Petrogenesis of carbonatites from the Eastern Himalayan syntaxis
Yongqing Liu,Dalai Zhong,Hans Joachim Massonne,R. Schweinhage,Yufeng Ren
IF: 3
2001-01-01
Journal of Asian Earth Sciences
Abstract:Carbonatites are carbonate-rich igneous rocks of mantle origin. Although carbonatites are rare among igneous rocks, they have received a considerable amount of attention in the recent years. Current experiments have shown that the primary carbonate melts can be produced directly by partial melting of mantle peridotite and their compositions are dolomitic, whereas the calcitic carbonatites are dominant at the Earth's surface on a global scale. Harmer and Gittins (1997) argued strongly that both calcitic and dolomitic carbonatites derived from primary magnesian mantle-derived carbonate melts. Experiments have also shown that calcitic magmas can be generated from dolomitic melts, consistent with deduction from studies on mantle nodules. Despite these various pieces of evidence, however, and old view that carbonates are secondary products of parental silicate magmas of nepheliniteic affinity is still popular. The details of carbonate melts' evolution are, therefore, still uncertain. The carbonatites of eastern Himalayan syntaxis are, however, dominantly dolomitic. Their phlogopite K-Ar ages are in range of from 3.6 to 5.5 Ma (Liu et al., 2000). It is generally accepted that the Indian plate has been moving northwards relative to Eurasian plate throughout the past 50 my to result in great Himalayas, and the mountain-building processes are still ongoing. Thus the carbonatites of the eastern Himalayas are important not only because they add to the number of occurrences of mountain-building setting, as most carbonatites are found within plates by anorogenic processes, but also because they have a few features which have significant implications for carbonatite genesis in general. Before the middle of 1990s, the eastern Himalayan syntaxis was the least-known segment of the Himalayas. Recently much work has been done to make this area known. One of the most important findings in this area is granulite facies complex of at least 600 km2, which was argued to experience high-pressure metamorphism and represent materials buried to upper-mantle depths (Liu and Zhong, 1997). To the south of the granulite facies complex is amphibolite facies complex. It is similar to the high-grade metamorphic complex, also known as the Greater Himalayan Sequence or the Central Crystallines, occurring to the west of the eastern Himalayan syntaxis. To the south of the amphibolite facies complex is greenschist facies complex which is analogous with the so-called Lesser Himalayan complex. To the south of the greenschist facies rocks are sedimentary rocks, belonging to the sub-Himalayan Siwalik molasse-type sediments. These pieces of evidence imply that the granulite facies complex here represent the lower crust of the eastern Himalayas. In the granulite facies complex, the carbonatites were discovered to intrude the granulite facies rocks, and form dyke swarms. Individual dykes vary from a few centimeters to several meters thick, are up to several kilometers long and have extensive metasomatic and/or contact metamorphic alteration halos. Harmer and Gittins (1997) found that the dolomitic carbonatites are abundant in old Precambrian cratonic regions which are regarded to represent the low crust of continents (Fountain et al., 1992). It seems, therefore, true that dolomitic carbonatites usually occurred in the low crust of continents. Additionally, several MgO-rich ultramafic dykes were found to occur in the northern dyke swarm. They are also banded by foliation of the granulite facies rocks. Common rock types are spinel olivine hornblendite, orthopyroxene olivine phlogopite hornblendite and orthopyroxene phlogopite hornblendite. Their compositions are similar to those of picrite and komaiite. No alkaline rocks (including floater) normally associated with carbonatite have been seen by us so far. The core parts of dolomitic dykes consist mainly of dolomite, with accessory magnesite and strontianite and very rare non-carbonate minerals. Dolomite defines a mildly porphyroclastic texture. About 0.5mm pockets of calcite and smaller dolomite with irregular shapes are distributed heterogeneously in the interstices of dolomite grains. From core to rim of dykes, the size of pockets increase, whereas dolomite grains shrink. This clearly shows that the calcite grows at the expense of dolomite in the dolomitic dykes. This clearly shows that the calcite grows at the expense of dolomite and confirms the view that dolomite can be replaced by calcite and the reverse is not true (Harmer and Gittins, 1997). Three types of alteration are distinguished. The most popular one is called as alkali altered rock, consisting of diopside + tremolite + potassium feldspar + albite + calcite with REE-rich allanite, apatite, epidote, and titanite as common