| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology | Volume 45 | Number 2 | Pages 253-273 | 2004
© Oxford University Press 2004; all rights reserved
Thermal History of the Horoman Peridotite Complex: A Record of Thermal Perturbation in the Lithospheric Mantle
DEPARTMENT OF EARTH AND PLANETARY SCIENCE, GRADUATE SCHOOL OF SCIENCE, UNIVERSITY OF TOKYO, TOKYO 113-0033, JAPAN
RECEIVED NOVEMBER 27, 2002; ACCEPTED SEPTEMBER 19, 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
The ascent history of the Horoman peridotite complex, Hokkaido, northern Japan, is revised on the basis of a detailed study of large ortho- and clinopyroxene grains
1 cm in size (megacrysts) in the Upper Zone of the complex. The orthopyroxene megacrysts exhibit distinctive M-shaped Al zoning patterns, which were not observed in porphyroclastic grains less than 5 mm in size described in previous studies. Moreover, the Al and Ca contents of the cores of the orthopyroxene megacrysts are lower than those of the porphyroclasts. The Upper Zone is inferred to have resided not only at a higher temperature than previously suggested but also at a higher pressure (1070°C, 2·3 GPa) than the Lower Zone (950°C, 1·9 GPa), in the garnet stability field, before the ascent of the two zones. The Horoman complex probably represents a 12 ± 5 km thick section of lithospheric mantle with an 10 ± 8°C/km vertical thermal gradient. The current thickness of the Horoman complex is 3 km, which is a result of shortening of the lithospheric mantle by 0·25 ± 0·1 during its ascent. The Upper Zone appears to have experienced a heating event during its ascent through the spinel stability field, with a peak temperature as high as 1200°C. The effect of heating decreases continuously towards the base of the complex, and the lowermost part of the Lower Zone underwent very minor heating at a pressure higher than 0·5 GPa. The uplift and associated deformation, as well as heating, was probably driven by the ascent of a hot asthenospheric upper-mantle diapir into the Horoman lithosphere. KEY WORDS: Horoman; P–T trajectory; thermal history; Al diffusion in pyroxene; geothermobarometry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
Records of the pressure–temperature history preserved in mantle peridotites provide important information to constrain the ascent rate of mantle materials and the mechanism of their emplacement into the lower crust. Particularly valuable is the spatial variation of P–T history preserved in a single peridotite body (e.g. MacGregor & Basu, 1976
; Obata, 1980; Ozawa & Takahashi, 1995), which provides information on the dynamics of heat and material transfer from the mantle to the crust when combined with geological mapping of the complex (Nicolas, 1989; Sawaguchi & Takagi, 1997).
Three types of P–T history (or trajectory) of mantle materials initially located at depths as deep as the garnet stability field have been recognized for mantle peridotites on the basis of petrological and mineralogical studies. The first type features nearly adiabatic ascent at high temperature and pressure, followed by a rapid drop in temperature at a certain depth (Obata, 1980; Tubía, 1994; Ozawa & Takahashi, 1995; Van Roermund et al., 2000; Drury et al., 2001; Paquin & Altherr, 2001a). The second type is characterized by a non-adiabatic ascent path [greater dT/dP than the first type (Kornprobst, 1969; Obata, 1980; Davies et al., 1993; Yang et al., 1993; Kadarusman & Parkinson, 2000; Tsai et al., 2000)], which suggests effective heat loss from the beginning of the ascent history. The third type involves a temperature rise by heating at a shallower depth, subsequent to either an adiabatic or a non-adiabatic ascent path (Van der Wal & Vissers, 1993; Ozawa, 1997; Nimis & Trommsdorff, 2001). More complex P–T trajectories, which can be combinations of the above three types, have also been proposed (Loubet & Allègre, 1979; Van der Wal & Vissers, 1993; Brenker & Brey, 1997). Such differences in the ascent history should reflect differences in the mode of thermo-mechanical interaction of the uplifting mantle with the adjacent lithospheric or asthenospheric mantle and lower crust. They could also indicate the thermal effects of a melt segregated from the asthenosphere or much deeper mantle.
The Horoman peridotite complex is ideal for study of such thermo-mechanical interactions. The complex is distinguished by its exceptional freshness, with very minor degrees of low-temperature alteration, and the presence of symplectic aggregates inferred to be a breakdown product of garnet (Kushiro & Yoder, 1966; Tazaki et al., 1972). These two features are closely related to its ascent history, particularly to the ascent rate or the cooling rate.
Early workers noted that the Horoman peridotite complex consists of upper and lower units with contrasting features in terms of lithology, structure and olivine fabric; these are called the Upper Zone and Lower Zone, respectively (Komatsu & Nochi, 1966; Niida, 1975, 1984; Table 1). Ozawa & Takahashi (1995) studied the thermal history of the Horoman complex for the first time. They examined the chemical zoning of orthopyroxene porphyroclasts (up to 5 mm) in plagioclase lherzolites and discovered M-shaped Al zoning patterns in the Lower Zone, but they found only bell-shaped Al zoning patterns in the Upper Zone. The Al contents of the cores of the bell-shaped orthopyroxenes from the Upper Zone are higher than the Al contents in the cores of M-shaped orthopyroxenes in the Lower Zone. Ozawa & Takahashi (1995) also noted that the Ca contents of the orthopyroxene cores gradually increase from the base to the top of the complex. These mineral chemical data clearly demonstrate that the Upper Zone followed a higher-temperature decompression path than the Lower Zone.
|
Ozawa & Takahashi (1995)
argued that the decompression path was nearly adiabatic up to the plagioclase–spinel facies boundary followed by a rapid cooling along the boundary and that the differences in the temperature of the ascent paths were principally due to differences in the initial temperatures within the garnet stability field. The initial depth of equilibration for the two zones was constrained at 2 GPa, but this was only a minimum estimate. The most important observation of the absence of M-shaped Al zoning patterns in orthopyroxene porphyroclasts with grain size up to 5 mm from the Upper Zone was simply explained by its initial higher temperature condition and subsequent adiabatic ascent path, which allowed the marginal Al peak in the zonation pattern to flatten out. In contrast, the M-shaped Al zoning patterns preserved in the Lower Zone were explained by a lower initial temperature and low-T adiabatic exhumation path.
Takazawa et al. (1996) estimated P–T conditions for the ‘primary’ and ‘secondary’ stages for peridotites from the upper horizon of the Lower Zone by applying several geothermobarometers to pyroxene cores and neoblasts. They estimated a ‘primary’ temperature 100°C higher than that of Ozawa & Takahashi (1995), even if the location of the horizon studied by Takazawa et al. (1996) is taken into consideration. The discrepancy must be reconciled to better constrain the absolute P–T conditions of equilibration in the mantle before the ascent of the Horoman complex.
In a subsequent paper, Ozawa (1997) modeled the development of M-shaped Al zoning patterns in the Lower Zone orthopyroxene porphyroclasts and demonstrated that the difference in orthopyroxene zoning patterns between the Upper and Lower Zones could be produced by two different model exhumation paths. The first involves an adiabatic decompression path for the Lower Zone with a significantly different initial temperature from the Upper Zone, in the garnet stability field. The second involves similar starting conditions in the garnet field, subsequently followed by adiabatic decompression with a temperature rise within the spinel field only for the Upper Zone. The second model provides an alternative way to explain not only the presence of M-shaped Al zoning patterns in orthopyroxene from the Lower Zone but also the common occurrence of a marginal hump in the Ca content of the orthopyroxene porphyroclasts in the Upper Zone, which is too high to be induced by an adiabatic path.
