax. Textural evidence shows that titanite (G97-3) is part of the peak metamorphic assemblage (Rötzler & Romer, 2001). It should be noted that the present U–Pb age data of zircon, monazite and titanite—in agreement with Nasdala et al. (1996) and Kröner et al. (1998)—are incompatible with earlier reported U–Pb zircon (SHRIMP) and monazite ages in the range 320–315 Ma (Baumann et al., 1997; Vavra et al., 1998).
Cooling and exhumation history
Numerical modelling demonstrates that the orogen-internal temperature distribution and variation with time strongly depend on the chosen boundary conditions, the most critical of which seems to be heat flux at the bottom of the modelled lithosphere and the exhumation rate (e.g. Stüwe & Sandiford, 1995; Grasemann et al., 1998). Assuming constant depth–constant heat flux as boundary condition (e.g. England & Thompson, 1984) not only implies lithospheric delamination, but also results in a temperature rise during exhumation (Stüwe & Sandiford, 1995). Such a temperature rise is not observed for the Saxon granulites (Rötzler & Romer, 2001, fig. 2). Assuming alternatively a constant depth–constant temperature boundary condition implies variable heat flow at the base of the lithosphere over time and does not require lithospheric delamination (Stüwe & Sandiford, 1995). This results in P–T paths that strongly depend on the exhumation rate (Grasemann et al., 1998): slowly exhumed rocks have plenty of time to lose their heat by conduction, and thus will cool distinctly during exhumation. In contrast, fast exhumation of high-pressure rocks will show little change in temperature along their P–T path for segments of fast exhumation. Fast exhumation rates and fast cooling rates seem little compatible with each other.
The inferred cooling history of the SGM strongly depends on the closure temperature of the various dated minerals, which among other factors also depends on the cooling rate. Most temperature values for isotopic closure are supposed to apply for scenarios of moderate to slow cooling, and therefore, may be too low for settings of fast cooling. Furthermore, the closing temperature of some minerals is also under debate because of the poorly known influence of fluids, deformation, distribution of crystal defects and recrystallization (see previous sections). The exhumation history of the SGM depends on the reliability of pressure estimates and their linking with the age data. All pressure estimates used for the calculation of exhumation rates were derived directly from the P–T path of dated samples (Figs 2 and 7), for which detailed petrological data exist (Rötzler & Romer, 2001). This avoids problems associated with spatially and temporally highly variable geothermal gradients in orogenically thickened crust.
|
Cooling and exhumation rates are calculated for discrete temperature values (for crystallization or cooling ages), the corresponding pressure being read from the P–T path (see Rötzler & Romer, 2001; Figs 2 and 7). For cooling ages, the assumed cooling temperature was used to derive the appropriate pressure from the P–T path (see Rötzler & Romer, 2001; Fig. 7). All rates are calculated taking into account the uncertainties of the age determinations. Although 1 uncertainties for pressure and temperature estimates were given by Rötzler & Romer (2001), they are not used for the calculation of the cooling and exhumation rates. Taking them into consideration would only slightly enhance the range of cooling and exhumation rates, but would not result in a qualitatively different solution. The extreme range in rates in the high-temperature segment of the exhumation path of the Saxon granulites is entirely due to the small difference between the ages of zircon/titanite and monazite, which allows even small age uncertainties to have a drastic effect on the derived rates. Cooling and exhumation rates are calculated for two extreme models (Fig. 7).
Model 1. The U–Pb ages of titanite and zircon are formation ages. Textural evidence indicates that they formed close to peak pressure conditions. Their age, therefore, reflects the pressure peak. Monazite, which on textural grounds is considered to have grown during peak metamorphism, is interpreted to yield a cooling age. Biotite yields a cooling age.
Model 2. Monazite, titanite and zircon U–Pb ages represent cooling ages. Any petrological argumentation to link the crystallization of these trace phases with the major phases, and thereby with P–T conditions, is fruitless as the age of the mineral is not related to the time of crystallization. The inferred post-peak cooling history depends strongly on the chosen values for the closure temperature of these minerals, a subject that in part is disputed (e.g. Mezger et al., 1989, 1991; Santos Zalduegi et al., 1996; Dahl, 1997; Mezger & Krogstad, 1997; Kamber et al., 1998; Villa, 1998). Age information about the metamorphic peak is not contained in the analysed systems and can be estimated only indirectly from the geological context.
Cooling history
From the cooling and exhumation rates estimated from the simple Models 1 and 2, it becomes clear that most combinations of closing-temperature estimates indicate geologically fast rates (Fig. 7). An age difference of 2–4 my between zircon and monazite results in cooling rates of 25–50°C/my for Model 1 and 75–150 to 25–50°C/my for Model 2, depending on the chosen values for the closure temperature of monazite and zircon (see Fig. 7). It should be noted that a significant contribution of excess 206Pb in monazite raises the monazite age, and thus would increase the derived average cooling rate (or would require a higher temperature of isotopic closure for monazite). The age difference between monazite and biotite, in combination with a range of closure temperatures for both minerals, yields average cooling rates ranging from 40 to 20°C/my (Fig. 7). This range, however, represents a minimum estimate, as the temperature of isotopic closure of biotite may be considerably higher than commonly thought [for an example, see Kühn et al. (2000)].
