Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.geol.msu.ru/deps/petro/Perchuk_publ/perchuk-Geol99.pdf Äàòà èçìåíåíèÿ: Mon Apr 22 18:43:13 2013 Äàòà èíäåêñèðîâàíèÿ: Thu Feb 27 23:30:18 2014 Êîäèðîâêà:

Rates of thermal equilibration at the onset of subduction deduced from diffusion modeling of eclogitic garnets, Yukon-Tanana terrane, Canada
Alexei Perchuk
IGEM, Russian Academy of Sciences, Staromonetny 35, Moscow, 109017, Russia, and CNRS-ESA 7058, Laboratoire de PÈtrologie, UniversitÈ Paris 6, T26-E3, 4 place Jussieu, 75005 Paris, France

Pascal Philippot
CNRS-ESA 7058, Laboratoire de PÈtrologie, UniversitÈ Paris 6, T26-E3, 4 place Jussieu, 75005 Paris, France

Philippe Erdmer
Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta T6G 2E3, Canada

Michel Fialin
CAMPARIS, UniversitÈ Paris 6, T26-E3, 4 place Jussieu, 75005 Paris, France ABSTRACT Well-preserved eclogitic rocks near Faro (Yukon-Tanana terrane, Canada) underwent a prograde evolution from ~510 °C and 1.1 GPa to 690 °C and 1.5 Gpa followed by nearisobaric cooling to ~540 °C. A remarkable feature of the garnet porphyroblast cores is the presence of minute garnet "inclusions" of distinctly different composition. Preservation of a sharp compositional gradient at the interface between the host and the inclusion garnets and results of diffusion modeling indicate that the counterclockwise pressure-time evolution took place on a very short time scale of about 0.2 m.y. Minimum rates of heating (950 °C/m.y.) and burial (7 cm/yr) calculated along the prograde part of the path are in good agreement with values extracted from thermal models of newly initiated subduction zones.

INTRODUCTION Considerable insight into the relationship between metamorphic rock pressure-time (P-T ) paths, thermal history, and tectonic evolution of orogenic belts has been obtained by combining metamorphic petrology techniques with thermal modeling (e.g., Oxburgh and Turcotte, 1974; England and Thompson, 1984). Garnets are of central importance in that respect, because they commonly display chemical and textural heterogeneities such as zonation or mineral inclusions that can be used to reconstruct quantitative P-T paths (Spear and Selverstone, 1983) and to calculate cooling rates of metamorphic terranes (e.g., Jiang and Lasaga, 1990). Applications of diffu-

sion modeling of growth-zoned garnet in the literature (e.g., Lasaga and Jiang, 1995; Perchuk and Philippot, 1997; Perchuk et al., 1998) have focused on reconstructing the retrograde P-T history following peak temperature of equilibration. Historical information on early stages of prograde P-T trajectories remains essentially unknown. This paper presents a new application of "geospeedometry" (Lasaga, 1983), allowing reconstruction of both the prograde and retrograde histories of an eclogite from Faro, in the northern Canadian Cordillera. We apply a diffusion model to remarkably minute Ca-rich garnet inclusions trapped within the core of growth-zoned eclogitic garnets. We show that the P-T trajectory is a

counterclockwise loop that relates to the postgrowth history of the Ca-rich garnet population, and we provide estimates of heating rates during burial. These results are examined in light of the modeled thermal state of accretionary wedges during early stages of subduction. GEOLOGIC OUTLINE The Yukon-Tanana terrane is the innermost suspect terrane of the western Canadian Cordillera. It includes both pericratonic rocks of likely North American affinity and allochthonous slices of volcano-sedimentary, oceanic, and granitic material inferred to have been emplaced onto the North American margin during Mesozoic collision (Tempelman-Kluit, 1979). In the Yukon, blueschist- and eclogite-facies rocks are known in at least six widely separated localities (see Erdmer et al., 1998). Although the relationships of high-pressure rocks with the country rocks are seldom clear, eclogite occurs generally as meter- to several hundred meter-sized lenses within siliceous or mafic and ultramafic rocks. The eclogitic lenses and their surrounding rocks are thought to represent remnants of trench material that have undergone middle Paleozoic, late Paleozoic, or early Mesozoic subduction. In this paper, we focus on one of the eclogitic lenses near Faro (Erdmer and Helmstaedt, 1983). FARO ECLOGITES Textural observations Eclogite near Faro occurs in three lenses of distinct petrologic character surrounded by