Interpretation of M-shaped Al zoning patterns in orthopyroxene from the Alpe Arami garnet peridotite body (Paquin & Altherr, 2001a) has recently provoked a controversy. Paquin & Altherr (2001b) argued for a nearly isothermal decompression path from the peak metamorphic conditions of 1180°C and 5·9 GPa followed by a rapid cooling at 1 GPa, whereas Nimis & Trommsdorff (2001) argued against this model and proposed a two-stage P–T history whereby the Alpe Arami body was equilibrated at 800°C and 3 GPa and was subsequently subjected to a short-lived thermal event at a temperature higher than 1100°C.
It is, therefore, still debatable if the apparent difference in the thermal history of the Upper and Lower Zones of the Horoman complex is due to a difference in the initial temperature within the garnet stability field or if the Upper Zone went through a subsequent heating event within the spinel stability field. The argument comes from ambiguity in the significance of the core composition of the orthopyroxene porphyroclasts, particularly those from the higher-temperature Upper Zone. The size of the largest porphyroclasts examined in previous studies (e.g. Ozawa & Takahashi, 1995) was less than 5 mm. Such porphyroclasts from the Upper Zone do not show the characteristic M-shaped Al profile and do not have a wide core region with uniform Al and Ca distribution. The size of 5 mm may not be large enough to have recorded any meaningful P–T conditions in the garnet stability field in the Upper Zone.
The purpose of this paper is further to constrain the ascent history of the Horoman peridotite complex by examining centimeter-sized orthopyroxene and clinopyroxene crystals (megacrysts) in plagioclase lherzolites of the Upper Zone. The new results suggest that the P–T history proposed by Ozawa & Takahashi (1995) should be revised. Moreover, this revised P–T trajectory allows discussion of the initial conditions in the Horoman peridotite complex mantle and the processes that controlled its ascent into the lower crust.
|
GEOLOGICAL BACKGROUND AND STUDIED SAMPLES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
The Horoman peridotite complex, Hokkaido, Japan is located at the southern end of the Hidaka metamorphic belt (Fig. 1), which is characterized by a low-P/high-T type of metamorphism in Paleogene to early Miocene times. The Hidaka belt represents an
23 km thick sequence including the upper-mantle and crustal section of a magmatic arc (Komatsu et al., 1983). The Horoman complex occupies the tectonically lowest horizon of the Hidaka belt and is approximately 8 km x 10 km in size.
|
The Horoman complex is stratified and its total thickness is estimated to be
3 km (Niida, 1984). Niida (1984) divided the complex into Upper and Lower Zones based on the contrasting geological characteristics of the two zones (Table 1). The Lower Zone is composed mainly of alternating layers of plagioclase lherzolite, spinel lherzolite (with or without two-pyroxene symplectites after garnet), and harzburgite with lesser amount of spinel-rich dunite–wehrlite layers. It shows cyclic layering whose half cycle starts from plagioclase lherzolite and ends in spinel-rich dunite, through symplectite-bearing spinel lherzolite, symplectite-free spinel lherzolite, and harzburgite (Fig. 1; Komatsu & Nochi, 1966; Niida, 1984; Obata & Nagahara, 1987; Takahashi, 1992). The lithological boundaries are gradational except for the sharp contact between dunite and harzburgite. The thickness of each cycle ranges from 200 to 600 m. Minor amounts of mafic layers (1·2% in volume; Takazawa et al., 1999) are also present.
The Upper Zone is composed of rock types similar to the Lower Zone, but plagioclase lherzolite is more abundant (Fig. 1). Mafic layers are more dominant in this zone (2·1 vol. %) as compared with the Lower Zone (Table 1; Takazawa et al., 1999). Plagioclase-rich segregation veins locally cut the plagioclase lherzolites, oriented parallel to the lineation but oblique to the foliation. The veins are composed mostly of Na-rich plagioclase with minor amount of Ti-rich pargasite and/or phlogopite and are absent in the lower 120 m above the boundary with the Lower Zone (Takahashi, 1997). This type of vein is frequently observed in the Nikanbetsu complex, located about 9 km to the east of the Horoman complex, which has lithological and petrographic features similar to the Upper Zone of the Horoman complex (Niida & Katoh, 1978).
Plagioclase lherzolite samples containing large pyroxene crystals (megacrysts) were collected from two horizons in the Upper Zone (Fig. 1). One locality is along the Horoman River, 200 m above the boundary with the Lower Zone. The other is on the North Ridge of Mt. Apoi, 700 m above the zone boundary. Two orthopyroxene and one clinopyroxene megacrysts were sampled from the Horoman River locality, and one orthopyroxene and one clinopyroxene from the North Ridge locality. The pyroxene megacrysts, which usually occur in a plagioclase-rich layer, are 1–2 cm in size (Fig. 2) with their long axes lying nearly parallel to the layering. They occur sporadically and never form clusters. The plagioclase-rich layer is often thinner than the pyroxene grains, showing a boudinage-like structure, and the megacrysts tend to have wide contact with the peridotite host.
|
Figure 2 shows the size of the studied pyroxene megacrysts in comparison with the size of the largest orthopyroxene porphyroclasts in both Lower and Upper Zone plagioclase lherzolite reported by Ozawa & Takahashi (1995)
. The size of the porphyroclasts in the normal plagioclase lherzolites is always less than 5 mm, and tends to decrease slightly with height above the base. In contrast, the megacrysts associated with the plagioclase-rich layers in the Upper Zone reach 8–17 mm in size and are 2–3 times larger than the largest porphyroclasts. |
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
Electron probe microanalyses were conducted with JEOL electron microprobe analyzers JXA-8900L and JCMA733-MkII at the Department of Earth and Planetary Science, Graduate School of Science, University of Tokyo. Analytical conditions for point and line analyses are 15 kV accelerating voltage and 1·2 x 10-8 A current measured with the Faraday cup. A focused beam was used except for line analyses of pyroxenes, for which a 20 µm diameter beam was used to obtain the bulk composition including exsolution lamellae. Counting time was 30 s for each element, but 10 s was used in the line analyses of pyroxenes to maximize the total number of analyzed points, from which averaged compositions were obtained for the core or marginal part. The ZAF correction scheme was used to reduce counts to oxide concentrations. In map analyses, conditions of 2·0 x 10-7 A current, 5 µm beam size, 5–20 µm pixel size, and 30–50 ms count time at each point were used.
|
MINERAL CHEMISTRY OF THE HOROMAN COMPLEX: NEW RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
Chemical zoning of large pyroxene porphyroclasts from the Upper Zone
Figure 3 shows the Al and Ca distribution in a large orthopyroxene megacryst in a plagioclase lherzolite from the North Ridge locality (Table 2). Marginal enrichment in Al and Ca is evident. The zoning patterns are concentric, although elongate clinopyroxene blebs near the center somewhat disturb the concentric zoning pattern of Ca. In line profiles, the Al and Ca contents have a marginal peak, from which the contents sharply decrease towards the rim and gradually decrease towards the relatively homogeneous core (Fig. 4). The variation in Al is counterbalanced by Si and Mg + Fe as shown in Fig. 4, and thus attributable to the Tschermak substitution, Si(Mg,Fe)
AlAl. A smaller grain (5 mm in size), present in the lower left corner in the Al and Ca maps (Fig. 3), displays zoning which monotonously decreases in Al and Ca from the center to the rim. Other orthopyroxene megacrysts from the Horoman River locality have similar Al and Ca zoning patterns.