Exhumation history
The exhumation history is intimately related to the cooling history and the chosen values for isotopic closure of the various geochronometers. The high-pressure segment of the exhumation path yields geologically fast average exhumation rates (Fig. 7). For instance, an age difference of 2–4 my between zircon and monazite results in exhumation rates of 9–18 mm/yr for Model 1 and 14–28 to 2–5 mm/yr for Model 2, depending on the chosen values for the closure temperature of monazite and zircon (see Fig. 7). It should be noted that a significant contribution of excess 206Pb in monazite raises the monazite age, and thus would increase the derived exhumation rate, or alternatively would indicate that the temperatures used here, 750°C and 950°C for isotopic closure of monazite, are distinctly too low. Fast early exhumation also is indicated by a U–Pb zircon age of 340 Ma (SHRIMP, no uncertainties stated, Vavra et al., 1999) from an anatectic lens in the Cordierite Gneiss Unit (Fig. 1). Cordierite gneisses occur between the Granulite Complex and the Schist Cover. They form part of the metamorphic zonation that is mainly due to thermal overprint from the hot Granulite Complex (e.g. Rötzler, 1992; Reinhardt & Kleemann, 1994). Their thermal peak was reached when the Cordierite Gneiss Unit, and thus the Granulite Complex, was at a middle-crustal level (4 kbar, Rötzler, 1992). Although it may not be possible to compare directly the thermal evolution of rocks from the interior of the granulite massif with rocks at its margin, the essentially indistinguishable ages of titanite and zircon from the peak assemblage of the granulite samples and the zircon age from the Cordierite Gneiss Unit require a very fast exhumation of the hot Saxon granulites to at least a middle-crustal level. For instance, if the granulite in the vicinity of the dated Cordierite Gneiss Unit had a P–T–t path comparable with that of the samples from Waldheim, the inferred exhumation and cooling rates are far in excess of 20 mm/yr and 100°C/my, respectively. The age difference between monazite and biotite, in combination with a range of closure temperatures for both minerals, yields exhumation rates ranging from 2 to <0·2 mm/yr (Fig. 7). This range is little dependent on the temperature of isotopic closure of biotite. Thus, once the granulites have reached the middle crust, independent of which model is chosen to interpret the data, the inferred exhumation rates invariably are moderate to low.
The two models thus constrain the exhumation of the Saxon granulites to have been fast during its initial phase and then markedly slower once the rocks had reached middle-crustal level. Such a slow-down of exhumation is also known from other areas with high-pressure rocks, such as, for instance, the Alps (Duchêne et al., 1997), the Gföhl area in the Bohemian Massif (O’Brien, 1997) and the Red River Shear Zone (Nam et al., 1998), and reflects the decrease of the driving force for exhumation with time. Independent of the chosen values for the temperature of isotopic closure of zircon, titanite and monazite, exhumation rates of the SGM are high (Fig. 7). They fall in the range known from other areas. For instance, exhumation rates as high as 15–39 mm/yr are inferred for the ultrahigh-pressure rocks from Dora Maira (Gebauer, 1997) and eclogites from Papua New Guinea (Hill & Baldwin, 1993). Exhumation rates in excess of 10 mm/yr have also been suggested for parts of the Bohemian Massif, high-grade rocks of the Alps and eclogites occurring along the Red River Shear Zone (e.g. Duchêne et al., 1997; Nam et al., 1998).
Comparison with thermal models
The exhumation and cooling rates do not conflict with rates known from other high-grade metamorphic areas. Little dependent on the chosen values for the temperature of isotopic closure of monazite, the obtained exhumation rates are high. In contrast, the obtained cooling rates vary strongly and become extremely high for closure temperatures as low as those commonly considered to be typical for monazite (e.g. Copeland et al., 1988; Zhang & Schärer, 1996) and biotite (e.g. Armstrong et al., 1966; Jäger, 1973). Fast average rates for the early phases of exhumation that are paralleled by fast average rates of cooling, however, seem little compatible with results from modelling (e.g. Grasemann et al., 1998). Instead, during initial rapid exhumation, there should be little cooling, as (1) the low thermal conductivity would not allow the loss of large amounts of heat within a short time and (2) the thermal contrast between the hot granulite and its wall rock would become large only after a significant amount of exhumation had occurred, i.e. the hot granulite had been placed in contact with rocks of the middle and upper crust. The thermal contrast between the granulite and its wall rocks, however, is the driving force for the cooling of the granulite. The apparent conflict between the estimated cooling and exhumation rates and the results of modelling may be resolved if either of the following two explanation applies. (1) In dry undeformed systems, the temperature of isotopic closure of monazite is distinctly higher than 750°C and possibly even 950°C. Such an explanation requires significantly lower rates of cooling during the initial stages of exhumation, which remains fast. (2) Thermal conduction was not the dominant process of cooling. Instead, there were alternative processes of cooling. The thermal contrast to the wall rock may have induced in the wall-rock dehydration reactions and possibly even partial melting. Both processes require large amounts of heat and, therefore, act as sinks (e.g. Fyfe et al., 1978; Walther & Orville, 1982; Haack & Zimmermann, 1996). Especially the presence of fluids, and, to a lesser extent, of melts, would allow for additional rapid heat loss through convection (e.g. Fyfe et al., 1978). Furthermore, fluids and melts reduce the rock strength, which is a prerequisite for models that suggest the Granulite Massif was hydraulically expelled and intruded into the NW foreland of the Variscan orogen (e.g. Henk, 1999; Franke & Stein, 2000).