Figure 1. Backscatteredelectron images showing the morphology of a Carich garnet inclusion preserving euhedral shape (arrowed) located in core (light gray) of a garnet porphyroblast.

20 µm
Data Repository item 9947 contains additional material related to this article. Geology; June 1999; v. 27; no. 6; p. 531­534; 4 figures.

531


blueschist-facies mica schist, micaceous gneiss, and quartzite. Samples for the present study were obtained from a single meter-sized occurrence (lens III of the nomenclature of Erdmer and Helmstaedt, 1983), the freshest eclogite among the three lenses. The eclogite displays a porphyroblastic texture characterized by euhedral garnet grains as much as 2 mm across homogeneously distributed within a fine-grained matrix of omphacite with minor quartz and rutile. The eclogite shows a foliation defined by the average flattening plane of omphacite and thin ribbons of quartz. Garnet contains inclusions of omphacite, clinozoisite, quartz, rutile, and titanite. A remarkable feature of garnet cores is the presence of tiny (20­50 µm in diameter) garnet grains displaying either a corroded outline when in direct contact with host garnet or a well-preserved euhedral shape where protected by mineral inclusions of clinozoisite and quartz (Fig. 1). This evidence suggests that the garnet inclusions represent an early generation predating the growth of the larger grains. The eclogite shows only local textural and mineral evidence of retrogression following peak equilibration. The main alteration feature is replacement of matrix omphacite and quartz by local segregations of clinozoisite ± paragonite, phengite, and amphibole. The segregations occur either as thin layers oriented parallel to the foliation plane or as randomly distributed "patches" overprinting the rock fabric. Garnet is surrounded by haloes of phengite and minor chlorite. All hydrous phases appear to have formed at the expense of an earlier anhydrous eclogite assemblage; they show no evidence of internal strain or preferred orientation and postdate the eclogite-facies foliation. This observation indicates that late-stage re-equilibration of the eclogite was controlled by a water-rich fluid. It is notable that no plagioclase + clinopyroxene symplectites at the expense of omphacite + quartz occur in partially altered domains. In

addition, Erdmer and Helmstaedt (1983) reported minute glaucophane needles randomly distributed within the omphacite groundmass. As for other hydrous phases, the glaucophane needles are devoid of strain and postdate the main eclogite-facies foliation. Mineral composition Mineral compositions were determined with a CAMECA SX50 electron microprobe at UniversitÈ Paris 6, CAMPARIS. Operating conditions were 15 kV accelerating potential, 10 nA sample current, and 10 s counting time. Host garnet shows prograde chemical zoning characterized by a marked increase in Mg, and decrease in Fe, and to a lesser extent Mn, from core to rim (Fig. 2A; chemical data provided upon request).1 In contrast, garnet inclusions are markedly enriched in Ca and depleted in Fe, and to a lesser extent Mg and Mn, compared to the host-garnet core composition (Fig. 2 A, B). In addition to the markedly different compositional trends, there is a distinction between the two types of garnets in the "plateau-like" shape of the garnet-inclusion chemical profiles. Although the composition of the garnet inclusions can vary (Fig. 2A), there is no apparent relationship between the composition of the inclusions, their morphology and their location within the garnet host. Omphacite in the matrix is generally jadeiterich (up to 47 mol%) with low acmite (up to 5 mol%) contents. Omphacite inclusions in garnet cores are generally enriched in acmite, whereas inclusions in garnet margins are almost acmite free and contain higher jadeite and magnesium, similar to matrix omphacite.
1GSA Data Repository item 9947, Chemical analysis, mathematical treatment of the diffusion model and uncertainties in calculated time-scale values, is available on request from Documents Secretary, GSA, P.O. Box 9140, Boulder CO 80301, editing@geosociety.org, or at www.geosociety.org/pubs/drpint.htm.