|
|
|
Figure 5 shows the Al and Na distribution in a large clinopyroxene megacryst in a plagioclase lherzolite from the Horoman River locality (Table 2). As in the case of the orthopyroxene megacryst, Al shows concentric zoning with a marginal maximum, from which the Al content sharply decreases towards the rim and gradually decreases toward the core (Figs 5 and 6). This clinopyroxene has large blebs of orthopyroxene (
1 mm across) and plagioclase in the center, where the Al content of the host tends to be broadly constant but with some scatter (Fig. 6). The presence of such blebs makes it difficult to determine the bulk compositional variation in the clinopyroxene.
|
|
The plagioclase blebs are low in An [100 x Ca/(Ca + Na)
46] compared with the plagioclase surrounding the clinopyroxene (An >60). A few domains consisting of several apparently isolated plagioclases with the same optical orientation were identified. The orthopyroxene blebs are optically continuous, suggesting that they were exsolved from the host clinopyroxene. The domain structure of the plagioclase blebs, intimately associated with the orthopyroxene blebs, suggests exsolution of plagioclase from a super-silicic clinopyroxene enriched in the Ca-Eskola component (Gasparik, 1985). The size of the blebs suggests that the exsolution started at an earlier stage of, or even before, the decompression to give enough time to grow to 1 mm in size. Clinopyroxene containing large plagioclase and orthopyroxene blebs is common as porphyroclasts in the Upper Zone, and fine-grained ‘inclusions’ of albite and quartz of <20 µm in size have also been reported in clinopyroxene porphyroclasts from the Lower Zone (Takahashi, 1997). These albite and quartz inclusions could also have been exsolved from a super-silicic clinopyroxene (Kitayama et al., 2000).
In addition to blebby orthopyroxene in the core, there are smaller and aligned elongate orthopyroxene grains in the marginal zone (Fig. 5). The morphology and uniform optical orientation, sharing the c-axis with the host clinopyroxene, clearly suggests that the orthopyroxene in this zone was exsolved from the host. The profile of Ca/(1 – Na), which is a good indicator of temperature for pyroxene (Brey & Köhler, 1990), is shown in Fig. 6b together with a smoothed curve obtained by taking a running average including the orthopyroxene exsolution. The margin of the megacryst with abundant exsolution of orthopyroxene (Fig. 5) is low in Ca/(1 – Na). Such zoning, characterized by a marginal minimum, is also observed in smaller clinopyroxene porphyroclasts (<5 mm in size) in plagioclase lherzolite from the Lower Zone.
The Na content of the clinopyroxene megacryst decreases towards the rim in the marginal zone. There is an abrupt change in the Na content at the boundary between the marginal zone and the blebby core, where the Na content is rather constant. The Cr content is high in the core, and gradually decreases towards the rim, where it shows an abrupt decrease to 0·1. The overall zoning pattern is asymmetric and the decrease is more significant towards the rim in contact with plagioclase and spinel aggregates (Figs 5 and 6d). Another clinopyroxene from the North Ridge locality shows similar chemical zoning and internal microstructures to those described above (Table 2).
Spatial variation of core Al and Ca contents in orthopyroxene
The average compositions of the cores of the studied pyroxene megacrysts from the Upper Zone are summarized in Table 3. The Al and Wo contents [100 x Ca/(Ca + Mg + Fe)] in orthopyroxene are plotted against the height from the base of the Horoman complex in Fig. 7. In the Al diagram, the compositions of the smaller orthopyroxene crystals (porphyroclasts less than 5 mm in size) are also shown for comparison. It is noticeable that the Al contents of the cores of the large orthopyroxene megacrysts are significantly lower than those of grains smaller than 5 mm within the Upper Zone (Fig. 7a). The Al contents of the megacryst cores are similar to or slightly lower than those of the small orthopyroxene porphyroclasts from the Lower Zone.
|
|
and 
, orthopyroxene grains <5 mm (
, orthopyroxene crystals >5 mm (On the other hand, the Wo content in the cores of the orthopyroxene megacrysts is similar to or lower than that of the smaller orthopyroxene porphyroclasts from the Upper Zone, but distinctly higher than that of porphyroclasts from the Lower Zone (Fig. 7b). If the high Wo values in the Upper Zone are ignored, there is a broadly linear increase in Wo content upwards in the complex from the middle of the Lower Zone, below which the Wo content is almost constant.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
Petrographic constraints on the thermal history of the Horoman complex
In this section, petrographic observations that constrain the earliest and latest stages of the ascent history recorded in the Horoman peridotites are summarized. The former comes from the presence of two-pyroxene + spinel symplectites in symplectite-bearing spinel lherzolites, and the latter from plagioclase-rich, fine-grained aggregates in plagioclase lherzolites. Petrographic, mineralogical and geochemical information that unequivocally constrains the deep origin of the Horoman complex is the occurrence of the two-pyroxene + spinel symplectites frequently present in slightly depleted spinel lherzolite and rarely in plagioclase lherzolites. These aggregates are interpreted to be the breakdown product of garnet through reaction between olivine and garnet induced by the decompression of the Horoman complex (Takahashi & Arai, 1989
; Takazawa et al., 1996; Obata et al., 1997; Yoshikawa & Nakamura, 2000). The plagioclase lherzolites of the Horoman complex are unusual in the mode of occurrence of plagioclase. Plagioclase is completely restricted to fine-grained aggregates, composed mainly of plagioclase, olivine and chromite spinel, in the Lower Zone. These aggregates are interpreted to be a breakdown product of garnet through a two-pyroxene + spinel assemblage. The occurrence of plagioclase in the lherzolites is attributable to their bulk chemical composition, particularly the Na content. Decompression reactions continued to shallower levels and produced plagioclase (An 60–90) in Na-rich, fertile lherzolites, but did not in less fertile lherzolites (symplectite-bearing spinel lherzolites). This implies that the formation and inter-mineral equilibrium in the fine-grained aggregates or rims of large crystals was reached only near the boundary of the spinel and plagioclase stability fields.
Two-pyroxene + spinel symplectite aggregates in orthopyroxene porphyroclasts in a plagioclase lherzolite sample from the Lower Zone also provide additional petrographic information on the latest P–T conditions. The spinel in this sample is very aluminous [Cr number = 100 x Cr/(Cr + Al) 10], and there is no sign of reaction to form plagioclase. The occurrence of an intact two-pyroxene + spinel symplectite in an Na-depleted local environment (inclusion in orthopyroxene) in a plagioclase lherzolite suggests a significant temperature drop at a pressure in the vicinity of the boundary between plagioclase and spinel facies (Ozawa & Takahashi, 1995).
P–T trajectory of the Lower Zone
In the Lower Zone, orthopyroxene and clinopyroxene porphyroclasts in plagioclase lherzolite and symplectite-bearing spinel lherzolite show Al zoning strongly dependent on grain size; large grains exhibit M-shaped Al zoning profiles, whereas smaller grains show Al contents monotonously decreasing from the core to the rim. The core Al content decreases with increase in grain size, and tends to attain a minimum value. In the lowermost horizon of the Lower Zone, a grain size of 5 mm is enough to attain the minimum value of 0·12 (Fig. 8). Large orthopyroxene porphyroclasts 5 mm in size from the lowermost Lower Zone are homogeneous in terms of their Ca content [0·029 in Ca (O = 6) or 1·5 in Wo].
|
Because isolated spinel grains occurring outside fine-grained aggregates in plagioclase and symplectite-bearing spinel lherzolites from the Lower Zone are notably more Cr-rich and Al-rich, respectively, than spinel present in the fine-grained aggregates, the peridotites attain neither chemical nor textural equilibrium within the plagioclase and spinel facies. Moreover, the Al content of the cores is too low to be in equilibrium with other phases in the spinel stability field at the temperature indicated by the Ca content of the orthopyroxenes (<800°C vs >900°C). Therefore, it is concluded that the cores of the large orthopyroxene porphyroclasts with low and uniform Al contents record an equilibrium condition in the garnet stability field.