Effect of reaction history on the initial isotopic composition of geochronometers
The U–Pb system of garnet gives a broad range of apparent ages, some of which are considerably older than those of zircon, monazite and titanite (see Figs. 3, 4 and 6) despite textural evidence of a common crystallization history and age. Such a discrepancy has been argued elsewhere to be due to inheritance of garnet from earlier metamorphism (e.g. Erambert & Austrheim, 1993), the presence of inherited zircon or monazite (e.g. Vance & Holland, 1993; Zhou & Hensen, 1995), or to reflect the pressure peak (e.g. Chen et al., 1998), whereby in the last interpretation the younger ages of other minerals are thought to be associated with the thermal peak. Textural and mineral-chemical evidence from sample G97-1 is not compatible with the presence of inherited garnet from an earlier high-grade metamorphic event (Rötzler & Romer, 2001). Garnet, however, shows a distinct compositional variation with higher Ca contents in its core (see Rötzler & Romer, 2001). An interpretation that the higher apparent age is related to the pressure peak of metamorphism is not compatible with the P–T history of the area and the regional tectonic scenario (e.g. Franke & Stein, 2000). Inherited zircon could significantly affect the U–Pb age of garnet if it originates from an earlier event (e.g. Vance & Holland, 1993; Zhou & Hensen, 1995). Although large zircon inclusions have been observed in garnet, there is no evidence for the presence of submicroscopic zircon inclusions in garnet. Therefore, the anomalously high U–Pb ages do not represent a memory from the early geological history of the investigated granulites. A wide range of U–Pb ages, as well as anomalously high ages, however, may also be due to initial isotopic heterogeneities originating from the reaction history. Garnet has grown over a wide range of pressure and temperature at the expense of various phases, such as biotite, aluminium silicates, plagioclase and ilmenite. During growth, the composition of garnet as well as the composition and relative contribution of the consumed phases change. Lead incorporated into garnet represents the weighted average (or a fraction thereof) of the lead made available from the consumed phases. All these phases may contain trace amounts of uranium and lead. Most importantly, the U/Pb concentration ratio will differ among the various phases. Thus, even if all consumed phases originally had the same lead isotopic composition, they all have developed individual and contrasting lead isotope compositions by the time of consumption through the formation of garnet. The diversity of lead composition thereby is a function of the range in U/Pb and the time elapsed between formation and consumption of the individual phases. Such a dependence on the reaction history of the initial isotopic composition of garnet has so far not been documented for lead, but is well known for the Sm–Nd system of garnet, for which—under unfavourable circumstances—it will yield grossly incorrect ages (e.g. Jagoutz, 1995; Romer & Smeds, 1996). The crucial point is (analogous to the example of incorrect Sm–Nd ages) that the source of initial lead in garnet varies with time, commonly is poorly known, and generally is not accessible for lead isotope analysis (as these minerals were consumed during garnet formation). The effect of heterogeneous initial lead becomes obvious only in samples with low parent/daughter ratio (P/D), i.e. those samples that even after a long time show little radiogenic lead. Depending on reaction history, age of the consumed phases and U/Pb of the consumed phases, heterogeneous initial lead may result in an age diversity of >100 my (see Fig. 4) or only a few million years. Depending on the variation of the initial lead isotopic composition, apparent ages for core and rim of garnet may differ. Thereby, it is possible that the core appears younger than the rim, just as there are cases where the core appears older than the rim. In the latter case, the age difference may be used improperly to deduce growth rates.
This problem of heterogeneous initial lead is also illustrated by the Th/U systematics of garnet. Calculated Th/U of garnet range from 0·5 to 6·0 (see Fig. 4b). Such a wide range cannot be obtained by changes in distribution coefficient for these two elements. Instead, it has to be due to contrasting availability of Th and U during garnet growth. The heterogeneous Th/U reflects the contrasting nature of reaction partners available for consumption during garnet formation, and thus indirectly argues for heterogeneous initial lead.
In a similar manner as for garnet, the scatter of the lead isotope data, the range in apparent Th/U and the range in 206Pb/238U age of rutile and apatite may be entirely due to heterogeneities in the initial lead isotope signatures of these minerals. For instance, attributing each rutile fraction an individual initial lead would result in concordant, internally consistent ages for all three samples without requiring extreme values for the initial lead isotope ratios. However, the resulting rutile age can be ‘adjusted’ (by choice of a suitable initial lead) to any value between the 342 Ma maximum age given by titanite and zircon (which would imply that rutile has a closure temperature for Pb that lies in the same range as that of zircon) and 323 Ma, i.e. the Rb–Sr age of biotite (which would imply that rutile remains an open system with respect to lead until the ambient temperature has fallen below the closure temperature of biotite). Similarly, there is textural evidence for two generations of apatite in sample G97-1. During crystallization, apatite incorporated lead from its precursors and, thus, inherited their isotopic signature. Although apatite is prone to lose radiogenic lead at high temperature, it is likely that initial lead and radiogenic lead behave differently. The radiogenic lead will occur in the crystal structure on sites that are damaged by the radioactive decay. The migration of lead from these sites will need less energy than migration of initial lead that occurs on undamaged structural sites suited for lead. Thus, it might be speculated that apatite retains a partial memory of its initial lead, i.e. the lead isotopic composition of several generations of apatite that cooled together below the closure temperature of apatite will not necessarily be homogeneous. This heterogeneity, which will also undoubtedly be present in minerals such as zircon, monazite and xenotime, will have an insignificant effect on the age of minerals with high P/D ratios, as the radiogenic lead contribution will overwhelm any initial heterogeneity given enough time. In contrast, for minerals that have low P/D ratios, the contribution of radiogenic lead will remain modest and the initial heterogeneities will remain apparent.
Age constraints for tectonic models
Early tectonic models for the formation of the Saxon granulites (e.g. Weber & Behr, 1983; Franke, 1993) discuss the high-temperature granulite-facies metamorphism in an Ordovician extensional setting (heat source through upwelling mantle) and subsequent exhumation as a metamorphic core complex in an extensional setting. These models rely heavily on (1) the wide range of age data obtained on granulites, whereby most minerals started to retain their radiogenic isotopes not before 340–350 Ma, and (2) the necessity of a mantle-derived heat source, which could become available in an extensional setting, to obtain the high temperature of granulite-facies metamorphism. However, we find no evidence for two metamorphic events or slow cooling from an Ordovician UHT event to a Variscan granulite-facies metamorphism. Instead, there is only evidence for one UHT event that lasted only a few million years, i.e. hardly longer than the geochronological resolution.