2.0

A
Fe

1.5 Fe

1.5

1.0 garnet inclusion cations 1.0 Ca 0.5 0.0 Mg cations

Ca

METAMORPHIC EVOLUTION Temperatures and pressures during garnet growth were estimated by using omphacite inclusions in garnet cores and rims, under the assumption that omphacite inclusions are in equilibrium with garnet. Calculations performed using the garnet-clinopyroxene Fe-Mg exchange equilibria of Krogh (1988) and the jadeite content of omphacite expression of Perchuk (1992) indicate a temperature increase of 510 ± 50 °C to 680 ± 50 °C for a minimum pressure increase from 1.1 ± 0.2 GPa to 1.5 ± 0.2 Gpa, respectively, during garnet growth. Similar equations applied to coexisting matrix omphacite and outer garnet rims yielded a temperature of 690 ± 40 °C at a minimum pressure of 1.4 ± 0.2 GPa. With respect to the retrograde path, the absence of clinopyroxene and plagioclase-bearing symplectite after omphacite + quartz strongly suggests that the rocks remained within the P-T stability field of matrix omphacite (Jd > 47%) + quartz. Because of the presence of hydrous minerals as reaction products, indicating that H2O was present as a catalyst, unfavorable kinetics are unlikely to account for the absence of clinopyroxene + plagioclase. An indication of the temperature of equilibration of the retrograde assemblage is given by Erdmer et al. (1998), who showed that the temperature of equilibration of Faro eclogites containing clinozoisite ranged between 420 and 490 °C at 1.5 GPa. Another indication comes from the stability field of glaucophane (i.e., < 540 °C at pressures of > 1 GPa; Maresch, 1977). These pressure-temperature constraints indicate a path (Fig. 3) in which the rocks reached peak P-T conditions following a progressive increase of both temperature and pressure. The shape of the trajectory after peak temperature is best approximated by a near-isobaric cooling curve extending from above the Jadeite 47 mol% isopleth down to ~540 °C. To convert the P-T path shown in Figure 3 to a pressure-temperature-time (P-T-t) path, the kinetics of diffusion in garnet must be explicitly treated. The sharp compositional gradients between the garnet inclusions and the garnet host (Fig. 2 A, B) are used below to derive rates of pressure and temperature variation along the reconstructed P-T trajectory. The latest part of the decompressional path, from about 500 °C and 1.5 GPa back to Earth's surface, shown as a dashed line in Figure 3, is taken from Erdmer et al. (1998); that P-T segment is not considered here. GARNET INCLUSIONS AS CLOCKS The specific mathematical treatment of the diffusion model in a sphere has been examined in detail elsewhere (see Perchuk and Philippot, 1997). The garnet inclusion­garnet host interface was modeled as a "stair-like" distribution function (see Perchuk et al., 1998; details of the mathematical treatment and of the uncertainties in the calculated duration of metamorphic event provided upon request). As compositional gradients
GEOLOGY, June 1999

0.5 Mn

B
0

Mg 20 40 distance, µm 60

Mn 0.0 0 200 400 600 800 1000 distance, µm

Figure 2. Compositional zoning profiles plotted in number of cations (Fe, Mg, Ca and Mn). A: Garnet host. B: Garnet inclusion.