The temperature and pressure conditions of the lowermost Lower Zone in the garnet stability field before the start of ascent are constrained by the core Ca [0·03 ± 0·002 in Ca (O = 6) or 1·54 ± 0·09 in Wo after subtracting non-quadrilateral components] and Al contents in orthopyroxene [0·12 in Al (O = 6) or 6·7 ± 0·2% in Al(M1)*, Al (O = 6) in M1 site extrapolated to the Cr-free system] at 900 ± 20°C and 1·75 ± 0·07 GPa using a single pyroxene geothermobarometer combining the methods of Lindsley (1983) and Gasparik (1987) (Fig. 9). The errors are 1
and were estimated from analyses of orthopyroxene cores >3·5 mm near the bottom of the Lower Zone. The lowest temperature and pressure conditions registered in the orthopyroxene rims are estimated by the same approach to be 800°C and 0·6 GPa (Fig. 9). Two-pyroxene or single-clinopyroxene geothermobarometry was not used for two reasons. First, clinopyroxene porphyroclasts commonly have more abundant and coarser exsolution and inclusions than orthopyroxene. Second, kinetic processes make it difficult to apply geothermobarometry based on more than one phase; it is necessary to decide which pairs are in equilibrium, but the existence of such pairs is not guaranteed.
|
Application of the Ca-in-orthopyroxene thermometer of Brey & Köhler (1990)
, which has been often used in the literature (e.g. Takazawa et al., 1996), gives higher temperature by 100°C and accordingly higher pressure by 0·4 GPa than those estimated above. The discrepancy is basically due to the difference in treatment of Fe2+, which increases the Ca solubility in orthopyroxene coexisting with clinopyroxene (Lindsley, 1983; Sack & Ghiorso, 1994). This effect is ignored in the thermometry of Brey & Köhler (1990), who argued that the higher Al content in natural system compensates for the Fe effect. Perkins & Newton (1980) demonstrated that the mutual solubility of pyroxenes widens (decrease in Ca for orthopyroxene) in the Al-bearing system by comparing experiments in the CMS and CMAS systems. Experimental results in the CMS and CMAS systems (Perkins & Newton, 1980; Nickel & Brey, 1984; Nickel et al., 1985) show that the Al effect is more notable at low pressure <3 GPa. Within this range of pressure, however, the decrease in Ca content (O = 6) in the CMAS system in the presence of garnet relative to the CMS system is <0·004 on average, which is far below the difference of 0·015 caused by a change in Mg number [100 x Mg/(Mg + Fe)] from 90 to 100. This Al effect corresponds to a rise in temperature by 50°C on average for the same Ca content at pressures <3 GPa and temperatures ranging from 900 to 1100°C, although the estimation contains significant errors because of the limited numbers of experiments and less tight bracketing (Nickel & Brey, 1984; Nickel et al., 1985). Application of this temperature correction to the above estimation yields 950°C and 1·9 GPa as the equilibrium conditions for the lowermost Lower Zone before the Horoman complex started to ascend.
The Fe effect is more significant than that of Lindsley (1983) in the quadrilateral pyroxene system thermodynamically calibrated by Sack & Ghiorso (1994), and gives temperatures lower by 50°C and pressure lower by 0·2 GPa than those estimated above (900°C and 1·75 GPa after correcting for the Al effect).
The Ca zoning in orthopyroxene porphyroclasts suggests a nearly adiabatic P–T trajectory for the lowermost Lower Zone. Orthopyroxene often shows a marginal hump in its Ca content higher than the core region by up to 0·007 (0·2 in Wo), which represents the minimum estimate for the peak temperature during decompression. The Ca zoning implies an isothermal trajectory (or a slight temperature rise) during the decompression because the isopleths for Ca in orthopyroxene with Mg number 91 are 0·002–0·004/GPa (0·1–0·22/GPa in Wo) at a temperature in the range of 800–1000°C (Sack & Ghiorso, 1994). Because of the constraint from the P–T condition recorded in the rim of orthopyroxene porphyroclasts (800°C, 0·6 GPa), there should be a rapid fall in temperature within a small decrease of pressure at a shallower level (Fig. 9).
Such a P–T trajectory is consistent with the M-shaped Al zoning in orthopyroxene as shown by Ozawa (1997), who modeled the development of Al zoning in orthopyroxene during decompression of peridotite (Fig. 10; see Appendix for details of modeling). The M-shaped zoning pattern can be formed when the decompression path is nearly isothermal from the garnet stability field to the boundary of the plagioclase and spinel fields after which temperature falls rapidly (Figs 9 and 10a). For grain size of 5 mm, the largest porphyroclasts in the Lower Zone, and a decompression rate of 1 kPa/year, no marginal peak in Al is produced if the P–T trajectory is too steep (paths 9–13 in Fig. 10a and b). Furthermore, only a very minor marginal peak is produced if the temperature fall at high pressure is significant, because rapid cooling restrains diffusion of Al into the marginal zone from the surface with high Al content (paths 6–8 in Fig. 10a and b). P–T trajectories that reproduce the typical M-shaped Al zoning must have a gentler slope in the P–T diagram (e.g. paths 1–5 in Fig. 10a and b).
|
Because the Al zoning pattern strongly depends on the decompression rate, the ascent rate can be determined by simulating the zoning profile if reliable Al diffusivity data in orthopyroxene are available (Fig. 11). Available Al diffusivity data (Smith & Barron, 1991
) give 1 kPa/year for the ascent rate with three constraints: the presence of a marginal Ca hump, the ratio of Al content in the marginal zone against that of the core <1·5, and the preservation of a wide region with constant Al content in the center (Fig. 8). To fulfill the second constraint, a steeper P–T path such as path 4 in Fig. 10 is more plausible, but this cannot be consistent with the first constraint requiring nearly isothermal conditions.
|
The observed lower marginal maximum in Al content than expected from Fig. 11 is attributable to a decrease in decompression rate at a pressure of
0·6 GPa, where the Horoman complex encountered the Hidaka crust (Ozawa, 1997). The estimated rate has large errors (probably more than one order of magnitude) because the diffusivity data obtained from Al zoning in natural orthopyroxene in contact with garnet contain large uncertainties, so they must be viewed with caution. The assumption of a constant decompression rate also introduces an additional uncertainty into the rate estimated above.
P–T history of the Upper Zone
In deriving the thermal history of the Upper Zone, previous investigators assumed that the core compositions of the orthopyroxene porphyroclasts represent that of the garnet stability field. Ozawa (1997), however, pointed out that the high Al and Ca contents in orthopyroxene porphyroclasts in the Upper Zone can also be explained by a heating event at 1 GPa during ascent from an initial condition similar to that of the Lower Zone (950°C and 2·0 GPa). He examined this possibility in detail by forward modeling of Al zoning in orthopyroxene during ascent of plagioclase lherzolite for various P–T paths and decompression rates (see Appendix for modeling details).
