THE PROBLEM OF DEPTH IN GEOLOGY: WHEN PRESSURE DOES NOT TRANSLATE INTO DEPTH
© 2013 E. Moulas", Y. Y. Podladchikov4, L. Y. Aranovichc,d, D. Kostopoulos6
aDepartment of Earth Sciences, ETH-Zürich, Sonneggstrasse 5, 8092 Zürich, Switzerland; e-mail: evangelos.moulas@erdw. ethz.ch bInstitute of Earth Sciences, University of Lausanne, CH-1015 Lausanne, Switzerland cInstitute of the Geology of Ore Deposits, Petrography, Mineralogy, and Geochemistry, Russian Academy of Sciences
Staromonetnyi per., 35, Moscow, 109017, Russia; e-mail: lyaronov@igem.ru dDepartment of Geology, University of Johannesburg, Auckland Park, Johannesburg 2006, South Africa eFaculty of Geology and Geoenvironment, National and Kapodistrian University of Athens, 15784 Greece
Received March 24, 2013
Abstract — We review published evidence that rocks can develop, sustain and record significant pressure deviations from lithostatic values. Spectroscopic studies at room pressure and temperature (P-T) reveal that in situ pressure variations in minerals can reach GPa levels. Rise of confined pressure leads to higher amplitude of these variations documented by the preservation of a-quartz incipiently amorphized under pressure (IAUP quartz), which requires over 12 GPa pressure variations at the grain scale. Formation of coesite in rock-deformation experiments at lower than expected confined pressures confirmed the presence of GPa-level pressure variations at elevated temperatures and pressures within deforming and reacting multi-mineral and poly-crystalline rock samples. Whiteschists containing garnet porphyroblasts formed during prograde metamorphism that host quartz inclusions in their cores and coesite inclusions in their rims imply preservation of large differences in pressure at elevated pressure and temperature. Formation and preservation of coherent cryptoperthite exsolution lamellae in natural alkali feldspar provides direct evidence for grain-scale, GPa-level stress variations at 680°C at geologic time scales from peak to ambient P-T conditions. Similarly, but in a more indirect way, the universally accepted 'pressure-vessel' model to explain preservation of coesite, diamond and other ultra-high-pressure indicators requires GPa-level pressure differences between the inclusion and the host during decompression at temperatures sufficiently high for these minerals to transform into their lower pressure polymorphs even at laboratory time scales. A variety of mechanisms can explain the formation and preservation of pressure variations at various length scales. These mechanisms may double the pressure value compared to the lithostatic in compressional settings, and pressures up to two times the litho-static value were estimated under special mechanical conditions. We conclude, based on these considerations, that geodynamic scenarios involving very deep subduction processes with subsequent very rapid exhumation from a great depth must be viewed with due caution when one seeks to explain the presence of microscopic ultrahigh-pressure mineralogical indicators in rocks. Non-lithostatic interpretation of high-pressure indicators may potentially resolve long-lasting geological conundrums.
DOI: 10.7868/S0869590313060058
INTRODUCTION
Since the pioneering publications by Eskola (1929) and Korzhinskii (1940; 1959) depth has become one of the major parameters to be considered in hard-rock geology. Korzhinskii (1940) based his "facies of depth" systematics on the succession of carbonation reactions in Ca-rich rocks. In the 1960's, pressure in metamorphic environments was constrained qualitatively through applying simple mineral equilibria that define stability of a few major index minerals (e.g., Ernst, 1963). After geothermobarometry was first introduced into petrology by Perchuk (1968; 1970), the quantitative determination of pressure-temperature conditions has become routine in metamorphic studies. Conventionally, pressure and depth are thought of
as synonymous due to the application of Archimedes' formula as a first-order approximation for the calculation of depth at the base of a rock column using pressure as input. This formula was derived from the force balance for static fluids in the gravity field and relates pressure with depth via:
P = pgh, (1)
where P is the pressure, p is the density, g is the acceleration of gravity and h is the height of the hydrostatic column (depth). In deforming regions of the lithosphere (at any length scale), stresses that develop due to deformation will have an additional effect on the value of pressure. In that case, we would have deviations from the lithostatic pressure formula Eq. (1) that can
0.5
0.4
<£ 0.3 ü
0.2
0.1
-
\-X
- 9--t )-
■1 >
0.1
0.2
0.3 0.4 P464, GPa
0.5
Fig. 1. Pressure dependence of the Raman spectra of quartz used to evaluate residual pressures from quartz inclusions in garnet from the Greek Rhodope. The quartz inclusions are ca 10 p,m in diameter and are enclosed in garnets with a radius of ca 0.5 mm. The shift of the two major Raman bands (206 and 464 cm-1) relative to an unstressed quartz single crystal is used to estimate pressure (P464 and P206). The calibration used is after Schmidt and Ziemann (200O). Note that the maximum pressure recorded in this sample (ca 0.5 GPa) roughly corresponds to the pressure of a rock situated in the middle crust if Eq. (1) is used.
be positive (overpressure) or negative (under-pres-sure). The magnitude of these deviations is crucial if depth is estimated from pressure and for this reason quantitative studies and different mechanical scenarios have to be considered.