532


identified along the garnet inclusion­garnet host boundary are mainly a function of grossularalmandine exchange, homogenization of garnet can be treated as a simple binary diffusion process. Because Ca displays the highest chemical gradient (Fig. 2B) and shows the smallest diffusion coefficient among divalent cations (Schwandt et al., 1996), it is considered to be the component controlling the rate of diffusion. Results of the calculation are shown in Figure 4, which plots grossular content as a function of the radial position in garnet. The curves represent the shape of the grossular concentration profile along the garnet interface calculated for a garnet-host radius of 550 µm and for different durations (0.2 and 1 m.y.) of metamorphic events; the calculations used the Ca diffusion coefficients of Schwandt et al. (1996). The plot shows that the grossular profile along the garnet interface can be preserved for a metamorphic history spanning about 0.2 m.y. (log t < ­0.7 ± 1.4). DISCUSSION In making a geodynamic interpretation of the P-T-t trajectory, it is necessary to stay within the constraints of model P-T-t relations that are provided by thermal simulation of convergent plate junctions (e.g., Oxburgh and Turcotte, 1974; Cloos, 1985; Peacock, 1987). All approaches share the topology of a downbowing of the isotherms near the top surface of the subducted slab. At initiation of subduction, heat must be rapidly withdrawn from the hanging wall and conducted into the descending slab, generating moderate P-T gradients in the underlying wedge. Ongoing plate convergence will increase depression of the isotherms and lead to a gradual cooling of the subducted slab over time until a steady-state thermal configuration is reached. Accordingly, because of the hot hanging wall, material that is subducted early will follow a P-T trajectory at a higher temperature for a given pressure compared to material subducted later. Numerical modeling by Cloos (1982) of flowmÈlange systems predicted that retrograde P-T paths for material in a mature mÈlange zone should be similar to the burial path. In contrast, because of the relatively high temperatures produced at the onset of subduction, the return P-T path for material in an immature system can be characterized by an initial stage of isobaric cooling attending continuous refrigeration of the hanging wall; material stored at depth experiences cooling without a decrease in pressure. The exhumation path of such early-accreted material will be at lower temperatures for a given pressure than the burial path. The counterclockwise P-T trajectory determined for the Faro eclogite is in agreement with this scenario. We suggest that the increase in both temperature and pressure revealed by the Faro eclogite was produced by proximity to the hot hanging wall during the initiation of subduction. As the blocks were stored at depth and the hangGEOLOGY, June 1999

C/km

1.8

Gln in

10

°

+Qtz Jd 45

Ab

high

?
1.4

50 km
+Qtz Jd 20

P, GPa

Ab

high

1.0
Ab
low
15
°C/km

Grtcore-Cpxinclusion Grtrim-Cpxinclusion Grtrim-Cpxmatrix

30 km

0.6 400 500 600 T, ° C 700 800

Figure 3. Pressure-temperature diagram showing counterclockwise P-T path for Faro eclogites determined from geothermobarometry and phase-equilibrium analysis. P-T estimates determined from garnet-clinopyroxene (Grt-Cpx) geothermometer of Krogh (1988) and the jadeite content of omphacite (Perchuk, 1992) are shown for omphacite inclusions and garnet cores (Grt core- Cpxinclusion , filled circles), omphacite inclusions and garnet rims (Grt rim- Cpxinclusion , open circles), and edge compositions of matrix phases (Grt rim- Cpxmatrix , filled diamonds). Decompressional P-T trajectory (dashed line) is from Erdmer et al. (1998). Abbreviations: Jd45 = omphacite with 45% of jadeite component, Qtz = quartz, Ab = albite. Gln in refers to stability field of glaucophane (after Maresch, 1977). Two geothermal gradients at 10 and 15 °C/km are shown.

ing wall cooled, the relatively high temperatures were succeeded by uniformly lower temperatures down to about 540 °C, resulting in the isobaric P-T segment shown on the P-T trajectory (Fig. 3). The diffusion model presented here provides quantitative constraints on the burial-heating and cooling rates experienced by the Faro eclogite. As shown above, the duration of the prograde and early isobaric cooling events would not have exceeded 0.2 m.y. Although the relative duration of these two events is unknown, it is possible to determine critical values of burial-heating and cooling rates above which the garnet-interface chemical profile cannot be preserved. Assuming that the Faro eclogite experienced a continuous P-T evolution, and considering a duration of 0.2 m.y., a P-T change of 0.4 GPa and 180 °C during the prograde metamorphic event yields minimum rates of burial and heating of 7 cm/yr and 950 °C/m.y., respectively. Conversely, a temperature change of 150 °C and a duration of 0.2 m.y. for the isobaric cooling event results in a minimum cooling rate of 750 °C/m.y. The minimum rate of burial determined above is consistent with rates of plate convergence of