One set of decompression paths examined by Ozawa (1997) is shown in Fig. 12. One of these is similar to the P–T trajectory inferred for the lowermost Lower Zone and features an isothermal decompression path followed by a rapid temperature drop in the plagioclase stability field (path 9). Other trajectories (paths 1–8) are those for a heating event in the spinel stability field for various heating duration starting from the same P–T conditions in the garnet stability field as that of path 9. The decompression paths 5–9 qualitatively reproduce the Al zoning in porphyroclastic orthopyroxene with grain size less than 5 mm from the Lower Zone, and paths 1–3 reproduce that of the Upper Zone. The heating model is coherent with the notable marginal hump in Ca content in orthopyroxene and the corresponding marginal minimum for Ca content in clinopyroxene commonly observed in the Upper Zone. Pyroxene megacrysts, for which data are presented above, provide critical information to substantiate the suggestion that the Upper Zone actually experienced a heating event.
|
The core Al content in the orthopyroxene megacrysts from the Upper Zone is significantly lower than that of the porphyroclasts (<5 mm in size) in the plagioclase lherzolites of the Upper Zone (Fig. 7a). The Al content of the orthopyroxene megacryst cores is nearly constant in the Upper Zone. The core Ca content in the orthopyroxene megacrysts is lower than that of the porphyroclasts, and shows a slight increase with stratigraphic height in the Upper Zone (Fig. 7b). The variation in the composition of the porphyroclasts within the complex as a whole is continuous without any gap between the two zones. The presence of wide core regions with fairly homogeneous composition for orthopyroxene (Figs 3 and 4) and clinopyroxene (Figs 5 and 6) suggests that the core compositions represent meaningful initial conditions at which the Horoman peridotites stayed for a fairly long time. The occurrence of pyroxene megacrysts in a thin plagioclase-rich layer <1 cm in thickness, surrounded by plagioclase lherzolite with the typical grain size, mineral mode and texture suggests that the megacrysts could have been in equilibrium with minerals in the host lherzolite in the garnet stability field.
After correcting for the Al effect, the equilibrium conditions are estimated to be 1070 ± 30°C and 2·3 ± 0·11 GPa for the uppermost Upper Zone based on 0·042 ± 0·004 in Ca (O = 6) (Wo = 2·2 ± 0·2) and 0·12 in Al (O = 6) [Al(M1)* = 6·22%; Table 3]. Because only one megacryst was analyzed at each locality, the maximum errors for porphyroclastic grains <5 mm in size are employed. These P–T conditions are obviously higher than those of the lowermost Lower Zone.
A maximum Al (O = 6) content in the marginal zone of the orthopyroxene megacrysts as high as 0·2 [12% in Al(M1)*; Fig. 4] can be explained by a nearly adiabatic decompression path. However, the maximum Ca content in the marginal zone attains 0·065 (0·033 in Wo; Fig. 4), which is similar to the Ca contents of smaller orthopyroxene cores from the uppermost Upper Zone (Fig. 3). These high Ca contents correspond to 1200°C at 1·0 GPa, suggesting that the Upper Zone experienced a heating event in the spinel stability field with a temperature rise of 130°C (Fig. 9). Application of the Ca-in-orthopyroxene thermometer of Brey & Köhler (1990) gives a much smaller temperature rise of 70°C.
Implication of the variation in the ‘initial P–T conditions’ in the Horoman peridotite complex
The P–T estimated conditions in the garnet stability field for the lowermost Lower Zone and the uppermost Upper Zone, based on orthopyroxene geothermobarometry, differ by 120 ± 50°C and 0·4 ± 0·18 GPa. Because the Al contents of the cores of the large orthopyroxenes are similar, the variation in the estimated conditions is principally attributable to the Ca content, the difference of which is statistically significant [0·03 ± 0·002 vs 0·042 ± 0·004 in Ca (O = 6) and 1·54 ± 0·09 vs 2·2 ± 0·2 in Wo; Fig. 7]. Although the absolute P–T estimates can vary by 150°C and 0·5 GPa depending on the applied geothermometers, the temperature difference is similar for any of the geothermometers.
Comparing the ‘initial conditions’ in the garnet stability field for the two zones, it is suggested that the Horoman complex was originally a part of the lithospheric upper mantle with the Lower Zone overlying the Upper Zone. A temperature gradient existed in the Horoman mantle, although the variation may not have been as large as that suggested by Ozawa & Takahashi (1995). The estimated vertical temperature gradient within the Horoman mantle is 10 ± 8°C/km. This value is comparable with the geothermal gradient in the lithospheric mantle estimated by applying geothermobarometers to lherzolite xenoliths (4–6°C/km; Boyd, 1973; Rudnick & Nyblade, 1999; Nimis & Taylor, 2000) if the large error is taken into consideration. It may represent the thermal gradient of the Horoman lithospheric upper mantle. If the pressure difference between the two zones is accepted, the distance between the top of the Upper Zone and the basal part of the Horoman complex is estimated to be 12 ± 5 km, which is much greater than the current thickness of the Horoman complex (3 km), by a factor of 4 ± 2. This can be interpreted as stretching that the Horoman complex underwent during its emplacement into the Hidaka lower crust.
Origin of the Upper and Lower Zones and its relevance to the thermal history
Table 1 summarizes geological, petrographic and mineralogical contrasts between the Upper and Lower Zones, compiled from previous studies of the Horoman complex (see references in the table legend). Any comprehensive model for the origin of the Horoman complex must explain all these features. A variable thermal history within the Horoman complex can account for most of the observed differences.
The higher-temperature path followed by the Upper Zone, irrespective of the exact ascent P–T path, can explain most of the petrographic and mineralogical distinctions. For example, the larger grain size of the two-pyroxene spinel symplectites and fine-grained aggregates after the symplectites in the Upper Zone peridotites can be attributable to rapid grain growth during a prolonged higher-temperature stage, assuming comparable strain rates. The more homogeneous composition of plagioclase and spinel in the Upper Zone peridotites can be explained by rapid elemental diffusion in the minerals and also along grain boundaries because of the high-temperature ascent path. The minor occurrence of symplectite-bearing spinel lherzolite in the Upper Zone may be partly attributable to more enhanced subsolidus reaction to produce plagioclase in the symplectite-bearing spinel lherzolites.
Ozawa & Takahashi (1995) argued that the Upper Zone might have partially melted where enough water was available. They further speculated that the plagioclase-rich veins oblique to the foliation were fractures filled by melt segregated from the surrounding partially molten peridotite (segregation veins; Table 1). Takahashi (2001) reported abundant plagioclase-rich segregation veins and also oscillatory Ca–Na growth zoning of plagioclase in the plagioclase lherzolites of the Nikanbetsu complex. These observations provide unequivocal evidence for partial melting and minor segregation in the Nikanbetsu complex, which is the high-temperature extremity of the Upper Zone of the Horoman complex. Therefore, it is plausible that the Upper Zone also underwent partial melting, and some of the geological contrasts with the Lower Zone (Table 1) may also be attributed to this melting event.
Isotopic ages for the mafic rocks (Takazawa et al., 1999) and peridotites (Yoshikawa & Nakamura, 2000) indicate that the mafic layers and the major lithological variation shown by the peridotites were not formed during the final ascent of the complex. The preservation of such old geological structures consistently overlapping the younger thermal structure can be explained by the overturned stratification of the two zones. The initially hotter and deeper Upper Zone may represent a lower part of the lithospheric mantle that now overlies the Lower Zone, which was a cooler and shallower part of the lithospheric mantle (Ozawa, 1997). Sawaguchi (2001) proposed that the current sequence of stratification within the Horoman complex could be explained by inverted thermal structures in the mantle wedge above a subducting slab. However, the estimated P–T conditions within the garnet stability field negate this possibility.