There are two types of garnet, relict Ca-rich garnet cores and re-equilibrated Ca-poor garnet. The chemical range between the two garnets reflects release of Ca in response to decompression (Rötzler & Romer, 2001). The high-pressure peak conditions in sample G97-1 are given by the assemblage garnet + kyanite + ternary feldspar + quartz, whereby the Ca-rich garnet is part of this assemblage. Zircon also is part of this assemblage. Furthermore, titanite from the garnet–clinopyroxene rock (G97-3) is part of the peak metamorphic assemblage and yields the same Variscan age as zircon in sample G97-1. Thus, there is no indication for a preserved Ordovician metamorphic history in the investigated granulite samples.
U–Pb zircon ages from the Saxon granulites fall in a small age range (e.g. von Quadt, 1993; Nasdala et al., 1996; Kröner et al., 1997, 1998; this paper) and do not require a long or complex metamorphic history. The large range in age data mainly originates from Sm–Nd mineral isochrons (470–340 Ma). As seen in the previous section, and argued by Jagoutz (1995), the initial Nd isotopic composition of the various metamorphic phases depends on the metamorphic reaction history and does not necessarily have to be the same for all metamorphic minerals. As a consequence, an ‘isochron’ defined by these phases initially may have a positive or negative slope (e.g. Jagoutz, 1995; Romer & Smeds, 1996) that translates into ages that are too old and young, respectively. Thus, the range in ages attributed to the UHT metamorphism possibly originates to a significant part from unfavourably suited samples. A comparison of age data from the UHT Saxon granulites with other Variscan granulites in the Bohemian Massif shows that U–Pb zircon age data fall in a small range [e.g. Gföhl assemblage: 337–345 Ma (van Breemen et al., 1982; Schenk & Todt, 1983; Friedl et al., 1994)], whereas age data obtained by other methods show a considerably larger spread (e.g. Beard et al., 1992; Becker, 1993; von Quadt, 1993). Again, we interpret this spread in Sm–Nd ages as due to the lack of initial isotopic homogeneity among the phases used for mineral isochrons. Thus, the anomalous Sm–Nd ages in the Saxon granulites and the other Variscan granulites of the Bohemian massif should not be interpreted in terms of a prolonged metamorphic history, memory of earlier stages of metamorphism, or low-temperature mobility of Sm and Nd (for the ages that are too young). Instead, these anomalous ages should be seen as what they are, the unfortunate analytical results from samples that do not fulfil the initial requirements for an isochron, i.e. the same initial isotopic composition at the same time for all phases used to define the isochron.
Petrological and geochronological data demonstrate that there was only one UHT event, and that the UHT rocks were exhumed quickly after 340 Ma. In contrast, sedimentological studies have shown a rather undisturbed sedimentary record in the orogenic foreland until 334 Ma, which implies that there was no significant change in topographic relief until 334 Ma (see Franke & Stein, 2000). This apparent conflict was resolved in a series of models that have in common fast later transport with subsequent diapiric rise of the Saxon granulites (e.g. Henk, 1999; DEKORP Working Group, 1999), but differed in the source of the protoliths. As the interpretation of the Saxon granulites as resurfaced material from the subducted Rhenohercynian plate does not work because of isotope geochemical reasons (Jaeckel et al., 1999), the model favoured at present suggests a derivation of the Saxon granulites from the Saxothuringian orogenic root to the SE of its present position by ‘hydraulic expulsion’ and subsequent diapiric rise (Henk, 1999; DEKORP Working Group, 1999; Franke & Stein, 2000).
The Saxon granulites are not the only Variscan UHT rocks within the Bohemian Massif. Furthermore, HT and UHT rocks within the Bohemian Massif do not represent small isolated, individual occurrences, but encompass huge volumes of rock, all of which are characterized by fast rates of exhumation to a middle- and upper-crustal level. Reliable U–Pb age data from these massifs fall in a narrow age range from 345 to 337 Ma for the HT and UHT metamorphism (van Breemen et al., 1982; Schenk & Todt, 1983; Friedl et al., 1994). Tectonic models for the evolution of the Saxon granulites, therefore, not only have to explain the fast exhumation history of this massif at the northern border of the Bohemian Massif, but also the occurrence and exhumation of coeval HT and UHT metamorphic rocks throughout the Bohemian massif.
|
SUMMARY |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Granulite-facies rocks in the structurally lower part of the Saxon Granulite Massif, Germany, experienced UHT–HP metamorphism during the Variscan orogeny. We sampled a sapphirine-bearing granulite (G97-1), a retrogressed felsic granulite (G97-2) and a garnet–clinopyroxene rock (G97-3), all collected within 1 km along a railway track near Waldheim, to reconstruct the P–T path and to constrain the age of the metamorphism and exhumation of this granulite complex.
P–T data derived from mineral assemblages of two samples in accordance with earlier estimates demonstrate that metamorphism in the Saxon Granulite Massif culminated in UHT–HP conditions of 1010–1060°C and 22 kbar (Rötzler & Romer, 2001). Mineral assemblages related to the UHT–HP metamorphism and retrograde assemblages related to the exhumation of the Granulite Complex define a clockwise P–T loop, where UHT conditions persisted even after a significant pressure drop. Retrograde hydrous phases are characteristic once the rocks have reached low pressure.