modern subduction zones. The heating rate is, to our knowledge, the first estimate obtained from high-pressure rocks. As such, it can only be compared with values extracted from thermal models of newly initiated subduction zones. Peacock (1987) and Peacock et al. (1994) provided a twodimensional model of the thermal structure of the downgoing slab during initiation of subduction. Using a 10 cm/yr convergence rate, 0.2 m.y. of convergence and a 10 °C.km­1 initial oceanic geotherm yielded heating rates ranging between 400 to 1300 °C/m.y. (see Figures 6 and 7 of Peacock, 1987 and Figure 9D of Peacock et al., 1994), which compares well with our estimates. With regard to the isobaric cooling history, transient thermal models of subduction zones predict that typical maximum cooling rates of the hanging wall are ~100 °C/m.y. at the initiation of subduction (Figure 9A in Peacock et al., 1994). Considering that the counterclockwise P-T trajectory shown in Figure 3 occurred in less than 1 m.y. indicates that the Faro rocks experienced cooling rates slower than their heating rates, which also agrees with results from the diffusion modeling. The range of heating rates calculated for the Faro
533


38

0.2 m.y.
34

0 m.y. 1 m.y.

Grossular, mol%

30

ACKNOWLEDGMENTS A. Perchuk acknowledges the tenure of a fourmonth invited professor position at UniversitÈ Paris 7. We thank Simon Peacock and Craig S. Schwandt for their constructive reviews. This work benefited from an Institut National des Sciences de l'Univers grant to P. Philippot (Contribution 98IT172) and Russian Foundation for Basic Research grant 97-05-64418 to A. Perchuk. Field work was supported by a Natural Sciences and Engineering Research Council of Canada grant to P. Erdmer. REFERENCES CITED Christensen, J. N., Rosenfeld, J. L., and DePaolo, D. J., 1989, Rates of tectonometamorphic processes from rubidium and strontium isotopes in garnet: Science, v. 244, p. 1465­1469. Cloos, M., 1982, Flow melanges: Numerical modelling and geologic constraints on their origin in the Franciscan subduction complex, California: Geological Society of America Bulletin, v. 93, p. 330­345. Cloos, M., 1985, Thermal evolution of the convergent plate margins: Thermal modeling and reevaluation of isotopic Ar-ages for blueschists in the Franciscan complex of California: Tectonics, v. 4, p. 421­433. England, P. C., and Thompson, A. B., 1984, Pressuretemperature-time paths of regional metamorphism I. Heat transfer during the evolution of regions of thickened continental crust: Journal of Petrology, v. 25, p. 894­928. Erdmer, P., and Helmstaedt, H., 1983, Eclogite from central Yukon: A record of subduction at the western margin of ancient North America: Canadian Journal of Earth Sciences, v. 20, p. 1389­1408. Erdmer, P., Ghent, E. D., Archibald, D. A., and Stout, M. Z., 1998, Paleozoic and Mesozoic high-pressure metamorphism at the margin of the ancestral North America in central Yukon: Geological Society of America Bulletin, v. 110, p. 615­629. Jiang, J., and Lasaga, A. C., 1990, The effect of postgrowth thermal events on the growth-zoned garnet: Implications for metamorphic P-T history calculations: Contributions to Mineralogy and Petrology, v. 105, p. 454­459. Krogh, E. G., 1988, The garnet-clinopyroxene Fe-Mg thermometer--Reinterpretation of existing experimental data: Contributions to Mineralogy and Petrology, v. 99, p. 44­48. Lasaga, A. C., 1983, Geospeedometry: An extension of geothermometry: Advances in Physical Geochemistry, v. 3, p. 81­114. Lasaga, A. C., and Jiang, J., 1995, Thermal history of rocks: P-T-t paths from the geospeedometry, petrologic data, and inverse theory of techniques: American Journal of Science, v. 295, p. 697­741.