Implication of the difference in ascent P–T path between the two zones
The petrographical and mineralogical observations on the Horoman complex may indicate that the upper part of the complex experienced a heating event in the spinel stability field during its ascent from the garnet stability field, and that the lowest part of the complex ascended almost adiabatically. The continuous variations of Ca and Al contents in orthopyroxene within the complex (Fig. 7) indicate that a thermal gradient was maintained up to the boundary between the spinel and plagioclase lherzolite stability fields.
The source of heat for the inferred heating event could be basaltic melt, which might be responsible for the formation of gabbroic rocks within the Hidaka metamorphic belt (Maeda & Kagami, 1996), passing through the Upper Zone itself or a region more closely adjacent to the Upper Zone than to the Lower Zone. Another possibility is that heating was induced by a hot asthenospheric, mantle diapir which was emplaced into the lithosphere adjacent to the Upper Zone (Van der Wal & Vissers, 1993; Kimura, 1996; Model 2 of Ozawa, 1997; Drury et al., 2001). Because the Upper Zone currently occupies the upper part of the complex, the sequence must have been overturned by the quicker ascent of the hotter, deeper part associated with heating. Overturning of the heated part cannot be caused only by magma ascent through fractures, and the hot ascending diapir model seems much more plausible (Ozawa, 1997). The hot Horoman complex itself might have acted as a significant heat source when it was emplaced into the lower crust of the Hidaka metamorphic belt, where it triggered metamorphism and partial melting of the lower-crustal rocks (Tagiri et al., 1995).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
The thermal history of the Horoman peridotite complex proposed by Ozawa & Takahashi (1995)
has been revised based on newly obtained data for the Ca and Al contents of pyroxene megacrysts from the Upper Zone. The variation in Al content of the cores of large orthopyroxene megacrysts in the Upper Zone is notably different from their predicted values; the Al content is almost constant or decreases slightly from the bottom to the top of the Upper Zone. The revised variation of the Ca content in the largest (megacrysts and porphyroclasts) orthopyroxene cores exhibits a linear increase upwards from the middle of the Lower Zone, beneath which the Wo content is nearly constant. These variations in mineral chemistry suggest that the structurally higher horizons (Upper Zone) of the complex initially resided at higher temperature and pressure conditions (1070 ± 30°C, 2·3 ± 0·11 GPa) than the lower horizon (Lower Zone; 950 ± 20°C, 1·9 ± 0·07 GPa) before ascent from the garnet stability field.
The higher Ca contents of the cores of porphyroclasts with 5 mm grain size than that of the megacrysts indicates that the Upper Zone experienced a heating event during its ascent through the spinel stability field, with peak temperatures higher than 1200°C. The effect of the heating event appeared to have decreased continuously toward the base of the complex, and the lower part of the Lower Zone underwent only a very minor heating event.
The thermal structure in the Horoman lithospheric mantle before the final decompression is comparable with that of the continental lithosphere. The vertical thermal gradient is estimated to be 10 ± 8°C/km. The current thickness of the Horoman complex (3 km) is a result of flattening by factor of 0·25 ± 0·1 during its ascent from depth; this deformation was probably driven by the ascent of an asthenospheric mantle diapir.
|
APPENDIX: KINETIC MODEL FOR Al ZONING IN ORTHOPYROXENE PORPHYROCLASTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
To model Al zoning of orthopyroxene in the peridotite system, we have to simplify the reaction processes, which involve transportation through grain boundaries, diffusion in an orthopyroxene grain, and migration of grain surface. If peridotites have porphyroclastic or mylonitic textures, consisting of a small number of large pyroxene crystals and a dominant fine-grained matrix, we can simplify the reaction system as illustrated in Fig. A1. In this case, we can divide the system into two domains: an equilibrium domain (matrix), in which the minerals are nearly in equilibrium because of their smaller grain size, and a kinetically controlled domain (porphyroclasts), in which the minerals cannot attain equilibrium because of their larger grain size. Rapid reactions involving net mass-transfer between the minerals in the fine-grained matrix suppress grain growth or dissolution of large porphyroclasts, which communicate with the fine-grained matrix at a rate limited by lattice diffusion in the porphyroclasts. The matrix may contain large minerals, such as garnet, but this does not affect the behaviour of two domains unless diffusion in such minerals affects the overall reaction rate in the equilibrium domain.
|
These conditions are mostly fulfilled in the studied Horoman plagioclase lherzolite at least after garnet became unstable, because of the presence of fine-grained aggregates. The current grain size of the fine-grained aggregate is of the order of 10–100 µm, which may have been much smaller during the main reaction stage, whereas the porphyroclastic orthopyroxene is 1–5 mm in size. This size contrast ensures quicker reactions in the fine-grained aggregate compared with diffusion in the orthopyroxene porphyroclasts. Because spinel and plagioclase are present mostly in the fine-grained aggregate, rapid grain boundary diffusion ensures that the surface Al contents of the orthopyroxene track near-equilibrium values as decompression proceeds. These conditions were not realized when garnet was stable, but diffusion within garnet crystals may not be so important in modifying the Al content in orthopyroxene in a fertile lherzolite. Rapid reaction and transportation through grain boundaries are sufficient to justify application of the model to decompression through the garnet stability field.
In accordance with the model, we may be able to assume: (1) no grain growth or dissolution of large orthopyroxene grains; (2) local equilibrium at the grain surface; (3) perfect equilibrium in the matrix; (4) high volume fraction of fine-grained matrix as compared with pyroxene porphyroclasts. Additional assumptions made in this paper are: (5) spherical orthopyroxene porphyroclasts; (6) initially homogeneous Al concentration; (7) CFMAS system with Mg/(Mg + Fe) = 0·9. The studied orthopyroxene porphyroclasts are not strongly elongated and the aspect ratio measured in thin section is <0·5.
The aforementioned model is cast into a spherical diffusion problem with a given surface Al concentration as a function of time, and is expressed by the following equation and initial and boundary conditions:
 | (1) |
 | (2) |
 | (3) |
 | (4) |
(Lasaga, 1983), which is defined by the equation
 | (5) |
 | (6) |
 | (7) |
 | (8) |
= 0 ', where 0 is the maximum compressed time and can be obtained by integrating (5) until the value does not change significantly. After dropping the primes, the non-dimensionalized equations become
 | (9) |
 | (10) |
 | (11) |
is a dimensionless parameter representing the ratio of the maximum compressed time 0, which corresponds to the effective cooling time scale against the diffusion time scale at the initial condition. This parameter is expressed as
 | (12) |
), but the corresponding finite element equation was solved because the boundary condition at r = 1 cannot be expressed in a simple integrable form.