The U–Pb age of concordant zircon (G97-1) at 341·5±0·8 Ma (2) and concordant titanite (G97-3) at 342·0±0·8 Ma (2) are interpreted as formation ages. As titanite in sample G97-3 is part of the mineral assemblage defining metamorphic peak conditions, this age corresponds to the metamorphic pressure peak. U–Pb monazite data (G97-1) are slightly discordant and yield a weighted 207Pb/206Pb minimum age at 338·0 ± 0·5 Ma (2). The U–Pb ages of zircon, titanite and monazite demonstrate (1) fast exhumation of the Granulite Complex during the first 2–4 my after the pressure peak at an average rate of 9–18 mm/yr and fast cooling at an average rate of 25–50°C/my and (2) that the temperature for isotopic closure of titanite in systems of low aH2O is in the same range as that of zircon and higher than that of monazite.
The Rb–Sr biotite age (G97-2) at 323·0 ± 2·3 Ma (2) demonstrates that the average exhumation and cooling rates of the Granulite Complex dropped significantly for the segment bracketed by the age of monazite and biotite. Depending on the values chosen for isotopic closure exhumation rates range from 2 to <0·2 mm/y and cooling rates range from 40 to 20°C/my.
Rutile, garnet and apatite (all G97-1) have little radiogenic lead and yield a broad range in apparent 206Pb/238U ages. This age range is a characteristic feature of metamorphic minerals with low parent/daughter ratio and is explained as being due to initial isotopic heterogeneity of these minerals. The initial isotopic heterogeneity is the result of the reaction history of the sample, may cause anomalously high or low ages, and eventually precludes the use of low-P/D minerals as precise geochronometers.
|
ACKNOWLEDGEMENTS |
|---|
We thank M. Boche and J. Müller for help with the mineral separation, and M. Dziggel (GFZ Potsdam) for drawing the final version of the geological map. R.L.R. thanks Ralf Hetzel (GFZ) for discussion and constructive comments on the manuscript. We gratefully acknowledge constructive reviews by W. Franke, N. Machado and A. von Quadt, as well as thoughtful editorial handling by P. Kempton. Sampling was financially supported by DFG-Grant KE 552/2-3. Analytical work was performed at GeoForschungsZentrum Potsdam.
|
FOOTNOTES |
|---|
*Corresponding author. Telephone: + 49-331-288-1405. Fax: + 49-331-288-1474. E-mail: romer{at}gfz-potsdam.de
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTINGMETAMORPHIC PETROLOGYGEOCHRONOLOGYDISCUSSIONSUMMARYREFERENCES |
|---|
Aranovich, L. Ya. & Newton, R. C. (1996). H2O activity in concentrated NaCl solutions at high pressures and temperatures measured by the brucite–periclase equilibrium. Contributions to Mineralogy and Petrology 125, 200–212.
Aranovich, L. Ya. & Newton, R. C. (1997). H2O activity in concentrated KCl solutions at high pressures and temperatures measured by the brucite–periclase equilibrium. Contributions to Mineralogy and Petrology 127, 261–271.
Armstrong, R. L., Jäger, E. & Eberhardt, P. (1966). A comparison of K–Ar and Rb–Sr ages on Alpine biotites. Earth and Planetary Science Letters 1, 13–19.
Baumann, N., Pilot, J., Werner, C.-D. & Todt, W. (1997). Zur Geochronologie und Isotopengeochemie des Sächsischen Granulitgebirges. Terra Nostra 97(5), 19–22.
Beard, B. L., Medaris, L. G., Jr, Johnson, C. M., Bruecker, H. K. & Misar, Z. (1992). Petrogenesis of Variscan high-temperature Group A eclogites from the Moldanubian Zone of the Bohemian Massif, Czechoslovakia. Contributions to Mineralogy and Petrology 111, 468–483.
Becker, H. (1993). Contrasting origin of Variscan high-P granulites and high-T peridotites (Bohemian Massif): evidence from Sm–Nd garnet ages. Terra Abstracts 5, 380.
Brooks, C. K. (1969). On the distribution of zirconium and hafnium in the Skaergaard Intrusion, East Greenland. Geochimica et Cosmochimica Acta 33, 357–374.
Chakraborty, S. (1997). Rates and mechanisms of Fe–Mg interdiffusion in olivine at 980 degrees–1300 degrees C. Journal of Geophysical Research B102, 12317–12331.
Chakraborty, S. & Ganguly, J. (1990). Compositional zoning and cation diffusion in aluminosilicate garnets. In: Ganguly, J. (ed.) Diffusion, Atomic Ordering and Mass Transfer. Advances in Physical Geochemistry, 8. Heidelberg: Springer, pp. 120–175.
Chen, F., Hegner, E. & Todt, W. (1998). Garnet dating of granulites from the Variscan foldbelt in Central Europe; evidence for early and pre-Variscan high grade metamorphism. Acta Universitatis Carolinae, Geologica 42, 221–222.
Cherniak, D. (1995). Lead diffusion in titanite and preliminary results on the effects of radiation damage on Pb transport. Chemical Geology 110, 177–194.
Cherniak, D. (2000). Pb diffusion in rutile. Contributions to Mineralogy and Petrology 139, 198–207.
Cherniak, D., Lanford, W. A. & Ryerson, F. J. (1991). Lead diffusion in apatite and zircon using ion implantation and Rutherford backscattering techniques. Geochimica et Cosmochimica Acta 55, 1663–1673.
Copeland, P., Parrish, R. R. & Harrison, T. M. (1988). Identification of inherited radiogenic Pb in monazite and implications for U–Pb systematics. Nature 333, 760–763.
Dahl, P. S. (1997). A crystal-chemical basis for Pb retention and fission-track annealing systematics in U-bearing minerals, with implications for geochronology. Earth and Planetary Science Letters 150, 277–290.
DEKORP Working Group (1999). Structure of the Saxonian granulites: geological and geophysical constraints on the exhumation of high-pressure/high-temperature rocks in the mid-European Variscan belt. Tectonics 18, 756–773.