26

22

0

10

20 30 distance, µm

40

Figure 4. Relaxation profiles of garnet inclusion­garnet host interface calculated for different time scales (0.2 and 1.0 m.y.) as a function of counterclockwise P-T trajectory shown in Figure 3. Initial (t = 0 m.y.) growth zoning profile was defined as a perfect step. Garnetinterface profile (plotted in mol% grossular, filled circles) is best described by relaxation profile labeled 0.2 m.y.

eclogites is in striking contrast with the range of 5 to 10 °C/m.y. deduced by Christensen et al., (1989) for regional metamorphic garnets. This suggests that the rates of metamorphism in subduction zones may be two orders of magnitude more rapid than rates of metamorphism in continent collision belts--an observation that clearly deserves further investigation. One final inference concerns the origin of the Ca-rich garnet inclusions. Because garnet strongly partitions Fe and Mn relative to other divalent cations, the fact that the garnet inclusions have higher Ca and lower Mn and Fe contents than the garnet host indicates that the bulk composition of the sample may have changed significantly prior to the high-pressure event. Although detailed chemical and textural analyses are necessary to elucidate the formation and preservation of the garnet inclusions (our work in progress), we suggest that the inclusions could represent relicts of an earlier metamorphic event affecting the Faro rocks. This possibility further supports the notion that the history of Cordilleran convergence is one of a succession of short and perhaps superposed subduction episodes (e.g., Erdmer et al., 1998) rather than of a single, prolonged subduction event (e.g., Cloos, 1985).

Maresch, W. W., 1977, Experimental studies of glaucophane: An analysis of present knowledge: Tectonophysics, v. 43, p. 109­125. Oxburgh, E. R., and Turcotte, D. L., 1974, Thermal gradients and regional metamorphism in the overthrust terrains with special reference to the eastern Alps: Schweizcerische Mineralogische und Petrographische and Mittceilungen, v. 54, p. 642­662. Peacock, S. M., 1987, Creation and preservation of subduction related inverted metamorphic gradients: Journal of Geophysical Research, v. 92, p. 12763­12781. Peacock, S. M., Rushmer, T., and Thompson, A. B., 1994, Partial melting of subducting oceanic crust: Earth and Planetary Science Letters, v. 121, p. 227­244. Perchuk, A. L., 1992, New variant of the omphacitealbite-quartz geobarometer, allowing the structural states of omphacite and albite: Akademii Nauk Doklady Rossiiskoi, v. 324, p. 1286­1189. Perchuk, A. L., and Philippot, P., 1997, Rapid cooling and exhumation of eclogitic rocks from the Great Caucasus, Russia: Journal of Metamorphic Geology, v. 15 , p. 299­310. Perchuk, A. L., Yapaskurt, V. O., and Podlesskii, S. K., 1998, Genesis and exhumation dynamics of eclogites in the Kokchetav massif near mount Sulu-Tyube: Geochemistry International, v. 36, p. 877­885. Schwandt, C. S., Cygan, R. T., and Westrich, H. R., 1996, Ca self-diffusion in grossular garnet: American Mineralogist, v. 81, p. 448­451. Spear, F. S., and Selverstone J., 1983, Quantitative P-T paths from zoned minerals: Theory and tectonic applications: Contributions to Mineralogy and Petrology, v. 83, p. 348­357. Templeman-Kluit, D., 1979, Transported cataclasite, ophiolite and granodiorite in Yukon; evidence of arc-continent collision: Geological Survey of Canada Paper 79-14, 28 p. Manuscript received October 23, 1998 Revised manuscript received February 16, 1999 Manuscript accepted March 5, 1999

534

Printed in U.S.A.

GEOLOGY, June 1999