In the model calculation, the Al concentration of orthopyroxene as a function of P–T is taken from Gasparik (1987). Grain diameter, activation energy and the pre-exponential term for Al diffusion in orthopyroxene used in the following calculation are 5 mm, 4 x 105 J/mol and 1·94 x 10-3 m2/s, respectively. The diffusion data are after Smith & Barron (1991). Because of the large uncertainty in the diffusivity data, the estimated absolute decompression rate and P–T path inevitably have large errors. Ozawa (1997) examined the effect of decompression rate and P–T path on the frozen-in Al zoning profile by performing a comprehensive parameter study.
|
ACKNOWLEDGEMENTS |
|---|
Discussions with M. Obata, E. Takazawa and T. Sawaguchi during the Samani conference and on many other occasions were very helpful. I thank H. Nagahara for her encouragement and critically reading the manuscript, and E. Nakamura for stimulating discussion on the Horoman complex. Reviews by R. Altherr and two anonymous reviewers are greatly appreciated.
|
FOOTNOTES |
|---|
* Corresponding author. Telephone: +81-3-5841-4509. Fax: +81-3-5841-4569. E-mail: Ozawa{at}eps.s.u-tokyo.ac.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL BACKGROUND AND...ANALYTICAL METHODSMINERAL CHEMISTRY OF THE...DISCUSSIONCONCLUSIONSAPPENDIX: KINETIC MODEL FOR...REFERENCES |
|---|
Boyd, F. R. (1973). A pyroxene geotherm. Geochimica et Cosmochimica Acta 37, 2533–2546.[CrossRef][Web of Science]
Brenker, F. E. & Brey, G. P. (1997). Reconstruction of the exhumation path of the Alpe Arami garnet-peridotite body from depths exceeding 160 km. Journal of Metamorphic Geology 15, 581–592.[CrossRef][Web of Science]
Brey, G. P. & Köhler, T. (1990). Geothermobarometry in four-phase lherzolites II. New thermobarometers and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353–1378.
Carslaw, H. S. & Jaeger, J. C. (1959). Conduction of Heat in Solids. New York: Oxford University Press, 510 pp.
Davies, G. R., Nixon, P. H., Pearson, D. G. & Obata, M. (1993). Tectonic implications of graphitized diamonds from the Ronda peridotite massif, southern Spain. Geology 21, 471–474.
Drury, M. R., van Roermund, H. L. M., Carswell, D. A., de Smet, J. H., van den Berg, A. P. & Vlaar, N. J. (2001). Emplacement of deep upper-mantle rocks into cratonic lithosphere by convection and diapiric upwelling. Journal of Petrology 42, 1131–1140.
Gasparik, T. (1985). Experimental study of subsolidus phase relations and mixing properties of clinopyroxene and plagioclase in the system Na2O–CaO–Al2O3–SiO2. Contributions to Mineralogy and Petrology 89, 346–357.[CrossRef][Web of Science]
Gasparik, T. (1987). Orthopyroxene thermobarometry in simple and complex systems. Contributions to Mineralogy and Petrology 96, 357–370.[CrossRef][Web of Science]
Green, D. H. & Hibberson, W. (1970). The instability of plagioclase in peridotite at high pressure. Lithos 3, 209–221.[CrossRef]
Kadarusman, A. & Parkinson, C. D. (2000). Petrology and P–T evolution of garnet peridotites from central Sulawesi, Indonesia. Journal of Metamorphic Petrology 18, 193–209.
Kimura, G. (1996). Collision orogeny at arc–arc junctions in the Japanese Islands. Island Arc 5, 262–275.
Kitayama, I., Parkinson, C. D., Okamoto, K., Nakajima, Y. & Maruyama, S. (2000). Supersilicic clinopyroxene and silica exsolution in UHPM eclogite and pelitic gneiss from the Kokchetav massif, Kazakhstan. American Mineralogist 85, 1368–1374.
Komatsu, M. & Nochi, M. (1966). Ultrabasic rocks in the Hidaka metamorphic belt, Hokkaido, Japan. I. Mode of occurrence of the Horoman ultrabasic rocks. Chikyukagaku (Earth Science) 87, 21–29.
Komatsu, M., Miyashita, S., Maeda, J., Osanai, Y. & Toyoshima, T. (1983). Disclosing of a deepest section of continental-type crust upthrust as the final event of collision of arcs in Hokkaido, North Japan. In: Hashimoto, M. & Uyeda, S. (eds) Acccretion Tectonics in the Circum-Pacific Regions, Tokyo: TERRAPUB, pp. 149–165.
Kornprobst, J. (1969). Le massif ultrabasique des Beni Bouchera (Rif Interne, Maroc): étude des peridotites de haute température et de haute pression, et des pyroxénolites, à grenat ou sans grenat, qui leur sont associées. Contributions to Mineralogy and Petrology 23, 288–322.
Kushiro, I. & Yoder, H. S. (1966). Anorthite–forsterite and anorthite–enstatite reactions and their bearing on the basalt–eclogite transformation. Journal of Petrology 7, 337–362.
Kushiro, I., Syono, Y. & Akimoto, S. (1968). Melting of a peridotite nodule at high pressures and high water pressures. Journal of Geophysical Research 73, 6023–6029.
Lasaga, A. C. (1983). Geospeedometry: an extension of geothermometry. In: Saxena, S. K. (ed.) Kinetics and Equilibrium in Mineral Reactions, Vol. 3: Advances in Physical Geochemistry. New York: Springer, pp. 81–114.
Lindsley, D. H. (1983). Pyroxene thermometry. Amrican Mineralogist 68, 477–493.
Loubet, M. & Allègre, C. J. (1979). Trace element studies in the Alpine type peridotite of Beni-Bouchera (Morocco). Geochemical Journal 13, 69–75.[Web of Science]
MacGregor, I. D. & Basu, A. R. (1976). Geological problems in estimating mantle geothermal gradients. American Mineralogist 61, 715–724.[Abstract]
Maeda, J. & Kagami, H. (1996). Interaction of a spreading ridge and an accretionary prism: implications from MORB magmatism in the Hidaka magmatic zone, Hokkaido, Japan. Geology 24, 31–34.
Nickel, K. G. & Brey, G. (1984). Subsolidus orthopyroxene–clinopyroxene systematics in the system CaO–MgO–SiO2 to 60 kb: a re-evaluation of the regular solution model. Contributions to Mineralogy and Petrology 87, 35–42.[CrossRef][Web of Science]
Nickel, K. G., Brey, G. P. & Kogarko, L. (1985). Orthopyroxene–clinopyroxene equilibria in the system CaO–MgO–Al2O3–SiO2 (CMAS): new experimental results and implications for two-pyroxene thermometry. Contributions to Mineralogy and Petrology 91, 44–53.[CrossRef][Web of Science]
Nicolas, A. (1989). Structures of Ophiolites and Dynamics of Oceanic Lithosphere. Dordrecht: Kluwer Academic.
Niida, K. (1974). Structure of the Horoman ultramafic massif of the Hidaka metamorphic belt in Hokkaido, Japan. Journal of Geological Society of Japan 80, 31–44.
Niida, K. (1975). Textures and olivine fabrics of the Horoman ultramafic rocks, Japan. Journal of Mineralogy, Petrology and Economic Geology 70, 265–285.
Niida, K. (1984). Petrology of the Horoman ultramafic rocks in the Hidaka metamorphic belt, Hokkaido, Japan. Journal of Faculty of Sciences, Hokkaido University, Series IV 21, 197–250.
Niida, K. & Katoh, T. (1978). Ultramafic rocks in Hokkaido. Monograph of Association for Geological Collaboration in Japan 21, 61–81 (in Japanese with English abstract).