Dodson, H. M. (1973). Closure temperature in cooling geochronological and petrological systems. Contributions to Mineralogy and Petrology 40, 259–274.[Web of Science]
Duchêne, S., Lardeaux, J.-M. & Albarède, F. (1997). Exhumation of eclogites: insights from depth–time path analysis. Tectonophysics 280, 125–140.[Web of Science]
England, P. C. & Thompson, A. B. (1984). Pressure–temperature–time paths of regional metamorphism. I. Heat transfer during the evolution of regions of thickened continental crust. Journal of Petrology 25, 894–928.
Erambert, M. & Austrheim, H. (1993). The effect of fluid and deformation on zoning and inclusion patterns in poly-metamorphic garnets. Contributions to Mineralogy and Petrology 115, 204–214.
Ewart, A. & Griffin, W. L. (1994). Application of proton-microprobe data to trace-element partitioning in volcanic rocks. Chemical Geology 117, 251–284.
Franke, W. (1993). The Saxonian Granulites: a metamorphic core complex? Geologische Rundschau 82, 505–515.
Franke, W. & Stein, E. (2000). Exhumation of high-grade rocks in the Saxo-Thuringian belt: geological constraints and geodynamic concepts. 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, 337–354.
Friedl, G., von Quadt, A. & Finger, F. (1994). 340 Ma U/Pb Monazitalter aus dem niederösterreichischen Moldanubikum und ihre geologische Bedeutung. Terra Nostra 94(3), 43–46.
Fyfe, W. S., Price, N. J. & Thompson, A. B. (1978). Fluids in the Earth’s Crust. Amsterdam: Elsevier.
Gebauer, D. (1990). Isotopic systems—geochronology of eclogites. In: Carswell, D. A. (ed.) Eclogite Facies Rocks. New York: Blackie, pp. 141–158.
Gebauer, D. (1997). 35 Ma old ultrahigh-pressure metamorphism and evidence for very rapid exhumation in the Dora Maira Massif, Western Alps. Lithos 41, 5–24.[Web of Science]
Grasemann, B., Ratschbacher, L. & Hacker, B. R. (1998). Exhumation of ultrahigh-pressure rocks: thermal boundary conditions and cooling history. In: Hacker, B. R. & Liou, J. G. (eds) When Continents Collide: Geodynamics and Geochemistry of Ultrahigh-Pressure Rocks. Dordrecht: Kluwer Academic, pp. 117–139.
Grew, E. S. (1986). Petrogenesis of kornerupine at Waldheim (Sachsen), German Democratic Republic. Zeitschrift für Geologische Wissenschaften 14, 525–558.
Grew, E. S. (1989). A second occurrence of kornerupine in Waldheim, Saxony, German Democratic Republic Zeitschrift für Geologische Wissenschaften 17, 67–76.
Haack, U. K. & Zimmermann, H. D. (1996). Retrograde mineral reactions: a heat source in the continental crust? Geologische Rundschau 85, 130–137.
Hagen, B., Hoernes, S. & Rötzler, J. (1995). Consistent ultra-high temperatures in the Saxonian granulite massif as deduced from cation exchange and O-isotope thermometry. Terra Abstracts 7, 340.
Henk, A. (1999). Modelling lower crustal extrusion in the Saxothuringian belt. Terra Nostra 99(1), 100.
Hill, E. J. & Baldwin, S. L. (1993). Exhumation of high-pressure metamorphic rocks during crustal extension in the D’Entrecasteaux region, Papua New Guinea. Journal of Metamorphic Geology 11, 261–277.
Hofmann, A. W. & Giletti, B. J. (1970). Diffusion of geochronological important nuclides in minerals under hydrothermal conditions. Eclogae Geologicae Helvetiae 63, 141–150.
Jaeckel, P., Anthes, G., Molzahn, M., Oncken, O. & Reischmann, T. (1999). Pb isotopic systematics of the central Variscan belt. Terra Nostra 99(1), 111–112.
Jaffey, A. H., Flynn, K. F., Glendenin, L. E., Bentley, W. C. & Essling, A. M. (1971). Precision measurement of half-lives and specific activities of 235U and 238U. Physics Review C4, 1889–1906.
Jäger, E. (1973). Die alpine Orogenese im Lichte der radiometrischen Altersbestimmung. Eclogae Geologicae Helvetiae 66, 11–21.
Jagoutz, E. (1995). Isotopic constraints on garnet equilibration. Terra Abstracts 7(1), 339.
Kamber, B. S., Frei, R. & Gibb, A. J. (1998). Pitfalls and new approaches in granulite chronometry—an example from the Limpopo Belt, Zimbabwe. Precambrian Research 91, 269–285.
Krogh, T. (1973). A low-contamination method for hydrothermal decomposition of zircon and extraction of U and Pb for isotopic age determinations. Geochimica et Cosmochimica Acta 37, 485–494.[Web of Science]
Kroner, U. (1995). Postkollisionale Extension am Nordrand der Böhmischen Masse: Die Exhumierung des Sächsischen Granulitgebirges. Freiberger Forschungshefte C457, 1–114.
Kröner, A., O’Brien, P. J. & Pidgeon, R. J. (1997). Carboniferous high, medium and low pressure metamorphism in granulites from the South Bohemia confirmed by SHRIMP zircon ages. Terra Nova (Abstract Supplement 1) 9, 495.
Kröner, A., Jaeckel, P., Reischmann, T. & Kroner, U. (1998). Further evidence for an early Carboniferous (340 Ma) age of high-grade metamorphism in the Saxonian granulite complex. Geologische Rundschau 86, 751–766.