accessory minerals. The altered rock occurs at inner side of alteration haloes around the dolomite dykes, varying in grain size from fine to coarse grained, and possesses granoblastic texture. Under microscope, the diopside grains coexist with potassium feldspar and albite, and REE-rich allanite occurs as fine-grained aggregates within or around the albite grains suggesting that the alkali alteration is accompanied with formation of REE-bearing minerals. The dolomite dykes have very high MgO, low SiO2, Al2O3, CaO and alkali contents relative to the average magnesiocarbonatite composition given by Woolley and Kempe (1989), specifically, the MgO and CaO contents are similar to those of stoichiometrical dolomite. Moveover, the dolomitic rocks have remarkably low concentrations of most incompatible elements. Recently the experiments and observations show that the dolomitic rocks with compositions of stoichiometrical dolomite and very low concentrations of most incompatible elements should likely be crystalline cumulates from melts, not quenched liquids. The calcite dykes have high SiO2 content, in the range of 7-13 wt%. The MgO and CaO contents of the calcitic dykes are also similar to those of end member calcite. Furthermore, these rocks also have very low REE concentrations. These rocks are also crystalline cumulates probably. The calcite rocks have higher REE concentrations than the dolomite dykes, but their chondrite-normalized patterns are perfectly parallel, suggesting that they came from similar origin. Experiment has also confirmed that if rising dolomitic melts percolate through metasomatic mantle rocks from about 70 km toward the Moho at 35-40 km, calcitic magmas with high silicate content can be produced. Here the calcitic rocks of the eastern Himalayan syntaxis have high SiO2 content, consistent with the experiment of Lee and Wyllie (2000). The alkali altered rock has much higher REE concentrations than the dolomitic rocks, but their chondrite-normalized patterns are perfectly parallel, suggesting that the REE were transported from the dolomitic rocks to the altered rock during precipitation of the dolomitic magmas. Compared with the country rock, the altered rock has relatively high MgO, CaO and loss of ignition contents, and low SiO2 and Al2O3 contents, suggesting that the fluid of carbonatite magma plays an important role in the formation of the altered rock. Compared with typical fenite, however, the altered rock has low contents of alkali components, low modal albite and K-feldspar and no aegirine or aegirinic pyroxene. Furthermore, the REE contents of the rock are also low relative to the average REE contents of carbonatites compiled by Woolley and Kempe (1989). These suggest that during emplacement of carbonatite magma, the fenitizing fluid coming from carbonatite magma is not sufficient, reacting with country rocks, to form the typical fenite. Moreover, the fenitizing fluid should contain low concentrations of most incompatible elements, such as alkali elements and REE. Concerning that the concentrations of most incompatible elements of eastern Himalayan dolomite carbonatites are also low, it is, therefore, expected that during emplacement, the magmas are crystal-rich with low concentrations of most incompatible elements. On the basis of these lines of evidence and in combination with experiments, a part of history of the carbonate magmas evolution is outlined as follows: The dolomites precipitated from parental magmas, probable magnesiocarbonatite magmas, to form cumulate mushes or dolomite-rich magmas, leaving REE-rich (?) residual liquid in mantle. The residual liquid may react with mantle rocks at first, releasing sufficient CO2 to enhance crack propagation at mantle. The cracks are likely channels for the dolomite-rich magmas rising. The dolomite-rich magmas may have minor non-carbonate phases, such as olivine and spinel, and their compositions are near to that of stoichiometric dolomite with low concentrations of incompatible elements. A small amount of the dolomite-rich magmas react with mantle country rocks when they rise through mantle along the cracks or faults of mantle, and to produce calcitic magma with high SiO2 content. Although a large amount of the dolomite-rich magmas were not changed into calcitic magma completely, the dolomites in the magmas have experienced the metasomatism to form calcite grains in the dolomite magmas. Additionally, it is expected that the dolomite-rich magmas should rise through the mantle along cracks quickly, otherwise the dolomite-rich magmas can not escape from mantle. When the dolomite-rich magmas and their metasomative products, calcic siliceous carbonatite magmas, intruded the granulite facies rocks, precipitation took place, the intercumulus liquids have, however, moved outside from the carbonatite bodies, brought most incompatible elements, and reacted with the granulite facies rocks to form the altered rocks.