Nimis, P. & Taylor, W. R. (2000). Single clinopyroxene thermobarometry for garnet peridotites. Part I. Calibration and testing of a Cr-in-Cpx barometer and an enstatite-in-Cpx thermometer. Contributions to Mineralogy and Petrology 139, 541–554.[CrossRef]
Nimis, P. & Trommsdorff, V. (2001). Comment on new constraints on the P–T evolution of the Alpe Arami garnet peridotite body (central Alps, Switzerland) by Paquin & Altherr (2001). Journal of Petrology 42, 1773–1779.
Obata, M. (1980). The Ronda peridotite: garnet-, spinel-, and plagioclase-lherzolite facies and the P–T trajectories of a high temperature mantle intrusion. Journal of Petrology 21, 533–572.
Obata, M. & Nagahara, K. (1997). Layering of Alpine-type peridotite and the segregation of partial melt in the upper mantle. Journal of Geophysical Research 92, 3467–3474.
Obata, M., Morisita, R. & Tanaka, K. (1997). The microstructure of symplectite in the Horoman peridotite and its formation processes. Memoir of the Geological Society of Japan 47, 163–171 (in Japanese with English abstract).
O'Neill, H. S. C. (1981). The transition between spinel lherzolite and garnet lherzolite, and its use as a geobarometer. Contributions to Mineralogy and Petrology 77, 185–194.[CrossRef][Web of Science]
Ozawa, K. (1997). P–T history of an ascending mantle peridotite constrained by Al zoning in orthopyroxene: a case study in the Horoman peridotite complex, Hokkaido, northern Japan. Memoir of the Geological Society of Japan 47, 107–122.
Ozawa, K. & Takahashi, N. (1995). P–T history of a mantle diapir: the Horoman peridotite complex, Hokkaido, northern Japan. Contributions to Mineralogy and Petrology 120, 223–248.[Web of Science]
Paquin, J. & Altherr, R. (2001a). New constraints on the P–T evolution of the Alpe Arami garnet peridotite body (central Alps, Switzerland). Journal of Petrology 42, 1119–1140.
Paquin, J. & Altherr, R. (2001b). ‘New constraints on the P–T evolution of the Alpe Arami garnet peridotite body (central Alps, Switzerland)’: Reply to comment by Nimis & Trommsdorff (2001). Journal of Petrology 42, 1781–1787.
Perkins, D., III & Newton, R. C. (1980). The compositions of coexisting pyroxenes and garnet in the system CaO–MgO–Al2O3–SiO2 at 900°C–1100°C and high pressures. Contributions to Mineralogy and Petrology 75, 291–300.
Rudnick, R. L. & Nyblade, A. A. (1999). The thickness and heat production of Archean lithosphere: constraints from xenolith thermobarometry and surface heat flow. In: Fei, Y., Bertka, C. M. & Mysen, B. O. (eds) Mantle Petrology: Field Observations and High Pressure Experimentation, A Tribute to Francis R. (Joe) Boyd. Geochemical Society Special Publication 6, 3–12.
Sack, R. O. & Ghiorso, M. S. (1994). Thermodynamics of multicomponent pyroxenes: II. Phase relations in the quadrilateral. Contributions to Mineralogy and Petrology 116, 287–300.[CrossRef][Web of Science]
Sawaguchi, T. (2001). Structural petrology of the Horoman peridotite: deformation history and exhumation process of the Alpine-type peridotite massif. Ph.D. thesis, Waseda University.
Sawaguchi, T. & Takagi, H. (1997). Inverted ductile shear movement of the Horoman peridotite complex in the Hidaka metamorphic belt, Hokkaido, Japan. Memoir of the Geological Society of Japan 47, 193–208.
Smith, D. & Barron, B. R. (1991). Pyroxene–garnet equilibration during cooling in the mantle. American Mineralogist 76, 1950–1963.[Abstract]
Tagiri, M., Tanaka, H. & Shiba, M. (1995). Melting of amphibolites and the form of melt-trap in amphibolite–migmatites of the southern Hidaka metamorphic belt, Hokkaido, Japan. Journal of Mineralogy, Petrology and Economic Geology 90, 50–63.[CrossRef]
Takahashi, E. & Kushiro, I. (1986). Melting of a dry peridotite at high pressure and basalt magma genesis. American Mineralogist 68, 859–879.
Takahashi, N. (1991). Origin of three peridotite suite from Horoman peridotite complex, Hokkaido, Japan; melting, melt segregation and solidification processes in the upper mantle. Journal of Mineralogy, Petrology and Economic Geology 86, 199–215.
Takahashi, N. (1992). Evidence for melt segregation towards fractures in the Horoman mantle peridotite complex. Nature 359, 52–55.[CrossRef]
Takahashi, N. (1997). Incipient melting of mantle peridotites observed in the Horoman and Nikanbetsu peridotite complexes, Hokkaido, northern Japan. Journal of Mineralogy, Petrology and Economic Geology 92, 1–24.[CrossRef]
Takahashi, N. (2001). Origin of plagioclase lherzolite from the Nikanbetsu peridotite complex, Hokkaido, northern Japan: implications for incipient melt migration and segregation in the partially molten upper mantle. Journal of Petrology 42, 39–54.
Takahashi, N. & Arai, S. (1989). Textural and chemical features of chromian spinel pyroxene symplectite in the Horoman peridotites, Hokkaido, Japan. Scientific Report of Institute of Geosciences of University of Tsukuba, Section B 10, 45–55.
Takazawa, E., Frey, F., Shimizu, N. & Obata, M. (1996). Evolution of the Horoman peridotite (Hokkaido, Japan): implications from pyroxene compositions. Chemical Geology 134, 3–26.[CrossRef][Web of Science]
Takazawa, E., Frey, F. A., Shimizu, N., Saal, A. & Obata, M. (1999). Polybaric petrogenesis of mafic layers in the Horoman peridotite complex, Japan. Journal of Petrology 40, 1827–1851.[CrossRef][Web of Science]
Tazaki, K., Ito, E. & Komatsu, M. (1972). Experimental study on a pyroxene–spinel symplectite at high pressures and temperatures. Journal of Geological Society of Japan 78, 347–354.
Tsai, C.-H., Liou, J. G. & Erns, W. G. (2000). Petrological characterization and tectonic significance of retrogressed garnet peridotites, Raobazhai area, North Dabie Complex, east–central China. Journal of Metamorphic Geology 18, 181–192.[CrossRef][Web of Science]
Tubía, J. M. (1994). The Ronda peridotites (Los Reales nappe): an example of the relationship between lithospheric thickening by oblique tectonics and late extensional deformation within the Betic Cordillera (Spain). Tectonophysics 238, 381–398.[CrossRef][Web of Science]
Van der Wal, D. & Vissers, R. L. M. (1993). Uplift and emplacement of upper mantle rocks in the western Mediterranean. Geology 21, 1119–1122.
Van Roermund, H. L. M., Drury, M. R., Barnhoorn, A. & de Ronde, A. A. (2000). Super-silicic garnet microstructures from an orogenic garnet peridotite, evidence for an ultra-deep (>6 GPa) origin. Journal of Metamorphic Geology 18, 135–147.[CrossRef][Web of Science]
Yang, J., Godard, G., Kienast, J.-R., Lu, Y. & Sun, J. (1993). Ultrahigh-pressure (60 kbar) magnesite-bearing garnet peridotites from northeastern Jiangsu, China. Journal of Geology 101, 541–554.[Web of Science]
Yoshikawa, M. & Nakamura, E. (2000). Geochemical evolution of the Horoman peridotite complex: implications for melt extraction, metasomatism, and compositional layering in the mantle. Journal of Geophysical Research 105, 2879–2901.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||