Kühn, A., Glodny, J., Iden, K. & Austrheim, H. (2000). Retention of Precambrian Rb/Sr phlogopite ages through Caledonian eclogite facies metamorphism, Bergen Arc Complex, W-Norway. Lithos 51, 305–330.[Web of Science]
Lasaga, A. C. (1981). The atomistic basis of kinetics: defects in minerals. In: Lasaga, A. C. & Kirkpatrick, R. J. (eds) Kinetics of Geochemical Processes. Mineralogical Society of America, Reviews in Mineralogy 8, 261–319.
Ludwig, K. R. (1994). ISOPLOT 2.71: a plotting and regression program for radiogenic-isotope data. US Geological Survey Open-File Report 91-445, 44 pp.
Manhès, G., Minster, J. F. & Allègre, C. J. (1978). Comparative uranium–thorium–lead and rubidium–strontium study of the Saint Séverin amphibolite: consequences for early solar system chronology. Earth and Planetary Science Letters 39, 14–24.
Mattinson, J. M. (1973). Anomalous isotopic composition of lead in young zircons. Carnegie Institution of Washington Yearbook 72, 613–616.
McKay, G. A., Wagstaff, J. & Yang, S.-R. (1986). Zirconium, hafnium, and rare earth element partition coefficients for ilmenite and other minerals in high-Ti lunar mare basalts: an experimental study. Proceedings of the 16th Lunar and Planetary Science Conference. Journal of Geophysical Research Supplement 97, D226–D237.
Mezger, K. & Krogstad, E. J. (1997). Interpretation of discordant U–Pb zircon ages: an evaluation. Journal of Metamorphic Geology 15, 127–140.
Mezger, K., Hanson, G. N. & Bohlen, S. R. (1989). High-precision U–Pb ages of metamorphic rutile: application to the cooling history of high-grade terranes. Earth and Planetary Science Letters 96, 106–118.
Mezger, K., Rawnsley, C. M., Bohlen, S. R. & Hanson, G. N. (1991). U–Pb garnet, sphene, monazite, and rutile ages: implications for the duration of high-grade metamorphism and cooling histories, Adirondacks Mts., New York. Journal of Geology 99, 415–428.
Nam, T. N., Toriumi, M. & Itaya, T. (1998). P–T–t paths and post-metamorphic exhumation of the Day Nui Con Voi shear zone in Vietnam. Tectonophysics 290, 299–318.
Nasdala, L., Gruner, T., Nemchin, A. A., Pidgeon, R. T. & Tichomirowa, M. (1996). New SHRIMP ion microprobe measurements on zircons from Saxonian magmatic and metamorphic rocks. Freiberger Isotopenkolloquium 1996, Proceedings. Freiberg: Freiberg University of Mining and Technology. pp. 205–214.
Nasdala, L., Wenzel, M., Vavra, G., Irmer, G., Wenzel, T. & Kober, B. (2001). Metamictisation of natural zircon: accumulation versus thermal annealing of radioactivity-induced damage. Contributions to Mineralogy and Petrology 141, 125–144.
O’Brien, P. J. (1997). Garnet zoning and reaction textures in overprinted eclogites, Bohemian Massif, European Variscides; a record of their thermal history during exhumation. Lithos 41, 119–133.[Web of Science]
Parrish, R. R. (1990). U–Pb dating of monazite and its application to geological problems. Canadian Journal of Earth Sciences 27, 1431–1450.[Web of Science]
Reinhardt, J. & Kleemann, U. (1994). Extensional unroofing of granulitic lower crust and related low-pressure, high-temperature metamorphism in the Saxonian Granulite Massif, Germany. Tectonophysics 238, 71–94.
Reinhardt, J. & Vavra, G. (1997). Petrological and geochronological constraints on the tectonometamorphic history of the Saxonian Granulite Massif. Terra Nostra 97(5), 137–139.
Romer, R. L. (1996). U–Pb systematics of stilbite-bearing low-temperature mineral assemblages from the Malmberget iron ore, northern Sweden. Geochimica et Cosmochimica Acta 60, 1951–1961.
Romer, R. L. & Smeds, S.-A. (1996). U–Pb columbite ages of pegmatites from Sveconorwegian terranes in southwestern Sweden. Precambrian Research 76, 15–30.
Rötzler, J. (1992). On the petrogenesis in the Saxon granulite massif (in German, with English abstract). Geotektonische Forschungen 77, 1–100.
Rötzler, J. & Romer, R. L. (2001). P–T–t evolution of ultrahigh-temperature granulites from the Saxon Granulite Massif, Germany. Part I: Petrology. Journal of Petrology 42, 1995–2013.
Rötzler, J., Budzinski, H. & Budzinski, G. (1994). Evidence for Early Variscan very high temperature (>1000°C)–high pressure metamorphism of garnet–clinopyroxene granulites in the Saxon granulite massif, Germany. European Journal of Mineralogy 6, Supplement 1, 234.
Santos Zalduegi, J. F., Schärer, U., Ibarguchi, J. I. G. & Girardeau, J. (1996). Origin and evolution of the Paleozoic Cabo Ortegal ultramafic–mafic complex (NW Spain): U–Pb, Rb–Sr and Pb–Pb isotope data. Chemical Geology 129, 281–304.
Schärer, U. (1984). The effect of initial 230Th disequilibrium on young U–Pb ages: the Makalu case, Himalaya. Earth and Planetary Science Letters 67, 191–204.[Web of Science]
Schenk, V. & Todt, W. (1983). U–Pb Datierungen an Zirkon und Monazit der Granulite im Moldanubikum Niederösterreichs (Waldviertel). Forschritte der Mineralogie 61, Beiheft 1, 190–191.