In this paper we review published data on the pressure variations recorded by metamorphic rocks. We subsequently present possible mechanisms that can be responsible for the formation and preservation of pressure variations in rocks and discuss the feasibility of predicting burial depth from pressure estimates.
PRESSURE INFORMATION FROM MINERAL ASSEMBLAGES
High-pressure mineral polymorphs like diamond and coesite are used as diagnostic features for the former existence of ultrahigh-pressure (UHP) mineral assemblages. These polymorphs are commonly found as inclusions in other minerals such as zircon, garnet and clinopyroxene (e.g. Sobolev and Shatsky, 1990; Chopin and Sobolev, 1995). The development of a pressure differential between inclusion and host has been used in fluid inclusion studies (e.g., Roedder and Kopp, 1975; Yamamoto and Kagi, 2008), and also to explain the preservation of coesite in metamorphic
rocks (e.g. Gillet et al., 1984; Smith, 1984; Van der Molen and Van Roermund, 1986). In certain cases, high residual pressures (up to 1—2 GPa) are still present in rocks at ambient conditions (e.g. Parkinson and Katayama, 1999; Parkinson, 2000; Ye et al., 2001; Nasdala et al., 2003; Enami et al., 2007; Howell et al., 2010). By using the pressure-dependent shift in the Raman spectra of coesite Parkinson (2000) was able to show that those coesite inclusions that did not display rupturing of the host garnet may be still subject to confining pressure of around 2 GPa. Employing the same method, Sobolev et al. (2000) estimated a confining pressure of 3.5 GPa for coesite inclusions in diamonds. Enami et al. (2007) demonstrated the existence of residual pressure by considering the pressure dependence of the Raman spectra of quartz (e.g., Fig. 1). Apart from the pressure variations documented for inclusions in garnet, the formation of chemically zoned plagioclase around kyanite has been used by TajC-manova et al. (2013; submitted) to explain the preservation of kyanite at low pressures and high temperatures, outside its stability field. The reason for these high residual pressures is that the host mineral acts as a "pressure vessel" inhibiting decompression of the inclusion (e.g., Chopin, 2003). The common assumption in such models is that the host phase had the same pressure and temperature (P-T) as the included phase at the time of enclosure.
Residual pressures are preserved not only during ascent and exhumation, as is commonly assumed. Parkinson (2000) documented the preservation of quartz crystals in garnet cores in a whiteschist from the Kokchetav massif (Kazakhstan) and of coesite crystals in their rims during peak metamorphism at UHP conditions (Fig. 2). Robin (1974) demonstrated that the stresses developed during the formation of coherent exsolution within cryptoperthitic alkali feldspars at approximately 680°C are of the order of GPa. These stresses were preserved after the decompression and cooling of the samples as revealed by TEM studies (Robin, 1974).
Palmeri et al. (2009) reported the finding of a-quartz incipiently amorphized under pressure (IAUP quartz) as inclusion in omphacite from eclog-ites of Lanterman Range, Antarctica, where regional high-pressure estimates were of the order of 3.3 GPa (Palmeri et al., 2007). The presence of IAUP quartz suggested a pressure range for the inclusion between 15 and 32 GPa (Palmeri et al., 2009; Godard et al., 2011).
Pressure variations are not specific only for inclusion-host relationships. Coesite has been experimentally produced in the intergranular region of deforming quartzite (Fig. 3) at confining pressures below its stability field under tri-axial stress conditions (confining pressure ct2 = ct3 ~ 1.2 GPa, T = 973 K; Hirth and Tullis, 1994).
(a)
(b)
5 mm
a P
Ü S2
3
1
200 400 600 800 Temperature, °C
Fig. 2. (a) — Sketch of garnet porphyroblast from a Kokchetav Massif whiteschist, showing the distribution of quartz and coesite inclusions in different growth zones (after Parkinson, 2000, simplified). (b) — P-T estimates by Parkinson (2000) place the formation of garnet core at P < 1 GPa and T = 380°C. Higher P-T conditions were estimated for the coesite-bearing garnet rims (P = 3.4—3.6 GPa, and T = 720—760°C). The persistence of monomineralic quartz in the coesite stability field (P > 2.7 GPa) suggests a mi
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.