Smith, H. A. & Giletti, B. J. (1997). Lead diffusion in monazite. Geochimica et Cosmochimica Acta 61, 1047–1055.
Steiger, R. H. & Jäger, E. (1977). Subcommission of geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters 36, 359–362.[Web of Science]
Stüwe, K. & Sandiford, M. (1995). Mantle–lithospheric deformation and crustal metamorphism with some speculations on the thermal and mechanical significance of the Tauern Event, Eastern Alps. Tectonophysics 242, 115–132.
Tilton, G. R. (1973). Isotopic lead ages of chondritic meteorites. Earth and Planetary Science Letters 19, 321–329.
van Breemen, O., Aftalion, M., Bowes, D. R., Dudek, A., Misar, Z., Povondra, P. & Vrana, S. (1982). Geochronological studies of the Bohemian Massif, Czechoslovakia, and their significance in the evolution of Central Europe. Transactions of the Royal Society of Edinburgh, Earth Sciences 73, 89–108.
Vance, D. & Holland, T. B. H. (1993). A detailed isotopic and petrological study of a single garnet from the Gassetts Schist, Vermont. Contributions to Mineralogy and Petrology 114, 101–118.
Vavra, G., Gebauer, D. & Schmid, R. (1994). Zircon growth and recrystallization during granulite facies metamorphism in the Ivrea Zone (Southern Alps): a combined cathodoluminescence and ion microprobe study. Journal of the Czech Geological Society 39(1), 114–116.
Vavra, G., Reinhardt, J., Todt, W. & Pidgeon, R. T. (1998). Dating of exhumation of a hot core complex (Saxonian Granulite Massif)—old prejudices and new perspectives of U–Pb dating of zircons and monazite. Freiberger Forschungshefte C471, 226–228.
Vavra, G., Reinhardt, J. & Pidgeon, R. T. (1999). Lower Carboniferous (340 Ma) HP/HT-metamorphism and rapid exhumation of the Saxonian Granulite Massif. Terra Nostra 99(1), 203.
Villa, I. M. (1998). Isotopic closure. Terra Nova 10, 42–47.
von Quadt, A. (1993). The Saxonian Granulite Massif: new aspects from geochronological studies. Geologische Rundschau 82, 516–530.
Walther, J. V. & Orville, P. M. (1982). Volatile production and transport in regional metamorphism. Contributions to Mineralogy and Petrology 79, 252–257.
Watson, E. B., Harrison, T. M. & Ryerson, F. J. (1985). Diffusion of Sm, Sr, and Pb in fluorapatite. Geochimica et Cosmochimica Acta 49, 813–823.
Weber, K. & Behr, H.-J. (1983). Geodynamic interpretation of the Mid-European Variscides. In: Martin, H. & Eder, F. W. (eds) Intracontinental Fold Belts. Berlin: Springer, pp. 427–469.
Zhang, L.-S. & Schärer, U. (1996). Inherited Pb components in magmatic titanite and their consequence for the interpretation of U–Pb ages. Earth and Planetary Science Letters 138, 57–65.
Zhou, B. & Hensen, B. J. (1995). Inherited Sm/Nd isotope components preserved in monazite inclusions within garnets in leucogneiss from East Antarctica and implications for closure temperature studies. Chemical Geology 121, 317–326.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Rotzler, B. Hagen, and S. Hoernes Geothermometry of the ultrahigh-temperature Saxon granulites revisited. Part I: New evidence from key mineral assemblages and reaction textures European Journal of Mineralogy, December 1, 2008; 20(6): 1097 - 1115. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Hagen, S. Hoernes, and J. Rotzler Geothermometry of the ultrahigh-temperature Saxon granulites revisited. Part II: Thermal peak conditions and cooling rates inferred from oxygen-isotope fractionations European Journal of Mineralogy, December 1, 2008; 20(6): 1117 - 1133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-J. Massonne, A. Kennedy, L. Nasdala, and T. Theye Dating of zircon and monazite from diamondiferous quartzofeldspathic rocks of the Saxonian Erzgebirge - hints at burial and exhumation velocities Mineralogical Magazine, August 1, 2007; 71(4): 407 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. ROTZLER, R. L. ROMER, H. BUDZINSKI, and R. OBERHANSLI Ultrahigh-temperature high-pressure granulites from Tirschheim, Saxon Granulite Massif, Germany: P-T-t path and geotectonic implications European Journal of Mineralogy, December 1, 2004; 16(6): 917 - 937. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Timmerman Timing, geodynamic setting and character of Permo-Carboniferous magmatism in the foreland of the Variscan Orogen, NW Europe Geological Society, London, Special Publications, January 1, 2004; 223(1): 41 - 74. [Abstract] [PDF]
|
|
|
|
 |
|
 M. Okrusch, R. Hock, U. Schussler, A. Brummer, M. Baier, and H. Theisinger Intergrown niobian rutile phases with Sc- and W-rich ferrocolumbite: An electron-microprobe and Rietveld study American Mineralogist, July 1, 2003; 88(7): 986 - 995. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. R. MARSCHALL, A. KALT, and M. HANEL P-T Evolution of a Variscan Lower-Crustal Segment: a Study of Granulites from the Schwarzwald, Germany J. Petrology, February 1, 2003; 44(2): 227 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Romer and J. Rotzler Effect of metamorphic reaction history on the U-Pb dating of titanite Geological Society, London, Special Publications, January 1, 2003; 220(1): 147 - 158. [Abstract] [PDF]
|
|
|
|
|
|
J. ROTZLER and R. L. ROMER P-T-t Evolution of Ultrahigh-Temperature Granulites from the Saxon Granulite Massif, Germany. Part I: Petrology J. Petrology, November 1, 2001; 42(11): 1995 - 2013. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||